UBC Faculty Research and Publications

Injection Drug Use Quality of Life scale (IDUQOL): A validation study Hubley, Anita M; Russell, Lara B; Palepu, Anita Jul 19, 2005

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

Item Metadata

Download

Media
52383-12955_2005_Article_194.pdf [ 517.75kB ]
Metadata
JSON: 52383-1.0223085.json
JSON-LD: 52383-1.0223085-ld.json
RDF/XML (Pretty): 52383-1.0223085-rdf.xml
RDF/JSON: 52383-1.0223085-rdf.json
Turtle: 52383-1.0223085-turtle.txt
N-Triples: 52383-1.0223085-rdf-ntriples.txt
Original Record: 52383-1.0223085-source.json
Full Text
52383-1.0223085-fulltext.txt
Citation
52383-1.0223085.ris

Full Text

ralHealth and Quality of Life OutcomesssBioMed CentOpen AcceResearchInjection Drug Use Quality of Life scale (IDUQOL): A validation studyAnita M Hubley*1, Lara B Russell1 and Anita Palepu2,3Address: 1Measurement Evaluation and Research Methodology, Dept of ECPS, 2125 Main Mall, The University of British Columbia, Vancouver, BC, Canada, 2Division of Internal Medicine, Department of Medicine, Faculty of Medicine, University of British Columbia, Vancouver, BC, Canada and 3Centre for Health Outcome and Evaluation Sciences, St. Paul's Hospital, Vancouver, BC, CanadaEmail: Anita M Hubley* - anita.hubley@ubc.ca; Lara B Russell - brussell@interchange.ubc.ca; Anita Palepu - anita@hivnet.ubc.ca* Corresponding author    Drug UseFactor AnalysisPsychometricsQuality of LifeReliabilityValidityAbstractBackground: Existing measures of injection drug users' quality of life have focused primarily onhealth and health-related factors. Clearly, however, quality of life among injection drug users isimpacted by a range of unique cultural, socioeconomic, medical, and geographic factors that mustalso be considered in any measure. The Injection Drug User Quality of Life (IDUQOL) scale wasdesigned to capture the unique and individual circumstances that determine quality of life amonginjection drug users. The overall purpose of the present study was to examine the validity ofinferences made from the IDUQOL by examining the (a) dimensionality, (b) reliability of scores, (c)criterion-related validity evidence, and (d) both convergent and discriminant validity evidence.Methods: An exploratory factor analysis using principal axis factoring in SPSS 12.0 was conductedto determine whether the use of a total score on the IDUQOL was advisable. Reliability of scoresfrom the IDUQOL was obtained using internal consistency and one-week test-retest reliabilityestimates. Criterion-related validity evidence was gathered using variables such as stability ofhousing, sex trade involvement, high-risk injection behaviours, involvement in treatment programs,emergency treatment or overdose over the previous six months, hospitalization and emergencytreatment over the subsequent six month period post data collection. Convergent and discriminantvalidity evidence was gathered using measures of life satisfaction, self-esteem, and social desirability.Results: The sample consisted of 241 injection drug users ranging in age from 19 to 61 years.Factor analysis supports the use of a total score. Both internal consistency (alpha = .88) and one-week test-retest reliability (r = .78) for IDUQOL total scores were good. Criterion-related,convergent, and discriminant validity evidence supports the interpretation of IDUQOL total scoresas measuring a construct consistent with quality of life.Conclusion: The findings from this study provide initial evidence to support the use of theIDUQOL total score. The results of the study also suggest the IDUQOL could be furtherstrengthened with additional attention to how some IDUQOL domains are described andPublished: 19 July 2005Health and Quality of Life Outcomes 2005, 3:43 doi:10.1186/1477-7525-3-43Received: 04 May 2005Accepted: 19 July 2005This article is available from: http://www.hqlo.com/content/3/1/43© 2005 Hubley et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.Page 1 of 10(page number not for citation purposes)satisfaction is measured.Health and Quality of Life Outcomes 2005, 3:43 http://www.hqlo.com/content/3/1/43BackgroundExisting measures of injection drug users' (IDUs) qualityof life (QoL) have focused primarily on health and health-related factors. The Opiate Treatment Index, the onlystandardized instrument designed specifically for IDUs, isessentially a symptom checklist [1]. The NottinghamHealth Profile [2,3] focuses exclusively on health. TheMOS surveys (MOS SF-36, MOS SF-20 and MOS-HIV)have been used in IDU populations but because they areconstructed to measure the range of health in the generalpopulation (with the exception of MOS-HIV), IDUs scorevery poorly [4,5]. It makes intuitive sense that IDUs havelower physical and psychological health relative to thegeneral population. It is not surprising that IDU scorestend to be clustered at the low end of the distribution andthat instruments devised for the general population maynot be particularly sensitive to change in the IDU popula-tion. For example, some studies found that, in workingwith HIV patients with a history of injection drug use,some scales