2.1. Study design:
A population-based cross-sectional study was conducted in the medical research unit of Kandahar University in Kandahar, Afghanistan. The entire study recruitment was carried out in seven months from November, 2018 to May, 2019. The recruitment process for the survey was announced to the public by distributing study brochures in Kandahar University and to each study participant. These brochures included all the information about the study in the local language and instructions for registration. The eligible criteria were: 1) male or female residents of Kandahar province (living for more than one year), 2) aged between 20 and 75 years, 3) able to sign informed consent and provide bio-specimens. Inability to provide informed consent, pregnant women, people with severe health diseases and visitors from other provinces were excluded from the study. We used stratified sampling to recruit equal number of normal weight, overweight and obese participants. A total of 729 participants were recruited from the population residing in the province and after excluding underweight participants and those with missing nutritional assessment data, 711 were finally included in the study for statistical analyses.
2.2. Sample size:
Sample size calculations demonstrated that a total of 691 subjects would be sufficient to detect a correlation of at least 0.3 with 80% statistical power (type I error 5%). Therefore, we will have >80% power at a significance level of 0.05 to detect a correlation of 0.3 or higher between i.e. dietary intakes or biomarker levels and obesity in 711 subjects. We have >80% power at a significance level of 0.05 to detect a correlation of 0.4 or higher between i.e. dietary intakes or biomarker levels and obesity on 355 subjects (when stratifying the analyses by gender) or on 178 subjects (when stratifying the analyses by gender and age). After correction for multiple testing (20 independent tests), we will still have >80% to detect a 0.4 correlation in 711 and 355 subjects and to detect a 0.5 correlation in 178 subjects. The calculations have been performed with the ‘Power’ statistical procedure of SAS 9.4”.
2.3. Ethics:
An informed consent form was developed in the Pashto language which included the information sheet and the consent certificate. This consent form, study protocol and all the questionnaires were submitted to the IARC Ethics Committee, Kandahar University’s institutional review board (IRB) and FHI 360’s IRB. A consent form was obtained from each participant. A project description leaflet was developed in Pashto language and given to all participants before their meeting with trained staff, which clearly explained the scope of the study, the extent of participation and the expected benefits of the study as a whole for better prevention of obesity.
2.4. Measures and data collection:
All the participants were provided with information about the study and were asked to provide written informed consent (in Pashto language). Face to face interviews and anthropometric measurements were conducted by trained health staff using standardized questionnaires. Training of health staff for interviews and study procedures was conducted in the medical faculty. Interviews and all the study procedures of the participants were undertaken in the medical research unit following Standard Operating Procedures (SOPs) specifically developed or adapted for the study. Table 1 summarises all the measurements of data collection in the study.
Table 1 Summary of measurements of data collection in KOR Study
Examination
|
Measurements
|
Details
|
Face to face interview
|
Demographics and socioeconomic information
|
Sex, age, ethnicity, year of birth, address, housing characteristics, education, marital status, occupation, household income
|
Lifestyle
|
Smoking, tobacco snuff, alcohol, physical activity
|
Medical history
|
Cardiovascular, respiratory, digestive and urogenital diseases, cancer, diabetes, allergy, and oral health
|
Family history of disease
|
High blood pressure, diabetes, cancer and others
|
Reproductive history (women)
|
Age at menarche, age at menopause, contraception, reproductive organs surgery, parity, and breastfeeding
|
Stress
|
Anxiety and depression symptoms (Hopkins Symptoms Checklist)
|
Dietary groups
|
Dairy; meat, eggs and fish; fruits; vegetables; legumes; whole and refined cereals; nuts, seeds and dried fruits; fats and oil; fast food and snacks; sweets and desserts; beverages; spices and condiments
|
Physical Examination
|
Anthropometric measures
|
Weight, height, sitting height, body mass index (BMI), waist and hip circumferences, blood pressure, imaging (liver ultrasound), bioelectric impedance analysis (BIA)
|
Clinical lab-based tests
|
Blood
|
Total cholesterol (TC), triglycerides (TG), high- and low-density cholesterol lipoprotein (HDL-C, LDL-C), and fasting blood glucose
|
Bio-sample collection
|
Blood, stool and urine
|
GenSaver, GenCollect, and BioSample TFN collection cards (Ahlstrom) to be stored at IARC for future biomarker analyses (i.e. metabolomics, inflammation markers, fatty acids, microbiome)
|
Non-dietary questionnaire and anthropometry:
A standardized questionnaire was developed from other studies conducted in collaboration with IARC (6,7) and administered by trained health workers in the research centre. It included information on socio-demographic and housing characteristics, income, occupation, ethnicity, personal and family medical history, stress, anthropometry, and physical activity. Socio-economic and demographic indicators include age, marital status, religion, number of siblings and children, education level, occupation, ethnicity (Pashtun, Tajik, Hazara, Uzbek, Baloch, others), residential address, type of housing, ownership of house and other house items. Personal and medical history include smoking status, diseases during childhood, vaccination history, previous diseases, present chronic diseases, trauma, anxiety and depression (using Hopkins Symptoms Checklist (HSCL)), previous hospitalization, recent treatment for a disease, regular intake of any medicine or dietary supplements, use of herbs and medicine plants, oral health, family history of cancer and other chronic disease in first degree relative, and for women, parity, age at first pregnancy, duration of breastfeeding, menstrual cycle history (age at menarche, menopausal status), use of oral contraceptive and hormone therapy.
