Does the funding source influence the results in economic evaluations? A case study in bisphosphonates for the treatment of osteoporosis

Background: Research sponsored by the pharmaceutical industry is often assumed to be more likely to report favourable cost-effectiveness results. Objective: To determine whether there was a relationship between the source of funding and the reporting of positive results. Methods: We conducted a systematic review of the literature to identify economic evaluations of bisphosphonates for the treatment of osteoporosis. We extracted the source of funding, region of study, the journal name and impact factor, and all reported incremental cost-effectiveness ratios (ICERs). We identified which ICERs were under the thresholds of $US20 000, $US50 000 and $US100 000 per QALY. A quality score between 0 and 7 was also given to each of the studies. We used generalized estimating equations for the analysis. Results: The systematic review yielded 532 potential abstracts; 17 of these met our final eligibility criteria. Ten studies (59%) were funded by non-industry sources. A total of 571 ICERs were analysed. There was no significant difference between the number of industry- and non-industry-funded studies reporting ICERs below the thresholds of $US20 000 and $US50 000. However, industry-sponsored studies were more likely to report ICERs below $US100 000 (odds ratio = 4.69, 95% CI 1.77, 12.43). Studies of higher methodological quality (scoring >4.5 of 7) were less likely to report ICERs below $US20 000 and $US50 000 than studies of lower methodological quality (scores <4). Methodological quality was not significantly different between studies reporting ICERs under $US100 000. Conclusions: In this relatively small sample of studies of bisphosphonates, the funding source (industry vs non-industry) did not seem to significantly affect the reporting of ICERs below the $US20 000 and $US50 000 thresholds. We hypothesize that methodological quality might be a more significant factor than the source of funding in differentiating which studies are likely to report favourable ICERs, with the higher-quality studies significantly less likely to report ICERs below $US20 000 and $US50 000 per QALY. Further research should explore this finding.