measuring the physical aspects of QoL wererelatively insensitive to change and that the effects of druguse tended to overshadow the impact of HIV on health[6,7]. Among crack smokers, most SF-36 subscales did notreflect the adverse health effects of crack cocaine use andtherefore appeared to have limited applicability with thispopulation [8].A meta-analysis of existing QoL studies indicated thatQoL and health status are distinct constructs that shouldnot be used interchangeably [9]. Of the instruments usedwith IDUs, only the MOS series examine QoL domainsother than health. With the exception of the MOS-HIV(which was adapted for use with HIV patients), the MOSdomains were chosen to measure QoL of the general pop-ulation. Existing QoL tools do not measure the QoL ofdrug users in a culturally-sensitive fashion [10]. Problemsarise with both the item content and methods of admin-istration. These measures clearly do not take into accountthe full complexity of drug dependence or account for theindividual factors that may compromise effective admin-istration. The social context in which drug injectors live islikely a key component of their QoL and most measuresdo not capture the chronic long-term impact of drug useon diverse domains such as social, psychological, physicaland occupational realms [11]. Even instruments such asthe MOS-HIV that are devised for HIV-infected individu-als are often not applicable to actively using IDUs becausethe effects of drug use tend to overshadow the impact ofHIV [6].QoL assessment continues to be widely used in clinical tri-als and observational studies of health and disease to eval-uate clinical interventions, treatment side effects, andcomponents of an individual's life that are deemed criticalto his/her QoL are needed. Clearly, QoL among IDUsencapsulates a range of unique cultural, socioeconomic,political, medical, and geographic factors that must beconsidered in measuring QoL. With these considerationsin mind, the Injection Drug User Quality of Life (IDU-QOL) scale, an instrument that captures the unique andindividual circumstances that determine QoL amongIDUs, was developed [13]. To be able to use an instrumentwith confidence, however, one needs to be able to provideevidence of the validity – that is, the meaningfulness, use-fulness, and appropriateness – of the inferences to bemade from scores obtained on the instrument with agiven population and in a given context [14-16]. The over-all purpose of the present study was to examine the valid-ity of inferences made from the IDUQOL. Several lines ofconstruct validity evidence were examined: (a) essentialunidimensionality supporting use of an IDUQOL totalscore, (b) internal consistency and test-retest reliability ofIDUQOL scores, (c) criterion-related validity evidence,and (d) both convergent and discriminant validityevidence.MethodsSampleParticipants consisted of a sub-sample of individuals par-ticipating in the Vancouver Injection Drug User Study(VIDUS), a longitudinal study of the incidence of HIVamong IDUs in Vancouver, Canada. The research designand methods of the VIDUS have been previouslydescribed [17]. In brief, this open cohort study was initi-ated in 1996 to clarify the socio-demographic and behav-ioural determinants of HIV sero-conversion among thisgroup. Eligibility for initial enrolment required currentinjection drug use (injected at least once within the lastmonth) and evidence of recent injection was required byinspection of needle tracks. Potential participants alsowere required to reside in the Lower Mainland of BritishColumbia and provide informed consent. Most partici-pants (82%) were recruited through word of mouth andstreet outreach programs. The remaining participants werereferred by the needle-exchange program (5%), otherstorefront agencies (10%), and clinics (3%). Participantswho have stopped injecting after the baseline visit are stilleligible for follow-up. Trained interviewers administer asurvey instrument every 6 months. Participants are askedabout their demographics, needle sharing, drug usingbehaviour, sexual behaviours, access to clean needles andsyringes, access to health care, service needs, and medicalservice use (e.g., self-reported visits to primary care/outpa-tient clinics, Emergency Department, detoxification,methadone maintenance, ambulance use and hospitaladmissions). Participants were reimbursed $20 CDN forPage 2 of 10(page number not for citation purposes)disease impact over time [12]. It has become evident thatpopulation-sensitive approaches that consider the manyeach study visit, at which time referrals were provided formedical care, HIV/AIDS care, available drug and alcoholHealth and Quality of Life Outcomes 2005, 3:43 http://www.hqlo.com/content/3/1/43treatment and counselling as needed. The VIDUS studyparticipants may not be representative of all IDUs becausethose in the lowest socioeconomic group are overrepre-sented in this study sample. However, it is this group thatis most in need of innovative interventions.In the present study, a total of 250 individuals wererecruited in the order in which they appeared for their