Abstract
Respirology
Children with asthma from lower socioeconomic status (SES), and those with poor adherence to controller medications have worse health outcomes, including higher rates of exacerbations and hospitalizations. Simpler medication regimens could potentially improve medication adherence and asthma control.
We aimed to determine if: (1) medication adherence (proportion of prescribed days covered [PPDC]) and primary non-adherence (PPDC = 0%) varied by SES; and if (2) there was an interaction between SES and daily dose frequency on adherence.
This retrospective cohort study included children 2-17 years old followed at the asthma clinic of a large Canadian pediatric hospital between 2011 and 2020. Patients were prescribed one of five medication regimens that included one or more of the following medication classes: inhaled corticosteroids (ICS), long-acting beta-agonists (LABA) and leukotriene receptor antagonists (LTRA). Primary outcome was medication adherence, using the PPDC: the number of days for which a medication was dispensed divided by the number of days for which it was prescribed. For regimens with two different medications, total PPDC was the average of the PPDCs for each medication. The main predictor was SES, measured by the Pampalon’s material deprivation index, a compounded index of income, employment, and education, based on postal codes. It is divided in quintiles, with quintile 1 being the least materially deprived. Comparison of PPDC between quintiles of SES was achieved by a Kruskal-Wallis test. Chi-square testing was used to assess the relationship between the proportion of patients with primary non-adherence (PPDC=0) and quintiles of SES. In the subset of patients who did not have primary non-adherence (n = 462), interaction between SES and daily dose frequency on log transformed PPDC was examined using linear regression models.
Among 551 patients, mean age was 7.1 years (SD 3.8) (Table1), and there was a similar number of patients in each quintile of SES based on a chi-square goodness of fit test (p=0.61). For the overall sample, mean PPDC was 38.1% (SD 27.6). PPDC was not statistically different between quintiles of SES (p=0.57) (Figure 1). Distribution of primary non-adherence (PPDC = 0) in the cohort was independent of the quintiles of SES (p=0.57). Keeping SES constant, twice-daily dose frequency was associated with a 1.1% decrease in PPDC (p=0.85). The interaction between daily dose frequency and SES was not statistically significant (p=0.66).
Medication adherence (PPDC) and primary non-adherence (PPDC = 0) did not vary significantly by SES, but twice-daily dose frequency was associated with a small decrease in mean PPDC. Quebec’s mandatory medication insurance could potentially explain the absence of difference in medication adherence across SES quintiles.
