UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

An evaluation of a community IV program Dixon, Beverly Lynn 2004

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_2004-0205.pdf [ 5.23MB ]
Metadata
JSON: 831-1.0091332.json
JSON-LD: 831-1.0091332-ld.json
RDF/XML (Pretty): 831-1.0091332-rdf.xml
RDF/JSON: 831-1.0091332-rdf.json
Turtle: 831-1.0091332-turtle.txt
N-Triples: 831-1.0091332-rdf-ntriples.txt
Original Record: 831-1.0091332-source.json
Full Text
831-1.0091332-fulltext.txt
Citation
831-1.0091332.ris

Full Text

A N E V A L U A T I O N OF A C O M M U N I T Y IV P R O G R A M by B E V E R L Y L Y N N DIXON B.S.N., The University of Victoria 1993 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF M A S T E R OF SCIENCE IN NURSING in THE F A C U L T Y OF G R A D U A T E STUDIES School of Nursing We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH C O L U M B I A April 2004 © Beverly Lynn Dixon April 2004 Library Authorization In presenting this thesis in partial fulfillment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Beverly L. Dixon 19/04/2004 Name of Author (please print) Date (dd/mm/yyyy) Title of Thesis: An Evaluation of a Community IV Program Degree: Master of Science in Nursing Department of School of Nursing The University of British Columbia Vancouver, BC Canada Year: 2004 11 Abstract The purpose of this study was to evaluate selected outcomes for the Simon Fraser Health Region Community IV Therapy Program specifically in terms of changes in hospital stay and the frequency of safety concerns. The outcomes will provide information to the decision-makers and stakeholders about the effectiveness of the Program in providing safe care and maximizing access to hospital beds. A Program Logic Model was developed and then retrospectively applied as the Program framework. Two hypotheses examined in this pre-experimental pre-test, post-test design, tested changes in length of hospital stay for patients with osteomyelitis and cellulitis two years before the Program started (April 1, 1999 to March 31, 2000), to two years after the Program was operating (April 1, 2002 to March 31, 2003). Research questions focussed on: 1) readmission rates for patients with osteomyelitis and cellulitis; 2) readmission rates for patients in the Program during the two years of operation (April 1, 2000-March 31, 2003); 3) safety concerns for patients in the Program; and 4) whether an association could be seen between safety concerns and readmission rates. Total sampling was used to collect data from the Canadian Institute of Health Information databases, which resulted in 20 pre-test, and 13 post-test patients with osteomyelitis, and 60 pre-test and 74 post-test patients with cellulitis. The hypotheses were not supported. There was no difference in the discharge pattern for patients in the Program and for patients not in the Program, leaving a question about whether and where the patients who were not admitted into the Program received IV antibiotics. Readmissions within one month of discharge for patients with cellulitis rose from 3.4% in the pre-test year to 10.8% in the post-test year. Over one third of patients entered the Program from the community, thus avoiding a hospital admission. These patients were significantly younger than patients who stayed in hospital for their IV therapy, suggesting that there may be barriers for older patients entering the Program. The readmission rate for patients in the Program was vvdthin the community benchmark. While there were 23 documented concerns regarding patient safety, no patient required a readmission due to an incident. The Program is believed to be delivering safe care. The findings of the study were discussed and then limitations, conclusions, and implications for nursing practice, education, and administration were identified. Recommendations for fiirther research were proposed. iv Table of Contents Abstract ii Table of Contents iv List of Tables vii List of Figure viii Acknowledgements ix CHAPTER ONE: INTRODUCTION 1 Background 1 SFHRIVT Program Description 2 Problem Statement and Study Purpose 4 Definition of Terms 5 Significance of the Study 6 Theoretical Framework 6 Orem's Self-Care Deficit Theory of Nursing 6 CHAPTER TWO: LITERATURE REVIEW 10 Issues that Influenced the Creation of Community IV Programs 10 Patient Selection and Admission to Community IV Programs 11 Settings in Which Community IV Antibiotics Are Delivered 13 Technology that Supports Community IV Therapy 14 Evaluations Conducted on Community IVT Programs 16 Readmission Rates 17 Adverse Drug Reactions and Complication Rates 18 Duration of Therapy while on Community IVT and Costing of Community IVT 19 Research Designs of the Outcome Studies 22 Logic Model 26 Research Hypotheses and Questions •. 29 CHAPTER THREE: METHODS 31 Evaluation Design 31 Research Design 32 Sampling Procedures 33 Data Collection Procedures 35 Data Collection for CIHI Information 35 Validity and Reliability of CIHI Data 37 Data Collection for Clinical Incidents 38 Validity and Reliability of Clinical Incident Data 39 Data Analysis 40 Study Limitations 42 Ethical Considerations 43 Summary of Methods 43 CHAPTER FOUR: FINDINGS 44 Hypothesis #1: Change in Hospital Length of Stay from Pre-Program to Post-Program Patients with Osteomyelitis 44 Hypothesis #2: Change in Hospital Length of Stay from Pre-Program to Post-Program Patients with Cellulitis 46 Additional Analysis to Determine if It Made a Difference to Hospital Stay When Some Patients Who Were Initially in Hospital Entered the Program 48 Research Question #1: Readmission Rates for Patients with Osteomyelitis and Cellulitis Who Were Initially Hospitalized 48 Description of Patients in the Simon Fraser Health Region Home IV Program 50 Research Question #3: Frequency and Types of Safety Concerns for Program Patients 51 Research Question #2: Readmission Rates for Program Patients 53 Research Question #4: Safety Concerns and Readmission Rates 54 Summary of Findings 54 CHAPTER FIVE: DISCUSSION 56 Hypothesis #1: Change in Hospital Length of Stay from Pre-Program to Post-Program Patients with Osteomyelitis 56 Hypothesis #2: Change in Hospital Length of Stay from Pre-Program to Post-Program Patients with Cellulitis 58 Research Question #1: Readmission Rates for Osteomyelitis and Cellulitis.... 60 Research Question #3: Frequency and Types of Safety Concerns for Program Patients 61 Research Question #2: Hospital Readmission Rates for Program Patients 64 Research Question #4: Safety Concerns and Readmission Rates 64 vii Usefulness of the Logic Model in Evaluating the Simon Fraser Health Region Community IV Program 65 Summary of Discussion 66 CHAPTER SIX: S U M M A R Y , LIMITATIONS, CONCLUSIONS, IMPLICATIONS, A N D RECOMMENDATIONS 68 Summary 68 Limitations 74 Conclusion 77 Study Implications 78 Implications for Nursing Practice 78 Implications for Education 79 Implications for Administrative Practice 80 Recommendations for Further Research 81 Concluding Remarks 83 References 84 Appendices Appendix I Study Matrixes 90 Appendix II SFHR Program Logic Model 103 Appendix III Data Collection Tool for Most Responsible Diagnoses of Osteomyelitis and Cellulitis Using CIHI Data 107 Appendix IV Patients in the Program Hospital Encounters 108 Appendix V Data Collection Tool for Program Patients 109 Appendix VI Data Collection Tool for Clinical Incidents 110 List of Tables Table 2.1: Outcomes for Comparison of Length of IVT in Hospital to Community. ...20 Table 4.1 Hospital Length of Stay for Patients with Osteomyelitis 1999/2000 and 2002/2003 45 Table 4.2 Hospital Length of Stay for Patients with Cellulitis 1999/2000 and 2002/2003 47 Table 4.3 Readmission Rates in 1999/2000 and 2002/2003 49 Table 4.4 Gender of Hospital Compared to Community Originating Program Patients.51 Table 4.5 Program Patient Readmissions to Hospital in 2001/2002 and 2002/2003 53 IX List of Figures Figure 3.1: Illustratation of Research Design 32 X Acknowledgements I would like to thank the many people who helped me through the process of writing this thesis. I wish to extend my deepest gratitude to Dr. Ann Hilton who convinced me to charge on despite unsupported hypotheses. Her faith in me has kept me moving this project forward through some difficult times; thanks for listening. To Dr. Carol Jillings and Dr. Anne Dewar thank you for your prompts and insightful comments. Thanks to the people who work with me and a special thanks to Lorraine Prysunka for acting as an ear. You kept me grounded and on track. Thank you to the teams I have worked with for your support and understanding as I rushed out the door for yet another meeting, or a trip to the U B C library, or for just one more article. Finally, but not least, thank you to my three children who helped with editing, who cooked yet another dinner, or who cleaned the bathrooms just one more time. To my husband David, without you this thesis would not have happened. You are my rock and my best friend. 1 CHAPTER ONE INTRODUCTION Background Community delivered antibiotic intravenous therapy is a well-established pathway for medically stable patients with serious infections. These patients are admitted to hospital, medically stabilized, and then discharged to their residences to complete their antibiotic intravenous therapy (IVT). Community outpatient IV programs were initiated to reduce health care costs, reduce patient length of stay in hospital, and improve utilization of hospital beds (Antoniskis, Anderson, Van Volkinburg, Jackson, & Gilbert, 1978; Kind, Williams & Gibson, 1985; Poretz et al., 1982; Stiver et al. 1978). The health care system in the United States has focused on reducing costs and improving utilization of acute-care beds (Antoniskis et al.; Kind et al.; Poretz et al.). The health care system in Canada has focused on improving utilization of acute care hospital beds, as more costly acutely ill patients replace discharged patients (Stiver et al. 1982). Community IV programs were identified as a means to assist patients to an earlier return to a home environment and to a return to work and school. Initially the programs treated patients with osteomyelitis and endocarditis, however as the safety and therapeutic efficacy of community IV therapy were validated, the programs expanded the type of infections treated in the community (Kind et al.). Therapeutic efficacy and safety of these treatments require that medication administration occurs with the correct dosage, frequency, and route. The IV line must be patent, in an appropriate vein, and free from phlebitis. The patient and family must be screened for suitability for the program, and then educated to perform the necessary self-2 care. Nurses screen patients and families for appropriateness for the program, educate the patient and family in how to safely manage self-care IV therapy, and monitor and support the patient and family in their home environment to ensure safe therapeutic IVT care delivery (Antoniskis et all978; Kind et al. 1985; Poretz et al. 1982; Stiver et al.1978). SFHR Community IV Program Description There are three basic models for community antibiotic IV therapy: 1) patients may attend a clinic that is staffed by health care professionals; 2) nurses may deliver the treatments at the patient's residence; or 3) patients may self-administer the medication at home (Nolet, 2000; Tice, 1996). In March 2001, the Simon Fraser Health Region (SFHR) initiated a SFHR IVT Antibiotic Program (The Program) in the New Westminster and Tri-Cities areas. The mandate for the Program is for acute and community care to collaboratively provide IV antibiotic therapy as a seamless process between hospitals and the community. The purpose of this program is to facilitate early patient discharge and prevent admissions to acute care hospitals, while providing safe and therapeutically effective IV antibiotic treatments (SFHR, 1999). The Program admits patients from the Royal Columbian Hospital (RCH) and Eagle Ridge Hospital (ERH) inpatient beds, from the emergency departments, and from physicians' offices. The Program's stakeholders include the nurses, physicians, and pharmacists working in the Program, the manager, director, and administrator who are responsible for the Program, and the patients and families who are treated in the Program. These stakeholders were consulted before the evaluation of this Program was initiated, and their input was used to provide direction of this evaluation. 3 The Program uses a team approach with an Infectious Disease Physician, Clinical Pharmacist, Home I V Nurse, and Community Intake Nurse working collaboratively to admit patients to the Program. Patients are admitted for self or caregiver administration o f I V antibiotics, or they may have a community nurse administer I V antibiotics within their residences. Primary care physicians refer their patients to one o f the Program's infectious disease physicians. The Infectious Disease Physician screens and admits medically stable patients who require I V antibiotics. This physician is responsible for directing medical care related to the patient's infectious disease process, while the referring physician is responsible for all other medical needs o f the patient. The Home I V Nurse screens the patient and family for suitability for the Program. Patients admitted to the Program must meet the selection criteria by being medically stable adults who are physically and mentally capable of participating in the Program. Patients must have a type o f infection that requires treatment with I V antibiotics for more than three days and be suitable for a community environment. The patient must have the correct venous access device for the type o f prescribed therapy. The patient and caregiver must be able to safely manage I V therapy at home and they must live in an environment that wi l l safely support community I V therapy. The home environment must have running water, a Auctioning telephone, a refrigerator, a freezer, and be situated in a neighbourhood that is considered safe for continuing care nurses to visit. Patients and caregivers are taught about the purpose and significance o f the I V therapy, the importance o f complying with the prescribed treatment, and how to maintain the venous access device. If the patient or caregiver is cognitively and physically able, then the nurse teaches the patient or caregiver how to administer the medication. I f the 4 patient or caregiver is not capable of administering the IV medication then a community nurse will visit the patient's home and administer the medication. Patients not suitable for the Program, stay within the acute care system and either remain in hospital or attend the ambulatory clinics in E R H or RCH. The pharmacist reviews the prescribed medication, checks the dosage in relation to serum drug levels as appropriate, and participates in the screening and education of patients and family members. Once the patient and caregiver have met the learning objectives for discharge to the community, the patient is referred to the Intake Nurse in the community for admission to the Continuing Care Division. The Intake Nurse reviews the patient information and ensures that the patient meets the eligibility criteria for community nursing. Patients return weekly to the hospital for re-assessment visits with the Infectious Disease Physician, Clinical Pharmacist, and Home IV Nurse. The Program's business proposal identified outcomes of improved bed utilization and increased access to other programs and services. It was predicted that by reducing acute care admissions to R C H and E R H and by reducing the length of stay for medically stable patients that there would be a 50% reduction in the hospital length of stay for the primary diagnoses of osteomyelitis, cellulitis, endocarditis, and pneumonia with 3000 saved patient days for R C H and E R H annually (SFHR, 1999). These outcome indicators address utilization issues, however the Program's purpose includes facilitation of a seamless process for delivering safe, therapeutically effective antibiotic IV therapy. Problem Statement and Study Purpose The SFHR Community IV Therapy Program has been operating in the New Westminster and Tri-Cities communities for two years and to date has had no evaluation. 5 There is a need to provide feedback to the Program's decision-makers and stakeholders about the effectiveness of the Program in maximizing the utilization of beds, and the effectiveness of the Program in delivering safe effective antibiotic IVT. The purpose of this study is to evaluate selected outcomes for the Program, specifically in terms of changes in hospital stay and the frequency of safety concerns that have arisen since the onset of the Program. Utilizing Canadian Institute for Health Information (CIHI) data, the average length of hospital stay at RCH and ERH for the Program's two most frequently diagnosed case mixed groups, osteomyelitis and cellulitis will be compared from before the Program began until after. Data from the two years before the community IV program started will be compared to data from the two years after the community IV program began operating. Continuing Care Nurses will identify the number of Program clients readmitted to hospital and the number of Program clients who experienced safety concerns as recorded by the region's clinical incident form. Definition of Terms In this study, the term of case mixed groups (CMG) is operationally defined as the method utilized by CIHI for combining acute care inpatients who have similar clinical and resource utilization characteristics. The patients are assigned into groups based upon the clinical and adrninistrative data collected from each hospital's data input into the Discharge Abstract Database (DAD) (CIHI, 2003, August). Readmission refers to a patient's unplanned admission to hospital within 28 days of discharge for further health problems or the need for further care (CIHI, 2003 August). Length of hospital stay refers to the number of days from the time the patient is registered in hospital as an inpatient until the day of discharge as entered into the DAD by the hospital (CIHI, 2003 August). 6 The clinical incident form, ENCON, is a paper-based form completed by nurses when patient's experience an unplanned change in care that has potential to alter the course of the patient's recovery. Significance of the Study This evaluation will assist in determining the effectiveness of the Program in achieving maximum utilization of acute care beds at RCH and ERH. Decision makers are reviewing all programs to determine which programs will receive financial support, and which programs will be eliminated. The Program needs to deliver safe and effective patient care and the outcomes from this evaluation will provide feedback to stakeholders and decision makers about the effectiveness of the Program. Theoretical Framework Orem 's Self-Care Deficit Theory of Nursing The Program's purpose is to provide safe and effective IV antibiotic therapy in the community. A program's goals should be matched with the most applicable theory (McKenzie & Smeltzer, 2001). The Program's goal is for patients or caregivers to self-administer the IV antibiotic, and when this is not possible, a nurse will administer the IV antibiotic in the patient's home. This goal fits Dorthea Orem's (1985) Self-Care Deficit Theory of Nursing, which focuses on the relationship between the individual and his or her environment, and the relationship between the individual and his or her ability to manage personal health care. When an individual requires and accepts nursing assistance to manage his or her health care, then a relationship is formed between the nurse and the individual. Orem's Self-Care Deficit Theory of Nursing has three parts, the Theory of Self-Care, the Theory of Self-Care Deficit, and the Theory of Nursing Systems. The 7 Theory o f Self-Care explains that people manage their personal health care either by control l ing the environment or by taking actions o f therapeutic self-care. A c t i o n s o f therapeutic self-care are specific actions the individual takes to meet personal health needs at a specific time (Orem). The individual ' s ability to personally meet this health care need is l imited by the individual ' s self-care agency, or his o r her capability for act ion (Orem). Factors such as knowledge , physical or intellectual ability may be limitations for self-care agency and the delivery o f safe care. Self-care deficit is the relationship between self-care agency, w h i c h is the individual ' s personal capability to safely meet health care needs, and therapeutic self-care, w h i c h are the actions the individual employs to meet personal health care needs (Orem, 1985). W h e n the health care need is greater than the individual ' s ability to meet the need, there is a self-care deficit, w h i c h O r e m describes as a relationship not a disorder. This self-care deficit is the catalyst for health professional intervention, and the relationship between the nurse and the individual . The Theory o f N u r s i n g Systems describes the actions that the nurse uses to help the patient or support persons to meet the patient 's self-care deficit demands (Orem). The nurse's ability to help the patient and or support person is l imited by the concept o f nursing agency, o r the nurse's capabili ty for interventions that assist the patient to meet his or her self-care deficits (Orem). The Theory o f N u r s i n g Systems supports three methods o f helping patients: 1) who l ly compensatory care, w h i c h is nurse delivered care; 2) partly compensatory care, w h i c h is nurse assisted care and; 3) supportive-educative care, w h i c h is care that supports the patient or the support person to learn to deliver personal or dependent health care (Orem). O r e m ' s Theory o f N u r s i n g Systems supports nursing individuals o f a l l ages and 8 abilities, and nursing the individual, the family, and the community. This range of patients, and the facilitation of the family and its interactions with the community, supports this theory in meeting the goals of the Program. Orem's Theory of Self-Care Deficit (1985) explains the relationships between a person's ability to meet his or her health care needs and the nurse's professional role in assisting the person to meet his or her unmet needs. Orem's theory provides the framework to select appropriate patients, to provide the appropriate level of intervention for patients, and to provide the appropriate level of education for patients and their support persons. Legitimization of the patient nurse relationship occurs when the patient consents to receive nursing care, and the nurse is able to provide the required care within the criteria of the Program. The environment can disqualify a patient from admission into the Program and Orem's theory acknowledges that environmental needs must be met before nurses can provide nursing care. The level of the patient's self-care deficit and the nurse's capability to help the patient meet the deficit will help to identify the appropriate level of intervention. Patients can receive wholly compensatory care when nurses administer IV antibiotics in the patient's home, or patients can receive supportive-educative care when the nurse supports the patient or the support person with learning to deliver the IV antibiotics. Orem's theory may support the nursing practice component of the Program, however this Program is founded on the principle of mterdisciplinary care. Orem's (1985) theory needs to support the professional practice of the physician and pharmacist. The Program's activities and members are implemented and the health care professionals respond to the needs of the patients and support persons based on the patients' identified self-care deficits. It appears that this nursing theory, which is implied 9 but not explicitly identified, is being used by non-nursing disciplines. Orem's theory is founded on the nature of relationships, and professional health care providers engage in therapeutic helping relationships with patients who consent to health care interventions. This theory should support and guide the activities of the Program's professionals in the safe delivery of I V antibiotic therapy. A concern in using this theoretical framework is that the theory is founded upon illness and deficits. The question remains as whether the theory can support the concept of wellness. Health care providers are responsible for maximizing the strengths and assets of their clients. A program review may highlight problems with this theoretical framework. 