Access to healthcare: underserved communities

Welcome to the final edition of the International Journal of Health Promotion and Education for 2024.In this issue, the Editorial team have looked back at the papers submitted across the Volume.In doing so, we found that we have had more papers published that employed quantitative methods than qualitative -and this surprised us!In terms of quantitative methods, most of the papers appeared to employ a mix of traditional inferential statistical tests (e.g.independent t-test or chi-square test) and logistic regressions.Importantly, good practice examples have emerged in relation to substantiating the choice of specific analytical aspects.This is the case of the article by Sui et al. in their analysis of Canadian university students' trait anxiety and nomophobia during COVID-19.In absence of theoretical grounds to guide the selection of the variables, the authors used backward stepwise selection to determine which independent variables would be eligible to enter the final model specification.Another way to guide model specification is adopted by Hall et al. by using hierarchical multiple regressions.This method underlines the authors' attention towards understanding the power of the independent variables of interest (i.e.specific drinking motives) in predicting drinking habits when socio-demographic variables are also controlled for.Besides establishing the model specification, considering how to best extract value from analysing the data on hand is equally important.In this sense, Gainer et al. enriched their analysis of depression and anxiety symptoms among U.S. physicians during the first phase of the COVID-19 pandemic by utilising multiple imputation to handle missing data.With this technique, the authors characterised the uncertainty around missing data points by estimating multiple plausible values derived from the distribution of the observed data.Overall, the variables' selection process guiding the model specification and the use of multiple imputation reflect the importance of examining and maximising the analytical value of the data.