Health and nutritional status of children hospitalized during the COVID-19 pandemic, Bangladesh

Abstract Objective To compare the health and nutrition of children younger than 5 years admitted to hospital during and before the coronavirus disease 2019 (COVID-19) pandemic in Bangladesh. Methods We collected data from hospital records of children 0–59 months admitted to the Dhaka Hospital of the International Centre for Diarrhoeal Disease Research, Bangladesh in March 2020–February 2021 (COVID-19 period; n = 2552) and March 2019–February 2020 (pre-COVID-19 period; n = 6738). Data collected included sociodemographic, anthropometric, clinical and biochemical characteristics. We compared these data for child admissions in the COVID-19 and pre-COVID-19 periods, including infants 0–11 months born during and before the pandemic and admitted to hospital. Findings Admissions of children as a percentage of total admissions were lower in March 2020 (2.47%; 63/2552) than March 2019 (8.30%; 559/6738), but increased to 20.61% (526/2552) in February 2021, three times greater than in the pre-COVID-19 period (6.69%; 451/6738). Children admitted during the COVID-19 period were significantly more likely to have dehydration, severe sepsis or septic shock, convulsions, hypernatraemia and raised creatinine than children admitted before the pandemic (P < 0.05). In infants < 6 months and those born during the pandemic, stunting and wasting were significantly higher than in infants in the pre-COVID-19 period (P < 0.05). The risk of death was higher in infants < 6 months during the pandemic (odds ratio: 1.66; 95% confidence interval: 0.95–2.92). Conclusion During the pandemic, children presented with more severe illness and poorer nutrition. Efforts are needed to reduce the adverse effects of the pandemic on the health and well-being of children.


Introduction
The growing global incidence of coronavirus disease 2019 (COVID-19) caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has led the World Health Organization (WHO) to acknowledge COVID-19 as a public health emergency. 1 The unprecedented global, social and economic crisis created by the pandemic has brought unpredictable threats to the nutritional status and survival of young children in low-and middle-income countries. 2 Childhood malnutrition, including wasting, is expected to increase as a result of sudden decreases in household incomes, food insecurity and disruption of the health-care system. 3 Before the start of the COVID-19 pandemic, 47 million children younger than 5 years in the world were wasted. 4 Estimates suggest that during the first year of the pandemic, a further 6.7 million children will have suffered from wasting and an additional 10 000 children will have died. 2 The COVID-19 pandemic is predicted to result in a reduction in health service coverage of essential pregnancy and newborn care in 132 low-and middle-income countries. 5 Recent research has shown that COVID-19 will continue to disrupt health and economic indicators worldwide, including progress in maternal and child nutrition. Based on the MIRAGRODEP model, a multicountry, multisector world economy equilibrium model, the COVID-19 pandemic will result in a 20% increase in global poverty. 6 To determine how much the COVID-19 pandemic will affect the health and development of children requires an evidence-based approach. 7 Currently, we lack evidence on the influence of COVID-19 on morbidity and nutrition in children younger than 5 years, including infants younger than 6 months and children born during the pandemic. These infants not only have the greatest growth velocity and a unique physiology, 8 but are also vulnerable to a lack of adequate nutrition. 9 In our hospital we observed that infants admitted to the hospital during the COVID-19 pandemic were sicker than infants admitted before the pandemic.
In this study, we aimed to assess changes in the health and nutritional status of children younger than 5 years and infants younger than 6 months admitted to hospital during the COVID-19 pandemic compared with the health and nutritional status of children of the same age admitted in the pre-COVID-19 period. In addition, we aimed to assess the severity of illness of children who were born and admitted to hospital during the COVID pandemic.

Study site
The Dhaka Hospital of the International Centre for Diarrhoeal Disease Research, Bangladesh (icddr,b) is the world's largest diarrhoeal disease treatment facility 10 and provides free treatment to about 200 000 patients annually. This hospital has advanced laboratory facilities capable of performing all types of routinely prescribed diagnostic tests. All units of the hospital were in operation during the COVID-19 pandemic. On admission to the hospital, children are assessed by triage nurses Research Children hospitalized during the COVID-19 pandemic, Bangladesh Sharika Nuzhat et al.
for dehydration and co-morbidities. Children without dehydration or comorbidities are moved to the outpatient department, and those with dehydration but without co-morbidities are shifted to the short-stay unit. Children with dehydration and co-morbidities, including electrolyte imbalance, enteric fever or severe malnutrition, are transferred to inpatient wards, which include the longer stay unit and the intensive care unit. The intensive care unit is equipped with invasive and non-invasive ventilation and other facilities for management of critically ill patients.

