A HEALTH TECHNOLOGY ASSESSMENT OF LOVAAS AUTISM TREATMENT: THE ROLE OF EVIDENCE IN LEGAL, HEALTH POLICY AND HEALTH CARE CONTEXTS by SANDRA LYNN BEARD B.S.N., The University of British Columbia, 1997 M.S.N., The University of British Columbia, 1999 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in The Faculty of Graduate Studies (Health Care and Epidemiology) THE UNIVERSITY OF BRITISH COLUMBIA APRIL 2007 © Sandra Lynn Beard, 2007 ABSTRACT In 1998 parents of autistic chi ldren l a u n c h e d a Charter of Rights and Freedoms c ha l l enge against the Province of BC for failing to fund Lovaas Autism Treatment (LAT) (Auton et al.). Al though initially successful, in 2004 the Supreme Court of C a n a d a overturned the lower courts' decisions a n d re jec ted the parents ' c laims for pub l i c funding. In add i t ion to the Charter issue, these lega l p roceed ings also highl ighted the discourse over judic ia l pol icy mak ing a n d the Courts ' interpretat ion of m e d i c a l e v i d e n c e - specif ical ly, the effect iveness of LAT. The use of m e d i c a l e v i d e n c e in law h a d b e e n identif ied as a n issue by both the A m e r i c a n Institute of M e d i c i n e (IOM) a n d the A g e n c y for Hea l thcare Research a n d Quality (AHRQ). This thesis was des igned to e x p a n d on the lOM/AHRQ's previous work by investigating the conceptua l iza t ions a n d processes used by law, health pol icy a n d health c a r e within the context of the Auton lega l p roceed ings in order to ga in a n understanding of h o w e a c h d o m a i n seeks, understands a n d appl ies e v i d e n c e . This was a c c o m p l i s h e d in two parts. First, the lega l d imension of a comprehens i ve heal th t e chno logy assessment (HTA) framework utilized a qual i tat ive g r o u n d e d theory m e t h o d o l o g y to examine part ic ipant interviews a n d lega l documents . This analysis resulted in a c o n c e p t u a l framework of scientific e v i d e n c e pathways that further de f ined , contextua l ized a n d dimensional ized the p h e n o m e n a of seeking, unders tand ing, a n d app ly ing e v i d e n c e within the three sectors. G r o u n d e d theory p roved to b e a n ef fect ive a p p r o a c h for exploring the lega l context a n d serves to b r o a d e n the s c o p e of e v i d e n c e HTA researchers c a n offer. S e c o n d , the effectiveness dimension of the HTA framework e m p l o y e d the methods of systematic review a n d crit ical appra isa l to investigate the current state of know ledge on the effect iveness of LAT. This analysis c o n c l u d e d that there remains a pauc i t y of rigorously des igned studies d u e to ongo ing me thodo log i c a l limitations. Overal l , the strength of the b o d y of e v i d e n c e on LAT was poo r a n d d id little to a d v a n c e its effectiveness c la ims. Together, these two analyses p rov ided a n u p d a t e on the effectiveness of Lovaas Autism Treatment; insight into h o w the e v i d e n c e of LAT was h a n d l e d in the Auton c a se ; a n d a dep i c t i on of h o w the sectors of law, health pol icy a n d heal th c a r e concep tua l i ze the evidentiary process in gene ra l . TABLE OF CONTENTS Abstract Table of Contents List of Tables List of Figures vi Acknowledgements vii Dedication ix CHAPTER 1 Background, Objectives and Rationale 1.1 Autism Spectrum Disorder 1 1.2 Early Intensive Behavioural Intervention/Applied Behavioural Analysis 3 1.2.1 Lovaas Autism Treatment 5 1.3 Precedent Lovaas Research 7 1.3.1 Lovaas, 1987 7 1.3.2 McEachin, 1993 8 1.4 Critiques of Lovaas & McEachin 11 1.5 BC Autism Intervention Programs 17 1.6 Policy Framework 20 1.7 Pertinent Legislative Frameworks 24 1.8 Legal Context 27 1.8.1 Auton : 27 1.8.2 Anderson 31 1.9 Canadian Legal System 37 1.9.1 Canadian Charter of Rights and Freedoms 39 1.10 Other Health Care Charter Cases 41 1.11 Legal Reviews of Auton 48 1.12 Objectives 52 1.13 Rationale 52 1.14 Research Questions.. 53 CHAPTER 2 Conceptual Framework 55 2.1 Understanding the Role of Evidence in Law, Health Policy and Health Care 55 2.1.1 Concept of Evidence 55 2.1.2 Opposing Conceptualizations of Evidence 57 2.1.3 Conceptual Background 57 2.2 Processes for Seeking, Understanding and Applying Evidence: Legal Context 63 2.2.1 Judicial Policy Making 63 2.2.2 Legal Discovery 64 2.2.3 Rules of Evidence 65 2.2.4 Burden of Proof 68 2.2.5 Standard of Care: Expert Witnesses & Clinical Practice Guidelines 69 2.3 Processes for Seeking, Understanding and Applying Evidence: Health Policy Context 72 2.3.1 Policy Process: The Policy Cycle 72 2.3.2 Evidence-based Decision Making: Health Technology Assessment 81 2.4 Processes for Seeking, Understanding and Applying Evidence: Health Care Context 86 2.4.1 Peer Review Process 86 2.4.2 Consensus Development 88 2.4.3. Evidence-based Medicine 90 184.108.40.206 Randomized Controlled Trials 98 220.127.116.11 Alternate Study Designs 99 18.104.22.168 Clinical Practice Guidelines 99 2.5 Conceptual Framework 101 CHAPTER 3 Research Design & Methodologies 103 3.1 Grounded Theory: The Legal Context 104 3.1.1 Grounded Theory Methodology 104 3.1.2 QSR N6 qualitative research software 113 3.2 Systematic Review and Critical Appraisal: The Effectiveness Context 115 3.2.1 Systematic Review 115 3.2.2 Study Quality Assessment 118 3.2.3 Critical Appraisal 122 3.3.4 Summary of Research Design 124 CHAPTER 4 Results 126 4.1 Analysis of the Role of Evidence 126 4.1.1 Descriptive Analysis of Conceptual Categories 127 22.214.171.124 Perspective of Evidence 127 126.96.36.199 Rules of Evidence 128 188.8.131.52 Process of Change 130 184.108.40.206 Adjudicators of Evidence 132 220.127.116.11 Decision-Makers of Evidence 135 18.104.22.168 Focus of Application of Evidence 139 22.214.171.124 Judicial Policy Making 140 4.1.2 Scientific Evidence Pathways 141 4.2 Analysis of Lovaas Effectiveness 142 4.2.1 Search Results and Study Selection 142 4.2.2 Study Quality Assessment 144 4.2.3 Critical Appraisal of Study Methods 146 CHAPTER 5 Discussion and Conclusions 164 5.1 Discussion 164 5.1.1 The Role of Evidence in Legal, Health Policy and Health Care Contexts: Seeking, Understanding and Applying Evidence 164 5.1.2 Application of Concepts to a Judicial Policy Making Case Study: Auton et al 171 5.1.3 Effectiveness of Lovaas Autism Treatment 174 126.96.36.199 Descriptive Data Synthesis 174 188.8.131.52 Clinical Heterogeneity 174 184.108.40.206 Methodological Heterogeneity 176 220.127.116.11 Heterogeneity in Results 178 18.104.22.168 The Current State of Knowledge on Lovaas Effectiveness Evidence 178 22.214.171.124.1 Advances and Improvements 179 iv 126.96.36.199.2' Ongoing Methodological Issues 179 188.8.131.52.3 Included Studies'Reported Limitations 180 184.108.40.206.4 Unanswered Research Questions 181 220.127.116.11 Strength of Body of Evidence 182 18.104.22.168.1 Quality of the Included Trials 182 22.214.171.124.2 Size and Significance of Observed Effects 182 126.96.36.199.3 Consistency of Effects Across Trials 184 188.8.131.52.4 Dose-response Relationships 184 184.108.40.206.5 Other Plausible Competing Explanations 184 5.2 Conclusions 186 5.3 Strengths and Limitations 189 5.4 Further Research 189 References 191 Appendices 205 Appendix A The Role of Evidence in Legal, Health Policy and Health Care Contexts ! 205 Appendix B Scientific Evidence Pathways: Initial Framework 206 Appendix C Participant Letter of Introduction 207 Appendix D Participant Consent Form 208 Appendix E Search Strategy 210 Appendix F Study Selection Form 219 Appendix G Study Quality Assessment Tool for RCTs: Modified Jadad Scale 221 Appendix H Study Quality Assessment Tool for Observational Studies: Reisch 223 Appendix I Data Extraction Form 228 Appendix J Seeking, Understanding and Applying Evidence: Key Concepts and Findings 237 Appendix K Interview Questions 243 Appendix L Scientific Evidence Pathways: Revised Framework 249 Appendix M Excluded Retrieved Papers 250 Appendix N Summary of Included Studies 254 Appendix O Analysis of Heterogeneity Between Groups 272 Appendix P Direction of Biases 279 Appendix Q UBC Research Ethics Board Certificate of Approval 281 LIST OF TABLES Table 1.1 Summary of legal cases 36 Table 2.1 Policy cycle 72 Table 2.2 Decision making strategies 78 Table 2.3 Comprehensive health technology assessment framework 83 Table 3.1 Study selection criteria 116 Table 3.2 AHRQ coding criteria 118 Table 3.3 Reisch quality assessment tool: revised criteria 122 Table 4.1 Summary of reasons for exclusion of studies 143 Table 4.2 Reisch quality assessment indices for the included observational studies... 144 Table 4.3 Jadad quality assessment indices for the included randomized controlled trials 145 Table 5.1 Statistically significant mean between-group differences at follow-up 183 vi LIST OF FIGURES Figure 2.1 The context, actors and level of application of the evidentiary process 101 Figure 3.1 The indicator-concept model 109 Figure 3.2 Study Design: The legal and effectiveness contexts of the comprehensive HTA framework 124 Figure 4.1 The context, actors and focus of application of the evidentiary process (revised) 126 vii ACKNOWLEDGEMENTS I would be remiss if I did not express my sincere gratitude to the key people who assisted me with this thesis. Many thanks to Mary-Doug Wright for her proficiency in systematic literature searches and Carolyn Green for her assistance as my second reviewer. Special appreciation also goes to my thesis committee - Anthony Sheppard, for guiding me through the unfamiliar maze of law and Ken Bassett, for his prowess with systematic reviews. And most importantly, my sincere thanks to Arminee Kazanjian, my thesis supervisor, for her kindness and patience as she guided me towards my vision for my research. DEDICATION It is with great love and admiration that I dedicate my thesis to my husband Jonathan and our children Jordan and Devon. During difficult times, both academically and personally, their unconditional love and support were what kept me focused. I am also extremely grateful to a late friend - Judi Moore - who sparked my interest in further education and continuously challenged me to reach for higher goals. And to my parents - Earle and Lilian Fleming - for their genuine interest in my work and their encouragement to never give upon a dream. CHAPTER 1: B A C K G R O U N D , OBJECTIVES A N D RATIONALE Autism Spectrum Disorder Autism Spect rum Disorder (ASD) is a constel lat ion of three pervasive d e v e l o p m e n t a l disorders (PDD) of c h i l d h o o d : (a) autistic disorder (AD), (b) Asperger syndrome, a n d (c) pervasive d e v e l o p m e n t a l disorder-not otherwise spec i f ied (PDD-NOS). Other condit ions that compr ise the b roade r definition of PDD inc lude Rett syndrome a n d ch i l dhood disintegrative disorder (CDD). Autism was first desc r ibed in 1943 by Dr. Leo Kanner, w h e n a.small g roup of chi ldren were found to display abno rma l character ist ics such as extreme aloofness, indi f ference to other p e o p l e , minimal eye c o n t a c t , severe l a n g u a g e deficits, a t yp i ca l responses to stimuli, minimal p re tend or imaginat ive play, a n d lack of desire to c o m m u n i c a t e (Amer ican A c a d e m y of Pediatrics, 2001). However , it was not until 1980 when the term "infantile aut ism" was first r e cogn ized as a d iagnost i c criterion in the Diagnostic and Statistical Manual of Mental Disorders (DSM). S ince then, the definition a n d criterion have b e e n e x p a n d e d to incorporate milder a n d more c o m m o n forms of this cond i t ion . PDD are lifelong neuro log ica l disabilities of unknown et io logy that general ly result in variations of impa i red social izat ion, c o m m u n i c a t i o n , a n d behav iour . Because autism is a spect rum disorder, symptoms are a he te rogeneous mix with distinct variations a m o n g chi ldren. There is no universal d e v e l o p m e n t a l de l ay or behav iour that classically dep ic ts autism. Very young chi ldren typical ly present with symptoms of d e l a y e d social izat ion such as poor eye c o n t a c t a n d a disinterest in their surroundings. They often a p p e a r to b e inwardly f o cused a n d resist phys ica l c o n t a c t . C o m m u n i c a t i o n is d e l a y e d a n d o n c e it appea r s it is rote, repetit ive, a n d lack ing purpose. These chi ldren se ldom interact with others a n d display behaviours such as rocking, f l app ing of their hands a n d rituals such as repetitively s tack ing blocks. Intell igence c a n vary from severe menta l retardat ion to superior intel lectual abilities. More specif ical ly , AD is cha rac te r i zed by d e v e l o p m e n t a l delays in social interact ion, c o m m u n i c a t i o n , a n d imaginat ive play. These chi ldren typical ly demonst ra te s tereotyped behaviours , interests a n d activities. A l though these behaviours o c c u r in late infancy, there are cases where d e v e l o p m e n t a l milestones that were o n c e a c h i e v e d regress or are lost in their entirety. Studies are currently underway to identify early signs of autism (Zwa igenbaum et al. , 2005). Some of this preliminary research indicates that by the a g e of 12 months infants w h o were later d i a g n o s e d as autistic dif fered from other infants in the fol lowing areas: (a) behav ioura l markers such as a typ ica l eye c o n t a c t , orienting to n a m e , socia l smiling; (b) p ro longed la tency to d i s e n g a g e visual at tent ion; (c) t empe ramen t (passivity fo l lowed by extreme distress react ions); a n d (d) d e l a y e d expressive a n d recept i ve l a n g u a g e . 1 Asperger syndrome, on the other h a n d , is limited to impairments in soc ia l interactions such as pee r relationships, a lack of empa thy , a n d obsessions abou t cer ta in topics. L a n g u a g e d e v e l o p m e n t is not a f f e c t ed a n d Asperger 's chi ldren are typical ly of a v e r a g e to a b o v e a v e r a g e inte l l igence. It is not yet known whether Asperger is ac tua l ly a higher funct ioning form of autism or i n d e e d a unique condi t ion of its own . The third ASD - pervasive d e v e l o p m e n t a l disorder not otherwise spec i f i ed (PDD-NOS) - is a n a typ ica l form of autism that is often used as a diagnosis w h e n chi ldren d o not c lear ly fall within the de f i ned criteria of the other ca tegor ies . With all three classifications of ASD, a multi-disciplinary t e a m c a n usually m a k e a diagnosis by the t ime a chi ld is two years o ld (Lord, 1995; Lord & M c G e e , 2001). The Diagnostic and Statistical Manual of Mental Disorders (DSM IV-TR) lists twelve criteria for autistic disorder. These criteria are d i v ided into: (a) impairments in qual i tat ive socia l interactions, (b) impairments in qual i tat ive c o m m u n i c a t i o n , a n d (c) restricted, repetit ive a n d stereotyped patterns of behav iour , interests a n d activities. Because the DSM IV-TR criteria were d e v e l o p e d for chi ldren 3 years of a g e a n d older, a t tempts to d i agnose younger chi ldren m a y b e difficult until all character ist ics are demons t ra ted . Therefore, a diagnosis of PDD-NOS m a y b e proffered until the chi ld "grows into" the full diagnosis of AD . However, assigning a preliminary diagnosis remains somewha t controversial . O n the o n e h a n d , while there is some consensus that the earlier intervention is b e g u n the more benef ic ia l the ou t comes , this must b e b a l a n c e d against the socia l a n d f inancia l impac ts of a false positive diagnosis. There are conf l ic t ing opinions on p reva l ence a n d i n c i d e n c e rates for autism. O n e review of twenty-three English l a n g u a g e ep idemio log i ca l surveys publ ished b e t w e e n 1966 a n d 1998 found that p r e va l ence rates increased e a c h successive publ i ca t ion year (Fombonne, 1999). In that review, the m e d i a n p reva l ence rate for autism rose from 5.2 per 10,000 for studies publ ished before 1989 to 7.2 per 10,000 for studies after 1989, while the i n c i d e n c e of autism r ema ined stable at 0 .2% (with rates increasing to 3-7% for siblings of autistic chi ldren). However , the author con tends that c la ims of an autism e p i d e m i c are u n f o u n d e d d u e to m e t h o d o l o g i c a l limitations of the existing d a t a , c h a n g e s in c a s e definition, improved c a s e recogni t ion, b roade r c o n c e p t of autism, a n d recogni t ion of autism amongs t normally intelligent p e o p l e (Fombonne, 2001). An u p d a t e to the 1999 F o m b o n n e review saw the p r e va l ence rate rise to 10.0 per 10,000 (Fombonne, 2003). However, a g a i n the author attributes the rise to improved c a s e definition a n d awareness of autism. In BC, approx imate ly 262 n e w cases of ASD are p r ed i c t ed e a c h year, y ielding a p r e va l ence of 0.65% (Wellington, 2003). Autism targets males more than females with a ratio of 3.8:1, a n d is not cor re la ted to a n y spec i f ic e thnic g roup nor to soc ia l factors such as a family's i n c o m e , lifestyle or educa t i ona l level. 2 Early Intensive Behavioural Intervention/Applied Behavioural Analysis Early Intensive Behavioural Intervention (EIBI) is a generic term referring to-comprehensive and intensive forms of behavioural interventions; another synonymous term is Applied Behaviour Analysis (ABA). With research dating back to the 1960s, EIBI/ABA claims to be one of the most efficacious treatment modalities for children with the life-long affliction of autism (Heflin & Simpson, 1998). ABA is the science of human behaviour. It is an objective, systematic process of applying interventions based on the principles of learning theory in an attempt to improve, socially significant behaviours (Association for Science in Autism Treatment, 2004). Behaviour modification enlists three general approaches to treatment: (a) operant conditioning, (b) respondent conditioning, and (c) cognitive conditioning (New York State Department of Health, 1999). Most of the current practices focus on operant conditioning to: 1. Increase certain behaviours 2. Teach new skills (e.g., communication, social skills) 3. Maintain behaviours (e.g., teaching self-control) 4. Generalize or to transfer behaviour from one situation to another (e.g., from the resource room to the mainstream classroom) 5. Restrict or narrow conditions under which interfering behaviours occur (e.g., modifying the learning environment) 6. Reduce interfering behaviours (e.g., self injury) (Association for Science in Autism Treatment, 2004) Since autistic children typically exhibit maladaptive behaviours that can be socially inappropriate, self-injurious, or physically and psychologically demanding on their parents, . immediate treatment that focuses on changing this behaviour is the priority. Behavioural intervention is usually the treatment of choice for autistic children. EIBI/ABA therapy is strictly structured and delivered at high intensity in order to reinforce learned behaviours. Interventions are designed to assist children to obtain and/or alter behaviours through a process of reinforcing adaptive responses and suppressing maladaptive behaviours. Two major assumptions underlie the foundation of EIBI/ABA. The first assumption is grounded in theories of neurobiological development, critical and sensitive periods, attachment theory, infant learning paradigms, prevention theory, and intervention research. This is the assumption that EIBI/ABA should be administered during the preschool years when the brain is most amenable to structural and functional change (Bailey, Aytch, Odom, Symons, & Wolery, 1999). The second set of assumptions is derived from theories of family systems, stress and coping, ecological development, parental roles, and cultural influences, and holds as its core principle the concept of family-centered practice (Bailey, et al., 1999; Guralnick, 2001). EIBI/ABA can be delivered via three different instructional methodologies: (a) discrete trial training (DTT), (b) naturalistic teaching or errorless teaching, and (c) incidental learning. DTT 3 involves breaking down skills and teaching them one step at a time with the child mastering the simple skills before progressing on to more complex ones. This takes place in the form of a "training trial" format and consists of four components: 1. The teacher or therapist presents a brief, distinctive instruction or question (stimulus). 2. The instruction is followed by a prompt, if the child needs one, to elicit the correct response. 3. The child responds correctly or incorrectly (the response). 4. The teacher or therapist provides an appropriate "consequence". (Association for Science in Autism Treatment, 2004) DTT is also known as the "ABC model," whereby each trial consists of: (a) an Antecedent (a "directive"); (b) a Behaviour (a "response"); and (c) a Consequence (a "reaction") (Autism Society of America, 1998). DTT is also referred to as the "Clinical/Prescriptive" method and as "Formal Compliance Training". The discipline of ABA focuses on reliable measurements and objective evaluation of observable behaviour. Outcomes are operationalized in quantifiable ways in order to measure frequencies and durations of specific behaviours. Evaluation of each individual trial is a critical part of therapy. If a response is not elicited within 5 seconds (or the response is incorrect) that trial ends and a new one begins. Children's response to therapy is monitored by detailed data collection. Once a skill is mastered (for example, a correct response achieved at least 80% of the time, as judged by two therapists), then that particular-skill is added to a maintenance schedule so that learned skills are revisited to prevent regression. In addition, these skills are adapted in order to ensure generalization to different people, settings, and materials (Autism Society of America, 1998). Behavioural programs can take place in a variety of settings. Typically, ABA begins in the home when children are young, as this is a natural learning environment. However, this approach involves a tremendous amount of input and support from the family. Other programs may advocate a school or clinic-based setting as principles of ABA can be more consistently adhered to. In any case, the program invariably transitions to community and educational settings once the children have learnt the basic behavioural and social skills. . Properly designed and executed ABA programs [should] contain many if not all of the components of effective treatment approaches... namely: individualized instruction tailor-made to address the specific needs of the child, behaviourally based methodology, low student-teacher ratio, early treatment, and family involvement. (Autism Society of America, 1998) Programs are supervised by an ABA consultant who typically receives preparatory training in a university program. It is important to note that there are many different interpretations and contexts of applied behavioural analysis. Many speech and language therapy programs, in 4 add i t ion to e d u c a t i o n a l interventions, are also g r o u n d e d in the principles of behav ioura l therapy. In add i t ion , some autism intervention programs c o m b i n e EIBI/ABA with other a p p r o a c h e s . Lovaas Autism Treatment While psychologist Dr. O. Ivar Lovaas ref ined the Lovaas Autism Treatment (LAT) p rog ram at the University of Cal i fornia Los Angeles (UCLA) during the 1980s, its origins are rooted in previous behav ioura l work with institutionalized chi ldren dat ing b a c k to 1963. An important distinction is that LAT is not synonymous with EIBI/ABA; it is a subset b a s e d o n similar principles. LAT employs the intensive, one-on-one DTT me thodo log i c a l a p p r o a c h with behav ioura l interventions carr ied out for 20-40 hours per week by a t e a m of therapists, parents a n d eventual ly, teachers . Its intensity makes it a costly intervention with estimates of b e t w e e n $45,000 - $60,000 C A D per chi ld per year (Ludwig & Harstall, 2000; Supreme Court of BC, 2000). Chi ldren beg in the p rogram as soon as possible after diagnosis (ideally before a g e 4) a n d cont inue for a two to three yea r pe r iod . The p rogram consists of approx imate ly 500 indiv idual tasks to b e learnt. The therapist initially focuses on t ea ch ing self-help a n d recept i ve l a n g u a g e sills, nonverba l a n d verbal imitation, a n d appropr ia te p lay behaviours. Parental invo lvement is c ruc ia l to prov ide consistency in m a n a g i n g behaviours. Next, chi ldren progress towards learning expressive a n d early abstract l a n g u a g e a n d interact ive play with peers eventual ly l ead ing towards a n a d v a n c e d app l i ca t ion , assimilating the skills for a c a d e m i c tasks, social izat ion, cause-ef fect relationships, a n d observat ional learning. In its original format, inappropr ia te behaviours were often m a n a g e d by aversive methods such as shouting " n o " or a s lap on the thigh, but, these have s ince b e e n r e p l a c e d with strategies such as ignoring a n d imp lement ing a time-out. Of interest, one study e x a m i n e d the ef fect of removing negat i ve verba l f e e d b a c k from the Lovaas p rog ram (White, 2000). Results actua l ly showed a trend towards nega t i ve f e e d b a c k increasing skill a c h i e v e m e n t a long with a decrease in the number of m a l a d a p t i v e behaviours. Lovaas attested that only a pract it ioner directly l inked with the original Lovaas program c o u l d c l a im to app l y LAT. Lovaas therapists often carry the des ignat ion of "Cert i f ied A p p l i e d Behaviour Analyst" through spec ia l ized training c o n d u c t e d at UCLA. These consultants are master's p r e p a r e d graduates w h o have under taken a 9-month internship consisting of 30 hours per w e e k of training. Lovaas purports that cert i f icat ion is a crit ical c o m p o n e n t in order to mainta in quality control of this part icular a p p r o a c h to ABA. It is c l a i m e d that, w h e n IBI is imp l emen ted at a n early a g e , some autistic chi ldren will c a t c h up with their normal peers by g r a d e one (Lovaas, 1987); however , to d a t e there is no rel iable e v i d e n c e to substantiate this c l a im (Basset et al . , 2000; ECRI, 2000; Smith, 1999). The centra l point of controversy surrounding IBI is the eff icacy/effect iveness c la ims of the various types of programs. While there are numerous publ ished articles on the top ic , there is a pauc i ty of methodo log i ca l l y val id studies. O n e t e chno logy assessment of comprehens i ve treatment 5 programs in the United States l o c a t e d 528 articles of w h i c h only 18 met the inclusion criterion of incorpora t ing a control group (ECRI, 2000). After a subsequent review of the internal, external, statistical conclusions a n d construct validity of these remaining 18 studies, 13 were found to b e "fatally f l awed" . Three of these final five studies were behav ioura l programs b a s e d o n the Lovaas m e t h o d . Al though the Lovaas g roup s h o w e d a n improvement in IQ scores a n d in some functional parameters , b e c a u s e of study design problems, this ga in c o u l d not b e at tr ibuted to the m e t h o d of t reatment. Another systematic review a n d crit ical appraisa l (Bassett, et a l . 