Technical ReportAssessment of health-related quality of life in spine treatment: conversion from SF-36 to VR-12
Introduction
Patient-reported outcomes (PROs) are essential to the evaluation of spine treatment. The Food and Drug Administration (FDA) requires that the PROs collected in Investigational Device Exemption trials include disease-specific and health-related quality of life (HRQoL) assessments [1]. Although several HRQoL are available, the Medical Outcomes Study (MOS) Short Form 36 (SF-36) has been the most frequently used HRQoL for spine surgery. The FDA specifically mentions the SF-36 in its recommendations for spine Investigational Device Exemption and an article devoted to the SF-36 was published in a spine journal [2]. Despite the creation of the SF-12 (a shorter version of the SF-36) and its validation for spine surgery [3], many spine studies and FDA clinical trials have continued to rely on the SF-36. As a result, SF-36 scores have accumulated for many years on large patient samples.
As will be described later, the short form surveys have a proprietary and a public domain version. The proprietary version uses the trademarked SF-36 and SF-12 names, whereas the public version (with copyright by Trustees of Boston University) is denoted by VR-36 and VR-12. Converting SF-36 to Veterans RAND 12 Item Health Survey (VR-12) scores would enhance the comparability of study results for several reasons. The VR-12 is included in the Medicare Health Outcomes Survey, which has sampled Medicare beneficiaries annually since 1998 [4]. The conversion between VR-12 and PROMIS Global Health is already available [5]. Recently, the Centers for Medicaid & Medicare Services (CMS) has recommended the use of PRO measures that are validated, non-proprietary, and relatively short [6]. The VR-12 was specifically recommended as HRQoL measure in the orthopedic field and will be used by CMS as part of the “Comprehensive Care for Joint Replacement Bundled Payment Model” under “Meaningful Use.”
The conversion of PRO scores to current measures creates several opportunities: the continuity of outcomes in long-term studies, the use of the converted outcomes as historical controls for current studies, and the contribution of HRQoL measures to comparative effectiveness research. Comparative effectiveness research compares the benefits and harms of different diagnostic, treatment, and monitoring methods [7]. Comparative effectiveness research converts HRQoL measures into utility scores, which are included in cost-effectiveness analyses. Two utility indices, the EuroQol instrument (EQ-5D) and the SF-36 derived SF-6D, have been used to assess the comparative effectiveness of the Spine Patient Outcomes Research Trial (SPORT) [8]. SF-6D scores have been predicted from spine-specific measures such as the Oswestry Disability Index and the Neck Disability Index [9], [10]. A utility index, the VR-6D, is also available for the VR-12 and has been found to be similar to the SF-6D [11].
The surveys yield eight scales (vitality, physical functioning, bodily pain, general health perceptions, physical role functioning, emotional role functioning, social role functioning, and mental health) and two summary measures: Physical Component Summary (PCS) and Mental Component Summary (MCS). Researchers at the Boston University School of Public Health [12] have developed and made available algorithms for the conversion of SF-36 to VR-12 scales and summary measures. The purpose of the present study is to investigate the conversion of the SF-36 to VR-12 PCS and MCS scores. Specifically, the present study will compare the average preoperative to postoperative score change, as well as the proportion of patients reaching the minimum clinically important difference (MCID) obtained with the SF-36 and VR-12, respectively. Second, the impact of missing answers will be assessed by comparing the VR-12 scores computed with and without imputation of missing answers (an imputation algorithm is included in the SF-36 to VR-12 conversion algorithms). Finally, SF-36 and VR-12 scores will be directly compared with each other to confirm the necessity of converting SF-36 scores to render them comparable with VR-12 scores.
Section snippets
Sample
The SF-36 scores used in this sample were collected from patients who underwent lumbar or cervical surgery from a single surgeon between August 1998 and January 2013. The sample includes all patients who had completed the SF-36 preoperatively and at 2-year postoperatively or 1-year postoperatively if the 2-year postoperative survey was missing.