Anthropometric measurements were performed by trained health personnel. Body weight was measured in all participants dressed in thin clothes with a digital electronic scale to the nearest 0.1kg. Both standing and sitting heights were measured without shoes with a stadiometer to the nearest millimetre. Body mass index (BMI) was calculated as weight (kilograms) divided by standing height (meters) squared. We used WHO’s classification for BMI categories: normal weight (BMI of 18.5-24.9), overweight (BMI of 25.0-29.9) and obese (BMI of ≥ 30). Waist circumference was measured midway between the lowest rib and superior border of the iliac crest at the end of normal expiration using a non-elastic tape to the nearest millimetre. Hip circumference was measured in standing position at the level of the most prominent part of the gluteus. Waist/hip ratio (WHR) was calculated from these measurements. For abdominal obesity, a cut-off point of 80cm and 94cm and for waist/hip ratio a cut-off point of 0.85 and 0.90 were considered for women and men, respectively.
Bioelectric impedance analysis (BIA) was used (Nutriguard Data Input device) to estimate body composition, particularly body fat. All the BIA measurements were performed while the person was lying in the supine position. To determine body fat (in kg and percent), lean mass (fat-free mass), body water, plus body cell mass (BCM) and extra-cellular mass (ECM), four electrodes were placed on the right hand, wrist, foot, and ankle and were connected to a generator applying an alternating electrical current of 0.8 mA and 50 kHz.
Participants were also asked about changes in their body weight during the last year, weight control methods and body silhouette at various ages. Pictograms with 9 options (from very thin to very fat), which are validated in different settings (8), were shown to both male and female participants to identify their body silhouette at different ages: Men, around ages 5 to 10 years, 15, 20, 30, 40, 50 years and at current age; Women, around ages 5 to 10 years, at menarche, 15, 20, before first birth, 30, 40, 50 years, at menopause and at current age.
Physical activity (PA) was measured by using the short version of International Physical Activity Questionnaire (IPAQ) (9). The questionnaire provided information about the duration of the physical activity in hours and minutes during a usual week. The intensity of physical activity was recorded as light, moderate and vigorous physical activities and later converted into METS (metabolic equivalent).
Dietary Assessment:
For a comprehensive assessment of dietary intake, a detailed country-specific food frequency questionnaire (FFQ) was developed which included all local food items relevant to the population. Data from previous dietary monitoring surveys have been used to select the food groups of interest (5). In addition, a research team of nutritional epidemiologists from IARC (International Agency for Research on Cancer) and Afghanistan checked the food list derived from these existing resources to optimize the food list and examples to be included further considering usual foods commonly consumed in Afghanistan. The food list of the final questionnaire consisted of different food items/categories (combining foods with similar food composition) and different consumption frequency categories (number of times per day, per week, per month, rarely or never). All foods consumed by the population in large and small quantity were included in the questionnaire. Different kinds of traditional and industrial beverages consumed by the population were also included in the questionnaire.
