UBC Faculty Research and Publications

The validity of the SF-12 and SF-6D instruments in people living with HIV/AIDS in Kenya Patel, Anik R; Lester, Richard T; Marra, Carlo A; van der Kop, Mia L; Ritvo, Paul; Engel, Lidia; Karanja, Sarah; Lynd, Larry D Jul 17, 2017

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

Item Metadata


52383-12955_2017_Article_708.pdf [ 807.6kB ]
JSON: 52383-1.0348888.json
JSON-LD: 52383-1.0348888-ld.json
RDF/XML (Pretty): 52383-1.0348888-rdf.xml
RDF/JSON: 52383-1.0348888-rdf.json
Turtle: 52383-1.0348888-turtle.txt
N-Triples: 52383-1.0348888-rdf-ntriples.txt
Original Record: 52383-1.0348888-source.json
Full Text

Full Text

RESEARCH Open AccessThe validity of the SF-12 and SF-6Dinstruments in people living with HIV/AIDSTrial registration: Clinical trials.gov identifier: NCT00830622. Registered 26 January 2009.Patel et al. Health and Quality of Life Outcomes  (2017) 15:143 DOI 10.1186/s12955-017-0708-7Westbrook Mall, Vancouver, CanadaFull list of author information is available at the end of the article1Department of Medicine, University of British Columbia, 828 West 10thAvenue, Vancouver, BC V5Z 1M9, Canada2Faculty of Pharmaceutical Science, University of British Columbia, 2405* Correspondence: anikpa@mail.ubc.caKeywords: Quality of life, Short-form 12, Kiswahili, HIV, Health state utility, SF6Dan assessment tool for physical health, mental health andAnik R. Patel , Richard T. Lester , Carlo A. Marra , Mia L. van der Kop , Paul Ritvo , Lidia Engel ,Sarah Karanja8 and Larry D. Lynd2,9AbstractBackground: Health-related quality of life (HRQoL) and health state utility value (HSUV) measurements are vitalcomponents of healthcare clinical and economic evaluations. Accurate measurement of HSUV and HRQoL requirevalidated instruments. The 12-item Short-Form Health Survey (SF-12) is one of few instruments that can evaluateboth HRQoL and HSUV, but its validity has not been assessed in people living with HIV/AIDS (PLWHA) in east Africa,where the burden of HIV is high.Methods: This cross-sectional study used baseline data from a randomized trial involving PLWHA in Kenya. Dataincluded responses from a translated and adapted SF-12 survey as well as key demographic and clinical data.Construct validity of the survey was examined by testing the SF-12’s ability to distinguish between groups knownin advance to have differences in their health based on their disease severity. We classified disease severity basedon established definitions from the US Center for Disease Control (CDC) and WHO, as well as a previously studiedviral load threshold. T-tests and ANOVA were used to test for differences in HRQoL and HSUV scores. Area underthe receive operator curve (AUC) was used to test the discriminative ability of the HRQoL and HSUV instruments.Results: Differences in physical component scores met the minimum clinically important difference amongparticipants with more advanced HIV when defined by CD4 count (4.3 units) and WHO criteria (compared to stage1, stages 2, 3 and 4 were 2.0, 7.2 and 9.8 units lower respectively). Mental score differences met the minimumclinically important difference between WHO stage 1 and stage 4 patients (4.4). Differences in the HSUV werestatistically lower in more advanced HIV by all three definitions of severity. The AUC showed poor to weakdiscriminatory ability in most analyses, but had fair discriminatory ability between WHO clinical stage 1 and clinicalstage 4 individuals (AUC = 0.71).Conclusion: Our findings suggest that the Kiswahili translated and adapted version of the SF-12 could be used asHSUV for Kiswahili-speaking PLHWA.1,2* 1 3 1,4 5 6,7in Kenya© The Author(s). 2017 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, andreproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link tothe Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.most East African nations [7], this validation is critic-domized controlled trial (RCT) in Nairobi, KenyaPatel et al. Health and Quality of Life Outcomes  (2017) 15:143 Page 2 of 9BackgroundMeasures of health-related quality of life (HRQoL) andhealth state utility value (HSUV) measurements are vitalcomponents of healthcare program and technologyevaluations. HRQoL is a multi-dimensional construct ofan individual or group’s perceived health status, whileHSUV ranks societal preferences for various states ofhealth [1, 2]. HRQoL is used to measured functionalchanges in health as a clinical outcome of health inter-ventions while HSUV describes the relative value that asociety places on living in this health state. Although thetwo measures are related, they are theoretically distinctin their derivation, application and interpretation. Accur-ate