Assessing Self-Care Perception in Patients Living with Type 2 Diabetes and Their Physicians

BACKGROUND: Type 2 diabetes mellitus (T2D) is chronic illness affecting millions in the United States. Patients living with T2D require highly individualized care and significant patient effort. This effort is comprised of the patient's self-care with regards to medication, diet, lifestyle, and mental health. Self-efficacy is a patient's ability to feel agency over their illness and therefore feel able to maintain self-care. Previous literature suggests that improving a patient's self-efficacy through various behavioral health interventions may improve a patient's ability to manage their T2D. Additionally, interventions on self-efficacy are thought to work regardless of health literacy level and might be a generalizable intervention. However, while validated surveys assessing patient diabetes distress, quality-of-life, social determinants of health, adverse childhood events, and more exist, no literature was found attempting to understand a patient's perspective on their self-care, and by extension, their self-efficacy. Under the premise that consistent beliefs between patient and physician regarding self-care are necessary to make meaningful plans promoting self-care and self-efficacy, the authors developed the term self-care perception consistency to assess relationship between patient and physician perceptions of a patient's self-care. OBJECTIVE: The objective is to assess the consistency between patient and physician perceptions of patient self-care through a biopsychosocial and structural/social determinants of health lens. METHODS: This study uses a cross-sectional, quantitative data set obtained by the Research Residency Network of Texas (RRNeT) through a 71-item survey study. This survey was completed across 12 Family Medicine residencies in Texas and included individuals between 18-75 who were living with T2D. Responses ranged from short free response to Likert-scale based questions and covered topics such as demographics, social determinants of health, patient self-care, diabetes distress, quality-of-life, adverse childhood events, and more. The physicians of each patient were asked to complete a shorter 10-item survey with broader analogous questions to the patient survey. This data was collected through RedCap and analyzed through RStudio. RESULTS: The term self-care perception consistency was coined to describe the relationship between the patient's and physician's perception of the patient's self-care. Self-care perception consistency was found to be lacking 31.2% of the time. Only HgA1c (p<0.01) was inversely correlated with self-care perception consistency in both the univariate and multivariate analyses of demographic factors and social determinants of health. Additional analysis was completed to assess the relationship of HgA1c control, patient diabetes distress, patient quality-of-life, and the physician survey with self-care perception consistency. Self-care perception consistent and inconsistent groups were found to have significantly different HgA1c control distributions (p < 0.01) in the subset of patients that rated their self-care positively, but no significant difference was found in the group that rated their self-care negatively. Patient self-care ratings were best correlated with their diabetes distress (p<0.01) and HgA1c (p<0.01) while physician ratings of patient self-care were best correlated with their perception of HgA1c, perception of patient diabetes distress, perception of patient quality-of-life, and perception of patient social connectedness (p<0.01 for all). Notably, trending diabetes distress, quality-of-life, the physician survey, and social determinants of health across patient self-care ratings in self-care perception consistent versus inconsistent groups revealed that only the physician survey showed opposite trends across the consistent and inconsistent groups. CONCLUSION: Self-care perception consistency was found to be lacking 31.2% of the time. Notably, HgA1c is correlated with patient and physician perceptions of patient self-care (p<0.01 for both); however, in instances of patient-physician self-care perception inconsistency, HgA1c is not correlated with patient self-care perception. Instead, diabetes distress remains predictive of patient self-care rating in all instances (p<0.01). Patient-physician self-care perception inconsistency is also associated with inconsistency in patient and physician perceptions of the patient's diabetes distress. As diabetes distress remains consistently correlated with patient self-care rating, using the validated diabetes distress survey-17 with an additional question regarding self-care may help physicians better understand patients and therefore target appropriate education and psychosocial interventions.