Each food item in the FFQ was assigned a portion size using standard local household units such as a plate, bowl, spoons of different sizes (tablespoon, teaspoon), tea-pot, tea-glass, and glass of water, as well as using photographs of foods and typical preparations of the local population, included in a food portion photograph book (FPPB) developed by the research team.
Regarding seasonal foods, participants were asked to answer the question based on intakes during periods/seasons when these foods were available. The daily intake was calculated according to the number of months per year that each seasonal food item was available.
The research team will also compile Food Composition Tables. For the compilation of the food composition tables we will use the indirect method based on pre-existing data from literature and local/neighbouring regions. Our data will be compiled according to international standards and guidelines for food composition data set by the FAO/INFOODS [FAO/INFOODS (2012). FAO/INFOODS Guidelines for Checking Food Composition Data prior to the Publication of a User Table/Database-Version 1.0. FAO, Rome]. The food composition table will be used to assess nutrient intakes from the FFQ.
The dietary intake assessment was performed by trained health staff who asked the participants to report their habitual consumption frequency of the different foods listed in the FFQ during the last year. The food portion photograph book showed life-size colour photographs of foods in four portion sizes with photographs of utensils. The frequency categories included in the questionnaire were: Never, < 1 time per month, 1-3 times a month, once/week, 2-4 times/week, 5-6 times/week, once/day, 2-3 times/day, 4-5 times/day and 6 times/day.
Daily food intakes were calculated by multiplying the frequency of consumption with the selected portion size, considering also the number of months per year that seasonal foods were available.
Clinical Non-invasive Assessment:
As no incentive was paid to the participants, the following non-invasive clinical measures were used to facilitate recruitment and as a compensation for the participant’s time and contribution. Blood pressure was measured by a mercury sphygmomanometer at right hand supported at heart level after sitting quietly for 15 minutes and recorded to the nearest 2 mmHg. Hypertension was defined as systolic blood pressure values of ≥140 mmHg and/or diastolic blood pressure values of ≥90 mmHg (10) or based on use of antihypertensive medicine. Abdominal ultrasound was performed by a trained physician. The purpose of the procedure was to scan the abdomen for any pathological changes. The liver was specifically scanned for the diagnosis of non-alcoholic fatty liver disease.
Participants were also evaluated for symptoms of anxiety and depression using Hopkins Symptoms Checklist (HSCL-25). This questionnaire contained 25 questions, where 10 and 15 questions were asked to assess anxiety and depression symptoms, respectively (11). A 4-point Likert scale was used to score each symptom and the total score was divided by 25 (total number of symptoms). Participants with a mean score of ≥1.55 were considered probable psychiatric cases and those with a mean score of more or equal to standard cut-off of 1.75 were considered symptomatic (12).
The results of the study were shared with the participants and on their request, they were also provided with feedback about their diet, physical activity level and blood lipid and glucose profile. In case of any abnormal findings, the participants were examined free of cost and treated similarly to other patients by a specialist according to the facilities present in the teaching hospital.
2.5. Biological Specimen Collection:
IARC’s standardized protocol for specimen collection was implemented as described in “Common Minimum Technical Standards and Protocols for Biobanks Dedicated to Cancer Research, IARC Technical Publication No. 44, 2017” (http://publications.iarc.fr/Book-And-Report-Series/Iarc-Technical-Publications/Common-Minimum-Technical-Standards-And-Protocols-For-Biobanks-Dedicated-To-Cancer-Research-2017)”. These protocols have been extensively used in epidemiological studies led by IARC in various settings (13).
Dried blood spot (DBS), dried urine strip (DUS) and faecal occult blood test (FOBT) cards are an easy and inexpensive means of collection and storage of biospecimens in settings where collection and storage of plasma, urine and stool is not optimal (e.g. due to poverty, logistic or environmental/climate constraints). DBS can be used for molecular biology techniques and other diagnostic assays. All of these collection cards can reduce the cost and difficulty of cold chain shipping of samples and can be shipped as non-dangerous goods at room temperature (14,15).