measurement of HSUV and HRQoL require vali-dated instruments. The limited number of validatedinstruments in East Africa has impeded studies evaluat-ing the health or economic impact of new treatments orprograms.The 12-item Short-Form Health Survey (SF-12) is oneof few instruments that can be used to evaluate bothHRQoL and HSUV [3, 4]; however, it has not been vali-dated for use in East Africa. The SF-12 measures eightdimensions of health to derive a physical componentsummary (PCS) and a mental health component sum-mary (MCS). Further, an algorithm has been developedthat converts SF-12 survey data into the preference-based Short Form 6D (SF-6D) score. The SF-6D pro-vides information about the HSUV (based on the SF-12),which can be used to calculate quality-adjusted life years(QALY) [1]. The SF-12 and SF-6D are commonly usedto collect HRQoL and HSUV for health technology eval-uations in resource-rich settings, but they are rarely usedin East Africa.The SF-12 HRQoL scores provide descriptive mea-sures of individual health, but not measures of economicvalue. The SF-12 PCS and MCS scores cannot be useddirectly to calculate quality-adjusted life years (QALY).Societal preferences of various health states are neededfor cost-effectiveness evaluation of new health technolo-gies, programs and interventions. Societal preferences ofthe general UK population have been elicited for anumber of health states generated by the SF-12 using atime-trade off method [1]. An algorithm has been createdbased on these preferences to generate the SF-6D score, avalue between 0.35 and 1. The SF-6D instrument scoresare typically used directly in cost-effectiveness evaluations,and it is one of the most widely used intruments toestimate QALYs [1, 5].HIV/AIDS is a progressive disease that results in acomplex array of health states ranging from asymp-tomatic to severe opportunistic infection or HIVwasting syndrome. Additionally, antiretroviral ther-apy (ART) is associated with adverse events that canimpact an individuals HRQoL. Kenya is an East(N = 538) (ClinicalTrials.gov number, NCT00830622)[8]. Baseline data were collected prior to initiatingART or receiving the intervention. Data from partici-pants in both trial arms were pooled to conduct theseanalyses. This multi-site trial involved three HIVclinics located in demographically and ethnographic-ally diverse settings [8].ParticipantsInclusion criteria were ART naïvety, aged 18 years orabove, access to a mobile phone, and the ability to textmessage or have somebody who could text message ontheir behalf. Individuals who met the inclusion criteriaand consented to participate were randomized toeither receive a cell-phone based adherence interven-tion or standard care only. The study protocol wasapproved by the University of Manitoba and KenyattaNational Hospital ethics review boards [8]. The samplesize calculation was based on primary trial outcomesincluding Viral load and adherence. While the trialwas not specifically powered to measure the secondaryHRQoL outcomes, a post-hoc sample size calculationally important for future use of the SF-12 to assess bothhealth and economic outcomes. The objective of thisstudy is to examine the performance of the Kiswahili-translated and adapted SF-12 survey, and the correspond-ing SF-6D scores, in Kenya. Particularly, the discriminativeability will be evaluated between well-defined severitygroups in a sample of PLWHA.MethodsStudy design and settingThis cross-sectional study, which took place betweenMay 2007 and October 2009, used data from a ran-African nation that has been seriously impacted byHIV with an estimated 1.5 million (1.3 million – 1.8million) PLWHA in 2015 [6]. Because the SF-12 candescribe a large range of health states, it can be aparticularly useful tool to evaluate the health ofpeople living with HIV/AIDS (PLWHA) at all stagesof the disease. To date, the discriminative abilities ofthe PCS, MCS and SF-6D have not been investigatedby HIV severity in East Africa. Given the twodistinct purposes of these instruments (i.e., measur-ing health (SF-12) versus valuing health (SF-6D)), avalidation of both instruments is needed regardlessof the administration of the same questionnaire.Since HIV/AIDS is the leading cause of death inrevealed the sample was adequate to detect the MCIDdifferences.Patel et al. Health and Quality of Life Outcomes  (2017) 15:143 Page 3 of 9Data and measuresThe variables were defined at study entry, which tookplace at ART initiation. Individuals had been receivingcare, but were ART naïve at the time of data collec-tion. A translated and adapted SF-12 version onesurvey was administered to participants at baselinealong with a survey that