Study design and data collection
This study was a cross-sectional study of the patient records of children admitted to the hospital from March 2019-February 2021. We collected data on sociodemographic characteristics (age, sex, type of delivery, immunization status and breastfeeding), anthropometric indices (stunting, wasting, and severe acute malnutrition), clinical characteristics (acute diarrhoea, dehydration, fever, convulsions, pneumonia, severe pneumonia, sepsis, severe sepsis or septic shock, hospital-acquired infection and death) and biochemical measures (hypernatraemia, hyponatraemia, hyperkalaemia, hypokalaemia and raised creatinine) of the children.
We assessed nutritional status based on z-scores, calculated according to the WHO 2006 growth standards. 11 We defined severe acute malnutrition in children of 0-59 months as weight-forlength or -height z-score < −3 standard deviations (SD) or by the presence of bilateral pedal oedema, irrespective of anthropometric indicators. We defined underweight as weight-for-age z-score < −2 SD, and stunting as length-or height-for-age z-score < −2 SD.
We categorized the children into two time periods: children admitted in the pre-COVID-19 period (March 2019-February 2020) and children admitted in the COVID-19 period (March 2020-February 2021). We compared the characteristics of the children in the two periods. Fig. 1 shows the numbers of children and infants included in the study in the pre-COVID-19 period and COVID-19 period.

Statistical analysis
We report categorical variables as numbers and percentages and normally distributed quantitative variables as mean and SD. We used the χ 2 test to compare categorical variables between the groups. We considered the COVID-19 pandemic period as the independent variable and the dependent variables as: anthropometric indices, acute watery and/or invasive diarrhoea, dehydration, fever, convulsions, pneumonia, severe pneumonia, sepsis, severe sepsis or septic shock, hospital-acquired infection, death and abnormal electrolyte status.
We used logistic regression analysis to assess the strength of associations between the COVID-19 pandemic and anthropometric indices, morbidities and mortality after adjusting for potential confounders (age and sex), reported as odds ratios (ORs) and 95% confidence intervals (CIs). We considered P < 0.05 statistically significant.
We used Stata version 13.0 IC (StataCorp, College Station, United States of America) for all analyses.

Ethical considerations
We retrieved data from the electronic database of patient records of the hospital. The institutional review board of the International Centre for Diarrhoeal Disease Research, Bangladesh approved the study. We de-identified data for analysis and publication.

Results
In total, 6738 and 2552 children younger than 5 years were admitted to inpatient wards during the pre-COVID-19 and COVID-19 periods, respectively. From March to August 2020, the hospital admission rate was significantly lower compared with the same time of the year in 2019. This decrease was due to the movement restrictions imposed during this period. From September 2020, with the removal of movement restrictions, hospital admissions gradually increased and in March 2021, admissions were three times higher than admissions in March 2019. Fig. 2 shows the monthly hospital admissions in the pre-COV-ID-19 and COVID-19 periods. As data were collected from hospital records, information was missing for some variables. We calculated percentage value on the basis of available records.

Children
Characteristics of the children younger than 5 years admitted to hospital during and before the COVID-19 pandemic are compared in Table 1. Children admitted to the hospital in the pandemic period were significantly younger than those admitted in the pre-pandemic period (11.11 months; SD: 8.54 versus 11.59 months; SD: 9.65; P = 0.020). Significantly more children admitted in the COVID-19 period had been delivered by caesarean section than children born before the pandemic (48.87%; 887/1815 versus 45.74%; 1875/4099; P = 0.026).
In age-and sex-adjusted logistic regression analysis, the odds of dehydration

Infants
During the COVID-19 period, 700 infants younger than 6 months were admitted to hospital, which was lower than the 1937 infants admitted in the pre-COVID-19 period (Table 1). Infants in the COVID-19 period were significantly younger than those admitted in the pre-COVID-19 period (3.44 months; SD: 1.48 versus 3.63 months; SD: 1.49; P = 0.004). The proportion of infants delivered by caesarean section was comparable between the two periods, 50.36% (279/554) for the COVID-19 period and 49.02% (600/1224) for the pre-COVID-19 period. Breastfeeding and immunization in these young infants were comparable in the pre-COVID-19 period and COVID-19 period.