PCS and MCS scoring
For the present study, SF-36 PCS and MCS scores were calculated following their original instructions [13], [14]. The SF-36 answers were then converted
Results
A total of 1,968 patients (1,559 lumbar and 409 cervical) had completed preoperative and postoperative SF-36. The postoperative SF-36 consisted of 1-year (n=946) and 2-year (n=1,022) intervals. Results are reported for combined 1-year and 2-year SF-36 as separate analyses yielded similar results. Incomplete answers to the survey resulted in up to 21 unscored SF-36 and up to 129 unscored VR-12, whereas imputing missing values captured all but 2 VR-12 scores. In all comparisons with the SF-36,
Discussion
These results indicate that the converted VR-12 scores are strongly correlated with the original SF-36 scores. When evaluating the treatment effect of spine surgery, the converted VR-12 scores, similar to the SF-36 scores, detect a significant postoperative improvement in PCS and MCS scores. The calculated MDC values and the proportions of patients whose score improvement reach MDC are similar for both SF-36 and VR-12. However, SF-36 scores should not be compared with VR-12 scores without
Conclusions
The conversion of SF-36 to VR-12 scores has been made possible by the development of a public domain algorithm. This score conversion facilitates the adoption of the VR-12 which will contribute to comparative effectiveness research and help participation in CMS regulatory schemes.
Acknowledgment
The authors would like to thank Professor Lewis Kazis for his help and support of this work.
References (34)
- et al.
The minimum clinically important difference in lumbar spine surgery patients. A choice of methods using the Oswestry Disability Index, MOS Short Form 36, and pain scales
Spine J
(2008) - et al.
Understanding the minimum clinically important difference: a review of concepts and methods
Spine J
(2007) Guidance document for the preparation of IDEs for spinal systems
(2000)SF-36 health survey update
Spine
(2000)- et al.
Reliability, validity, and responsiveness of the short form 12-item survey (SF-12) in patients with back pain
Spine
(2003) Health Outcomes Survey (HOS)
- et al.
Linking Physical and Mental Health Summary Scores from the Veterans RAND 12-Item Health Survey (VR-12) to the PROMIS((R)) Global Health Scale
J Gen Intern Med
(2015) Patient-Reported Outcomes Summit for Total Joint Arthroplasty Report
J Arthroplasty
(2015)- et al.
Comparative effectiveness research in spine surgery
Neurosurg Focus
(2012) - et al.
Comparative effectiveness evidence from the spine patient outcomes research trial: surgical versus nonoperative care for spinal stenosis, degenerative spondylolisthesis, and intervertebral disc herniation
Spine
(2011)
Predicting SF-6D utility scores from the Oswestry disability index and numeric rating scales for back and leg pain
Spine
Predicting SF-6D utility scores from the neck disability index and numeric rating scales for neck and arm pain
Spine
A preference-based measure of health: the VR-6D derived from the veterans RAND 12-Item Health Survey
Qual Life Res
VR-36, VR-12, and VR-6D
36-Item Short Form Survey (SF-36)
SF-36 Physical and mental health summary scales: a user's manual
Imputing the physical and mental summary scores (PCS and MCS) for the MOS SF-36 and the Veterans SF-36 Health Survey in the presence of missing data
Cited by (0)
FDA device/drug status: Not applicable.
Author disclosures: MFG: Royalties: RTI (F); Stock Ownership: Bonovo (D), International Spine & Orthopedic Institute, LLC (F); Nocimed (A); OuroBoros (C); Viscogliosi Bros Venture Partners LLC (F); Consulting: K2M (E); Medtronic (E); Aesculap (E), outside the submitted work. AGC: Nothing to disclose. KMS: Nothing to disclose. FWS: Nothing to disclose.
The disclosure key can be found on the Table of Contents and at www.TheSpineJournalOnline.com.
The Veterans RAND 12-Item Health Survey (VR-12) was developed from the Veterans RAND 36-Item Health Survey (VR-36) which was developed and modified from the original RAND version of the 36-item Health Survey version 1.0 (also known as the “MOS SF-36”).
SF-36 and SF-12 are registered trademarks of the Medical Outcomes Trust. The VR-36 and the VR-12 are copyrighted by the Trustees of Boston University.