collected data on gender,age, income and rural/urban residence. The SF-12was administered on the same day that the WHOstage, CD4 count and viral load measures were taken.CD4 count was collected (FACScan, Becton Dickinson,Sunnyvale, CA, USA) as part of routine clinical care andviral load (Amplicor, Roche Diagnostics, Mannheim,Germany) was assessed as part of the trial protocol [8].Research clinicians administering the baseline surveyassessed the World Health Organization (WHO) clinicalstage of HIV infection [8].Theoretical foundationA longer form of the SF-12, the SF-36, has beentranslated and adapted for use in 40 countries as partof the International Quality of Life Assessment(IQOLA) project [9]. Kiswahili, the primary languagein many East African nations, was not among theoriginal IQOLA project translations. However, twosubsequent studies (Wagner et al. and Wyss et al.)translated and evaluated a Kiswahili translated SF-36survey [10, 11]. Wagner et al. evaluated content, qual-ity and scaling of the translated survey in a generalKenyan population, demonstrating that the SF-36 sur-vey performed comparably to the UK counterpart[10]. Wyss et al. extended this work by assessing thevalidity of the SF-36 using a method of known groupvalidation [11]. They demonstrated that the SF-36could discriminate health status between groups withknown differences in health based on theory orevidence. The discriminative ability of a HRQoLsurvey is an important validation step to ensure thesurvey can adequately capture outcomes of interest[12]. The SF-36 is cumbersome to administer in re-search settings, so the briefer SF-12 was created [3].The SF-12 has been shown to retain much of the de-scriptive ability and validity of the SF-36, but has notbeen validated in East Africa.Translation and adaptation processAn international team of healthcare professionals andresearchers translated the English SF-12 (Version 1)into Kiswahili based on IQOLA recommendations.The survey was reviewed by a multidisciplinary focusgroup of English and Kiswahili speaking healthcareproviders and researchers for relevance, ease ofunderstanding, and cultural appropriateness. Wherenecessary, items and response options were slightlymodified and culturally adapted to make the ques-tionnaire relevant and appropriate for use in a Ken-yan context. Literature reviews and expert opinionwere used to inform changes to the survey. For ex-ample, ‘climbing stairs’ in the original SF-12 waschanged to ‘climbing a hill’, based on a previous studyusing the SF-36 in Tanzania [10, 11]. After translatingthe survey into Kiswahili, it was back translated intoEnglish and assessed by a focus group of Englishspeaking healthcare researchers to ensure consistency.The survey was pre-tested on a sample of 20 Kenyanindividuals and healthcare staff to evaluate culturalappropriateness and understanding.ValidationWe investigated the construct validity of the survey usingknown group validation [11]. This method involvesdemonstrating that the PCS, MCS or SF-6D surveyscores are able to discriminate scores between groupsknown a priori to have differences in their health sta-tus. We used three established criteria to classify HIVseverity: CD4 cell count, viral load, and WHO clinicalstage of HIV infection.We hypothesized that the HRQoL and HSUV would belower in more advanced HIV disease stages independentlyof how severity was defined. Further, since HIV is pre-dominantly a physical disease, we hypothesized physicalscores would show greater differences than mental healthscores. Our specific hypotheses were: 1. MCS, PCS andSF-6D scores would be lower in individuals withCD4 < 200; MCS, PCS and SF-6D scores would be lowerin individuals with viral load >55,000 copies/ml; and MCS,PCS and SF-6D scores would be lower in individuals inWHO stages 2, 3 & 4 compared to individuals in WHOstage 1. Since WHO stage 1 individuals are asymptomatic,we suspected that there would be a bigger difference inHRQoL and HSUV between these individuals and moresymptomatic individuals [13].Severity threshold definitionsWe used the United States (US) Center for DiseaseControl (CDC) severity stages, based on CD4 cellcount, as our first definition of disease severity [14].Stage 1 includes individuals with a CD4 count ≥500cells/mm3; stage 2 includes individuals with a CD4count between 200 and 499 cells/mm3; and stage 3includes individuals with CD4 count <200 cells/mm3.The vast majority of individuals initiating ART haveCD4 near or below 350 cells/mm3, as that was theART treatment guidelines in Kenya at the time. Fur-ther, presentation to care with advanced HIV care hasbeen defined as having a CD4 count below 200 [15].To maintain an adequate sample