PUSHME (Primary Care USage of Health Promoting Messages): A Text Message-based Intervention in Primary Care Patients With Hypertension: a Randomized Controlled Trial

This is a randomised controlled multi-centre study.The study will involve 400 patients from 9 primary health care centres located in four different regions in Sweden (Region Skåne, Region Kronoberg, Region Stockholm, Västra Götalandsregionen).

Patients in the intervention group will receive SMS messages aiming to remind, encourage and motivate patients to pursue healthy lifestyle changes. After baseline measurement, participants in the intervention group will receive four semi-personalized SMS messages per week for six months, in addition to their usual anti-hypertensive treatment. Each week, the participants will receive SMS from each of the following groups: A. Physical activity, B. Tobacco use, C. Dietary habits, and D. Cardiovascular health in general, except for non-smokers who, instead of the tobacco use-SMS, will get one extra randomly selected SMS.

Included patients that consent to take part in the study will be invited to their primary health care centre for a baseline visit. The following measurements will be assessed by a research assistant: blood pressure (in sitting position after 5 minutes rest; mean of two measurements in a standardized procedure with validated electronic BP devices), BMI and waist-hip circumference. Furthermore, the patients will complete a short questionnaire for evaluation of medical history, medication, tobacco and alcohol use, physical activity level, self-rated health and health-related quality of life. Blood samples for HbA1c and cholesterol will be drawn. Randomization will be performed after completion of baseline assessments and questionnaires. A follow up control will be performed after 6 months with the same assessments as at the baseline visit.

The primary endpoint is change inblood pressure (mmHg). Secondary endpoints are changes in Cholesterol (total cholesterol, high-density lipoprotein [HDL], low-density lipoprotein [LDL]) (mmol/l), Tobacco and alcohol use, BMI (kg/m2), waist circumference, HbA1c (mmol/mol), Self-rated health (five-graded Likert scale), Health related quality of life, as measured by EQ5D-5L, Self-reported physical activity.

The power analysis indicates a sample size of 189 patients in each arm. The calculation is based on an assumed statistical power of 80%, a two-sided test, using a significant level of 5% with a difference of 4 mm Hg between the groups, a standard deviation of 13 mm Hg and a drop out rate of 15%. Data will be analyzed according to the intention-to-treat principle. Differences in mean change of endpoints between intervention and control groups will be calculated by ANCOVA, with baseline values as covariates. Correlation between behavioral change (smoke cessation,increased level of physical activity) and behavioral predictors will be analyzed with logistic regression analysis.