10 CHAPTER TWO LITERATURE REVIEW In this section, an overview and synthesis of theoretical and empirical literature will be presented, with the purpose of guiding the evaluation questions and methods. The literature will be reviewed to identify: 1) the issues that influenced the creation of community IV programs; 2) the client selection and admission to community IV programs; 3) the technology that supports the community IV programs; and 4) the prior studies that have evaluated community IV programs. These studies will be presented, analyzed, and critiqued. The review included a search of electronic databases, utilizing the lirniters of humans and English language, included CINAHL, Medline, EBSCO Host, Academic Search Premier and International Pharmaceutical Abstracts with the key words, intravenous therapy, antibiotics, outpatient, and infusion between 1970 to the present. Issues that Influenced the Creation of Community IV Programs Early American and Canadian studies cite cost pressures as the primary reason for initiating their programs (Antoniskis et al., 1978; Stiver et al., 1978). Later studies cite the need to efficiently utilize acute care hospital beds as a reason to operate community IV program (Poretz et al., 1982). American studies initially focused on cost savings (Balinsky & Nesbitt, 1989; Eisenberg & Kitz, 1986; Kind, Williams, & Gibson, 1985) and later, on the generation of revenue (Tice, 1996). In contrast, Canadian studies identified the need to reduce hospital length of stay. In Canada, patients have universal access to hospital care, resulting in full occupancy, with no identified savings gained from early discharges, as costs increase when acutely ill patients are admitted into the 11 empty beds (Stiver et al. 2000). As health care reform decreases the number of acute care beds, Canadian programs seem to be seeking to maximize the utilization of the acute care beds in a cost effective and therapeutically effective manner. Recent Canadian studies reflect this pressure as outcome measures include saved inpatient days, therapeutic effectiveness of treatments, and identification of the cost per patient while on the community IV programs (Choudhri, 2000; Delorme et a l , 2000; Pennie, Glavin, & Smaill, 2000; Stiver, Trosky, Cote & Oruk, 1982; Stiver et al. 2000). Studies from other countries, including the United Kingdom, state that providing long-term community IV therapy was the reason for instituting their programs (Kayley et al., 1996). A French study describes the need to evaluate the clinical efficacy of therapy, and the cost of outpatient parenteral antimicrobial therapy as the reasons for their study (Bernard et al., 2001). Reducing costs for patient care and increasing access to acute care hospital beds are the two primary issues that created the environment that supports community IV programs. However creating a program with a structure that supports safe effective therapeutic care is necessary to ensure that programs are sustainable. Patient Selection and Admission to Community IV Programs In this section a review and discussion of patient selection criteria, the process for admitting patients to the program and the environment in which community IV therapy is delivered will be presented. The literature is consistent in identifying the characteristics of an appropriate patient for a community IV antibiotic program (Choudri, 2000; Kayley et al. 1996; Nolet, 1998; Stiver et a l , 1982; Tice, 1998). Patients must be medically stable, have a patent, Sanctioning venous access device, and be able to leave the hospital except for the administration of an IV antibiotic. Patients and their families need to understand the need for the IV therapy, the guidelines for receiving the therapy, and to consent to treatment. Patients and caregivers require the physical and cognitive abilities to ensure safe administration of the IV antibiotic therapy, and to ensure medical assistance is sought when required. The home environment must be clean, must have a telephone, a refrigerator, and a freezer for storing medications. Patients must be able to accept delivery of medications and be able to travel for visits for medical re-assessments. The home IV team has responsibility for selecting patients for the program and for discharging patients from the hospital to the community. The literature refers to a home IV team as a physician, a pharmacist, and a nurse (Choudhri, 2000; Delorme et al. 2000; Mortlock & Schleis 1998; Nolet 1998; Stiver et al. 1982; Stiver et al. 2000; Williams et al. 1997). The health care providers screen the patient and caregiver for suitability for the program, and ensure that the patient and the caregiver have the necessary skills and resources to return to their community residence. The physician determines the patient's medical suitability for the Program. The Physician provides a diagnosis, orders a form of treatment, admits the patient into the program, and monitors care and evaluates outcomes (Delorme et al.; Stiver et a l , 2000; Williams et al., 1997)). Nurses also screen patients for suitability for the program by reviewing; the venous access availability, the suitability of home environment, and the educability of the patient and or caregiver. Nurses develop nursing treatment plans, monitor ongoing care, provide the education related to mamtauiing the IV therapy device, and ensure that the patient or the caregiver acquires the necessary skills for self-administration or caregiver administration of the IV antibiotic (Delorme et al., 2000; Stiver et al., 2000; Williams et al., 1997). Pharmacists screen the drug regime for suitability for a community program, 13 and educate the patient and caregiver in relation to drug information and possible adverse effects. Pharmacists monitor and evaluate outcomes and order and review serum drug levels to monitor possible drug toxicity (Williams et al.). In this literature review, once a patient is admitted into the community, the patient becomes known as a client. Clients may access outpatient antibiotic IVT in ambulatory clinics or in their homes. Settings in Which Community IVAntibiotics Are Delivered In this section, antibiotic IVT delivery in residences and in ambulatory clinics and the advantages and limitations of each delivery system will be discussed. Clients may receive IV antibiotics in their homes or in an ambulatory clinic, which are usually located in a hospital or in the community. In Canada, ambulatory clinics are generally located in hospitals and clients receiving medication and treatment in these clinics are considered hospital outpatients, not community IV clients (Choudhri, 2000; Delorme et al., 2000; Pennie, Glavin, & Smaill, 2000). Providing community IV therapy within a clinic environment allows each nurse to provide care for a number of clients at one time. As a means of reducing nursing labour costs, Alberta and British Columbia initiated trials of community IV ambulatory clinics that are operated in community health offices (Aspen Regional Health, 2001; SFHR, 2003; Vancouver Coastal Authority, 2002). These clinics are in the early developmental stages and provide care to a limited number of medically stable, ambulatory clients who have transportation to and from the clinic. Limitations to this setting include the requirement that clients be mobile and have transportation to the clinic, and that clients attend the clinic during hours that may be in conflict with work or school schedules. This setting may be a strain for clients who are fatigued from an 14 infection. The advantage to this environment is that the client and support person may access other members of the health care team. Community IV program clients who do not attend clinics receive their medications in their residences. Clients or their caregiver either self-administer the medications or receive a visit from a nurse who administers the medications. Self-administration requires that the client or the caregiver is able to master the concepts, principles, and technical skills related to his or her specific IV device and medications (Choudhri, 2000; Nolet 1998; Stiver et al. 1982). This method provides the greatest freedom for clients to have the therapy fit into their lifestyle and to have greater ability to attend work or school (Stiver et al. 1982). Clients with impaired vision or impaired fine motor movements may not be able to master the skills. Nurses may provide the IV therapy in the client's home, however the client is required to be at home at specific times to receive the medication. The advantage of this model is that the nurse has time to assess the client's response to therapy, assess the home environment, and answer questions. Nursing home visits may be difficult and expensive to sustain as the number of clients participating in these programs can be expected to increase by 10% per year, due to advances in technology with IV access devices, extended release medications and locking pumps (Williams et al., 1997). Technology that Supports Community IV Therapy In this section, an overview of changes in technology that have facilitated an increased number of clients receiving community IVT will be presented. Over the past 25 years, changes in technology related to IV antibiotics, IV access devices, and infusion devices have increased the number of individuals who can enter community IV programs. 15 This increase in community IV therapy clients is anticipated to continue due to a combination of increasing cost pressures on the health care system, the evolution of antibiotics that require daily or twice daily dosing, and new technology for vascular access and uifusion devices (Choudhri, 2000; Delorme et a l , 2000; Williams et a l , 1997). Venous access devices provide a safe means for delivering antibiotics that are irritating to the veins, but due to their effective antimicrobial activity, are important for outpatient IV programs (Hammond, 1998; Poretz, 1998; Williams et al). The development of new venous access devices that can safely remain mdweiling for weeks or months has increased the number and type of antibiotics that can be administered in the community IV programs. Venous access devices include short peripheral catheters, mid-line catheters, and peripherally inserted catheters (Gilbert, Dworkin, Raber, & Leggett, 1997; Williams et al. 1997). Short peripheral catheters are easy to insert, however they are not suitable for a community IV program because of the limited number of medications that can be administered through these lines. Mid-line catheters can accommodate hyperosmolar solutions and can remain indwelling for two to four weeks (Gilbert et al). Peripherally inserted central catheters (PICC) are indicated when venous access is difficult, or when the osmolality of the drug or the length of therapy exceeds the recommended range of therapy for short peripheral catheters or mid-line catheters (Gilbert et al.; Hammond, 1998; Williams et al.). Midline and PICC lines can accommodate a variety of medications and can be left indwelling for an extended time, however they are complex to insert and require careful observation because of potential complications. PICC lines are a central venous catheter and require the care and attention associated with other types of central lines. Potential complications include bleeding, infiltration, air emboli, infections, thrombosis, pulmonary emboli, dislodgement of the catheter from its original placement, and breaks or tears in the external catheter that require repairs or removal (Gilbert et al., 1997; Hammond, 1998). Infusion devices provide an added safety factor for PICC lines and midlines. These devices can facilitate a continuous infusion that eliminates the need to stop the infusion, flush the line, and reconnect for the next dose. Electronic computerized pumps are re-useable and user-friendly devices that require little programming (Hammond) and may facilitate an early discharge for patients who are unable to self-administer their medications (Williams et al., 1997). The evolution of technology has facilitated a greater number of individuals who can safely enter community IV programs. In response to this demand for community IV programs, pharmaceutical companies marketed longer lasting antibiotics and manufacturers of health care technology marketed venous access devices and infusion devices that supported patient or caregiver administration of medications. It is important to identify outcome measures to ensure that programs are performing the way they state they will perform, and to ensure that patients are receiving therapeutically effective care. This next section will provide a review of outcome studies performed on community IV programs. Evaluations Conducted on Community IV Programs The literature was examined for studies that evaluated community IV programs. This review of 13 studies published between 1978 and 2001, includes seven Canadian studies, three studies from the U.S.A., and one study each from France, the United Kingdom, and Australia. The following discussion will focus on the findings of these studies, the evaluation methods used by these studies, and the strengths and weaknesses 17 of these studies. The studies are surnmarized in a matrix in Appendix I. Many of the studies measured the outcomes, readmission rates, adverse drug reactions, complications of therapy, and length of therapy while on community IVT compared to length of therapy while in hospital. Nine studies measured cost of therapy on community IVT compared to cost of therapy while in hospital. These outcomes are discussed in the next section, however the discussion on costing is limited, as this study will not utilize costing as an outcome measure. Readmission Rates Six out of the thirteen studies calculated the readmission rates in their programs by identifying the number of clients who required admission to hospital due to lack of resolution of their infectious process while on the community IVT program. Three programs reported readmission rates below 5%. The McMaster community IVT program reported 1 out of 44 or 2% of the clients required readmission (Penner et al., 2000). The Vancouver Hospital program reported 4 out of 117 or 3.4% of the clients required readmission due to issues specifically related to infections (Wai, Frighetto, Marra, Chan, & Jewesson, 2000). The Australian study reported that 5 out of 133 or 3.8% of the clients required readmission to hospital during a 15-month period for clinical treatment (Montalto, 1997). Three studies reported readmission rates above 5%. In the U .S.A., a large study reported 9 out of 150 or 6% of the clients required readmission to hospital for surgical interventions related to their infections (Poretz et al, 1982). Bernard et al. (2001) reported that 4 out of 39 or 10% of the clients required readmission to hospital due to medical complications and one acute allergic reaction. Antoniskis et al. (1978) reported 18 that 3 out of 13 or 23% of the clients required readmission to hospital due to difficulty with venous access and because one caregiver withdrew support. The small sample size utilized by Antoniskis and Bernard et al. reduced the rigor of their studies and reduced the validity of their programs' readmission rate outcomes. The identified readmission rates ranged between 2% to 23%. There were however, differences in how programs were operationalized, and differences in how readmissions were reported. The Australian program admitted individuals from an emergency department, however a nurse administered the IV medications in the clients' residences (Montalto, 1997). Programs in the US utilized seiT-administration or caregiver administration (Antoniskis, 1978; Poretz, 1982). The Vancouver Hospital program and the program from France provided self-adrninistration and nurse administration of IVT (Bernard et al., 2001; Wai et al. 2000). These differences in delivery may have affected the outcomes. Individuals who self-administer must have the physical and cognitive ability to perform the tasks, however they may not be monitored as closely as individuals who receive IVT by nurses. Wai also reported readmissions for issues related to infections separately from readmissions for non-infection issues. The Vancouver program readmitted 12 community IV clients out of 117 or 10.2%, however there is no clear information on why these clients required readmission. There is no reported standard for readmission rates for community IV programs, and the reported readmission rates did not reflect the combination of self administered and nurse administered systems. Adverse Drug Reactions and Complication Rates Studies reported adverse drug reactions and complication rates. Stiver et al. (1978) discussed complication rates and types by measuring laboratory work. The 19 authors compared clients on the home IV program to a matched case group obtained by retrospective chart audit. Results indicated that the home IV clients experienced significantly lower incident rates of phlebitis and slightly higher rates of benign eosinophilia, alopecia, ototoxicity, and nephrotoxicity. The authors did not discuss reasons for the differences between the groups. Kayley et al. (1996) and Pennie et al. (2000) reported on the number of incidents of blocked catheters and the number of IV restarts. Kayley et al. had a sample of 67 clients over two years and reported problems with IV lines, drug allergies, and diarrhoea within the complication category. Pennie et al. reported blocked and leaking catheters, and restarts on 44 patients over 153 days. The studies varied in how outcomes were classified. For example, Kayley et al. included drug allergies as a complication, making comparison between studies difficult. While drug allergies can be identified through a retrospective chart audit, other outcomes that are identified through laboratory values need to be established as outcomes before individuals enter the program to ensure all subjects receive the laboratory work. Duration of Therapy while on Community IVT and Costing of Community IVT Ten of the thirteen studies measured the duration of therapy on community IV programs (Appendix A). Duration of therapy was consistently calculated by tabulating the number of days an individual received IVT in the hospital or in the community. Four studies compared the duration of therapy while on hospital IVT, to the duration of therapy while on community IVT as outlined in Table 2.1 below. Due to new antibiotics that can be given less frequently, and new technology related to electronic pumps and central lines that support stronger antibiotic therapy in the community, it is reasonable to expect that patients would be able to be safely discharged to the community IV programs 20 earlier in their hospital stay. Reviews conducted more recently should demonstrate fewer days of therapy in hospital compared to therapy in the community. Wai et al. (2000) and Bernad et al. (2001) support this expectation. The Australian study by Montalto fails to support this expectation, however this program utilized 24 hour on call staff, which makes their program expensive to operate and less cost efficient. Table 2.1 Outcomes for Comparison of Length of IVT in Hospital to Community Study Patients Percent of Therapy Percent of Therapy In Hospital in Community / / (%) / (%) Antoniskis et al., 13 22 days (51) 21 days (48) (1978) Bernard et al., 39 39 days (41) 55 days (54) (2000) Montalto, (1997) 133 5.4 days (53) 4.8 days (47) Wai et al. (2000) 107 10.9 days (32) 26.6 days (68) 21 Most studies used the duration of therapy data to calculate avoided bed days and to calculate savings related to providing care in the community rather than providing care in the hospital. Costs for therapy were inconsistently calculated, with differences related to the process each study utilized to calculate the costing. Each study found significant savings for delivering IVT in the community compared to delivering IVT in the hospital. Three Canadian studies reviewed costing per client (Choudri, 2000; Delorme et al., 2000; Wai et a l , 2000). Choudri calculated the cost per client as $92, however there is no explanation for how that cost was calculated. Delorme et al. used the hospital per diem rate to calculate cost of hospital stay for outpatient versus inpatient stays. This program provided the outpatient IVT within an ambulatory clinic that was situated within the acute care centre. Costs for the drugs do not appear to be included in this calculation. Wai et a l , from the Vancouver Hospital Health Sciences Centre, performed a treatment cost analysis that assumed clients would otherwise receive these treatments in an acute care hospital. The authors acknowledge that this assumption is a limitation because hospital patients may otherwise have received these treatments on an outpatient basis (Wai et al.). Resources for each treatment course and actual direct medical costs for the care of the hospital patients and community clients were included, however no costs were calculated for individuals missing work or the cost of day care. Mean cost per client for community therapy was calculated at $6,326, and mean cost per client if the community therapy had been delivered in the hospital was calculated at $14,271. The results of this study indicate that by using strict selection criteria, this community program is substantially less expensive than providing IVT in the hospital. The findings from these studies suggest that community IVT programs are safe and cost effective. 