The aim of the study: To evaluate the effect of nutrition education intervention (NEI) on glycemic control and other diabetes-related outcomes of patients with type 2 diabetes mellitus (T2DM), at a tertiary health Center, in Basrah, Iraq. Specific objectives: (i) To describe and compare the glycemic control (HbA1c), socio-demographic status (age, gender, marital status, and employment), medical characteristics (duration of diabetes, family history, and blood pressure measures), nutritional status (anthropometric measurements, metabolic parameters, and dietary intake). Besides, lifestyle characteristics (physical activity level, smoking, and alcohol drinking), health belief model (HBM) constructs (susceptibility, severity, benefits, barriers, and self-efficacy), diabetes knowledge (DK), and health literacy (HL) level of patients with T2DM at baseline (T0), between intervention and control groups. (ii) To evaluate the effectiveness of NEI on glycemic control (HbA1c), medical characteristics, nutritional status, lifestyle characteristics, HBM constructs, DK, and HL level of patients within and between intervention and control groups at visit time 1 (T1), and visit time 2 (T2).
Participants and setting: The study was a parallel, randomized design, was conducted from June to December 2019, and involved 208 participants (20-64 years) diagnosed with T2DM at least 6 months before the study and had poorly controlled diabetes (HbA1c ≥ 7%). The study setting is the outpatient clinics at Faiha Specialized Diabetes, Endocrine, and Metabolism Centre (FDEMC), Basrah, Iraq. This study was involved in three phases based on the Generalized Model for Health Program Planning. Phase I included a cross-sectional study for need assessment among 280 participants with T2DM. The data that were collected included socio-demographic status, medical characteristics, nutritional status, lifestyle behavior, and health literacy. Phase II included the preparation and development of nutrition education materials based on HBM. Phase III was the randomized controlled trial among 208 participants with T2DM who attended the program were selected according to specific criteria. Participants aged between 20 and 64 years old, without hearing or vision problems and severe illnesses like cancers and renal failure were included in the study. The eligible participants signed the study consent form after the screening. The data were collected at baseline (T0), visit time 1 after 12 weeks of intervention (T1), and visit time 2 after 10 weeks of follow-up (T2). The institutional ethics committees approved the study and all participants provide their written consent before study enrollment.
Intervention: The participants were allocated using simple randomization to either intervention group (IG), or control group (CG). A total sample size was 208 participants with T2DM (104 per group) to detect a 5% reduction in glycemic control (HbA1c), at visit time 1 (3 months after intervention) and visit time 2 (3 months after following up) and allowing 30% drop out rate. The CONSORT report was followed up, wherever possible. The allocation was concealed using serially numbered opaque closed envelopes until intervention is assigned. The intervention period was 12 weeks long with the following mechanisms: the curriculum (twelve weekly classes, 1 to 1.5 hrs. each); follow-up classes (one monthly each lasting 1 to 1.5 hrs.); nutrition education materials (pamphlets) and reminder calling before each session for IG. The pamphlets were distributed for use at home as individual goal-setting activities. The participants were bucked up to bring a friend or a family member. While the follow-up period was 10 weeks long with the following mechanisms: monthly follow-up class, and reminder calling before each class for IG. The participants in the CG received classes about diabetes and its complications with some advice related to physical activity on different days on IG. Besides, they received the same educational materials (pamphlets) at the end of the study. Both groups continued with the usual medical care at respective Center. All the classes are offered in the meeting hall in the Center by face- to- face using the local Arabic language, which utilized the lecture method using PowerPoint presentations and whiteboard, lecture-discussion groups with household measurements to estimate the portion size and related videos or animations. The classes were concentrated on plate method and eating vegetables before carbohydrates as a simplified manner for participants with low HL. it were administered by a nutritionist and a team of endocrinologists, and diabetes experts. The content and strategies utilized in the intervention were based on Health Belief Model theory (HBM). Standardized questionnaires for data collection are considered as appropriate measures for the study outcomes and suitable for use it with the target patients.
Evaluation of intervention: The evaluations of the effectiveness of NEI were conducted at T1 and T2, on the following outcomes: (i) Clinical outcomes: HbA1c (%), lipid profile in mmol/L, and blood pressure in mmHg. (ii) Nutritional status: BMI in kg/m2, and dietary intake (energy intake in kilocalories per day, macronutrient intake in gram per day, vegetable and fruit intake in servings per day). (iii) Physical activity level in MET-min/week, and (iv) Potential behavior mediators: HBM constructs (susceptibility, severity, benefits, barriers, and self-efficacy), DK, and HL level. The outcome objective is to evaluate the changes between groups at T1 and T2 for HbA1c, lipid profile, and blood pressure, BMI, and dietary intake, physical activity level, HBM constructs, DK, and HL level. The study hypothesis: (i) intervention will reduce the HbA1c at T1 and these levels will be significantly lower in IG than CG, and these levels will remain significantly lower at T2. (ii) IG will have significantly better results in other diabetes-related outcomes that including nutritional status, physical activity level, blood pressure; HBM constructs, DK, and HL scores at T1, and these significantly improved results will maintain at T2 compared to CG. (iii) IG compared to CG will have significantly more patients who achieve a good HbA1c at T1 and will remain significantly lower at T2.
Data analysis: The general linear model (GLM) repeated measures ANCOVA was performed to measure the changes over time with the adjustment of significant differences at baseline as covariates. An intention-to-treat analysis (ITT) of all the outcomes of the study was achieved for non-adherence and missing outcomes. The level of significance was set at p \< 0.05 for a two-tailed test.