in both groups, wedichotomized individuals above and below CD4 countdifferences between the two groups. For the WHO clin-different threshold scores as cut-offs. The area underPatel et al. Health and Quality of Life Outcomes  (2017) 15:143 Page 4 of 9of 200 cells/mm3, reflecting a comparison of individ-uals with advanced HIV infection to those withoutadvanced HIV infection.Our second definition of severity was based on aprevious US study that used viral load threshold toclassify individuals [12]. Viral load is associated withdisease progression: an increased viral load indicatesadvanced disease and predicts progression to AIDSor death [16]. We classified individuals above orbelow 55,000 copies/ml to assess differences in thescores and draw descriptive comparisons to theprevious US sample [12].Our third definition of severity was the WHO HIVclinical staging system, which is based on physicalsymptoms. The WHO clinical stages are particularlyuseful in limited-resource settings, as CD4 cell countsare not always available. Symptoms have been groupedinto four stages. Stage one individuals are asymptom-atic; stage two individuals have mild symptoms suchas rash or upper respiratory tract infections; stage 3individuals have moderate to severe symptoms such asunexplained chronic diarrhea for greater than 1 month;and stage 4 individuals have severe to life-threateningsymptoms such as extreme weight loss or opportunisticinfections.Based on our three definitions of severity, we cate-gorized our sample into two groups based on theirCD4 count or viral load threshold and four groupsaccording to WHO clinical stages. We assessed thePCS, MCS and SF-6D, compared scores between eachgroups, and determined the discriminative ability ofthe scores.Statistical analysisWe conducted a descriptive analysis of the baselinecharacteristics of the study population, and stratifiedthe results by the severity groups we defined. Wecalculated individual PCS and MCS scores using cor-related weights from the US and SF-6D scores basedon UK weights [1, 3, 17, 18]. The SF-12 was designedto give a population mean MCS and PCS of 50 with astandard deviation of 10 in a disease-free US popula-tion [3]. The minimum clinically significant difference(MCID) for both PCS and MCS scores has been sug-gested to be in the range 3–5 points; however, MCIDfor HRQoL scores are not well-established [19]. Weused a change of 3 to interpret the clinical significanceof differences that we observed, but caution is sug-gested in interpreting the MCID since a 1-pointchange can be meaningful if it came at no additionalcost [19]. The MCID for the SF6D has been suggestedto be 0.033 (95% CI 0.029 to 0.037) [20].We calculated mean PCS, MCS and SF-6D scores ineach of the severity categories. For CD4 and viral loadthe ROC curve (AUC) is a measure of signal to noiseof an instrument [21]. An AUC of 1 indicates perfectdiscriminatory ability; an AUC of between 0.8 to 1shows good to excellent ability to discriminate; anAUC of between 0.7 to 0.8 shows fair discriminativeability; an AUC of between 0.60 and 0.70 shows weakability to discriminate; an AUC below 0.60 indicates afailure to discriminate between groups; and an AUCof 0.50 suggests the instrument is no more useful topredict the group to which an individual belongs thanflipping a coin [21].ResultsThe sample had 538 participants, with greater representa-tion by females (n = 350/538; 65%) and urban residents(n = 436/538, 81%). Table 1 shows the characteristics ofour sample separated by severity category. CD4 count datawere complete; however, 9 (1.7%) participants had missingSF-12 responses; 43 (8.0%) were missing viral load data;and 72 (13.3%) were missing WHO clinical stage. Table 2summarizes the mean scores by severity group andTable 3 lists the AUC results of each score. We ob-served statistically and clinically significant differencesin PCS scores in several comparisons. The MCS had aweak signal in some comparisons, indicating that had amodest ability to discriminate across groups. The SF-6Dscores also show monotonic trends in the hypothesizeddirection in all analyses and there were statistically signifi-cant differences in several comparisons (Table 2).Results by CD4 count threshold severity definitionical stage analysis, we used analysis of variance analysis(ANOVA) with a post-hoc analysis to test for differencesin scores between the four groups. Participants withmissing CD4 counts, viral load or WHO stage wereexcluded from the respective analysis.We used receiver operator characteristic (ROC)curves as a second test of the discriminative ability ofthe instruments [12, 21]. Traditionally, a ROC plotsthe sensitivity by 1-specificity of a diagnostic