2000) was also successful in uncover ing 1200 EIBI abstracts. (This search strategy was not limited to the United States a n d also inco rpo ra t ed a sea rch for fugitive literature). After the a p p l i c a t i o n of the Nat ional Institute of Health criteria (for eva lua t ion of primary effectiveness da ta ) , the s ame primary Lovaas studies that h a d b e e n u n c o v e r e d by the ECRI report r ema ined . A g a i n , it was d e t e r m i n e d that "while m a n y forms of intensive behavioura l therapy clearly benefit chi ldren with autism, there is insufficient, scientifically-valid effectiveness e v i d e n c e to establish a causa l relationship b e t w e e n a part icular p rog ram of intensive, behav ioura l treatment, a n d the a c h i e v e m e n t of 'normal funct ioning ' " (p. ix). Smith (1999) c o n d u c t e d a systematic review of early intervention therapies b e t w e e n 1980 a n d 1999 a n d u n c o v e r e d nine "behaviour analyt ic t reatment programs", five of w h i c h were Lovaas -based (Anderson et. al . , 1987; Birnbrauer & L e a c h , 1993; Lovaas , 1987; M c E a c h i n et al . , 1993; Sheinkopf & Siegel, 1998). Only the Lovaas a n d the follow-up M c E a c h i n studies were b a s e d on the original U C L A Lovaas project; the others were partial repl ica t ion studies that inco rpora ted varying intensities of t reatment a n d levels of therapist training. Al though Smith was a co-author of the M c E a c h i n study, he general ly c o n c u r r e d with previous reviewers' criticisms of the m e t h o d o l o g i c a l flaws of these studies. W h e n discussing the repl icat ion studies he c o n c l u d e d that under some c i rcumstances early intervention might not yield positive results. However , w h e n speak ing a b o u t the original U C L A studies, he no t ed that to d a t e this p rog ram h a d the most favourable ou t comes a n d the strongest me thodo logy . Yet, he a p p e a l e d to the research communi ty to design future studies that were scientifically rigorous. 6 Precedent Lovaas Research Lovaas, 1987 In 1987 Lovaas publ ished the results of his study on intensive behav iour modi f i ca t ion in the Journal of Consulting and Clinical Psychology. This report (a long with its follow-up in 1993 by M c E a c h i n et al.) went on to b e c o m e p re ceden t research studies for the field of a p p l i e d behav ioura l analysis. Lovaas ' research was the cu lminat ion of a behav ioura l intervention project that b e g a n in 1970 at the University of Cal i fornia, Los Angeles . Lovaas ' study c o m p a r e d two groups: (a) exper imenta l group (>40 hrs/week of 1:1 intensive treatment) ; a n d (b) control #1 (<10 hrs/week of the same 1:1 intensive treatment, plus a variety of other treatments). Control #1 was used to determine the rate of spontaneous improvement in autistic ch i ldren. Lovaas also inco rpora ted a s e c o n d control g roup of chi ldren w h o h a d b e e n se l ec ted from a cohor t of chi ldren that h a d previously b e e n studied by F reedman et a l . (1985). This g roup was assembled to counter potent ia l criticisms that chi ldren referred to the study might h a v e const i tuted a subgroup of chi ldren with either favourab le or unfavourable ou t comes . Lovaas desc r ibed the inclusion criteria for this control group, a n d stated that these chi ldren were t rea ted like Contro l G roup 1 subjects with the excep t ion be ing that the intervention was administered outside of the Young Autism Project. Parents of chi ldren in both groups r e ce i ved extensive training a n d were act ive ly involved in their chi ld 's treatment p lan . Physical aversives were only utilized on the exper imenta l group. Treatment intensity was r e d u c e d to 10 hrs/week in kindergarten for those chi ldren progressing normally; others con t inued to rece ive 40 hrs/week forrnore than 6 years. Assignment to t reatment groups was not r andomized , " d u e to parent protest a n d eth ica l cons idera t ion" (p. 4). Instead, assignment was b a s e d on the avai labi l i ty of therapists. Lovaas reported, " b e c a u s e fluctuations in staff availabil ity were not assoc i a ted in any w a y with client character ist ics it was assumed that this assignment wou ld p r o d u c e unb iased groups" (p. 4). The inclusion criteria consisted of: (a) an i ndependen t diagnosis of autism from a m e d i c a l doc to r or PhD psychologist (based on DSM-III criteria), (b) chrono log i ca l a g e of less than 40 months if mute a n d less than 46 months if echo la l i c , a n d (c) prorated menta l a g e of 11 months or more at a chrono log i ca l a g e of 30 months. Lovaas reported that the latter inclusion criterion resulted in 15% of referrals be ing e x c l u d e d . Pre-treatment measures c o n c l u d e d that the exper imenta l g roup a n d control #1 were similar on all measures at intake, with the excep t ion of ch rono log i ca l a g e in w h i c h the control subjects we re 6 months o lder on a v e r a g e (this was later shown to b e not significant). Assessments were c o n d u c t e d in a b l inded fashion w h e n the chi ldren were b e t w e e n 6 a n d 7 years of a g e (when they wou ld normally h a v e c o m p l e t e d g r a d e one) . O u t c o m e measures consisted of IQ scores a n d first-grade p l a c e m e n t d a t a . A chi ld r e ce i ved a score of 3 (and were classif ied as normal functioning) if they a c h i e v e d a normal IQ score , h a d c o m p l e t e d 7 g r a d e one in a normal class in a school for normal chi ldren a n d h a d b e e n m o v e d forward to g r a d e two. A score of 2 was ass igned to those chi ldren w h o a t t e n d e d g r a d e one in a n aphasia class ( l anguage d e l a y e d , l a n g u a g e h a n d i c a p p e d or learning d isabled) . A n d finally, a score of 1 was g iven to chi ldren w h o h a d b e e n p l a c e d in autist ic/retarded classes or if the chi ld 's IQ score fell into the severely re ta rded range . Lovaas repor ted that at follow-up the exper imenta l group was significantly higher than both control groups on the variables of c lassroom p l a c e m e n t a n d IQ. Out of the 19 exper imenta l chi ldren, 9 (47%) were reported to h a v e successfully c o m p l e t e d g r a d e one in a normal c lassroom a n d a d v a n c e d into g r a d e two. In add i t ion , these chi ldren a c h i e v e d an a v e r a g e or a b o v e a v e r a g e score on IQ tests (M=107). Eight chi ldren (42%) passed first g r a d e in a n aphas i c c lassroom with IQ scores within the mildly re ta rded range (M=70). A n d , two chi ldren (10%) a t t e n d e d classes for autist ic/retarded chi ldren with profoundly re ta rded IQs <30. There were no dif ferences repor ted b e t w e e n the two control groups at either intake or follow-up. C o m b i n e d follow-up d a t a from the control groups showed that only 1 ch i ld (2%) r e a c h e d normal funct ioning, 18 chi ldren (45%) h a d b e e n taught in aphas i a classes, while 21 subjects (53%) were in classes for autist ic/retarded chi ldren. Lovaas c o n d u c t e d a study within a study w h e n he tested the hypothesis that aversives (shouting " N o " , or a slap on the thigh), in response to self-stimulatory or aggressive a n d noncomp l i an t behaviour , wou ld p r o d u c e significant changes . Four chi ldren from the •experimental g roup a n d another four from Control #1 were initially t reated without aversives a n d then later, aversives a d d e d to the treatment p ro toco l . Behavioural o u t c o m e s were measured a n d c o m p a r e d at both these time points. Lovaas repor ted that both groups m a d e " sudden a n d stable reduction[s] in inappropr ia te behaviours a n d sudden a n d stable increases in appropr i a te behav iours" (p. 7). Add i t iona l findings of the study were that pro-fated menta l a g e was significantly re lated to o u t c o m e for bo th the exper imenta l a n d control #1 groups a n d that a b n o r m a l s p e e c h was significantly re la ted to o u t c o m e in control #1. Chrono log i ca l a g e at onset of intervention was not re la ted to o u t c o m e (which ruled out the issue of di f ferences in ch rono log i ca l a g e b e t w e e n groups at intake). A n d finally, Lovaas c l a i m e d it was possible to pred ic t the nine chi ldren that h a d a c h i e v e d normal funct ioning b a s e d on an analysis of eight pre-treatment variables. McEach in , Smith & Lovaas, 1993 A two-part follow-up of the original chi ldren from Lovaas'' 1987 study was publ ished in 1993 in the American Journal on Mental Retardation. For. the first part of the research the investigators e x a m i n e d whether the gains m a d e by the exper imenta l g roup h a d b e e n ma in ta ined after t reatment h a d e n d e d . For the s e c o n d part, the nine chi ldren in the exper imenta l g roup that h a d a c h i e v e d "best o u t c o m e s " were e x a m i n e d for signs of residual 8 autistic symptomatology (as compared to a cohort of children without behavioural problems -"non-clinical comparison group"). For McEachin et al.'s study, the mean age of the experimental group was now 13 years and the children had been out of treatment for an average of 5 years (range: 0-12 years). More specifically, the group of best outcome children had been without treatment for 5 years on average (range: 3-9 years). In comparison, the mean age of the control group was 10 years with a mean time out of treatment of 3 years (range: 0-9 years). The researchers acknowledged that their experimental group was older and had been out of treatment longer than the control group and explained that the age difference was due to the original treatment assignment procedure. With the Lovaas study, initially all referrals were assigned to the experimental group because therapists were available. However, the authors of McEachin et al. determined statistically that there'was no association between the order in which the children were referred and intake or outcome IQ. Measures obtained for this follow-up study included an assessment of school placement, IQ, the Vineland Adaptive Behaviour Scales, and the Personality Inventory for Children. Then, to compare the best outcome subjects with the non-clinical comparison group, a comprehensive clinical rating scale was developed that captured the typical areas of problems for autistic children. All these tests were administered and scored by blinded assessors for the experimental group, while staff members of the program and outside agencies tested and scored the control group. Statistical tests indicated that there was no significant differences between children who had been evaluated by project staff and those evaluated by outside agencies. The authors reported that at follow-up, one of the nine best outcome children in the experimental group had regressed and subsequently p laced in a special education class. However, one of the other ten experimental children (who had not reached best outcome) had in fact improved and was now in a regular class. The remaining experimental children had maintained their school placement. Therefore, the figure of 47% of children in the experimental group attaining normal classroom placement was upheld. In comparison, none of the nineteen children in the control group were in regular classes. In regard to IQ, the experimental group had a significantly higher mean IQ than did the control group (30 points higher) prompting the investigators to conclude that the experimental group had maintained its gains in intellectual functioning. In addition, the experimental group showed a higher level of functioning, more adaptive behaviours, and less maladaptive behaviours than the control group. And finally, related to personality functioning, the two groups did not differ on overall scale elevation, however, there were differences between groups on two specific scales - the psychosis and the somatic concerns scales - with the control group displaying higher levels of psychosis and lower levels of somatic complaints. 9 Upon contrasting the best outcome children against the non-clinical comparison group, the best outcome subjects had IQs in the high end of the normal range. In addition, most of these children also scored above average on the Vineland Adaptive Behaviour Scale. However, although there were three best outcomes children that had achieved only marginal scores on some of the Vineland subscales, on average, the group achieved composite scores within the normal range. On the Vineland Maladaptive Behaviour Scale, three children displayed clinically significant maladaptive behaviour, but as a group, the best outcome children did not display clinically significant aberrant behaviour. And finally, in regard to personality functioning, some deviations from average were noted on the personality test and the clinical ratings scales for the best outcomes children. However, the authors contributed this finding to the extreme scores of one child as opposed to problems with the entire group. Although they acknowledged that group averages were not usually interpreted in this manner, they backed up their position by stating that statisticians concur that, "there are many times when group averages represent the performance of few or no subjects within the group" (p. 370). In this situation, they attributed this variance almost exclusively to one subject. In sum, eight children (42% of the experimental group) "may be judged to have made major and enduring gains and may be described as 'normal-functioning'" (p. 368) as opposed to none of the control group achieving this goal. 10 Critiques of Lovaas & McEachin Lovaas ' original 1987 research study, a l ong with M c E a c h i n et al.'s (1993) follow-up, cont inue to b e c i t ed a n d cr i t iqued within a c a d e m i c , legal , a n d heal th c a r e forums. A l though e a c h respect ive discipline possesses their own unique criticisms, this next sect ion will discuss some of the m e t h o d o l o g i c a l issues within an a c a d e m i c context , as this should b e the initial step before any heal th or lega l pol icy dec is ion is cons ide red . Rebuttals by Lovaas a n d his proponents are also i n co rpo ra t ed . Lack of Randomization The most often c i ted criticism of Lovaas ' 1987 study was the lack of randomizat ion (Gernsbacher , 2003; Mes ibov , 1998; Schopler , Short & Mesibov, 1989). Lovaas ' subjects were ass igned to either the treatment arm or one of two control groups b a s e d on the availabil ity of Lovaas-trained therapists. This was d o n e d u e to "parent protest a n d eth ica l considerat ions" (p. 4). Therefore, Lovaas d e v e l o p e d a "wait-list" control group, wh i ch was b a s e d on the first-come-first-served principle. If staff members were ava i l ab le when the chi ld was brought in, the chi ld went into the exper imenta l group; if not, the chi ld was assigned to a control g roup . However , this was not en fo r ced . Patients were not ass igned sequential ly; it was d e p e n d a n t u p o n the therapist determining whether they were ab l e to take another ch i ld . At that point, the families h a d a n opt ion not to par t i c ipa te . Lack of randomizat ion brought abou t conce rns regard ing the representativeness a n d comparab i l i t y of the sample . For examp le , arguments were m a d e that there were fewer higher funct ioning chi ldren in the control groups. However, Eikeseth (2001), a researcher from one of the Lovaas repl icat ion sites in Norway, d isputed this by c la iming that the exper imenta l g roup a n d control g roup #1 were similar in regard to 19 out of 20 important var iables before the t reatment was a p p l i e d . A n d , as brought u p in the Auton l ega l p roceed ings , a schoo l of thought exists that while quasi-random assignment cou ld m a k e the treatment adminis tered to the exper imenta l g roup look better than it actua l ly was, it c o u l d not m a k e a n ineffect ive t reatment look ef fect ive (Baer, 1993): Inclusion Criteria Prorated Mental Age Schopler , Short a n d Mesibov (1989) took issue with the use of the prora ted menta l a g e (PMA) as a select ion criterion in determining the intel lectual funct ioning of the chi ldren. (Lovaas utilized the Bayley, Binet, a n d Cattrell IQ tests on 9 0 % of his research subjects for pre-testing). "To adjust for variat ion in M A [mental age ] scores as a funct ion of the subject 's C A [chronologica l age ] at the t ime of test administration, P M A scores were c a l c u l a t e d for a C A at 30 months ( M A / C A x 30)" (Lovaas, 1987, p. 3). Schopler a r g u e d that a PMA typical ly dep ic ts chi ldren as lower funct ioning than the ratio IQ a n d by using this criterion some chi ldren, besides the profoundly re ta rded , were unnecessari ly e x c l u d e d from Lovaas ' study (15% to b e exact ) . 11 Schopler po in ted out that Lovaas ' repor ted m e a n P M A of 18.8 for his t reatment g roup translated into a ratio IQ of 63, wh i ch was significantly higher than any previously repor ted r a n d o m samp le of autistic chi ldren. Jo rdan , Jones, a n d Murray (1998) desc r ibed the use of the P M A as "psychometr ica l l y dub ious " (p. 111). However , this c l a im was o n c e a g a i n d isputed by Eikeseth (2001) by expla in ing that Lovaas used ratio scores at intake a n d dev ia t ion scores at follow-up. This was a conservat ive measure as it apparent l y c o m p a r e d the highest scores at intake with the lowest scores at follow-up. Chronological Age Lovaas i n c l uded chi ldren <40 months if mute a n d less than 46 months if they demons t ra ted the autistic trait of echo la l i a . Echola l ia is r ecogn ized as a character is t ic of chi ldren with a better prognosis (Schopler, et al . , 1989). Lovaas a c k n o w l e d g e d that the echo la l i a criterion was e m p l o y e d to select a sample that wou ld progress more quickly, but asserted that d a t a analysis s h o w e d no greater improvement for e cho l a l i c ch i ldren. Sex Distribution Due to the lack of randomizat ion , more girls e n d e d up be ing enrol led into one of the Lovaas control groups. This c l a im of bias was b a s e d on the fac t that females are known to have poorer prognoses than males. Eikeseth (2001) a g r e e d that the exper imenta l g roup might have h a d a better prognosis than the two control groups, however , Smith (1997) po in ted out that two of the three girls in the exper imenta l g roup actua l ly a c h i e v e d the status of best o u t c o m e , thus demonstrat ing that the gende r bias criticism was not relevant. However , unequa l sex distribution c a n o c c u r e ven with randomizat ion , if the study is not large e n o u g h . Delay in Assessment In McEach in ' s 1993 follow-up study, the exper imenta l g roup was 3-5 years older than the control g roup at final assessment. Therefore, gains may h a v e b e e n d u e to add i t iona l d e v e l o p m e n t a n d e d u c a t i o n . Also, the treatment group h a d b e e n out of therapy longer. However, Eikeseth (2001) a rgued this point by c la iming that there was no e v i d e n c e that maturat ion or non-behavioural e d u c a t i o n p roduces large gains in individuals with autism. He went on to suggest that relapse might o c c u r w h e n behav ioura l t reatment is w i thdrawn. The length of t ime out of treatment cou ld in fac t bias the study against the t reatment group. Measurements Use of Different Measures Before and After Treatment Researchers have also po in ted out that Lovaas e m p l o y e d different measures before a n d after t reatment (Howlin, 1997). However, Smith a n d Lovaas (1997) rebut ted this assertion with the a rgument that this was a c c e p t a b l e p rac t i ce since no single measure of intel lectual funct ioning c o u l d b e val id a n d rel iable across an entire range of ages . In fact , Lovaas (2000) r e sponded " f ew wou ld suggest that 3 year olds should rece ive the same assessment as 13 year olds" (p. 32). Adjustments for the basel ine var iable c a n b e m a d e utilizing analysis of c o v a r i a n c e . 12 Measures Not Reflective of Important Areas of Difficulties in Autism Another criticism, m a d e by Howlin (1997) a n d Jo rdan e t a l . (1998) was that Lovaas used gross o u t c o m e measures instead of key measures of spec i f ic autism features. Specif ical ly , Howlin no ted a pauc i t y of measures on socia l interact ion, friendships, c o n c e p t u a l abilities, socia l c o m m u n i c a t i o n , obsessional a n d ritualistic behaviours, a n d disturbances of m o o d . Eikeseth (2001) coun te r ed this c l a im by stating that the 1993 follow-up study by M c E a c h i n et a l . c o n t a i n e d 33 o u t c o m e measures, all of wh i ch ref lected important areas of difficulty in autism. Control Groups Jo rdan et a l . (1998) c o n t e n d e d that Lovaas ' control g roup #2 was a n opportunist sample . However , this was d isputed by Eikeseth (2001) by justifying that this g roup was used to rule out re fe r ra l and select ion bias. Mes ibov (1998) a rgued that different cut-off ages were used for e cho la l i c a n d mute chi ldren; that the control group h a d fewer higher funct ioning chi ldren than wou ld b e e x p e c t e d (typically 20-30% of autistic chi ldren are higher functioning); a n d that e a c h control g roup r ece i ved different testing protocols. However, Lovaas dismissed the c l a im of fewer high funct ioning chi ldren in the contro l g roup by a rgu ing that, e v e n with all his exper i ence , he c o u l d not predic t wh i ch chi ld wou ld s u c c e e d b e c a u s e high funct ioning adults may not have a p p e a r e d as high funct ioning chi ldren (Johnson, 1994). Treatment Delay in Treatment It was no ted by Jo rdan et a l . (1998) that control g roup #1 was d e l a y e d in rece iv ing their t reatment mak ing the exper imenta l group 6 months older than the control g roup . (This was the only var iab le out of 20 assessed that was not consistent b e t w e e n the groups). Eikeseth (2001) a g r e e d with these findings, but a rgued that Lovaas d id not find a relationship b e t w e e n a g e at intake a n d treatment o u t c o m e w h e n the best o u t c o m e chi ldren were c o m p a r e d to those chi ldren in the study w h o d id not a c h i e v e normal intel lectual a n d a c a d e m i c funct ioning. Due to the young sample (M=35 months), a g e at intake may not have b e e n a n influential var iab le . However, it does not n e g a t e the fac t that it may still b e an important var iab le for future studies. Timing of Initiation of Treatment While intuitively it makes sense that the earlier t reatment is initiated the better the prognosis for a n autistic ch i ld , Howlin (in 2003) po in ted out that in fac t there was no research that substant ia ted this assumption. Lovaas a d v o c a t e d t reatment be fore the a g e of four, while the Autism Society of A m e r i c a (1998) states, "A l though not conf i rmed by research , it is believed [italics a d d e d ] that the best a g e to beg in intensive ABA therapy is b e t w e e n 24-42 months or before 3 V2 year of a g e " (p. 4). While research has demonst ra ted improvements in chi ldren w h e n intervention programs are init iated before the a g e of four, they h a d fa i led to inc lude a systematic compar i son with chi ldren of other ages . Conversely, there was no e v i d e n c e to support the hypothesis that d e l a y e d initiation of treatment wou ld b e of little benefit . Instead, the ' 1 3 belief was "bet ter late than never" as o p p o s e d to "early intervention or no th ing " (p. 256). S ince this t ime, add i t iona l research has b e e n c o n d u c t e d wh i ch reported gains for 4-7 year olds after one year of Lovaas treatment (Eikeseth et al . , 2002). Treatment Fidelity/Integrity Ensuring that a treatment is imp l emen ted as it was originally c o n c e p t u a l i z e d is a n issue that is not only re levant to Lovaas therapy but has p l a g u e d researchers in the fields of psychiatry, psycho logy a n d e d u c a t i o n for years. Measurements of t reatment integrity were not c o m m o n w h e n