reg-ularly scheduled appointment for VIDUS. A subsequentappointment for the quality of life study was scheduledand participants were paid $10 CDN in each session of thepresent study. Data from nine participants were excludedbecause of missing data or because they were deemed, atthe time of data collection, to be too impaired to focus onthe research tasks. The final sample consisted of 241 IDUsranging in age from 19 to 61 years (M = 39.4, SD = 9.5years). There were more males (63%) than females (37%)and most participants (85%) had completed high school.There were no significant socio-demographic or drugusing behaviour differences between the 250 recruitedindividuals and the other VIDUS participants.The first 50 participants were invited to return for a secondsession within 6–8 days to collect test-retest reliabilitydata. All 50 participants returned for the second session asscheduled. In the test-retest group, 58% were male and42% were female. These participants ranged in age from22 to 59 years (M = 41.7, SD = 9.2). Most of the partici-pants (90%) had completed high school.MeasuresInjection Drug User Quality of Life Scale (IDUQOL)The present version of the IDUQOL consists of 21 lifedomains and builds on the original version first pub-lished by Brogly et al. [13]. Many of these domains (e.g.,Being Useful, Drugs, Drug Treatment, Harm Reductionand Neighbourhood Safety) are particularly relevant tothe physical, social, psychological, occupational, and geo-graphical reality of IDUs' lives. Life domains are each rep-resented on a 5 by 5 inch card, with the name of thedomain and a simple representative picture on the frontof the card and a description of the domain on the back ofthe card. Graphic representations were used so that thismeasure would be more accessible to respondents who donot speak English as a first language or have low literacyskills.Although the administration of the IDUQOL permittedrespondents to subjectively weight the importance of thelife domains to his/her quality of life, a review of the liter-ature on importance ratings and weighting [18] as well asan empirical comparison of the utility of weighted versusunweighted scores with the IDUQOL showed that weight-present study, wherein the respondent simply assigned asatisfaction rating for each life domain using a 7-pointLikert-type scale ranging from 1 (very dissatisfied) to 7(very satisfied) and illustrated with seven stylised frown-ing and smiling faces. Again, visual representation wasincluded as a guide for respondents with limited Englishor literacy skills. Domain scores were summed and aver-aged to obtain an overall quality of life score ranging from1 (very dissatisfied) to 7 (very satisfied).Satisfaction with Life Scale (SWLS)The SWLS is a 5-item global measure of life satisfaction[20]. Scores range from 5 to 35, with higher scores repre-senting greater life satisfaction. This measure was selectedbecause life satisfaction was seen as a related construct toquality of life.Rosenberg's Self-Esteem Scale (RSES)The RSES is a 10-item measure of global self-esteem [21].Total scores range from 10 to 40, with higher scores repre-senting greater self-esteem. This measure was selectedbecause self-esteem was seen as a related construct toquality of life.Marlowe-Crowne Social Desirability Scale Short Form X2 (MC X2)The MC X2 [22] is a 10-item short form version of theMarlowe-Crowne Social Desirability Scale (MC SDS) [23].Strahan and Gerbasi reported that it correlates .80 orhigher with the MC SDS. The MC X2 provides an estimateof socially desirable responding as a potential source ofmeasurement error. Total scores range from 0 to 10, withhigher scores representing higher social desirability inresponding. The MC X2 was selected because measures ofpervasive characteristics such as social desirability arestrongly recommended to assess discriminant validity[24,25].Demographic InformationIn examining criterion-related validity, the followingdemographic variables were used to create groupsexpected to differ in their quality of life: stability of hous-ing, sex trade involvement, high-risk injection behaviours(i.e., lending or borrowing needles, daily use of heroin,cocaine, speed, or crack), involvement in a methadonemaintenance program or drug treatment program, report-ing hospitalization and emergency department attend-ance or overdose within the previous six months.Predictive criterion variables included: hospitalizationand emergency treatment over the six-month period post-data collection. All variables were measured and codeddichotomously.Page 3 of 10(page number not for citation purposes)ing does not improve upon the use of simpler unweightedscores [19]. Thus, unweighted scores are used in theHealth and Quality of Life Outcomes 2005, 3:43 http://www.hqlo.com/content/3/1/43ProceduresEthics approval for this study was obtained from the Uni-versity of British Columbia and Providence Health CareResearch Ethics Boards. Participants met one-on-one withone of three trained VIDUS staff members for a single ses-sion lasting approximately 25–30 minutes. Participantswere identified only by their