Study design, setting and participants
This study was a cluster randomized controlled community trial with a two-arm parallel design that lasted for one year. This study was conducted in the West Arsi Zone, Oromia region, Central Ethiopia; 250 km from Addis Ababa, the nation’s capital. There are 16 districts in the West Arsi Zone (13 rural districts and three towns). A total of 2,929,894 people were estimated to live there by mid-2022. A total of 12,556 km2 make up the zone, which has a climate of 45.5% highland, 39.6% medium land, and 14.9% lowland36. According to the Zonal report of 2022, there are 417 public health facilities, 5 of which are hospitals, 324 of which are health posts, 88 of which are health centers, and 203 of which are private medium and higher clinics, including one nongovernmental hospital and two private hospitals providing health services. At least one sexual and reproductive health (SRH) service was utilized in the zone; in 2019, it was 58.6%37. The study was conducted between August 2022 and July 2023. This trial included pregnant adolescents before 16 weeks of pregnancy who intended to remain in the study region until delivery. Adolescents who refused to give their informed consent were not included in the study.
Sample size determination and sampling technique
The sample size was calculated using G*Power 3.0.10. The required sample size was determined using the following assumptions: a 95% confidence interval, a 5% margin of error, 80% power, and an intra-cluster correlation (ICC) of 0.03 (based on a comparable published study ICC)38 for the difference between two independent means (two groups). A 10% loss to follow-up was considered, and a design effect of two was employed. A total of 488 people constituted the computed sample sizes. As a result, both intervention and control groups of 244 pregnant adolescents were included. Twenty-eight clusters were used, and the average cluster size was 19. A single-stage cluster sampling technique was applied in this study.
Recruitment, randomization and intervention allocation
Five of the zone’s 16 districts had nutritional education interventions; thus, they were excluded from the study. Four districts, namely, Dodola Rural, Adaba, Gedeb Hasasa, and Siraro, were chosen from among the remaining eligible districts by simple random sample (SRS) technique. Samples of nonadjacent kebeles (clusters) from the four districts were chosen using SRS. Six clusters each from the Dodola Rural and Adaba districts, as well as eight clusters each from the Gedeb Hasasa and Siraro districts, were selected based on their proportion to size allocation and considering cluster allocation in the intervention and control groups. Kebeles (the smallest administrative entity in Ethiopia) were used as the randomization unit (clusters). Clusters for the intervention and control groups were distributed using the lottery (SRS) approach. The Consolidated Standards of Reporting Trials (CONSORT) guidelines were used to report the results (Fig. 1).
![figure 1](https://i0.wp.com/media.springernature.com/lw685/springer-static/image/art%3A10.1038%2Fs41598-024-55709-y/MediaObjects/41598_2024_55709_Fig1_HTML.png?resize=685%2C845&ssl=1)
Flow of the study participants through the trial according to the criteria recommended in the CONSORT guideline.
A cluster randomized trial was used to avoid message contamination because pregnant teenagers in the same cluster were likely to communicate and discuss intervention messages. To avoid information leakage, all pregnant adolescents who met the criteria in one cluster were enrolled in the same arm (either the intervention or control arm). Buffer zones (none selected clusters) were also positioned39,40.
The study included all pregnant teenagers who met the inclusion criteria. The study participants were screened and enrolled by nurses, and the clusters were randomly assigned. A house-to-house survey was performed, and pregnant teenagers who met the criteria were screened by first date of their last menstrual period and urine HCG test was used to confirm their pregnancy. Urine-based pregnancy tests and urine HCG levels were used. The procedure involved dipping a test strip into a urine sample. The test strip contained chemicals, monoclonal antibodies that reacted to the presence of the pregnancy hormone HCG (human chorionic gonadotropin). The results are displayed as lines on the test strip.
Intervention
The intervention strategy used in this study was a community-based nutritional behavioral change communication intervention (NBCC) based on the Health Belief Model (HBM). The recommendations of the World Health Organization, the blended training module on nutritional counseling developed by the Federal Ministry of Health of Ethiopia, EFDRE/MOH, and related interventional studies41,42,43,44 served as the basis for preparing the intervention package. Additionally, the baseline study, conducted at the start of this study, served as a direction for the development of intervention tools; the NBCC included the husbands of the pregnant teens and a demonstration of how to prepare meals. The intervention items included a training manual for nutrition counsellors, leaflets with key messages for pregnant adolescents and their families, and counseling checklist cards.
The intervention method was tested for one week in an environment similar to the research site, and adjustments were made in light of the results. The counseling manual’s core contents included eating a range of foods, especially iron-rich foods, animal products, fruits, and vegetables, and increasing the meal frequency and portion size as gestational age increased. The main components of counseling guidance also included taking iodized salt and iron/folic acid supplements. Additional messages about the core components included reduced workload, day rest, the use of impregnated bed nets, and the use of medical services.