Lovaas c o n d u c t e d his study. However, c laims against the integrity of Lovaas ' p rogram (Jordan, et al. , 1998) were disputed by Eikeseth (2001) w h o a r g u e d that, a l though formal t reatment integrity was not moni tored , speci f ic measures were taken . These i nc luded the establ ishment of t reatment protocols a n d manuals , weekly cl inic meet ings (with t e a c h i n g demonstrat ions), monthly g r and rounds, a n d use of individual logbooks, d a t a sheets a n d written reports. Treatment Intensity Jo rdan et a l . (1998) suggested that the ou tcomes of Lovaas ' study might b e d u e to the intensity of the p rogram as o p p o s e d to the ac tua l t reatment p rov ided . However , this was later d isputed by Eikeseth, Smith, Jahr a n d Elde.vik (2002) w h o c o m p a r e d the effects of 28 hours per week of behav ioura l t rea tment to 29 hours per week of e c l e c t i c spec ia l e d u c a t i o n t reatment for 4-7 year olds. As previously discussed, the behavioural ly t reated chi ldren a p p e a r e d to m a k e more progress on measures of intel lectual , l a n g u a g e a n d adap t i v e funct ioning after one year of treatment. Lovaas a d v o c a t e d the impor tance of 40 hours per week of therapy b a s e d on the fac t that "there are no d a t a to support that a n intervention of less than 40 hours per week will result in 4 7 % rate of normal funct ion ing" (Lovaas, 2000, p. 30). However, at a c o n f e r e n c e of the Autism Society of C o n n e c t i c u t in 1998, Dr. John M c E a c h i n recogn ized the barriers f a c e d by many parents trying to access LAT. His position was that there was no "all or no th ing " abou t the program a n d if parents c o u l d obta in 25 hours per week that wou ld b e sufficient (Hultgren, 1998). Use of Aversion Therapy O n e of the most content ious issues surrounding Lovaas ' earlier work was the use of aversives. S ince these methods are no longer used, o n e a rgument is that it inval idates the results of Lovaas ' 1987 study (Jordan et al. , 1998). Eikeseth (2001) d i sagreed a n d c o n c l u d e d that while it wou ld m a k e it very difficult to rep l i ca te a n d c o m p a r e studies, the results remain va l id . Outcomes Claims of Recovery or Cure M u c h controversy has also c en te r ed on Lovaas ' c l a im of ach i ev ing "normal intel lectual a n d educa t i ona l funct ion ing" in 4 7 % of the exper imenta l group (Lovaas, 1987, p. 7). A l though Lovaas was cogn izan t of the d e b a t e a b o u t whether or not normal functioning e q u a t e d to 14 recovery from autism, he still refers to his best o u t c o m e subjects as recovered. Somewhere a long the line, however , recovered has b e e n translated into cured. Lovaas has publ ic ly re jec ted the cla ims of a " c u r e " by stating "I don ' t c l a im a cure b e c a u s e w e haven ' t got ten to the organ ic var iab le that is caus ing aut ism" (Johnson, 1994, p. 9). Schopler , Short a n d Mes ibov (1989) jo ined this a rgument w h e n they po in ted out that the Lovaas study fa i led to use typica l soc ia l , behav ioura l a n d c o m m u n i c a t i o n o u t c o m e measures to support the hypothesis of atta ining normal functioning. They c o n t e n d e d that w h e n IQ measures a n d progress in schoo l are used a lone , they are poor indicators of t reatment e f f i cacy . Schoo l p l a cemen t s m a y have more to d o with administrative policies surrounding specia l-needs chi ldren. In add i t ion , improvements in IQ scores m a y b e d u e to improved c o m p l i a n c e rather than in cogni t i ve funct ioning. They took issue with Lovaas ' statement that " the r e cove red chi ldren show no p e r m a n e n t . . . behav ioura l deficits a n d their l a n g u a g e appea r s to b e normal " (Lovaas, 1987, p. 8) as the c l a im was unsupported by d a t a . Mes ibov (1998) also po in ted out that there are m a n y high-functioning autistic p e o p l e with near-normal IQ's a t t end ing regular publ ic schools w h o wou ld still b e cons ide red severely h a n d i c a p p e d . Settings Lovaas autism treatment, as originally c o n c e p t u a l i z e d , was de l i vered in a structured c l in ical setting with dai ly therapist supervision. Since then, inc reased d e m a n d for LAT has fo r ced the Lovaas Institute to design workshop-based services that c a n b e transferred into commun i t y a n d h o m e settings. Lovaas wa rned , however , that these workshop-based t reatment programs h a d very different ou t comes , a n d est imated that d a t a d o c u m e n t i n g normal funct ioning in these settings wou ld only b e abou t 20 percent (Lovaas, 2000). Lovaas suggested this was d u e to factors such as high staff turnover, less frequent supervision, lower t reatment intensity, a n d utilizing therapists w h o h a d less a c a d e m i c preparat ion in learning-based theory a n d research. An e ven greater c o n c e r n was that/should these workshop-based services b e prov ided by therapists not tra ined by UCLA, treatment e f f i c acy wou ld d rop to less than 10%. Therapists Parents seeking behav ioura l therapists to administer Lovaas therapy are vulnerable to the various c la ims of expertise m a d e by these practitioners. Lovaas was a d a m a n t that only UCLA-trained therapists were qual i f ied to deliver Lovaas autism treatment (Lovaas, 2000). Lovaas therapists h a d to h a v e a master's d e g r e e be fore they were cons ide red for a 9-month internship that consisted of 30 hours per week of training. However, critics felt that others c o u l d b e trained in the techniques of discrete trial training (DTT) (Donnelly, 1996). Of interest is that Lovaas ' perspect ive on behav ioura l training appea r s to h a v e c h a n g e d over the years. In an interview publ ished inThe A d v o c a t e , Lovaas espoused the virtues of be ing ab l e to hire a n d train volunteers stating that, " m a n y p e o p l e c a n d o this without years of expensive training" (Johnson, 1994, p. 20). In fact , he p romoted the two-day UCLA workshop (at 15 a cost of $1,400) followed by a second workshop two months later (for $600 or $700). Yet, in 2000 while refuting some of the misunderstandings surrounding the UCLA Young Autism Project, Lovaas stated, "reading the teaching manual, attending a workshop led by UCLA certified consultants, practicing behavioural therapy on several families or spending a short time at the UCLA affiliated site, does, not make a person qualified to provide UCLA based treatment" (Lovaas, 2000, p. 30). Costs With the lack of empirical guidance on the most effective intensity of treatment, LAT expenses can fluctuate dramatically. Lovaas estimated the cost of treatment at $60,000 USD per year, or $120,000 for the recommended two years of treatment to achieve "normal functioning", at which point the best-outcome children would not require any additional treatment. He contrasted this against an estimate of more than two million dollars for life-long protective care (Lovaas, 2000). Appropriateness For All Autistic Children Lovaas determined that his therapy, which is language-based, does not work on children who are visual learners. In fact, his research focus has shifted towards this group of children in order to understand how they learn. "Skill in auditory matching" was the only predictor of success at the end of three months (e.g., whether the children can imitate sounds like words) (Johnson, 1994, p. 19). Those that appeared to benefit the most from LAT were nonverbal and non-compliant (Donnelly, 1996, p. 6). 16 British Columbia Autism Intervention Programs Historically, the government of British C o l u m b i a used to offer a myriad of services for. autistic chi ldren such as respite, h o m e support services, ch i ldcare , physical a n d o c c u p a t i o n a l therapy, s p e e c h / l a n g u a g e therapy, hear ing services, a n d behav ioura l support. However , the focus of these programs was on supportive c a r e as o p p o s e d to therapeut ic services. Therefore, in 1995, families of autistic chi ldren b e g a n a letter-writing c a m p a i g n in a n a t tempt to lobby the government for Lovaas Autism Treatment. By 1996 the parents h a d c o l l e c t e d the names of sixty-three psychiatrists on a petition in support of Lovaas a n d by 1998 over e ight thousand citizens h a d s igned a petit ion urging the Province to fund effect ive autism treatment . However , in response, the government p l a c e d a morator ium on any further autism therapies; this a c t i on prec ip i ta ted the Auton lega l p roceed ings (to b e desc r ibed in a later section). In M a y 1999, one year after the Auton c a s e h a d c o m m e n c e d , the BC Ministry of Chi ldren a n d Families a n d the Ministry of Educat ion l a u n c h e d its Autism Action Plan a n d the Autism Action Implementation Plan to address concerns expressed by the publ ic in regard to autism services for chi ldren, youth a n d their families. This p lan was c o n c e p t u a l i z e d after consultat ion with parents, service providers, communi ty a d v o c a t e s a n d government staff. The p lan was gener i c in that it p rov ided "early intervention a n d treatment" ; it d id not speci f ica l ly outline a program of early intensive behav ioura l intervention. However, imp lementa t ion was h a m p e r e d by both funding constraints a n d the equity v iewpoint that autistic programs h a d to b e b a l a n c e d against services to other spec ia l needs chi ldren. In M a r c h of 2001, in response to the Auton BC Supreme Court ruling (July 2000), a n inter-Ministerial c ommi t t e e des igned the Provincial Centre for Autism and Related Disorders p rogram (P-CARD). The Early Intensive Behavioural Intervention (EIBI) initiative for chi ldren under the a g e of 6 started in M a y 2001 a n d enta i led cont rac t ing three chi ld d e v e l o p m e n t centres l o c a t e d in Ladner, the Thompson O k a n o g a n region, a n d Victor ia, to prov ide the first official EIBI p rogram in the Province. This p rogram offered EIBI for 20 hours per week on a 1:1 basis in add i t ion to speech- l anguage a n d o c c u p a t i o n a l therapy. Each program h a d opportunit ies for the chi ldren to integrate with typica l peers, involved parents in the training a n d intervention, a n d utilized only positive behav ioura l re inforcement to hand le behav ioura l problems (Mirenda, 2005). In total , 75 chi ldren were enrol led in these centres. A c o m p o n e n t of the program was to inc lude training of add i t iona l EIBI therapists in addi t ion to an eva luat ion project to measure the i m p a c t of the therapy on the chi ldren, their families a n d the communi ty . In M a y 2002, instead of e x p a n d i n g the EIBI program, the Ministry of Ch i ldren a n d Family Deve lopmen t (MCFD) initiated a n interim early intensive intervention (IEII) funding opt ion for autism treatment (the program n a m e has subsequently c h a n g e d to: Autism Funding: Under Age 6). Under this p rogram, families of chi ldren with ASD under the a g e of six are el igible to rece ive a max imum of $20,000 per year for autism interventions, training, a n d equ ipment . However, this is a 17 parent-administered program that requires families to assume a range of c l in ical a n d administrative responsibilities for their chi ldren's behav ioura l t reatment programs. Therefore, they must enter into formal funding agreements with the government . C l in ica l responsibilities inc lude select ion of a behav ioura l therapist, a n d in consultat ion, d e v e l o p m e n t of a behav ioura l p lan of intervention. In regard to the administrative responsibilities, there are two options for the I Ell p rogram. The first is the invoice payment option whe reby the family appl ies to the M C F D for a billing number so that treatments c a n b e invo i ced directly to the government . The s e c o n d opt ion is the direct funding option where the family sets up a trust a c c o u n t for the chi ld a n d the M C F D deposits funds into the a c c o u n t on a monthly basis. Both these options require de ta i l ed record keep ing a n d a c c o u n t i n g procedures to verify services r e ce i ved . In add i t ion , the behav ioura l therapists b e c o m e emp loyees of the families, thus necessi tat ing full payroll responsibilities such as the requisite submission of emp loyment insurance premiums, taxes, a n d Work Safe BC contributions. A l though administrating the IEII p rogram c a n b e cha l l eng ing for some families, the benefit is that the parents have the opt ion of choos ing a n d customizing a p rogram that is spec i f ic to their chi ld 's needs. However, to b e eligible for funding, the therapists must b e se l ec ted from a list of provincial ly a p p r o v e d service providers. This list is ma in ta ined by a third party provider - Autism Communi t y Training (ACT) - wh i ch also provides information, training a n d support services to these families. The autism funding p rogram was further e x p a n d e d in April 2003 to inc lude schoo l-aged chi ldren (initially c a l l ed the Extended Autism Intervention Funding (EAI) p rogram a n d then c h a n g e d to Autism Funding: Ages 6-18). Four Ministries are responsible for co l l abora t ing on this p rog ram: (a) Health Services/Health Planning (for diagnosis a n d assessment); (b) Chi ldren a n d Family Deve lopmen t (intervention funding); (c) Educat ion (educat ion programs); a n d , (d) the overarch ing Provincial Health Services Authority. For chi ldren 6-18 years, diagnosis does not have to b e establ ished by a multi-disciplinary t e a m (unlike the under 6 year olds). A g a i n , the family is free to c h o o s e the type of behav ioura l t reatment as long as it is administered by an a p p r o v e d service provider a n d is b a s e d on the chi ld 's d o c u m e n t e d intervention p lan . However, for these schoo l a g e chi ldren, the funding is only p rov ided for intervention services de l ivered outside of regular schoo l hours (e.g., the behav ioura l therapist w o u l d not b e a l l owed into the classroom). Currently, the max imum funding ava i l ab le for chi ldren 6-18 year o ld is $6,000 per year. However , recently the BC government a n n o u n c e d that it wou ld b e increas ing its funding to the schoo l districts to $16,000 per year for e a c h student d i a g n o s e d with autism spect rum disorder (as o p p o s e d to the current funding a g r e e m e n t wh i ch only i n c l uded chi ldren d i agnosed with autism) (Lee, 2006). The gove rnment also establ ished the BC Autism Assessment Network (BCAAN), under the jurisdiction of the Provincial Health Services Authority (PHSA), wh i ch is responsible for the 18 assessment a n d diagnosis of all chi ldren with suspec ted autism. The Ministry of Health Planning's Standards and Guidelines for the Assessment and Diagnosis of Young Children with Autism Spectrum Disorder in British Columbia (BCMOHP, M a r c h 2003) provides the criteria for wh i ch these chi ldren are assessed. In M a y 2003, the government o p e n e d the Provincial Autism Resource Centre (PARC) in Sunny Hill Health Cent re for Chi ldren in V a n c o u v e r fo l lowed by centres at the Q u e e n A lexandra Cent re for Chi ldren's Health in Victor ia a n d at Chi ldren a n d Women ' s Hospitals in Vancouve r . These multidisciplinary teams of nurses, psychologists, socia l workers, pediatr ic ians, psychiatrists, o c c u p a t i o n a l therapists, a n d speech- l anguage pathologists are responsible for the assessment a n d diagnosis of autism. In order to rece i ve funding, parents must either use the r e c o m m e n d e d clinicians from the PARC, or if they c h o o s e a private consultant, the c l in ic ian must use the same guidelines that PARC follows. In August 2003, the government ' s c l i n i c-managed EIBI p rogram a n d its pa r en t-managed IEII p rogram were formally e va lua ted (Mirenda, 2005). In total, 39 chi ldren from the EIBI p rogram a n d 31 from the IEII p rogram were e va lua t ed for up to two years for the IEII p rogram a n d up to 32 months for the EIBI p rogram. Assessment tools were administered at base l ine , 6 months, 1 year, 2 years, a n d 32 months. The study e va l ua t ed the fol lowing: 1. Treatment hours a n d its relationship to progress m a d e by the ch i ld . 2. Impact of types of service p rov ided , diagnosis, materna l e d u c a t i o n , ch rono log i ca l a g e , a n d chi ld "testability". 3. Rate of c h a n g e prior to a n d over 2 years of intervention. 4. Parental percept ions a n d satisfaction. Study results i nd i c a t ed that there was no significant relationship b e t w e e n a chi ld 's progress a n d the hours of treatment r e ce i ved ; however , some predictors of progress we re ident i f ied. There also was no significant d i f ference in ou t comes b e t w e e n the EIBI a n d the IEII services nor d id materna l e d u c a t i o n or the child's a g e at treatment initiation predic t the chi ld 's progress. However, chi ldren w h o were " tes tab le " before initiation of treatment, 6 months later on , or both h a d m a d e more measurab le gains than chi ldren w h o were un-testable. For examp le , the testable chi ldren demonst ra ted a n increase in IQ from 60 to 83.7 (untestable chi ldren h a d a small significant increase from 45.8 to 49.7). In addi t ion , both groups h a d a significant reduct ion in autism behaviours. Overal l , both programs assisted the chi ldren to m a k e more progress per month than they m a d e before they started treatment. A n d , in regard to parenta l percept ions a n d satisfact ion, results showed that parenta l stress was greatly r e d u c e d in both types of programs a n d that they were very satisfied overal l . In order to grasp the multitude of complexit ies that i m p a c t the des ign a n d implementa t ion of government programs such as the autism project, it is necessary to step b a c k a n d examine the underlying pol icy a n d legislative frameworks that i m p a c t pol icy deve lopmen t . Therefore, a brief summary of these main c o n c e p t s follows. 19 Policy Framework The field of policy science e vo l ved after World War II w h e n pol i t ical s c i ence students b e c a m e interested in investigating the relationships b e t w e e n governments a n d citizens. During this time, students sought frameworks that i nco rpora ted the c o n c e p t s of justice, equity, a n d the pursuit of soc ia l , e c o n o m i c , a n d polit ical d e v e l o p m e n t into their existing constructs of normat ive or mora l dimensions of government . Research of this e ra focused on the purposes of gove rnment a n d the activities required of governments to ascerta in a high quality of life for their peop l e . However , it was general ly r ecogn ized that a g a p in k n o w l e d g e existed b e t w e e n polit ical theories a n d a c tua l pol it ical pract ices . This vo id was filled with the d e v e l o p m e n t of new a p p r o a c h e s to study polit ical p h e n o m e n a ; h e n c e the evolution of the discipl ine of policy science (Howlett & Ramesh, 1995). Harold Lasswell was one of the original pioneers of an a p p r o a c h that went b e y o n d focusing on the structure of governments , the behav iour of pol it ical actors, or wha t governments ought to d o . Instead his perspect ive was d i r ec ted towards wha t governments actua l ly d o - the pub l i c pol ic ies a n d po l i cymak ing processes. Lasswell asserted that po l i cy s c i ence dif fered from previous methodo log ies in that it was multi-disciplinary, p rob lem solving, a n d normat ive in its a p p r o a c h . Multi-disciplinary refers to the fac t that pol icy s c i ence goes b e y o n d its historically narrow focus on gove rnment institutions to inc lude perspect ives from fields such as soc io logy, e c o n o m i c s , law, a n d politics. P rob lem solving means that research should b e re levant with a v iew to solving a c tua l socia l problems as o p p o s e d to contr ibuting to more a c a d e m i c deba tes . A n d , the criterion of normative assumes that, "po l i cy s c i ence should not b e c l o a k e d in the guise of 'scientific objectivity ' , but should recogn ize the impossibility of separat ing goals a n d means, or values a n d techniques , in the study of gove rnment ac t ions" (Howlett & Ramesh , 1995, p. 3). A l though these three character ist ics of pol icy s c i ence have b e e n revised somewha t over the years, essentially Lasswell's early work remains the framework of mode rn d a y studies. Public policy has many c o m p e t i n g definitions, but is essentially " a course of a c t i on or inact ion chosen by publ ic authorities to address a g iven p rob lem or interrelated set of prob lems" (Pal, 1997, p. 1). Its first feature - a course of a c t i on - implies that policy-makers d e v e l o p a n d work within a pol icy framework by app ly ing a pol icy to a speci f ic context . Also of s igni f icance to this definition is the assertion that q government 's inaction on a n issue c a n also b e de f ined as pol icy in of itself. However , for inact ion to b e v i e w e d as a publ ic pol icy, a government wou ld have h a d to consciously c o m e to the dec is ion not to ac t , or to mainta in the status quo. The definition's s e c o n d feature - referring to problems a n d interrelated sets of problems - recognizes that publ ic po l i cymak ing is a m e t h o d of dea l ing with problems; therefore policies are not v i ewed as ends in themselves but as instruments or tools to address publ ic conce rns . "Policies are menta l constructs, strings of phrases a n d ideas . The text of a pol icy s ta tement a n d the programs a n d act ions that follow it are simply e v i d e n c e for the menta l construct " (Pal, 1997, 20 p. 12). Public pol icy goes b e y o n d the everyday p rac t i ce of dec is ion-making; instead pol icy results from decis ion-making c o n d u c t e d within a framework or pattern (Pal, 1997). Sometimes this occurs quite insidiously with governments c o m i n g to the real ization that decisions m a d e over t ime may in f ac t n o w constitute a pol icy. While individuals, private organizations a n d governments alike all h ave policies, a pol icy is d e e m e d a public policy b a s e d on the source of the pol icy, not on the target of the pol icy, a n d c a n originate from legislature or another d e l e g a t e d authority. Theorists such as William Jenkins a n d Thomas Dye d e b a t e d whether public policy was a process or a c h o i c e . Jenkins's conceptua l i za t ion of publ ic pol icy was , " a set of interrelated decisions taken by a pol i t ical ac to r or group of actors conce rn ing the select ion of goals a n d the means of a ch i e v ing them within a spec i f ied situation where those decis ions should, in principle, b e within the power of those actors to a c h i e v e " (Howlett & Ramesh, 1995, p. 5). This definition a c k n o w l e d g e s that rarely does a gove rnment solve a p rob lem with o n e single dec is ion . In reality, a number of decisions wou ld have b e e n m a d e a long the w a y to compr ise a publ ic pol icy. It also recognizes a government 's c a p a c i t y to implement a pol icy by c o n c e d i n g that pol icy cho i ces might b e limited by factors such as f inancia l constraints or lack of resources. A n d finally, Jenkins' conceptua l i za t ion supports the goal-oriented behav iour of pol icy programs. Here, governments def ine a g o a l for a p rob l em a n d d e v e l o p the mechan isms in wh i ch to a c h i e v e it. Alternatively, Thomas Dye de f ined publ ic pol icy as be ing "anyth ing a government chooses to d o or not to d o " (Howlett & Ramesh, 1995, p. 4). While somewha t v a g u e , it d id stipulate that the agen t of the pol icy must b e a government (therefore possessing the legit imate authority to impose policy) a n d that to b e cons ide red publ ic pol icy it must consist of a de l ibera te dec is ion . A n d finally, J ames Anderson 's definition desc r ibed pol icy as " a purposive course of a c t i on fo l lowed by a n ac tor or a set of actors in dea l ing with a p rob lem or matter of c o n c e r n " (p. 6). This definition in t roduced the possibility of multiple sets of pol icy actors as policies were often formulated not only by multiple decisions, but also by multiple sets of decision-makers. It also r ecogn ized the link b e t w e e n government ac t i on a n d the pe r cep t ion of the existence of a prob lem or issue requiring attent ion. There are five categor ies of policy actors that par t i c ipate in the pol icy process: (a) e l e c t e d officials, (b) a p p o i n t e d officials, (c) interest groups, (d) research organizations, a n d (e) the mass m e d i a . Elected officials inc lude members of the government ' s executive (also known as the Cab inet ) a n d the legislature. C a b i n e t ministers p lay a significant role in pol icy format ion as they are bes towed with the constitutional authority to govern the country/Province. This authority is suppor ted by their legit imate control over information dissemination a n d f inancia l resources, in add i t ion to their a c cess to the m e d i a a n d bureaucra t i c expertise. The legislature, 21 on the other h a n d , is a forum in wh i ch publ ic problems are identif ied a n d policies d e b a t e d . It is a process that holds a government a c c o u n t a b l e to its p e o p l e , as o p p o s e d to a venue for deve lop ing pol ic ies. While legislators have the opportunity to question bills a n d pol ic ies, in majority governments the authority of the execut ive overshadows legislators' input into the a c tua