CLINICAL DECISION SUPPORT TO PROMOTE SAFE PRESCRIBING TO WOMEN OF REPRODUCTIVE AGE: DIFFERENTIAL EFFECTS BY SUBGROUP

Purpose:  Primary care providers (PCPs) frequently prescribe medications which may lead to birth defects if used during pregnancy. Electronic medical records (EMR) with clinical decision support (CDS) may help alert physicians to this risk and promote counseling, but the impact of CDS may vary by visit subgroup.

Method:  We developed two CDS system to promote communication about medication risks and evaluated their effect on the practices of 41 PCPs. PCPs were randomized to receive either a “simple” or “multifaceted” CDS system for 6 months. We abstracted EMR data from visits made by female patients aged 18-50 during the 10 months prior to implementation and the 6 months during the intervention; one visit was randomly selected from those made by each patient. Women with documentation of sterilization or infertility were excluded. Our primary outcome was the proportion of visits with a teratogenic prescription that had evidence of family planning services (e.g. counseling, contraceptive prescriptions, pregnancy tests or referrals to specialists). Using mixed effects logistic regression models adjusted for covariates and clustering, we compared the change in this proportion by type of CDS, patient age, physician gender, and whether the visit was with the patient’s usual primary care provider. To test each comparison, an interaction term between time period and the factor was included.

Result: Study PCPs were 40+/-10 years old on average; 49% were women.  A total of 805 EMR records were analyzed. Introduction of CDS was associated with a slight increase in the proportion of visits with provision of family planning services when a teratogen was prescribed (+2.6%points, 95%CI: -3.3 to +8.4). There was no difference in effect by CDS type (-1.4%points, 95%CI: -13.5 to +10.8). However, provision to women 30 years and older appeared to improve more than provision to women under 30 (+5.6%points, 95%CI: -9.3 to +20.5); visits with female physicians experienced a larger increase in provision than visits with male physicians (+3.9%points, 95%CI: -8.0 to +15.7); and provision appeared to increase more during visits with usual primary care providers than visits with less familiar providers (+9.8%points, 95%CI: -2.4 to +22.0).

Conclusion:  Overall, these CDS systems slightly increased provision of family planning services when a teratogenic medication was prescribed. Subgroup analysis revealed that CDS was more effective for certain patient and provider subgroups.