22 However, the validity of the findings is limited by the rigor of the research design that formed the framework for collecting the data. Research Designs of the Outcome Studies Eleven of the studies used a descriptive research design while the remaining two studies used a non-equivalent control group design (Appendix I). Limitations of descriptive research include lack of a rigorous design and therefore less ability to claim that the results are due to the intervention (Polit & Hungler, 1999). Five of these studies strengthened their design by performing their reviews over two or more years (Choudri, 2000; Kayley et al., 1996; Kind et al., 1985; Stiver et al., 1982; Stiver et a l , 2000). This longitudinal process adds to the rigor, however the lack of randornization and lack of control groups leaves the studies with limited ability to demonstrate that the results of the studies are the result of the programs' effects. Of the studies that used non-equivalent control group designs, only one study had a reasonably similar comparison group. Stiver et al. (1978) compared 23 community clients receiving home IV to 23 matched hospital patients selected through a retrospective chart review. The comparison group was matched to the home IV group using a matched case by diagnosis, age, and therapy. Home IV clients experienced 50% more incidents of benign eosinophilia, and small increases in ototoxicity, alopecia, and nephrotoxicity, and 50% fewer incidents of phlebitis than the hospital group. The authors did not provide an explanation for these differences. Costs for the home IV group were significantly less than for the hospital group. Cost per day for home IV was identified as $40 and the estimated hospital IVT was identified as $137 had the entire course of treatment been delivered in the hospital environment. 23 Antoniskis (1978) utilized a non-randomized design for 20 patients admitted to hospital for IV antibiotics to treat their infections. He selected 13 of these patients for the outpatient IV program and used the seven hospital patients who did not meet selection criteria for the outpatient IV program to act as a control group. Antonisikis found the two groups quite comparable when he measured results of therapy. Two out of thirteen home IV clients had a recurrence of their infections and one out of seven hospital patients had a recurrence of the infection. Like Stiver, Antoniskis found home IVT to be significantly less expensive than inpatient IVT. Both studies concluded that home IVT is therapeutically effective at a significantly reduced cost. Since both these studies utilized small sample sizes, non-randomized, non-equivalent designs, the results may be due to a factor other than the home IV program, however since both studies came to similar conclusions there is added support for their outcomes. The 13 studies in this review utilized convenience sampling, which is appropriate because the studies measured outcomes from all available participants. Generalization of results can be made to the individuals who were in the study. The sample sizes for five studies were greater than 100 subjects. The largest sample, from Manitoba, compared eligibility criteria and types of infections for 748 clients treated from 1978 to 1989 to 1095 clients treated from 1996 to 1998 (Choudri, 2000). Delorme et al. (2000) reviewed 343 clients over one year, Kind et al. (1982) reviewed 315 clients over 10 years, Montalto (1997), reviewed 133 clients over 15 months, and Stiver (2000) reviewed 107 clients over 3 years. The studies with larger sample sizes provided better data, however these were non-randomized samples utilizing descriptive designs with no power analysis to determine appropriate sample size. Larger sample sizes are more representative of the 24 population to which the study results are to be generalized, and reduce the rate of sampling error (Polit & Hungler 1999). The strength of this review is the number of studies with consistent outcomes that demonstrated community IVT is safe, effective and incurs less cost than delivering IVT in the hospital. The lack of homogeneity among subjects is a concern because 12 out of the 13 studies included a variety of diagnoses in their evaluations. The exception was Bernard et al. (2001), which evaluated one diagnostic group, osteomyelitis. The concern about lack of homogeneity is reduced because the 13 studies demonstrated consistent outcomes between studies. The strength for this review is that the selection criteria for participants, other than diagnosis, was consistent between studies. Subjects were medically stable, cognitively and physically able to seiT-administer or caregiver administer the IV medication utilizing the protocols set out prior to discharge. When those conditions were not met, clients received IVT from community nurses, attended ambulatory clinics, or remained in hospital. Delivery systems differed between the studies. Most studies reviewed clients and caregivers who self-administered the IV medications, however programs from the United Kingdom (UK), Australia, and the provinces of Manitoba and British Columbia, utilized a nurse-administered system when self delivery was not possible (Appendix I). Studies differed in the type of venous access devices and uifusion pumps that were utilized. This is expected since this review extends over a 23-year period and technology has changed the way IV medications are delivered. Programs moved from using short peripheral catheters that require changing approximately every 3 to 5 days, to peripherally inserted central catheters, which can remain intact for months. Infusion pumps changed the 25 ability of programs to offer home IV to individuals who required multiple medications or medications more frequently than twice per day. New antibiotics facilitated less frequent dosing and more freedom for patients to return to work or school. There were design problems in the studies. Non-randomization, inclusion of a wide range of client diagnoses, and lack of a rigorous experimental design and reliance on descriptive statistics are limitations for these studies. The strengths of the studies are uniform patient selection criteria other than diagnoses and collection of similar outcome data. The studies described the criteria utilized to determine an outcome measure, and how the outcome measures were collected. These studies consistently reported that home IV programs were safe and effective The SFHR proposal (1999) directed that the expected days avoided be set at 50%. This benchmark is comparable with the studies in this review and appears feasible. Vancouver Hospital (Wai et al., 2000) experienced an average hospital stay of 10.9 days and a home IV length of stay of 23.6 days. This is below the 50% benchmark set by the steering committee. There is no reliable standard for readmission rates for home IV programs, and the Program will document readmission rates to identify a possible benchmark for other programs that utilize both the client administered system and nurse administered system. The Vancouver Hospital program experienced a 10.2% readmission rate total for all community home IV clients over three years. The Program should expect a total readmission rate of no higher than about 10% as the Program utilizes similar selection criteria and delivery systems. The literature indicated the selection criteria and the delivery system of home IV programs are important factors that 26 affect a program's outcomes. As a means of evaluating the Program, a logic model is presented as a framework for viewing the program's goals, objectives, and activities. Logic Model In this section, an overview on program logic models and a description of the Program in terms of the logic model will be presented. The Program's logic model is described in Appendix II. A program logic model is a visual representation of a program that illustrates the linkages between the program's activities, goals, objectives, and short-term and long-term outcomes (Dwyer & Makin, 1997; Julian, Jones & Deyo, 1995). This framework facilitates the program's evaluation. The program logic model links a program's theory, goals, components, processes, and outcomes with the organization's mission, vision, and values, which facilitates consistency between the program's purpose and the organization's purpose (Hernandez, 2000). To achieve this consistency, the program's goals, which drive the program's activities, must philosophically fit with the organization's mission, vision, and values. Dwyer and Makin place the program's outcome goals at the top of the logic model, which means the program objectives and activities evolve out of the program's goals. The logic model helps program planners and evaluators articulate how the program will operate, how resources will be utilized, and how program outcomes will be measured (Dwyer & Makin; McLaughlin, & Jordan, 1999). Program managers are responsible for demonstrating that program activities and outcomes provide accountability and quality improvement information; logic models facilitate communicating this information to stakeholders (McLaughlin & Jordan, 1999). Outcome objectives drive outcome evaluations, process objectives assist with formative evaluations, and indicators aid with data collection (Julian et al., 1995). The Program utilized the SFHR mission, vision, and values as the foundation for developing the Program's goals, objectives, and indicators. This mission is working together for health, and the values and principles are respect, trust, integrity, universal accessibility, and citizen participation (SFHR, 2001). The quality indicators are safety, satisfaction, accessibility, efficiency, cost effectiveness, utility, and value (SFHR, 1997). The mission and values provide the principles upon which this program was established, and the quality indicators provide the foundation upon which the program can be evaluated. The logic model provides the framework within which to evaluate the program. This logic model was developed as a framework for the Program's evaluation. The business proposal was reviewed to identify the Program's primary goals and associated indicators, the target groups, and the program's components and processes. The two primary program goals were placed at the top of the logic model as these goals drive the program (Appendix II). The target groups are the individuals upon whom this program's activities and purpose are focused. The Program's purpose is to deliver safe, therapeutically effective, accessible, cost effective IVT in a satisfactory manner to the Program's target groups. In providing this service, it is believed the community will experience increased access to acute care beds and services. The target groups, or intended recipients of the service, are medically stable adults who require IVT for antibiotic therapy. The caregivers or families of these adults comprise the other target group. The Program's components or activities are grouped into categories labelled admission and referral, delivery of medication and monitoring of patient response to 28 therapy, and education for client and support persons. The admission and referral component addresses the activities of screening clients for appropriateness for the Program, and the admitting of clients into the community. The Home IV Team in the hospital and the Intake Nurse in the community perform these functions. The delivery of medication and monitoring of client response to therapy is performed by the client and support persons, continuing care nurses who visit clients in their homes, and the Home IV Team at reassessment visits. The Home IV Nurse and Continuing Care Nurses perform the education component. The long-term outcome objectives for each component evolve out of the program goals. The short-term outcome objectives for each component evolve out of the long-term outcome objectives. In the short-term outcome objectives within the delivery of medication and client response to therapy component, 100% of the clients are expected to report adverse reactions related to their IVT to the health care provider. Clients are expected to report to the Emergency Department if significant IVT related adverse reactions occur. If these short-term outcome objectives are achieved, then the long-term outcome objectives of community clients successfully resolving their IVT health related issues is more likely to occur. Clients who experience successful resolution of their IVT related health illnesses while in the community, should result in clients remaining in the community and a decrease in the hospital length of stay for hospital patients with osteomyelitis and cellulitis. Process objectives describe the program activities that support the outcome objectives. The process objective that health care providers will record clients' adverse reactions to IVT on clinical incident forms describes how health care providers will track adverse reactions. 29 Resources and supports are required before activities can be implemented. The resource section of the logic model identifies the environmental, human, and technical resources required for the program to operate. There is a need for home IV staff to access computers with appropriate software so the referral process can occur. The logic model illustrates the program's flow of activities and how the activities will be measured. The program indicators, objectives and goals direct the program's evaluation process. As noted in the purpose of this proposal, there is a need to provide feedback about the effectiveness of the Program in maximizing utilization of beds, and providing therapeutically effective care. This evaluation will focus on these objectives, which are stated in the delivery of medication and client response to therapy section of the logic model (Appendix II). Research Hypotheses and Questions The primary goal of the Program is to increase the community's access to hospital beds. It is believed that by delivering IVT in the community rather than in the hospital, that R C H and E R H will have increased ability to admit patients. The following research hypotheses and research questions will be used to measure the effectiveness of the Program in attauiing the primary goal. The hypotheses to be tested are: 1. The length of hospital stay for patients with osteomyelitis at R C H and E R H is lower after the second year of the Program's operation (April 1, 2002-March 31, 2003) compared to two years before the Program began (April 1, 1999-March 31, 2000) as measured by Canadian Institute for Health Information (CIHI) data. 30 2. The length of hospital stay for patients with cellulitis at R C H and E R H is lower after the second year of the Program's operation (April 1, 2002-March 31, 2003) compared to two years before the Program began (April 1, 1999-March 31, 2000) as measured by CIHI data. To address questions related to the effectiveness and safety concerns related to the Program, the following research questions were asked: 1. What is the readmission rate for patients with osteomyelitis and cellulitis at R C H and E R H two years after the Program's operation (April 1, 2002-March 31, 2003) compared to two years before the Program began (April 1, 1999-March 31, 2000)? 2. What was the readmission rate to R C H and E R H for individuals who participated in the Program between April 1, 2001 and March 31, 2003? 3. What is the frequency and type of safety concerns identified by critical incident forms generated from the Program for the two years the Program has been operating (April 1, 2001-March 31, 2003)? 4. Is there any association between the occurrence and the nature of critical incidents and the need for readmission for individuals who participated in the Program? 31 CHAPTER THREE METHODS In this chapter the process used to evaluate the hypotheses and research questions is identified by outUning and describing the evaluation design, research design, sampling procedures, data collection, and analysis procedures. The validity and reliability of CIHI data and the clinical incident data are discussed. At the end of this chapter, the limitations of this study are discussed and the ethical considerations and approval are identified. Evaluation Design A participatory evaluation approach was facilitated in this study by having the stakeholders identify the issues to be addressed in the evaluation. A participatory evaluation is collaborative and seeks contributions from all individuals involved in the program (Health Canada, 1996). The intent is for the evaluation outcomes to be useful and valuable for the stakeholders who perform the daily operations of the program and for the decision-makers who have authority related to funding the program (Health Canada). As a means of facilitating a participatory evaluation, an electronic message was sent to RCH and ERH staff asking for the identification of areas of concern or problems with the Program. All staff with access to Meditech, which is the universal electronic mail system for both facilities, should have received this memo. Feedback was received, reviewed, and organized using the logic model components. Responses indicated staff concerns were related to the admission and referral activities and the delivery of medication and client response to therapy components of the Program. During the Program's planning phase, key stakeholders, who were responsible for providing funding, directed the Program's goals and expected outcomes. These goals and expected outcomes are addressed in the two hypotheses statements. Research Design A pre-experimental pre-test post-test design was used to test the directional research hypotheses, and a descriptive design was used to answer the research questions (Polit & Hungler, 1999). The evaluation hypotheses compared for each diagnosis of osteomyelitis and cellulitis: a) the length of stay for patients admitted to R C H and E R H two years before the Program started (April 1, 1999 to March 31, 2000), to b) the length of stay for patients admitted to R C H and E R H two years after the Program began operating (April 1, 2002 to March 31, 2003). It is appropriate to use this design to compare outcomes before an intervention is implemented to after an intervention is implemented (Polit & Hungler). Length of stay data were collected into one database for osteomyelitis and one database for cellulitis. Data were collected using the pre-experimental research design illustrated in Figure 3.1. April 1,1999 Program Started April 1, 2002-March 31, 2000 April 1, 2001 March 31, 2003 O, x 0 2 Pre-test: Oi data collection Intervention: x Post-test: 0 2 data collection Figure 3.1: Illustration of Research Design 33 The study's research questions address the frequency of Program patients' safety concerns, readmissions, and the possible association between the two variables. Information on the type and frequency of safety concerns was described and counted. Readmissions within 28 days of discharge from hospital were calculated for: a)hospital patients with osteomyelitis and cellulitis for the fiscal year April 1, 1999-March 31, 2000 and for the fiscal year April 1, 2002-March 31, 2003, and for b) all Program patients for the two fiscal years April 1, 2001 until March 31, 2003. This research strategy is consistent with descriptive research design, a non-experimental approach that allows the researcher to observe, describe and document events (Polit & Hungler, 1999). This descriptive research provides information about the effects of the Program and provides direction for further evaluation. Sampling Procedures The sampling procedure for the hypotheses was total sampling, which included all known individuals who met the inclusion criteria. The samples that addressed the hypotheses included all patients over the age of 14 years, with the primary diagnoses of osteomyelitis and cellulitis admitted to R C H and E R H between April 1, 1999 and March 31, 2000 and again between April 1, 2002 and March 31, 2003 and entered into the CIHI database for each identified fiscal year. The exclusion criteria used in this study excluded patients 14 years of age and younger because the Program does not accept patients within this age range and therefore the Program cannot be expected to influence any access to the beds occupied by these children. Research question #1 reviewed the readmission rate for all patients over the age of 14 years admitted to R C H and E R H with osteomyelitis and cellulites from April 1, 1999-March 31, 2000 and again from April 1, 2002-March 31, 2003. Research question #2 reviewed the readmission rate for all patients in the Program and included all patients admitted into the Program between April 1, 2001 and March 31, 2003. Research question #3, which addressed the frequency and type of safety concerns identified by clinical incident forms generated from the Program for the time April 1, 2001 through to March 31, 2003, included all Program patient safety concerns that were identified while patients participated in the Program. A l l samples were limited to individuals recorded in the D A D by Health Record Technicians employed by Fraser Health Authority (FHA). Patients were subject to the Program's selection criteria of medically stable adults over the age of 14 years who require intravenous antibiotic therapy for more than three days. Sample size is a factor in determining the validity in rejecting a false null hypothesis. A power analysis assists in determining if the sample size is large enough to correctly test the hypotheses (Huck, 2000). Total sampling was used in this study and therefore a power analysis was not performed. The alpha level for this study was set at 0.05, which balances the risk of Type I and Type II errors while mamtaining an acceptable chance of error, and power was set at the traditional level of 0.80 (Huck). The effect size is the strength of the community IV program in decreasing the patients' length of hospital stay. No study in the literature review identified an effect size, however the literature review did establish that similar programs resulted in a reduction in the length of hospital stay for Home IV patients by about 50%. The purpose of the Program is to reduce hospital stay for patients with selected diagnoses by 50%. A moderate effect size of 0.50 (Huck) was set for this study. 