Inappropriate Ordering of Multitarget Stool DNA Tests for Colon Cancer Screening

Discussion

Our study assessed the rate at which mt-sDNA was inappropriately ordered, the most common reasons for why mt-sDNA was not indicated but ordered anyway, which specialties ordered the most mt-sDNA, and whether rates of inappropriate testing varied by race, sex, insurance type, and the specialty of the ordering provider. This research was not designed to determine why patients and physicians selected mt-sDNA over colonoscopy.

This is the second study to evaluate the inappropriate use of mt-sDNA. In a sample of 902 patients, Agarwal et al. found 17.7% of all mt-sDNA tests were inappropriately ordered tests, most commonly because of being up to date with screening colonoscopy.⁵ Similarly, in our sample, more than 20% of patients had inappropriate ordering of mt-sDNA, most commonly because patients were up-to-date on screening colonoscopy or were above average risk for CRC. This trend could be explained by several factors. Patients may refuse colonoscopy, and mt-sDNA may be substituted. Physicians may also offer mt-sDNA as a low-risk alternative for patients at high-risk for adverse events after colonoscopy. Lastly, providers may be unfamiliar with guidelines for appropriate use of mt-sDNA.

Significant race differences existed in the rate of inappropriate testing, with White race being more likely to receive testing when contraindicated. This finding underscores differences in health care access and quality among race groups, as well as the burden of unconscious bias which may be at play in the landscape of mt-sDNA testing. In contrast, we found no significant sex differences in the rates of inappropriate testing, which is similar to previous findings by Agarwal et al.⁵ In addition, differences in insurance type did not significantly correlate to differences in inappropriate testing, which contextualizes our findings within the larger health care ecosystem. Our findings also demonstrate that the frequency of inappropriate testing significantly varies by the specialty of the provider, with Internal Medicine being the specialty with the highest rate of inappropriate testing. Future research assessing the trend of inappropriate mt-sDNA testing should consider these specialty-specific differences to better understand the reasons why physicians and patients would elect for mt-sDNA over colonoscopy in situations where mt-sDNA would be inappropriate.

If our findings are confirmed, they have cost implications as the direct cost of mt-sDNA is $650.⁶ Furthermore, patients at increased CRC risk would benefit from a more sensitive procedure such as a colonoscopy, rather than mt-sDNA.⁷ Future studies could understand the motivation to order testing outside approved indications through provider surveys and interviews.

Limitations include a limited number of specialists, though the vast majority of tests are ordered by primary care providers. In addition, this study was conducted at a single institution within an outpatient setting; future studies spanning multiple institutions or geographic locations are required to more reliably assess the findings of this study. Our study was conducted during the COVID-19 pandemic, when there was increased use of noninvasive CRC screening; it is possible that our data captured an increased number of inappropriate mt-sDNA testing as a result.

We suggest that increasing provider education on the appropriate ordering of noninvasive screening may help decrease the rate of inappropriate referrals, save costs, and improve clinical outcomes in patients at higher risk for CRC.