l po l i cymak ing process. Hence , legislators are not significant actors in the pol icy process (Howlett & Ramesh, 1995). Appointed officials are me thodo log i c a l a n d substantive experts (known as "bureaucrats" ) who , in theory, are hired by the execut ive to assist with government ' s publ ic pol icy a n d administrative duties (not so m u c h in a c tua l pract i ce ) . This is a he te rogeneous , non -partisan co l l ec t ion of multiple disciplines, e a c h with their own unique perspect ive , values a n d ideals. The present d a y reality of comp l ex , multi-sectoral governments has impar ted greater responsibilities on the role of the bureaucra t . These a p p o i n t e d officials n o w assume the po l i c ymak ing a n d implementat ion functions o n c e reserved for the execut i ve a n d legislature. S ince po l i cymak ing decisions o c c u r in isolation within these bureaucra t i c commit tees , other pol icy actors are not privy to the process. However, while bu r eauc r a c y m a y a p p e a r to reign supreme, it still remains the ult imate responsibility of the execut ive to a p p r o v e all policies imp l emen ted . Typically, the more content ious the pol icy, or the more divisive the bu reauc r a c y on the pol icy 's merit, the more execut ive input is proffered (Howlett & Ramesh, 1995). Interest groups are outside of the preserve of the government a n d are typical ly socia l organizations that represent groups such as labour forces (e.g., union workers, businesses, professional associations), individuals with spec ia l needs (e.g., Families for Effective Autism Treatment), a n d consumer activists (e.g., BC Taxpayers' Associat ion). Interests groups c a n play key roles in the pol icy process, as they are often the purveyors of the information that is required by polit icians a n d bureaucrats for po l i cymak ing . Knowing this, interest.groups m a y e l ec t to only release information to the polit ical party that is most sympathet ic towards their cause . Alternatively, polit icians a n d opposit ion parties may seek out interest groups that support their pol icy platform. C o u p l e d with the promise of information, interest groups also wie ld power in the form of f inancia l incentives (e.g., c a m p a i g n contributions) a n d votes. However , how m u c h power a n interest group has will d e p e n d upon its organizat ional structure, for e x a m p l e , its size, f inancia l a n d h u m a n resources, a n d coalit ions with other organizations (Howlett & Ramesh, 1995). Research organizations are also i ndependen t of government a n d inc lude university researchers a n d i n d e p e n d e n t "think tanks". University research tends to b e non-partisan a n d spec ia l ized , focusing on e x p a n d i n g theoret ica l frameworks of publ ic pol icy, while private research organizations h a v e a more comprehens i ve , a p p l i e d a p p r o a c h , p roduc ing r e commenda t i ons for policy-makers. Think tanks may b e a l i gned with the ideologies of certa in 22 polit ical parties a n d will seek them out w h e n disseminating their findings (Howlett & Ramesh, 1995). A n d finally, the role of the mass media in the pol icy process remains somewha t controversial with some p e o p l e feel ing they play a marginal role to others stressing their impor tance . However , wha t is a g r e e d upon is that the m e d i a links governments to society. Investigative reporters have the m a n d a t e of d iscover ing problems, discerning their causes a n d suggest ing remedies . Thus, they often bring problems to the forefront of the pol i t ical s cene . "The role of the m e d i a in the pol icy process lies in the fac t that in reporting problems they c o m b i n e the roles of passive reporter with ac t i ve analyst as well as a n a d v o c a t e of a solution" (Howlett & Ramesh, 1995, p. 59). Interest groups often take a d v a n t a g e of the med ia ' s power by ga in ing government ' s at tent ion through sensationalist news reporting. Alternatively, pub l i c officials will also seek out the m e d i a to release select information that may promote their pol i t ical platform. Either way , pol icy actors are wise to the inf luence of m e d i a on popu la r opin ion a n d will opt to use it w h e n strategical ly war ranted . 23 Pertinent Legislative Frameworks The overarch ing m a n d a t e , under wh i ch governments must adhe re to, is the Canada Health Act. The Hospital Insurance and Diagnostic Services Act of 1957 a n d the Medical Care Act of 1966 fo rmed the framework for the Canada Health Act as w e know it today . In the late 1950s the federa l gove rnment offered to c o m p e n s a t e the Provinces 5 0 % of their cost for insured heal th services if they a g r e e d to insure all hospital a n d physic ian services a n d c o m p l y with some s tandard principles. A l though these Acts d id not prevent the Provinces from extra-billing patients for services, it was not to the Provinces' a d v a n t a g e to d o so as their federa l contr ibut ion wou ld b e r e d u c e d proport ionately . In 1977 a m o d e l of b lock funding that was no longer l inked to direct provincia l expenditures for health c a r e r e p l a c e d this formula for shared costs. In response, many Provinces imp l emen ted direct pat ient charges , wh i ch the federa l gove rnment v i ewed as a threat to the free a n d universal phi losophy of health ca re . The Canada Health Act was e n a c t e d on April 1st 1984 thereby merg ing the two previous Federal Acts. In add i t ion , it a d d e d add i t iona l restrictions a n d penalt ies aga inst extra-billing of patients. The primary object ive of the Canada Health Act is " to protect , p romote a n d restore the physica l a n d menta l well-being of residents of C a n a d a a n d to faci l i tate reasonab le a ccess to health services without f inancia l or other barriers" (Baker a n d Bhabha , 2004, p. 25). The Provinces are required to cove r "insured health services", de f ined as all "med i ca l l y necessary" hospital services, "med i ca l l y requ i red" phys ic ian services, a n d surgical-dental p rocedures per formed within a hospital. In order to b e el igible for the cash transfers a n d tax point transfers the Provinces h a d to c o m p l y with the five principles of the Act: (a) universality, (b) comprehens iveness , (c) portability, (d) publ ic administration, a n d (e) accessibi l i ty (Flood & Choudhry , 2002; M a d o r e , 2004). Of speci f ic r e l evance to the Auton c a s e were the principles of universality a n d comprehensiveness. Universality is a fundamenta l C a n a d i a n va lue that ensures that qualified provincia l residents rece ive insured health c a r e services p rov ided on uniform terms a n d condit ions. Qualified, as de f ined under the Medicare Protection Act, des ignates that a benef ic iary of provincia l health c a r e services must b e a C a n a d i a n Citizen (or have pe rmanen t res idency status), a n d must reside within BC for a minimum of 6 months per c a l e n d a r year. Comprehensiveness refers to the provision of all insured health services p rov ided by hospitals, m e d i c a l practitioners (or dentists within hospitals); or, if the provincia l l aw permits, add i t iona l services de l ivered by other listed health c a r e practitioners. A cruc ia l issue that unfo lded in Auton per ta ined to the fac t that the C H A does not specif ical ly def ine "med ica l l y necessary" or "med ica l l y requi red" ; nor are these terms operat ional ly de f i ned in provincial legislation. Typically, wh i ch services are i n c l uded in publ ic ly f unded plans are nego t i a t ed b e t w e e n the government a n d m e d i c a l associat ions (this is also the c a s e w h e n services get de-listed). Critics a rgue that this process is f l awed in that the process for determining med ica l l y necessary services is too intimately l inked to determining c o m p e n s a t i o n 24 rates for physicians. A n examp l e of the limitations of comprehensiveness is in regard to prescription drugs. A l though med i ca t i on is med ica l l y necessary, prescription drugs in C a n a d a are only insured at a rate of 3 6 % c o m p a r e d to 91 .1% for hospital services a n d 9 9 % for physic ian services. This p laces C a n a d a a longs ide the United States, Mex i co , a n d Turkey as far as providing a publ ic ly f unded universal drug program (Flood & Choundry , 2002). Also of r e l evance to Auton is the C H A ' s definition of "hea l th c a r e pract i t ioner" as, " a person lawfully entit led under the l aw of a Province to prov ide health services in the p l a c e in wh i ch the services are prov ided by that pe rson" (Supreme Court of C a n a d a , 2004, A p p e n d i x A). This was at issue w h e n it was de te rmined that Lovaas behav ioura l therapists were not classified as health c a r e practitioners. The next legislative framework - the BC Medicare Protection Act (MPA) - guides the administration of health c a r e services. The purpose of the Act is, " to preserve a publ ic ly m a n a g e d a n d fiscally sustainable health c a r e system for British C o l u m b i a in wh i ch a c ces s to necessary m e d i c a l c a r e is b a s e d on n e e d a n d not a n individual's ability to p a y " (Government of BC, 2004, p. 1). It enshrines the principles of the Canada Health Act while attesting to the conv i c t ion that, " the p e o p l e a n d government of BC bel ieve that m e d i c a r e is one of the def ining features of C a n a d i a n n a t i o n h o o d " (BC Health Coa l i t ion , 2005). It a c k n o w l e d g e s , " the judicious use of m e d i c a l services in order to mainta in a fiscally sustainable heal th c a r e system for future generat ions" (Government of BC, 2004, p. 1) a n d prohibits patients from b e i n g c h a r g e d for a benefit or for "materials, consultations, procedures , use of a n off ice, c l inic or other p l a c e or for any other matters that relate to the rendering of a benef i t " (Health C a n a d a , 2006, p.. 11). The /Vied/care Protection Act establishes a n d regulates the 8rit/'sh Columbia Medical Services Plan. Qual i f ied residents are entit led to rece ive health c a r e benefits with the paymen t for such services m a d e to the service provider b y the government . The M P A differentiates b e t w e e n services p rov ided by a " m e d i c a l pract i t ioner" a n d those p rov ided by a "hea l th c a r e pract i t ioner" a longs ide the distinction m a d e in the Canada Health Act b e t w e e n co re a n d non-core services. " M e d i c a l practit ioners" deliver co re services a n d inc lude phys ic ian services a long with services p rov ided within hospitals; these benefits must b e fully f u n d e d a n d ava i l ab le to all residents. C o r e services c a n also b e supp l emen ted by partially f u n d e d , non-core services at the discretion of the Province. These inc lude services prov ided by "hea l th c a r e pract it ioners" such as chiropractors, dentists, optometrists, podiatrists, a n d by regulat ion, physica l therapists, massage therapists, a n d nurses. Therefore, a service c a n n o t b e a benefit under the Medical Services Plan unless it is p rov ided by a m e d i c a l practit ioner or by a health c a r e pract i t ioner listed in sect ion 13 of the Medicare Protection Act or in a regulat ion (Health C a n a d a , 2006; Supreme Court of C a n a d a , 2004a). The Medical Services Commission is the regulatory a g e n c y c h a r g e d under the Medicare Protection Act with implement ing the Medical Services Plan. The Commiss ion has the authority 25 to determine whether a service is cons ide red a co re benefit a n d whether a speci f ic individual is d e e m e d either a m e d i c a l pract it ioner or a health c a r e practit ioner. Sect ion 15(2) of the Medicare Protection Act also grants the commission power to c a n c e l a pract i t ioner 's enrolment. The Commiss ion does not h a v e the authority to prov ide funding for a serv ice not p rov ided by a m e d i c a l pract i t ioner (e.g., non-core services). In these situations, a spec ia l request for funding is m a d e to a Supplementary Practit ioner Spec ia l Commi t t e e , spec i f ic to the type of heal th c a r e pract i t ioner administering the service. However, neither the Commiss ion nor the Supplementary Practit ioner Spec ia l C o m m i t t e e has the authority to fund non-core benefits if the p roposed provider is not a l r eady listed in s. 13 of the Medicare Protection Act. Judicious funding of non-core benefits is in c o m p l i a n c e with the principles of the Canada Health Act. Provinces are free to limit or restrict benefits in terms of cost a n d number of treatments (e.g., restrictions on the number of insured eye exams) ; no service by a "hea l th c a r e pract i t ioner" is fully insured. In add i t ion , a l though the Medicare Protection Act requires that a service b e "med i ca l l y requ i red" , as previously men t i oned , this term is not d e f i n e d . The inspiration for this thesis or ig inated from the Auton lega l p roceed ings that p l a yed out in both the trial a n d appe l l a t e Courts of British C o l u m b i a a n d the Supreme Court of C a n a d a . The fol lowing sections will provide the b a c k g r o u n d of these trials. 26 Legal Context Aufon v. British Columbia (Ministers of Health. Children & Families. Education) (Supreme Court of BC, Mar 31, 1999) In M a y 1998, on behal f of her infant son, Miche l le Auton l a u n c h e d a n app l i c a t i on to serve as the representat ive plaintiff for autistic chi ldren in BC in a class ac t i on suit l a u n c h e d against the Province for failure to fund Lovaas Autism Treatment (LAT). The c l a iman t c i t ed a violation of s. 15(1) of the Canadian Charter of Rights and Freedoms (to b e discussed in a future section). Sect ion 15 of the Charter is frequently n a m e d in lega l app l i ca t ions as it speaks to the right to equal i ty: "every individual is equa l before a n d under the l aw" (Morris, 1996, p. 279). More specif ical ly , s. 15 prohibits discrimination b a s e d on r a ce , nat ional or ethnic origin, colour, religion, sex, a g e or menta l or physical disability" (pg. 279). In the Auton c a s e the c la im was b a s e d on menta l disability. The plaintiff was seeking relief for past a n d future costs of c a r e incurred for LAT. The c o m m o n issues in the class ac t ion suit identif ied by M a d a m Justice Mar ion Al lan (Allan J.) were : 1. Whether Lovaas was an appropr ia te form of autism treatment. 2. A n d , if so, whether the plaintiffs were entit led to funding (both past a n d future). 3. Whether the defendants fa i led to exercise discretion by refusing funding for Lovaas. 4. Whether the de fendants ' failure to fund Lovaas v io la ted s. 15(1) of the Charter . 5. Whether the defendants ' failure to inc lude Lovaas as a n insurable benefi t under s. 1 of the Medicare Protection Act v io la ted s. 15(1) of the Charter. The d e f e n d a n t Ministries of Health, Chi ldren & Families, a n d Educat ion o p p o s e d the class ac t ion app l i c a t i on a n d instead p roposed that the plaintiff's c a s e b e heard v ia a judic ia l rev iew process. A judic ia l review is, " a review by a Court of law of some ac t , or failure to ac t , by a government official or entity, or by some other legally a p p o i n t e d person or organ ized b o d y " (Gifis, 2003, p. 278). Al lan J . a g r e e d with the government a n d dismissed the class p r o c e e d i n g app l i ca t ion . Auton et al. v. Attorney General of BC and the Medical Services Commission of BC (Supreme Court of BC, July 26, 2000) The objec t i ve of this subsequent p r o c e e d i n g was to determine liability in regard to the Charter v iolat ion c l a im. If the C rown was found to b e at fault, appropr i a te remedies were to b e addressed in a future hear ing. Three add i t iona l chi ldren were a d d e d to the petitioners' original c l a im in order to represent the full spectrum of autism. The c l a im sought the fol lowing relief: (a) to rece ive a dec la ra t ion from the C rown that failure to fund Lovaas c o n t r a v e n e d s. 7 a n d s. 15 of the Charter, a n d (b) entit lement to funding for all past a n d future Lovaas expenses. While, s. 15 of the Charter deals with discrimination, S. 7 states: " eve ryone has the right to life, liberty a n d security of the person a n d the right not to b e depr i ved thereof e x c e p t in a c c o r d a n c e with the principles of fundamenta l justice" (para. 84) (also to b e discussed in a 27 future sect ion) . The C r o w n r e sponded by c la iming that the chi ldren were not d iscr iminated against. However , the C rown also stated that if there were a violation, it wou ld b e justifiable under s. 1 of the Charter (Section 1 provides Courts with the authority to potential ly limit con tes ted rights a n d freedoms) . O n e of the centra l issues of this c a se was the Lovaas e f f i cacy d e b a t e . The petitioners a r g u e d that with Lovaas treatment many autistic chi ldren show improvement in l a n g u a g e , social ization a n d intellect. On the other h a n d , the C rown c o n t e n d e d that the e f f i c acy of Lovaas h a d yet to b e establ ished a n d as such should not b e ruled a "med i ca l l y necessary serv ice" p rov ided under the Medicare Protection Act. Both counse l d rew u p o n autism experts a n d publ ished research studies on Lovaas therapy. In add i t ion , reports were commiss ioned to support e a c h party's position. O n behal f of the C rown , the British C o l u m b i a Of f i ce of Health Techno logy Assessment (BCOHTA) c o n d u c t e d o n e such study (Bassett et a l , 2000). This health t e chno logy assessment study per formed a systematic review a n d crit ical appra isa l of the e v i d e n c e on Lovaas therapy a n d c o n c l u d e d that, d u e to identif ied me thodo log i c a l flaws, there was a lack of scientific e v i d e n c e to support the effectiveness c l a im of this t rea tment opt ion . The petitioners, on the other h a n d , coun te r ed by stating that the go ld s tandard of r andomized contro l led trials in autism was often not feasible d u e to a lack of resources. They c i t ed expert opin ion that the "quasi-random assignment" me thodo logy utilized in these studies essentially a c c o m p l i s h e d the same o u t c o m e as r a n d o m assignment (Baer, 1993). In add i t ion , the petitioners f u n d e d a cost-benefit analysis of implement ing a Lovaas t reatment p rogram (Hi ldebrand, 1999). In the e n d , the j udge d iscredi ted the BCOHTA report, stat ing: "[that it] exh ib i t ed ] a n obvious bias towards support ing the Crown's position in this litigation that d e t r a c t e d ] significantly from its usefulness" (para. 48). In add i t ion , the Court c o n c l u d e d that it was b e y o n d its jurisdiction to direct government to specif ical ly fund Lovaas; therefore, the e f f i cacy d e b a t e was v i ewed as be ing purely a c a d e m i c a n d irrelevant to the o u t c o m e of the p roceed ings . The judge also dismissed the petitioner's cost-effectiveness report stating that it was impossible to est imate both the immed ia t e costs of implement ing a Lovaas p rogram a n d the " inev i table savings" (para. 145) to the government in the future. The Court ruled that the C rown h a d fai led to mee t its constitutional ob l igat ion of providing a "med i ca l l y necessary serv ice" (para. 102) a n d in do ing so h a d c o n t r a v e n e d s. 15 of the Charter (paras. 139 & 158). This violation was also d e e m e d unjustifiable under s. 1. The judge c o m m e n t e d that, " the respondents ' a rgument that they are unab le to prov ide ef fect ive t reatment for autism b e c a u s e of constraints in the legislation govern ing m e d i c a r e attempts to erect a false barrier" (para. 154). In fac t , the judge proffered two potent ia l funding models by wh ich to c o m p l y with the ministries' constitutional obl igations. The Court 's first suggestion was to inc lude EIBI in the MSP program a n d a d d behav ioura l therapists to the s c h e d u l e d list of health 28 c a r e providers; the s e c o n d suggestion was to p a y for the treatment through b lock funding (para. 154). Auton et al. v. Attorney General of BC (Supreme Court of BC, Feb 6, 2001) • At the subsequent r emedy hear ing, Al lan J . d i rec ted the government to fund gener ic Early Intensive Behavioural Intervention (EIBI). In addi t ion , s ince the.Ministry of Chi ldren a n d Families, be ing a socia l services ministry, l a c k e d the m a n d a t e a n d the expertise to deliver this treatment, the Court ruled that the p rogram should b e de l ivered through the Ministry of Health. The specif ics of the petitioners' c l a im for r emedy under the Charter i n c l uded the fol lowing: 1. A dec la ra t ion from the government that they b r e a c h e d the chi ldren's equal i ty rights. 2. An order of mandamus (such a n order wou ld c o m p e l the government to a b i d e by their statutory duty by initiating act ions to rectify their Charter violation). 3. F inancial compensa t i on for all past a n d future expenses for Lovaas t reatment (future t reatment durat ion a n d intensity to b e de te rmined by either a m e d i c a l pract i t ioner or psychologist) . While the ma in points of the Crown's submission inc luded the fol lowing arguments : 1. Their r e c o m m e n d a t i o n that the only r emedy should b e a dec l a ra t ion of their Charter violat ion, in add i t ion to the provision of funding for EIBI therapy for chi ldren with autism a n d autism spect rum disorder. In other words, they opposed the fol lowing petitioners' requests: a . Funding EIBI "whenever , a n d to the extent that, a m e d i c a l pract i t ioner or psychologist r e c o m m e n d s it" (para. 4). (Citing the reason that unlimited access to treatment wou ld restrict pol icy options with respect to diagnosis a n d assessment). b. Reimbursement for past a n d future Lovaas treatment. c. An order of mandamus (as the C rown felt the p roposed P-CARD program was e v i d e n c e of their intent to right a constitutional wrong.) d . Funding for EIBI past the a g e of 6 (stating that the a g e limit is b a s e d on e v i d e n c e , the Reasons for Judgment , a n d the petitioners' o w n c la im as to the effectiveness of t reatment for y o u n g chi ldren [para. 39]). In add i t ion , the C r o w n also sought a de lay , (a " remedia l transition per iod" ) , to submit the dec l a ra t ion of a Charter violation in order to provide sufficient t ime to implement the P-CARD program. However , the petitioners c h a l l e n g e d the validity of this p rogram. They a r g u e d that the Crown h a d not revea led the speci f ic t reatment protocols a n d staffing qual i f icat ions, a n d that the p rogram itself was nothing more than a sca led-up version of a n earlier pilot project. They d i s app roved of the a g e restriction of 6 years (instead of upon the discretion of a physic ian) ; the amoun t of therapy set at 20 hours/week (instead of 40 hours); assessment a n d diagnosis 29 per formed by a multidisciplinary t e a m (as o p p o s e d to a physic ian or psychologist) ; a n d the lack of a Lovaas c o m p o n e n t to the program (para. 24). In her dec is ion , Al lan J . c o n c l u d e d that: 1. The Court still c ou ld not direct government to fund or prov ide a spec i f ic treatment, nor c o u l d the Court d i c t a te how to a l l oca te scare health c a r e resources. 2. The e v i d e n c e showed that Lovaas h a d b e e n incorpora ted into other provinc ia l , U.S. a n d international autism programs a n d was suppor ted by many physicians in BC. 3. The Court still c o u l d not direct the government to prov ide EIBI, b a s e d on the a d v i c e of a phys ic ian or psychologist (as o p p o s e d to a multi-disciplinary team) . 4. The government was ob l i ga t ed to provide ef fect ive t reatment to autistic chi ldren to al lev iate their d i s a d v a n t a g e d state; this t reatment c o u l d not b e d e l a y e d or d e n i e d d u e to cost. 5. There was e v i d e n c e to support the c l a im that early ef fect ive t reatment of autism wou ld result in large cost savings to the government in the long term. 6. The judge was unab le to rule on whether the government b r e a c h e d its obl igat ions if it fa i led to fund EIBI past the a g e of 6 years. 7. A n order olmandamus was not war ran ted as the government was showing intent by implement ing the P-CARD program. Therefore, the Court ruled that (paras. 65-67): 1. The C rown must prov ide a dec la ra t ion to the petitioners that its failure to prov ide ef fect ive t reatment was a violation of their rights under s. 15 of the Charter. 2. The C rown must fund EIBI. 3. Each of the adult petitioners was entit led to "symbol ic d a m a g e s " of $20,000 (The judge de te rmined that a n order to c o m p e n s a t e the petitioners for all past a n d future Lovaas expenses c o u l d set a p r e c e d e n t for all subsequent petitioners). Symbol ic d a m a g e s were not b a s e d on costs incurred (or to b e incurred in the future); it merely r e cogn ized the role these petitioners undertook on behal f of all autistic chi ldren in BC. 4. P roceed ings against the M e d i c a l Services Commiss ion b e dismissed, as it was the government that was in violation of the Charter, not the MSC . 