VIDUS study ID code on allresearch forms utilized for this project. All participantsprovided informed consent and then completed the studymeasures in the same order (IDUQOL, MC X2, SWLS,RSES). Demographic and predictive criterion data wereobtained from VIDUS using the participants' VIDUS IDcodes, a use that was disclosed to participants as part oftheir informed consent. Retest sessions for the sub-sampleof 50 participants followed the same consent process, for-mat, and tasks as the initial session.ResultsEssential unidimensionality and use of IDUQOL total scoreTo be able to use a summed total score on a measure suchas the IDUQOL, it is important to demonstrate that themeasure shows either strict or essential unidimensionality[26,27]. Strict unidimensionality denotes the presence ofa single common factor whereas essential unidimension-ality indicates the presence of a reasonably dominantcommon factor along with secondary minor dimensions[28,29].An exploratory factor analysis using principal axis factor-ing in SPSS 12.0 was conducted on the 21 items of theIDUQOL to determine whether essential unidimensional-ity was present and supported the use of the IDUQOLtotal score. According to Gorsuch's guideline of 5 to 10cases per item [30], the sample size for the present study(n = 241) was considered adequate for factor analysis ofthe 21-item IDUQOL. The data met the criteria for Bar-tlett's Test of Sphericity, χ2(210) = 1488.02, p < 0.001 andthe Kaiser-Meyer-Olkin criteria for sampling adequacy,KMO = .88 [31]. The first factor had an eigenvalue of 6.40and explained 30.5% of the variance in participants'responses. The ratio of the first to the second eigenvaluewas 4.3 which exceeds the strict criterion of a ratio greaterthan 4.0 for evidence of unidimensionality [30,32,33].These results, in addition to a visual examination of thescree plot (see Figure 1) indicated an essentially unidi-mensional factor structure for the IDUQOL, which sup-ported the use of a total score [32-35]. Factor loadingsranged from .31 to .71 for all IDUQOL items on a singlefactor.Mean performance and reliabilityTable 1 displays the inter-item correlation matrix for theIDUQOL. The mean inter-item correlation was .26, whichscores obtained by the sample on the IDUQOL, SWLS,RSES, and MC X2. Given the focus in the present study onthe IDUQOL, gender differences on scores from thismeasure were also examined. No statistically significantdifferences in performance on the IDUQOL were foundbetween men (M = 4.25, SD = 0.96) and women (M =4.10, SD = 1.02), t (239) = 1.14, p = .20, and the effect size(d = 0.15) is considered small according to Cohen [37].In addition to an internal consistency reliability estimate,the one-week test-retest reliability estimate for the IDU-QOL scores was also computed. Based on the sub-sampleof 50 participants who completed the measure twice, thetest-retest reliability estimate was .78, with correlations foreach domain across the two sessions ranging from .32 to.67. Table 3 shows the test-retest correlations for alldomains.Criterion-related validity evidenceTable 4 shows the correlations of the IDUQOL total scoreswith the dichotomously scored criterion variables. Of thestatistically significant correlations, all were in theexpected direction. That is, lower IDUQOL scores wererelated to unstable housing, sex trade involvement, bor-rowing and lending needles, daily use of heroin andspeed, and overdose in the past six months. The IDUQOLscores did not correlate significantly with daily use ofcocaine or crack, methadone or drug treatment, emer-gency treatment, or hospitalization within the six monthsprior to, or following, the initial test session.IDUQOL scores are based on a wide range of domainsthat encompass social, physical and emotional realms,and therefore, as a total score, might not correlate signifi-cantly with specific criterion variables. To explore thispossibility, analyses were carried out at the domain level,matching available criterion variables with relevant IDU-QOL domains. For example, the criterion variables ofengaged in sex trade and Rosenberg Self-Esteem Scalescores were correlated with the Feeling Good about Your-self IDUQOL domain score. Table 5 shows the correla-tions of selected IDUQOL domain scores andcorresponding criterion variables.Convergent and discriminant validity evidenceTable 6 shows the correlations of IDUQOL total scoreswith the SWLS, RSES, and MC X2. The convergent meas-ures (SWLS, RSES) showed moderately high correlationswith the IDUQOL as would be expected between con-structs that are related but not the same. The correlationbetween the IDUQOL and the discriminant measure (MCX2) was in the low to moderate range and thus acceptable[38]. As expected, the convergent measures were bothPage 4 of 10(page number not for citation purposes)Clark and Watson [36] describe as acceptable. Table 2shows the mean performance and internal consistency ofmore highly correlated with the IDUQOL total score thanwas the discriminant measure. Correlations were alsoHealth and Quality of Life Outcomes 2005, 3:43 http://www.hqlo.com/content/3/1/43conducted