Infants aged 0-11 months
Of the children under 5 years admitted to hospital, 1195 were born in the CO-VID-19 period and were aged 0 to 11 months. We compared these infants with 4467 infants of the same age group (0-11 months) admitted in the pre-COVID-19-period (Table 3).
Infants 0-11 months admitted during the COVID-19 period were significantly younger at admittance than infants of the same age admitted in the pre-COVID-19 period (5.59 months; SD: 2.80 versus 6.48 months; SD: 2.95; P < 0.001). Significantly greater proportions of infants 0-11 months admitted in the COVID-19 period than those admitted in the pre-COVID-19 period were stunted and wasted, dehydrated, had convulsions, had sepsis and severe sepsis, and had hypernatraemia and raised creatinine (P < 0.05) ( Table 3). In logistic regression analysis, with adjustment for age and sex, infants born during the COVID-19 period compared with those born in the pre-COVID- 19 Table 4). However, the odds of severe pneumonia (OR: 0.77; 95% CI: 0.62-0.94) and hyponatraemia (OR: 0.50; 95% CI: 0.39-0.64) were significantly lower in the infants born and admitted in the COVID-19 period. The likelihood of death was not significantly higher in the infants in the COVID-19 period compared with infants in the pre-COVID period (OR: 1.12; 95% CI: 0.67-1.86).