test andhelps to determine the ability of the test to discrimin-ate between a diseased and non-diseased population.It has also previously been used to determine theconstruct validity of an instrument by evaluating ifthe instrument can correctly discriminate two groupsknown to have differing HRQOL [12]. We used ROCcurves to assess whether the scores could correctlycategorize a participant into a severity group usingthreshold analyses, t-tests were used to test for statisticalMean PCS and SF-6D scores were significantly lower in in-dividuals with CD4 < 200 cells/mm3 than in individualsTable 1 Characteristics of sample separated by severity categoryCD4 < 200N = 364N (%)CD4 ≥ 200N = 169N (%)VLa >55,000N = 281N (%)VLa ≤55,000N = 214N (%)Stage 1N = 114N (%)Stage 2N = 126N (%)Stage 3N = 204N (%)Stage 4N = 22N (%)Male Gender 136 (37) 51 (30) 114 (41) 62 (29) 30 (26) 48 (38) 72 (35) 7 (32)Age20–29 62 (17) 37 (22) 46 (16) 43 (20) 30 (26) 19 (15) 39 (19) 4(18)30–39 184 (51) 88 (52) 148 (53)40–49 89 (24) 32 (19) 68 (24)50+ 30 (8) 12 (7) 19 (7)Income (Schillings)≤ 2000 93 (29) 43 (29) 58 (23)Patel et al. Health and Quality of Life Outcomes  (2017) 15:143 Page 5 of 9above that threshold. The PCS was 4.3 units lower and theSF6D was 0.05 units lower suggesting a clinically significantdifference based on the MCID. The mean MCS score was2.4 units lower in individuals with CD4 < 200 cells/mm3, sothe difference was not clinically significant. We also com-pared mean values of PCS and MCS scores to a US sampleand scores from of our sample were comparable to the pre-viously reported estimates (Table 4) [12]. The AUC for allthree scores were in the weak to poor range (Figs. 1 and 2),indicating that they had some ability to distinguish these se-verity groups (0.57–0.61). Floor and ceiling effects were ob-served for the SF6D scores, but not for the PCS and MCS2001–10,000 140 (43) 71 (48) 114 (45)10,001–40,000 75 (23) 30 (20) 64 (25)> 40,000 14 (4) 5 (3) 15 (6)Urban Res. 295 (81) 139 (82) 238 (85)aVL viral loadscores (Fig. 3).Results by viral load thresholdThe SF-6D score was statistically significantly lower inindividuals with a viral load >55,000 copies/ml and metthe MCID difference we specified. The PCS and MCSTable 2 Mean HRQoL scores by severity subgroupSub Group PCS (SDa) MCS (SDa) SF6D (SDa)CD4 < 200 N = 364 41.1 (11.0)* 43.4 (10.7)* 0.67(0.15)*CD4 ≥ 200 N = 169 45.4 (10.3)* 45.8 (11.0)* 0.72(0.15)*Viral Load >55,000 N = 281 41.5 (10.6)* 43.8 (10.9) 0.67 (0.15)*Viral Load ≤55,000 N = 214 43.7 (11.3)* 44.5 (10.8) 0.71 (0.16)*WHO Stage 1 N = 114 46.7 (8.7)** 46.0 (11.0)** 0.73 (0.15)**WHO Stage 2 N = 126 44.7 (10.3) 44.6 (10.3) 0.71 (0.15)WHO Stage 3 N = 204 39.5 (11.3)** 42.7 (11.0) 0.66 (0.16)**WHO Stage 4 N = 22 36.9 (11.3)** 41.6 (10.2)** 0.61 (0.13)**aStandard Deviation*Statistically significant difference between severity group p < 0.05**Statistically significant difference between severity group p < 0.05 based onANOVA with post-hoc Tukey’s procedurescores were also lower, but were not clinically significantaccording to the MCID we specified. The average scoreswithin these viral load categories were comparable to apreviously reported US sample (Table 4). The AUC waspoor, indicating that the survey could not discriminatewell between these populations (Figs. 1 and 2).Results by WHO stageBoth the PCS and SF-6D had a statistically significantmonotonic downwards trend as severity increased. Thedifference in PCS scores between stage 1 and stages 2, 3and 4 was 2.0, 7.2 and 9.8 units respectively, indicating a104 (49) 63 (55) 59 (47) 95 (47) 10 (45)48 (22) 16 (14) 24 (19) 51 (25) 8 (36)19 (9) 5 (4) 3 (2) 19 (9) 0 (0)65 (35) 26 (27) 29 (25) 57 (32) 6 (30)80 (43) 41 (43) 59 (51) 84 (48) 4 (20)36 (19) 25 (26) 24 (20) 27 (15) 10 (50)4 (2) 3 (3) 3 (2) 8 (5) 0 (0)170 (79) 107 (94) 116 (92) 137 (67) 18 (82)clinically significant difference in physical health as HIVprogresses from stage 1 through 4. The AUC of the PCSand SF-6D were 0.71 and 0.68, respectively, indicating thatthe scores had fair discriminate ability between WHOstages one and four (Fig. 1). The MCS means were statisti-cally different between Stage 1 and Stage 4 patients andthe AUC by this comparison was 0.71 suggesting ability todiscriminate between these two groups.DiscussionOur study shows that HRQoL and HSUV scores de-rived from a Kenyan modified and translated SF-12Table 3 Area under the ROC curve comparisonsComparison Groups PCS AUC MCS AUC SF6D AUCCD4 < 200 vs CD4 ≥ 200 0.61 0.61 0.61Viral Load ≤55,000 vs >55,000 0.56 0.54 0.57WHO stage 1 vs stage 2 0.55 0.58 0.55WHO stage 1 vs stage 3 0.67 0.67 0.64WHO stage 1 vs stage 4 0.72 0.71 0.68survey can discriminate HIV disease severity