between the MC X2 and both the SWLS (r =.35) and RSES (r = .41). Because the relationship betweenthe IDUQOL and the convergent measures could be dueto the common influence of social desirability bias, partialcorrelations between the IDUQOL total scores and theSWLS and RSES, controlling for MC X2 scores, were con-ducted. These are reported in Table 6.DiscussionThe IDUQOL was developed to be a more appropriateand sensitive measure of quality of life for IDUs withintheir unique context of social, psychological, physical,occupational, and geographical factors. This study wasdesigned to examine the construct validity of inferencesmade from the IDUQOL by exploring the factor structure,reliability, criterion-related validity evidence, and conver-gent and discriminant validity evidence. The exploratoryfactor analysis using principal axis factoring indicates thepresence of essential unidimensionality, which, in turn,consistency and one week test-retest reliability estimatesfor the IDUQOL total score were satisfactory.Criterion-related validity evidence for inferences madefrom IDUQOL total scores is weak. That is, althoughlower IDUQOL total scores were statistically significantlyrelated to unstable housing, involvement in the sex trade,borrowing and lending needles, daily use of heroin andspeed, and overdose in the previous six months, the corre-lations were low (r = -.14 to -.26). Moreover, IDUQOLtotal scores did not correlate significantly with daily use ofcocaine or crack, methadone or drug treatment, emer-gency treatment within the previous six months, or hospi-talization within the following six months (r = -.12 to.07). These results may not be too surprising, however,given that the IDUQOL measures numerous life domains.When specific criterion variables were correlated withindividual IDUQOL domains, some showed considerablyScree Plot Showing Eigenvalues for Each Possible Factor of the IDUQOLFigure 1Scree Plot Showing Eigenvalues for Each Possible Factor of the IDUQOL.1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21Factor Number01234567EigenvaluePage 5 of 10(page number not for citation purposes)supports the use of a total score for the IDUQOL. Internal stronger correlations (e.g., Rosenberg Self-Esteem Scalecorrelated .58 with the Feeling Good about YourselfHealth and Quality of Life Outcomes 2005, 3:43 http://www.hqlo.com/content/3/1/43domain; instability of housing correlated -.30 with theHousing domain; drug treatment program and metha-done treatment correlated .21 and .19, respectively, withthe Drug Treatment domain). The fact that the correla-tions between other criterion variables and specificdomains did not change appreciably or even declined(e.g., daily use of specific drugs with the Drugs domain)suggests that there may be some lack of consistency inhow participants interpreted the IDUQOL domains. Forexample, when rating their level of satisfaction with theDrugs domain, it is not clear whether individual partici-pants may have indicated dissatisfaction because of a lacklow or near-zero correlations between criterion variablesand some IDUQOL domain ratings. In other cases, corre-lations may be low because of low variability (e.g., mor-tality) or reduced information (e.g., dichotomous (yes/no) rather than continuous (actual number) measure-ment of overdoses in previous six months) in the criterionvariables. In future criterion-related validity researchinvolving the IDUQOL, some criterion variables mayneed to be measured differently to improve the variabilityin scores.These results suggest that improvements can be made toTable 1: Inter-item Correlations on the IDUQOLItemsBU DR DT ED FA FG FR HR HE HC HO IN LA MO NS PA RC SX SP TRBU 1.00DR 0.27 1.00DT 0.14 0.37 1.00ED 0.36 0.13 0.14 1.00FA 0.35 0.12 0.15 0.13 1.00FG 0.54 0.43 0.26 0.25 0.34 1.00FR 0.31 0.22 0.20 0.28 0.35 0.52 1.00HR 0.21 0.21 0.22 0.17 0.10 0.15 0.15 1.00HE 0.29 0.38 0.25 0.13 0.25 0.43 0.29 0.13 1.00HC 0.17 0.20 0.40 0.19 0.23 0.29 0.27 0.18 0.35 1.00HO 0.23 0.30 0.21 0.16 0.30 0.33 0.40 0.10 0.40 0.37 1.00IN 0.39 0.34 0.31 0.25 0.28 0.48 0.37 0.22 0.31 0.40 0.31 1.00LA 0.45 0.35 0.23 0.28 0.22 0.38 0.37 0.22 0.16 0.30 0.31 0.31 1.00MO 0.35 0.26 0.16 0.30 0.20 0.34 0.33 0.14 0.28 0.31 0.28 0.31 0.41 1.00NS 0.31 0.18 0.24 0.19 0.22 0.32 0.42 0.26 0.29 0.36 0.36 0.37 0.34 0.34 1.00PA 0.22 0.14 0.18 0.12 0.26 0.29 0.27 0.04 0.18 0.14 0.26 0.24 0.15 0.11 0.21 1.00RC 0.18 0.09 0.18 0.24 0.17 0.20 0.29 0.25 0.12 0.27 0.13 0.26 0.25 0.26 0.24 0.11 1.00SX 0.24 0.18 0.12 0.14 0.36 0.29 0.33 0.06 0.26 0.19 0.25 0.24 0.24 0.23 0.22 0.60 0.20 1.00SP 0.27 0.25 0.28 0.25 0.29 0.45 0.37 0.17 0.21 0.16 0.18 0.29 0.30 0.24 0.17 0.18 0.11 0.20 1.00TR 0.29 0.18 0.19 0.34 0.19 0.27 0.34 0.14 0.21 0.22 0.34 0.30 0.34 0.36 0.25 0.10 0.28 0.19 0.21 1.00TO 0.39 0.29 0.29 0.24 0.25 0.46 0.50 0.13 0.39 0.37 0.34 0.43 0.35 0.31 0.40 0.17 0.24 0.18 0.31 0.30Items: BU = Being Useful; DR = Drugs; DT = Drug Treatment; ED = Education; FA = Family; FG = Feeling Good; FR = Friends; HR = Harm Reduction; HE = Health; HC = Health Care; HO = Housing; IN = Independence; LA = Leisure Activities; MO = Money; NS = Neighbourhood Safety; PA = Partner(s); RC = Resources in the Community; SX = Sex; SP = Spirituality; TR = Transportation; TO = Treatment by Others.Table 2: Mean Performance and Reliability on the IDUQOL, MC X2, SWLS, and RSESPossible Score Range Actual Score Range Mean (Standard Deviation) Internal ConsistencyIDUQOL 0 – 7 1.9 – 6.7 4.19 (0.98) .88SWLS 5 – 35 5 – 32 14.44 (7.17) .85RSES 10 – 40 11 – 40 27.39 (4.96) .82MC X2 0 – 10 0 – 10 4.53 (2.10) .62IDUQOL = Injection Drug User Quality of Life Scale, SWLS = Satisfaction with Life Scale, RSES = Rosenberg Self Esteem Scale, MC X2 = Marlowe-Crowne Social Desirability Short Form X2. Internal consistency reliability estimates were obtained using Cronbach's coefficient alpha.Page 6 of 10(page number not for citation purposes)of availability of drugs or because of the impact of drugsin their lives. As a result, this lack of clarity may producehow (a) some IDUQOL domains are described, and (b)satisfaction is measured that would strengthen the utilityHealth and Quality of Life Outcomes 2005, 3:43 http://www.hqlo.com/content/3/1/43of this measure. Individual qualitative interviews withIDUs to explore how individuals are interpreting the IDU-QOL domains and assigning satisfaction ratings wouldprovide important guidance on the types of modificationstively as an outcome measure in intervention studies inwhich programs addressing specific aspects of quality oflife (e.g., housing, health) are evaluated.Table 3: One Week Test-Retest Reliability Estimates for the IDUQOL Domain and Total ScoresIDUQOL Domain Reliability EstimateBeing Useful .60**Drugs .59**Drug Treatment .32*Education .44**Family .43**Feeling Good .67**Friends .66**Harm Reduction .47**Health .44**Health Care .44**Housing .63**Independence .57**Leisure Activities .62**Money .55**Neighbourhood Safety .65**Partner(s) .64**Community Resources .34*Sex .52**Spirituality .57**Transportation .59**Treatment by Others .49**IDUQOL Total Score .78*** p <.05, ** p <.01; N = 50Table 4: Correlations of IDUQOL Total Scores with Criterion MeasuresCriterion Variable IDUQOL Total ScoreHousing (stable = 0/ unstable = 1) -.16*Engaged in sex trade (no = 0/yes = 1) -.17**Currently borrowing needles (no = 0/yes = 1) -.19**Currently lending needles (no = 0/yes = 1) -.25**At least once daily use of heroin (no = 0/yes = 1) -.26**At least once daily use of cocaine (no = 0/yes = 1) -.11At least once daily use of speed (no = 0/yes = 1) -.14*At least once daily use of crack (no = 0/yes = 1) -.12Currently on methadone treatment (no = 0/yes = 1) .07Drug treatment program in last 6 months (no = 0/yes = 1) .01Overdose in last 6 months (no = 0/yes = 1) -.14*Visited ER in last 6 months (no = 0/yes = 1) -.06Hospitalized in last 6 months (no = 0/yes = 1) -.06Visited ER in subsequent 6 months (no = 0/yes = 1) -.05Hospitalized in subsequent 6 months (no = 0/yes = 1) .01*p < .05, **p < .01; N = 241Page 7 of 10(page number not for citation purposes)to be made. More importantly, further considerationneeds to be given to how the IDUQOL can be used effec-Convergent and discriminant validity evidence for theIDUQOL was strong. Convergent measures (SWLS, RSES)Health and Quality of Life Outcomes 2005, 3:43 http://www.hqlo.com/content/3/1/43correlated more highly with the IDUQOL total scoresthan was the discriminant measure (MC X2). The moder-ate (r = .54 to .59) correlations between the IDUQOL totalscores and the measures of related, but not identical, con-structs of life satisfaction and self-esteem are to beexpected. The finding of a significant but low moderatecorrelation of .35 between the IDUQOL total scores andthe MC X2 provides evidence to support discriminantvalidity but also suggests social desirability plays somerole in participants' responses. A similar relationship wasfound between the MC X2 and both the SWLS and RSES.ence of social desirability bias, partial correlationsbetween the IDUQOL total scores and the SWLS andRSES, controlling for MC X2 scores, were examined. Theresults showed that, although the magnitude of these cor-relations declined slightly, the relationships between theIDUQOL and both the SWLS and RSES were not due tosocial desirability bias.ConclusionThe findings from this study provide preliminary evidenceto support the meaningfulness, usefulness, and appropri-Table 5: Correlations of Selected IDUQOL Domain Scores with Selected Criterion VariablesCriterion Variable IDUQOL DomainsDrugsAt least once daily use of heroin (no = 0/yes = 1) -.13At least once daily use of cocaine (no = 0/yes = 1) -.07At least once daily use of speed (no = 0/yes = 1) -.12At least once daily use of crack (no = 0/yes = 1) -.07Drug TreatmentCurrently on methadone treatment (no = 0/yes = 1) .19**Drug treatment program in last 6 months (no = 0/yes = 1) .21**Feeling Good About YourselfRosenberg Self Esteem Scale .58**Engaged in sex trade (no = 0/yes = 1) -.20**HealthCurrently on methadone treatment (no = 0/yes = 1) .06Drug treatment program in last 6 months (no = 0/yes = 1) .01Visited ER in last 6 months (no = 0/yes = 1) -.14*Hospitalized in last 6 months (no = 0/yes = 