Factors associated with compliance with weekly iron and folic acid supplementation among school adolescent girls in Debub Achefer district, northwest Ethiopia: school-based cross-sectional study

Iron deficiency anemia is a public health problem among adolescents that could be addressed by weekly Iron Folic Acid Supplementation (IFAS). The Ethiopian government piloted weekly IFAS in schools, where its effectiveness depends on compliance. We assessed the determinants of compliance with the weekly IFAS in Ethiopia. A school-based survey was conducted in 506 adolescent girls on weekly IFAS. Compliance was considered when girls reported WIFAS for at least three months without discontinuation. Bivariable and multivariable logistic regression models were modeled, with odds ratios reported. Out of 506, 25.8% had limited access to educational resources, and 79.4% had no information on IFAS. Among these, 47.9% (95% CI: 45.5–49.9%) had poor compliance with weekly IFAS. Non-compliance was mainly due to school absenteeism (55.9%). Important predictors of poor compliance were adolescent girls’ marital status (AOR = 5.21; 1.55–17.6), academic standing (AOR = 4.37; 2.20–8.70), family income (AOR = 1.85; 1.09–3.15), access to health education materials (AOR = 1.57; 1.02–2.40), problems with IFAS (AOR = 2.44; 1.26–4.74), a discouraging home environment for the program (AOR = 2.27; 1.54–3.34), and a lack of knowledge of the IFAS program (AOR = 1.40; 0.97–2.03). Compliance with weekly IFAS is optimal, which could be improved via strong adherence support and feasible supplementation schedules.


Factors associated with compliance to weekly IFAS
We run separate bivariable and multivariable (adjusted model) regressions to identify possible risk factors for poor compliance with weekly IFAS.All the proposed variables were run using cross-tabulation, and only those relevant or significant contributors were included in the final adjusted model.In the non-adjusted model, married adolescent girls, attending grades 1-4, girls from public and private schools, those who had experienced some problems (side effects) with IFAS, a discouraging home environment, and poor knowledge of IFAS were important determinants of compliance with weekly IFAS.

Question Category Frequency Percent (%)
For how long is IFA tablet should be taken within a month?
For one week 7.6 For three weeks 10.4 For three weeks 5.2 For four weeks and above 417 76 How many tablets should be taken within a year?were positively associated with poor-compliance with the weekly IFAS.The odds of poor compliance were significantly higher among those adolescent girls without access to health education materials (AOR = 1.57; 95% CI: 1.02-2.40)and who encountered some problems with IFAS (AOR = 2.44; 95% CI: 1.26-4.74)as compared to their counterparts.The odds of having poor compliance with IFA in the school were higher among those without discussion with family members than among those who had discussion with family members.Moreover, discouraging home environments for the IFAS program (AOR = 2.27; 95% CI: 1.54-3.34)and having poor knowledge of the IFAS program (AOR = 1.40; 95% CI: 0.97-2.03)were associated with higher odds of poor compliance with the weekly IFAS (Table 5).

