The All of Us Research Program provides an efficient platform for remote training medical/health science students in science methodology – Showcase: Peripartum cardiomyopathy and its relationship with proxies of socioeconomic status.

Philadelphia, PA

Background: Medicine is a science-based discipline. Scientific method investigates a problem by identifying it, formulating a question and a hypothesis to explain it, testing the hypothesis, and collecting and analyzing data that can potentially reject it. Scientific method also applies to clinical problems providing a systematic approach for diagnosis and treatment. It is an iterative process requiring repetition to achieve proficiency. Iterating processes need time that must integrate with busy schedules within medical school curricula. We explored the “All of Us” NIH-research program to introduce formal training in science methodology to medical students by conducting a retrospective analysis on peripartum cardiomyopathy (PPCM). PPCM is an idiopathic cardiomyopathy with systolic dysfunction occurring during the last month of pregnancy or within months after childbirth in women with previously normal cardiac function. The estimated global PPCM prevalence is 0.05% live births. A retrospective study conducted at the University of Pennsylvania Health System showed that black race and socioeconomic proxies (like neighborhood disadvantage index (NDI)) were independently associated with sustained cardiac dysfunction. The study further demonstrated that among the components of NDI (education, high-rental occupied housing, annual income below poverty line, female-headed household, adult unemployment, adults on public assistance), low-education and high-rental occupied housing were significantly associated with sustained cardiac dysfunction. Here we assessed the effect of educational level on the prevalence of sustained cardiac dysfunction from PPCM across the US using the “All of Us” repository.

Methods: We met with the M4 student and set a schedule to interact virtually during the 4-week research rotation. Once the student enrolled in the program (https://allofus.nih.gov/), she interrogated the relationship between educational level and PPCM prevalence within the “All of Us” repository (retrospective cohort study). We conducted a Chi-square analysis using the All of Us Jupiter Notebook to evaluate if there was a significant association between educational level and PPCM prevalence. Educational level was categorized as college graduate or advanced degree, college one to three, twelve or General Educational Development (GED), less than a high-school degree or equivalent. Statistical significance p< 0.05.

Results: A cohort of 88 patients with PPCM was identified. Using the Chi-square test, we found a significant relationship between PPCM prevalence and educational level. The Chi-square statistic (23.87) was greater than the critical value (7.81) and the p-value (p< 0.001) was less than alpha (0.05). The relative risk (RR) was 4.31 for patients that did not receive a college degree, compared to those who completed college (confidence interval [CI] 2.56-7.26). Among the 88 patients in the PPCM cohort, 80% did not have a college degree.

Conclusion: 1. Low-level education (no college degree) has a significant impact on PPCM prevalence; 2. Identifying socioeconomic factors influencing PPCM prevalence may improve PPCM prevention, early diagnosis, and management; 3. We were able to integrate the research project using the “All of Us” program as an introductory platform for remote formal training in science methodology within the constraints of M4 timeline, demonstrating the program’s flexibility for integrating scientific training into medical education.