It was also emphasized how undernutrition can harm a person’s development and how vulnerable pregnant adolescents need to eat. The benefits of eating enough meals that are diverse and the challenges in maintaining a balanced diet were also highlighted in the NBCC guidebook. Throughout her pregnancy, each pregnant adolescent attended four counseling sessions. Personalized NBCC based on trimester was provided during home visits. Counselors used a client-centered approach to identify specific dietary preferences and needs. Before allowing the teenagers to choose easily accessible, agreeable, and affordable guidance at their location, counselors considered the needs of the pregnant adolescents, their household income, and any gaps they had discovered. Counseling was conducted using the GALIDRAA approach (Greet, Ask, Listen, Identify, Discuss, Repeat, Agree, and Appoint)45.
A counseling guide that included the necessary information was used to conduct each counseling session for the NBCC, which lasted 45–60 min. The first appointment concentrated on the basics of nutrition, food groups, selecting foods that provide a balanced diet, showing how to prepare meals, the frequency of meals, and the use of iodized salt before 16 weeks of pregnancy. The second and third counseling sessions, which covered the whole of the counseling manual, were offered during the second trimester. Final counseling, which focuses on weight gain and incorporates all the module’s important messages, was provided based on the gaps that were discovered during the previous trimesters of pregnancy.
Each pregnant adolescent in the intervention arm received a leaflet containing the key themes in Afan Oromo and Amharic (local languages) and appropriate images. Anyone at home who could read was asked to read the leaflet to the pregnant adolescent if the adolescent could not read it.
Health extension workers chose 14 AFD [formerly known as WDAs] counselors based on their performance and prior involvement with public health services. The 14 AFDs were carefully supervised by four BSc nurses. Role-playing exercises and fieldwork using the training handbook were part of rigorous one-week training for counselors and supervisors. After the intervention had been in place for 2 months, the supervisors and counsellors received three additional days of training to ensure that the providers continued to adhere to standardized practices.
Intervention fidelity
Criteria were created to assess the integrity of the intervention based on the best practice suggestions developed by the National Institutes of Health Behavioral Change Consortium46. The criteria47, included checklists to assess the intervention design, counselor training, counseling process, receipt of the intervention, and implementation of the skills picked up during the intervention. Nonadjacent clusters were selected to prevent information contamination. The intervention and control groups had an equal number of clusters from each district to balance differences.
The intervention strategy was tested before the experiment. Additionally, each pregnant adolescent received the same number and frequency of counseling sessions, and the lengths of interactions within the intervention group were comparable to standardize the method. Counselor training was given in a group environment using a training booklet, role-playing, and simulated counseling sessions. Tests administered before and after training, as well as a practical evaluation, were utilized to evaluate counselors’ skills and knowledge. The process observer graded the counselors using a “yes/no” rating system and looked at things such as using a counseling guide, covering the whole subject, the duration and frequency of counseling, preparation, accuracy, and the counselor’s ability to respond to questions appropriately. Pregnant adolescents’ understanding of food throughout pregnancy was assessed through interviews via checklists to determine their knowledge of the main components of the intervention.
Participants, counselors, and data collectors were blinded to the study’s objective; participant allocation concealment was impractical given the nature of the intervention. Until the analysis was complete, the groups were given a unique nonidentifiable number that also served to blind the data entry clerk. The counseling process was supervised by the main investigator and counseling supervisors.
Data collection procedure and measurements
The primary outcome of this study was nutritional status as measured by the MUAC, while the secondary outcome was GWG. The mid-upper arm circumference (MUAC) is the recommended assessment tool for nutritional status because of its simplicity and sensitivity in detecting undernutrition. In low-resource settings, where girls have minimal subcutaneous fat, it is the preferred measurement method because changes in the MUAC are more likely to reflect changes in muscle mass48. It has been demonstrated that a low maternal MUAC is useful for detecting unfavorable delivery outcomes, such as intrauterine growth restriction, preterm birth, and asphyxia at birth49. The left mid-upper arm circumference (MUAC) was measured at the anatomical landmark at the midpoint of the acromion and olecranon processes of the nondominant hand, with the palm facing upward and the women’s elbows flexed to 90°. The measurements were taken twice by employing inelastic MUAC tape and interpreting the measurements to the nearest 0.1 cm.
