OBJECTIVES

Several studies reported both QWB and SF-6D utility scores head to head. However, no one attempted in a systematic way to address the following question. Do QWB and SF-6D produce the same utility scores and ultimately the same conclusions? This study investigated to address that gap.

METHODS

Studies that employed the QWB and SF-6D to measure HRQOL published in 1998-2018 were searched using PubMed, Embase, and Cochrane. An exploration of the relationship between QWB and SF-6D utility scores using Histogram and Scatter plot was performed. Further, the mean difference of QWB and SF-6D and its possible implications in cost-effectiveness analysis (CEA) was discussed.

RESULTS

14 studies were identified that met our inclusion and exclusion criteria. A total of 82 group mean utility scores of QWB and SF-6D were extracted for our analysis. QWB utility scores data was symmetric and more likely followed the normal distribution attributes compared to SF-6D. Analysis of 82 paired group mean scores showed, SF-6D had higher utility score values compared to QWB (100%), with 90% of the observations had a mean difference ≥0.05. We took a case of one study to show the implications in CEA. We calculated the incremental effect for QWB (δQWB=0.03), SF-6D (δSF-6D=0.04) and assumed the incremental cost ( ψ) as $1500≤ ψ<$2000 and λ (willingness to pay) $50,000 per QALY. The Calculated ICERs were {$50,000≤ ICER (QWB) < $66,667} and {$37,500 ≤ ICER (SF-6D) < $50,000} per QALY for QWB and SF-6D respectively. The intervention is cost effective when SF-6D is used as effect measure while decision changes when QWB is used (not cost effective). QWB and SF-6D utility scores are different may have inconsistent decision in CEA.

CONCLUSIONS

QWB and SF-6D scores for the same population are not identical. Therefore, full exploration of these findings in CEA is warranted.