The Jordan MOH conducted its third BRFSS in 2007. We used a multistage sampling design to select households in which the survey was administered. We used the 2004 Jordan census to identify census enumeration blocks for the master sampling frame, and we selected households from a sample of blocks, or primary sampling areas. This sampling frame was stratified by governorate, major city, other urban area, and rural area into 30 strata that fit within 3 regions, north, middle, and south. Geographic ordering of the blocks in the frame provided implicit stratification. In each stratum, we systematically selected a sample of 461 blocks with probability proportional to Jordan's total population (Appendix
). We selected 8 households from each block. In each household, we randomly selected 1 adult aged 18 years or older and interviewed that person in Arabic. We conducted interviews between June 1, 2007, and August 23, 2007.
Of 3,688 households selected, 3,654 adults (99%) were successfully interviewed. The survey instrument (available on request) included questions on demographics (eg, sex, age, educational status), health status, health care access, tobacco use, physical activity, nutrition, hypertension and cholesterol awareness, and prevalence of heart disease, diabetes, and asthma.
We defined respondents who smoked 100 cigarettes in their lives and who currently smoked as current smokers. We assessed participation in moderate physical activity with the question, "Do you do any moderate-intensity sports, fitness or recreational (leisure) activities that cause a small increase in breathing or heart rate such as brisk walking and lifting light and moderate weights for at least 10 minutes continuously?" To assess consumption of fruits and vegetables we asked, "How many cups of fresh or cooked vegetables did you have yesterday?" and "How many cups of fruits or fresh juices did you have yesterday?" We considered people who responded yes to the question, "Have you ever been told by a health professional that you have high blood pressure?" to have hypertension. We considered people who responded yes to the question, "Have you ever been told by a health professional that your blood cholesterol is high?" to have high blood cholesterol. We considered people who responded yes to the question, "Have you ever been told by a health professional that you have diabetes?" to have diagnosed diabetes. Type of diabetes was not assessed. Women who reported having gestational diabetes only were considered not to have diabetes.
We asked respondents the following questions related to health-related quality of life (HRQOL): "Would you say in general your health is excellent, very good, good, fair, or poor?" On the basis of the response to this question, we defined a dichotomous variable for fair or poor self-rated health status. We also asked respondents, "Now, thinking about your physical health, which includes physical illness and injury, for how many days during the past 30 days was your physical health not good?" and "Now, thinking about your mental health, which includes stress, depression, and problems with emotions, for how many days during the past 30 days was your mental health not good?" We did not ask respondents for specific underlying reasons of any reported unhealthy days. These questions and their construct validity are described elsewhere (9
). We calculated overall unhealthy days as the sum of physically and mentally unhealthy days, not to exceed 30 days. We defined a dichotomous HRQOL variable as fewer than 14 or 14 or more unhealthy physical days, unhealthy mental days, and unhealthy days (mental or physical). A total of 14 unhealthy days is a meaningful cut point for participants reporting substantially impaired HRQOL.
All questions were translated from English into Arabic and then back translated to ensure accuracy. To ensure consistency, we conducted pilot tests of the Jordan BRFSS for 2002, 2004, and 2007 under realistic field conditions and used the same trained interviewers who were recruited to conduct the actual survey to implement the pilot tests. The testing process accounted for all survey activities: approaching potential participants, seeking and obtaining informed consent, making arrangements/appointments for data collection, preparing the site, collecting all data, identifying follow-up cases, and avoiding double data entry. Twenty male and female participants of different educational and socioeconomic levels and varied ages were used for each pilot test. We summarized participants' comments into a single report and made modifications to the survey instrument, ensuring intended meanings were retained.
To compare self-reported health information to actual medical measurements, we selected a sample of 116 of the total 461 blocks and invited survey respondents to participate in a standard medical examination. Participants completed a consent form, and the study design was approved by the Jordan MOH. Of the 889 survey respondents who were invited to participate in the medical examination, 765 (86%) agreed to participate. Participants were evaluated at local health clinics, where height, weight, waist circumference, and blood pressure measurements were obtained. A fasting blood sample was obtained from each participant and sent to a central laboratory where total cholesterol and blood glucose were measured. Standardized training was provided to the attending physicians of the participating local health clinics, and all participating physicians used the same standard equipment for blood testing and for measuring height and weight.
For participants of the medical examination, we computed BMI as weight divided by the square of height (kg/m2). Participants were classified as normal weight (BMI <25.0), overweight (BMI 25.0-29.9), and obese (BMI ≥30.0). We defined high blood pressure as 140/90 mm Hg (systolic/diastolic), high blood cholesterol as ≥240 mg/dL, impaired fasting glucose as 100 mg/dL to 125 mg/dL, and diabetes as ≥126 mg/dL. We considered the presence of antihypertensive medications for high blood pressure, lipid lowering medications for high blood cholesterol, and insulin or oral hypoglycemic medication for impaired fasting glucose and diabetes for these classifications.
For respondents who participated in the medical evaluation, we estimated the relative odds of overweight and of obesity associated with selected health risk factors by using logistic regression analysis adjusted for sex, age, education, smoking, physical activity, and fruit and vegetable consumption. We used STATA statistical software (STATA Corporation, College Station, Texas) in all analyses to accommodate the complex survey sampling design.