Blood samples were collected in TFN and GenSaver cards, stool samples were collected in GenSaver and GenCollect cards and urine samples were collected in GenCollect cards. 99.5% of the participants provided with both types of blood cards. For GenCollect 82% and for GenSaver 83% of the participants provided stool samples. Urine cards were prepared only for the first hundred participants as a sample. Before sample collection the following data were recorded from all subjects: date and time of sample collection, date and time of preparing and packing biospecimen collection cards, fasting status (blood samples were only collected during fasting), time since last meal or drink, diet on previous day given that intake of some foods may affect biological measurements, smoking status, last menstrual cycle (women only), and use of any type of drugs or dietary supplements (i.e. multivitamins) in past week. All DBS, DUS and FOBT samples were stored in a clean, dry, temperature and humidity-controlled area of the laboratory, not exceeding 30°C and with no direct sunlight exposure before their shipment to IARC for storage and biomarker analyses.
Blood samples:
Blood samples were obtained by venipuncture from fasting participants seated in a semi upright position into two vacutainers tubes of 5 ml containing heparin anticoagulant. One tube was analysed for lipid and glucose levels and the second tube was used for the preparation of DBS on one GenSaver and two BioSample TFN collecting cards (Ahlstrom). Using a pipette with a disposable tip, 125 μl and 50 μl of blood (for Gensaver and TFN respectively) was transferred by the laboratory technician to the centre of one circle on a labelled filter card without touching the filter paper directly with the tip of the pipette. This procedure was repeated to fill all circles of the cards and three cards were prepared per participant. The paper cards were then let dry for four hours at room temperature and later placed in a labelled plastic bag with a desiccant sachet with a humidity indicator. DBS were stored at room temperature before shipment to IARC.
Stool samples:
All stool samples were collected at the study recruitment site, using a standard protocol: stool collection paper was placed inside the toilet bowl. The participant collected a sample of the stool with a spatula, placed it in a labelled stool collection cup and returned the container to the laboratory. There, a small portion of stool was scraped by the laboratory technician using a wooden stick and all four cells of the card were smeared thinly with stool and the flap was closed. For each participant two stool cards, a GenSaver and a GenCollect, were collected. When completed, the stool cards were inserted into a gas-impermeable labelled plastic bag containing a desiccant pack with a humidity indicator.
Urine samples:
Participants provided urine samples in separate clean plastic containers along with stool samples. GenCollect urine card was soaked with urine in the container until it was completely saturated. The card was then suspended in a secure place and allowed to dry for around 6 hours. Finally, the dried card was covered with the flap and sealed in a labelled plastic bag with a humidity indicator desiccant pack. Two DUS cards were prepared for each participant. Before their shipment to IARC, these samples were kept in a temperature not exceeding 30°C and humidity-controlled (no more than 22%) room, safe from direct exposure to sunlight.
Biomarker analyses to measure metabolic health:
Glucose and lipid profile (total cholesterol, triglyceride, HDL-cholesterol and LDL-cholesterol) measurements were performed in plasma from fasting participants in Kandahar University Teaching Hospital. Samples were centrifuged and analysed using an enzymatic photometric test method within one hour of sample collection. Participants with a fasting plasma glucose level of >115 mg/dL or on antidiabetic medicine were considered diabetics. Dyslipidaemia was defined as triglycerides >200 mg/dL, total cholesterol >200 mg/dL, LDL-C >130 mg/dL or HDL-C <40 mg/dL (16–18).
Future biomarker analyses:
The establishment of a biobank, which includes DBS, DUS and dried stool samples will be of great value to perform high-quality biomarker-based research, such as metabolomics and microbiome analyses on stool and blood samples and DNA extraction from DBS for genetics and epigenetics studies.
All questionnaire data and biological samples were centralized at the International Agency for Research on Cancer in Lyon, France.
2.6. Data handling and statistical analysis:
Data cleaning
All data of the questionnaire were entered into Epi Info 7. The electronic version of the database was double-checked with paper forms which were completed manually during interviews. A unique numerical identifier was given to each study participant. Data variables were properly coded and labelled for statistical analyses. Duplicate observations and invalid data points were removed and spelling errors were corrected.
Statistical analyses
We will assess the association between demographic risk factors, dietary patterns and biological markers and metabolic health. All potential confounders available will be adjusted for in each analysis. Logistic regression and standard normal regression will be used to study the association between the exposures of interest and the risk of obesity. Statistical analyses are performed at Kandahar University and at IARC by STATA version 14, SPSS version 23 and SAS 9.4.