Discussion
The current study was to investigate the level of compliance with weekly IFAS and important factors affecting compliance among adolescent girls in Debub Achefer district, Ethiopia, where national weekly IFAS was piloted.According to this study, only 66.1% of adolescent girls had good compliance with IFAS which indicates a promising compliance yet to be improved.The national guideline recommends adolescent girls to take at least twelve tablets of IFA within three months for better effectiveness with universal compliance 20 .However, the higher noncompliance (33.9%) was reported in the current study, which could limit its effectiveness in reducing the burden of anemia.On the contrary, the current compliance rate is very similar to studies reported from pilot conducted in Ethiopia, where 62.9% of them consumed the full dose (24 tablets) during the six-months period 22 .On the other hand, the current non-compliance rate is lower than studies conducted at Bahour commune in rural Puducherry public schools (47.2%) 27 , and rural Pondicherry (85.8%) 28 .
Additionally, 93.2% of adolescent girls have taken IFA at least once in the first semester of the academic year.This experience agrees with the technical report of operational research conducted by AAU that showed 89.3 and 73.8% were on the weekly IFAS program 22 .This high percentage of adolescent girls who ever started taking IFA in the aforementioned study could be due to collaborative efforts of stakeholders involved in the program.It is also obvious that activities like training, supply provision, and close follow up at the beginning of a new  20 .But if the same effort is maintained throughout the program implementation, persistently higher compliance and better outcomes could be achieved.
Related to this, one-fifth of girls interrupted IFA, mainly due to school absences, side effects, or a lack of IFA supply.In different studies, the reasons for interrupting IFA were the occurrence of side effects;-nausea or vomiting, bad taste; misconceptions, and being absent from school on the date of supplementation 18,19,29,30 .Also, thirty seven girls did not completely accept IFA, which was mainly linked to a lack of clear information and misconceptions about the IFA tablets.This implies the need for enhanced community engagement and BCC intervention to increase adoption and address context-specific misconceptions 3,28,31 .Scaling weekly IFAS schedules usually requires strong collaborative work by creating community engagement and making adolescents understand the temporary nature of these incidents.
The study shows that those girls who had experiencing common side effects on weekly IFAS had lower compliance to IFAS.Although the common side effects of IFAS-heart burn, nausea, metallic taste, and discomfort are common, the weekly administration could significantly reduce their occurrence and improve compliance 32 .However, side effects may terrify girls and testify to some misconceptions in the community.It is important to have a strong behavioral change intervention focusing on the transitory nature of these side effects 15,33 .
In the current study, adolescent girls who were married and being from nongovernmental worker families were more likely to have poor compliance with weekly IFAS compared to their counterparts.These could be related to a better home environment and knowledge on the issue among girls among governmental workers as compared to farmers and private workers.Hence, it could be a result of better education and a favorable attitude toward health issues.However, unlike the findings of this study, another study indicated that there is an association between compliance with IFA supplementation and increasing age 19 that could also be associated with the grade level.
The study also showed that the presence of a discouraging at home environment and poor knowledge of IFAS were associated with noncompliance among adolescent girls.For instance, a study indicated that modeled behavioral interventions could improve mean knowledge scores, attitude, subjective norm, perceived behavioral control, and behavioral intention for positive behaviors 34 .Therefore, the IFAS program should establish functional BCC and community engagement packages through the use of influential local leaders to improve the home environment for better adoption.This would increase their awareness and knowledge and help their family understand the advantages of compliance with IFA tablets.It should be noted that public hesitance usually exists in the first few months of the program and might get better thereafter 32 .
Adolescent girls with limited access to SBCC material were 57% higher odds of noncompliance with weekly IFAS compared with their counterparts.Thus, poor access to educational materials could limit their exposure to information about the IFAS program and make them absent from school during supplementation schedules.In addition, access to BCC materials further promotes open discussion with family members for better compliance.For example, a study showed that noncompliance and supplement interruption were related to parental recommendations 21 .

Limitations of the study
The results from this study were self-reported by the adolescent girls, so there is a possible tendency to recall bias, and girls may tend to overestimate their compliance.However, we assessed compliance using three separate yet interconnected questions to capture the real experience and cross-validate their response.This could reduce haphazard responses to compliance with IFAS.

Conclusion
Compliance with the weekly IFAS was found to be promising yet to be enhanced via various targeted behavioral change interventions and better program designs.Also, non-compliance to weekly IFAS was found to be associated with poor knowledge, access to SBCC materials, a discouraging home environment, awareness, and school absenteeism.This can further be improved by increasing access to health education materials, flexible supplementation schedules, and the participatory involvement of stakeholders.Strong adherence support and feasible supplementation schedules could help increase compliance with the weekly IFAS.Hence, it would be very helpful to increase access to educational materials and deliver health education to increase compliance with supplementation.The findings of this study could implicate things to foster while scaling this program at a larger scale.

Study setting, and intervention settings and modality
The school-based cross-sectional study was done in randomly selected schools in northern Ethiopia's Debub Achefer district, West Gojam Zone, in Amhara Regional State.The district is located at an average elevation of 2300 m with an annual temperature of 16.7-37.6°C.According to the Ethiopian Statistical Service's population projection for 2022, the district had a total population of 176,088 people, with 88,990 men and 87,098 women, in a total area of 1102 square kilometers 35 .The district has diverse agricultural practices, mainly mixed cropping and animal keeping.Maize is the main staple crop produced and consumed in the area.A study showed that about 11% of pregnant women had anemia 36 .Moreover, there is evidence of prevailing intestinal parasitosis and it affects causes anemia, which is 37% of pregnant women (hookworm, 18.2%) 37 .About 32 and 11% of school children were undernourished and anemic, respectively 38 , indicting the prevailing malnutrition in the study area.The district has 35 schools that started the weekly IFAS program at the time of this study.According to the national recommendation, 6668 of the 14,833 adolescent girls in the area were taking the IFA supplement The current study focuses on adolescents who participate in a weekly IFAS and nutrition education program that is largely delivered by National Nutrition Program (NNP II) implementing sectors and their institutions.These interventions were implemented in selected schools, primary health care units (PHCUs), and community structures such as the Health Development Army in the districts chosen for this initiative.The service was provided using the school's weekly IFAS platform in collaboration with the district health and education office, the health facility's integrated outreach platform, and the health facility's routine weekly IFAS and nutrition education delivery platform.
The previous service was provided in the form of an observed weekly IFAS or taking the supplement at home.However, the current intervention employed an integrated supplementation and behavioral change intervention through nutrition education delivered by service providers.The interventions were delivered as pilot government initiative by trained staffs, and school teachers along with targeted behavioral interventions.The NNP II was in charge of carrying out this program.The current study targeted only adolescent girls attending school who have been taking weekly IFAS; they were included for the sake of physical accessibility and a large share of the program.

