Gender Differences in Prevention Strategies and Therapeutic Adherence After Acute Myocardial Infarction

OBJECTIVES

General objective: To evaluate if there are differences in the level of adherence to recommended secondary prevention therapies (pharmacological and non-pharmacological) between women and men surviving a type 1 acute myocardial infarction (with obstructive coronary artery disease), its potential consequences, and the potential factors related to that difference, if present.

Specific objectives: To compare between women and men:

1. The adherence to recommended secondary prevention therapies at 6 months and 12 months measured as the:

1a. Proportion of patients adherent to all recommended pharmacological secondary prevention therapies at 6 and 12 months.

1b. Proportion of patients adherent to all recommended non-pharmacological secondary prevention therapies at 6 and 12 months.

2. The relationship of the adherence with control of risk factors, clinical outcomes, and use of healthcare resources.

2a. The proportion of patients with optimal control of all cardiovascular risk factors at 6 and 12 months.

2b. The incidence of clinical outcomes at 6 and 12 months. 2c. The cumulative use of healthcare resources and cost after discharge.

3. To assess the factors associated with non-adherence to recommended secondary prevention therapies, with special emphasis to socioeconomic factors and gender issues.

STUDY DESIGN

DESIGN:

Prospective, matched cohort study of patients hospitalized for a type 1 acute myocardial infarction with evidence of obstructive coronary artery disease who are discharged home alive. Women will be enrolled first, ideally in a consecutive manner, as they are the focus of the study. Men will be recruited subsequently as the comparison group, with 1:1 matching for age and ECG presentation. Matching will be performed locally, in each study site (hospital). All patients will undergo 1-year follow-up with clinical and therapeutic adherence evaluation.

RECRUITMENT:

Participant centers:

25 hospitals managing routinely acute myocardial infarction and representing different regions from Spain will be invited to participate.

Patient recruitment:

All consecutive women fulfilling all inclusion criteria and without exclusion criteria surviving the index hospitalization will be invited to participate. Subsequently, men with inclusion criteria and without exclusion criteria, matched for age and ECG presentation, will be recruited.

STATISTICAL ANALYSIS:

A minimum sample size of 820 participants (410 per group) was estimated for an expected relative difference in recommendations compliance of the all-or-none composite primary outcome of 20%. An α error of 0.05 and a β error of 0.20 were considered for the sample size. The recommendations compliance was estimated as 42% for women and 52% for men. With the consideration of 20% of lost to follow-up (discontinuations) the total sample size is 984 patients, which is rounded out to 1000 patients (500 women and 500 men). To increase regional representation and allow exploration of potential regional variability, 25 hospitals from all regions in Spain (17 Autonomous Communities) will be invited to participate.

For the description of continuous variables, mean and standard deviation, or median and interquartile range will be used for Gaussian and non-normal distributions, respectively. For describing categorical variables, frequencies and percentages per category will be used. Categorical variables will be compared using the chi2 test, whilst continuous variables will be compared using the student t test.

Logistic regression models will be used to evaluate differences in binary outcomes between women and men (in hospital acute myocardial infarction management, post discharge myocardial infarction management). All subjects will be assumed to have a fixed follow-up (12±1 month). Multivariate adjusted models will be used to address any potential confounding in the associations between sex and each of the outcomes. Covariates will be selected based on their pre-defined clinical value and the unbalances observed across groups in the univariate analyses. Linear, multinomial or ordinal regression models would be used in case of continuous, categorical, or ordinal outcomes, respectively.

In addition to the models evaluating the association between sex and outcomes, a predictive model will be conducted to set predictors for the following outcomes:

1. a composite of all preventive therapies (all-or-nothing for diet, physical activity, P2Y12 inhibitors, statins, and beta blockers or angiotensin converting enzyme inhibitors [if prescribed at discharge]), and

2. each group of preventive interventions (drugs, cardiac rehab, diet, physical activity) Candidate predictors are classified as biological factors (age, sex, risk factors, comorbidities), disease-related (type of myocardial infarction: ST-segment elevation acute myocardial infarction (STEMI) / non-ST-segment elevation acute myocardial infarction (NSTEMI), left ventricular ejection fraction, number of vessels…), pharmacological factors (daily number of drugs, daily number of doses, specific drugs), socio-economic factors (zip code, education level, employment status, wages…) and specific gender issues (family responsibilities, work conciliation…). Special consideration will be given to age due to the high figures of in-hospital mortality in young women.

For women-specific analysis, female specific factors (menarche age, pregnancies, gestational diabetes, menopause…) will be considered.

It will be used the Clinical Outcomes, HEalthcare REsource UtilizatioN, and relaTed costs (COHERENT) model to study the hospital-related healthcare resources (emergency department visits, specialist visits, re-hospitalizations and urgent procedures) and costs.