Data collection

The present research was conducted on the campus of the University of Tabuk, Saudi Arabia. The data were collected at various locations which had been specifically chosen for their convenience to students across all the university’s colleges. A team comprising a medical doctor, a dietician, 2 nurses, and 2 laboratory technicians performed the dietary and anthropometric data collection and blood sample extraction.

Each participant was interviewed to complete the structured questionnaire which had been developed by the researchers based on the large body of relevant prior literature to meet the objectives of the research.^{10–12,15–17} The questionnaire covered 5 main topics: the participants’ personal sociodemographic data (eg, their age, marital status, and monthly household income); dietary information relating to their intakes of iron-rich foods and of iron absorption–inhibiting or enhancing foods, and whether or not they followed particular dietary regimens; their obstetric and gynecological history (ie, their menstrual history, including its frequency and duration; their usual menstrual flow during their cycle; whether or not they encountered blood clotting; their use of oral and other contraceptives, and if so, how long they had done so; and their number of children and the birth intervals between them); any current or past diagnoses of medical conditions (eg, chronic diseases, blood disorders, or blood transfusions); their history of smoking (if any); and, finally, their personal and family histories of IDA.

Before the participant interviews were conducted to complete the questionnaires, a pilot study was performed with a sample of 20 students to test the design of the questionnaire and to gather feedback which could be used to make any alterations which might be needed. This pilot study also established the time required for questionnaire completion, which was measured at an average of 15 minutes.

A trained dietician administered a 24-hour dietary recall to estimate each participant’s nutrients intake by asking them to report all the foods, drinks, condiments, and sauces they had consumed during the day prior to the interview. In each interview, different serving plates, cups, glasses, and spoons were shown to the participant to assist them in estimating the portion sizes of the foods and drinks they had consumed. Before completing the 24-hour dietary recall, the dietician checked whether the participant regarded the previous 24 hours as having been typical in terms of their diet. If they did not (due, eg, to having attended a party or family gathering or to having dined out at a restaurant rather than at home), then they were instead asked to report on their diet 2 days prior to the interview date to capture data for a typical day. Tinuviel, WISP version 4.0 nutritional analysis software (Warrington, UK) was then used to analyze the dietary data captured by the questionnaires. In instances where nutritional information was unavailable in the software, other sources (eg, manufacturer product labels or information supplied by fast-food companies) were consulted. The adequacy or otherwise of each participant’s food consumption was then determined using dietary reference intake.¹⁸

Each participant’s weight was measured to the nearest 100 g using medical weighing scales (Seca Ltd, Hamburg, Germany) while they were barefoot and wearing minimal outer clothing. To ensure accuracy of measurement, the scales were calibrated at regular intervals as per the manufacturer’s guidelines and checked for zero readings prior to each measurement. Every participant’s height was also measured to the nearest 0.5 cm using a portable height measure (Seca Ltd), and their body mass index was worked out via the formula: body mass (kg)/height (m²) in line with the World Health Organization’s (WHO) criteria for the classification of overweight and obesity.¹⁹

On the same day when the dietary and anthropometric data were collected, a nurse took a 10-mL sample of venous blood from each participant for hematological and biochemical screening tests. The laboratory technicians in the research team used 2 types of vacutainers to perform venipunctures; one contained EDTA for hematological tests, including hemoglobin (Hgb), hematocrit, mean corpuscular hemoglobin, mean corpuscular volume, and mean corpuscular hemoglobin concentration. These were performed using a Beckman coulter LH750 machine (Beckman Coulter Inc., Miami, FL, USA) which was regularly checked and calibrated during the study according to the standard quality assurance protocols set at the beginning of the research. In the case of the biochemical screening, the participants’ blood was collected using vacutainers without any additional anticoagulant, so each blood sample was allowed to clot as it was assigned to serum separation. Serum iron, ferritin, and total iron capacity were each measured using a modular machine (Hitachi, Finchampstead, UK), which, again, was regularly checked and calibrated during the study as per the standard quality assurance protocols set at the beginning of the research.