Prevalence of obesity in panama: some risk factors and associated diseases

Study design and area

The data used for this article was obtained from database of the first Panamanian
Survey on Risk Factors Associated with Cardiovascular Disease (PREFREC, by its acronym
in Spanish). It was a cross-sectional, descriptive study conducted in the trans-isthmian
zone of the Republic of Panama, between October 2 010 and January 2 011, developed
by researchers from the Gorgas Memorial Institute (GMI) and from the Panamanian Ministry
of Health.

The study area encompassed the provinces of Panama and Colon, which contains the capital
of Panama, and the health regions of Western Panama, San Miguelito, East Panama, Colon,
and Metropolitan health region (Fig. 1) 14]. This provinces lodge the highest population in the country, where 60.4 % of all
Panamanians 18 years or older reside. According to the National Institute of Statistics
and Census (INEC, by its acronym in Spanish) the distribution of age, sex and other
socio-demographic factors are similar in the remaining 39.6 % of the Panamanian population,
not addressed in this research.

Fig. 1. Republic of Panama. Political and administrative divisions

Study sample population

The study included individuals aged 18 years and older who resided in private houses
(according to maps produced by the national census for the year 2 000). Houses were
sampled, employing a single-stage, probabilistic, and randomized sampling strategy
with a multivariate stratification, developed by the INEC. This strategy was conducted
separately for urban, rural and indigenous areas 14], 15]. Each census unit comprised private occupied houses (8–30 houses were reached per
unit), which were stratified, on the first hand, by the Administrative Political Code
of the Republic of Panama. Subsequently, they were stratified according to population
size and the educational level of the study population 14], 16].

This survey was conducted applying a structured form (questionnaire), that consisted
of 14 parts: parts 1 and 2 were disposed for socio-demographic variables; parts 3
through 6 were destined to learn the behavior of the participants towards the consumption
of tobacco, intake of alcohol, consumption of foods and performance of physical activity,
respectively; parts 7 through 11 were destined to learn about the personal pathologic
history of hypertension, T2DM, high cholesterol, obesity, and stroke; part 12 was
disposed to learn the participant’s family history of non-communicated chronic diseases;
and part 13 was destined for the anthropometric measurements and for the results of
fasting blood samples.

Professionals and students in their senior year of Health Sciences education (medical
doctors, nurses, and nutritionists) administered the questionnaire. These collaborators
were trained by the authors of the research in interviewing and survey management
in order to guarantee the quality of the data collection process. Nutritionists (licensed
and students) that collaborated in the survey were trained in anthropometric measurements.
In indigenous areas, were dialects are spoken, the survey administrators were supported
by interpreters who spoke these indigenous dialects 16].

It is important to stand out that nutritionists (licensed and students) were in charge
of administrating parts 4 and 13 of the survey; the other parts were administered
my medical doctors, and nurses (licensed professionals and students).

Fifteen days before the survey started, the population segments were visited (pre-screen
procedure), in order to guarantee an adequate response rate and that the participants
would be fasting. Using a spiral technique, a random survey was conducted between
people aged 18 year and older residing in occupied houses that agreed to participate
in the research (maximum 2 adults per household).

Potential participants were provided with material explaining the objective of the
study, confidentiality of their participation, voluntary nature of the research, potential
risks and benefits from their participation, fasting requirements, institutions to
call to solve their doubts and were provided with a written form (Informed Consent)
for them to sign in case they agree to participate in the study.

A total of 3 590 individuals aged 18 years and older who agreed to participate in
the study were evaluated from the sampled houses. Pregnant women, individuals without
weight and height values, and without waist circumference values, were excluded from
this analysis, resulting in a sample size of 3 507.

Variables

Obesity is our main variable, and was defined as all individuals with BMI???30 kg/m
²
(men and women) 5], 7], 8]. For the purpose of calculating the BMI, height and weight measurements were done
twice. The height was taken setting the participant in the Frankfurt plane; the participant
was weighted being in light clothes, bare-foot, with both arms resting at each side
of the body, and in the Frankfurt plane. If the difference between the two measurements
was greater than 0.5 cm or 0.5 kg, a third measurement was taken. SECA® height measuring
instruments and weighing scales were used.

Age was defined as the years from the time of the individual’s birth until the day
the survey was conducted; sex, by the phenotypical characteristics that distinguish
men from women; area, by the geographic domain where the respondent usually lives
(urban, rural or indigenous); and ethnic group, by the participant’s self identification
as an afro-Panamanian (Panamanian of African descent), mestizo, white, native American
(Amerindian), or of Asian descent.

Physical activity was defined as any activity that requires energy expenditure under
aerobic respiration, like sports and physical exercises and was classified into two
categories: insufficient physical activity (less than 150 min/week) and physically
active (150 min and more/week).