Discussion
Our study findings show that CO-VID-19 is already affecting the health of children, especially young infants, with a significantly higher percentage of critically ill children being treated at our facility in the COVID-19 period compared with during the pre-COVID-19 period. Moreover, compared to the pre-COVID-19 period, these infants had a higher proportion of deaths, although the difference was not statistically significant.
We found a marked reduction in admission of children between April 2020 and August 2020. After the detection of the first COVID-19 case in Bangladesh, 12 the country imposed a complete stay-at-home order for 2 weeks from 26 March 2020. 13 The stay-at-home order was extended to 30 May 2020, 14 and was then followed by restrictions on movement and limited business hours. 15 Restrictions on movement were officially lifted on 1 September 2020. As a result, admission of children to hospital gradually increased from September 2020 onwards in the COVID-19 period. Admissions of infants in February 2021 were three times higher than in the corresponding pre-COVID-19 period of 2019.
An inpatient survey in an Australian paediatric hospital suggested that more than one third of parents delayed medical care visits for their children from fear of COVID-19. 16 Other studies from different parts of the world have also reported such delays. 17,18 The COVID-19 pandemic is likely to have affected health-care systems in many ways. The government of Bangladesh is concerned about reduced coverage and quality of maternal and child health services, but little evidence is Children hospitalized during the COVID-19 pandemic, Bangladesh Sharika Nuzhat et al.
available on health service provision, use or adaptation during COVID-19. 19 The most important factors identified for the disruption of the health system are disturbances to the livelihood of people from the stay-at-home order and related restriction measures, and the lack of protection measures for health-care workers. 20 A recent study documented a decline in visits to health facilities for regular antenatal care during April and May 2020 in Bangladesh compared with the same months in 2019. 20 Likewise, visits to family planning clinics and child immunization centres also declined. 20 Our findings on children born during the COVID-19 period also showed lower proportions of immunization in  21 A study on contextual factors influencing maternal, neonatal and child health care in Bangladesh, Nigeria and South Africa also reported that during the initial period of the pandemic, the use of health facilities for normal deliveries in Bangladesh decreased, which was attributable to more home births. 20 However, we observed a higher proportion of children delivered by caesarean section during the pandemic period than those born during the pre-COVID-19 period, although the difference was not statistically significant. Our analysis also showed that during COVID-19, breastfeeding improved in children younger than 5 years compared with the pre-COVID-19 period. We had no data on the duration of exclusive breastfeeding, but the use of breast-milk substitutes in Bangladesh reportedly increased during COVID-19. 21 Health services for children younger than 5 years have declined significantly in Bangladesh due to the COVID-19 pandemic. 21 The use of health services for children of this age group in March 2020 fell by 25% compared with service use in March 2019. 21 Research on the indirect effect of COVID-19 in 118 low-and middle-income countries, based on the worst of three scenarios, suggested that even a minor disruption to health-care services could increase childhood wasting by 10%. 22 Our study supports these findings. Among the young infants, wasting and stunting was significantly higher in the pandemic period and a greater proportion of children born during the pandemic were wasted compared with children in the same age group born in the pre-COVID-19 period.
National nutrition strategies had been implemented in Bangladesh to reduce the rates of maternal and child undernutrition, but problems with coordination resulted in poor delivery of nutrition services. 23 The uptake of maternal and newborn health services decreased by about 19% during the pandemic. 21 Various initiatives have been adopted by the public and private sectors to deal with the adverse effects of the COVID-19 pandemic on the healthcare system. 19 Development partners including local nongovernmental organizations supported the National Nutrition Services of Bangladesh to develop the national guidelines on continuing essential nutrition services during the COVID-19 pandemic. 19 It was anticipated that disruption to essential services may result in a 37% increase in child mortality in Bangladesh by 2021. 21 In our analysis, although admissions of children to hospital fell in the pandemic period, a significantly higher percentage of the children admitted were critically ill with dehydration, sepsis, convulsions and electrolyte imbalances. During the pandemic there was a higher mortality than during the pre-pandemic period, although the difference was not statistically significant. This reduction in hospital admissions has raised concerns about late presentation of critically ill children. 24 The COVID-19: coronavirus disease 2019; aOR: adjusted odds ratio; CI: confidence interval; SD: standard deviation. a OR adjusted for age and sex. b Length-or height-for-age z-score < −2 SD. c Weight-for-length or -height z-score < −2 SD. d Weight-for-length or -height z-score < −3 SD or by the presence of bilateral pedal oedema, irrespective of anthropometric indicators. e Serum sodium ≥ 150 mmol/L. f Serum sodium < 135 mmol/L. g Serum potassium > 5.5 mmol/L. h Serum potassium < 3.5 mmol/L. i Serum creatinine > 35 μmol/L for < 12 months and > 65 μmol/L for 12-59 months. Note: Reference categories are children admitted to hospital before the COVID-19 pandemic, March 2019-February 2020.
higher incidence of severe illness among children in the COVID-19 period than in the previous year may be a result of delays by caregivers in bringing their children to hospital because of strict movement restrictions, the nationwide stay-at-home order and the fear of contracting COVID-19. In an electronic survey on 24 April 2020 of 752 British paediatricians working in emergency departments and paediatric assessment units, 241 (32.05%) reported that they had observed delayed presentations. Sepsis was the second most common condition found in the children who presented late to the hospital. 25 While COVID-19 and paediatric inflammatory multisystem syndrome temporally associated with SARS-CoV-2 or multisystem inflammatory syndrome in children have attracted particular attention, we should not forget non-COVID-19 sepsis, the incidence of which is still higher in children than sepsis associated with COVID-19. 26 Our findings concur with this observation as we did not encounter any COVID-related illness in children in our hospital. In 2005 in the United States, there were 75 255 paediatric hospitalizations involving severe sepsis indicating that sepsis was common before COVID-19. 27 With regard to electrolyte imbalance, we observed that hypernatraemia and raised serum creatinine levels were the most common imbalances in the COVID-19 period. Hypernatraemia in children with diarrhoea has several causes. In our study, hypernatraemia was probably a result of delayed presentation of the children, because caregivers tried at home to treat the child with oral rehydration salts before coming to the health facility; these preparations may have been incorrectly made up or given Length-for-age z-score < −2 SD c Weight-for-length z-score < −2 SD d Weight-for-length z-score < −3 SD or by the presence of bilateral pedal oedema, irrespective of anthropometric indicators. e Serum sodium ≥ 150 mmol/L. f Serum sodium < 135 mmol/L/. g Serum potassium > 5.5 mmol/L. h Serum potassium < 3.5 mmol/L. i Serum creatinine > 35 μmol/L for < 12 months and > 65 μmol/L for 12-59 months. Notes: The COVID-19 period was March 2020-February 2021 and the pre-COVID-19 period was March 2019-February 2020. As data were collected from hospital records, information was missing for some variables. We calculated percentage value on the basis of available records.
too often or in too large a volume. 28 Another study reported that children with hypernatraemia were more likely to have convulsions 29 than children without hypernatraemia. This observation could explain why more children in our study had hypernatraemia in the COVID-19 period than the pre-COVID-19 period. Hyponatraemia is common in people with cholera because of the high loss of sodium in the stool. This condition is more common in invasive diarrhoea because of the syndrome of inappropriate antidiuretic hormone secretion. 30 Dur-ing the COVID-19 period, the higher prevalence of handwashing might be associated with the lower incidence of cholera and invasive diarrhoea compared with the pre-COVID-19 period and hence the lower incidence of hyponatraemia.