Acute lower respiratory tract infections, particularly pneumonia, are the leading cause of childhood morbidity and death in developing countries such as Bangladesh.1 Acute respiratory tract infection causes more than 2 million child deaths worldwide each year, mostly from pneumonia, and 90% of them occur in less-developed countries.2^–4 Recent estimates suggest that 1.9 million (95% CI 1.6 million to 2.2 million) children died from acute respiratory tract infection throughout the world in 2000, and 70% of them occurred in Africa and Southeast Asia.5 In Bangladesh, acute lower respiratory tract infections account for 25% of deaths in the under 5 age group and 40% of all infantile deaths.6 Depending on clinical presentation, pneumonia can be classified as very severe, severe or non-severe, with specific treatment guidelines available for each.7^–10 It is recommended that children with severe or very severe pneumonia be hospitalised for supportive treatment, including suction, oxygen therapy for hypoxaemia, fluid and nutritional management, and close monitoring.7^–10 In Bangladesh, there are not enough hospital beds for admission of all severe and very severe cases of pneumonia. In addition, hospitalisation may not be possible because of the inability of parents to visit the hospital. It is therefore important to provide institutional care for children who cannot be hospitalised. A prospective observational study was conducted to examine the feasibility of day-care-facility-based, modified primary care as an alternative for children denied hospital admission who would otherwise be sent home.

METHODS

Setting

This was a collaborative study involving the ICDDR,B, Radda Maternal and Child Health (MCH)-Family Planning (FP) Centre (Radda Clinic), and the University of Basel, Switzerland. The study was approved by the research and ethical review committees of ICDDR,B, and was conducted from June 2003 to May 2005 at the Radda Clinic, Mirpur, Section 10, Dhaka, about 5 km from the Dhaka Hospital, ICDDR,B. The Radda Clinic is a non-governmental organisation which has provided MCH services since 1974 in the Mirpur area of the metropolitan Dhaka City Corporation with a population of about 1.5 million. It operates from 08:00 to 16:00 daily, except for weekends and holidays, and provides primary care for common childhood illnesses, including pneumonia, on an outpatient basis. After triage, children with severe or very severe pneumonia are referred to either the Institute of Child Health and Shishu Sasthya Foundation Hospital (ICHSH) or the Dhaka Shishu Hospital (DSH) for admission; however, many cannot be admitted because of shortage of beds. The ICHSH and DSH are about 1 and 3 km, respectively, from the Radda Clinic. Adequate space for 12 beds in two separate rooms was established at the clinic. Additional staff, comprising a doctor, two nurses, and four health workers, were hired to operate the clinic every day of the week to provide care for children from 08:00 to 17:00, remaining on call beyond these hours. The health workers were trained to prepare and administer diets to the children, and educate and motivate mothers to comply with treatments and follow-up. Provisions for oxygen therapy and electric suction therapy, a pulse oximeter and a weighing scale were made available at the clinic.

Inclusion and exclusion criteria

Children of either sex aged 2–59 months with severe or very severe pneumonia according to WHO criteria (box 1),7^–10 who had been refused admission to ICHSH and DSH because of lack of beds after proper referral, were enrolled in the study after parental consent had been obtained. Health workers accompanied the children during referral to ICHSH/DSH, and also brought them back to the Radda Clinic after refusal of admission. Children with a history of taking antibiotics for pneumonia during the illness and those with associated co-morbidities such as tuberculosis, congenital heart disease (CHD), bronchiolitis, bronchial asthma, severe malnutrition (<−3 weight-for-age Z score), sepsis, hypoglycaemia, convulsion and meningitis were not enrolled. Children who lived a long distance (>5 km) from the clinic and those who presented late in the day (after 14:30) were also excluded from the study.

Case management

Parents brought their children to the Radda Clinic at 08:00 and took them home at 17:00 seven days a week. Thick secretions in the throat and nostrils were removed by gentle suction with an electric suction machine. Oxygen saturation was routinely measured in every child after enrolment in the study before any oxygen therapy was given. Oxygen was administered via nasal cannulae to all hypoxaemic children with oxygen saturation <95%11 in room air, as recorded by the pulse oximeter. Oxygen saturation was routinely monitored in every hypoxaemic child receiving oxygen therapy at intervals of 30 min to 2 h, depending on the patient’s condition, until oxygen saturation remained stable at ⩾95% in room air. It was rechecked during oxygen therapy as well as after removal of oxygen for 2–5 min. Children who were still hypoxaemic at 17:00 were referred to Dhaka Hospital, ICDDR,B or ICHSH for admission and continued care.

