Elsevier

Annals of Vascular Surgery

Volume 34, July 2016, Pages 164-170
Annals of Vascular Surgery

Clinical Research
Comparison of Use of Short Form-36 Domain Scores and Patient Responses for Derivation of Preference-Based SF6D Index to Calculate Quality-Adjusted Life Years in Patients with Intermittent Claudication

Presented at the European Society of Surgical Research International Congress, Geneva, Switzerland, June 2010.
https://doi.org/10.1016/j.avsg.2015.12.028Get rights and content

Background

The short form 36 (SF36) questionnaire is used for assessment of generic quality of life. Responses to the individual question in SF36 are also used for calculation of the SF6D index score. This score is used for calculation of quality adjusted-life years (QALYs) in economical analyses. As the individual patient questionnaires are not always available for performing systematic reviews and meta-analyses, a new formula has been developed for derivation of SF6D index score from the reported SF36-domain scores. This study aimed to evaluate the validity of this formula for use in patients with intermittent claudication.

Methods

A retrospective review of a prospectively collected database of a randomized controlled trial was performed. A total of 178 patients were recruited. Clinical indicators of ischemia were recorded. All patients completed SF36 questionnaires. Response and domain-based SF6D scores (R-SF6D and D-SF6D) and QALYs were calculated. Correlation and agreement analysis were performed.

Results

Response rate was 88% (n = 781) over a 1-year follow-up period. Domain-based SF6D score (mean, 0.684; standard deviation [SD] 0.110) was significantly higher (paired t-test, P = 0.001) than the response-based score (mean, 0.627; SD, 0.110) with a mean difference of 0.056 (95% confidence interval, 0.053–0.060). Mean QALY calculated using D-SF6D score (0.503; SD, 0.116) was also significantly higher than the QALY calculated from the R-SF6D score (0.467; SD, 0.121). Bland-Altman comparison showed strong agreement (limit of agreement −0.167 to 0.054) between the 2 methods with equal variances (Pitman's test, P = 0.629). D-SF6D scores showed stronger correlation with clinical indicators of ischemia (r = 0.246–0.602) compared with that of R-SF6D scores (r = 0.233–0.549).

Conclusions

Domain-based estimation of SF6D score is a valid and reliable method with strong agreement to the gold standard response–based scores in claudicants. However, adjustments may be required in studies using a mixture of D-SF6D and R-SF6D scores for QALY calculation.

Introduction

Peripheral vascular disease affects 5% of UK population aged older than 55 years.1 Intermittent claudication (IC) is the commonest presentation characterized by tightness on walking in the calf, thigh, or buttock muscles that is relieved by rest and returns on walking again. IC is initially treated with best medical treatment including smoking cessation, antiplatelet treatment, lipid-lowering agents, and modification of risk factors. Further treatment options include supervised exercise programs (SEP), percutaneous transluminal angioplasty (PTA), vasodilator medication, and surgery.2 Clinical indicators such as ankle brachial pressure indices, walking distances, restenosis rates, reintervention rates, and limb loss rates have been used for outcome reporting in clinical studies and randomized trials involving patients with IC. However, the primary goal of treatment for patients with IC is an improvement in their quality of life (QOL).3

Improvement in the QOL has been the main aim of treatment for patients with claudication for a long time.3, 4, 5 Historically, clinical trials for patients with peripheral vascular disease reported QOL in secondary outcomes.6, 7 But this trend has changed and more recent trials are including QOL outcomes as the primary end point.8, 9, 10 There are many instruments available to measure health-related QOL outcomes and preference-based health profiles in general population and patient groups.11 Most of these instruments have been used in patients with IC.12 Both generic and disease-specific QOL instruments have been validated for use in claudicants.4, 13 Short form 36 (SF36) is the most common instrument used for generic QOL measurement.4, 14 SF36 produces a profile of health reporting 8 domain scores and 2 summary scores. SF36 profile and summary scores have been widely reported in claudication trials, systematic reviews, and meta-analyses.5, 6, 8, 9, 10, 15, 16, 17

A further instrument was subsequently developed by the restructuring of SF36 to yield a single index score known as SF6D.18 It was developed for the measurement of utilities and generation of quality-adjusted life years (QALYs) that are used in economical analyses for comparison of treatments. However, individual responses from the completed SF36 questionnaires were required to generate the SF6D score. This posed a problem for subsequent reviews and meta-analyses as retrieval of individual patient responses from historical studies is extremely challenging, and the data are often unavailable. Therefore, clinicians, researchers, and economists are dependent on the data reported in literature. As these studies routinely present the domain scores of SF36 in their results, a domain-based method for consolidation and synthesis of new data can prove to be invaluable. Recently, a new formula has been developed for derivation of SF6D score from SF36 domain scores obviating the need for individual patient responses.19 This formula has not been validated for use in patients with IC. This study was designed to test this new formula and compare it with the gold standard method of SF6D score generation in these patients.

