Interventional Study to Decrease Cardiovascular Risk in Women in Primary Health Care in South Croatia

The study took place in family medicine (FM) offices from 1 March to 1 September 2014. Two FM offices were in the city of Split, the capital of the Split-Dalmatian County, one office in the neighboring Adriatic island of Brač and one in the southern Adriatic island of Korčula.

All participants reported their most recent menstrual period. To assess the impact of reproductive status, we used the categorization proposed by the "Stages of Reproductive Aging Workshop" (STRAW): 1) premenopausal (PRE) women - last period reported within 61 days before the intervention; 2) perimenopausal (PERI) women - last period reported 61-365 days before the intervention; and 3) postmenopausal (POST) women - last period reported more than 365 days before the intervention.

All participants were exposed to the intervention. The intervention consisted of a 60-minute lecture titled "Change of lifestyle and nutrition habits to reduce cardiovascular risk". The lecture was delivered in family medicine offices by four specialists of family medicine individually to the groups of 6-8 participants. The lecture was verbal, harmonized among four family physicians, and contained instructions for changing nutrition habits, smoking habits and instructions for increasing physical activity. At the end of the lecture, each participant received a personally tailored decision aid, which included the list of her risk factors, assessment of the 10-year risk of fatal CVD (based on the data provided by the participants before the lecture), and instructions on what she should do in terms of changing the lifestyle habits.

The study participants first completed the Pre-study questionnaire (Questionnaire 1), which included: a) demographic data, b) attitudes and knowledge questionnaire about CV risk factors, c) decisional conflict scale (DCS), d) integrative hope scale (IHS), and e) eating habits questionnaire (20).

Attitudes and knowledge questionnaire about CV risk factors was created according to the model of "Ottawa Decision Support Tutorial".

The DCS consist of 16 items rated in a 5-point Likert-type response format, and measures individual's uncertainty toward a course of action. There are five subscales: uncertainty, informed, values clarity, support and effective decision. The scores on the total scale and subscales are calculated as a sum of items, divided by the number of items and multiplied by 25, allowing for a score range from zero (no decisional conflict) to 100 points (extreme decisional conflict). The Croatian version of the scale was previously validated.

The IHS is 23-items scale, a self-rating instrument with items being rated on a six-point Likert scale from 1, strongly disagree, to 6, strongly agree. It provides an overall score and four dimension scores, obtained by summing up the individual item scores, with negative items being rated inversely. This produces possible overall hope scores ranging from 23 to 138 with higher scores representing higher hopefulness. The scores for the sub-dimensions vary according to the number of items. The Croatian version of the scale was previously validated. Hope to be healthy at 70 and hope to reduce CV risk was assessed by a visual analogue scale from 0 to 100.

EPAT is a simple, quick, self-administered tool using an easy scoring method for accurately assessing fat and cholesterol intake. It is a reliable and valid substitute for more time-consuming food records. EPAT also provides an efficient way to monitor eating patterns of patients over time and is arranged to provide an educational message that reinforces the consumption of recommended types and numbers of servings of low-fat foods. The questionnaire was translated into Croatian by the authors and then back translated by an independent language expert to confirm the translation validity.

Ten-year risk of fatal CVD was estimated using the ACC/AHA (American College of Cardiology/American Heart Association) guidelines, based on the following data collected from the study participants: age, gender, race, total and HDL (high-density lipoprotein) cholesterol, systolic blood pressure, data about antihypertensive therapy, diabetes mellitus and smoking status.

Immediately after the lecture, the participants filled the Post-lecture questionnaire (Questionnaire 2), which included: a) attitudes and knowledge about CV risk factors, b) decisional conflict scale (DCS) and c) integrative hope scale (IHS).

Three months after the lecture, the participants filled in the last questionnaire (Questionnaire 3), which included: a) attitudes and knowledge about CV risk factors, b) decisional conflict scale (DCS), c) integrative hope scale (HIS) and d) eating habits questionnaire (EPAT). Ten-year risk of fatal CVD was also calculated at this time point.

BMI, waist and hip circumference, systolic and diastolic blood pressure, blood cholesterol, triglycerides, physical activity and smoking status were measured at each of three time points.

Six months after the intervention, we contacted the participants for the final assessment, along with their feedback on the intervention in general and their personal opinions on further improvements in their lifestyle. This feedback was in the form of a semi-structured interview conducted by the authors, the interview included 13 structured questions and 4 open-ended questions.

We preformed the thematic analysis of the answers, grouping them into theme categories. Categorization of the answers was made by the two independent assessors. After determining the categories, each answer was marked s 1 if matching to specific category or as 0 if not. Answers were used as predictors in further analysis.

After the study, we grouped the participants into those who reduced the CV risk and those that did not. We used logistic regression for all parameter we measured to identify factors contributing to the reduction in CV risk.

Chi square test was used to compare categorical variables (presented as absolute numbers and percentages). For continuous variables, presented as means with 95% confidence intervals, we used Mann-Whitney U test. P values ≤0.05 were considered statistically significant. Logistic regression was used to describe the relationships between variables. Odds ratio for each outcome were reported with 95% confidence interval. All statistical analyses were conducted using the MedCalc statistical program (version 16.2.1 MedCalc Software, Ostend, Belgium).