All children received an intramuscular injection of ceftriaxone once a day at a dose of 75–100 mg/kg for at least 5 days, as it has been used successfully for outpatient treatment of the most severe bacterial pneumonia in children and because of its single daily dose which can easily be administered during daily clinic visits.12

Feeding of the children had two components, one at the clinic and the other at home during the night. Infants (7–11 months) and children received a locally produced milk-based diet (milk-suji; 67 kcal/100 ml and 1.4 g protein/100 ml) every 2 h with at least four feeds administered between 08:00 and 17:00 at the clinic. Breast feeding was continued for breastfed children, and infant formula (68 kcal/100 ml and 1.5 g protein/100 ml) was given to non-breastfed infants aged 2–6 months. Mothers were provided with three or four feeds of milk-suji/infant formula in a hot pot to feed their children at night. No child received nasogastric tube feeding at the clinic or at home during the night.

Children who failed to attend in the morning were visited at home by a health worker, who brought them back to the clinic.

The above management continued every day until there was clinical improvement, defined as the child becoming afebrile, no fast breathing, no lower chest wall indrawing, no danger signs, and no rales on auscultation. After successful management, children were discharged from the clinic with advice for follow-up. Those who failed to improve with day-care management were referred to Dhaka Hospital, ICDDR,B or DSH for admission.

Follow-up

Parents were asked to bring their children to the Radda Clinic every 2 weeks for a period of 3 months for examination by the study doctor. Children who failed to attend a follow-up date were visited at home by a health worker and brought back to the clinic. During follow-up visits, morbidity (respiratory, diarrhoeal, or other) data were collected, treatment advice was given, and anthropometric indices were recorded. Any child who developed pneumonia, diarrhoea, or other complications requiring hospitalisation during follow-up visits were referred to Dhaka Hospital, ICDDR,B or DSH.

Data analysis

Success of day-care management was defined as improvement in clinical condition without referral to hospital, full compliance with the day-care management until recovery, without premature discontinuation of the study for any reason, and not dying during the study. All data were collected on case report forms, edited, entered into a personal computer, and analysed using statistical software (SPSS V10; SPSS Inc, Chicago, Illinois, USA). The main outcome measures of the study were the proportion of successes and failures of day-care management with 95% CI. Other outcome measures were proportion (with 95% CI) of patients requiring referral to hospital, and proportion (with 95% CI) discontinuing the study prematurely without fulfilling the criteria for success.

RESULTS

A total of 557 children were screened, 306 of whom were not enrolled in the study for various reasons (fig 1). Of the 306 children not enrolled, seven did not have pneumonia, 92 had pneumonia (non-severe), 70 had severe pneumonia, 57 had very severe pneumonia according to the WHO criteria,7^–10 and the remaining 80 had bronchiolitis according to the clinical assessment of the study doctor and the paediatricians working at the ICHSH and Radda Clinic. The 70 children with severe pneumonia were not enrolled because 13 refused to give consent, 14 had contagious diseases such as measles and pulmonary tuberculosis, 13 had previously received antibiotic treatment, nine had CHD, nine lived a long way from the clinic, eight presented late in the day, two were less than 1 month old, and two had no identified cause. The 57 children with very severe pneumonia were not enrolled because 48 were severely malnourished, three were less than 1 month old, one presented late in the day, three had CHD, and two had no specific reason.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1 Trial profile. VSD, ventricular septal defect.

A total of 251 children with severe or very severe pneumonia were enrolled at the Radda Clinic from the Mirpur and surrounding communities. Tables 1 and 2 show the baseline and clinical characteristics, respectively, of the study children. On examination, 143 (57%) children were hypoxaemic, with a mean (SD) oxygen saturation of 93 (4)% in room air, which was corrected by oxygen therapy to 98 (3)% (table 2). The mean (SD) duration of oxygen therapy required to correct hypoxaemia was 3 h 40 min (1 hour 45 min). Of the 143 children with hypoxaemia, only three required referral to the Dhaka Hospital, ICDDR,B, and one to the ICHSH, as they remained hypoxaemic at the end of the first day of enrolment (17:00) at the Radda Clinic.

