Time to Recovery from Severe Acute Malnutrition and Its Predictors among Admitted Children Aged 6-59 Months at the Therapeutic Feeding Center of Pawi General Hospital, Northwest Ethiopia: A Retrospective Follow-Up Study

Background Ethiopia is one of the countries in sub-Saharan Africa with the highest rates of severe acute malnutrition. Early recovery is a performance indicator for severe acute malnourished children for the therapeutic feeding. Despite the available interventions to tackle nutritional problems, there is scarce information on time to recovery and its determinants among children with SAM in Ethiopia. Objective The study is aimed at assessing time to recovery from severe acute malnutrition and its predictors among admitted children aged 6-59 months at the therapeutic feeding center of Pawi General Hospital, northwest Ethiopia, from January 2013 to December 2017. Methods An institution-based retrospective follow-up study was conducted among 398 children aged 6-59 months. The data were collected by using data extraction sheet. The data were cleaned and entered using EpiData version 4.2.0.0 and exported to Stata version 14 statistical software for further analysis. Kaplan-Meier survival curve was used to estimate median nutritional recovery time after initiation of inpatient treatment, and log-rank test was used to compare time to recovery between groups. The Cox proportional regression model was used to identify the predictors of recovery time. Adjusted hazard rate with its 95% CI was reported to show strength of relationship. Results The recovery rate was 5.3 per 100 person-day observations, and the median recovery time was 14 days (95% CI: 13–15). The lower chance of early recovery was found among children who were not fully vaccinated (AHR: 0.73 (95% CI: 0.56, 0.96)), while high chance of recovery was found among children who had no anemia (AHR: 1.66 (95% CI: 1.23, 2.23)), TB (AHR: 2.03 (95% CI: 1.11, 3.71)), and malaria infection (AHR: 1.54 (95% CI: 1.09, 2.17)) at admission. Conclusion and Recommendation. The overall nutritional recovery rate was below the accepted minimum standard. Children not fully vaccinated and children without malaria, anemia, and TB comorbidities at admission had a higher chance of recovering early from severe acute malnutrition. Hence, treating comorbidities is vital for prompt nutritional recovery.


Introduction
Severe acute malnutrition (SAM) is defined as very low weight for height < 70%, visible severe wasting, or the presence of nutritional edema. In children aged 6-59 months, mid-upper arm circumference (MUAC) less than 115 mm is also indicative of severe acute malnutrition [1,2].
Globally, 17 million children are living with severe acute malnutrition; 1 to 2 million die every year; 25 to 30% of the deaths are in many poor countries [2]. Most of the children with SAM live in South Asia and sub-Saharan Africa. Furthermore, it is the most common cause of hospital admissions, accounting for 20% of pediatric hospital admissions in Ethiopia [3].
Ethiopia is still one of the countries with the highest rates of severe acute malnutrition despite a small decline from 12% to 10% over the last 15 years. The prevalence of severe acute malnutrition among under five-year-old children in the country is estimated at 2.9%; in the study where the setting was Benishangul-Gumuz region, it was 3.1% [4]. The annual cost associated with SAM is estimated to be US $262.62 per child in Ethiopia [3]. Besides, delayed discharge from severe acute malnutrition increases the number of children to be treated, medicine cost, and human resource demand, thus imposing economic burden on resource-poor settings [5,6]. Furthermore, if left untreated, it might result in the impairment of physical, intellectual, social, and adaptive behavior [7,8].
Studies reported inconsistencies in the rate of recovery from SAM which ranges from 22.1 to 95.36%. In Ethiopia, the rate of recovery is known to be below the accepted international minimum standard of 75%. For instance, Mekele, Bahir Dar, and Gondar reported 22.1% (9), 58.4% [9], and 68.5% [10], respectively. Likewise, studies conducted in Ethiopia on the time to recovery from SAM ranged from 14 to 26 days [9,[11][12][13][14][15][16][17][18]]. Shorter recovery time shows an acceptable performance of the treatment and caring process. Recovery time is affected by medical comorbidities like anemia, malaria, dehydration, hypoglycemia, HIV, TB, hypothermia, and lack of child full vaccination. Though these predictors are assumed to delay the time to recovery, some investigators show conflicting results in listing the possible causes of prolonged recovery time.
Although Ethiopia adapted the WHO SAM guideline and has been working for Sustainable Development Goals 2025 under its nutritional strategy goal in order to end hunger, achieve food security, and improve nutrition, only limited information exists regarding the outcome of the SAM treatment provided at the therapeutic feeding center. Ethiopia Demographic and Health Survey (EDHS) reported that the Benishangul-Gumuz region had a high prevalence rate. However, no studies have been conducted in the study area on nutritional recovery. Therefore, this study is aimed at determining the time to recovery from severe acute malnutrition and its predictors among children aged 6-59 months admitted to Pawi General Hospital, northwest Ethiopia.