1) -.16*Health CareVisited ER in last 6 months (no = 0/yes = 1) .03Hospitalized in last 6 months (no = 0/yes = 1) .003HousingHousing (stable = 0/ unstable = 1) -.30**How Others Treat YouEngaged in sex trade (no = 0/yes = 1) -.15**p < .05, **p < .01; N = 241Table 6: Correlations and Partial Correlations of IDUQOL Total Scores With Convergent and Discriminant MeasuresCorrelations with IDUQOL Total Scorea Partial Correlations with IDUQOL Total ScorebConvergent MeasuresSatisfaction With Life Scale .59** .54**Rosenberg Self Esteem Scale .54** .47**Discriminant MeasureMarlowe-Crowne Social Desirability Scale (MC X2).35****p < .01; aN = 241; bPartial correlations control for MC X2 scores, N = 238Page 8 of 10(page number not for citation purposes)Because the relationship between the IDUQOL and theconvergent measures could be due to the common influ-ateness of inferences made from IDUQOL total scores.Factor analysis supports the use of a total score. BothHealth and Quality of Life Outcomes 2005, 3:43 http://www.hqlo.com/content/3/1/43internal consistency (Cronbach alpha = .88) and one-week test-retest reliability (r = .78) for IDUQOL totalscores are good. Convergent and discriminant validityevidence supports the interpretation of IDUQOL totalscores as measuring a construct consistent with quality oflife and yet distinctive from life satisfaction, self-esteem,and social desirability bias. The criterion-related validityevidence is weak, but also suggests that the utility of theIDUQOL could be further improved with greater atten-tion to how some IDUQOL domains are described, howsatisfaction is measured, and how the IDUQOL and itsdomains may be applied in both the development andevaluation of various interventions (e.g., drug treatmentprograms, health and clinical interventions, and socialprograms).List of abbreviationsIDUQOL injection drug user quality of life scaleQoL quality of lifeIDUs injection drug usersVIDUS Vancouver Injection Drug User StudyHIV Human immunodeficiency virusSWLS Satisfaction with Life ScaleRSES Rosenberg's Self-Esteem ScaleMC SDS Marlowe-Crowne Social Desirability ScaleMC X2 Marlowe-Crowne Social Desirability Scale ShortForm X2Authors' contributionsAH obtained funding, designed the study, directed the sta-tistical analyses, prepared the initial draft of the manu-script and conducted revisions. LR assisted in preparingthe data, performed statistical analyses and assisted withrevisions. AP conceived of the study, obtained funding,coordinated data collection, and conducted revisions ofthe manuscript. All authors read and approved the finalmanuscript.AcknowledgementsThis research was supported by an operating grant from the Canadian Insti-tutes of Health Research (CIHR) to Dr. Anita Palepu and Dr. Anita Hubley. Additional support was provided through a Canadian Institutes for Health Research New Investigator Award and a Michael Smith Foundation for Health Research Senior Scholar Award to Dr. Anita Palepu. The authors would like to thank Kathy Li, Nancy Laliberte, Dave Isham, and Robin Brooks and the participants at the Vancouver Injection Drug User Study for References1. Darke S, Hall W, Wodak A, Heather N, Ward J: Development andvalidation of a multidimensional instrument for assessingoutcome of treatment among opiate users: Opiate Treat-ment Index.  British Journal of Addiction 1992, 87:733-742.2. Torrens M, San L, Martinez A, Castillo C, Domingo-Salvany A, AlonsoJ: Use of the Nottingham Health Profile for measuring healthstatus of patients in methadone maintenance treatment.Addiction 1997, 92:707-716.3. Puigdollers E, Domingo-Salvany A, Brugal MT, Torrens M, Alvaros J,Castillo C, Magri N, Martin S, Vazquez JM: Characteristics of her-oin addicts entering methadone maintenance treatment:Quality of life and gender.  Substance Use & Misuse 2004,39:1353-1368.4. Ryan CF, White JM: Health status at entry to methadone main-tenance treatment using the SF-36 health surveyquestionnaire.  Addiction 1996, 91:39-45.5. Stein MD, Mulvey KP, Plough A, Samet JH: The functioning andwell being of persons who seek treatment for drug and alco-hol use.  Journal of Substance Abuse 1998, 10:75-84.6. Carretero MD, Burgess AP, Soler P, Soler M, Catalan J: Reliabilityand validity of an HIV-specific health-related quality-of-lifemeasure for use with injecting drug users.  AIDS 1996,10:1699-1705.7. Dalgard O, Egeland A, Skaug K, Vilimas K, Steen T: Health-relatedquality of life in active injecting drug users with and withoutchronic hepatitis C virus infection.  Hepatology 2004, 39:74-80.8. Falck RS, Wang J, Carlson RG, Siegal HA: Crack-cocaine use andmental health status as defined by the SF-36.  AddictiveBehaviors 2000, 25:579-584.9. Smith KW, Avis NE, Assmann SF: Distinguishing between qualityof life and health status in quality of life research: A meta-analysis.  Quality of Life Research 1999, 8:447-459.10. Metzger DS, O'Brien CP: Substance Abuse: The Challenge ofAssessment.  In Quality of Life Assessments in Clinical Trials Edited by:Spilker B. New York, Raven Press, Ltd.; 1990:237-246. 