Change in Physical Activity after Diagnosis of Diabetes or Hypertension: Results from an Observational Population-Based Cohort Study

Abstract

Background: Chronic diseases like diabetes mellitus or hypertension are a major public health challenge. Irregular physical activity (PA) is one of the most important modifiable risk factors for chronic conditions and their complications. However, engaging in regular PA is a challenge for many individuals. The literature suggests that a diagnosis of a disease might serve as a promising point in time to change health behavior. This study investigates whether a diagnosis of diabetes or hypertension is associated with changes in PA. Methods: Analyses are based on 4261 participants of the population-based KORA S4 study (1999-2001) and its subsequent 7-and 14-year follow-ups. Information on PA and incident diagnoses of diabetes or hypertension was assessed via standardized interviews. Change in PA was regressed upon diagnosis with diabetes or hypertension, using logistic regression models. Models were stratified into active and inactive individuals at baseline to avoid ceiling and floor effects or regression to the mean. Results: Active participants at baseline showed higher odds (OR = 2.16 [1.20;3.89]) for becoming inactive after a diabetes diagnosis than those without a diabetes diagnosis. No other significant association was observed. Discussion: As PA is important for the management of diabetes or hypertension, ways to increase or maintain PA levels in newly-diagnosed patients are important. Communication strategies might be crucial, and practitioners and health insurance companies could play a key role in raising awareness.

Keywords: chronic disease; diabetes; diagnosis; health behavior; hypertension; physical activity.

Figure 1

Study population and analysis flow chart. * Of the baseline S4 sample, 174 died, 204 moved outside the study region or to an unknown location, and 12 refused to be further contacted. Of the remaining 3871 eligible participants, 176 could not be contacted, 220 were too ill or busy, and 395 refused to participate further. Thus, 3080 participants (80% response rate) took part in the first follow-up F4 (2006-2008) examination. Of the F4 sample, 168 died, 97 moved outside the study region or to an unknown location, and 67 refused to be contacted further. Of the resulting eligible 2748 participants, 48 could not be contacted, 332 were too ill or busy, and 207 refused to participate further. Adding 118 participants from S4 without F4 information, 2279 (79% response rate, excluding 118 participants without F4 information) participants took part in the second follow-up FF4 (2013–2014) examination. Annotations: n = number of participants, Nobs = number of observations.

Figure 2

Overview of the cases with incident diagnosis of diabetes or hypertension for the active and inactive strata by PA change. See Figures S1 and S2 in the supplement for a detailed overview of how the presented numbers were derived. Numbers of observations for the combination of both follow-up periods: Active-Diabetes = 2722; Active-Hypertension = 2174; Inactive-Diabetes = 2298; Inactive-Hypertension = 1679. Deviations from the numbers of Figure 2 and Figure 3 are due to missing values in covariates for which the models of Figure 3 were adjusted. Abbreviations: PA = physical activity; S4 = baseline study (2000); F4 = first follow-up (2007); FF4 = second follow-up (2014).

Figure 3

The odds-ratios for changing PA after a diabetes/hypertension diagnosis for both baseline PA strata. ORs for the active stratum display the odds of changing to inactive. ORs for the inactive stratum display the odds of changing to active. All models were adjusted for the following factors: sex, age, education, family status, baseline BMI, change in BMI, and change in physical and mental health-related quality of life. Abbreviations: Nobs = Number of observations, OR = Odds ratio, 95% CI = 95% Confidence interval.