(1024) Changes In Use Of Progesterone After The PROLONG Trial: A Physician Survey.

Following a pilot study to determine optimal sample size, we conducted a vignette-based (see Section D in Figure 1) survey in eight academic hospitals in the NYC area. We used a Likert scale to measure counseling tendencies regarding use of IM and Vaginal progesterone before and after the PROLONG trial (-5 to +5, with -5= advise strongly against, +5 = strongly recommend). We determined that a 90% power for a 2-sided paired t-test with a significance level (SL)=0.05 would be achieved with 13 participants in each of the eight institutions. P</span>articipants were asked to describe which governing institutions influenced practice changes the most (-5 to +5, with -5= did not influence my habits, +5 = completely changed my habits). Recruitment occurred on Labor and Delivery units, and included residents, generalists, and MFMs. 

Results:

Of the 157 providers approached, 145 (92%) responded to the survey. Of the 145, 136 (94%) completed all sections and were included in final analysis. A comparison of pre-to-post PROLONG trial practices revealed that respondents were less likely to recommend IM progesterone and more likely to recommend vaginal (p < 0.001). Neither level of training nor costs to the patient predicted this change. The estimated mean change was −2.1 Likert points for prescribing IM progesterone following the PROLONG Trial publication (95% CI [−3.5, −0.7], p=0.003). Participants reported that practice changes were most influenced by the FDA’s statements. (Spearman correlation test, p=0.019 r=0.2 [Section E in Survey, Figure 1]).

Conclusion:

Obstetricians and MFMs in our cohort made significant changes in their counseling regarding progesterone use for patients at risk for preterm birth. This paradigm shift came rather quickly after the publication of the PRLONG trial, showing little lag time between publication and change in practice.