usingthree severity definitions. These findings suggest con-struct validity of the modified SF-12 and may have im-portant implications for the use of the instrument inKenya and other east African nations. We confirmthat the SF-12 may be used as a tool to measure phys-ical and mental health as part of program and inter-vention evaluations. Furthermore, the SF-12 surveycan be scored to derive an SF-6D preference-basedmeasure that can be used to calculate QALYs. The SF-present or absence of a variety of symptoms based onHIV severity. Data were collected by highly trainedresearch nurses as part of an internationally fundedrandomized trial adding a level of scrutiny to datacollection and accuracy of classification. As would beexpected, we observed the largest differences in PCS,MCS and SF-6D between WHO stage 1 and stage 4.We were unable to find a similar comparison in theliterature, but the findings confirm the survey’s abilityto discriminate between groups known to have differ-Table 4 Comparison of mean scores to a US sample of HIV patientsPCS Kenya Mean (SDa) MCS Kenya Mean (SD) PCS USA [11] Mean (SD) MCS USA [11] Mean (SD)CD4 ≥ 200 cells/mm3 45.4 (10.3) 45.8 (11.0) 45.3(11.3) 42.6 (9.6)CD4 < 200 cells/mm3 41.1 (11.0) 43.4 (10.7) 40.1 (11.4) 43.3(9.8)Viral load ≤55,000 copies/ml 43.7 (11.3) 44.5 (10.8) 44.5 (11.6) 42.9 (9.5)Viral load >55,000 copies/ml 41.5 (10.6) 43.8 (10.9) 40.2 (11.5) 41.6 (10.2)aStandard DeviationPatel et al. Health and Quality of Life Outcomes  (2017) 15:143 Page 6 of 96D scores declined with increased severity of diseaseand could theoretically rank health states in a validorder in practice. These instruments could be particu-larly important to support the increasing demand formeasurement and evaluation of HIV/AIDS programs.Additionally, our results have described the mean anddistribution of HRQoL or HSUV scores for a variety ofHIV health states, and the results could be used inmathematical models to calculate QALYs, estimatedisease burden and/or conduct economic evaluationsin Kenya.The WHO stages were perhaps the most accurateindication of HRQoL since the system relies on theFig. 1 The PCS and SF-6D ROC curves when comparing WHO stage one to(AUC) is a measure of signal to noise of an instrument. The signal appearsincreases. This indicates discriminatory ability of both survey scores and givit was designed to measureences in HRQoL. We also observed differences ingroups by established CD4 and viral load thresholds.We were able to confirm the survey scores coulddiscriminate between these alternate classifications asfurther confirmation of the discriminatory ability ofthe survey scores. Our findings were strengthenedthrough the consistent findings across multiple cri-teria of HIV severity.Our results were consistent with previous studies ofHRQoL and HSUV in PLWHA. Delate et al. reportedmean SF-12 summary scores in a sample of USPLWHA [12]. The mean PCS and MCS scores we ob-served in a Kenyan population have mostly similarmore advanced stages. Caption: The area under the ROC curveto improve as the severity gap between the comparison groupses face validity to them since the survey is correctly measuring whatPatel et al. Health and Quality of Life Outcomes  (2017) 15:143 Page 7 of 9means and standard deviations as the US sample. In asystematic review of HIV/AIDS focused HSUV studies,Tengs et al. pooled utility values for three HIV healthstates: asymptomatic HIV; symptomatic HIV; andAIDS; they reported HSUVs of 0.94, 0.82 and 0.70respectively [13]. The mean HSUVs in our samplewere generally lower (0.61–0.73) than those reportedin the systematic review (Table 2). However, the re-view summarized evidence of HSUV of a broad sampleof PLWHA, while we assessed a cohort at a particu-larly vulnerable time: ART initiation. Within severityFig. 2 The ROC curves of SF-12 derived PCS and MCS using CD4 and viralpartly because of the more general definitions of severity. However, both Pthreshold comparisongroups, the average HSUV may have improved overtime due to adaptation to disease and due to drugtreatment [22].There were several limitations to this study. First,normative data from the United States was used tocalculate the PCS and MCS and scoring data from theUK was used to calculate the SF6D scores. Externalscoring was used due to a lack of a local scoring algo-rithm for the SF-12 or SF6D in Kenya or a similar set-ting. Previous studies in Africa have used scoring datafrom other settings as a surrogate to overcome thisload thresholds. Caption: The signal was weaker in this comparison,CS and MCS showed some signal by CD4 severityng ePatel et al. Health and Quality of Life Outcomes  (2017) 15:143 Page 8 of 9limitation, however future studies are needed to evalu-ate these important measures in Kenya and otherFig. 