Brief mindfulness coaching enhances selective attention in medical scientists: A pilot study

Medical scientists have dual commitments to medicine and research that create hectic and stressful work schedules that can impact well-being as well as cognition. In this study, we evaluated whether mindfulness coaching alongside wearable-based lifestyle monitoring can benefit this population. We conducted a waitlist-controlled intervention study (n = 43) that included participation from pre-clinical students, graduate students, and medical scientist faculty. Quantitative outcomes assessments included subjective measures of burnout, mindfulness, self-compassion, and wellbeing, as well as objective cognitive assessments. Results showed no impacts on subjective measures (p > 0.2). Yet, notably objective performance on attentive cognition was improved at post-intervention (p = 0.003). Extent of improvement in selective attention was correlated with suppression of visual alpha oscillations – a neural marker for distractibility - measured using electroencephalography (EEG) (r = -0.32, p < 0.05).

Additionally, we obtained qualitative feedback from all participants after they had all received the intervention, including the waitlist arm. Thematic analysis of this feedback showed that participants in both groups equally rated the overall experience as very good (3.7 ± 0.98 out 5), appreciated that the intervention paid attention to lifestyle factors, and contributed to mindfulness, compassion and sense of community. Majority (56.8%) of all participants reported that they expect to change their well-being related behaviors in the future as a result of the intervention. Overall, the study suggests utility of mindfulness coaching for improving attention skills in medical scientists, but that more needs to be done to enhance subjective well-being in this population.

Methods
Methods

A total of 43 physician-scientists participated in the study (mean age: 28.95 ± 4.36 years, range: 23-43 years, 41.86% female). Most (76.7% ) participants were recruited from the dual-degree MSTP at the University of California San Diego (UCSD), and 23.3% had graduated an MD PhD program and were now physicians/junior faculty. Recruitment occurred directly through an in-person, annual retreat as well as over email. All participants provided written informed consent for study participation in accordance with the Declaration of Helsinki, and all experimental procedures were approved by the UCSD Institutional Review Board

All data collection and intervention took place during Fall 2021 through Summer 2022. Participants provided demographic information at the beginning of the study. Of the total 43 participants, 22 were randomly assigned to the well-being intervention group and 21 were assigned to a waitlist control group. The sample size within each group was powered to detect medium effect size pre/post differences (Cohen’s d >0.6), at beta power of 0.8 and alpha level of 0.05. Between-group differences met criteria for investigating only large effect size outcomes (Cohen’s d >0.8) at beta power of 0.8 and alpha level of 0.05. Effect sizes were calculated a priori using the G*Power software (Faul et al., 2007).

The well-being intervention group received the intervention during Fall-Winter quarter, while the waitlist group received no intervention during this time. The Waitlist control group eventually received the intervention during Spring-Summer quarter. All participants completed subjective and objective neuro-cognitive assessments before and after the first wave of the intervention, i.e., Fall-Winter quarter. At the end of Summer 2022, all participants in both groups also completed an experience feedback survey.

Regarding missing data, there were no missing data for subjective assessments. Two participants in the waitlist control group were missing cognitive and neural data, while one participant in the well-being intervention group had corrupted neural data that could not be analyzed, hence, was missing. For the end of year qualitative survey, six participants had missing data (5 from the well-being intervention group and 1 from the waitlist-group).