Reliability and Validity of the Medical Outcomes Study Short Form-12 Version 2 (SF-12v2) in Adults with Non-Cancer Pain

Abstract

Limited evidence exists on how non-cancer pain (NCP) affects an individual's health-related quality of life (HRQoL). This study aimed to validate the Medical Outcomes Study Short Form-12 Version 2 (SF-12v2), a generic measure of HRQoL, in a NCP cohort using the Medical Expenditure Panel Survey Longitudinal Files. The SF Mental Component Summary (MCS12) and SF Physical Component Summary (PCS12) were tested for reliability (internal consistency and test-retest reliability) and validity (construct: convergent and discriminant; criterion: concurrent and predictive). A total of 15,716 patients with NCP were included in the final analysis. The MCS12 and PCS12 demonstrated high internal consistency (Cronbach's alpha and Mosier's alpha > 0.8), and moderate and high test-retest reliability, respectively (MCS12 intraclass correlation coefficient (ICC): 0.64; PCS12 ICC: 0.73). Both scales were significantly associated with a number of chronic conditions (p < 0.05). The PCS12 was strongly correlated with perceived health (r = 0.52) but weakly correlated with perceived mental health (r = 0.25). The MCS12 was moderately correlated with perceived mental health (r = 0.42) and perceived health (r = 0.33). Increasing PCS12 and MCS12 scores were significantly associated with lower odds of reporting future physical and cognitive limitations (PCS12: OR = 0.90 95%CI: 0.89-0.90, MCS12: OR = 0.94 95%CI: 0.93-0.94). In summary, the SF-12v2 is a reliable and valid measure of HRQoL for patients with NCP.

Keywords: SF12-v2; non-cancer pain; psychometric properties; reliability; validity.

Figure 1

Derivation of study sample.

Figure 2

The SF-12 Physical and Mental Component Scores (PCS12 and MCS12) of the study sample, categorized by number of chronic conditions.