Methods and Materials
2.1. Study Design, Setting, and Period. A hospital-based retrospective follow-up study was conducted on SAM cases enrolled at the hospital from 2013 to 2017. Data was collected from March to April 2018. The study was conducted in the Benishangul-Gumuz regional state, Pawi General Hospital. Pawi is found in the Metekel zone of the Benishangul-Gumuz regional state, 570 km away from Addis Ababa, the capital of Ethiopia. The study site is geographically located at 11.009′N latitude and 36.003′E longitude and at an altitude of 1050 meters above sea level. The mean annual rainfall and maximum temperature of the area are 1555.1 mm and 32°C, respectively. The hospital has different departments that provide specialized services in outpatient, inpatient, and operation theatre departments that serve an estimated 300,000 people in the region, have 200 beds, and have around 1000 inpatients admitted per month on average. The average patient flow to the therapeutic feeding center was 300/year and 25/month for 37 beds ( Figure 1).
All children aged 6-59 months and treated for severe acute malnutrition at the therapeutic feeding center of Pawi General Hospital enrolled from January 2013 to December 2017 were included in this study. Children in affected areas were screened for signs of SAM and diagnosed based on anthropometric measurement and examination of the feet for bilateral pitting edema. Children who fulfill the criteria for admission were admitted to the TFC, for example, if the W/H was <70% of the median WHO growth standard, if the MUAC was found to be less than 11 cm, or if the children had bilateral pedal edema. The data were obtained by reviewing records from the inpatient therapeutic feeding registration book and individual follow-up chart.
2.2. Sample Size Determination. From a total of 420 children with SAM admitted and treated at the therapeutic feeding center (TFC) of the hospital from January 2013 to December 2017, 398 were selected. After evaluation, 22 of them were excluded as they did not have admission and discharge date.

Data Collection
Tool and Procedure. Data were extracted for one month by trained nurses using data extraction sheet prepared in the English language after checking all available data on the severe acute malnutrition inpatient registration book and medical patient charts. The data extraction sheet crosschecked with preestablished known source study variables. Lists of charts of study participants during the study period were taken from the card room, and the charts were coded. Finally, charts which have complete baseline data were selected and variables were recorded. The data extraction form was developed after reviewing relevant literature from the standard treatment protocol for the management of severe acute malnutrition, SAM registration log book, and monitoring multicharts. It includes child sociodemographic characteristics (age, sex, and residence), anthropometric measurements (height, weight, and MUAC), comorbidities, and treatment outcomes of the child.

Variables of the Study.
The dependent variable of the study is time to recovery from SAM (i.e., the event of interest is recovery and the response variable is rate of recovery). The independent variables considered are child sex, age, residence, comorbidities (TB, HIV, cough (pneumonia), anemia, malaria, diarrhea, vomiting, hypoglycemia, dehydration, hypothermia, septic shock, vaccination status, and dermatitis (skin lesion)), type of malnutrition (marasmus, kwashiorkor, and marasmus-kwashiorkor), and routine drugs (folic acid supplement, amoxicillin, and deworming medication).

Operational
Definitions. Recovered: weight for height greater than or equal to 85% of the median WHO growth chart reference, absence of bilateral pitting edema, and no medical complication [21] Time to recovery: time to recovery was obtained by calculating the difference (in days) from the start of treatment until the child recovered [13] Censored: those who were referred, were nonresponders, were defaulted, and died Defaulters: those who leave the treatment before the child is cured/lost with unknown status, when the patient discontinued before recovery [20] 2.6. Data Management and Analysis. Data were entered into EpiData version 4.2.0.0 after checking for completeness, and then the data was cleaned and exported to Stata version 14 for analysis. Descriptive analyses for continuous and categorical data describing the cohort characteristics at baseline and during follow-up were made.
Schoenfeld residual analysis (global test) was used to show the Cox proportional hazard model assumption valid at P value = 0.2536. Predictors of time to recovery were identified using bivariable and multivariable Cox proportional hazard models (CPHM). All independent variables that had P value less than 0.25 in the bivariable model were considered candidate variables for the multivariable model. The output of the multivariable CPHM is presented using adjusted hazard ratios (AHR) with the respective 95% confidence intervals (CI). Kaplan-Meier survival estimator and log-rank tests were used to estimate median recovery time during the treatment period and to compare time to recovery between groups, respectively. The Cox-Snell plot was used to check the overall model fitness. Multicollinearity was checked using the variance inflation factor.  (Figure 2).
However, the recovery rate and time were varied with predictors, including anemia and tuberculosis (Table 4).