Section snippets

Materials and Methods

The study was conducted in the Academic Vascular Surgical Unit of a tertiary-care university hospital. It was designed as a retrospective review of a prospectively collected database and was approved by the local research ethics committee. Patients were recruited as a part of a randomized controlled trial of PTA, SEP, and combined treatment (PTA + SEP).8, 20, 21 All patients with IC due to femoropopliteal atherosclerotic disease were included in the study. Patients with critical limb ischemia

Results

The median age for study participants was 70 years (interquartile range, 63–75 years). Sixty-one percent (n = 108) of patients were males. The overall response rate for completion of SF36 was 88% (n = 781) over 1 year. The clinical indicators, SF6D scores, and QALYs are summarized in Table I. D-SF6D score was higher than the R-SF6D score with a mean difference of 0.056 (95% confidence interval [CI] 0.053 to 0.060), and this was statistically significant (Paired t test, P < 0.001). QALYs

Discussion

Different preference-based instruments result in significantly different results when calculating QALYs.29, 30, 31, 32, 33 This has to be taken into account by the funding bodies when performing cost-utility analyses for comparison of different clinical problems and population groups.31 National Institute of Health and Care Excellence recommends this approach for health technology appraisals and has standardized the use of EuroQoL EQ-5D index score in their comparative analyses.34 EQ-5D and

Conclusions

Estimation of SF6D score using the domains of SF36 questionnaire (D-SF6D) is a valid and reliable method for use in patients with IC. Response-based estimation (R-SF6D) remains the gold standard instrument and should be used if appropriate data are available. However, D-SF6D score can be reliably used for pooling of data in systematic reviews, retrospective analyses, meta-analyses, and calculation of QALYs in economical analyses where individual responses are not available. Adjustments may be

References (41)

  • F.A. Mazari et al.

    Early outcomes from a randomized, controlled trial of supervised exercise, angioplasty, and combined therapy in intermittent claudication

    Ann Vasc Surg

    (2010)
  • J. Brazier et al.

    The estimation of a preference-based measure of health from the SF-36

    J Health Econ

    (2002)
  • S.A. Kharroubi et al.

    Modelling SF-6D health state preference data using a nonparametric Bayesian method

    J Health Econ

    (2007)
  • J.M. Bland et al.

    Comparing methods of measurement: why plotting difference against standard method is misleading

    Lancet

    (1995)
  • C.A. Marra et al.

    Not all “quality-adjusted life years” are equal

    J Clin Epidemiol

    (2007)
  • J.A. Kopec et al.

    A comparative review of four preference-weighted measures of health-related quality of life

    J Clin Epidemiol

    (2003)
  • M. Joore et al.

    The impact of differences in EQ-5D and SF-6D utility scores on the acceptability of cost-utility ratios: results across five trial-based cost-utility studies

    Value Health

    (2010)
  • J. Hanmer

    Predicting an SF-6D preference-based score using MCS and PCS scores from the SF-12 or SF-36

    Value Health

    (2009)
  • F.G. Fowkes et al.

    Edinburgh Artery Study: prevalence of asymptomatic and symptomatic peripheral arterial disease in the general population

    Int J Epidemiol

    (1991)
  • R.M. Greenhalgh et al.

    The adjuvant benefit of angioplasty in patients with mild to moderate intermittent claudication (MIMIC) managed by supervised exercise, smoking cessation advice and best medical therapy: results from two randomised trials for stenotic femoropopliteal and aortoiliac arterial disease

    Eur J Vasc Endovasc Surg

    (2008)
  • Cited by (3)

    Financial support: Academic Vascular Surgical Unit, University of Hull, Hull, UK (no external funding).

    View full text