5. The petitioners should b e financial ly c o m p e n s a t e d for the costs of litigating the Charter c l a im . Auton et al. v. Attorney General of BC (Court of A p p e a l for BC, Oc t 9, 2002) At the Court of A p p e a l , a l though the Crown d id not a p p e a l the symbol ic f inancia l a w a r d , they d id a p p e a l the s. 15 Charter violation ruling. They also c o n t e n d e d that, should the appe l l an t Court ag r ee with the lower Court 's f inding, any. b r e a c h wou ld b e justified under s.l of the Charter. 30 Reciprocal ly , the petitioners c ross-appea led the r emedy stating that the symbol ic amoun t was insufficient to cove r their d a m a g e s . They also c ross-appea led the gener i c nature of the EIBI order b e c a u s e they were specif ical ly seeking Lovaas therapy. Upon review of the c a s e , one dissenting j udge (Justice Lambert) d i sag reed with the lower Court 's ruling on several points. With regard to the f inancia l remedy , Justice Lambert re jec ted the symbol ic a w a r d amount , instead proffering a formula that w o u l d best address the ac tua l costs incurred by the plaintiffs. Pertaining to the intensity a n d durat ion of the treatment, Justice Lambert d i sag reed with the a g e cut-off for Lovaas at six years. Instead, he r e c o m m e n d e d a n a g e limit of 14 years. In add i t ion , Lambert J .A. felt that instead of the Crown's proposa l for a multi-disciplinary t e a m assessment, t reatment should b e f u n d e d u p o n written d o c u m e n t a t i o n from the family physic ian a l ong with either a neurologist or a psychologist; these m e d i c a l practitioners wou ld b e the ones to determine the speci f ic type, intensity a n d durat ion or the p rogram. However , in the e n d , two of the three appe l l an t Court judges (Justices Hall a n d Saunders) uphe ld the lower Court 's ruling in regard to the Crown's Charter v iolat ion a n d the c h o i c e of f u n d e d treatment opt ion (generic EIBI). However, they c o n c l u d e d that, s ince Lovaas h a d proven ef fect ive for the four chi ld petitioners, the Ministry of Health should prov ide future funding for LAT for these spec i f ic chi ldren only (irrespective of their a g e , p rov ided that it was still de te rmined to b e ef fect ive by a m e d i c a l practit ioner). They also dismissed the petitioners' cross-a p p e a l . However , instead of a thorough assessment of the factors that may have contr ibuted to a s. 1 c l a im , the majority of the Court invoked its parens patriae jurisdiction (a duty bes towed on the Courts to protect the rights of those p e r c e i v e d to b e d i s a d v a n t a g e d , e.g., chi ldren). The C rown subsequent ly a p p l i e d for a n d was g ran ted l eave to a p p e a l to the Supreme Court of C a n a d a . Anderson et al. v. Attorney Genera/ of BC (Supreme Court of BC, 2003) In the Auton trial, Al lan J . reasoned that there was a g o o d c h a n c e that a dec l a ra t ion of unconstitutionality in favour of the plaintiffs wou ld p r o d u c e significant c h a n g e s in government pol icy that wou ld indirectly benefit other autistic chi ldren whose rights h a d b e e n v io la ted . However, s ince the eventua l o u t c o m e of Auton saw funding for LAT for only the four chi ld petitioners (and gener ic EIBI for the rest of the autistic children), the petitioners from the d e n i e d Auton class ac t i on suit l a u n c h e d their own Charter cha l l enge . In this trial the petitioners sought the s ame relief a w a r d e d to the plaintiffs in Auton. However, the Auton Court of A p p e a l j udgment was not a " conso l i da ted order" (para. 24) (e.g., i napp l i c ab l e to all autistic chi ldren in BC). In Anderson, Justice Pitfield ruled that the Crown h a d also v io la ted these petitioners' rights a n d , s ince the Court d e e m e d it unnecessary to litigate the same issues any further, he 31 declared that the petitioners were also entitled to LAT. However, Pitfield J. dismissed their claim to the $20;000 symbolic award. The Attorney General of BC subsequently appealed the Anderson decision while the child petitioners cross appealed (Court of Appeal for BC, 2004). The Crown requested an extension of the deadline to file their appeal factum until after the outcome of Auton's appeal to the Supreme Court of Canada . They argued that the two cases were intertwined and Auton's outcome would impact their case. Levine J.A. denied the Crown's request based on the child petitioners' contention that, since the Attorney General had historically elected to litigate each autism case individually, there would be no reason to delay the appeal . However, the Auton Supreme Court of Canada ruling was handed down before the Anderson appeal was heard. Therefore, In February 2005, on the initiative of the Crown, the Court of Appeal for BC allowed the Crown's appeal (and dismissed the petitioners' cross appeal), thereby setting aside the lower Court's ruling that granted Lovaas treatment to the remaining children from the original Auton case. Attorney Genera/ of BC, ef af., v. Auton. et al. (Supreme Court of Canada, 2004a) On June 9-10th, 2004, the Supreme Court of Canada heard the Auton appeal . In addition to submissions from the Attorney General of BC and counsel for Auton, the Court heard arguments from various interveners from across the nation such as other provincial Attorneys General, along with various social actors and autism groups such as Women's Legal Education and Action Fund (LEAF), the DisAbled [sic] Women's Network of Canada (DAWN), and Families for Early Autism Treatment (FEAT). Each stakeholder spoke to specific findings of fact from the lower Courts, and/or conceptual and constitutional inconsistencies. The Attorney General of BC's submission focused on whether the respective definitions of "benefits" and "health care practitioners" found in the BC Medicare Protection Act infringed on the petitioner's s. 7 and s. 15 rights as set out in the Charter. The Crown's position was that their funding decision was not discriminatory and that "a decision to delay, refuse, ration, or experiment with public funding for a new treatment [was] not a decision about the worth or dignity of those suffering the disabling condition" (Supreme Court of Canada , 2004b [argument of Cowper, G. - counsel for the Attorney General of BC]). It was further contended that a judicial process essentially usurped an essential governmental policymaking framework that takes into account a myriad of polycentric values. One of the Crown's central arguments was the fact that the Supreme Court of BC determined the therapy was efficacious and in doing so, inferred that a delay or refusal to provide an efficacious treatment constituted discrimination. The Crown did not contest that "behaviour therapy" was effective; the issue revolved around the inference of discrimination. In fact, the Crown admitted that their funding decision may have been "wrong, cheap, unwise, or late" (Supreme Court of Canada , 2004b [argument of 32 C o w p e r , G.]) but not discriminatory. They a r g u e d that the nature of the Auton c a s e was for funding, not equal ity. Another main a rgument presented by the C rown per ta ined to the lower Courts ' error in judicial reasoning in regard to the Charter discrimination analysis. This error was seen as be ing pe rpe tua t ed by Al lan J.'s incorrect app l i ca t ion of the Canada Health Act's principles of universality a n d comprehensiveness: The exclusion of effect ive t reatment for autistic chi ldren undermines the primary object ive of the M e d i c a r e legislation, wh i ch is to prov ide universal [italics a d d e d ] heal th ca re . (Supreme Court of BC, 2000, pa r a . 151) As previously men t i oned , in the Canada Health Act, universality refers to a c c e s s to the health c a r e system, whereas comprehensiveness refers to speci f ic services de l i vered by physicians a n d dentists within hospitals. The C rown po in ted out that there was no constitutional or statutory right to ef f icac ious treatment. In fact , they w a r n e d the Supreme Court that a ruling in support of a violation of s. 15 of the Charter wou ld in ef fect c rea te a "const i tut ional ly-mandated right to publ ic fund ing " for all e f f icac ious programs. The submission by Chris Hinkson, counse l for Auton, h ighl ighted the d isputed facts of the case . First of all , he no ted that the C rown d id not a p p e a l Al lan J.'s ruling that the Lovaas issue resided within the s c o p e of the Ministry of Health's jurisdiction (as o p p o s e d to the Ministries of Educat ion or Chi ldren a n d Families). S e c o n d , the C rown d id not initially a p p r o a c h the c a s e using a f inancia l a rgument (as he contes ted they were now doing) . Instead, the Crown 's strategy h a d b e e n to d e b a t e the Lovaas e f f i cacy c l a im . However, Mr. Hinkson a r g u e d that it wou ld cost b e t w e e n thirteen a n d thirty-three million dollars per year not to treat autism in BC. Third, Chris Hinkson s tated that there was no e v i d e n c e that the C rown w e i g h e d the effect iveness of Lovaas treatment, or ente red into a cost-effectiveness analysis. This is a n interesting s ta tement in v iew of the fac t that the C rown h a d commiss ioned the BCOHTA report, wh i ch was g iven significant attent ion by Al lan J . in her reasonings. A n d finally, Mr. Hinkson c o n t e n d e d that there was no e v i d e n c e that suppor ted a systematic a n d transparent a p p r o a c h to this pol icy opt ion on the part of the government . To this remark Justice Binnie proffered a suggestion that an appropr ia te order from the lower Courts might have b e e n to direct the government to c o n d u c t a proper pol icy analysis in order to determine entit lement. Some of the submissions from the interveners in the c a s e presented slightly var ied perspect ives. For examp le , the Autism Society of C a n a d a (ASC) h a d a more systems a p p r o a c h , focusing on government ' s lack of a "constitutionally appropr ia te process a n d criteria to support its dec is ion not to fund Lovaas " (Supreme Court of C a n a d a , 2004b [argument of Crol la , D.A. & O'Brien, M.K.]) They stressed the n e e d for accountab i l i t y a n d t ransparency in pol icy deve l opmen t . The C a n a d i a n Associat ion for Commun i t y Living (CACL) a n d the Counc i l of C a n a d i a n s with Disabilities (CCD) on the other h a n d reasoned that a c o m p a r a t o r group analysis 33 was not appropr i a te for a s.15 Charter analysis. Instead, they felt that the pressing question was: "D id they [autistic children] get wha t they n e e d e d ? " (Argument of Shilton, E.J;, Fa raday F.C. & C h a d h a , E.) Their position was that Courts n e e d e d to supervise the frameworks that governments use for resource a l locat ion b a s e d on equal ity rights; they a r g u e d that current decis ion-making processes d o not inc lude a n i m p a c t analysis of these rights. FEAT of Ontar io a n d Alberta a rgued that a crit ique of the Auton c a s e written by D o n n a Greshner a n d Steven Lewis (2003) was in fac t a "discriminatory report". This pape r , wh i ch was frequently r e f e renced by various interveners in this case , e x a m i n e d the cla ims m a d e in Auton a n d the judic ia l reasoning of the lower Courts that resulted in pol icy be ing d i c t a t e d by law. FEAT felt that the Auton c a s e was not abou t establishing a constitutionally g u a r a n t e e d right to health c a r e but rather ensuring access to a core , med ica l l y necessary a n d publ ic ly f unded heal th t reatment for vulnerable autistic chi ldren. A n interesting submission was presented on behal f of an adult with autism - Miche l le Dawson . Her counse l s tated that she d id not support either side in the c a s e a n d resented the assumption that autistic chi ldren required treatment in order to b e "no rma l " (Ms. Dawson h a d refused EIBI/ABA therapy) . In add i t ion , her lawyer d isputed the often s tated statistic of only 1:64 chi ldren a ch i e v ing "normal funct ion ing" without some form of intervention stating that the research was publ ished in 1970. Their a rgument c en te r ed on the fac t that neither party h a d taken the t ime to consult an autistic person. O n the respondents ' side, counse l for the chi ldren c l a i m e d that the government was funding med ica l l y necessary treatment for non-disabled C a n a d i a n chi ldren a n d adults with menta l illness, yet autistic chi ldren were be ing d e n i e d the same benefi t . McLach l i n C J C found that such a n assertion e q u a t e d to funding for gjl medica l l y required treatment. O n N o v e m b e r 19, 2004 the Supreme Court of C a n a d a h a n d e d d o w n its judgment (Supreme Court of C a n a d a , 2004a). The Court a l l owed the a p p e a l , a n d ruled that the BC government ' s c o n d u c t h a d not infringed upon the respondents ' equal i ty rights under s. 15 of the Charter. It found that both the Canada Health Act a n d the Medicare Protection Act fa i led to support the respondents ' assumption that all med ica l l y required or necessary non-core services must b e f u n d e d . Instead, the Court po in ted out that the Acts confer only funding for co re services p rov ided by m e d i c a l practitioners; any funding for non-core services was at the discretion of e a c h Province. The Acts were not i n t ended to mee t all m e d i c a l needs of the p e o p l e ; in fact , McLach l i n C J C c a l l ed the legislative s c h e m e " a partial heal th p l a n " (para. 43). However, some non-core services c a n b e funded through the M e d i c a l Services Commiss ion (MSC) of BC (the administrative b o d y that governs the Medicare Protection Act). Nevertheless, at the t ime of the original trial, behav ioura l therapists were not a r ecogn ized health c a r e practit ioner, nor d id the MSC have the authority to c r ea te this c lassif icat ion. Therefore, ABA/IBI therapy c o u l d not b e f u n d e d under the p lan . Consequent ly , the Court 's ruling was that BC's law 34 govern ing non-core benefits d id not support the respondents ' c a s e . Therefore, b a s e d on this analysis, the Supreme Court held that funding for all med ica l l y required services was not a constitutional right p rov ided for by the law. A po int of interest in McLach l in ' s s. 15 analysis was her specu la t ion on the o u t c o m e of the inquiry if the respondents h a d f r amed their c a s e differently: "as a c l a im to the benefi t of equa l app l i c a t i on of the l aw by the M e d i c a l Services Commiss ion" (para. 45). Her hypothesis be ing that the health c a r e s c h e m e itself was discriminatory as it p rov ided funding for some non-core therapies whi le refusing to fund a n "equa l ly necessary" therapy such as ABA/IBI. However this c l a im wou ld have b e e n cont ingent on showing that there was a benef i t p rov ided by law. This means that, at the t ime of the c a se , there n e e d e d to b e a non-core service (the "benef i t " ) be ing p rov ided under legislation. Therefore, if behav ioura l therapists h a d b e e n listed as health c a r e practitioners then the M e d i c a l Services Commiss ion wou ld have b e e n ob l i ged to prov ide LAT. Another issue centra l to the s. 15 analysis was the c h o i c e of a c o m p a r a t o r g roup . Here the respondents h a d to prove that they were d e n i e d a benefi t that was typical ly ava i l ab le to a similar g roup of p e o p l e . The respondents chose their c o m p a r a t o r g roup to b e non-disabled chi ldren a n d their parents, a l ong with mental ly ill adults. However, the Court d i sagreed a n d d e c i d e d that a more appropr ia te c o m p a r a t o r group w o u l d b e a "non-disabled person or a person suffering a disability other than a menta l disability . . . seeking or rece iv ing funding for a non-core therapy important for his or her present a n d future heal th, wh i ch is emergen t a n d only recently b e c o m i n g r e cogn ized as med ica l l y requ i red" (para. 55). The Court a c k n o w l e d g e d the " emergen t " nature of Lovaas therapy a n d d e c i d e d , " p e o p l e rece iv ing well-established non-core therapies are not in the s ame position as p e o p l e c la iming relatively n e w non-core benefits" (para. 55) a n d that governments may have legit imate reasons for de lay ing or deny ing funding. McLach l in C J C po in ted to the fac t that e v i d e n c e on how the gove rnment r e s p o n d e d to requests for non-core services from non-disabled or otherwise d i sab led p e o p l e wou ld have b e e n benef ic ia l in determining whether the government discr iminated against the ch i ld petitioners, or whether it p r o c e e d e d similarly with all requests. Based on this lack of e v i d e n c e , the Supreme Court of C a n a d a c o u l d not support the c l a im of discrimination. A n d finally, the petitioner's also c l a i m e d a violation of s. 7 of the Charter. However , McLach l i n C J C po in ted out that the respondents d id not identify the pr inciple of f undamenta l justice that h a d b e e n c o n t r a v e n e d by the refusal to fund Lovaas therapy. Principles of f undamenta l justice are general ly a g r e e d u p o n lega l principles that are "vital or fundamenta l to our soc ieta l notion of just ice" (Supreme. Court of BC, 1993, pa ra . 590). Therefore, b a s e d on a lack of e v i d e n c e , there was no validity to c l a im a s. 7 v iolat ion. 35 A g a i n , to better understand the context in wh i ch the Au ton a n d Anderson trials p l a yed out, a n in-depth look at the C a n a d i a n legal system a n d some of its re levant statues, a l ong with some re levant cases , is required. Table 1.1 be l ow provides a summary of the pertinent lega l cases. Table 1.1 Summary of Legal Cases a n d Judgments Case Auton (Mar 31, 1999) Supreme Court of BC Auton etal. (July 26, 2000) Supreme Court of BC Auton etal. (Feb 6, 2001) Supreme Court of BC Auton etal. (Oct 9, 2002) Court of A p p e a l for BC Anderson et al. (Aug 22, 2003) Supreme Court of BC Auton etal. (Nov 19, 2004) Supreme Court of C a n a d a Issues Class ac t ion app l i ca t ion against BC government for failure to find Lovaas Autism Treatment (LAT) [Charter cha l l enge ) . Determinat ion of liability in regard to Charter v iolat ion. C l a im for r emedy hear ing. Parents sought funding for all past a n d future LAT expenses. Province a p p e a l e d the Charter s. 15 violation. Petitioners a p p e a l e d the $20,000 remedy a n d the gener ic nature of EIBI services. Similar c a se brought against the prov ince by the remaining chi ldren left out of the original Au ton p roceed ings . Whether a provincial health p lan is ob l i ga t ed under the Canada Health Act to prov ide treatments outside of the co re services, a n d if so, whether failure to fund these services const i tuted a violation of the Charter. Judgments Class ac t i on d e n i e d . C a s e p r o c e e d e d v ia judic ia l review. Province found in violation of the Charter for not providing a "med i ca l l y necessary" service. O rde red to fund Early Intensive Behavioural Intervention (EIBI). Parents only g ran ted $20,000 as a "symbol ic a w a r d " . Lower court rulings uphe ld . However, LAT a w a r d e d to the four ch i ld petitioners. Province a g a i n found in violation of the Charter. Chi ldren g r an ted funding for LAT, however d id not rece ive the $20,000 "symbol ic a w a r d " . Province of BC w o n their a p p e a l . Court ruled that the Canada Health Act a n d the Medicare Protection Act does not prov ide a legal right to all med ica l l y required or necessary non-core services. 36 Canadian Legal System Canada 's judicial system (with the exception of Quebec) is derived from the English structure, which is based on judge-made law called common law. During the early 11 t h century the English government appointed judges to settle disputes. Lacking formal laws upon which to guide their decisions, these early cases were tried based.on custom, tradition, societal norms and common sense. These historical judgments set precedent for future trials thus providing the basic legal principles, concepts, language, methods of legal reasoning, and divisions of law for the modern day legal system. Since confederation in 1867, Canadian laws have evolved principally from federal parliamentary and provincial legislative laws called statutes. The Constitution Act of 1982 (Canada Act) granted authority to the Provinces to enact their own statutes within the legislative scope defined in the Act. Statutory laws prevail over common law (Sneiderman, et al, 1995). In the common law system, either the judge or jury is charged with the determination of facts and the application of the relevant law to those facts. If a trial is by jury, the judge is responsible for determining the legal issues while the jury decides the factual matters. If the trial is by judge alone, their role is to provide objective arbitration over specific disputes. Objective arbitration is a process that initially seeks out the truth and then applies specific rules of procedure and rules of evidence in order to determine the outcome of the case (Gall, 2004). Health law is a relatively new discipline that has risen from tort law. The law of torts is a branch of common law that deals with injuries or damages inflicted upon a person by another. A number of factors have contributed to health law becoming a discipline unique unto itself. These include the scientific explosion in areas such as genetic testing and biotechnology; complex ethical dilemmas brought about by end of life issues such as euthanasia; unsatisfactory patient care outcomes as a result of health care reform policies; and privacy issues over the advancement of information technology, just to name a few (Downie & Caulfield, 1999). The function of the Canadian Courts is, "to provide fair and just resolution of the various problems and conflicts that are brought before them" (Gall, 2004, p. 209). This judicial decision-making takes place within the context of the adversarial system where the assumption is that the truth is most likely to be uncovered. One of the key features of judicial decision-making is the process of adjudication. The role of adjudication is in settling disputes between private individuals or groups, or the government and the individual. Usually these are not future debates of general policy issues; instead, they are immediate controversies that occur when opposing interests intersect. Gall (2004) describes the fundamental objectives of the Canadian judicial system as the following: 37 ur Courts must entertain a search for truth, a n d that search for truth must b e c o n d u c t e d in a manner a n d with the result that might, broadly speak ing , b e cha rac te r i zed as the dispensat ion of justice. In turn, the dispensat ion of justice must not only b e d i rec ted at those persons a p p e a r i n g before our Courts, but also at the s ame time it must b e d i rec ted to the best interests of society at large, (p. 212) The C a n a d i a n judiciary is d i v ided into a hierarchy of Courts e a c h with their own unique jurisdictional responsibilities. Provincial Courts are gove rned by their respect ive provincia l statutes a n d consist of Courts of inferior jurisdiction (e.g., provincia l Courts) to Courts of superior jurisdiction (Supreme Court a n d Court of Appea l ) . However, while Provinces appo in t lower provincia l Court judges (e.g., Youth Court , Family Court, Small Claims Court a n d the criminal jurisdiction of the Provincial Court) , the higher Courts are under a . federa l constitutional limitation whereby judges must b e federal ly a p p o i n t e d . While e a c h Province shares similar fundamenta l object ives, the manne r in wh i ch they are const i tuted varies a m o n g Provinces. This results in different responsibilities for apparent l y similar Courts across the country. Procedures are in p l a c e whereby parties c a n dispute the o u t c o m e of trials. If in counsel 's opin ion the j udge has erred in either the finding of fac t or in the app l i c a t i on of law the opt ion to a p p e a l the dec is ion is ava i l ab le through the provincia l Court of Appeal. A pane l of the Court of A p p e a l usually consists of three judges; majority dec ides the o u t c o m e of the a p p e a l . A l though error in f ac t is grounds for a p p e a l , the m a n d a t e of the appe l l an t Court is primarily to j udge the app l i c a t i on of l aw to the facts as outl ined in the Reasons for Judgment. The unsuccessful litigant seeking to over-turn the trial dec is ion is c a l l ed the appellant while the successful party is t e rmed the respondent. If the successful party also disputes the judgment (for e x a m p l e , asserts that the f inancia l sett lement was not sufficient) they are c a l l e d the appellant on cross appeal a n d the other party is the respondent on cross appeal (Sneiderman, et a l , 1995). The doctrine of precedent d ictates that o n c e a higher Court rules on a speci f ic issue the lower Courts must app l y the same ruling (provided that the facts of the c a s e are similar in nature). H e n c e , d e p e n d i n g on the position e a c h counse l is taking, the lawyers' m a n d a t e is to careful ly examine the facts of p r e c e d e n t cases to determine whether they app l y to their, c a s e before the Court or whether there are distinguishing facts that wou ld render the doctr ine of p r e c e d e n t not a p p l i c a b l e (Sneiderman, et a l , 1995). The federa l Courts consist of the Federal Court of C a n a d a , the Tax Court of C a n a d a a n d the Supreme Court of C a n a d a . These Courts are gove rned by federa l statues with federal ly a p p o i n t e d judges. The Supreme Court of Canada is the highest Court