3 Histogram of survey scores. Caption: The PCS and MCS scores didthe SF6D may have had both a floor effect at a score of 0.3 and a ceilinAfrican settings [23, 24]. Second, we were missingWHO stage and viral load data for several partici-pants. We had an adequate sample size to show statis-tically significant differences between groups; however,the direction of the potential bias due to missing datais uncertain. Since the missing data may have beendue to administrative errors, there would likely be nosystematic pattern in missing individuals. Finally, thesurvey had been modified from its original questions, sotheoretical constructs may have been affected. The surveyappears to perform as designed in main scores derivedfrom the survey, but more nuanced measures of healthstatus were not assessed in this study.ConclusionWe found that a Kiswahili translated and adapted SF-12survey could discriminate between HIV severity groupsin Kenya. The SF-12 is widely used in clinical trials inthe US and Europe as an objective measure of HRQoLassociated with new drug therapies and health interven-tions. The SF-12 could accompany clinical trials beingconducted in Kenya and in other areas in East Africa tohelp quantify HRQoL and HSUV that have previouslygone unmeasured. Further research is needed to showthe ability of the SF-12 survey to detect changes in qual-ity of life over time as individuals’ health status changes.Further research is also needed to determine Kenyaspecific scoring for both the SF-12 and SF-6D instru-ments, and to test the survey in a broad range of diseases.ot appear to have any floor or ceiling effects in this sample. However,ffect at a score of 1This study is a fundamental step towards increased use ofthe SF-12 and other HRQoL instruments in east Africa.AbbreviationsAIDS: Acquired immunodeficiency syndrome; ART: Antiretroviral therapy;AUC: Area under the curve; HIV: Human immunodeficiency virus;HRQoL: Health related quality of life; MCID: Minimum clinically importantdifference; MCS: Mental component summary; PCS: Physical componentsummary; ROC: Receiver operating characteristic; SF 12: Short form 12questionnaire; SF6D: Short form 6 dimension score; US: United States;WHO: World Health OrganizationAcknowledgementsWe would like to acknowledge Maja Grubisic for her assistance in thisevaluation.FundingThis project was funded by the US President’s Emergency Plan for AIDSRelief, grant number PHEKE.07.0045. The funding agency had no role in thedesign or conduct of this research. Anik Patel was supported in part by theNational Institute of Mental Health of the National Institutes of Health undergrant number R01MH097558–01 and by the Canadian HIV Trials Network.Richard Lester was supported in part by a Michael Smith Foundation awardfor Health Research Scholars.Availability of data and materialsN/A.Authors’ contributionsAP conceived the study, participated in design of the study, conducted dataanalysis and drafted the final manuscript. RL Secured funding for the study,participated in design of the study and helped to draft the final manuscript.CM, MLV, PR, LL participated in design of the study and helped to draft thefinal manuscript. LE participated in data analysis and helped to draft the finalmanuscript. SK participated in data collection, participated in study designOutcomes. 2007;5:54.19. Hays RD, Morales LS. The RAND-36 measure of health-related quality of life.Ann Med. 2001;33:350–7.20. Walters SJ, Brazier JE. What is the relationship between the minimallyimportant difference and health state utility values? The case of the SF-6D.Health Qual Life Outcomes. 2003;1:1.21. The Area Under an ROC curve. [cited 2016 May 9]; Available from: http://gim.unmc.edu/dxtests/roc3.htm.22. Menzel P, et al. The role of adaptation to disability and disease in health statevaluation: a preliminary normative analysis. Soc Sci Med. 2002;55:2149–58.23. Badri M, et al. When to initiate highly active antiretroviral therapy in sub-Saharan Africa? A South African cost-effectiveness study. Antiviral Therapy.2006;11:63.24. Francois C, et al. Analysis of health-related quality of life and costs based ona randomised clinical trial of escitalopram for relapse prevention in patientswith generalised social anxiety disorder. Int J Clin Pract. 