Assessing the Impact of Electronic Ordering on Stool Parasite Testing Practices in an Academic Medical Center

Abstract

Abstract Introduction While microscopic stool ova and parasite examination (MSOP) is the traditional testing method for intestinal parasites, it has low diagnostic yield when used inappropriately. To promote stewardship at our institution, MSOP was only orderable utilizing paper requisitions. The only stool parasite testing orderable in our electronic ordering system was the Giardia / Cryptosporidium immunoassay (available as stool O&P screen), which has adequate sensitivity to detect the most common regional intestinal parasites. This was changed during two regional cyclosporiasis outbreaks, when MSOP was added to our electronic ordering system (EOS) to facilitate Cyclospora detection (available as stool O&P examination). The stool O&P screen was also changed to Giardia and Cryptosporidium antigen to better reflect the testing performed. We aimed to measure the impact of this change in terms of test utilization. Methods We quantitated the amount of testing performed pre- and post-intervention over 12-month periods, outside of the outbreak period. We performed chart reviews of 141 patients who received stool parasite testing (Giardia/Cryptosporidium immunoassay and/or MSOP) and compared the proportion of appropriate vs. inappropriate stool parasite testing before (n = 73) and after (n = 68) MSOP was added to the EOS. We used the following criteria to define appropriate MSOP testing; the presence of (1) any travel history outside the continental US, (2) immunocompromised status or (3) intestinal parasite mentioned as a differential diagnosis in medical record. MSOP was considered overordered in patients who failed to fulfill criteria. Testing was deemed underordered if only Giardia/Cryptosporidium immunoassay was ordered on patients who fulfilled criteria for MSOP. Both underordering and overordering were considered inappropriate testing. Statistically significant differences were calculated using Fisher’s exact test (p<0.05). Results The number of patients tested per month by MSOP increased from an average of 10 pre-intervention to 56 post-intervention (p<0.0001). The proportion of appropriate MSOP utilization increased from 44% to 58% of (p=0.13). Inappropriate testing happened in both periods, though underordering was significantly higher pre-intervention (89% vs. 10% of inappropriate testing, p<0.0001), while overordering was significantly higher post intervention (90% vs. 11% of inappropriate testing, p<0.0001). Pre-intervention, 34 of 38 patients in whom underordering occurred were immunocompromised patients tested by the Giardia/Cryptosporidium immunoassay only. Notably, all patients who had intestinal parasite mentioned as a differential diagnosis in the electronic medical record were tested by MSOP, regardless of its availability in the ordering system. Conclusion These findings suggest that omission of MSOP from the electronic ordering system can prevent appropriate testing, particularly in immunocompromised patients. Transitioning to an electronic ordering format can promote overall test utilization and decrease the amount of inappropriately narrow testing, though this also promotes inappropriately broad testing which may represent an opportunity for future clinical decision support interventions.