in the country a n d functions as the ult imate appe l l a t e Court for the Court system. The Supreme Court is compr i sed of nine judges (one Chief Justice of C a n a d a a n d eight "Puisine Justices") that preside in an o d d number in order to prevent dec is ions 'be ing split evenly. The number of judges that hear the p roceed ings is b a s e d on the severity of the ease (Sneiderman et a l . , 1995). Be tween 1984 a n d 38 1997 the Supreme Court of C a n a d a nullified thirty-four federa l a n d seventeen provincia l statutes (Manfredi & Ma ion i , 2002). Canadian Charter of Rights and Freedoms The Canadian Charter of Rights and Freedoms was e n a c t e d in 1982 as part of the " n e w " Constitution of C a n a d a of the Canada Act. The Constitution of C a n a d a is the supreme l aw of the nat ion. The Charter r e p l a c e d the 1960 8/7/ of Rights, wh i ch at the t ime was powerless over provincia l statutes as it was not a constitutional d o c u m e n t . The Charter conta ins essentially the same elements as the original 8/7/ of Rights, but individuals' rights a n d f reedoms were more clearly cod i f i ed . In add i t ion , the Charter is a p p l i c a b l e to both federa l a n d provincia l levels of government a n d is i n t ended to protect the p e o p l e from arbitrary or il legal legislation. If a statute appea r s to con t r avene the Charter it c a n b e c h a l l e n g e d in the Courts a n d potential ly d e e m e d unconstitut ional (Sneiderman, et al , 1995). The only excep t ion is found in s. 1 "where the government c a n show reasonab le a n d demonst rab le justification in a free a n d d e m o c r a t i c soc iety" (Mykitiuk & Wal lrap, 1999, p. 320). Therefore, rights a n d f reedoms are not absolute, a n d Courts are c h a r g e d with we igh ing the c l a im against all other c o m p e t i n g soc ieta l interests. Accusat ions of violation of the Charter are but o n e mechan i sm by wh i ch plaintiffs may seek access to spec i f ic health c a r e services. Be tween 1984 a n d 1997 the federa l gove rnment was ab l e to d e f e n d against Charter cha l l enges in 62.5% of the cases , while provincial governments have b e e n successful 72.3% of the t ime. The nullified provincia l statutes were significantly 'younger ' than the annu l led federa l statues with a m e a n ' a g e ' of ten years as o p p o s e d to twenty-three (Manfredi & Ma ion i , 2002). Section 15 Sect ion 15 of the Charter is frequently c i t ed in lega l appl icat ions since it speaks to equal i ty rights: "every individual is equa l before a n d under the l aw" (Morris, 1996, p. 279). More specif ical ly, s. 15 prohibits discrimination b a s e d on race , nat ional or ethnic origin, co lour , religion, sex, a g e or menta l or physical disability" (pg. 279). S. 15 claims in heal th c a r e are typical ly f o cused on e x p a n d i n g the s c o p e of insured services. In these cases, patients will a l l ege that a spec i f ic heal th c a r e pol icy either exc ludes them from c o v e r a g e or reduces their share of resources, thus violating their rights to equa l protect ion, a n d equa l benefi t of the law. While the potent ia l for s. 15 c la ims is large, in ac tua l fac t only a few cases have b e e n successful. S ince Auton represents a s c o p e of c o v e r a g e case , in order to prove a violation of equal i ty rights, (under the steps establ ished in Law v. Canada ), the plaintiff must prove three things: (a) differential t reatment b e t w e e n groups b a s e d on a personal character ist ic ; (b) the differential t reatment must b e on a g round that is e n u m e r a t e d (powers g ran ted by the Constitution) in s. 15 or ana logous to an e n u m e r a t e d g round ; a n d , (c) the differential t reatment must b e so substantive that it offends the plaintiff's essential human dignity (Greschner, 2002). 39 However, rationing of health services d u e to cost , risk, safety a n d low effect iveness wou ld not constitute a violat ion of rights. Section 7 Sect ion 7 of the Charier states that, "everyone has the right to life, liberty a n d security of the person a n d the right not to b e depr i ved thereof e x c e p t in a c c o r d a n c e with the principles of fundamenta l just ice" (Supreme Court of BC, 2000). However, proving this in the heal th c a r e context has b e e n less successful. A s. 7 c l a im requires that the plaintiffs initially show that either "l iberty" or "security" incorporates health or heal th c a r e services. For e x a m p l e , Rodriguez v. British Columbia (a case on euthanasia) was successful in arguing that a n individual 's right to refuse m e d i c a l t reatment fell within the Charter's jurisdiction of "security". Then, o n c e the plaintiff has establ ished the liberty/security criteria, they must prove that the depr ivat ion of the right to liberty a n d security con t r a vened the principles of fundamenta l justice; this means that the bas ic tenets of the lega l system wou ld have to have b e e n b r o a c h e d . S. 7 cha l l enges in health c a r e are incredibly hard to prove as this rule typical ly appl ies more to criminal cases. Section 1 A n d finally, s. 1 of the Charter conta ins the "limitations c l ause " , wh i ch provides Courts with the authority to potential ly limit con tes ted rights a n d f reedoms. The framework for judic ia l reasoning a n d analysis of a s. 1 c l a im is referred to as the Oakes test as the criteria for support ing the limitations c lause was d e c i d e d in R. v. Oakes (1986). S. 1 allows governments to justify violations to the Charter as reasonab le limits. In order to d o this though , two criteria must b e satisfied: (a) " the i m p u g n e d law must pursue an object ive that is sufficiently important to justify limiting a Charter right", a n d (b) " the government must demonstrate that the means chosen to attain it are reasonab le a n d demonstrab ly justified" (Hartt & M o h a h a n , 2002, p. 23). Satisfying the first criteria is not difficult as governments c a n argue that their pol icy object ive is to protect the M e d i c a r e system by effect ive use of resources. However, the s e c o n d criterion is somewha t harder to demonst ra te a n d requires c lear e v i d e n c e that other pol icy options have b e e n cons ide red , that the pol icy impa i red the right or f r eedom as little as possible, a n d that it should not disproport ionately ef fect the ta rge ted popula t ion (Greschner, 2002). In essence , the s. 1 requirement of demonstrabi l i ty resembles the ev idence-based a p p r o a c h taken by pol icy -makers. In fac t : In the majority of Charter cases Courts perform their most important task not in def ining the substantive mean ing of rights or liberties, or in measuring gove rnment ac t i on against those definitions, but in determining the s c o p e of " reasonab le limits" on rights under s. 1 of the Charter. (Maioni & Manfred i , 2003, p. 928) 40 Other Health Care Charter Cases While Auton certainly c a p t u r e d the med ia ' s at tent ion, some other health c a r e cases that both p r e c e d e d a n d fo l lowed Auton h ave also contr ibuted to this ongo ing d e b a t e abou t judic ia l po l i cymak ing a n d Charter cha l l enges be ing used to access health c a r e services. CR. v. Alberta (Director of Child Welfare) (Alberta Court of Queen's Bench, 1996) Facts and Issues This c a s e was the first to consider funding for treatment for autistic ch i ldren. However, as o p p o s e d to determin ing wha t type of t reatment to fund, it focused on wh i ch gove rnment depa r tmen t was a c c o u n t a b l e for del ivering it. Here, the parents of a n autistic ch i ld sought government funding for Lovaas therapy. In this part icular situation, the chi ld was d i a g n o s e d later than usual at the a g e of 5. He went on to rece ive Lovaas treatment beg inn ing at the a g e of 6 for 20 hours per w e e k outside of schoo l hours. After personally funding the therapy for nine months, the family reques ted f inancia l assistance from their school district. The app l i c a t i on was refused b a s e d on the c la im that the district was not responsible for support ing programs outside of normal schoo l hours. Then, pursuant to the Child Welfare Act, the family a p p r o a c h e d Alberta 's H a n d i c a p p e d Chi ldren's Services (a division of A lber ta Family a n d Socia l Services). O n c e a g a i n , their request was d e n i e d cit ing the ruling that spec ia l e d u c a t i o n programs were within the rea lm of the School Act a n d as such superseded the Ch/7d Welfare Act. The parents p r o c e e d e d to file l ega l a c t i on against the Director of Ch i ld Welfare c l a im ing a violat ion of s. 15 of the Charter of Rights and Freedoms. Three of the key issues for these p roceed ings were as follows: (a) whether the Director of Chi ld Welfare erred w h e n determining that Lovaas was under the exclusive jurisdiction of the School Act, (b) whether the Director of Chi ld Welfare v io la ted s. 15, a n d (c) if the Charter was v io la ted , whether the Court should order funding pursuant to s. 83 of the Child Welfare Act. Reasons Justice Deyel l ruled that the Director of Ch i ld Welfare h a d erred by charac ter iz ing Lovaas as a n e d u c a t i o n a l p rog ram; therefore, there was no justifiable reason to withhold t reatment . In add i t ion , d e p e n d i n g u p o n the part icular c i rcumstances of e a c h individual c a se , the j udge c o n c l u d e d that Lovaas c o u l d also b e appropr ia te ly f unded under the School Act. The Court 's dec is ion on this key issue m a d e the question of a Charter violation irrelevant; therefore that c l a im was dismissed. Justice Deyell o rdered the Director of Chi ld Welfare to fund 9 0 % of the costs of therapy for a one-year per iod a n d to reimburse the parents 9 0 % for expenses incurred over the previous year. The judge d e c i d e d that Lovaas therapy was to b e f u n d e d b y the Ministry of Chi ldren's Services a n d not through health insurance. However , subsequent cases have d e c l i n e d to follow the legal reasonings in CR. In one part icular c a s e (Alberta Court of Queen ' s Bench , 1999) Justice Rawlins c o m m e n t e d on these p roceed ings stating that the Courts should not h a v e interfered with the government 's right to d e c i d e whether or not to fund speci f ic 41 services, nor should it have ordered a certa in p e r c e n t a g e of the services to b e f u n d e d (paras. 9 & 22). In any regard , the o u t c o m e of CR. was that A lber ta instituted a pilot project utilizing ABA interventions; Ontar io, Prince Edward Island, Newfound land a n d M a n i t o b a all fo l lowed suit shortly thereafter with their own adapta t ions of ABA programs (Supreme Court of BC, 2000, paras. 73-81). DJ.N. v. Alberta (Child Welfare Appeal Panel) (Alberta Court of Queen's Bench, 1999) Facts and Issues In this c a se , D.J.N, a p p e a l e d a dec is ion by the Ch i ld Welfare A p p e a l Panel that struck d o w n a request to prov ide f inancia l assistance for a variety of add i t iona l programs for her autistic son B.N. These programs inc luded s p e e c h therapy, o c c u p a t i o n a l therapy, a Read ing Foundat ion p rogram, socia l skills a n d compu te r l a n g u a g e training. B.N. was enrol led in a spec ia l e d u c a t i o n p rogram, but was on the wai t ing list for some add i t iona l services such as s p e e c h therapy. The appe l l an t c l a i m e d the chi ld was unab le to k e e p up with his peers, requiring services b e y o n d wha t the school was provid ing. The Director of Chi ld Welfare d e n i e d these requests on the basis that it was not within his jurisdiction to grant these services. The appe l l an t a p p e a l e d the dec is ion to the Chi ld Welfare A p p e a l Panel w h o in turn uphe ld the dec is ion stating that the requested services were under the jurisdiction of the School Board under the School Act. The questions for the Court were : (a) Did the Panel fail to app l y the l aw equal ly to B.N.? a n d , (b) Did the Panel err in f inding it d id not have jurisdiction over the reques ted services? (para. 7). Reasons Justice Rawlins found that the Panel d id not violate s. 15(1) of the Charter, as the e v i d e n c e on whether the appe l l an t r ece i ved inconsistent t reatment was inconc lus ive . In add i t ion , entit lement to services under the Ch/7d We/fare Act was at the discret ion of the Director. A n d , in regard to the s e c o n d quest ion, the Court de te rmined that the Panel h a d not erred in f inding it d id not have jurisdiction over the requested services. Eldridae v. British Columbia (Attorney General) (Supreme Court of Canada, 1997) Facts and Issues This was the p r e c e d e n t c a s e for all future health c a r e Charter l itigation. Here, three dea f appel lants sought a dec la ra t ion that the failure to provide sign l a n g u a g e interpreters as a n insured benefi t v io la ted s. 15 (1) of the C a n a d i a n Charter of Rights and Freedoms. At the time of l itigation, neither the Hospital Insurance Act nor the Medical and Health Care Services Act (as it was known at the time) c o v e r e d l a n g u a g e interpreters. The appel lants asserted that, " the a b s e n c e of interpreters impaired] their ability to c o m m u n i c a t e with their doctors a n d other health c a r e providers, a n d thus increase [d] the risk of misdiagnosis a n d ineffect ive t reatment" (para. 3). 42 Eldridge rose to the Supreme Court of Canada in 1997 where it was debated whether the definition of "benefits" in s.l of the Medicare Protection Act, in addition to various sections of the Hospital Insurance Act, infringed on s. 15(1) of the Charter. If the Court determined there was a violation, then the question remained whether it was saved by s. 1 of the Charter. In addition, there was the issue in regard to the extent that decisions made by private entities become subject to Charter review and the extent to which governments must provide equal access to public services for the disabled (Cornish & Faraday, 1997). Reasons The Supreme Court of BC ruled against the s. 15(1) Charter violation with Justice Tysoe's reason being that "sign language interpretation [was] ancillary to medically required services in much the same way as [was] transportation to a doctor's office" (para. 11). In his view, the Charter did not stipulate which programs a government should implement, only that the implemented programs should be distributed equally. In 1995 the Court of Appeal for BC upheld the lower Court ruling stating that, "the lack of interpreting services in hospitals [was] not discriminatory because the Hospital Insurance Act [did] not provide any "benefit of the law" within the meaning ofs. 15(1) of the Charter" (para. 13). The absence of interpreters was not caused by the legislation but by the hospital's discretion as to how their global budget was distributed. And, since hospitals were not "government" as defined in s. 32 of the Charter their failure to provide these services did not constitute a violation of s. 15(1). The Court also found that the Medical and Health Care Services Act also did not violate s. 15(1) because there was no distinction between the deaf and hearing populations in that both populations had the benefit of free medical services. On appeal to the Supreme Court of Canada (1997), the majority overturned the lower Courts' rulings and found that the government's actions violated s. 15(1) of the Charter. They concurred that, "the evidence clearly demonstrate^] that, as a class, deaf persons receive[d] medical services that [were] inferior to those received by the hearing population" (para. 94). By providing a benefit, the government had to ensure its accessibility to all. Even though hospitals are private entities, their responsibility to uphold the Charter is extended by way of their contract to implement government policies. Eldridge therefore broadened the range of entities and activities that could theoretically be subject to a Charter challenge. Chaoulliv. Quebec (Attorney General) ( 2 0 0 0 , 2 0 0 2 , 2 0 0 5 ) Facts and Issues George Zeliotis was a Quebec resident that sought out private health insurance due to delays he had experienced in receiving treatment through the publicly funded health care system. However, s. 15 of the Quebec Health Insurance Act and s. 11 of the Quebec Hospital Insurance Act prevented him from doing so. 43 J a c q u e s Chaoul l i was a physic ian w h o sometimes worked outside of the publ ic health c a r e system. He h a d a p p l i e d to the Q u e b e c Health Insurance Board to ope ra t e a private "opted-out " hospital but was d e n i e d . In add i t ion , his m e d i c a l p rac t i ce of prov id ing h o m e visits on a 24-hour basis was also in quest ion. His c o n c e r n was that timely heal th c a r e wou ld not b e ava i l ab le should he or his family require it. He, too, felt he should b e ab l e to purchase private insurance to cove r m e d i c a l services currently p rov ided within the pub l i c system. Chaoul l i a n d Zeliotis brought a motion before the Superior Court of Q u e b e c seeking a dec la ra t ion that s. 15 of the Heath Insurance Act a n d s. 11 of the Hospital Insurance Act v io la ted ss. 7, 12 a n d 15 of the Charter in addi t ion to ss. 1, 4, 5 a n d 24 of the Q u e b e c Charter of Human Rights and Freedoms. However, the real issue that was identi f ied by the Superior Court j udge was the introduct ion of a private health c a r e system paral lel to the publ ic system (Manfredi & Maion i , 2005; Supreme Court of C a n a d a , 2005; T i edemann , 2005a). Reasons The Superior Court of Q u e b e c (2000) d id not analyze the c a s e in respect to the Q u e b e c Charter. Instead, M a d a m Justice P iche identif ied the key issue as whether the provisions of the two provincia l Acts v io la ted the app l i cants ' rights under s. 7 of the Canadian Charter. She de te rm ined that the right to obta in private insurance was an "anci l lary or inc identa l e c o n o m i c right" (T iedemann, 2005a, p. 4) s ince in this c a s e it was l inked to life, liberty or security of the person. She a g r e e d that the Q u e b e c Acts c r e a t e d a barrier to a c ces s required heal th services. However, a s. 7 violation of the Canadian Charter cou ld only b e d e c l a r e d if the publ ic health system c o u l d not gua ran tee a ccess to services, wh i ch was not the situation in this c a se . Therefore, no violation of s. 7 resulted a n d the Court ruled that the app l i can ts ' rights were not v io la ted under either ss. 12 or 15 of the Canadian Charter. The c a s e p r o c e e d e d to the Q u e b e c Court of A p p e a l in 2002 whe re it was dismissed. All three Justices c o u l d not concu r on whether the right was purely a n e c o n o m i c right or a n ancil lary or inc identa l e c o n o m i c right. A n d finally in 2005, four of the seven justices of the Supreme Court of C a n a d a a g r e e d that prohibit ing the purchase of private insurance v io la ted s. 1 of the Q u e b e c Charter a n d that such a violation was not justified under s. 9(1). The Court, however , was equal ly split in regard to whether this prohibition v io la ted s. 7 of the C a n a d i a n Charter s ince one Justice only ana l yzed the c a s e from the perspect i ve of the Q u e b e c Charter. M a d a m Justice Deschamps c o n c l u d e d that the e v i d e n c e d id not support the hype surrounding the content ion that a l lowing private insurance to cove r services normally de l ivered by the publ ic system wou ld result in a paral lel system that wou ld b l e e d the publ ic ly f unded o n e . The dissenting justices were cr i t ical of the majority dec is ion in that they felt the Court h a d not de f ined " h o w m u c h health c a r e is ' r easonab le ' e n o u g h to satisfy s. 7 of the Canadian Charter of Rights and Freedoms . .. a n d s. 1 44 of the Charter of Human Rights and Freedoms" (Supreme Court of C a n a d a , 2005, p a r a . 163) (Flood & Sullivan, 2005). Barclay (Guardian ad litem of) v. British Columbia (Attorney General) (Supreme Court of BC, 2005) Facts and Issues This ac t i on was initiated in M a r c h 2002 in response to the o u t c o m e of Auton. Patrick Barclay was a n 11 year-old chi ld with autism w h o d id not rece ive any autism treatment until the a g e of 7'/2. At that time he b e g a n a program of a p p l i e d behav ioura l analysis, p a i d for by his family, until such t ime as they were no longer ab le to afford treatment. This a c t i on also c l a i m e d a b r e a c h of s. 15 of the Charter a n d sought re imbursement for all past a n d future treatment. This c a s e never p r o c e e d e d to trial as the Auton j udgment from the Supreme Court of C a n a d a de te rmined the. inevitable o u t c o m e of any relevant cases in progress. Instead, the plaintiffs sought a n order from the Supreme Court of BC for spec ia l costs arising from the proceed ings . In response, the de fendants o p p o s e d the plaintiff's app l i c a t i on a n d in turn sought an order to c o v e r their own costs. Some of the issues d e b a t e d i n c l uded : (a) whether the de fendants were cons ide red the successful party in light of the Supreme Court of C a n a d a ruling a n d as such, their costs wou ld b e c o v e r e d ; (b) whether the plaintiff's dec is ion to p r o c e e d with the trial before the Supreme Court of C a n a d a judgment in Auton wou ld b e cons ide red a waste of time a n d money ; a n d (c) whether the plaintiff was cons idered a "pub l i c interest litigant" (a party that has no personal o u t c o m e in a trial) (para. 16). Reasons The Supreme Court of BC found that the plaintiffs were entit led to their costs, while the de fendants were d e n i e d theirs. The de fendants subsequently a p p e a l e d to the Court of A p p e a l for BC (2005). R. v. Wynberg (Ontario Superior Court of Justice, 2005) Wynberg. R. et al v. H.M.Q. in Right of Ontario and Deskin. M. et al v. H.M.Q. in Right of Ontario (Court of Appeal for Ontario, 2006) Facts and Issues This litigation represented 35 chi ldren from 30 families w h o were d i a g n o s e d with autism. It was b a s e d on ac ces s to early intervention therapy after the a g e of 6 as a right under the Ontar io Education Act. The Ontar io Superior Court hea rd the plaintiffs' a l legat ions that the Province's Intensive Early Intervention Program (IEIP), wh i ch p rov ided or f u n d e d treatment for chi ldren ages 2 to 5, v io la ted s. 15 of the Charter as it d iscr iminated against the chi ldren on the basis of a g e . They a r g u e d that their s. 7 rights were also v io la ted. The plaintiffs were not cha l l eng ing the constitutionality of any legislation or regulations; instead, they were cha l l eng ing the government ' s act ions or inactions relevant to the design a n d implementa t ion of the IEIP. In add i t ion , o n e of the families (Deskin) also filed a neg l i gence c l a im. 45 The government ' s response was that the a g e restrictions were b a s e d on f inancia l constraints a n d were in no w a y i n t ended to b e discriminatory. Reasons Justice Kiteley ruled that the government h a d v io la ted s. 15(1) of the Charter on the basis of a g e a n d disability a n d that this violation was not justified under s. 1. In add i t ion , the Minister of Educa t ion was found to h a v e v io la ted his duty under s. 8(3) of the Education Act by failing or refusing to ensure the appropr ia te programs for school a g e chi ldren under s. 15 of the Charter. The Court d e n o u n c e d the same violat ion under s. 7 of the Charter a n d also d e n i e d the neg l i gence c l a im by the Deskin family. Justice Kiteley also found that "IBI/ABA [was] not an 'emergent ' therapy or t reatment" (para . 27); this is particularly interesting in light of the earlier Supreme Court of C a n a d a ' s assertion to the contrary [Auton, 2004). As a remedy , Justice Kiteley o rde red a dec l a r a t i on of violat ion of the Charter in add i t ion to f inanc ia l c o m p e n s a t i o n to the petitioners for past a n d future IBI/ABA treatment. The gove rnment of Ontar io a p p e a l e d the dec is ion (T iedemann, 2005b). O n July 7 2006, the Court of A p p e a l for Ontar io struck d o w n the lower Court 's dec is ion a n d unanimously ruled that the Ontar io IEIP was not discriminatory a n d therefore d id not con t r avene the Charter of Rights and Freedoms. In response to the c l a im of a g e discrimination, the Court f o u n d that a l though autistic ch i ldren a g e six a n d older h a d b e e n t rea ted differently b a s e d on a g e , no e v i d e n c e was presented to establish that this differential t reatment const i tuted discr iminat ion. Pertaining to the c h a r g e of disability discrimination, a g a i n , the Court of A p p e a l d i sagreed with the lower Court 's f inding of violation of the c la imants ' equal i ty rights on the basis of disability. The Court of A p p e a l c o n c l u d e d that the c la imants h a d fa i led to demonst ra te that their clients h a d rece i ved differential t reatment by not rece iv ing the IEIP. The Court of A p p e a l obse r ved that it h a d not b e e n demons t ra ted that a p rog ram of this type c o u l d e ven b e adminis tered in a school setting (due to its intensity), nor d id the c la imants proffer e v i d e n c e of the effect iveness of the existing programs for autistic ch i ldren or for ch i ldren with other disabilities. The A p p e a l Court 's reasons be ing that this add i t iona l e v i d e n c e wou ld have b e e n essential to establish that the IEIP was the only ef fect ive alternat ive a n d that by deny ing it wou ld cu lminate in discrimination. The Court of A p p e a l also revoked the d a m a g e s o rdered by the lower Court o n the basis that in genera l , d a m a g e s are not a f f o rded w h e n dec la ra tory relief is p rov ided . In add i t ion , they d id not find lega l g round for the s. 7 c l a im dec l a r ing that since the IEIP was not shown to b e the only ef fect ive p rog r am for autistic chi ldren, there was no fac tua l basis to c o n c l u d e that it was fundamenta l to their pe rsonhood a n d deve l opmen t . The lawyer for the plaintiff was seeking l eave to a p p e a l to the Supreme Court of C a n a d a . Of part icular interest in the ruling of the Ontar io Court of A p p e a l was its s. 1 analysis (even though a f inding of discrimination was not m a d e ) . The Court of A p p e a l d i sag reed with the trial j udge by stating that the a g e cut-off of 6 years was a reasonab le limit under s. 1 of the Charter. 