To identify the intake of predisposing foods to the development of obesity, a questionnaire
of food consumption was developed and validated by licensed dietitians. In this questionnaire
a variety of foods were listed, allowing the interviewee to indicate their daily frequency
of food intake during a regular week. According to the frequency of weekly consumption
of food reported by the interviewed individuals, the average daily intake was estimated.

The total of listed foods were classified into 4 groups, two “cardiovascular protective
food” groups (a group of consumption of vegetables and fruits, and a group of consumption
of non-fried fish and tuna fish) and two “cardiovascular risk food” groups (a group
of consumption of sugary foods and a group of consumption of fried foods).

The method used to measure the frequency of consumption of protective foods and risk
foods allowed us to identify the types of foods for which caloric density is associated
with obesity and it permitted us to quantify the number of times they were consumed
and by whom 15]. On this matter, excessive consumption of high fat foods was defined as the intake
of 2 or more daily fried foods or foods high in fat, such as: “tortilla”, puff pastry,
patty, chips, fried plantain, fried yucca, fried sausage, fried nuggets, fried chicken,
fried meat, fried fish, fried pork and pork crackling; snacks, as “kaprichitos”, “doritos”
and “taquitos”; and use of coconut milk or coconut oil for cooking.

Intake of foods or beverages with an equivalent of 6 teaspoons of sugar per day or
more, such as sodas, artificial drinks (like Tang, Kool-Aid, etc.), milk chocolate
bars, sweets, and candies in quantities of 6 or more, were considered as excessive
consumption of beverages or foods high in sugar.

The biological variables included for this analysis were: the family history of obesity,
abdominal obesity, T2DM, Hypertension (high blood pressure), Low Density Lipoprotein
Cholesterol (LDL), and High Density Lipoprotein Cholesterol (HDL).

Family history of obesity was defined as the background of obesity in parents, siblings,
aunts, uncles or grandparents in first or second degree of relatedness.

Waist circumference was measured placing the measuring tape around the patient’s abdomen
(the patient should be standing with his arms separated from his core) between the
uppermost border of the iliac crest and the lower border of the costal margin. The
participant was requested to make a deep inspiration followed by a normal expiration;
at this moment the tape was placed in the position discussed above and locked. The
reading was taken twice. If the difference between these two was more than 1 cm, a
third measurement was performed. SECA® measuring tape was used. In this sense, abdominal
obesity was defined as a waist circumference??88 cm in women and??102 cm in men.
Additionally, for men, we used the waist circumference cutoff point??94 cm 16].

T2DM, was defined as those individuals who reported to be under pharmacological treatment
for T2DM, and those who weren’t under pharmacological treatment but who presented
fasting blood glucose values???126 mg/dL and/or glycated hemoglobin percentage (HbA1c)???6.5 %
14].

The measurement of blood pressure (BP) was performed with calibrated automatic sphygmomanometers
made by American Diagnostic Corporation Model 6013, which had cuff sizes for people
with normal weight and for people who were obese. Three BP measurements were made
in the right arm of each person with the participant seated and with the minimum of
5 minutes between the start of the survey and the first measurement, and between the
2
^nd
and 3
^rd
measurements. An average of these 3 BP’s was used to define the participant’s BP 16]. On this matter, hypertension, involved those individuals who reported to be under
pharmacological treatment for hypertension, and those who weren’t under pharmacological
treatment, but who presented systolic blood tension values???140 mmHg and/or diastolic
blood tension values???90 mmHg 16].

LDL and HDL values were obtained through analysis of blood samples, which were draw
after the survey was applied. Persons with high levels of LDL included individuals
with LDL values???100 mg/dL. Persons with low levels of HDL included women with HDL
values of??50 mg/dL and men with HDL values of??40 mg/dL 17].

Analysis plan

The general and specific prevalence in this study were estimated using percentages
with 95 % confidence intervals (CI) and comparisons were made with the age-adjusted
rates for the Panamanian population in 2014 9]. Standard deviation (SD) was used to report averages.

For risk analysis Odds Ratio (OR) and P values were calculated using contingency tables
for all variables. A P value??0.05 was considered statistically significant. Multivariate
logistic regression analysis was applied to the data set of the explanatory variables
of obesity in men and women only to identify statistically significant variables 18], 19].

The data were processed using SPSS (version 19) (SPSS Inc, Chicago, IL), Microsoft
Excel 2010 (Microsoft Corporation, Washington DC), and Epi Info, version 7.0 (Centers
for Diseases Control and Prevention, Atlanta, GA).

All of the participants signed an informed consent form. The National Bioethics Committee
of the Republic of Panama approved the research.