Study participants
This study primarily targeted all adolescent girls aged 10-19 who receive weekly IFAS at schools in the Debub Achefer district of northern Ethiopia.Then, a random sample of schools and adolescent females from each school were assessed to determine compliance and its determinants for weekly IFAS at the study location.Hence, all 506 adolescent girls who were taking weekly IFAS from the chosen schools, were included in the study.

Sample size and sampling procedures
The sample size was estimated using the single population proportion formula, assuming a prevalence of compliance (66%) 22 , 5% level of significance at 95% confidence level, 5% non-response rate, and a design effect of 1.5 to account for possible clustering and heterogeneity in the outcome measured.A marginal error of 5% around the point prevalence of compliance rate was assumed where a total of 543 adolescents were required for this study.Among these a total of 506 adolescent girls were actually included in the study.
The district was carefully chosen because it is one of the Ministry of Health's experimental districts for the weekly IFAS intervention area.Six schools were chosen at random from a total of 35 that were targeted for this intervention.Then, the sample was proportionally allocated to each randomly selected schools where adolescents girls were randomly selected for interview.As a consequence, 62 adolescents (from Aferida school), 88 (from Nifasa school), 65 (from Korench school), 99 (from Lalibela school), 135 (from Ashuda school), and 94 (from Lihudi school) were picked for this study.

Data collection procedures and quality assurance
The data was obtained using a standardized, pretested, and translated questionnaire via the Open Data Kit (ODK) offline data collection tool.The questionnaire was created using previous material and in Amharic, the local language.A team of skilled and experienced data collectors was assembled to conduct face-to-face interviews with adolescent girls.Compliance was measured through self-reported absences from the weekly scheduled IFAS tablets.This method is more valid and less prone to recollection bias and respondent burden because the tablets are normally given weekly for around three months as opposed to daily doses.This would suggest complete program compliance, but noncompliance could lead to IFAS's ineffectiveness.Hence, compliance to weekly IFAS was defined as taking or not missing any weekly schedules for consecutive three months.If a girl takes a consecutive 12 tablets over the three month period without interruption, she is classified as compliant to weekly IFAS.
Six trained data collectors with university degrees and extensive experience in data gathering techniques and field activities were hired.Data collectors and supervisors received one day of training on the tools, primarily the manner to capture data and the ODK system to capture data.A week before beginning data collecting in the selected schools, a pre-test of the questionnaire was completed on 10% of the samples.Throughout this data collection, the principal investigator (the first author) ensured that all data collection methods were followed.

Variables in the study
The explanatory variables considered in this study were socio-demographic and socio-economic factors, knowledge of adolescents and adolescents' attitudes toward weekly IFAS, availability of water in the school, sources of information, support from family and teachers, peer support, and other program-related factors.This study's outcome variable was adolescent girl compliance with weekly IFAS, which was divided into two categories: good compliance and not complaining about the supplementation regimen.Weekly IFAS was defined as providing 60 mg of elemental iron and 2.8 mg of folic acid to teenage girls aged 10-19 years for three months.As a result, adolescent girls were deemed to be in good compliance with weekly IFAS if they took the supplement weekly for at least three months in the first semester of the academic year; otherwise, they were regarded as non-compliant [20][21][22] .

Data processing and analysis
The data gathered using ODK was cleaned and reviewed for completeness by the data collectors and the supervisor on a daily basis before being sent to the server.SPSS 20 and STATA 14 were used for the analysis.The information was presented in the form of numbers, tables, graphs, medians, and means, as well as a measure of dispersion.Logistic regression assumptions were validated, and binary logistic regression analysis was utilized to discover characteristics associated with IFAS compliance in teenage girls.A bivariable binary logistic regression

Table 1 .
Socio-demographic characteristics of school-age girls on weekly IFAS in Debub Achefer district, northern Ethiopia, 2019.

Table 2 .
Knowledge of adolescent girls on weekly IFAS at Debub Achefer district, Amhara region, northern Ethiopia, 2019.

Table 3 .
Level of awareness of IFAS and utilization patterns of weekly IFAS among adolescent girls from the selected schools in Debub Achefer district, northern Ethiopia, 2019.a, refers to due to age restriction, misconception with the IFAS will be contraception and might cause infertility and b, indicates sickness, menstruation, taking other medications and misconceptions with IFAS tablets.

Table 4 .
Facilitators or barriers to weekly IFAS among adolescent girls from the selected schools in Debub Achefer district, northern Ethiopia, 2019.a, includes abdominal pain, headache, chills, and vomiting; b, includes absent from school, age limit, and other misconceptions.