35 Sample size for the study was estimated from the number of patients admitted in the first year of the Program's operation; 20 patients with osteomyelitis and 33 patients with cellulitis. The number of admissions for both these diagnoses in the fiscal year April 1, 2002 to March 31, 2003 was expected to be slightly lower due to individuals being admitted directly to the Program and avoiding hospital admission. Data Collection Procedures Data Collection for CIHI Information Data that were used to test the hypotheses, to identify the number of Program patients, and to calculate the readmission rates were collected from the CIHI database. CIHI is a merger of health information systems from the Management Information Systems Group, and the Hospital Medical Records Institute with the Health Information Division of Health Canada, Statistics Canada, and the National Health Information Council (CIHI, 2003). The purpose of this merger was to produce health information that would support health policy decision-making in Canada (CIHI). The CIHI web site acknowledges that while 100% of acute care discharges were reported from most Canadian provinces and territories, Manitoba reported only 34% and Quebec reported no discharges. Hospitals complete CIHI abstracts for all separations including discharges, stillbirths, and deaths, and submit the data to CIHI using electronic transmission or using paper (CIHI, 2003). The CIHI centre enters and processes the data to detect and correct discrepancies and edits for validity and consistency before reports are generated and published. This study utilized data from the CIHI Health Services Databases and Registries, which collects the D A D information. D A D data are collated into categories by diagnosis to create the C M G category information (CIHI). 36 In December 2003, a Health Record Technician employed by the FHA, electronically collected all CIHI data for this study. The study's purpose, hypotheses, and research questions were explained and discussed with the technician to ensure that appropriate and correct data were collected. The technician collected R C H and E R H data by C M G to identify the patients with the most responsible diagnoses of osteomyelitis and cellulitis who were discharged from R C H and E R H for the fiscal year April 1, 1999-March 31, 2000 and for the fiscal year April 1, 2002-March 31, 2003. These data, which were used to address the study's two hypotheses, included information for each patient on age, gender, the C M G code for the primary diagnosis of either osteomyelitis or cellulitis, the length of hospital stay, and the chart number. The data were entered into a database that serves as the study's hypotheses data collection tool (Appendix III). The technician collected location, admission date, age, and sex data related to patients in the Program (Appendix V). The Health Record Technician also collected the data that addressed the readmission rates for patients with osteomyelitis and cellulitis in R C H and ERH, and the readmission rates for patients in the Program. The readmission rates were calculated for patients with osteomyelitis and cellulitis for the fiscal year April 1, 1999-March 31, 2000 and for the fiscal year April 1, 2002-March 31, 2003. Readmission rates for patients in the Program were calculated for the fiscal years from April 1, 2001 through March 31, 2003. These data included a description of the readmission code, patient age, gender, chart number and or Personal Health Number, and the primary diagnosis by C M G code for all actively registered Program patients who had a hospital admission within the 37 Fraser North Region. The data were entered into a database that serves as the study's data collection tool (Appendix IV). Validity and Reliability of CIHI Data This study relies on the accuracy and consistency of CIHI data. CIHI recently published a report with the descriptive results of a review that examined the extent of variations in the D A D data (CIHI, 2003). This review was conducted due to concern that variation in abstracting and coding practices between hospitals or regions could compromise the comparability and therefore the reliability and validity of the data. A variation in data between provinces and between hospitals within provinces was identified, however the results were described with descriptive statistics, with no reliability or validity tests reported (CIHI). The differences were related to the interpretation of coding and reporting guidelines, the availability of qualified health record technicians, the influence of hospital needs for planning and funding, and the documentation practices between hospitals within provinces. Data collected by individual hospitals were considered one data set, and therefore variations within individual hospital data were not reported. The process for collecting data is standardized within facilities and the standards for collecting data is regulated by the B.C. Health Information Standards Council (Ministry of Health, 2000). Comparing data that are collected by the same group of health record technicians, utilizing the same collection techniques, the same policies and protocols that comply with the standards set by CIHI and provincial standards will support valid and reliable results. Although no statistical validity or reliability tests were found in the literature, R C H and E R H employ the same Health Record Technicians to collect data and the hospitals use a standard 38 collection process, which is believed to reduce variations in the data (K. Swinnard, personal communication, August 28, 2003). One Health Record Technician who is an employee of the FHA and works within R C H compiled the data for this study. In 2001 through 2002, CIHI introduced a new classification system called ICD-10-CA. In July 2003, CIHI produced a report that quantified the impact of their new classification system on historical trending of statistics (CIHI, 2003, July). In this report, there was no statistical evidence to indicate that the osteomyelitis C M G differed from historical trend. CIHI reports that infected skin abrasions may be coded into the C M G for trauma of skin, subcutaneous tissue, and breast tissue rather than the C M G for cellulitis. This discrepancy in coding is related to patient volume, not length of hospital stay (CIHI, 2003, July). The number of patients with infected skin abrasions that may impact the volume of cellulitis patients in this study is believed to have been minimal and will be considered a limitation of this study. Data Collection for Clinical Incidents The next set of data was collected to evaluate the frequency and type of safety concerns experienced by patients in the Program. Two clerks, one employed at R C H and one from Continuing Care collected this data as part of their regular duties. These data were obtained from the clinical incident forms, called E N C O N Forms, which were completed by Continuing Care nurses who provided home nursing visits to Program patients during the two years April 1, 2001 until March 31, 2003. The completed forms were submitted for risk management review and then after patient identification was removed, the forms were forwarded to a data management centre where information was collated and categorized. Categories include equipment failure, falls, and medication or treatment problems. After the data management centre completed the report, the forms were returned to and stored at the sending hospital, either R C H or ERH. As part of their regular duties, two clerks, one who worked at R C H and one who worked at E R H reviewed the clinical incident forms and completed the standard database. The information transcribed onto this database included the date of the incident, the number from the clinical incident form, gender, the type of incident, the comments documented on the clinical incident form, and the nurse's assessment of the degree of injury and outcomes if significant. The clinical incident forms do not provide a space for diagnosis, or chart number. The data were collected from the databases. A copy of the data collection tool is in Appendix VI. Data to address the fourth research question related to association between the occurrence and nature of critical incidents and the need for readmission for patients in the Program were collected from the databases. If a patient was to require readmission due to a clinical incident, the nurse would document that action on the clinical incident form. Validity and Reliability of Clinical Incident Data A l l data collection is subject to bias at the point of collection. Clinical incident forms are subjective and biased to the individual nurse who believes the circumstance requires documentation on the form. A form should be completed when an incident occurs that either causes harm to a patient or has the potential to cause harm to a patient, and managers have the responsibility to ensure forms are completed for these incidents. However, this data collection method is open to error. There are no statistical validity or reliability results available for the clinical incident forms. 40 Data Analysis The two study hypotheses were tested with independent t-tests to compare the length of hospital stay for patients with osteomyelitis and cellulitis for the fiscal year April 1, 1999-March 31, 2000 compared to the fiscal year April 1, 2002-March 31, 2003. The analyses utilized data provided by CIHI statistics, which consisted of each patient's diagnosis, age and gender, and the length of stay per patient in days. These data were entered into SPSS Student Version 12.0 for Windows (SPSS) to obtain descriptive statistics as a means to gain familiarity with the data (SPSS, 2003). The number of patients, the mean, the standard deviation, and the minimum and maximum number of days for length of hospital stay for patients with osteomyelitis and cellulitis were reviewed. Data were reviewed with the descriptives of age and gender to identify possible associations between length of stay and gender or age. Outliers were examined and reviewed to determine if there was an error in the recording or collection of the data, or if there was an individual who presented an extreme value (Munro, 2001). Following the review of the descriptive statistics, the hypotheses for length of stay for patients with osteomyelitis and the hypothesis for length of stay for patients with cellulitis were tested with independent t-tests. Independent t-tests were appropriate for these continuous data as the measurement scale for these length of stay data was ratio with an absolute zero and an equal distance between each day an individuals stayed in hospital (Munro, 2001). An independent t-test can be used when the question being asked is whether the means for two independent groups are different from one another (Munro, 2001). The two means are independent of one another as the comparisons were between the mean length of hospital stay for individuals admitted to hospital, (April 1, 41 1999-March 31, 2000) and the mean length of hospital stay for individuals admitted to hospital two years later, (April 1, 2002-March 31, 2003). The assumptions associated with the t-test are that each subject and group must be independent of one another, the population should meet the normality assumption, and the samples should have equal variance in characteristics (Munro, 2001). These assumptions need to be met or the outcomes attached to the statistical results may be invalid (Munro). The assumption of independence was met, as each hypothesis compared two mutually exclusive groups of subjects. These groups consisted of patients who were admitted into the hospital two years before the Program started and did not have the opportunity to receive community IVT, compared to patients who were admitted to the hospital two years after the Program started and may have received community IVT. The normality assumption was expected to be met as total sampling procedures included all individuals admitted to R C H and E R H with the diagnosis of either osteomyelitis or cellulitis. This procedure provided samples that were expected to be representative of the targeted population. The equal variance assumption was expected to be met, as the groups are comprised of patients with a single primary diagnosis of either osteomyelitis or cellulitis, however the sample sizes between groups were unequal. Variability was tested utilizing Levene's test for variability as supported by the statistical program SPSS student Version for Windows (SPSS, 2003). The two hypotheses were directional, and this study performed one-tailed tests with a level of significance of (d = 0.05) (Huck, 2000). The two research questions that addressed readmission rates utilized descriptive statistics including the total number of patients, their age, gender, and the number of 42 readmissions including the mean, the standard deviation, and the category for each readmission. For the question related to patients with osteomyelitis and cellulitis, the readmission rate for each diagnosis in the fiscal year April 1, 1999-March 31, 2000 was compared to the readmission rate for each diagnosis in the fiscal year April 1, 2002-March 31, 2003. For the question related to readmission rates for patients in the Program, the readmission rate was calculated for the period April 1, 2001 until March 31, 2003. The results are intended to provide quality improvement information and were measured against the community benchmark of a 10% readmission rate as identified in Chapter 2 in the literature review. Data were reviewed to identify possible associations between readmission and gender or age. For the clinical incident research question, the descriptive statistics addressed the two years of the Program's operation April 1, 2001 through March 31, 2003. These data included all patients in the Program regardless of diagnosis. Descriptive statistics included the total number patients, the total number of incidents, the type of incident, the degree of patient injury in each incident, and whether any patient required readmission due to the incident. These data were reviewed to determine whether or not the Program is meeting its mandate to deliver safe care. Study Limitations There are a number of limitations for this study. This study is limited in rigor due to utilizing a pre-experimental design, with generalization of the results limited to the Program population and not to other community IV programs in the FHA. This limitation may be significant as policy makers and decision makers are comparing community IV programs within the Authority to determine which program model is most 43 effective and efficient. Another limitation is the lack of validity and reliability evidence for CIHI data. Unidentified extraneous variables may be influencing the number of discharges from these two hospitals and it can be expected that some of the patients could receive their IV therapy at the ambulatory clinics if they were not participating in the Program. This design's limitations include lack of rigor due to lack of a control group. The only potential control group identified is the national measurement of these two CMGs for the same times. This non-equivalent control group lacks rigor, as there is no way to control for extraneous variables that influence the national C M G outcomes (CIHI 2003). There is no randomization, however total sampling procedures are used, which adds strength to the study's design. Ethical Considerations Permission for this study was obtained from the U B C Screening Committee for Research and Other Studies Involving Human Subjects and from The Clinical Investigation Committee chairman Dr. M . R. Foulkes, M D at RCH. This project is considered a Quality Assurance Project involving review of existing therapy, and medical records review, with no need to obtain patient consent. Qualified clerical staff performed the data collection and patient identifiers were removed. Summary of Methods This chapter presented the evaluation design, research design, sampling procedures, data collection and analysis procedures. The validity and reliability of CIHI data and the clinical incident data were discussed. The study's limitations and considerations for ethical approval were outlined. 44 CHAPTER FOUR FINDINGS In this chapter, a description of the sample and report of the findings are presented. The results are presented according to the hypothesis being tested and the research question being asked. When the samples were small, non parametric procedures were used for the analyses. Hypothesis #1: Change in Hospital Length of Stay from Pre-Program to Post-Program for Patients with Osteomyelitis It was predicted that hospital length of stay for patients admitted with osteomyelitis during the pre-test year, i.e. before the Program was offered (April 1, 1999 through March 31, 2000) would be longer than hospital length of stay for patients admitted with osteomyelitis during the post-test year after the Program was implemented (April 1, 2002 through March 31, 2003). Information on the equivalence of the two groups in terms of their age and gender is presented first. Patients at the two time points were similar in terms of age and gender. In the pre-Program year, there were 20 patients and there were 13 patients in the post-Program year. While this represents a 65% decrease in admissions, the difference between years was not statistically significant (X 2 = 1.49, p = 0.22). In the Pre-program period, patients ranged in age from 28 years to 88 years ( M = 59.4, SD = 18.33) and in the post-Program year ranged in age from 31 years to 82 years (M = 62.27, SD = 19.25). Sixty percent of the pre-Program group and 53 % of the post-Program group were 64 years or younger. There was no significant difference in gender between the two groups with 65% of the pre-Program group being men and 38.4% were men in the post-Program period ( X 2 = 0.04, p = 0.84). Hospital stays ranged from 1 to 46 days in the pre-Program period (M=15.25, SD 12.55) and from 1 to 47 days in the post-Program period (M=15.53, SD 15.6) (Table 4.1). There was no significant difference in length of stay between the two periods (t = 0.06, p = 0.48; Mann Whitney U = 114, p = 0.57), therefore the hypothesis was not supported. Table 4.1 Hospital Length of Stay for Patients with Osteomyelitis 1999/2000 and 2002/2003 1999/2000 2002/2003 Length of n = 20 n=13 stay in days / (%) / (%) 1 1 (5) 4 (30.7) 2 1 (5) 0 3 1 (5) 0 4-7 6 (30) 2 (15.3) 8-14 2 (10) 0 15-21 2 (10) 4 (30.7)* 22-29 5 (25) 1 (7.6) 30-50 2 (10) 2 (15.3) Total 20 100 13 99.6 * One 44-year-old female patient admitted into the Program after 16 day stay in hospital. Hypothesis #2: Change in Hospital Length of Stay from Pre-Program to Post-Program for Patients with Cellulitis It was predicted that length of hospital stay for patients with cellulitis would be less for the patients admitted during the post-test year (April 1, 2002 to March 31, 2003) compared to hospital length of stay for the patients admitted with cellulitis during the pre-test year (April 1, 1999 to March 31, 2000). The two groups were similar in terms of age and gender. The 60 pre-test inpatients ranged in age from 21 years to 87 years (M = 58.06, SD = 19.9) and the 74 post-test inpatients ranged in age from 26 years to 94 years (M = 62.27, SD = 19.25). The groups were similar in terms of gender with the pre-test group including 58% females and the post-test group including 53% females. Hospital stay ranged from 1 to 83 days ( M = 10.40, SD = 13.43) for the pre-test group and 1 to 106 days (M = 11.95, SD = 17.54) for the post-test group (Table 4.2). There was no significant difference in hospital stay between the groups (t = -0.57, p = 0.28; Mann Whitney U = 2201, p = 0.93). The early pattern of discharge between the groups was similar with 53.3% of the pre-test group and 55.4% of the post-test group being discharged within 7 days of hospital stay, however the post-test group had one person who stayed 106 days, 23 days longer than any patient in the pre-test period. The hypothesis was therefore not supported. Furthermore, when the patient with a 106-day hospital stay was considered an outlier and was removed from the post-test group, the post-test range became 1 to 66 days, however there was still no significant difference in the length of hospital stay between the pre-Program and post-Program groups. 47 Table 4.2 Hospital Length of Stay for Patients with Cellulitis 1999/2000 and 2002/2003 1999/2000 2002/2003 Whether Patients went into Home IV Program or not in 2002/03 In Program Not in Program Length of n = 60 n=74 n= 14 n = 60 stay in days / (%) / (%) / (%) / (%) 1 9 (15) 12 (16.2) 4 (28.6) 8 (13.3) 2 6 (10) 7 (9.4) 1 (7.1) 6 (10) 3 7 (11.6) 6 (8.1) 1 (7.1) 5 (8.3) 4-7 10 (16.6) 16 (21.6) 3 (21.4) 13 (21.6) 8-14 15 (25) 16 (21.6) 4 (28.5) 12 (20) 15-21 7 (11.6) 8 (10.8) 1 (7.1) 7 (11.6) 22-29 3 (5) 2 (2.7) 0 2 (3.3) 30-60 2 (3.3) 4 (5.4) 0 4 (6.6) en- 1 (1.6) 3* (4.0) 0 3 * (5) Total 60 (99.7) 74 (99.8) 14 (100) 60 (100) * One patient in 2002/2003 stayed 106 days. 48 Additional Analysis to Determine if It Made a Difference to Hospital Stay When Some Patients Who Were Initially in Hospital Entered the Program To further explore possible reasons for there being no significant difference in the hospital length of stay, the cellulitis patients who were initially hospitalized but entered the Program (April 1, 2002 to March 31, 2003) were separated from those patients who remained in hospital for their treatment. Because there was only one patient with osteomyelitis who was initiaUy hospitalized and then entered the Program, the comparison was limited to the patients with cellulitis. Of the 74 patients hospitalized for cellulitis, 14 were initially hospitalized and then entered the Program. The outlier patient who was in hospital for 106 days was removed for further analyses. The cellulitis patients who remained in hospital for their treatment in 2002-03 were then compared for hospital length of stay to the sample in 1999 to 2000. There was no significant difference in hospital length of stay between the two periods (t = -0.55, p = 0.58). The patients were no different in terms of age (t = -1.65, p = 0.10). Research Question #1: Readmission Rates for Patients with Osteomyelitis and Cellulitis Who Were Initially Hospitalized Readmission rates were calculated for the patients initially hospitalized for their osteomyelitis or cellulitis and compared between 1999-2000 to 2002-2003 (Table 4.3). In 1999-2000, 35% of the patients were readmitted compared to 46.2% in 2002-2003. Of these, 5% required readmission for their osteomyelitis in 1999-2000 compared to none requiring readmission for osteomyelitis in 2002-2003. Other admissions accounted for 30% and 46.2% for each period. There was no significant difference between the two periods for readmissions related to osteomyelitis (X = 0.41, p = 0.52). 