NEW INSIGHTS FOR STATISTICAL ANALYSIS OF BLOOD PRESSURE RESPONSE TO RESISTANCE TRAINING IN ELDERLY HYPERTENSIVE WOMEN

NOVAS PERPECTIVAS PARA A ANÁLISE ESTATÍSTICA DA RESPOSTA DA PRESSÃO ARTERIAL AO TREINAMENTO RESISTIDO EM MULHERES IDOSAS HIPERTENSAS

ABSTRACT

The main goal was to present statistical procedures for a better data interpretation of responsiveness, explain how to deal with RTM effect, and describe how to determine clinically important changes in BP from significant real difference (SRD). Twenty-seven hypertensive elderly women were included, and RT consisted of a periodized linear model. The RT lasted 10 weeks, with two sessions performed per week. Responders were classified on the basis of SBP differences between time-points T1 (first 3 weeks) and T4 (weeks 9-10). Statistical analyses were performed using One-Way Repeated Measures ANOVA, an analysis of covariance (ANCOVA), the linear mixed model (LMM) was used in the present study, and SRD was also calculated. In conclusion, when one-way repeated measure ANOVA was conducted to determine whether there was a statistically significant difference in SBP levels over the course of 10-week RT, results showed a non-significant reduction of -2.24 mmHg, while classifying subjects by responsiveness provides a different perspective of the results. Furthermore, initial SBP was the more powerful predictor of post-exercise SBP response, as analyzed by the regression to the mean effect. Finally, the reductions of -2.24 mmHg was not statistically significant nor clinically meaningful, but fell within the measurement error of the SBP measurements.

Keywords:
Responsiveness; Regression to the mean; Standard error of measurement; Clinical difference; Blood pressure