Quality of life scores using SF-12 and SF-36v2 questionnaires for patients with multidrug-resistant tuberculosis in Vietnam

TL;DR

This study compares two questionnaires for measuring quality of life in patients with drug-resistant tuberculosis in Vietnam, finding that the shorter version overestimates scores in some groups.

Contribution

The study evaluates SF-12's performance against SF-36v2 for quality of life in MDR-TB patients and identifies overestimation in specific demographics.

Findings

SF-12 scores were higher than SF-36v2 for both physical and mental component summaries.

SF-12 showed moderate correlation with SF-36v2 but tended to overestimate quality of life in older and male patients.

The AUC values indicated strong discriminative ability of SF-12 for mental component scores.

Abstract

The health-related quality of life (QoL) assessed by the 12-item Short Form (SF-12) offers a time-efficient alternative to the 36-item version 2 (SF-36 v2). This study aimed to compare the performance of SF-12 and SF-36 v2 among patients with rifampicin-resistant/multidrug-resistant tuberculosis (RR/MDR-TB) in Vietnam.

A cross-sectional survey was conducted among RR/MDR-TB patients treated in seven provinces between October 2020 and March 2023. Participants completed the SF-36 v2 questionnaire at enrollment. Physical (PCS) and mental component summary (MCS) scores were compared between SF-12 and SF-36 v2. Linear regression assessed the ability of PCS-12 and MCS-12 to predict PCS-36 and MCS-36. Discriminative ability was assessed via receiver operating characteristic (ROC) curves.

This study included 565 participants with a median age of 45.4 years (IQR 44.2–46.5) and a male proportion of 71.7%. 64.1% (362/565) had PCS-36 and 88.3% (499/565) had MCS-36 scores well below and below population norms. The mean scores of PCS-12 and MCS-12 were higher than those of PCS-36 (45.1; 95% CI 44.1–46.1 vs. 43.6; 95% CI 43.1–44.1), and MCS-36 (45.2; 95% CI 43.8–46.6 vs. 34.6; 95% CI 33.9–35.4; p < 0.001), respectively. The Intraclass Correlation Coefficients (ICCs) between PCS and MCS scores of SF-12 and SF-36 v2 were 0.6 (95% CI: 0.5–0.6) and 0.5 (95% CI: 0.4–0.5) respectively. The AUC values for comparing the performance of PCS and MCS of two scales were 0.89 and 0.99, respectively. We found the excellent linear correlation between PCS-12 and PCS-36 (r = 0.8; p < 0.001) and between MCS-12 and MCS-36 scores (r = 0.9; p < 0.001).

SF-12 had moderate correlation to SF-36 v2 for QoL assessment in RR/MDR-TB patients, though it tends to overestimate QoL in older and male individuals.

The online version contains supplementary material available at 10.1186/s12955-026-02479-y.

Figures