46 The Court acknowledged the government's role in allocating scare resources to where the greater good could be done, and based on the evidence, this was dec ided to be in the autistic population under the age of 6. They reasoned that spreading limited financial and human resources over a larger age cohort would dilute the effectiveness of treatment to children under the age of 6, which it is purported to be the most efficacious "window of opportunity". The Court of Appeal directed lower Courts to consider program objectives and competing social interests when confronted with claims against allegedly under inclusive government programs. They also deferred to the expertise of the government when it comes to making difficult policy choices. Hewko v. Attorney General of BC and School District No. 34 (Abbotsford) et al. (Supreme Court of BC, 2006) Facts and Issues This case was centred on a school district's response to a child's disability and resultant special needs requirements to access a public'education. Darren Hewko was a 9 year-old boy who was diagnosed with autism at the age of 3. He received home-based Lovaas Autism Treatment for two years until he started half-day kindergarten. At this point he continued to receive LAT at home while at kindergarten he was assigned a special education assistant. The school district permitted the Lovaas behavioural therapist into the school in order to transition Darren into the classroom. However, Darren's parents wanted the special education assistant to be a member of the home therapy program. In addition, they requested Lovaas therapy be supplied in the school setting (the school program did incorporate principles of applied behavioural analysis, but not pure Lovaas). The Hewko's request was not granted, and within approximately six months Darren began to regress. The school district recommended that Darren be removed from the integrated classroom and placed in a resource room at an alternate school, however, his parents refused and withdrew him from the public education system (Bakan, personal communication, Nov 22, 2005). The Hewko's sued the Province and the Abbotsford School District seeking relief for alleged discrimination pursuant to s. 15 and s. 7 of the Charter of Rights and Freedoms along with a claim of negligence and/or breach of duties provided for under the School Act. Reasons The Abbotsford School Board was found liable for breach of the Board's statutory duty "to consult with the plaintiffs" (para. 8) while the Court exonerated the Province from liability pursuant to statutory duties. In addition, the plaintiffs failed to establish a breach of either s. 15 or s. 7 of the Charter. In summary, the trial judge held that, "the District failed to carry out its duty to consult by failing to meaningfully consult with the parents" (para. 11) pursuant to the School Act. 47 Legal Reviews of Auton Greschner & Lewis (2003) Greschner a n d Lewis' art icle was o n e of the most c i t ed papers during the Au ton Supreme Court of C a n a d a a p p e a l . Their report e x a m i n e d the cla ims m a d e in the lower Courts ' p roceed ings in add i t ion to critiquing the various componen t s of the judic ia l reasoning process. The authors' rat ionale for cha l l eng ing the o u t c o m e of Auton i n c l uded : (a) its significant f inancia l implicat ions for society, (b) its role as setting p r e c e d e n t for future judic ia l invo lvement in po l i cymak ing , a n d (c) the n e e d to find a p p r o a c h e s for gove rnment that wou ld m a k e their policies incontes tab le under the Charter. Greschner & Lewis' review was b a s e d on the propositions that governments are better suited to determine c o m p l e x po l icy issues rather than the Courts; that hea l th c a r e decisions should b e ev idence-based ; a n d , that Courts c o u l d b e useful in draw ing attent ion to problems with government programs a n d gove rnmenta l dec is ion mak ing . Greschner a n d Lewis's critique b e g a n with a n examinat ion of the c a s e within the contexts of heal th c a r e a n d the Canada Health Act (CHA). S ince Auton was a s c o p e of c o v e r a g e c l a im , w h e n v i e w e d in this perspect ive , it was essentially cha l l eng ing the constitutionality of the C H A itself. To beg in with, Greschner a n d Lewis po in ted out that the principles of the C H A were frequently misused in the Au ton trial. For e x a m p l e , Al lan J. spoke on the primary impo r t ance of the m e d i c a r e legislation be ing the provision of universal hea l th c a r e however , in actual i ty, it was the principle of comprehensiveness that was be ing l i t igated. For examp le , if Lovaas were an insured health service (principle of comprehensiveness) then all qual i f ied provincia l residents that required this treatment wou ld have a c ces s to it (principle of universality). The authors a rgued that the principle of comprehensiveness was never i n tended to cove r every med ica l l y necessary treatment. Next, Greschner a n d Lewis took excep t ion to the lower Courts ' d isregard for the implicat ions of ignoring the C H A criteria for determining insured health services (def ined as hospital services, phys ic ian services a n d surgical-dental services). Lovaas therapy is neither administered by a physic ian nor p rov ided within a hospital setting. As a n ana logy , they po in ted out that, under this d i rect ion, Provinces that d o not have a prescription drug p rogram wou ld in f ac t b e in violat ion of the Charter (since prescription drugs are med ica l l y necessary). Greschner a n d Lewis also a r g u e d that there was a lack of e v i d e n c e before the Courts that suppor ted the conc lus ion of Al lan J . that Lovaas therapy was a hea l th issue (as o p p o s e d to an e d u c a t i o n a l or socia l service issue). They po in ted out ' that the determinat ion of wh i ch c a m p Lovaas fell was integral to the procedures subsequently taken for the constitut ional analysis. They r easoned that, d u e to the " d e e p e r pocke t s " of the hea l th c a r e system, "it w o u l d presumptively b e better for plaintiffs to a p p r o a c h the s. 15 analysis as patients, not as pupils or poor p e o p l e " (p. 519). 48 The quality and type of evidence admitted was also identified as a major concern. Affidavits provided by the petitioners' parents were used to support their claim of treatment efficacy. However, Greschner and Lewis indicated that this anecdotal evidence was questionable due to the fact that these particular children may have improved anyway (due to maturation), or they may have been an atypical representation of autism. The authors also took exception to the Courts discounting what they perceived to be relevant research on the effectiveness of Lovaas. Their argument was that although the Supreme Court of BC ruled that the Court could not specifically order Lovaas therapy (hence, the question of its effectiveness became moot), the Court did order generic EIBI therapy in its place without an evidential foundation of its general effectiveness rate. Then, at the level of the Court of Appeal, the effectiveness evidence on Lovaas became especially relevant since the remedy-granted to the four child petitioners was for Lovaas therapy. The authors also questioned the utility of employing the general acceptance principle in judicial reasoning; the fact that other Provinces had developed IBI programs did not necessarily mean that IBI was effective. Essentially, policy precedence had become a surrogate marker for an effectiveness (or cost-effectiveness) claim. Instead, Greschner and Lewis argued that the Courts should have requested evidence from the government that supported their reasoning behind their program development. They also noted the conspicuous absence of any arguments about the relative effectiveness of either Lovaas or EIBI as compared to other treatment options for autism. And finally, Greschner and Lewis delved into a detailed account of issues pertaining to the constitutional analysis by the Courts. However, as the purpose of this thesis centres on evidentiary issues, the constitutional analysis will not be discussed, with the exception of noting that Greschner and Lewis were highly critical of the lack of attention given to the Oakes analysis of the s.l claim. Selick (2003) In Selick's article for the Fraser Forum (a series of reviews of public policy issues by the Fraser Institute), the author lashed back at the Court of Appeal for BC's decision in Auton (2002) by declaring that the judges failed to consider all the ramifications of their decision. Specifically, Selick spoke to the Court's negligence of not reviewing opportunity costs for such a ruling. She contended that by funding Lovaas the additional money required would either have to be re-directed away from other programs or alternatively, taxes would have to be increased. By re-allocating money, Selick maintained that other faceless citizens would then become the next discriminated group. The author also challenged the Court's rationalization of its parens patriae jurisdiction (as defined in a previous section, this is a duty bestowed on the Courts to protect the rights of those perceived to be disadvantaged, e.g., children). Selick contended that, in its duty to protect 49 a n d a d v a n t a g e chi ldren, the government wou ld inadvertent ly harm other chi ldren w h e n funds were r ea l l o ca t ed and/or taxes raised; she raised the question of whether the Courts should consider those chi ldren as well . Baker & Bhabha (2004) The article by Baker a n d B h a b h a focused o n the definit ion of medically necessary hea l th services a l o n g with the Canada Health Act's (CHA's) pr inciple of universality within the context of the Ontar io heal th c a r e system. The authors initially po in ted out that the term medically necessary h a d not b e e n clearly de f i ned in either the Canada Health Act or in other provinc ia l health insurance legislation. Consequent ly , determining the s c o p e of med ica l l y required services is a pol icy dec is ion as o p p o s e d to a legally enforceable-rule. Another important distinction m a d e was in regard to wha t const i tuted experimental treatment. The authors c i t ed except ions to the list of insured services in the Ontario Health Insurance Plan (OHIP) if a t reatment for a m e d i c a l cond i t ion was general ly a c c e p t e d to b e exper imenta l within Ontar io (even if it is med ica l l y necessary a n d administered by a physic ian in a hospital). In a c a s e that dea l t with c a n c e r treatments, the Province's Health Services A p p e a l a n d Review Board out l ined the fol lowing criteria in order for a treatment to b e a c c e p t e d as non-experimental (p. 27): 1. The p rocedu re is a c c e p t e d as appropr ia te in a number of r e spec t ed a c a d e m i c a n d c l in ica l c a n c e r centres. 2. It is suppor ted by publ ished pee r rev iewed articles in r e spec t ed journals. 3. It is publ ic ly f unded or f u n d e d by private insurers. 4. It is merely an extension of pre-existing a n d widely a c c e p t e d therapeut ic modalit ies. When Ontar io Courts are f a c e d with decis ions to fund procedures that are not listed under OHIP, the fol lowing types of e v i d e n c e are used to determine whether or not the t reatment should be cons ide red exper imenta l (p. 27): 1. A c a d e m i c research not ing the positive effects of the therapy. 2. The extent of the therapy 's usage a n d the length of time it has b e e n p r a c t i c e d . 3. If the p rac t i c e enjoys l imited use, whether this is d u e to the high cost of the therapy a n d not doubts abou t its effect iveness. 4. The number of c l in ica l trials. Baker a n d B h a b h a then p r o c e e d to examine the C H A principle of universality stressing that a l though everyone is c o v e r e d , not everything that is d e e m e d med ica l l y necessary is insured. However, similar to Al lan J . in Auton, the authors a p p r o a c h e d the failure of gove rnment to cove r Lovaas therapy through the lens of universality as o p p o s e d to comprehensiveness stating, " a [health] p l an wh i ch provides necessary services to some, but not to o t h e r s . . . is not only f l awed in respect of the C H A , but also in respect of the Charter" (p. 32). They fa i led to recogn ize that Lovaas was not a n insured hea l th benef i t that was be ing a c c e s s e d by other groups. 50 Svrett (2005) This article c o m p a r e d a n d contrasted Auton with a similar c a s e l i t igated within the English Court system (T. v. Special Needs Educational Tribunal and Wiltshire County Council  ELR 704 [Administrative Court]). Spec i f ic to Auton, the author ident i f ied m a n y of the s a m e issues discussed in previous critiques (confusion over the C H A principles of universality a n d comprehensiveness; lack of attent ion g iven to the e f f i cacy d e b a t e ) . Syrett po in ted out that the lower Court 's re l iance on parenta l affidavits of Lovaas treatment e f f i c acy was a n e x a m p l e of the lega l system inappropr iate ly general iz ing individual ou t comes to a n entire popu la t ion . He p r o p o s e d that the judgments of the lower Courts may h a v e b e e n in f luenced by parenta l pressure a n d that some observers, "wou ld a c co rd ing l y b e less p r e p a r e d to r ega rd this as a legit imate basis for funding the treatment than a dec is ion wh i ch met standards of scientific rationality" (p. 350). A l though Syrett suppor ted the o u t c o m e of the Supreme Court of C a n a d a , he c o n t e n d e d that the e f f i cacy d e b a t e o n c e a g a i n d id not rece ive the attent ion it d e m a n d e d . He c h a l l e n g e d the Court 's line of reasoning in determining the m e d i c a l necessity of Lovaas b a s e d on the f a c t that t reatment h a d only b e e n recent ly f u n d e d in two other Provinces in the past yea r (and h e n c e , de te rmined that it was still in the emergen t phase) . Syrett s tated, " a gove rnmenta l dec is ion to fund or refuse a part icular t reatment may b e taken for a number of reasons wh i ch are u n c o n n e c t e d to the latter's e f f i c acy " (p. 350). He con t i nued by saying that the lower Courts put too m u c h weight on the e v i d e n c e p rov ided by those that h a d vested interests in the o u t c o m e of the trial (the parents), while at the level of the Supreme Court of C a n a d a , the focus a p p e a r e d to shift to the possible cost c o n s e q u e n c e s of ruling for a constitutional entit lement to Lovaas therapy. A n d in conc lus ion, Syrett hypothes ized that parents of autistic chi ldren wou ld have a c c e p t e d the ruling of the Supreme Court of C a n a d a better if the c a s e h a d b e e n f r amed within a cl inical a n d cost-effectiveness context as o p p o s e d to a compar i son of services b e t w e e n Provinces [general acceptance principle). 51 Objectives The object ives of this thesis are, (a) to examine the role of e v i d e n c e within the contexts of law, heal th po l icy a n d health ca re , a n d (b) to determine the current state of k n o w l e d g e in regard to the effect iveness claims of Lovaas therapy for autism spect rum disorder. Rationale The Auton c a s e d rew attent ion to the content ious interplay b e t w e e n law, heal th pol icy, a n d heal th c a r e in regard to the Courts ' invo lvement in health pol icy decis ions. This p l a y e d out through the process of judic ia l po l i cymak ing . While originally c a l l e d u p o n to d e c i d e cases of m e d i c a l ma lp rac t i c e , the Court 's role has e vo l v ed to inc lude issues regard ing the regulat ion of m e d i c a l p rac t i ce , phys ic ian supply m a n a g e m e n t , hospital restructuring, a n d ac ces s to speci f ic treatments (as was the c a s e in Auton) . Proponents of judic ia l po l i c ymak ing suggest that the process protects minority rights, the promot ion of h u m a n e condit ions in institutions, restricts bureaucra t i c arbitrariness a n d in genera l , promotes positive soc ia l c h a n g e (Anderson, 1992). O n the other h a n d , critics c i te limitations inherent in the lega l system such as: (a) judges ' a n d lawyers' l ack of e d u c a t i o n a l b a c k g r o u n d or expertise to hand le c o m p l e x po l icy issues; (b) the nature of the adversarial system whereby information c a n b e wi thheld ; (c) the Courts ' re l iance on p r e c e d e n t cases to d e c i d e future p roceed ings ; (d) the limited focus of a judic ia l review process; a n d (e) the Courts ' inability to foresee the long-term c o n s e q u e n c e s of their decisions (Anderson, 1992). Manf red i a n d Maion i (2002) concu r with Anderson in regard to the adversarial structure of ad jud i ca t ion imped ing comprehens i ve information gather ing . However, they also c o n t e n d that within the context of rights discourse, judic ia l rev iew narrows the range of pol icy alternatives a n d systematical ly favours nat iona l norms a n d standards. In law, ad jud ica t i ve facts prov ide the basis of determin ing historical cause-and-effect relationships; however , many a c a d e m i c s c o n t e n d that Courts are ill e q u i p p e d to ext rapo la te the process to present d a y p h e n o m e n a . For examp le , in Auton, the systematic rev iew a n d crit ical appra isa l of the e v i d e n c e c o n c l u d e d that there was a lack of scientif ic rigor to support the effect iveness c l a im for LAT (Bassett et a l , 2000). However, despite these findings, five judges (two trial judges a n d three a p p e a l judges) p r o c e e d e d to a w a r d Lovaas t reatment to the petitioners. However, of part icular interest to this thesis is the reversal of the lower Courts ' decisions by the Supreme Court of C a n a d a . How is it the same e v i d e n c e p r o d u c e d such a different o u t c o m e ? What was go ing on in the lower Courts that y i e lded findings of fac t (formal lega l truth) that d e v i a t e d from the ac tua l truth (substantive truth)? These a re important questions as the ramifications of a judic ia l po l i cymak ing process c a n extend far b e y o n d the original c a s e at h a n d by redirect ing publ ic money to speci f ic stakeholders. O n e of the key issues in a judic ia l review is that the domains of law, pol icy a n d heal th c a r e differ in r ega rd to their respect ive wor ld views, conceptua l iza t ions , interpretations a n d use 52 of e v i d e n c e . This p h e n o m e n o n has b e e n the focus of d e b a t e (Eisenberg, 2001; Havighurst et a l . 2001) a n d g a i n e d m o m e n t u m as Auton a d v a n c e d to the Supreme Court of C a n a d a . This thesis was speci f ica l ly des i gned to invest igate the conceptua l iza t ions a n d processes used by e a c h of the doma ins (law, hea l th pol icy a n d hea l th care) in order to ga in a n unders tanding of h o w e a c h seeks, understands a n d appl ies e v i dence . This is a c c o m p l i s h e d in two parts. First, the lega l dimension of the heal th t e chno logy assessment (HTA) framework utilizes a qual i tat ive g r o u n d e d theory me thodo logy to examine part ic ipant interviews a n d lega l documents . This provides early conceptua l iza t ions of e a c h sector 's perspect ives on e v i d e n c e a long with provid ing HTA researchers with a n a p p r o a c h to investigate the soc ia l (legal) context . A n d s e c o n d , the effectiveness dimension of the HTA framework employs the methods of systematic review a n d crit ical appra isa l to investigate the ongo ing Lovaas Autism Treatment effect iveness d e b a t e . Together, these two componen t s prov ide insight into h o w the Lovaas effect iveness e v i d e n c e was h a n d l e d in the Auton c ase , a n d h o w e a c h sector conceptua l i zes the evident iary process in genera l . The goals of this research are : 1. To b e ab l e to prov ide pol icymakers with a current, comprehens i ve b o d y of e v i d e n c e o n Lovaas Autism Treatment to a i d in the d e v e l o p m e n t a n d eva lua t ion of provinc ia l EIBI t reatment programs. 2. To b e a b l e to prov ide lawyers with a current, comprehens i ve b o d y of e v i d e n c e on Lovaas Autism Treatment for potent ia l future lit igation. 3. To c r ea t e a better understanding within a n d b e t w e e n the domains of law, heal th po l icy a n d hea l th c a r e in regard to h o w e a c h concep tua l i za t ion seeks, understands a n d appl ies e v i d e n c e . Research Questions This thesis sets out to answer the fol lowing questions: 1. How d o the sectors of law, health pol icy a n d health c a r e seek, unders tand, a n d a p p l y e v i d e n c e ? 2. What is the current state of know ledge in regard to the effect iveness claims of Lovaas therapy? C h a p t e r 1 has p rov ided a n introduct ion to Autism Spect rum Disorder a n d Early Intensive Behavioural Intervention. It e x a m i n e d o n e speci f ic form of a p p l i e d behav ioura l analysis -Lovaas Autism Treatment - a n d highl ighted the o n g o i n g effect iveness d e b a t e . The c h a p t e r p r o c e e d e d to situate this intervention within the l oca l heal th pol icy context of the British C o l u m b i a n autism intervention program a long with the lega l context of the Charter c ha l l enge to access this therapy [Auton et al.). 53 C h a p t e r 2 goes o n to identify a n d summarize the re levant multi-disciplinary c o n c e p t u a l perspect ives on e v i d e n c e u n c o v e r e d in the literature a long with the oppos ing conceptua l iza t ions that a c c o m p a n y e a c h view. Processes for seeking, unders tanding a n d app ly ing e v i d e n c e in e a c h of the lega l , heal th pol icy a n d heal th c a r e domains are d iscussed. From this literature review, preliminary c o n c e p t u a l maps a n d frameworks are d e v e l o p e d to prov ide a n ana ly t i ca l lens to examine the role of e v i d e n c e within the lega l contex t of the Heal th Techno logy Assessment framework. C h a p t e r 3 provides an in-depth descr ipt ion of the research methodo log ies utilized in this thesis: (a) g r o u n d e d theory, (b) systematic review, a n d (c) crit ical appra isa l . It also provides b a c k g r o u n d information on the qual i tat ive research software p rog ram - QSR N6 - that was used to assist in the analysis of the interview d a t a a n d lega l documents . In sum, this c h a p t e r outlines the processes a n d rat ionale beh ind the design, c o n d u c t a n d analysis of this thesis. C h a p t e r 4 presents the results of the analyses of both the interviews a n d the l ega l documents , a l o n g with the ou t comes of the crit ical appra isa l of the Lovaas e v i d e n c e . The qual i tat ive d a t a findings from the interviews a n d l ega l documents are f r a me d within the context of the c o n c e p t u a l maps d e v e l o p e d for this thesis. The findings from the systematic rev iew a n d cr it ical appra isa l of the Lovaas effect iveness e v i d e n c e are a p p l i e d to ob jec t i ve criteria to discern quality a n d validity. A n d finally, C h a p t e r 5 provides a discussion a n d conc lus ion to the analyses of the role p l a y e d by e v i d e n c e in lega l , health po l icy a n d heal th c a r e contexts a l o n g with the findings from the cr i t ical appra isa l of the effect iveness e v i d e n c e on Lovaas Autism Treatment. In add i t ion , this chap t e r highlights how lega l issues are identif ied a n d ana l yzed within a hea l th t e chno logy assessment framework. The chap te r also draws conclusions on the current state of know ledge a n d strength of the b o d y of e v i d e n c e in rega rd to Lovaas Autism Treatment. A n d finally, this chap t e r c o n c l u d e s by proffering suggestions for further research. 