Predictors of Time to Recovery from Severe Acute
Malnutrition. The multivariable Cox regression analysis showed that malaria, TB, anemia, breast feeding status, and vaccination status were found to be the predictors of time to recovery from SAM. Those who had no malaria at baseline had increased chance of recovery from SAM by 1.54 (AHR: 1.54 (95% CI: 1.09, 2.17)) compared to those who had malaria.
At admission, those who had vaccination status not appropriate for the age had 27% (AHR: 0.73 (95% CI: 0.55, 3 International Journal of Pediatrics 0.95)) less chance of recovery than their counterparts. Children admitted for SAM with no anemia had 1.66 (AHR: 1.66 (95% CI: 1.23, 2.23)) times more chance of recovery compared to those who had anemia at baseline.
Children without TB infection at admission were 2.03 (AHR: 2.03 (95% CI: 1.11, 3.71)) times more likely to recover from SAM compared to those with TB infection at baseline. Children who had mixed feeding were less likely to recover as compared to those who had exclusive breast feeding (AHR: 0.59 (95% CI: 0.44, 0.81)) ( Table 5).
The median recovery time of 14 days (95% CI: 13, 15) was within the acceptable minimum international standard set for therapeutic feeding centers which is less than 28 days [22]. The finding is in line with those of studies conducted in Woliso and Gedeo zones [13,23], but it was lower than that in Tigray (21days) [11], Yirgalem (18.6 days) [24], Jimma (21 days) [14], Bahir Dar (16 days) [20], and Karat referral hospitals (26 days) [25]. The discrepancy could be due to differences in the study setting as the latter studies were conducted in referral and specialized hospitals where children with the most severe SAM cases are referred.
In this study, children who had no anemia at admission were more likely to recover early compared to their  [9,26]. The reason might be that malnourished children are more likely to have micronutrient deficiency particularly iron deficiency anemia and that intervention was not initiated as early as possible as iron damages cell membranes and makes infections worse. Besides, other medical complications also exacerbate time to recovery [27,28]. Similar to other studies reported elsewhere [9,24,29], children without TB infection at admission were 2.03 times more likely to recover from SAM as compared to those with TB infection at baseline. This could be explained by the fact that the presence of malnutrition facilitates the progress of TB infection to an active disease due to its immunosuppressive effects. Moreover, tuberculosis creates a potentially lethal cycle of worsening illness and deteriorating nutritional status, ultimately resulting in a longer recovery time [30]. This finding revealed that severely malnourished children with no malaria infection recovered early compared to children with malaria infection. This is supported by studies conducted in Niger, Amazon, and Rwanda [25,31,32]. The reason might be that malaria impairs gluconeogenesis and increases energy consumption, induces malabsorption, predisposes one to loss of immunity, and results in inability respond to standard treatments [32,33]. Severely malnourished children who had vaccination status not appropriate for the age were 27% less likely to recover earlier than their counterparts. The finding is supported by results in developing countries [34]. The impact of severe acute malnutrition will be worsened as innate and adaptive immune dysfunction leads to immune dysregulation [34,35].
Children who had history of mixed feeding were 41% less likely to recover earlier than children who had history of exclusive breast feeding. The finding is in line with a study reported by World Health Organization previously [36] and a finding from Dire Dawa, eastern Ethiopia [37]. Mixed feeding may impose additional pathogen and complicate the management of children with SAM.

Strengths and Limitations
The strength of the study was the use of the advanced statistical analysis model. Further, this study is one among few studies in Ethiopia particularly novice for the study setting that were conducted to assess treatment outcomes and factors affecting time to recovery among 6-59-month-old children with SAM admitted at the Pawi General Hospital stabilization center. Due to the nature of the retrospective and secondary data, incomplete records were seen in some variables. We were unable to explore other sociodemographic characteristics, such as maternal and paternal educational status.

Conclusion
In the study area, the overall recovery rate was low. Children whose vaccination status was not appropriate for age and who had malaria, anemia, TB comorbidities, and history of mixed feeding at admission were predictors associated with time to recovery from severe acute malnutrition. Thus, emphasis should be given to prompt and timely management of malaria, anemia, and TB and more importantly to universal vaccination coverage appropriate for age.

Abbreviations
AHR: Adjusted hazard ratios CPHM: Cox proportional hazard models CHR: Crude hazard ratio CI: Confidence intervals GDP: Gross domestic product EDHS: Ethiopia Demographic and Health Survey HIV: Human immunodeficiency virus IQR: Interquartile range IV: Intravenous MUAC: Mid-upper arm circumference NG: Nasogastric tube SAM: Severe acute malnutrition SD: Standard deviation TB: Tuberculosis TFC: Therapeutic feeding center WHO: World Health Organization χ 2 : Chi-square.

Data Availability
The datasets used and/or analyzed and in which conclusions are drawn during the current study are included in the manuscript.

Ethical Approval
Ethical clearance was obtained from the Ethical Review Committee of School of Nursing on behalf of the Institutional Review Board of University of Gondar. Information in the data abstraction tool was nonidentifying and anonymous. The files of entered data in the software and the final result of this study were tied with passwords. Confidentiality and privacy of the information were kept safe throughout the whole process of the research work.