In this study, a Bayesian logistic model is used to re-examine the findings of Pimenta et al. [5] for the association between demographics and hyperaldosteronism (H-Aldo). A data set on 251 patients from the University of Alabama at Birmingham (UAB) Hypertension Clinic for resistant hypertension was used. Applying a Markov Chain Monte Carlo (MCMC) approach, the authors focused on both age and gender as possible correlates of hyperaldosteronism. Using normal priors for the age and gender parameters, both younger age (posterior OR = 1.04) and males (posterior OR = 4.33) were more likely to experience aldosterone excess. The 95% posterior credible intervals did not contain unity. Posterior density profiles also indicated marked departure from posterior expected unit odds ratios.

There was no difference in mean office BP between H-Aldo and normal aldosterone status (N-Aldo) patients. Daytime, night-time, and 24-h systolic and diastolic BP were significantly higher in H-Aldo compared to N-Aldo males. Daytime, night-time, and 24-h systolic BP were significantly higher in H-Aldo compared to N-Aldo females. Multivariate analysis indicated a significant interaction between age and aldosterone status. The effects of aldosterone on ambulatory BP levels were more pronounced with increasing age. The authors demonstrate how the model performs under relevant clinical conditions. The conditions were all tested using a Bayesian statistical approach allowing for the robust testing of the model parameters under various stress conditions which we introduce into the model. The convergences of the parameters to stable values are seen in trace plots which follow the convergence patterns. This allows for precise estimation for determining clinical conditions under which the logistic pattern will change. Further, numerical and graphical examples of our results are given.

aldosterone, Bayesian, demographics, logistic, hypertension quality.