True difference or something else? Problems in cost of osteoarthritis studies.

OBJECTIVES

To identify the commonalities and discrepancies among the published cost of osteoarthritis (OA) studies and to further propose some potential improvements for this type of research.

METHODS

A systematic literature review was performed on MEDLINE (1966 to April 2006) by using 2 overlapping search strategies to identify cost of illness studies in OA based on the predefined eligible criteria. Direct and indirect costs per patient per annum were separately summarized across studies and countries. The sample size-weighted mean was calculated when there were 2 or more studies from the same country. Discount rates of 3% and appropriate exchange rates were used in conversion of the costs reported in different years and different currencies to 2005 US dollars.

RESULTS

Ten articles fulfilling the eligible criteria were included in the literature synthesis. Of these, 4 were from the USA, 2 each from Canada and France, and 1 each from Italy and Hong Kong. After adjusted to 2005 US dollars, annual direct costs per patient (in descending order) were $9147 in Hong Kong, $4792 in USA, $2878 in Canada, $1271 in Italy, and $345 in France. In contrast, indirect costs were only reported by 5 studies in 4 countries. The highest indirect costs were $9847 per patient per annum in Canada and the lowest were $864 in Hong Kong.

CONCLUSIONS

The observed substantial variations in costs of OA across studies and countries may not reflect the true differences among them. The comparability across these identified studies is quite limited, which highlights the importance of standardization in cost of OA studies.