Day-care management was successful in 234 children (93% (95% CI 89% to 96%)). Of the remaining 17 children (7% (95% CI 4.3% to 10.6%)), 11 (4.4% (95% CI 2.5% to 7.7%)) required referral to hospital and six (2.4% (95% CI 1.1% to 5.1%)) discontinued treatment (table 3). Reasons for referral to hospital were mostly respiratory distress with hypoxaemia in 10 and heart failure secondary to ventricular septal defect (VSD) in one (fig 1). Reasons for withdrawing from the study were: two families left the city; one family moved to live >5 km from the clinic; one family had a sick child at home; one mother could not spare time from work; one family withdrew for personal reasons (fig 1). There were no deaths during the day-care period, but four children (1.6% (95% CI 0.6% to 4%)) died during the 3-month follow-up period: two at the Dhaka Hospital, ICDDR,B, one each due to very severe pneumonia with hypoxaemia and hospital-acquired sepsis; the cause of death of two children could not be determined as they died at home and the information was collected long after the events occurred. Of the 234 children successfully discharged, 11 (4.7% (95% CI 2.6% to 8.2%)) were referred to hospital during the 3-month follow-up period because of respiratory distress with hypoxaemia in five, VSD with cyanosis in two, and one child each with severe pallor (haemoglobin = 7.8 g/100 ml), pulmonary tuberculosis, vesical calculus, and pneumonia with severe malnutrition. The VSD cases were not diagnosed clinically on enrolment as the children had additional respiratory sounds due to severe respiratory distress that masked the cardiac murmurs, which became evident on subsequent days. Diagnosis of pulmonary tuberculosis in one child and development of severe malnutrition in another occurred during the follow-up period.

DISCUSSION

The results clearly show that a select group of children with severe or very severe pneumonia, without associated co-morbidities such as severe malnutrition, can be safely managed on a day-care basis in resource-poor countries where hospital beds are scarce, such as Bangladesh.

The main advantages of the day-care model evaluated in this study are: (a) it is an attractive alternative because of easy replication at most urban and rural outpatient clinics and day-care centres, with slight modification of the existing staffing, including adequate training, motivation and provision of some logistic support; (b) lower cost than hospitalisation. However, the logistics and acquisition of supplies would require additional funding, which clinics may find difficult to acquire.

That none of the children who were managed solely on a day-care basis died is very reassuring; however, this may be, at least in part, related to exclusion of children with associated co-morbidities such as severe malnutrition, sepsis, hypoglycaemia, convulsion, meningitis, CHD and tuberculosis according to protocol guidelines. Children were recruited from the outpatient department of the Radda Clinic, leading to a selection bias, as the more sick children reported directly to a hospital. Our study children were managed by adequately trained research staff who were under greater supervision than normal and thus likely to have been more motivated and worked with greater dedication. A better staff to patient ratio than normal may also have played a role in a country where the ratio is usually sub-optimal.

Our high success rate can be explained by the possibility of less severity of illness, as only 57% of the children were hypoxaemic. These received oxygen therapy for a few hours, and most were no longer hypoxaemic by the end of the first day (17:00) and went home.

A major limitation of our study is that this was not a randomised controlled trial, and thus the efficacy of the management could not be directly compared with hospital care. The absence of a true control group weakens any conclusions drawn, but the use of a control group would not have been ethical under the circumstances.13 14 Our encouraging results suggest the need for a randomised, controlled clinical trial to prove or disprove them. Any future randomised, controlled clinical trial should include a component to assess the cost-effectiveness of the interventions, as this would be important in selecting the intervention for wider implementation in national programmes. Such a study is ongoing at the Radda Clinic and ICHSH as the two primary sites for day-care management and inpatient management, respectively, of children with severe pneumonia.

CONCLUSIONS

The results of this study indicate that severe and very severe pneumonia without associated co-morbidities such as severe malnutrition in children can be successfully managed on a day-care basis at established day-care clinics, if adequately trained and motivated staff and the necessary logistic support can be made available. The death of four children and referral of an additional 11 children during the 3-month follow-up period indicates the importance of follow-up for early detection of medical problems to prevent deaths.