49 Table 4.3 Readmission Rates in 1999/2000 and 2002/2003 Patients with Osteomyelitis 1999/2000 2002/2003 Total n = 20 n = 13 n = 33 Readmission / (%) / (%) / (%) No planned readmission 13 (65) 7 (53.8) 20 (60.6) Readmitted within week 0 (0) 0 (0) 0 (0) Readmitted within month 1 (5) 0 (0) 1 (3.0) Other 6 (30) 6 (46.2) 12 (36.4) Patients with Cellulitis 1999/2000 2002/2003 Total n = 60 n = 74 n=134 Readmission / (%) / (%) / (%) No planned readmission 17 (28.3) 31 (41.8) 47 (35) Readmitted within week 1 (1.7) 4 (5.4) 5 (3.7) Readmitted within month 1 (1.7) 4 (5.4) 5 (3.7) Other 41 (68.3) 35 (47.3) 76 (56.7) 50 In terms of readmission for cellulitis, 71.7% required readmission in 1999-2000 compared to 58.1% in 2002-2003. Of these, 3.4% required readmission because of their ceilulitis in 1999-2000 compared to 10.8% in 2002-2003. This increase in readmissions is disturbing, however the difference is not statistically significant (X 2 = 2.68, p = 0.10). Readmissions for other diagnoses were calculated at 68.3% in 1999-2000and 47.3% in 2002-2003. Description of Patients in the Simon Fraser Health Region Home IV Program The remaining research questions focus on information about the 248 patients who were part of the Home IV Program during the two year period, April 1, 2001 to March 31, 2003. Patients were admitted into the Program from hospital (n = 156) or from the community (n = 92). This included 134 males and 114 females and their ages ranged from 17 to 94 (M = 55.08, SD = 18.09). Those admitted from hospital were older (Mann-Whitney U = 5812, p = 0.01). Patients admitted to the Program from hospital ranged from 19 years to 94 years ( M = 57.2) and those admitted from the community ranged from 17 years to 88 years ( M = 51.5). Gender breakdown for both groups was similar (Table 4.4). There was no reason to expect that the clientele would differ from year to year but to examine the distribution, both periods are described. There was little variation other than the number of clients in each year. There were 81 patients admitted in the first year (2002), and 167 in the second year (2003). In 2001 to 2002, patients ranged from 20 to 91 years of age ( M = 54.92, SD = 18.67) and in 2002 to 2003, they ranged from 17 to 94 years (M = 55.16, SD = 17.86). The proportion of men to women was similar (Table 4.4). 51 Table 4.4 Gender of Hospital Compared to Community Originating Program Patients Hospital (h = 156) Community (n = 92) / (%) / Males 89 (57) 45 (49) Females 67 (43) 47 (51) 156 (100) 92 (100) Research Question #3: Frequency and Types of Safety Concerns for Program Patients One of the research questions addressed the frequency and type of safety concerns identified by clinical incident forms generated from the Program for the two years the Program had been operating, which is the time period from 2001 to 2003. Patient safety concerns were identified through a review of the clinical incident form. A total of 23 clinical incident forms were generated during the two-year period. A l l of the 23 clinical incident forms were related to issues involving IV therapy. There were 10 incidents related to equipment failures or problems, 6 related to staff knowledge or skill deficits, 4 related to family or support person knowledge or skill deficits, 2 related to patients experiencing adverse reactions to the IV therapy, and one medication error by an outside pharmacy. Of the 10 equipment failure incidents, 7 related to IV tubing and 3 to IV pumps. IV tubing incidents were consistently related to the extension tubing breaking or coming apart. Pump difficulties were related to two incidents where the alarms beeped with no identified cause and one where a pump started before the pre-set time on the pump. Of the 6 incidents related to staff knowledge and skill deficits, 2 pertained to patients missing their medications, 2 where nurses failed to follow protocol when managing IV lines, and 2 where nurses failed to correctly manage the IV pumps. Of the 4 incidents related to family and support person knowledge and skill deficits, 2 occurred when the support person appeared to lack the knowledge or skills required to solve problems when the IV pump was alarming. Another occurred when a family member forgot to connect the pump to the IV medications, which resulted in two bags of medication uifusing too rapidly, and in an additional incident, a family member did not correctly close off the IV tubing. During home visits, nurses helped the patients solve these problems. There were two incidents related to adverse effects of IV therapy, one in which a patient developed phlebitis (inflammation of the vein) and another when there was a blocked peripherally inserted central catheter line (PICC). Both reactions required medical intervention, however readmission was not required. In addition to the above, there was an incident related to a medication error made by the private company that provides medications and supplies. The private company delivered an IV medication that was labelled with the ordered medication, however the name on the medication did not match the name of the client who received the medication. The company sent the correctly labelled medication when notified of the incident and then the company completed a follow up investigation. There were therefore 23 documented incidents related to safety concerns during 53 the two years that the Program delivered care to 248 patients. This represents a 9.2% rate for documented safety incidents per client admission. No incident was documented as serious and no client was documented as requiring a readmission. Research Question #2: Readmission Rates for Program Patients Another research question related to the readmission rate for Program patients during the two years of the Program's operation (April 1, 2001 to March 31, 2003). During this two year period, there were 10.4% of the patients readmitted to hospital (Table 4.5) and of these, 4% were readmitted with diagnoses related to their primary diagnosis for which they received treatment in the Program. Of those readmitted, 1.2% were readmitted within one week and 2.8% within one month of discharge from the hospital. The majority of the Program patients, who required readmission, were hospitalized for health problems other than the infection that required community IVT. There was no additional documentation to identify the reasons for these readmissions. Table 4.5 Program Patient Readmissions to Hospital in 2001 /2002 and 2002/2003 2001/2002 2002/2003 Total n=81 n=167 n = 248 Readmission (%) / (%) / (%) / (%) No readmission 72 (88.8) 150 (89.8) 222 (89.5) Readmitted within week 2 (2.5) 1 (0.6) 3 (1.2) Readmitted within month 3 (3.7) 4 (2.4) 7 (2.8) Readmitted for other diagnosis 4 (4.9) 12 (7.2) 16 (6.4) Research Question #3: Safety Concerns and Readmission Rates An additional research question asked if there was an association between the occurrence and nature of clinical incidents and readmissions for patients in the Program. Two patients did require medical intervention, one for phlebitis and one to unblock a central venous catheter, however there was no documentation that these patients required admission to hospital. None of the remauiing 21 E N C O N forms indicated that the patients required hospitalization as a result of the incidents. There was therefore no data to substantiate that any of the readmissions were related to incidents documented on the E N C O N forms. Summary of Findings In this chapter, the findings were described. The hypotheses were not supported: there was no difference in length of hospital stay for patients with osteomyelitis or the patients with cellulitis between the pre-Program period to the period after the Program began. There were also no significant differences in the readmission rates between the pre-test period and the post-test period for either the osteomyelitis group or the cellulitis group. The readmission rates for patients with osteomyelitis patients were 35% for the pre-Program group, and 46.2% for the post-Program group. However, in the Pre-Program group, only 5% of the patients were readmitted due to osteomyelitis and in the post-Program group, no one was readmitted due to osteomyelitis. The readmission rates for the cellulitis patients were 71.7% for the pre-Program group and 58.2% for the post-Program group, however of importance is that only 3.4% required readmission related to their cellulitis in the pre-Program period but 10.8% required readmission for their cellulitis during the post-Program period. 55 When examiriing those patients who were part of the Home IV Program, most were admitted into the program from hospital versus directly from the community (63% versus 37%). Although almost 90% of the Program patients did not require any readmission to hospital, 4% were readmitted within a month because of the primary diagnosis that they were treated for in the Program while 6.4% were admitted for other reasons. When safety concerns were examined, there were 23 listed during the two years of the Program operation. Of these, 10 related to equipment failure or problems, 6 related to staff knowledge or skill deficits, 4 related to family or support person knowledge or skill deficits, 2 related to clients experiencing adverse reactions to the IV therapy and an additional incident related to a medication error. When readmissions related to safety concerns were reviewed, there was no evidence to suggest that any incident that was documented on a clinical incident form resulted in a patient readmission. The findings will be discussed and interpreted in Chapter 5. 56 CHAPTER FIVE DISCUSSION In this chapter, the study findings are interpreted and discussed. The discussion is presented according to the hypothesis being tested and the research question being asked. Limitations of the study will also be examined. Hypothesis #1 Change in Hospital Length of Stay from Pre-Program to Post-Program for Patients with Osteomyelitis The data for the pre-Program year (April 1, 1999 through March 31, 2000) and the post-Program year (April 1, 2002 through March 31, 2003) were collected using CIHI data from patients who had been admitted to two hospitals with the primary admitting diagnosis of osteomyelitis. Total sampling was used, however the sample sizes were small, with 20 patients for the pre-Program group and 13 patients for the post-Program group. Although the groups were small, they were similar in terms of age and gender. There was no significant difference in hospital length of stay between the two groups with both having average length of stays about 15 days. About 45% of patients in both the pre-Program and the post-Program groups were discharged in 7 days or less of their hospitalizations, however, the post-Program patients seemed to be discharged a little more rapidly as 30% were discharged within the first 3 days of admission while only 15% of the pre-Program patients were discharged during that same time frame. Interestingly, only one of the post-Program patients was discharged into the Program. One would have expected that more of those patients might have been discharged into the Program. This result was unexpected as patients with osteomyelitis were identified as potential community IV clients in the Program's business proposal and these patients were expected to reduce their hospital length of stay by 50% (SFHR, 1999). Osteomyelitis is indicated as appropriate for community IV therapy due to the duration of IV antibiotic therapy that is required (Bernard et al., 2001; Choudhra, 2000; Gilbert et al., 1997; Pennie, et a l , 2000; Poretz, 1998; Tice, 1998). Since a large number of patients with osteomyelitis were discharged within one week of admission to hospital, it is reasonable to question whether and where they received their community IV antibiotic therapy. As noted in the program description in Chapter One, patients who were discharged and continued to require IV antibiotics may attend the ambulatory clinics in E R H or R C H . A practice of referring patients to these ambulatory clinics may have become established before the Program became available and physicians may not have adjusted their referral patterns to include the Program. The Program's logic model identifies that patients must be referred to the Program's Infectious Disease Physician for management of the infection. This requirement may be a barrier to physicians referring to the Program, if the attending physician wishes to continue to manage the care of the patient. The findings from this hypothesis indicate that the length of hospital stay for patients with osteomyelitis did not decrease between the pre-Program group and the post-Program group. The findings imply that patients are being discharged with an infection from hospital into the community but not into the Program for community IV antibiotic therapy. The Program will need to address where these patients are being treated because i f the patients are being treated in the ambulatory units at E R H and R C H then the Program is not fully meeting the goal of increasing access to hospital resources. The sample size for this group was small and it is possible that an unidentified variable was 58 uifluencing the outcome of this hypothesis. Furthermore, the number of patients hospitalized with osteomyelitis decreased from 20 to 13 between the two periods, which may indicate that the decrease in the number of hospitalized patients in the post-Program period was related to the home IV program. Patients may have avoided hospitalization by admitting directly into the home IV program from the community. It is possible that those hospitalized patients during the post-Program period were sicker and were hospitalized because they needed more services and community IV support would not be appropriate for them. Since the patients in the hospital were older when compared to the patients in the Program, hospitalized patients may have required more assistance, and it is not unreasonable that the length of stay did not change. Hypothesis #2: Change in Hospital Length of Stay from Pre-Program to Post-Program for Patients with Cellulitis CIHI data were collected for all cellulitis patients admitted to E R H and R C H who met the Program selection criteria for the post-Program year (April 1, 2002 through March 31, 2003) and for the pre-Program year (April 1, 1999, through March 31, 2000). This total sampling method resulted in 60 patients in the pre-Program year and 74 patients in the post-Program year. The samples were similar in terms of age and gender. In terms of length of stay, there was no significant difference between groups. Both groups remained in hospital about 10 to 12 days although there was considerable variation. Since three patients stayed longer than two months in the post-Program group compared to one patient in the pre-Program group, it is possible that more patients in the post-Program group were waiting for another level of care. This difference could have impacted the length of stay outcome. Patients in the home IV program had similar length 59 of hospital stays compared to patients in the Vancouver Hospital home IV Program, which documented an average hospital stay of 10.6 days before discharge to the Vancouver IV program (Wai et al., 2000). The pattern of discharging patients remained relatively unchanged despite the opportunity to discharge into the Program, as 55% of the post-Program patients and 53% of the pre-Program patients were discharged within 1 week. One post-Program patient stayed 106 days in the hospital, which was 40 days longer in hospital than the next longest hospital stay for a post-Program patient. The results were still not significant in terms of change in hospital length of stay after this outlier was removed from the data set. During the post-test period, of the 74 people hospitalized with celluhtis, only 14 patients with cellulitis were admitted from the hospital into the Home IV Program within 21 days of their hospital admission. Since the majority of patients with ceUulitis who were admitted to hospital were discharged within 1 week of hospitalization, and only a small number of those patients were discharged into the Program, there may have been barriers to patients entering the Program. The admission criteria of the Program's logic model, identifies that patients and support persons must possess the physical and cognitive abilities to participate in the Program, and that patients must live in environments that support safe adiriinistration of IVT. These selection criteria are consistent with the criteria required of other programs (Choudri, 2000; Kayley et a l , 1996; Nolet; 1998; Tice 1998). The patients who entered the Program were younger than the patients who did not enter the Program and it is possible that the older patients did not enter the Program due to their diminished cognitive or physical ability. 60 While some patients likely did not enter the Program due to cognitive and physical abilities, some patients may have been discharged to the community for IV therapy to an undetermined location. The discharge pattern for the pre-Program and post-Program groups was similar, which implies that a system for discharging patients into the community may have existed before the Program became operational. Community IV programs in Canada became popular in the mid 1990's (Choudri, 2000; Delorme et al., 2000; Pennie et a l , 2000; Stiver et al. 2000; Wai et al. 2000). The Program did not become operational until March 2000, after other Canadian programs were well established. In the literature, cellulitis was identified as an appropriate medical condition for community IV therapy due to the length of time that IV antibiotic therapy is required to resolve the infection (Bernard et al., 2001; Choudhra, 2000; Gilbert et al., 1997; Pennie, et al., 2000; Poretz, 1998; Tice, 1998). It is possible that the early discharges that were identified in the pre-Program group were due to an alternate system for delivering community IV antibiotics that became established before the Program became operational. Research Question #1: Readmission Rates for Patients with Osteomyelitis and Cellulitis One patient was readmitted within one month for osteomyelitis in the pre-Program year and no patients were readmitted within one month for osteomyelitis in the post-Program year. Overall, readmission rates for patients with osteomyelitis increased from 35% in the pre-Program year (April 1, 1999 to March 31, 2000) to 46% in the post-Program year (April 1, 2002 to March 31, 2003). These readmissions were primarily due to patients requiring readmission for conditions other than the primary admitting diagnosis of osteomyelitis. It is reasonable to assume that the readmissions may be related to patients' co-morbidities. Patients with cellulitis experienced the reverse, as readmissions related to the primary diagnosis of cellulitis increased from 1.7% in the pre-Program year, to 10.8% in the post-Program year. More patients with cellulitis entered the community IV program compared to patients with osteomyelitis, and it is possible that the increase in these readmissions is partially due to the cellulitis patients participating in the Program. The readmission rate related to the primary diagnosis of cellulitis (10.8%) is comparable to the readmission rate for the Program patients (10.4%). This dramatic increase in readmission rate may be due to patients being discharged before they are medically stable. While the overall readmission rate for patients with cellulitis decreased in the post-Program year compared to the pre-Program year, a readmission rate of 47% warrants further investigation. The patients with osteomyelitis and the patients with cellulitis experienced high readmission rates, which were not explained with the data collected in this study. Research Question #3: Frequency and Types of Safety Concerns for Program Patients During the period April 1 2000, to March 31, 2003, there were 248 patients admitted to the Home IV Program: There were 92 patients admitted from the community and 156 patients were admitted from hospital. The percentage of patients (37%) admitted from the community was unexpectedly high, and this result may indicate that the Program is working well at directly admitting patients from the community and was likely influential in decreasing the numbers of patients who were hospitalized. The Program logic model does not address admitting patients from the community and there are no indicators to track the effects that admitting patients directly from the community will have on the outcomes of the Program. However, admitting patients directly from the community will likely help the Program meet the goal that is described in the logic model, of having patients on the Program avoid 3000 days spent in hospital. There were 23 clinical incident forms completed, therefore indicating a rate of 9.2% for safety concerns. None of the safety concerns resulted in the patient requiring hospitalization. There were 10 incidents of equipment failure of which 7 were incidents related to IV tubing breaking and coming apart and 3 incidents were related to IV pump failures. The private provider of the IV pumps promptly responded by replacing the pumps upon request. The 7 incidents of tubing failure for 248 patients (0.3% per patient) in two years is comparable to the United Kingdom community IV experience which noted 2 broken line clamps for 67 patients (0.3% per patient) in two years (Kayley et al., 1996). The Program's Team Leader followed up on the difficulty with the IV tubing by notifying the manufacturer who changed the manufacturing process used to produce the tubing (L Dowsley, private communication, December, 2003). There were 4 incidents related to support persons and 6 incidents related to staffs knowledge and skill. The problems appeared to revolve around IV line management and pump management for both families and staff. These deficits were addressed with additional education as appropriate. The outcomes documented on the clinical incident forms were noted as no degree of injury related to the incidents, however these safety issues need to be prevented. These safety concerns have a common theme that is related to the nurses', patients', and support persons' skill and knowledge related to working with the pumps and managing IV lines. Working with IVs was a new skill for some of the nurses and there may have been times when nurses had trouble due to their limited skills and experience working with IVs and pumps. These nurses often work alone in patients' homes with limited access to clinical manuals or other resources that can support the mastery of certain skills. There may not have been the expert resources to assist the nurses, especially if the nurses were working in the evening. Adverse reactions were noted as one case of mechanical phlebitis that required removal of an intravenous and one case of a PICC line that was cleared. These adverse reactions required medical intervention, however there was no documentation to indicate that either patient required readmission. This is a rate of .04% complication per patient for both the phlebitis and blocked PICC line. Vancouver Hospital's community IV program, which has similar patients, experienced 1.4% rate of mechanical phlebitis and 3% rate of blocked PICC lines (Stiver et al., 2000). The data related to adverse reactions were collected from nurses documenting on the clinical incident forms and therefore there may be more cases of phlebitis and blocked central catheters than is reflected in this study. The private pharmaceutical company that delivers medications and supplies to clients' homes made an error in labelling an IV antibiotic. However, there were no serious repercussions and the company followed-up up the incident for quality improvement purposes and to prevent a recurrence. This review of clinical incident forms