11. Koch T: Life quality vs the 'quality of life': Assumptions andunderlying prospective quality of life instruments in healthcare planning.  Social Science and Medicine 2000, 51:419-427.12. Globe DR, Hays RD, Cunningham WE: Associations of clinicalparameters with health-related quality of life in hospitalizedpersons with HIV disease.  AIDS Care 1999, 17:71-86.13. Brogly S, Mercier C, Bruneau J, Palepu A, Franco E: Towards moreeffective public health programming for injection drug users:Development and evaluation of the Injection Drug UserQuality of Life Scale.  Substance Use & Misuse 2003, 38:965-992.14. APA, AERA, NCME: Standards for educational and psychologi-cal testing.  Washington, DC, American Psychological Association,American Educational Research Association, National Council onMeasurement in Education; 1985. 15. Hubley AM, Zumbo BD: A dialectic on validity: Where we havebeen and where we are going.  J Gen Psychol 1996, 123:207-215.16. Messick S: Validity of test interpretation and use. ResearchReport No. 90-11.  Educational Testing Service; 1990. 17. Strathdee SA, Patrick DM, Currie SL, Cornelisse PG, Rekart ML,Montaner JS, Schechter MT, O'Shaughnessy MV: Needle exchangeis not enough: lessons from the Vancouver injecting drug usestudy.  AIDS 1997, 11:F59-65.18. Russell LB, Hubley AM: Importance ratings and weightings: Oldconcerns and new perspectives.  International Journal of Testing2005, 5:105-131.19. Russell LB, Hubley AM, Palepu A, Zumbo BD: Does weighting cap-ture what's important? Revisiting subjective importanceweighting with a quality of life measure.  Social IndicatorsResearch . in press20. Diener E, Emmons RA, Larsen RJ, Griffin S: The Satisfaction WithLife Scale.  Journal of Personality Assessment 1985, 49:71-74.21. Rosenberg M: Society and the adolescent self-image.  Princeton,NJ, Princeton University Press; 1965. 22. Strahan R, Gerbasi KC: Short, homonegeous versions of theMarlow-Crowne Social Desirability Scale.  Journal of ClinicalPsychology 1972, 28:191-193.23. Crowne DP, Marlowe D: A new scale of social desirability inde-Page 9 of 10(page number not for citation purposes)their assistance in collecting and preparing the data for this research. pendent of psychopathology.  Journal of Consulting Psychology 1960,24:349-354.Publish with BioMed Central   and  every scientist can read your work free of charge"BioMed Central will be the most significant development for disseminating the results of biomedical research in our lifetime."Sir Paul Nurse, Cancer Research UKYour research papers will be:available free of charge to the entire biomedical communitypeer reviewed and published immediately upon acceptancecited in PubMed and archived on PubMed Central Health and Quality of Life Outcomes 2005, 3:43 http://www.hqlo.com/content/3/1/4324. Campbell DT: Recommendations for APA test standardsregarding construct, trait and discriminant validity.  AmericanPsychologist 1960, 15:546-553.25. Foster SL, Cone JD: Validity issues in clinical assessment.  Psy-chological Assessment 1995, 7:248-260.26. Nandakumar R, Ackerman T: Test modeling.  In The SAGE Hand-book of Quantitative Methodology for the Social Sciences Edited by: Kap-lan D. Thousand Oaks, CA, SAGE Publications; 2004:93-105. 27. Pedhazur EJ, Schmelkin LP: Measurement and scientific inquiry.In Measurement, design, and analysis: An integrated approach Hillsdale,NJ, Lawrence Erlbaum Assoc. Pub.; 1991. 28. Stout W: A nonparametric approach for assessing latent traitunidimensionality.  Psychometrika 1987, 52:589-617.29. Stout WF: A new item response theory modeling approachwith applications to unidimensionality assessment and abil-ity estimation.  Psychometrika 1990, 55:293-325.30. Gorsuch RL: Factor analysis.  2nd edition. Hillsdale, NJ, LawrenceErlbaum; 1983. 31. Pett MA, Lackey NR, Sullivan JJ: Making sense of factor analysis:The use of factor analysis for instrument development inhealth care research.  Thousand Oaks, CA, SAGE Publications;2003. 32. Hattie J: Methodology review: Assessing unidimensionality oftests and items.  Applied Psychological Measurement 1984, 20:1-14.33. Hattie J: An empirical study of the various indices for deter-mining unidimensionality.  Multivariate Behavioral Research 1985,19:49-78.34. Fabrigar LR, Wegener DT, MacCallum RC, Strahan EJ: Evaluatingthe use of exploratory factor analysis in psychologicalresearch.  Psychological Methods 1999, 4:272-299.35. Russell DW: In search of underlying dimensions: The use (andabuse) of factor analysis in Personality and Social PsychologyBulletin.  Pers Soc Psychol Bull 2002, 28:1629-1646.36. Clark LA, Watson D: Constructing validity: Basic issues inobjective scale development.  Psychological Assessment 1995,7:309-319.37. Cohen J: A power primer.  Psychological Bulletin 1992, 112:155-159.38. Netemeyer RG, Bearden WO, Sharma S: Scaling procedures:issues and applications.  Thousand Oaks, CA, SAGE Press; 2003. yours — you keep the copyrightSubmit your manuscript here:http://www.biomedcentral.com/info/publishing_adv.aspBioMedcentralPage 10 of 10(page number not for citation purposes)

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.52383.1-0223085/manifest

Comment

Related Items