54 CHAPTER 2: CONCEPTUAL FRAMEWORK Understanding the Role of Evidence in Law, Health Policy and Health Care Increasingly, law is relying on scientific e v i d e n c e to inform its fact-finding process. This has o c cu r r ed d u e to the t echno logy explosion, adop t i on of s tandard ized rules of e v i d e n c e , a n d the increasingly more c o m m o n litigious a p p r o a c h to p rob lem resolution. While intuitively inviting s c i ence into the ev ident ia l forum seems like a natural course, the process has not b e e n without its problems a n d controversies. As desc r i bed by Kaye (1997), l aw has "harnessed the powerful eng ine of scientific investigation to procedures c ra f ted in the days of the horse-drawn car t " (p. xvii). In order to examine the role e v i d e n c e p l a yed within the three contexts of law, health po l icy a n d hea l th c a r e in the Auton a n d Anderson l ega l p roceed ings , I d e v e l o p e d a c o n c e p t u a l f ramework that pul led from relevant c o n c e p t s within e a c h of these three domains . These c o n c e p t s were se l ec ted b a s e d on their ability to provide insight into the processes used by e a c h sector to seek, understand a n d apply e v i dence . This c h a p t e r descr ibes these concep t s , presents the current discourse on e a c h topic , a n d then proffers a c o n c e p t u a l f ramework utilized for the analysis in this thesis. The Concept of Evidence The c o n c e p t of evidence has different meanings a n d standards d e p e n d i n g u p o n wh i ch disciplinary c a m p one is from. From the health c a r e perspect ive e v i d e n c e is b a s e d on empir ica l observations of real events (Eddy, 2001) . Ideally this plays out in the form of r andomized contro l led trials. However, w h e n this level of e v idence-based d a t a are not ava i l ab le , a l ternate processes such as pee r rev iew a n d consensus d e v e l o p m e n t are e m p l o y e d to gather , assess a n d disseminate other types of e v i d e n c e (to b e discussed in a later section). However , the utility of e a c h strategy c o m e s with its o w n un ique strengths a n d limitations. In the pol icy context , the c o n c e p t of e v i d e n c e has b e e n classified as either colloquial or scientific (Lomas, Culyer, M c C u t c h e o n , M c A u l e y & Law, 2005) . Colloquial evidence is the non-scientific contextua l side of a pol icy, "anyth ing that establishes a fac t or gives reason for bel iev ing someth ing " (p. 3). Alternatively, scientific evidence is seen as, " information g e n e r a t e d through a prescr ibed set of processes a n d procedures r ecogn ized as scientif ic" (p. 8); it is know ledge that is explicit (codif ied a n d propositional); systematic (utilizes transparent a n d explicit methods) ; a n d rep l i cab le (utilizing the same methods on the same samp le will p r o d u c e the same results). This systematic review identif ied differences b e t w e e n these two types of e v i d e n c e ; the first was in regard to its usage . The authors desc r ibed colloquial evidence as be ing incorpora ted into pol icy decisions o n c e decision-makers a g r e e d u p o n its r e l evance , while the inclusion of scientific evidence was methodo log ica l l y de te rmined . The authors also identi f ied two different views pertaining to the role of scientific e v i dence . The first perspect ive , wh i ch has 55 its genesis within e v idence-based med i c ine , was that s c i ence reveals "universal truths" (p. 10) b a s e d on the assumption that its role is i ndependen t of context [context-free evidence). This v iew addresses the quest ion " c a n it work? " a n d "provides a gl impse of wha t might b e a c h i e v e d under idea l c i r cumstances " (p. 10). The s e c o n d point of v iew proffered by the researchers or ig inated from the socia l sc iences a n d holds as its conv ic t ion the bel ief that, " e v i d e n c e has little m e a n i n g or impor t ance for dec is ion mak ing unless it is a d a p t e d to the c i rcumstances of its a p p l i c a t i o n " (p.l 1). This type of e v i d e n c e seeks to answer the questions "will it work? " a n d "is it worth it?" a n d is te rmed context-sensitive evidence. The authors c o n c l u d e d that: To c r e a t e context-sensitive g u i d a n c e [to policy-makers], context-free s c i e n c e needs to b e in tegra ted into the s c i ence on " l o c a l " variables such as publ ic attitudes, pat ient preferences , professional proclivity, manager i a l c a p a c i t y , e c o n o m i c feasibility, g e o g r a p h i c loca t ion , a n d so on , in search for consensus a round what might b e a c h i e v a b l e rather than wha t might b e a universal c l inical truth, (p. 11) Lomas et a l . (2005) also summarized the different categor ies or dimensions of e v i d e n c e u n c o v e r e d in the literature. These consisted of three genera l a p p r o a c h e s to def in ing e v i d e n c e : (a) by m e t h o d of co l l ec t ion , (b) by general-purpose, a n d (c) by source. Method of collection inc ludes e v i d e n c e c o m p i l e d through research designs such as exper imenta l or observat iona l ; the general-purpose a p p r o a c h takes into considerat ion the reason for its app l i c a t i on , such as prob l em ident i f icat ion as o p p o s e d to measur ing effectiveness; while, source differentiates research from co l loquia l forms of e v i d e n c e . The authors g o on to descr ibe six dimensions of context to b e cons ide red a longs ide the co re c o n c e p t s of health ou t comes a n d appropr iateness. These contextua l dimensions inc lude : (a) imp lementa t ion e v i d e n c e , (b) organizat ional e v i dence , (c) att itudinal e v i d e n c e , (d) forecast e v i d e n c e , (e) economic/ f i nanc i a l e v i dence , a n d (f) ethics e v i d e n c e . While the co re c o n c e p t s are o b t a i n e d through exper imenta l , quasi-experimental , or observat ional methods , their literature rev iew suppor ted the a rgument that existing research methods c o u l d b e easily e m p l o y e d to obta in add i t iona l information on e a c h of the six contextua l ca tegor ies . A n d finally, understandably , evidence is the most concep tua l l y d e v e l o p e d within the discipl ine of law. Evidence is broadly de f ined as, "al l the means by wh i ch any a l l e g e d matter of fact , the truth of wh i ch is submitted to investigation at judicial trial, is establ ished or d i sp roved " (Gifis, 2003, p.182). Evidence c a n b e de f ined by its source [expert evidence); t ype [hearsay evidence); d e g r e e of proof [conclusive evidence); process for obta in ing [illegally obtained evidence); quality [incompetent evidence); a n d weight [preponderance of evidence), to n a m e but a few. Law schools offer numerous courses d e d i c a t e d strictly to this c o n c e p t while lega l libraries are s t a cked with authoritative texts on the subject. However , some policy-makers c o n t e n d that l ega l definitions of e v i d e n c e were not very helpful w h e n it c o m e s to gu id ing po l icy decisions (Lomas, et al . , 2005). 56 Opposing Conceptualizations of Evidence It was inevi table that w h e n the domains of pol icy a n d health c a r e intersected within the judic ia l pol icy mak ing context of the Auton a n d Anderson trials that e a c h sector's di f ferences in regard to their concep tua l i za t ion of evidence wou ld b e c o m e apparen t . Policy's c o n f i d e n c e in their commiss ioned health t e chno logy assessment of the e v i d e n c e on Lovaas was usurped by the appa r en t we ight p l a c e d u p o n the parents' affidavits by the Courts. Lega l rules of e v i d e n c e , tools for constitutional analysis, a n d c o n c e p t s such as the burden of proof cont ras ted sharply against processes e m p l o y e d by the pol icy a n d heal th c a r e sectors to support their e v i d e n c e in the Lovaas e f f i c acy d e b a t e . Conceptual Background for Thesis In April 2000, the Amer i c an Institute of Med i c i ne (IOM) a n d the A g e n c y for Hea l thcare Research a n d Qual i ty (AHRQ) s p e a r h e a d e d a d e b a t e c a l l e d "Evidence: Its Meanings in Health Care and in Law" (Havighurst, et al . , 2001; Peterson, 2001). This forum brought together l ead ing experts in the fields of law, ep idemio logy , hea l th services research, a n d hea l th p l an administration to explore the differing professional constructs of e v i d e n c e . The impetus for the workshop was the p a r a d i g m shift in hea l th c a r e towards e v i dence-based m e d i c i n e a n d the o u t c o m e of a p r e c e d e n t U.S. Supreme Court c a s e that saw judges assuming a greater role in screen ing admissible expert e v i d e n c e [Daubert v. Merrell Dow Pharmaceuticals, 1993). The workshop part ic ipants shared a c o m m o n c o n c e r n that, " lega l uses a n d interpretations of sc i ence-based m e d i c a l e v i d e n c e . . . m a y d iverge substantially from the uses a n d interpretation of that e v i d e n c e by the m e d i c a l a n d health c a r e researchers w h o p r o d u c e it a n d of the practitioners a n d health plans that use it in mak ing c l in ical decisions a n d pol ic ies" (Havighurst, et al . , 2001, p. 195). Through a series of reports commiss ioned before the workshop, in add i t ion to a c a d e m i c deba tes throughout the sessions, the IOM/AHRQ part ic ipants d i scovered that oppos ing constructs of e v i d e n c e existed amongst various professional bodies . A l though the part ic ipants were unab le to reconc i l e their differences, six spec i f ic c o n c e p t u a l areas of discourse were ident i f ied: (a) popu la t ion probabil it ies vs. individual causat ion , (b) pre h o c vs. post h o c e v i d e n c e , (c) c l in ical progress vs. lega l reform, (d) ad jud ica t ion of di f ferences, (e) rules of e v i d e n c e ; a n d , (f) decision-maker. The workshop c o n c l u d e d , " open-minded discussions a n d thoughtful c o n c e p t u a l a n d empir ica l analyses c a n br idge our fields a n d identify areas of m iscommun ica t i on . " (Eisenberg, 2001, p. 380). The commiss ioned reports, a l o n g with the findings from this workshop, were publ ished col lect ively as a series of peer-reviewed papers in a spec ia l feature in Journal of Health Politics, Policy and Law (Vol. 26, No. ). For ease of in-text referenc ing, this b o d y of know ledge will b e c i ted as Eisenberg (2001). 57 Population Probabilities vs. Individual Causation To beg in with, the IOM/AHRQ forum dete rmined that law, pol icy a n d heal th c a r e differed in their concep tua l i za t ion of the patient/cl ient by taking either a popu la t ion or individual perspect ive . In tort l aw (neg l igence litigation), the Court seeks to de te rmine whether harm has c o m e to a n individual by w a y of a n ac t ion that has a l ready o c c u r r e d , or whether the individual was d e n i e d a potent ia l benefit from a n ac t ion that was not car r ied out. Eisenberg (2001) c o i n e d this v iewpoint, " e v i d e n c e of the ins tance " (p. 375) as it focuses on the tenets of individual rights, wrongs, a n d harms a n d utilizes e v i d e n c e in a n effort to determine causa t ion under part icular c i rcumstances (However, in Auton the lower Courts were assuming a pol icy role, as o p p o s e d to the familiar " indiv idual causa t ion " perspect ive) . A n e x a m p l e of law's individualist perspect ive c a n b e found in a n A m e r i c a n c a s e in wh i ch expert e v i d e n c e perta ining to the use of fac i l i ta ted c o m m u n i c a t i o n (FC) to quest ion a n autistic ch i ld in Court was d e e m e d inadmissible (Cande lo ra , 1995). ("Faci l i tated c o m m u n i c a t i o n involves hand-over-hand or hand-on-forearm support of students by a f a c i l i t a t o r . . . to offer physical support while typing on a k e y b o a r d " [p. 753]). Because the Court v i e w e d FC as scientific, it was subject to the same admissibility criterion of reliability. Critics felt that, "by requiring genera l quant i tat ive testing [of FC], the C o u r t s . . . shifted the focus from a n individual's ' [italics a d d e d ] ability to c o m m u n i c a t e to the ability of a class of p e o p l e to c o m m u n i c a t e " (p. 99). Within the po l icy forum, decision-makers are c o m p e l l e d to discern the fairest w a y of equi tab ly distributing finite resources amongst a population of individuals. Within this popu la t ion perspect ive , many c o m p e t i n g needs must b e w e i g h e d through various dec is ion mak ing processes. A n d , in regard to health ca re , in response to the e v i d e n c e d - b a s e d med i c i ne movement , physicians are increasingly relying on populat ion-level statistics b a s e d on probabil it ies in order to m a k e c l in ica l decisions; this equates to " e v i d e n c e of the genera l i zab le " (Eisenberg, 2001, p. 375). Pre Hoc vs. Post Hoc Evidence Another differing perspect ive that arose was how the different disciplines c o n c e p t u a l i z e d e v i d e n c e in regard to the timing of utilization of e v i d e n c e relative to the o u t c o m e . In tort law, e v i d e n c e is used post-hoc to discern whether an o u t c o m e that has occu r r ed in the past was c a u s e d by a n appropr ia te , harmful or inappropr ia te a c t i on . It is used to " judge responsibility a n d render just ice" (Eisenberg, 2001, p. 376). While in Auton, e v i d e n c e involved pre hoc assessments by the Courts. In pol icy forums, e v i d e n c e c a n b e used pre hoc to inform decision-makers on populat ion-level issues prior to their ou t comes . In heal th ca re , e v i d e n c e is also general ly used in a pre h o c probabil ist ic fashion to support a c l in ica l dec is ion 58 prior to a pat ient o u t c o m e (however, Eisenberg no ted that med i c i ne often reflects b a c k on past decisions to learn from its mistakes [e.g., morbidity a n d mortality rounds]). Clinical Progress vs. Legal Reform The w a y in wh i ch c h a n g e occurs in law a n d health c a r e also impac ts the concep tua l i za t ion of e v i dence . In law, Courts, legislatures, or regulators perform the c h a n g e agen t role, while in health ca re , c h a n g e m a y o c c u r w h e n new scientific e v i d e n c e is a d o p t e d by key op in ion leaders a n d disseminated to other practit ioners. Within the po l i cy context , c h a n g e occurs through processes of pol icy d e b a t e . Adjudication of Differences Another identif ied d i f ference was the w a y in wh i ch experts were used in e a c h profession to ad jud i ca t e dif ferences. The judic ia l system operates within two different sets of ideals a b o u t the c o n d u c t of fact-finding proceed ings . The first is the tradit ional adversarial system a n d the s e c o n d is the g a t e k e e p e r a p p r o a c h (Shuman, 2001). These two processes represent different methods to address the c o r e values inherent in the judic ia l system: a c c u r a c y , fairness, ef f ic iency, consistency, a n d accessibi l i ty. Which a p p r o a c h is used often d e p e n d s on the type of c a s e (e.g., criminal vs. civil) a n d the e v i d e n c e at issue. A n understanding of these two different perspect ives is important in order to discern how law assesses expertise a n d h o w e v i d e n c e is admi t t ed into Court . In the adversarial approach, lawyers d e c i d e on what e v i d e n c e to submit a n d judges d e c i d e wh i ch e v i d e n c e c a n b e presented to a n d ultimately admi t t ed by the Court . The adversar ia l process assigns responsibility to the individual parties ( and not the judge) for identifying.relevant issues, presenting proof, a n d c h a m p i o n i n g the c a s e through the judic ia l system (Brooks, 1991). E ach party is responsible for f inding a n d submitting e v i d e n c e in addi t ion to f inding limitations or weaknesses in the opposit ion's c a se . The underlying assumption is that truth is more likely to b e u n c o v e r e d by a contest b e t w e e n those that ho ld the greatest vested interest in the o u t c o m e . Critics of the adversarial system a rgue that both lay a n d expert witnesses c a l l e d by e a c h respect ive party will favour their attorney-employer 's perspect ive . Historically, admissibility of expert opinion was b a s e d on a n assessment of the expert 's qual i f icat ions leav ing an eva luat ion of the reliability a n d weight of their p resented e v i d e n c e up to the trier of fact . However , b e c a u s e the individual parties control the e v i d e n c e , there is a risk that the trier of f ac t may b e exposed to a b iased samp le of experts. It has b e e n p roposed that one alternat ive for civil cases wou ld b e Cour t-appo inted experts, a process similar to a journal editor request ing experts to review a manuscript. In theory, this wou ld prov ide impartiality with judgments b a s e d on expertise rather than a d v o c a c y (Eisenberg, 2001). However , this suggestion has b e e n met with m u c h controversy a n d as a result, is rarely used . In add i t ion , judges may also b e unab le or unwilling to w e e d out poor experts, as there exists a large poo l of "professional experts" more than willing to m a k e a ca ree r out of testifying. 59 The gatekeeper model arose from the o u t c o m e of three p r e c e d e n t A m e r i c a n cases : Daubert v. Merrell Dow Pharmaceuticals (1993); General Electric Co. v. Joiner (1997); a n d , Kumho Tire Co. v. Carmichael (1999); within a C a n a d i a n context , the m o d e l is rooted in R. v. Mohan (1994). A l though this a p p r o a c h to ad jud ica t ion will b e desc r ibed in more deta i l in a subsequent sect ion, suffice it to say at this point that the o u t c o m e of these cases resulted in judges assuming a far greater role in ensuring that e v i d e n c e presented to the Court was rel iable a n d re levant (Eisenberg, 2001). In the health c a r e context , scientific e v i d e n c e is general ly in tegrated o n c e it has b e e n through a peer review process; one such m e t h o d is publ icat ion in a peer-reviewed journal. Here, potent ia l manuscripts are distributed amongst substantive a n d me thodo log i c a l experts for crit ique with the final dec is ion on their worthiness for publ i ca t ion residing with the journal's editor. In this scenar io editors b e c o m e the ad jud ica tor of e v i d e n c e for their profession. The IOM/AHRQ workshop d id not report on the c o n c e p t of ad jud ica tor for the pol icy context , however , this writer is hypothesizing this role to b e assumed by bureaucrats a n d looks to discover this in the d a t a . Rules of Evidence O n e of the main focuses of the IOM/AHRQ c o n f e r e n c e was the c h a n g e s to evident iary rules brought abou t by the three p r e c e d e n t Amer i c an cases. A l though Daubert cont r ibuted to the cod i f i ca t ion of the rules of e v i d e n c e , subsequent cases such as Kumho Tire Co. v. Carmichael demons t r a t ed that judges h a d flexibility in app ly ing the test of reliability to the e v i d e n c e a n d substantial discretion in determining wh i ch e v i d e n c e to admit . Despite great expecta t ions of raising the evidentiary bar for admissibility of m e d i c a l e v i d e n c e , the i m p a c t of Daubert (and Mohan in the C a n a d i a n context) was not consistent across the Courts. In fact , lega l rules of e v i d e n c e are in constant flux. Within heal th ca re , rules of e v i d e n c e focus on the crit ical appra isa l of the ava i l ab le scientific e v i d e n c e . This c a n o c c u r in a c o u p l e of different ways. At the level of the individual physic ian, this wou ld involve an eva luat ion of a part icular article in respect to its r e l e vance to a patient 's spec i f ic c l inical p rob lem. Ideally, the article wou ld b e a systematic review of a number of wel l-designed randomized contro l led trials (RCTs), wh i ch also i nco rpora ted a n assessment of the level or quality of evidence. A review of the level of e v i d e n c e is a process that allows for the qual i tat ive ca tegor iza t ion of research e v i d e n c e in add i t ion to def in ing spec i f ic g rades of recommendation, while a review of a study's quality reveals the measures researchers have taken in their des ign, c o n d u c t , a n d analysis in order to limit bias (to b e discussed in more detai l in the Methodology c hap t e r of this research). It is r e cogn i zed that this systematic process of apprais ing research does not a lways o c c u r in a c tua l m e d i c a l p rac t i ce . Some m e d i c a l specialt ies, such as surgery, m a y not b e c o n d u c i v e to eva luat ion by RCTs (unless the principle of equipoise exists b e t w e e n two surgical procedures ) . 60 Other reasons for the difficulties in des igning surgical RCTs inc lude select ion a n d observer bias, bl inding, learning curve, effectiveness versus e f f i cacy a n d standardizat ion of t e chn ique (Meakins, 2002). Thus, in these situations a weake r study des ign (such as a n observat ional study) may b e used. However , e ven for research questions that are c o n d u c i v e to be ing e x a m i n e d with a n RCT des ign, factors such as costs, length of follow-up, a n d eth ica l issues surrounding randomizat ion , m a y often prevent it. A l though peer-reviewed e v i d e n c e in med i c ine is held in high regard , there does not prevai l c o m m o n criteria ("rules of ev idence " ) u p o n wh i ch to eva lua te its quality a n d strength. Instead, multiple frameworks exist u p o n wh i ch c l in ical p rac t i ce guidel ines are often d e v e l o p e d . O n e e x a m p l e is the me thodo logy d e v e l o p e d by the C a n a d i a n Task Force on the Periodic Health Examinat ion in 1979 for determining strength of e v i d e n c e a n d levels of r e c o m m e n d a t i o n (1994). This me thodo logy was subsequently a d a p t e d by the U.S. Preventive Services Task Force (Harris, et al . , 2001) a n d since its incept ion the two countries have c o l l a b o r a t e d on their r e commenda t i ons . The IOM/AHRQ workshop d id not report on the c o n c e p t of rules of e v i d e n c e within a pol icy context . Decision-maker While both l aw a n d med i c ine va lue the credentia ls of experts, some dif ferences exist in regard to the source of the e v i d e n c e a n d w h o dec ides on its validity; essentially, w h o should determine wh i ch e v i d e n c e should b e fo l lowed? In law, the source of e v i d e n c e is often times a primary source (e.g., the author of a research pape r , or a c red ib le expert in the field). However , the eva luat ion of the strength of the e v i d e n c e c a n rest on the shoulders of a less we l l -educa ted jury (e.g., civil trials in BC) or on a judge 's assessment of the credibility of a n expert witness or their eva luat ion of the scientific merit of a study. Determining credibility of heal th c a r e practitioners c a n b e especia l ly difficult in that, b e y o n d their gener ic l icense to p rac t i ce , there is no formal mechan i sm to eva lua te their qualif ications a n d credentia ls . When c a l l e d u p o n as an expert witness, the practit ioner's assertion of their qual i f icat ions is taken at f a c e va lue ("prima fac ie" ) (Supreme Court of BC, A u g 30, 1993). In med i c ine , those w h o are d e e m e d the most expert j udge the e v i d e n c e . Physicians infrequently will r e ad the primary source of the e v i d e n c e (e.g., the research paper ) ; instead, relying on protocols , dec is ion trees, consensus statements, c l in ical pa thways a n d c l in ica l p rac t i ce guidel ines. Physicians also utilize s econda ry reviews a n d syntheses of research, such as C o c h r a n e reports (The C o c h r a n e Co l l abora t ion is a n international, non-profit organizat ion that p roduces a n d disseminates systematic reviews of health c a r e interventions). A n d , o n c e a g a i n , the IOM/AHRQ workshop d id not specif ical ly identify the dec is ion -maker in heal th pol icy but is assumed to be the bureaucra t and/or e l e c t e d off icial . 61 Definitions of Effectiveness Another differing concep tua l i za t ion that d id not originate from the IOM/AHRQ workshop, but instead b e c a m e a p p a r e n t through this literature review, was the conf l ic t ing definitions of effect iveness. Members of the health c a r e t e a m wou ld v iew a t reatment as ef fect ive if it m a t c h e d the spec i f ic requirements of an individual patient, while health pol icy researchers wou ld v iew effect iveness in light of increas ing heal th utility indicators as demons t r a t ed through r andomized contro l led trials (Biller-Andorno, Lie, & ter Meu len , 2002). In add i t ion , these authors also ex tend this c o n c e p t to inc lude differing conceptua l iza t ions of cost effect iveness arguing that r e commenda t i ons p r o d u c e d from cost effectiveness analyses m a y conf l ic t with physicians' pursuit of opt ima l individual efficiency. These two views also link with the previous population probabilities vs. individual causation d e b a t e . 62 Processes of Seeking, Understanding and Applying Evidence: Legal Context Judicial Policy Making Auton took p l a c e under the jurisdictional context of judicial policy making. This is a process whe reby the Courts ascerta in l ega l jurisdiction over d e c i d i n g a pub l i c po l i cy issue with • the purpose of p roduc ing socially desirable results (Feeley & Rubin, 1998). While not exclusive to rights-based judic ia l review brought abou t by Charter cha l lenges , within the rea lm of heal th c a r e these cases have con tes ted issues such as regulat ion of m e d i c a l p rac t i c e , phys ic ian supply m a n a g e m e n t , hospital restructuring, a n d a c c e s s to spec i f ic treatments. It is within this context o
UBC Theses and Dissertations
A health technology assessment of Lovaas Autism Treatment : the role of evidence in legal, health policy… Beard, Sandra Lynn 2007
Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.
- 831-ubc_2007-266779.pdf [ 16.51MB ]
- JSON: 831-1.0100323.json
- JSON-LD: 831-1.0100323-ld.json
- RDF/XML (Pretty): 831-1.0100323-rdf.xml
- RDF/JSON: 831-1.0100323-rdf.json
- Turtle: 831-1.0100323-turtle.txt
- N-Triples: 831-1.0100323-rdf-ntriples.txt
- Original Record: 831-1.0100323-source.json
- Full Text