2008;62:1693–702.Patel et al. Health and Quality of Life Outcomes  (2017) 15:143 Page 9 of 9and helped to draft the final manuscript. All authors have read andapproved the final manuscript.Ethics approval and consent to participateThe study protocol was approved by the University of Manitoba andKenyatta National Hospital ethics review boards. Research Ethics wasapproved by the UBC Behavioural Research Ethics Board under applicationnumber H10–00392.Consent for publicationN/A.Competing interestsRichard T Lester is Executive and Scientific Director of the WelTelInternational mHealth Society and WelTel Incorporated, which develop andimplement mobile health solutions. The remaining authors declare they haveno competing interests.Author details1Department of Medicine, University of British Columbia, 828 West 10thAvenue, Vancouver, BC V5Z 1M9, Canada. 2Faculty of Pharmaceutical Science,University of British Columbia, 2405 Westbrook Mall, Vancouver, Canada.3School of Pharmacy, University of Otago, Dunedin, New Zealand.4Department of Public Health Sciences/Global Health (IHCAR), KarolinskaInstitutet, Widerströmska Huset, Tomtebodavägen 18A, 171-77 Stockholm,Sweden. 5School of Kinesiology & Health Sciences, York University, 160Campus Walk, North York, ON M3J 1P3, Canada. 6Faculty of Health Sciences,Simon Fraser University, 8888 University Drive, Burnaby, BC V5A 1S6, Canada.7Deakin University, Geelong Australia, Faculty of Health, School of Health andSocial Development, 221 Burwood Highway, Burwood, Victoria 3125,Australia. 8Monitoring, Evaluation and Research Unit, Amref Health Africa inKenya, P.O. Box 30125-00100, Nairobi, Kenya. 9Center for Health Evaluationand Outcomes Science, St. Paul’s Hospital, 588 – 1081 Burrard Street,Vancouver, BC V6Z 1Y6, Canada.Received: 14 October 2016 Accepted: 21 June 2017References1. Brazier JE, Roberts J. The estimation of a preference-based measure ofhealth from the SF-12. Med Care. 2004;42:851–9.2. Health-Related Quality of Life (HRQOL). [cited 2016 May 29]; Available from:http://www.cdc.gov/hrqol.3. Ware JE Jr, Kosinski M, Keller SD. A 12-Item Short-Form Health Survey:construction of scales and preliminary tests of reliability and validity. MedCare. 1996;34:220–33.4. Brazier J, Roberts J, Deverill M. The estimation of a preference-basedmeasure of health from the SF-36. J Health Econ. 2002;21:271–92.5. Räsänen P, et al. Use of quality-adjusted life years for the estimation ofeffectiveness of health care: a systematic literature review. Int J TechnolAssess Health Care. 2006;22:235–41.6. Country Report: Kenya. 2015 [cited 2016 10 January]; Available from: http://www.unaids.org/en/regionscountries/countries/kenya.7. GBD Compare. 2013 [cited 2016 May 25]; Available from: http://vizhub.healthdata.org/gbd-compare/.8. Lester RT, et al. Effects of a mobile phone short message service onantiretroviral treatment adherence in Kenya (WelTel Kenya1): a randomisedtrial. Lancet. 2010;376(9755):1838–45.9. Wagner AK, et al. Cross-cultural comparisons of the content of SF-36translations across 10 countries: results from the IQOLA project. J ClinEpidemiol. 1998;51:925–32.10. Wagner AK, et al. A Kiswahili version of the SF-36 health survey for usein Tanzania: translation and tests of scaling assumptions. Qual Life Res.1999;8:101–10.11. Wyss K, et al. Validation of the Kiswahili version of the SF-36 health surveyin a representative sample of an urban population in Tanzania. Qual LifeRes. 1999;8:111–20.12. Delate T, Coons SJ. The discriminative ability of the 12-item short formhealth survey (SF-12) in a sample of persons infected with HIV. Clin Ther.2000;22:1112–20.•  We accept pre-submission inquiries •  Our selector tool helps you to find the most relevant journal•  We provide round the clock customer support •  Convenient online submission•  Thorough peer review•  Inclusion in PubMed and all major indexing services Submit your next manuscript to BioMed Central and we will help you at every step:13. Tengs TO, Lin TH. A meta-analysis of utility estimates for HIV/AIDS. MedDecis Mak. 2002;22:475–81.14. Terms, definitions, and calculations used in CDC HIV surveillancepublications. 2012 [cited 2016 11 July]; Available from: http://www.cdc.gov/hiv/statistics/surveillance/terms.html.15. Antinori A, et al. Late presentation of HIV infection: a consensus definition.HIV Med. 2011;12:61–4.16. Mellors JW, et al. Plasma viral load and CD4+ lymphocytes as prognosticmarkers of HIV-1 infection. Ann Intern Med. 1997;126:946–54.17. Hays, R. Programs and Utilities. [cited 2015 Dec 10]; Available from: http://gim.med.ucla.edu/FacultyPages/Hays/utils/.18. Farivar SS, Cunningham WE, Hays RD. Correlated physical and mental healthsummary scores for the SF-36 and SF-12 health survey, V. 1. Health Qual Life•  Maximum visibility for your researchSubmit your manuscript atwww.biomedcentral.com/submit


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



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"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items