highlights the need to develop a reliable system to track patients' adverse reaction to therapy and to ensure there is a reliable system to monitor and follow up safety concerns. Research Question #2: Hospital Readmission Rates for Program Patients The readmission rate to hospital for Program patients was calculated at 10.4%. This readmission rate is comparable with the Vancouver Hospital program experience of 10.2% (Wai et al., 2000). Readmissions from studies reviewed in the literature ranged from 2% to 23%, however due to the similarity between the design of the Vancouver Hospital program and the SFHR Program, the outcome of 10% was set as a benchmark for this study. The majority of Program patients required readmission for health problems coded as other, which means a health problem other than the infection being treated in the Program. The readmission rate was within the benchmark, however there was an increase in the rate of patients being readmitted with other health problems; from 4.9% of the patients in the first year to 7.2% of the patients in the second year. This trend is disturbing and leaves the question of whether or not patients are being discharged from hospital before they are medically stable. Research Question #4: Safety Concerns and Readmission Rates There was no indication that any of the safety concerns identified on the clinical incident forms resulted in patients' readmissions. There was a potential source for harm to patients that originated from equipment failures related to IV tubing and IV pumps. This harm was reduced by the prompt action of the team leader in addressing the tubing issues with the manufacturer, and the manufacturer changing the process used to produce the tubing. The information from the forms indicated that education for staff and patients and support persons was an issue, however no patient required readmission due to these 65 issues. While it is possible that a patient in the program may have been readmitted due to a clinical incident, but the incident was not documented on the clinical incident form, it is more likely that a patient experienced an adverse reaction to therapy and required admission into hospital for treatment. Clinical incidents related to practice issues are closely followed and documented, however adverse reactions to therapy are recorded in the patient's chart and not necessarily followed as closely for prevention. Usefulness of the Program Logic Model in Evaluating the Simon Fraser Health Region Community IV Program The Program logic model provided a visual representation of the program identifying the connections between the program's activities and expected outcomes, giving a clear description of the program. This structure facilitated the development of the research hypotheses and the research questions. The logic model guided the interpretation of the findings and the benchmarks for the research questions. For example, the logic model outlined the expected readmission rate for patients in the Program as 10%, and the length of hospital stay for patients with osteomyelitis and cellulitis was expected to drop by 50% from the pre-Program period to the post-Program period. The logic model was used as a guide to identify potential barriers to admitting patients into the Program, such as physician practice in referring patients, and the Program's admission criteria, and to address the frequency and types of safety concerns. The logic model outlined the process for educating the patients, their support persons, and the nurses. Indicators in the Program's logic model required a review of the clinical 66 incident form to identify specific incidents of adverse reactions to therapy. The evaluation identified that the logic model did not address the admission of patients directly from the community into the Program. These patients may have specific needs that are not identified in the logic model and therefore are not addressed in the daily activities. Summary of Discussion This evaluation measured both outcome and process activity indicators, which are considered important when planning a program evaluation (Posavac & Carey, 1997). There were no significant changes in hospital length of stay for either diagnosis of osteomyelitis or cellulitis. The distribution of hospital length of stay for the pre-test and post-test osteomyelitis and cellulitis patients were similar, which suggested that a possible method of discharging patients into the community had developed before the Program became operational. This raised the question of whether there were barriers to admitting patients into the Program, related to physician referral patterns, and admission criteria related to the age of patients in the hospital. The four research questions reviewed process objectives within the delivery and client response to therapy component of the Program's logic model. An unexpected percentage (37%) of Program patients were admitted from the community. While the hypotheses were not supported in relation to decreasing length of hospital stay, there were 92 patients who avoided hospital admission, which helped the Program meet its goal of increasing access to hospital beds. Readmission rates for the cellulitis and osteomyelitis inpatient populations were noted to be high and related to medical conditions not related to their infections. Clinical incident forms indicated that Program patients, support persons, and staff experienced difficulty with problem-solving skills and knowledge related to IV lines and pumps. The readmission rate for Program patients was within the levels identified in the literature and there was no evidence that any safety issue initiated a patient's hospitalization. There were areas of concern related to the Health Information data, the reliability of data on the clinical incident forms, the l'imited scope of the evaluation and the theoretical framework for the Program. The next section will address the study's conclusion, recommendations related to nursing practice, and nursing administration. 68 CHAPTER SIX S U M M A R Y , LIMITATIONS, CONCLUSIONS, IMPLICATIONS, A N D RECOMMENDATIONS The purpose of this study was to evaluate selected outcomes for the Program, specifically in terms of changes in hospital stay from two years before the Program was implemented to two years after the Program was operating. The frequency of safety concerns related to the Program's activities and rates of readmissions were measured. A summary of the study will be presented, limitations of the study will be discussed, conclusions of the study will be identified, and implications related to nursing practice, education, and administrative practice will be identified. Recommendations for further research will be also be proposed. Summary Community delivered IV antibiotic therapy is an accepted method for medically stable patients to be discharged home where they can complete their IV antibiotic treatments. Patients diagnosed with osteomyelitis and cellulitis are candidates for community IV therapy (Bernard et al., 2001; Choudhra, 2000; Gilbert et al., 1997; Pennie, et al., 2000; Poretz, 1998; Tice, 1998). The Program has been providing community IV antibiotic therapy for two years and the purpose of this study was to provide feedback to the Program's decision-makers and stakeholders with respect to the effectiveness of the Program in providing safe care and maximizing the utilization of hospital beds. The review of the literature identified readmission rates and safety concerns related to the IV therapy as two key variables to measure. Due to concerns related to the 69 need for patients and support persons to posses the cognitive and physical ability and the skills to manage IV antibiotic therapy, clear patient selection criteria and technology to support patients at home was identified. Outcome studies provided benchmarks for the expected readmission rates, and for the expected number of adverse reactions and safety concerns related to community IV patients (Kayley et al., 1996; Stiver et al. 2000). The Vancouver Hospital program readmission rate was used as a benchmark for the Program due to the similarities in the two programs' designs and the similarities in the types of patients admitted into the two programs (Stiver et al., 2000; Wai et al., 2000). A program logic model was retrospectively developed for the Program using the Program's original goals, objectives, and processes as a framework. A logic model facilitates consistency between a program's purpose and the organization's purposes because the model links a program's theory, goals, components, processes, and outcomes with the organization's mission, vision, and values (Hernandez, 2000). The Program's logic model is based upon the Dwyer and Makin (1997) template because this model places the outcome goals at the top, which means the program objectives and activities evolve out of the program's goals. The theoretical framework that supported the logic model was identified as Orem's Self-Care Deficit Theory of Nursing (Orem, 1985). This theory was identified after reviewing the Program's processes, goals, and activities and notching the goals with the most applicable theory. While this theory is from 1985, it remains relevant because the theory's focus is on the ability of the patient to manage personal health care. While the label self-care deficit implies limitations, the theory supports the patient's strengths and assets by focusing on the capabilities of the individual to perform self-care. The 70 health care provider initiates a professional relationship only when the patient demonstrates an inability to perform self-care. The health care provider will first try to teach the individual to perform self-care, and if unsuccessful, then the professional delivers the health care activity. Therefore, the theory supports the patient and nurse in a relationship whether the patient requires the nurse to deliver the IV therapy, or whether the patient can self-administer the IV therapy. The needs of the patient direct the level of nursing intervention that is required to support the patient. The evaluation hypotheses predicted that the hospital length of stay for patients with osteomyelitis and cellulitis that had been admitted to E R H and R C H two years before the Program began (April 1, 1999 to March 31, 2000) would be longer when compared to osteomyelitis and cellulitis patients that had been admitted two years after the Program was operating (April 1, 2002 to March 31, 2003). The evaluation research questions focused on the readmission rates of the patients with osteomyelitis and cellulitis in the pre-Program compared to post-Program time, and the readmission rates of patients in the Program during the two years of the Program's operation (April 1, 2000 to March 31, 2003). Reviewing the Program's clinical incident form revealed safety concerns related to patients during the two years of the Program's operation. The final research question addressed whether the safety concerns related to patients in the Program influenced the readmissions of these patients. A participatory evaluation design was used to address the hypotheses and research questions. Stakeholders identified the issues that focused primarily on the admission and referral component and the delivery of medication and patient response to therapy component in the logic model, however only the component related to the delivery of 71 medications was evaluated. The research design was a pre-experimental pre-test post-test design, which addressed the hypotheses, and a descriptive design, which addressed the research questions. Total sampling was used to collect data from all patients who met the selection criteria and had been admitted to either R C H or E R H within the pre-test and post-test periods identified in the hypotheses. A Health Record Technician collected CIHI data on length of hospital stay for all patients with osteomyelitis and cellulitis who had been admitted into the hospitals within the pre-Program period and the post-Program period. This Health Record Technician also collected the data relating to patient readmissions and data relating to patients in the Program. Information from the clinical incident forms was collected into a data base prepared by a clerk. The hypotheses were tested with independent t tests, and the research questions were tested with descriptive statistics. Non-parametric tests were used when sample sizes were small or when categorical data was reviewed. Data collection resulted in 60 patients with cellulitis in the pre-Program year, and 74 patients in the post-Program year and 20 patients with osteomyelitis in the pre-Program year and 13 patients in the post-Program year. Patients with cellulitis ranged in age in the pre-Program group from 21 years to 87 years and in the post-Program group from 26 years to 94 years. Patients with osteomyelitis ranged in age in the pre-Program group from 28 years to 88 years and in the post-Program group from 31 years to 82 years. No significant differences were identified in the sample characteristics of age and gender between the pre and post cellulitis or the pre and post osteomyelitis groups. It was noted that patients who entered the Program were significantly younger than patients who did 72 not enter the Program. There may be barriers for older patients entering the Program, such as diminished cognitive or physical abilities. The directional hypotheses predicted that the length of hospital stay would decrease for the post-Program groups compared to the pre-Program groups. The length of stay for patients with osteomyelitis was not different (p = 0.57) and the length of stay for patients with cellulitis was not different (p = 0.28). The distribution of the pre-Program groups and the post-Program groups were similar with 45% of the pre-test and post-test patients with osteomyelitis discharging within 7 days and 53% of the pre-test and 55% of the post-test patients with cellulitis discharging within 7 days. There were more patients with cellulitis in the post-Program group who stayed longer than two months compared to the pre-Program group. These longer stay patients were possibly waiting in hospital for an alternate level of health care for example, long-term care. In the post-Program year, 14 patients with cellulitis entered the Program, however the pattern of when patients were discharged from hospital was found to be similar for patients in the Program and for patients not in the Program. The consistent pattern of hospital length of stay for all groups may mean that another form of community IV therapy was in place before the Program began, and continues to be an option. The alternate locations for community IV therapy are believed to be the ambulatory clinics within Eagle Ridge Hospital and Royal Columbian Hospital, and use of these clinics may be due to physician referral patterns or to the Program's patient selection criteria. Readmission rates for cellulitis and osteomyelitis patients indicated that patients were admitted with health issues unrelated to either cellulitis or osteomyelitis. The readmission rates were believed to be high. Program patients numbered 248 over two 73 years (April 1, 2001 to March 31, 2003) composed of 134 males and 114 females ranging in age from 17 years to 94 years. Patients who were admitted to the Program from the community (n = 92) and from the hospital (n = 156) were similar in gender, however as noted before, the patients in the Program patients were significantly younger than patients not in the Program. Safety concerns for patients in the Program were determined by reviewing the Program's clinical incident forms. There were 23 documented incidents on the clinical incident forms, which represents a 9.2% rate for incidents per patient admission. No incident was documented as serious, and no incident was documented as resulting in a readmission to hospital. There were 10 incidents related to equipment issues and there were 10 incidents related to nurses and support persons having difficulty managing IV lines or IV pumps. Two patients had adverse reactions that required medical intervention but not readmission, one patient presented with phlebitis and one patient presented with a blocked peripheral central venous catheter. The private pharmaceutical company that supplies medications and supplies made a medication dispensing error, corrected the error, and completed a follow up report. The community Team Leader managed the equipment problems, and corrective action was implemented before this evaluation. Nurses who visited the patients in their homes provided the patients and support persons with additional education and support as required. Readmission rates for Program patients were within the benchmarks determined by the literature review. This study used the Vancouver Hospital benchmark of 10% due to the similarities in the two programs. The Program patients experienced a 10.4% readmission rate for all reasons including readmission to issues unrelated to the infection 74 that resulted in the community IV treatments. The Program patients experienced a 4% readmission rate for issues related to their infections. No documentation indicated that any of the 23 incidents identified on the clinical incident forms resulted in a patient readmission. Limitations This study had four areas related to limitations: the use of Health Information data related to the lack of reliability and validity testing and the type of data requested; the use of the clinical incident data, which has a potential lack of reliability and validity; the limited scope of the evaluation; and the retrospective determination of the Program's theoretical framework. The use of Health Information data reflects two areas of concern. There was insufficient Health Information data requested at the time that the study proposal was developed, particularly related to where and whether non-Program patients with osteomyelitis and cellulitis received IV antibiotics after discharge. The data did not provide information about patients' discharge locations, and at the proposal phase, no request was made for this information. The second area of concern related to Health Record data is the potential lack of validity and reliability of CIHI data, which was discussed in Chapter Two. The same group of Fraser Health Authority (FHA) Health Record Technicians used provincial standards to guide the collection of health record data, however no statistical reliability tests are available for R C H or E R H CIHI data. Coding errors may have occurred, and discharge and admission data may be inaccurate. A limitation related to the clinical incident forms is the potential lack reliability and validity related to the collection of the data. Nurses collected these data by completing clinical incident forms, however the decision to complete the form depends on the judgement of the nurse. Incidents may have occurred, but forms may not have been completed due to time restraints, or there have been differences in the judgement between nurses about whether a form should be completed. During the time of September 2002 through March 31, 2003, the team leader position was reduced and this reduction in the leadership structure may also have influenced the consistency in how the forms were completed and how the incidents were followed up. The data related to safety concerns for the Program is gathered from nurses documenting on the clinical incident forms and therefore the data may lack reliability and there may be more cases of phlebitis and blocked central venous catheter lines than was reflected in this study. The data used in this study were gathered from two time points, the first from April 1, 1999 to March 31, 2000, and the second from April 1, 2002 to March 31, 2003. This two-year period subjects the study to a possible threat to internal validity related to history and maturation, because changes may have occurred that influenced the hospital length of stay from the pre-test period to the post-test period. While the outcomes between the two periods were not significant, unidentified factors may have influenced the outcomes. The evaluation of this program was limited in scope to the delivery of medication and the patient response to therapy component of the Program Logic Model. There may have been unidentified variables within the other two program components, the referral and admission component and the education of the patient and the support person 76 component that influenced the outcomes. The unexpected finding that 37% of the Program patients were admitted into the Program from outside the hospital may be related to variables within the admission and referral component of the logic model. The outcome indicators in the Program Logic Model may need to be revised to address the specific needs of patients admitted from the community. More information from the admission and referral component may have helped identify why patients with cellulitis and osteomyelitis were not being admitted to the Program, as there may be unidentified barriers to admitting hospital patients into the Program. For this study, the theoretical framework for the Program was identified after the Program was in operation. Ideally, a theoretical framework guides the development and implementation of a program. McKenzie and Smeltzer advise program planners to choose a theory or model that will address the specific problem that the program is seeking to address, and then develop the program around the theory or model (2001). Orem's Self-Care Deficit Theory of Nursing was retrospectively matched to the Program's activities and processes (Orem, 1985). This is a problem, as the theory or model drives the processes and activities. The Program is attempting to help patients and families achieve independence, problem-solving skills, and a sense of wellness. Although this theory is almost 20 years old, and the Program's activities and processes appear to be supported by this theory, a review of wellness models may identify a model that further supports the patients' independence and self-care capabilities. Conclusion In this section, the study's conclusions and the extent to which the study has evaluated the Program will be discussed. Due to the pre-experimental design used in this evaluation, the results can be generalized only to the groups studied in this evaluation. One goal of the Program was to increase access to hospital beds. The study findings did not support the hypotheses that patients with osteomyelitis and cellulitis would have longer hospital stays in the pre-test group when compared to the length of hospital stay for patients with osteomyelitis and cellulitis in the post-test groups. The anticipated decrease in hospital length of stay would have resulted in increased access to hospital beds. There were more patients in the post-Program group who experienced hospital stays of two months or longer than did patients in the pre-Program group. However, 92 patients in the Program avoided hospital admission altogether, which