A pretested structured questionnaire was used to obtain the data. Sociodemographic characteristics, meal frequency, use of medical services, types of diet, and Household Food Insecurity Access Scale (HFIAS) scores were collected. Six clinical nurses and two master of public health (MPH) holders worked as data collectors and supervisors, respectively. Pregnancy tests were performed by three female laboratory technologists. The data collectors conducted in-person interviews with the participants at their homes to administer the questionnaire. To the best extent possible, the adolescent’s privacy was protected by prohibiting access to the site where the interviews took place.
According to the Food and Nutrition Technical Assistance (FANTA) III recommendation from the United States Agency for International Development (USAID) and the Food and Agricultural Organization (FAO), dietary intake was determined using 24-h recalls50. The objective was to ascertain whether the meals consumed by pregnant teenagers varied. Ten food groups were used to compute dietary diversity; grain, dairy, meat, white roots and tubers; nuts and seeds; eggs; dark green leafy vegetables; poultry and fish; plantains; other fruits and other vegetables; and other fruits and vegetables high in vitamin A are among the ten food groups listed in the recommendations. Adequate dietary diversity is achieved if pregnant adolescents consumed food from 5 or more food groups within 24 h of day/before data collection; minimum dietary diversity for women (MDD-W) was used51,52. The participants were asked to remember every meal they consumed during the previous 24 h; both inside and outside the home. The participants were also asked if they could remember any between-meal snacks they may have consumed. Food items received a “1” rating if consumed during the reference period and a “0” rating if not.
The HFIAS (Household Food Insecurity Access Scale) Guideline53 was used to assess food security. The HFIAS was used to evaluate the households’ level of food security and included nine questions. Prior to this, the questions were validated for use in developing nations54. Food-secure households experienced fewer food insecurity indicators than did the first two food insecurity indicators. Households that experienced between two and ten, eleven to seventeen, or more than seventeen food insecurity indicators were considered mildly, moderately, or severely food insecure, respectively.
The household’s wealth index was derived using principal component analysis (PCA), taking into account access to a latrine, a water source, household durable assets, and agricultural land. The responses to the none dummy variables were all split into three groups. A code of 1 is allocated to the highest rating. The two smaller digits, however, were given a code of 0. Using variables with a commonality value larger than 0.5, PCA factor scores were produced. The first primary component score for each family is kept to calculate the wealth score. Quintiles of the wealth score were created to classify households as poorest, poor, medium, rich, or richest55.
The autonomy of the pregnant adolescents was evaluated using eight questions. Code one was provided for each question when a decision was made by the girl, by herself or jointly with her husband; otherwise, code zero was provided. The mean score was used to determine the pregnant adolescent’s decision-making capacity44.
Data management and analysis
The data were entered using the Kobo Toolbox and exported to SPSS version 25 for analysis. The baseline variations in the two groups’ sociodemographic characteristics were examined using a Chi-square test. Paired t tests and independent t tests were used to determine within- and between-group differences.
The effect of the intervention on changes in the nutritional status and gestational weight gain of pregnant adolescents over time was estimated using a linear mixed-effects model. This model was chosen because of the repeated assessments (pre- and postintervention) and the clustering of individuals, which allowed us to explain how the results were correlated. The Akaike information criterion (AIC) was used to help us choose the best statistical model. The model that displayed the lowest AIC was selected. The bivariate linear mixed regression model’s variables with p 0.2 were chosen as potential candidates for the multivariate linear mixed model analysis. By analyzing how time and the intervention interacted, the effectiveness of the intervention was evaluated.
Participants were examined as random effects during model fitting. A linear mixed-effects model also makes it possible to manage the impact of various confounding variables. The intra-cluster correlation coefficient of the final model was 0.04, indicating that constructing a third-level model was not necessary. The intercept-only model was created initially. To take into consideration time-invariant variables at the individual level, a two-level model was fitted. The effect of intervention was estimated by testing the interaction term between treatment allocation and time.
Ethical considerations
Ethical approval was obtained from the Jimma University IRB/ethics committees with reference number JUIH/IRB/194/22, and the Oromia Regional Health Office provided support. All methods were performed in accordance with the relevant guidelines and regulations; the study was performed in accordance with the Declaration of Helsinki. Each study participant received a thorough description of the study’s title, goal, protocol, and duration, as well as the potential risks and benefits, prior to providing informed consent. Each teenager provided verbal, written, and signed informed consent prior to any interview or measurement. Informed consent was obtained from the LAR (legally authorized representative) for study participants aged 18 years and younger. Participants were made aware of the publication of their anonymous comments. Informed consent was obtained from participants prior to the commencement of interviews. The researcher remained truthful to the academic and ethical requirements. Finally, the researcher kept the data in a locked file cabinet in a safe place after the completion of the study. Informed consent was obtained from both the adolescent and their husband or parents. Finally, any ethical issues that arose during this research were resolved through discussion between the researcher and JU’s IRB.
Credit: Source link