A high degree of geographic variation in the use of medical interventions is usually considered as a sign for inappropriate use. However, the plain geographic variation has the disadvantage that the variation might also be appropriate due to differences in the regions. Hence, we conducted a more comprehensive analysis on Magnetic Resonance Imaging (MRI) use and the relationship to hip or knee replacements. We evaluated whether there is evidence that guideline recommendations regarding hip replacements and total knee replacements are being followed. Additionally, we tried to assess whether the use of MRI is related to subsequent interventions.

We extracted recommendations of the American College of Radiology (ACR) on the use of MRI relevant to hip replacements and total knee replacements. Subsequently, we created three hypotheses on MRI for hip or total knee replacements on what to expect from the data when these recommendations are being followed.

For each hypothesis we calculated a multiple linear regression to analyze Organisation for Economic Co-operation and Development (OECD) data. This was necessary to control for other important variables that might have had an influence on the number of interventions despite the MRI use (for example, healthcare spending, or Computed Tomography (CT) use).

The initial results on (primary) hip replacement and secondary hip replacement were heavily influenced by outliers. After the exclusion of the outliers (Turkey and Belgium), (primary) hip replacements were related to MRI use but not secondary hip replacements. The results on MRI and (primary) hip replacement suggest that the relationship between MRI and hip replacement in Turkey is lower than in the other OECD nations.

Regarding knee replacements, we detected a relationship between the MRI use and total knee replacement. An increase of 10 MRI examinations per 1,000 population would, according to our model, result in 9.8 additional total knee replacements per 100,000 population.

The relationship of MRI and (primary) hip replacement hints to inappropriate use of MRI in Turkey since the data shows a substantial deviation in the relationship compared to the other OECD nations. Apart from this, we found no evidence for inappropriate use of MRI in connection with hip replacements.

However, our results suggest that MRI is inappropriately used in relation to total knee replacements. This might contribute to potentially unnecessary total knee replacements.