represents avoided bed days that helped the Program to meet its goal to increase access to hospital beds. Patients with cellulitis were readmitted for cellulitis more frequently in the post-Program group compared to the pre-Program group. This jump is unexpected and raises concern regarding the medical stability of patients at the time of discharge. Program patients experienced a readmission rate that is within the benchmark identified in the logic model. There were some safety concerns for patients in the Program while staff or support persons delivered the IV therapy. There is a need to review the process used for educating staff, patients, and support persons before patients are discharged and before nurses care for patients. The rates for adverse reaction to therapy were consistent with 78 the rates found in the literature and are considered acceptable. The incidents were documented as no degree of injury to a minor degree on the patient, and no patients were documented as requiring admission to hospital due to a clinical incident. Overall, the Program is providing safe care. Patients admitted into the Program were significantly younger than patients who did not enter the Program, and it is possible that the older hospitalized patients are sicker and required more care than can be delivered by the Program. Additionally the older patients who remained in hospital may have had diminished cognitive and physical abilities that would have excluded them from being admitted into the Program. Study Implications Implications for Nursing Practice Patients in the Program were readmitted within the benchmarks identified in the literature, which provided evidence that patients were appropriately selected and appropriately and effectively treated for their infections. Nursing practice had a key role in ensuring that patient care was safe and effective by ensuring that appropriate patients were selected, and in ensuring that the delivery of the IV antibiotics was safe and accurate. When the Program decides to expand, then this selection and screening criteria and the monitoring activities could serve as a template for an expansion. The Program experienced few documented safety concerns however; the identified incidents reflect a need for the Program to ensure that nurses have access to clinical support from expert nurses. The safety concerns identified a need for nurses to monitor their practice standards and for the Program to ensure there is a clinical leadership role in place, which supports the development and maintenance of practice 79 standards. Safety concerns related to support persons and patients indicate that there is a need to review the effectiveness of the patient resource material. The nursing standards and the patient resource packages used by the Program were based upon community standards that were set by other community IV programs within the FHA, and from professional practice standards. This may be an opportunity to review all patient resource materials and practice standards for the Program, and compare outcomes with other programs. This type of review will provide information for quality improvement for community IV programs. Implications for Education The clinical incident forms indicated there were safety concerns related to nurses' and support persons' management of IV lines and IV pumps. Since the nurses teach the patients and families, it is reasonable that the first step is to review the process used to educate the nurses, and then review the learning outcomes that the nurses will use to assess the patients' and support persons' competency. The learning packages for Program nurses and Program patients and families were adapted from learning packages from other community IV programs within the FHA. The Program could look to the other community IV programs within the F H A to compare outcomes and to share updated teaching plans. This would be beneficial for the other programs as well as for the Program. This study identified incidents related to nurses' knowledge about equipment, however no reasons for the lack of knowledge were identified. There is an opportunity to raise awareness about the Program and the type of patients that receive treatment in the Program. It may be beneficial to include a short discussion about the Program during orientation for health care providers and since 80 physicians refer patients to the Program, it may be useful to provide physicians with updated information about when and how to access the Program. The Program needs to increase community awareness about community IV programs and what that means to patients and families. There is an opportunity to increase public awareness about all the community IV programs through the FHA newsletter that is posted on the FHA web and is available throughout the F H A hospitals. Implications for Administrative Practice The results of the study indicate that there is a need to review where the IV therapy is delivered, how clinical incident forms are monitored and followed up, how to link with other Programs and compare outcomes, and how to track the patients from admission to discharge. The results of the hypotheses indicate that inpatients are being discharged within a few days of admission to hospital. Some of theses patients are possibly receiving community IV antibiotic therapy, however not all these patients are discharged to the Program. Vancouver Coastal Health Authority and Aspen Regional Health Authority #11 have initiated community clinics where patients can receive their IV therapy (Aspen Regional Health, 2001; Vancouver Coastal Authority, 2002). Clinics that are situated in the community may be an option that will increase the patient capacity of the Program, and provide an alternative for patients who do not meet the patient selection criteria. The results of this study indicated that 37% of the Program patients originated from the community and not from the hospital. The Program Logic Model requires revision to accommodate the admission and treatment of these patients, including identification of outcomes related to the amount of bed days saved. One of the Program's goals is to improve access to hospital beds, and avoided bed days need to be calculated. This can be accomplished by estabUshing a system that will track patients in the Program from referral through admission and discharge. Developing this system will require Information Services and Health Record involvement. The development of an electronic tracking system may facilitate other community IV programs gathering data with the same indicators. If this is possible then community IV programs within the F H A will be able to compare outcomes, which could enhance quality improvement initiatives. The current system for tracking the patients' adverse reactions and safety concerns is manual and unreliable. The Program Logic Model identifies the adverse reactions that are to be recorded. There is an opportunity to develop a system that will facilitate the recording of all Program patients' adverse reactions. This system needs to be accessible to all members of the team and the outcomes should be followed up and reported for quality improvement purposes. Recommendations for Further Research This study was limited by using a quantitative approach to review one component of the Program. The study findings indicated that a review of the referral and admission component of the Program might be beneficial in answering the outstanding questions from this evaluation. The review of the hypotheses resulted in unexplained and unexpected outcomes of high readmission rates for inpatients with cellulitis and osteomyelitis. Additional evaluations that focus on the admission and referral component of the Program could help identify possible barriers to admitting patients into the Program. , During the pre-evaluation phase, the stakeholders identified issues that related to two 82 components, the referral and admission component and the delivery of medications and response to therapy component. Further study on the admission and referral component could help identify what, if any, barriers there are for the older population to be discharged into the Program and how to expedite individuals to be admitted directly from the community into the Program. The methodology used in this evaluation was restricted to a quantitative approach, however a qualitative approach that uses focus groups of Program stakeholders and patients may provide rich data that a quantitative approach does not support. A focus group related to nurse's knowledge might be helpful in identifying why nurses had some difficulty managing the IV equipment in the patients' homes. This information might be useful to all the community IV programs in the FHA. The results of the study indicated that there is a need to determine why patients with osteomyelitis and cellulitis experience a high rate of readmission for medical issues unrelated to their admitting diagnose. A prospective study focusing on readmissions for patients with these two diagnoses may address this question. The readmission data that were collected during this evaluation does not provide the depth of information required to determine if the readmissions indicate that patients are being discharged before they are medically stable. Additionally, the results of this evaluation leave a question of where patients in the hospital with cellulitis and osteomyelitis are discharged within a few days of admission. A descriptive review of discharge practices for patients who are discharged with active infections that require IV antibiotic therapy may address this question. Information from this study could help identify an increased number of options for physicians to discharge patients into the appropriate setting based upon specific patient care needs. Concluding Remarks Community IV antibiotic therapy programs have been an option for Canadians since 1978. The Simon Fraser Community IV Program was started in 2000, and benefited from the work of earlier programs. This study was to evaluate selected outcomes for the Program and report these outcomes to the Program's stakeholders and decision makers. The outcome of this evaluation did not support the hypotheses, however the study findings did reveal unexpected outcomes related to the origin of the Program patients. The evaluation highlighted the need for systems to support gathering and analyzing data for quality improvement. The results of this study have identified the need for continued evaluation of this program. 84 References American Psychological Association. (2001). Publication manual of the American Psychological Association (5 t h ed.). Washington DC: Author. Antoniskis, A. , Anderson, B. , Van Volkinburg, E.J., Jackson, J .M., & Gilbert, D.N. (1978). Feasibility of outpatient seiT-administration of parenteral antibiotics. The Western Journal of Medicine, 128(1), 203-206. Aspen Regional Health Authority #11. (2001, June). Coordinated community care in rural Alberta: Home visit audit findings. Best practices conference. Calgary, Alberta. Balinsky, W. & Nesbitt, S. (1989). Cost-effectiveness of outpatient parenteral antibiotics: A review of the literature. The American Journal of Medicine, 87(3), 301-305. Bernard, L. , El-hajjj, Pron, B. , Lotthe, A. , Gleizes, V . , & Signoret, F., et al. (2001). Outpatient parenteral antimicrobial therapy (OPAT) for the treatment of osteomyelitis: evaluation of efficacy tolerance and cost. Journal of Clinical Pharmacy and Therapeutics, 26(6), 445-451. Canadian Institute for Health Information. (2003, May). Coding variations in the discharge abstract database (DAD) data. Retrieved August 24, 2003, from http ://www. cehip. org/DataArea/Datalnfo/cihi. htm Canadian Institute for Health Information. (2003, July). Coping with the introduction of ICD-10-CA and CCI. Retrieved August 30, 2003, from http: //www, cehip. org/DataArea/Datalnfo/cihi. htm 85 Canadian Institute for Health Information. (2003, August). Hospital Separations. Retrieved August 4, 2003 from http ://www. cehip. or/DataArea/Datalnfo/chih. htm Choudhri, S.H. (2000). The Manitoba community intravenous therapy program: Then and now. The Canadian Journal of Infectious Diseases, 11 (Supplement A). 4A-5A. Delorme, L. , Frenette, C , LeCorre, I., Duchesne, J., Delorme, C , & Plourde, P. (2000). Ambulatory intravenous antibiotic therapy in Quebec: The Hopital Charles LeMoyne experience in 1996. The Canadian Journal of Infectious Diseases, 11 (Supplement A) 6A-1 OA. Dwyer, J .M. & Makin, S. (1997). Using a program logic model that focuses on performance measurement to develop a program. Canadian Journal of Public Health, 88(6), 421-425. Eisenberg, J .M., & Kitz, D.S. (1986). Savings from outpatient antibiotic therapy for osteomyelitis: Economic analysis of a therapeutic strategy. Journal of the American Medical Association, 255(12), 1584-1588. Gilbert, D .M. , Dworkin, R.J., Raber, S.R., & Leggett, J.E. (1997). Outpatient antimicrobial-drug therapy. The New England Journal of Medicine, 337(12), 829-839. Hammond, D. (1998). Home intravenous antibiotics: The safety factor. Journal of IV Nursing, 21(2), 81-95. Hernandez, M . (2000). Using logic models and program theory to build outcome accountability. Education & Treatment of Children, 23(1), 24-41. Huck, S.W. (2000). Reading statistics and research (3 r d ed.). New York: Longman. Julian, D.A., Jones, A. & Deyo, D. (1995). Open systems evaluation and the logic model: Program planning and evaluation tools. Evaluation and Program Planning, 18(4) 333-341. Kayley, J., Berendt, A.R., Snelling, J .M., Moore, H. , Hamilton, H.C., Peto, T.E.A., Crook, D.W.M. , & Conlon, C P . (1996). Antimicrobial practice: Safe intravenous antibiotic therapy at home: Experience of a U K based programme. Journal of Antimicrobial Chemotherapy, 37(4), 1023-1029. Kind, A.C. , Williams, D.N. , & Gibson, J. (1985). Outpatient intravenous antibiotic therapy: Ten years'experience. Post Graduate Medicine, 77(2), 105-111. McKenzie, J.F., & Smeltzer, J.L. (2001). Planning, implementing, and evaluating health promotion programs (3 r d ed.). Boston: Allyn and Bacon. McLaughlin, J.A. & Jordan, G.B. (1999). Logic models: a tool for telling your program's performance story. Evaluation and Program Planning, 22(1), 65-72. Ministry of Health. (1996). Guide to project evaluation: A participatory approach. Population Health Directorate: Health Canada. Ministry of Health Services. (2003). International statistical classification of diseases and related health problems: Tenth revision, Canadian Modification. Retrieved August 24, 2003, from http://healthnet.hnet.be.ca/hds/approved Montalto, M . (1997). An audit of patients admitted for home intravenous therapy directly from the emergency department. International Journal of Clinical Practice, 51(7), 433-436. Mortlock, N.J. , & Scheis, T. (1998). Outpatient parenteral antimicrobial therapy technology. Infectious Disease Clinics of North America, 12(4), 862-878. No let, B.R. (1998). Outpatient antimicrobial therapy: Patient selection in outpatient parenteral antimicrobial therapy. Infectious Disease Clinics of North America 12(4), 836-847. Nolet, B.R. (2000). Office and clinic-based ambulatory infusion programs: Opportunities for the infusion nurse. Journal of IV Nursing, 23(5S) (Supplement). S32-S41. Orem, D.E. (1985). Nursing: concepts of Practice (3 r d ed.). New York: McGraw-Hill Book Company. Pennie, R.A., Glavin, V . , & Smaill, F . M . (2000). Experience with a special outpatient antimicrobial program at McMaster University Medical Centre in 1997. The Canadian Journal of Infectious Diseases, 11 (Supplement A), 15A-16A. Polit, D.F., & Hungler, B.P. (1999). Nursing research principles and methods (6 t h ed.). New York: Lippincott. Poretz, D .M. , Eron, L.J . , Goldenberg, R.I., Gilbert, A.F. , Rising, J., Sparks, S., & Horn, C E . (1982). Intravenous antibiotic therapy in an outpatient setting. Journal of American Medical Association, 248(3), 336-339. Poretz, D . M . (1998). Evolution of outpatient parenteral antibiotic therapy. Infectious Disease Clinics of North America, 12(4), 828-834. Posavac, E.J. & Carey, R.G. (1997). Program evaluation: Methods and case studies (5 t h ed.). New Jersey: Prentice Hall. 88 Simon Fraser Health Region. (1997). Planning for improvement: A framework for Simon Fraser Health Region. Unpublished Manuscript. Simon Fraser Health Region. (1999). Simon Fraser Health Region community intravenous therapy program proposal. Unpublished Manuscript. Simon Fraser Health Region. (2001). Mission/vision/values. Unpublished Manuscript. Simon Fraser Health Region. (2003). Home health care ambulatory clinic guidelines. Unpublished Manuscript. Stiver, H.G., Gordon, O., Telford, Mossey, J.M., Cote, D.D, & Middlesworth, E.J. et al. (1978). Intravenous antibiotic therapy at home. Annals of Internal Medicine, 89(Part 1), 690-693. Stiver, H.G., Trosky, S.K., Cote, D.D., & Oruck, J.L. (1982). SeiT-adrninistration of intravenous antibiotics: an efficient, cost-effective home care program. Canadian Medical Association Journal, 127(3), 207-211. Stiver, G., Wai, A. , Chase, L. , Frighetto, L. , Marra, C , & Jewesson, P. (2000). Outpatient intravenous antibiotic therapy: The Vancouver Hospital experience. Canadian Journal of Infectious Diseases, 11(Supplement A), 11 A-14 A. SPSS. (2000). SPSS student version 10.0 for Windows. Prentice Hall. Tice, A .D. (1996). Alternate site infusion: The physician-directed, office based model. Journal of Intravenous Nursing, 19(4), 188-193. Tice, A .D. (1998). Outpatient parenteral antimicrobial therapy for osteomyelitis. Infectious Disease Clinics of North America, 12(4), 903-919. Vancouver Coastal Health Authority. (2002). Evergreen Ambulatory Care Clinic liaison referrals: Guideline draft 3. Unpublished Manuscript. Wai, A.O., Frighetto, L. , Marra, C.A., Chan, E., & Jewesson, P.J. (2000). Cost analysis of an adult outpatient parenteral antibiotic therapy (OP AT) Programme. PharmacoEconomics, 18(5), 451-457. Williams, D.N. , Rehm, S.J., Tice, A.D. , Bradley, J.S., Kind, A .C . , & Craig, W.A. (1997). Canadian Infectious Diseases, 25(4), 787-801. > CL o" CD <S 0 0 3 s I 2 3 CD ^ «-t-> o £ 3 -n 5 s s : 0) to 1 3 -P 5*B; c OQ 3 ad cr o 3 3 cr B; 55' i — • o" f-f-ere o' a. . P o c o o o e a. cr CO T3 o cs p CL CC 3 o rc O o c T 3 ?T ° 2? ~ o o e •a an O S £L ~ W co S- o 3- S w >< w ft 3 CD P 8 * O o" o B* o „ c a. -a cs Q o 5. 3-a § i 3 2 B" E. r o O -t CO O 3 P • • cc a to a CD i § ~ 2 3 3 3 o CD 3 ^ < ft" T3 CD S <" co o cr • P 3-cr p P 3 C L CTQ P 8. a P <5 o 3 n p Z 3 2 CL 3 3 a 3 •§ N 3 . 2 < p. p. o co O 3 3. ^ 3 o o — s 00 ^ to -1 3 3 3 |" 3 3 3 3 cr cr cr £L VI CO CO a "i O 3 T3 P 3 O O 3. c o 3 w o P a - 8 P O 5* w o o c n £ o <-> n CO I * 5 3 3 § cr ft p T3 3 5. 2. =r °^ o ^ Cu C L . E ,5 £0 _S» > ! 2 2 w 1  1  £ — to - l _ a a §' to to ft o 5-SL - . 2 == ft 3 -o s s o cr •~r o 3 1-4 ft !» CO a. p p 3 a. C L < » 2. ft o o' 3 8-3 OQ 2 * S1 3- p 3 3 | 1 I» C> 3- 3 O O 1-3 _ to _ 55 t u ft - 1 o » m 3 3 o o c -a a -t o c •a 3 OQ o o 3 T3 8 1^ S a CL-O S 3 =f 5 ' S S' I I" oo t-1 5° 3. S. o C L — c c 0 1 o to CO P 3 B" OQ tc to f s a-i 3 0Q s, r1 o 3 B. 5<-O 2 3 £ 2 » 9 0 ci c« co co 01 Ci o is cn Q 3" 05 I P C 3 Ii 3.<B S < P B a | | S 5. S £ 2". cn p c 13 Q . (5 O » CT-i-t, o- n ^ 3 5 o> rt P • o co cr. *-< ^ < " B1 S rt CO CO 2" 3 © w o ^ •a CO O 3 P 3 3 3 o^qq 5' o CO T3 & P rt" 2 <-»- >£ 3 K> o> <^> a-a w •a o> o 3 S . — Oi -* S 5 ' CO P l l o fr CO § , 3 s 3 w « p a 8 f 5 KS 2 * . o o g> — 3 « ° o g-p 3 — p * ' cr 3 I p O) to « . co ?r 3 o-3 3* *< co 5" P •o P S" 3 3 3" C O co < oi tt CO < tt 2 3-3 o „ 3 O O) 3* O o P CO /-> 2- o 3 p- a Q. 3 3 § « "3 " «-+ tt" 3 P TS o- a. ° £ o> (g S o 3 co ° - 55-a." 3 S» 3 -a o $> 3 Z P o §. ? B" 0) S Q. C 0 3* p S CO JZ . 01 tt. Oi CO ii' frl a. cr o" 9t 5" CO o O 3 3 P A) o co _ 2. 8 Z o -i oi P o. 3 o 3 sr o t r p 3 OQ Oi O 3 co a-c Oi o o_ o 3 N" p r+ 5' 3 0\ c o c r ^ " S Oi P C« Oi p • • 3 2. o CO -Q P pj 0 0 o o o o c cr to n P c_ 5" 55' o '" 3 R" P « s 3 ^ O 3" o " p ' n o o o 3 3 5 I. o o C 3 "O co 21' , 3 co 2- n 2 tt o o Z o P •a tt o o> p cr B Q . 3:v: <5 CO Ol tt T3 3" on f-1 >^ ^ 03 » o a, £ ^7* tJ> B3 2 cn g. p ~* r-h r-t »• o tS: x-rt — oi a a tt " a 8 " 5 N 3 tt p ' P S.-6 C ta-rt >-f O tt a. o o • T 3 P P i " oi CL a. tt 05 Ol 3 «s 3" « CO co y5 05 Ol , 3 3" co fr I"1 Xtf 73 Q O 9 | c o § 1 |: 11: 3 ! cf W 3 a > 5 oo to o SL O H £5 S B » OQ ?V T3 • cb"*< 3 _ CO 52. '"I a- P> cf. 3 ^ o c co < fa » g C O fr 5" P 3 S . O P B- co < =t g . ( g 8 P CO CO CD r=r , . CD » 3i a O 3 S ^ SL cr CD Cu 5* a 5" co OQ O • • 3 P C L CD 3 o CD CO O o ' o 3 3 " O CD T3 3 § 3 | £ g q g g- 3 o O • O co "2. a-CL C F> g. ST g. « TO R- o £ 8 . £ cT S CD P 3 T3 [1 0=3 CD' P 4 CD fr s ,5 OQ O < C to i-t P t 0 3 0 r T c L C D 5 C D C D 8. 2, cT cr S> to 55' °> c OQ I-I CD P O o' 3 P V CD 3 P ' o 3 CD C L < P p' cr cT CO § C L CO 3 " O T3 8. 3 £ B P 2 . o CD CO P 3. CO p 3 C L " O 3 3 — O CD 3" ** O co </2 P 3 TJ_ CD" O •-t o c T 3 co CD o c o o 3 2 £ < CD — . CO I B OQ JI-T S P T 3 p 3 C L CD 3 P T3 P o C L to' o c CO CO 1' o C to o C L P CO cr o o 3 o *+> cr Si P CO o 3" a to_ o • • 2 to o p P a 3' C L ,3' OQ co P •o CD , , 3 CO 3" <' O CD •o p £• 3' C L . § ft P 3- a CD " — T3 P 9 c^ B-3 « 5 ^ !» g- OQ a . g . ^•8 1 3 g CO CD CO CO 5 35 to B' £ O rt-CD o > g o 3 P 3 a . p p — o 3 o CD -a rt o P 3 o o O") l i 2. CT" =. -> St. rt e- 3* rt O K> r5 3 ft< 3^ O co O O " " 3 to p rt a 5 3 GO a . o 2> 3 -o 3 ^ P 3- a - a < P 3 O O (J P S. » 2*. 3 S° P Z? Ct 55 s 8 a | S S §. g- a- <-i- ~ a o p rt rt O CD O o a s co S3 3 rt f) g. CO P O » § gL 3 - P 2 CTQ 3 rt co O. 4^ £ 1 rt CD O*J s. =t a . ta-rt 3 3 0Q o • • CD —• co 3 • a CO O 3 P 2. £r P 2. T! P o tT 3 " 8 rt T3 co p ft) jj-3 VO I § •a < "=*• 2 3 5 CJQ §• 3 o f t t/5 P i t CO fr s CTQ a 3 3 o P co c o 3 5 co o ? 2. ft* co *a T3 3 ' P <* cr 0 1 5" CO 3 a . 5-P c i f fr s B* •-t "O i—• o 5 * co " 3 Et. 2 ^ «> ft -a E ^ cr ^ i« » fs 00 -* ^ • o 2 ^ <^  H 3 O 3 Q co •• tr. g -co 3 ,5" OTP O CTQ w r-1 3. 2 e g-^ co a- 3. -1 *< ft S'| I" 0 -T •' 1 s-s? 3 J*j» • co <T> jo "O <3 2 o 0 1 S •5 o 8 a rt Ct •"d o rt aN H 3* P CT* CO O-c CO rt o ^ i 3 2, co a -o 3 CO. 3 S? co £ q-ct co 3^ CO p 3 P - a j o rt SS 3 rt s: rt 2 p I I 3 3' c o O rt> o -1 PT Ct <*> Ct 3" p 3* rt rt, CO CO CO •n ~ *3 ~ CO P 3 rS^ 3 rt ^ 5 . o O rt O rtj S2. Bs co ps p 0 < v : s-" 00 8* 5-a . o co o 5 3 S. rt a . co 3 o rt S1 3-rt 1 <L « 31 p P rt CT* co 3t*CT r -1 0 5 f t < r t O p co O K- 3 " CT" 10 rt g' a - § CO X 8 rt rt 3 °rt 3" co fr r S -< H > 5 - 9 3 2? 2 » " 3 3 fr C ^ W CD 3 o CD CD rt rt „ % a | s• O CO C0 O-l**< a o CD <-< CD P rt to 3 P CD 3 o CD CD 2" a. s o CD C/5 s 5" OQ 5" -a P rt CD" 3 CD o o 3 rt i-> rt St* » ?S" 3 P p a 09 3-8 CD rt P " S T 3 Eo CD* o 2 . < N 2. P. rt. U J CD o o CD 3 CD" 3 o CD as p i t CD fr s ,B" OQ o a a o co 3 CD rt T3 P CD rt 3 o" 3 2 o rt rt1 < 3" P CD rt 2 P •a 2" v: cT CO O 3 1 co o o 3 CD _ co T3 • P o CD 3 r o 3 OQ. rt* 3 Ou If fr a 3* a. "O o rt 5" % CX. /*"•> /I*. O 3 3" £ 3" O co "2. rt* P rt O 3 I-O •be fee co ^ o 5" "O rt CD rt o o 3 D. P ' < o o CO a p_ o* 3 CD 0-Q . ,5' OQ CD CD 35 ft o 5' 8 I * 5" OQ H 2 3* 3 . CD v< CO " 3 Co' 3 3" .9* 2 3 rt P "O g. CO C CD 3 <c-r 3 CO CD a ' o" ful for c rt CO 3 g o. °* CD OQ rt CD ° 5 co a — P 2 5' *o a-, a. o 5 =• CD co O- " CD f " 0 p 2 **• rt. o* CD Si co 3 P "*' 3 o —S 3 co — R" o-» 5" P 2 CD 3 CD P 3 3 1' CO C0 8 fa I 3* co P 3 o a p ? 3 n p a 3 a S ro 3 »2 o 3 o 3. -! 2 O 3 o c= 3 ^ 5 a. C/5 CL cr 5' o CO <i 3 fl fl a 3 3- e cr o I—I CO 5T " 3 < o 3 OQ 3 P rO fl| CC t/3 C L e • rc 3 3 OQ o _. ro 3 •a ro co O 3 P P fl ON ^1 fl P c? J co 7 : 3" e-rc g co rc • C L O 3 P 3 cr o' CO p 3 C L O J OO C L -EL 3 t Q 2 3 fl I' < ; - | — OO < 5" rc o rc o P 3 fr fr I 3 OQ o ffi a § 8-8 J3 fl rc fl =• 3 CL s o- SL £ CL P o 3 co P < P cr *< p rc p fl co O P 3 o < rc •2' ? cr. CL rc g 3 to rc o fl) f frl o , CL o o o s C L O Q P " P cr 8 ^ P • • to to r to + 5* o 5- 3. ^1 Ii 3 OQ P o CL O 3 3 5 " g Et rc cr rc •1 1—1 2 ~-5. 3 2 § 2 CL a s. o rc fl <5 p 3 O co 1' o c rc" ft g.§ 3.1 00 CO cr Q §1 8.8 .ro cr: = C g. ro St o' L? Q 3 rc o p 3 8, <& cr CO p a. o 3 o p 3 p p O 3" O 3 a. OO 2" -3 •a 3 5' s « 3 £. CD to s o CD » 13 ^ CD 03 VO — ' . rt 2 -5' § 05 vo VD OO o vo OO VO P ^ 3 £ CD" 5 «> Q- • • vo oo vo II OO O o 3 < I rt 3 C3 rt oo p 3 •a^ 5" P fr OO CTQ O - B » O - P 3 T3 CD o o a. c c = " ^ 3 3 3 CD rt s-- - o o 3 2 H". to 3 ~* 3 O £L O q- 3 SjS ci rt P P -a o £ B CD ,—i 3 05 to * §••§ O rt g - ^ ^ & & .2" I-H 3 rt "O Q g ' S £§ . 3 rt 3 5. P § 3 p a. oo P a. P rt o' 3 vo vo rt - " vo oo a-! vo 3 O a. vo vo oo rt rt 3 vo oo p 3 o -o <5 P 3 o 3 8 -s a o r ~ o g o g o 3 o o rt < P a. 5" P rt CD ™ o •-+5 f frl 2 vo II VO Os H J -O s VO OS VO "a" 9P ^ H- Q. z 5>i vo P 3" i—• II Cjv: S to oo to 3 u i i o c r Lr> "3 3 rt ^ rt a -a - I O A CD . 3 P rt o ON rt 'vo P B . C/> 3 0Q vo VO vo vo 0 0 OS 3 •3 3 oo O vo vo 0 0 VO to c£ p ^ o ^-4 CO OO rt-1 rt) rt. VO rt 0 0 rt, VO rt p O a; p **\ 3 , a. si OQ a. CD 2 rt_ o_ O* tt' *o 5 o 2 P- rt' to o o 3 P rt ^ 3 rt CD 8,1. rt < 3" rt rt 03 T 3 to rt rt O 3 -CT* to rt rt » 3* 2. « rt rt p 3,* rt 3 rt rt c X> 3 2 2 5 g <-> p P *S S 2 3 p" o -a rt 3> I » 8 a "O Q3 p rt 2. rt' to 3 O rt 3 T3 3 5 Q. z s: z tr o rt o s 9 B. 2. ft •S » 5 ' r t 3 P rt 5' 3 P 3 O C o o 3 rt CD & o rt a. 03 o rt -a P 8. p 5' § rt 3 5-U c l rt rt r 1 oo OQ § 5' " p p 3 Q . rt rt 3 , <rt 3" co fr r 55" g. g S § Si rt 3. o 3 I g. E g FJ'I p a 3 3 o p o o. 3 CD 3 CD 8 2. & < o 2 P co dp p P p' CD 3 o CD C/5 3 P O P * < 3* fr fl CD CD SSI o o w crfl P o & cf § a 3 3 g 8 S - i a f l 3 SL P 3 1 & & 2 3 CO 3 CD H o CD < 3 * fl O fl CD -! o 3 P t/3 a -3 OQ C L s o CD — K> C>5 P £h 3 2. 3 cT a. P C L fl fl »• ft SL K • — CD O o 3 < I CD 3 O CD co - I 3 >ft ft fl co S t 3 OQ OO P 3 fl_ CD fr a 5" OQ o ~^ _ S 45 "r2 2^. CD 3 0 ? 3/ p 3 2 P C L CD 2 CO V< CO "> 5" co 3 p CO < a P o If o" SL O 3 o o a- o 3 I 1 o P SE 3". 5 CD o 3 CD P O 3 3 0Q 2 D C CD 3 7 3 3- n ST » 2 C L 2 ° 3 C L < p p1 *< =. » p-8 s J CD O P ^ f l a CD 2 ?r o 3 o 3 - fl s 3 § C L O 5 3 o o CD CO f fr 2. 3 " o C L co P < CD C L 3> OS -I Os -a 1—' >-t C L , £ « 0^  Q " o 4^ P CO p C L P v; CO fl~ p. SL a P CO P < CD" 2 3 H -CD P CO CO 5* 3 cr < § b-0Q p CD 5 5 S3 OQ l CD &. a O P I p o 3 3 e ft1 CD 2! C L O 3 • O 0 O CD O U> g CD Z D !* gg'S co 3 3* > C L < 2 3 0Q CD P 7T ET 0Q fl n o o — o ft cr o CD O C L P 3 5" & S 5' " CD 5-C L 5-0Q co fl 3 3 S C L 3 CD 0Q C L C« 3 fl fl a 55" I" » * P S CD ft ^ ^ p o 3 3 3 " M 3 CD 3 3 3 ! S-8L co 3 CD O CS 3 C L O S T C fl O CD CD ^ C l O ° 3 3 -a 5. P ^ O » • 3 3 « 3 a P cr CD f—1 0 0 co a P 3 It CO ™ g p fl 111 1 a & 3 co ft ft R- 5" O * 3 3 O 3 c r ^ r . rt- C L cT. ° 8 g CD ^ £L ft-PS" ff CD c: A cr 8 « o o f & CD 3 O oo o cr 3 CD Eft 3 R OQ 3 B-Vi Vi CD B o p p 3 C L CD 3 . cr' CO fr r P 5' o p s P C L P 2 p o so .£ g 5 O S D T J 2 p o os 2 - C L p u-O 3 « 5 fl) 2 1 cr o> c 5" p -o « fl> — w £T P 8 SL 8 | cr 3 fl) cr c P * o a a c o cr cn 8 ft o o !-« fl> fl) 3 a JO fl) T3 £5 T3 cn c r O H 3 % 8 §_ g c? ^ a-g a e. s 3 3 g s a * s § o o ^ cr co ro 2i •3 o co o o < co fD P PtJ 3 ' ft P o T3 cr. 3 =r. S. CO a CO SL o r o 3 0 0 f 1 3 " > OO 0 0 o o b cr to 5 0 0 3 o ci § in t p 5' 3 3 ft cr o 3 3 fl) (0 C L » p o o f o ' 5" sr OQ cr S| P -a ° 3. C L § P 3 v; 3 co ^ O P OO 0 0 , . 0 0 3 o U 3 CO 0 Q • ' T3 P O O 3 < fl) fl) 3 o fl) co P 5' OQ oo p fr oo a . 3 0 Q o ffi a S 8 -8 O p C L -i 3 . ro 13 a S P P fl) «> p 3 co 2 3 FT & 8 s a 2 8 " i fl> 3 fl> fr 2 8.. W C" o It, 3 SOQ - J <-f 3 fl) so fl) 3 P ft' g fK' rt- fl) ft 3 CO w o O T3 2 T3 P 2 p" II ^ U> co CO 0 0 cr o 3 p ft a 8 " fl> 3 3 p - o cr cr 8 2 rt 3 2 p" II V ) SO co P fl) C L 2? I I1 0 ? •a P S-1 £11 § 3 I § s-1 a » 3 a ^ 3 - O ft 3" o =T °' => fl> P o 3. 3" 2 ^ ft a K" 3 co C ST 8 o o P g H fin OQ fl) 3 cr a o 3 p o* 3 3 ft O X) 3 3 P C L O p 3 ft co P • • O 3 o f o p o <. a OQ 5 a OQ 3 co p ° I g S* OQ p-fl) O § «5 3" — o 3 ' 3 3. fo ° ~v < cr fo p 8 ^ al o n> o o fl> 3 cr cr O rt-3 "= ^ 63 fl) 3 3 3 C L P v: CO oo ,8 cr CO fr O ^ W X < H > 2 § 2 $ o g ffg Sr fl J« 3 O O 2 55' S CD P p e-l to O c 3 P C L „ P O B CD fl o § EL 5 ^ CD ft P St 3 9 S3" 3 3 * O CD CD CD 3 2 3 3 H ~ Ct fr P O £ 3 " 5". P < SS 2 CD ^ P r<1 cr P P CD Q O C L 3 ? 1 S 5 3 - f l 3 " P I £ gfl P ° 3 " no P CD ' to P fl| o to CD 3 cf 00 ca. CD 3 o CD fl o p << u> oo p1 o u> p 3 g Q P fl fl 2 ° " 3 O fl to O 3 P CD fl X o CD ca. £• o P CD 3 CD ca. 5' oo oo P fr 00 s 00 5 CD O o 3 fl_ o ' p 5" 3 O fl CD 3 ca. ft < p *i p" cr CD to CD ^ o 2 oo 3 ctqs a--cr 3* p T2. g 8 § I fr ^ f . Eh 2 £L - 2 2 - S 3 < £ ^ 2 ^ 3 ? CD 2 . D CD to If fr a 3 " ca. *•* o P ft-p" o o O 8 70 fl o 3 o o to cr CD 3 % B" i — 3 S 4 o. ^  8 c " 5 fl CD o § • 2 8. 3 " to 8r a U c 2 P oo 2 . 3 o CD P " ft CD 3 " to CD ft. p 3 fl £ ' ^ 8 ST CD O » 2 CD O to O o 3 i " P 5' 3 O CT CD O 3 O 3 3 P fl 5 8 3 3 -" p O 3 O 1 I < i—. ) — co rj" co r-, ^ 8. e. to S . 3 -CD ' -to 3 " cr 4J 8' ct ft 2T 3 * O 2 CD a. CO O 5' CD O o B J ;oo 2. CD ' 3 o' 3 g p fl * ° • , fl c 3 3 r4- co K> ; ? C L CD s. ° 3 3 * co 5' p 2. 55• g o 3 2 ° O 3 " — . CD B CD b' ft 3 O. cr 8 P •n CD fl to v; C P CD to j j . to rt O ' 3 C L O -o o o 3 I-cr 1* s ci. CD o o 3 ~ s co co 3 ft' O " « 2 3 ^ 8 ^ 8 & 3 cl 2r 3 -p ^ —. CD o ^ 3 CD O cr <; P § • CD <: fl CD ^ CD 3 - s- a 3 ft CD q 55 S-2 "cl 5 ? P 5 O g.§ * 3 & . 2L -3 ^ B o W 3 _ * ' sr CD C L Q- cr 2 f i a 8 o §• CD e - 3 . Hi 3 - & fi CD B' f 3 ( E. ft 3 §P 3 H 3 " 5" p i O a C L 3* P -CT. o P P to' p 8 C L » 5 ' CD p £ cr. C L o C L P f 00 p a; ^ 8 n 00. co cL p cr 3 co 1" R' CD oo CD , 3 Fft cr co fr r P o p 3 P o. 03 03 TS •< *0 rt O 03 3 03 rt fD 3 o o P o z 3 03 3 03 to §-. Sf- K ' 2. § o rt 03 rt' 3 2 ^ 03 p 03 g f P 3 P CO rt 3 a o o C3 o 3 g-o p •< 3 CD Q . P rt o -a CD O *-a > o' o-o P 3 P s* p" 3 2" -3 o t o CD cr CD P 3 O0 8-5-3 CTQ o • • CD rt' to 3 T3 CD rt co O 3 o o 3 •S O0 O Si -rt CD-'S*- 1—• CD 3 ft 2 §' 3 *Q OQ 00 P fr 3' 0Q P O . CD < 8 rt . CD g - Cf «> Lo rt 2 O g 0Q "o rt o O 3 H 3-Q-*0 rt _ a-O Q rt rt rt to » rtT"° rt *0 rt ^ a a S- **' 3. rt p. ° n g o 5 tn rt " O rt- CD cT " > rt 3 — J P OQ __l —1 rt-rt CO ' *a o o co to o rt> p i t s ct 3 8 2 • i ° T3 co P " a ST ° a rt n 3 Q. rt B < P ^ P p 3 P co r ? 1 t co If fr a o rt. rt 3 E- a rt n> rt J - * , CO rt. O -rt CD P O 3 - ' £ a © 0 Q 5- >^> o 3 CD 3 p* O. 00 rt) •a sf 2* o rt* 3 3 6 5 111 o a " 5 3 CO rt w 3 l*0 to 5' - 3 o a. j i . rt (S3 CO rt Q . cn rt g " °- rt K) CO ON "S 5 ft rt> O n rt S *5 3- 3 S *2 O co « CO -a H i CJ Q 2 CJ ro tO 1_> l r t^ | i a O rt i-i , 1 1 J3 rt I , ON rt T J rt u j 3 ft. to g 3 - ^ 3 o o CO n " O 2! P 2 3 5* ^ rt a rt 3 P OQ T5 E ± sr •• o i - i rt) T O to 0^ to > L-i H o. P E5* sr a § P OQ •< sr ** o 5> rt, M sr II o Q_ 03 CO ^1 111 , 3 OQ 1 CO O rt CO 03 O p cT ° < 3 3 - 5 -P «g.<B ^rt^3 U) rt rt O CD 5 CO -- « sr 0 3 rt rt g> « *? rt P* 2 *t •" fr CD B' r »• S 8-g w 9 CD T3 1 / 5 O - CD CO co rt P rt CO 2. o* o " rt. rt p rt rt cr o £ rt 3 03 O rt i£ sr^> a I 3 a rt p to rt« CO 5 a 3 a 5. rt rt O rt /rt rt " sf. o <; CD rt > 5 -rt o gr ft § • s- a CD P rt- 3 r-T o* o s UJ #T] CO rt « CD T ) P S- rt rt 2 p rt 3 g 3-sr 2 -3 P co — ' rt 3 _ O i — i rt rt S - $ "-" -T* C3 P rt '™ f i? "o "* 2 P 5 c J? a Er is. P p s 5-3* O 5 r 3- ft rt 5 -03 3 to CO P 5d r 3* U J rt P g S 1 2 f ft T3 2 rt"S. ^  sr rt o n . ft. 3 ^ ' rt, "~ 2% sr ^ rt g 3" < rt « > co P 3 rt "0 3 * D. O- co a « « o < rt « « rt rt rt) rt n K S 3 rt 9' 8 P < 9-' W P -3* a* 3 rt 2. T3 P 5* o rt CO 03 CD o a s | . ? §:« rt to p -5 a S* *T3 OO 3 , rt rt p rt Sp A O 3 g 0Q P £ 1 rt CD 3 , ft co fr 3' 3 2 o ft) o 3. 3 3 £L P ^ 3 P - J P " 3 § s ^ 3* rt o 3 s p 2 S 8 3 <-6" 3" a ro to o. > 3 O. O 2" 1^ ' 5 3 - § O ^ S - g t ^ B'S lit? co «> g Cf5 rt » O 3" O 3 ft £ O rt rt & S3 B= G rt 3. o p o 3 ^ ft co ^ rt rt oo rt 3 3 OQ rt • • -a 3 ft ft rt 3 00 rt £> a 3 co _ SL P. rt £ o* "-*-> 3 p rt C T rt oo P |fr oo rt ,5" OQ OQ 3 o » 3 rt  rt) co *^ 8 2-- • ft o 3 O rt) CO 3. «> rt o 3 rt co <*> 3 o p •a P •a *-< rt p rt 32 rt* O *o 3 3* rt O CO g rt p &-p "O a* rt rt Q . P. & P 3 3 3 P H r t r t J a. <: rt 5- B- % a OQ B5 * g rt T 3 rt <^ rt " p-CT* O P rt S\ "* < US' rt 3 3 o B-3 5 rt rt ft" I o P g. 3 a. 1 1 § 3/ 3* 2. ft & & 2 g 5" rt 3 •O ° Z -I- p bo "* a. CO p H- 3* f a g 5 *-< a p 2 3 ^ 11 S-y U l rt I ^ rt "3. . - (Z) I—' a < p < p OQ P &. rt 3 OQ i" rt S 5 3 ^ 3 P 3 P 1 3 f P rt 3* a s_ rt Q-rt P rt 3* P 3 rt rt ft co" to 00 9 CO o 3* a 3 rt , | |0Q rn rt rt CO 3 3* ^ - B* p °> a* rt rt w 3 -O <g a-8 I • Ift 2 rt 3 K» OQ -rt rt-3- rt P 3 = Ct rt 3^  P <» 3 {s-rt cr P o C= 3 co "* P 3 rt) rt 3 co <' rt 3- 3 rt 2. ' rt -s a-. a § a* ? rt P 3* P 3 rt rt- C rt rt CO 22 B • OQ P 3 3 _ &. rt « 2 0 rt T P p rt p-=* ft 3 co 32 3* • «. g. E? S tu a rt %. e-o ^ 3 » a rt I* 3 2 O se. 3 3 N rt ^ - » p ^ 3 rt 3 p ^ rt) p P 3 CO rt n ° 3 rt rt rt r P rt OQ rt co P 3 rt rt 3 <rt 3* co fr r T l p 3 o fl> H A o ^ o y f f s 2 S o 2 S3 p. § 2 a 12 § 8 o ft V o" g co 3 C L 3' 3 3- 3 to o H W " — O '-h 2 3 o p P 3 £ B C L W O g i H > A O 00 o fl C <-3' « r * ~ ft 8 3 -3 o 2. « 5 ft-3- 3 o 2 fl 3 g A> cr. "« fl) SL g ft » t o ft P CD 2 3 3 " P CT S , 5 s o es- o fl> a, H *t 8 31 « ° 3- «• o « o 0 5 fl 03 > 3 8*3 2. ft 2". § 2 OO V 5- 3 • ^ 3 OQ ft S " fl) 3- co co 3 1 12 2 — ° S3 o 3 3 •a. 3 2 3. u> 2' >^ 3 I 8 ° | fl - 2. 5 H: to fl q-OQ fl> O co _ fl) t*£ - ^ 3 cT 3 8 " ; 5 . >-*) CO S A g-8 - 2 3- 2' 2. OQ a co C 3\ 3 8 0 2 § S 3-P fl O cr H 8 ft-§ o'x 3 O c o 3 o fl > H > 8 § 8 o 2-3 3" OQ Q fl) P 3-' 2. P r-f-— o" 3 fl o fl) fl 8 Q-fl) 3, < P !-» P ' cr ST 3 P C L 00 fl> *?> P C L 5 o A) g 3 § o 3 P v: to o o © o < i cr fl> — o to o so ffi. co fl fl) fl P O SL 3 si 8. o > ^8 o 53. co fl fl) O A SL ft 8 8 CO fl cr 3 C L 3 fl) CL 3* O fl r - o o OO o o i^s 00 p 00 < v> g M , 0 Q O co —! • S O 00 W SL a. v> B g, 3 3 50 ft 3 -1 3 o o co P 5" 3 SO 5 2- • • •~r so co -t «= 2 OQ 3 " 5' SL 8. 2- S3 P 3 S ' A § ft OQ p fl) <•+ g' to cr 00 & Q ft S 3 8.8 o o g CL fl) C L g g. Er 2 « 3 p CO 3> P fl) < 3- O o p fl) co O srfl> >-i P fl cr fl> p fl o 3 p 0" 3 8 o o" 3 I fl> o fl p cr o 3 so G J 3" so O co 3> C L f l CO — 2-3 0Q fl 19 P 3 C L O 3 O i-t-P cr-8 C L «-+ cr f t cr o 3 i l l 5 ' l i sr fl -S 5 g U co cr. fl> ^2. 8 « SL Ef g" 2 OQ cr. »- S". 5 cr 2 cr p 3 3- & 8. & 1 0 2 C L => 8 « C L -1 3 - o 2t Fl » S. a x ft cr P P CO P O C L f l 3 cr. 3-fl o CO CO CO fl) SL 3 r> 3 P cr 2- % 1- o 3 . co' fl fl g -5 3 O C L S. to C L O fl> cr co O C L 2. O p • P ft fl fl cr p ~> ri cr. 0 p fl> < 3-3 3. 8 o 3 cr o 3 <W ^ w 3. 5:  So 2 O C L . -> g < " 8 5 2 0 3 3 o P 3 o cr-a ^ 3/ ft ?r 0 P* ^ 3 ? o tt f-+ * * z 2 3 s-3. 3 Ct. fl) co C L S3, ft ft> P 2 00 A l -3 O co fl) 00 p fr 00 8 5 0Q fl) co f frl cr o C L 3 0Q fl) 3 , 1 co frl P «-»-5' O 103 Appendix II PROGRAM LOGIC MODEL FOR THE SIMON FRASER HEALTH REGION IVT PROGRAM GOALS: 1. SFHR Community IVT Program clients will experience safe, effective accessible, IVT in a cost effective manner that is satisfactory to clients, families and service providers. 2. SFHR services and programs will have increased access to hospital beds. INDICATORS OF SUCCESS: 1. Clients will resolve their IVT health related problem in their homes. 2. After one year of operation, clients participating in the Program will avoid 3000 hospital days. TARGET GROUPS: 1. Medically stable clients over 14 years of age who require antibiotic IVT for more than three days who are in the community or who have been initially admitted to hospital. 2. Family and support persons who will provide home support as indicated by client need. COMPONENTS: Admission and Referral to the Program Delivery of Medication and Client Response to Therapy Education for Client and Support Person The logic model has been revised to reflect the outcomes of the study. The revisions within the Admission and Referral to the Program component, are identified by italics and have been added to reflect the need to admit clients directly from the community. 104 ADMISSION AND REFERRAL Long Term Outcome Objectives 1. 95% of medically stable clients who are over the age of 14 years, who require antibiotic therapy for more than 3 days, and who can safely be discharged to the community will enter the Program. 2. Community clients who meet Program selection criteria will admit directly into the Program. Long Term Outcome Indicators 1. After 1 year of operation a prospective chart review of discharged patients will demonstrate that 95% of eligible hospital patients entered the Program. 2. A prospective chart review of discharged hospital patients will demonstrate that patients who met admission criteria for directly admitting into the Program from the community did not admit into the hospital. 3. A prospective chart review will identify that patients in the community who qualified for direct admission into the Program are not admitted into hospital. DELIVERY OF MEDICATION AND CLIENT RESPONSE TO THERAPY Long Term Outcome Objectives 1. Clients with a primary diagnosis of, and treated for, one of the top four case mixed groups admitted into the SFHR Community IVT Program will reduce their hospital length of stay by 50%. 2. 95% of Program clients will resolve their IVT related health problem without adverse effects. 3. Clients will have an improvement in their quality of life from the time of admission into the Program until 4 weeks after their IVT experience. 4. Clients and support persons will experience their IVT in a manner that is satisfactory for them. Long Term Outcome Indicators 1. After 1 year of operation a Health Record review at RCH and ERH for case mixed groups of the most frequent diagnoses will demonstrate a 50% reduction of hospital length of stay compared to the hospital length of stay for the previous year. 2. Readmission rate for IVT clients will be 10% or less as evidenced by CIHI data. ENCON forms will be reviewed for incidents of: Phlebitis, allergic reaction, blood stream infection, blocked central venous catheters, and other IVT related adverse reactions 3. Client information from Short Form 36 Health Survey that is collected: at discharge; after 24 hours at home; and 4 weeks after discharge from the Program. 4. Clients and support person completed satisfaction questionnaires. EDUCATION FOR CLIENT AND FAMILY Long Term Outcome Objectives 1. Clients who are physically and mentally able will self-administer or have support person administered IVT within 3 days of admission to the Program. 2. All clients or support persons who administer the IVT will possess the knowledge and skills to deliver safe effective IVT. Long Term Outcome Indicators 1. Nursing documentation will evidence that 95% of capable clients will self-administer or have support person administered IVT within 3 days of admission to the Program. 2. No client experienced adverse effects as a result of a break in technique or improperly managed IVT as evidenced by chart documentation and ENCON forms. 105 Intermediate Outcome Objective Clients and support persons use correct techniques for managing IV devices. Intermediate Outcome Indicator Clients and support persons demonstrate correct IVT management during nursing home visits and during reassessment clinics. ADMISSION AND REFERRAL Short Term Outcome Objectives 1 . All clients either return to or enter into a home environment that contains the necessary supports to deliver safe administration of community IV. 2. Clients and support persons possess the physical and cognitive abilities to participate in the Program. Short Term Outcome Indicators 1. Review ENCON forms for safety issues detected in clients' homes. 2. Review the documentation that indicates clients or support persons failed to successfully participate in the Program. DELIVERY OF MEDICATION AND CLIENT RESPONSE TO THERAPY Short Term Outcome Objectives 1. Clients experience resolution of adverse effects from treatment. 2. Clients or support persons report IVT adverse effects to their health care provider. 3. Clients seek medical aid when adverse effects from rVT are consistent with instructions to seek medical aid. Instructions are contained in the client information package provided at the time of admission into the Program. Short Term Outcome Indicators 1. Documentation on the clients' chart for the number of days the client required to achieve resolution of the adverse reaction to IVT. 2. The number of clients who did not report adverse effects compared to the number of clients who did report adverse effects to the health care provider. 3. The number of r V T clients who reported to the Emergency Department as captured by admission and discharge data. EDUCATION FOR CLIENT AND FAMILY Short Term Outcome Objectives 1 . All clients and support persons demonstrate correct techniques and provide correct rationale for medication delivery before self-management of IVT or IV devices. Short Term Outcome Indicators 1. Clients and support persons successfully complete the education skills as documented on the teaching tools. 106 ADMISSION AND REFERRAL Process Objectives 1. Within 1 year of operation, there will be SFHR standardized regional client screening and referral processes. This process will be facilitated by a seamless electronic information system. 2. 95% of eligible clients will be assessed for the Program within 1 working day of referral to the Program. 3. 95% of eligible clients will enter the Program within 2 working days of referral. 4. All clients initially admitted to hospital and referred to the Program will meet selection criteria. 5. The Program admission process will identify a process for admitting clients directly from the community DELIVERY OF MEDICATION AND CLIENT RESPONSE TO THERAPY Process Objectives 1. Policies and procedures for client care delivery will be regionally standardized and meet professional standards for practice. 2. ENCON forms, will be completed for known client adverse reactions to IVT. 3 FHR Community IVT will be provided within the allotted budget or be appropriately justified. 4. Clients will receive medications supplies and equipment before medication delivery times are due. 5. Equipment and supplies will be regionally standardized. EDUCATION FOR CLIENT AND FAMILY Process Objectives 1. Educational material for clients support persons and staff will meet best practice and professional standards for practice. 2. Standards for assessing client and support person knowledge and skill level will be regionally standardized. 107 ADMISSION AND REFERRAL Process Indicators of Success 1. Operational electronic referral and screening tools that are standardized throughout the region. 2. The number of days for clients to be assessed as evidenced by the time between the order entry date and the referral date. 3. The number of days for clients to enter the Program as evidenced by the time between the referral date to the community and admission to the community. (This is manual and may not be accurate.) 4. The number of clients referred to the community that meet the Program's criteria compared to the number of clients that do not meet the criteria. DELIVERY OF MEDICATION AND CLIENT RESPONSE TO THERAPY Process Indicators of Success 1. Clinical practice standards meet professional standards and are consistent between acute care and the community. 2. ENCON forms will be signed off and followed up by the manager within 2 working days. 3. Review of financial reports for supplies, medications, and staffing actual costs against budgeted costs. 4. The number of clients who are missing medications, supplies, or equipment at the time their medications are due. 5. There are regionally standardized equipment and supplies for the Program. EDUCATION FOR CLIENT AND FAMILY Process Indicators of Success 1. Education material meets professional practice standards and best practice as evidenced through a community review of practice and a literature review. 2. Client assessment tools will be standardized and meet professional practice standards and best practice as evidenced through a community review of practice and a literature review. Resources Canadian Intravenous Nurses Guidelines for Practice. SFHR Standards Policies and Procedures for Patient Care. Infection Control Guidelines: a) CDC b) APIC c) SFHR Standards for Infection Control Registered Nurses Association of BC Standards for Practice College of Pharmacists Standards for practice Canadian Institute of Health Information Short Form 36 Health Survey Quality of Life Indicator Education Handouts and Tools: a) SFHR Home IV Program Manual prepared by SFHR Home IV Program: PE 456 revised August 2001, Adapted with permission from South Fraser Health region. b) Home Run 6060IV Pump prepared by L. Dowsley, CRN Community IVT Tri-Cities/New Westminster and D. Westendorp CRN Burnaby Home Health and adapted from Baxter/Sabratek Homerun 6060 Patient Information Manual (August 2001). Calea the Private provider for medications and IV supplies related to Community IVT 108 Appendix III Data Collection Tool for Most Responsible Diagnoses of Osteomyelitis and Cellulitis Using CIHI Data ^ _ C M G Code Most Responsible Diagnosis Sex Age LOS Readmit Code Readmit Code Description Chart # Disch Date 109 Appendix IV Patients in the Program Hospital Encounters Data collection for hospital encounters and readmissions of patients in the Program who were admitted to a hospital within the Fraser North during the time point of April 1, 2001 through to March 31, 2003. Location E A C O M / I V E R H Hosp -Assessment Clinic HACOM/IV R C H Hosp Assessment Clinic P H N Chart # Hospital Name Admit Date Disch Date C M G C M G Name LOS Age Sex 110 Appendix V Data Collection Tool for Program Patients Patients registered in the SFHR Community IV Program April 1, 2001 through to March 31, 2003. Location Admit Date PHN Chart # Age Sex EACOM/IV HACOM/IV I l l Appendix VI Data Collection Tool for Clinical Incidents Data Collection Tool for E N C O N Forms generated by clinical incidents in the community related to Program patients Date Gender Type of Incident Comment on form Degree of Injury Re-Admission 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0091332/manifest

Comment

Related Items