Skip to main content
Download PDF

Characteristics of children readmitted with severe pneumonia in Kenyan hospitals

BMC Public Health volume 24, Article number: 1324 (2024) Cite this article

Abstract

Background

Pneumonia is a leading cause of childhood morbidity and mortality. Hospital re-admission may signify missed opportunities for care or undiagnosed comorbidities.

Methods

We conducted a retrospective cohort study including children aged \(\ge\)2 months-14 years hospitalised with severe pneumonia between 2013 and 2021 in a network of 20 primary referral hospitals in Kenya. Severe pneumonia was defined using the 2013 World Health Organization criteria, and re-admission was based on clinical documentation from individual patient case notes. We estimated the prevalence of re-admission, described clinical management practices, and modelled risk factors for re-admission and inpatient mortality.

Results

Among 20,603 children diagnosed with severe pneumonia, 2,274 (11.0%, 95% CI 10.6–11.5) were readmitted. Re-admission was independently associated with age (12–59 months vs. 2–11 months: adjusted odds ratio (aOR) 1.70, 1.54–1.87; >5 years vs. 2–11 months: aOR 1.85, 1.55–2.22), malnutrition (weight-for-age-z-score (WAZ) <-3SD vs. WAZ> -2SD: aOR 2.05, 1.84–2.29); WAZ − 2 to -3 SD vs. WAZ> -2SD: aOR 1.37, 1.20–1.57), wheeze (aOR 1.17, 1.03–1.33) and presence of a concurrent neurological disorder (aOR 4.42, 1.70-11.48). Chest radiography was ordered more frequently among those readmitted (540/2,274 [23.7%] vs. 3,102/18,329 [16.9%], p < 0.001). Readmitted patients more frequently received second-line antibiotics (808/2,256 [35.8%] vs. 5,538/18,173 [30.5%], p < 0.001), TB medication (69/2,256 [3.1%] vs. 298/18,173 [1.6%], p < 0.001), salbutamol (530/2,256 [23.5%] vs. 3,707/18,173 [20.4%], p = 0.003), and prednisolone (157/2,256 [7.0%] vs. 764/18,173 [4.2%], p < 0.001). Inpatient mortality was 2,354/18,329 (12.8%) among children admitted with a first episode of severe pneumonia and 269/2,274 (11.8%) among those who were readmitted (adjusted hazard ratio (aHR) 0.93, 95% CI 0.82–1.07). Age (12–59 months vs. 2–11 months: aHR 0.62, 0.57–0.67), male sex (aHR 0.81, 0.75–0.88), malnutrition (WAZ <-3SD vs. WAZ >-2SD: aHR 1.87, 1.71–2.05); WAZ − 2 to -3 SD vs. WAZ >-2SD: aHR 1.46, 1.31–1.63), complete vaccination (aHR 0.74, 0.60–0.91), wheeze (aHR 0.87, 0.78–0.98) and anaemia (aHR 2.14, 1.89–2.43) were independently associated with mortality.

Conclusions

Children readmitted with severe pneumonia account for a substantial proportion of pneumonia hospitalisations and deaths. Further research is required to develop evidence-based approaches to screening, case management, and follow-up of children with severe pneumonia, prioritising those with underlying risk factors for readmission and mortality.

Peer Review reports

Background

In many low-and-middle-income-countries (LMICs), children and adolescents disproportionately face the double burden of acute respiratory infectious diseases and chronic comorbidities, resulting in long-term impairment of lung function [1, 2], reduced quality of life [3], and mortality [4]. Presence of chronic comorbidities and prior severe pneumonia episodes increase the risk of pneumonia re-admission and is associated with substantial costs [5, 6]. Almost 10% of children with community acquired pneumonia have recurrent pneumonia [7, 8]. Recurrent pneumonia is defined as at least two episodes of pneumonia in one year or three episodes ever, with interval radiographic clearing of opacities; whereas persistent pneumonia is characterized by persistence of symptoms and radiographic abnormalities for more than one month [9]. Recurrent and persistent pneumonia have varied aetiologies that are known antecedents to bronchiectasis or childhood interstitial lung disease [8, 10, 11], including rare genetic conditions that are increasingly being reported in sub Saharan Africa [12,13,14]. Obstructive airway disease in children hospitalised with severe pneumonia may be underdiagnosed, and other differential diagnoses may not be considered [15, 16]. While predictors of repeated acute hospitalisation for paediatric asthma which tends to co-exist with pneumonia are known and relatively modifiable, those for severe pneumonia readmission in sub-Saharan Africa are lacking [17].

The World Health Organization (WHO) guidelines for pneumonia used in many LMICs comprise a syndromic approach to disease classification and management to enable health workers at the first level referral hospitals to provide basic care for young children [18]. However, the validity of the syndromic approach to pneumonia classification and management has yet to be tested among children presenting for severe illness after a previous episode of hospitalisation. Robust data on inpatient management of children with repeated hospitalisation for severe pneumonia are needed to guide appropriate policy recommendations on preventive, diagnostic, treatment, and follow-up measures to optimize care and outcomes.

The aims of this study were to: (i) establish the proportion of children aged 2 months to 14 years who were re-admitted with severe pneumonia in an established network of Kenyan public hospitals; (ii) determine the factors associated with severe pneumonia re-admissions in this population; (iii) describe the investigations and treatment provided to children re-admitted with severe pneumonia; and (iv) explore the risk factors associated with mortality among these children.

Methods

Reporting

The reporting of this observational study follows the Strengthening of Reporting of Observational studies in Epidemiology (STROBE) statement [19].

Ethics, consent, and permissions

The Kenya Medical Research Institute (KEMRI) Scientific and Ethics Review Unit approved the collection of the deidentified data analysed in this study and waived the requirement for individual informed consent from patients whose data were used for this research. The Clinical Information Network (CIN) is run in collaboration with the Kenya Paediatric Association, Ministry of Health, and participating hospitals with the aim to improving routine clinical documentation for quality improvement, and for observational and interventional research [20].

Study design and setting

We undertook a retrospective cohort study involving patients admitted to CIN hospitals. The CIN was initiated in October 2013 and comprised 20 purposefully selected county hospitals distributed across the densely populated central and western regions of Kenya during the period of this analysis [21, 22]. Patients are admitted by a junior clinician under the supervision of one or more hospital paediatricians. CIN implements the use of structured clinical forms for all admitted children, capturing biodemographic data, caregiver-reported medical history, including whether the patient was readmitted and vaccination status according to the national routine schedule, clinical signs, diagnoses, diagnostic tests ordered, treatment plan, and inpatient outcome. Decisions on clinical and nursing management are made at the discretion of hospital staff with access to basic laboratory and radiological services. Case management is based on the Kenya national paediatric guidelines [23] adapted from WHO guidelines [18] as a reference (Table 1). Patients are reviewed at least once daily for clinical progress. At hospital discharge, caregivers receive instructions on correct administration of continuing treatments and a follow up outpatient clinic visit is scheduled typically within two weeks.

Table 1 WHO pneumonia classification for children aged 2–59 months with cough and/or difficulty in breathing
Full size table

Study participants, data sources and variables

The study population comprised all children admitted to paediatric wards in CIN hospitals. Children hospitalised with a documented diagnosis of pneumonia were included. Children aged less than 2 months or more than 14 years, and those without severe pneumonia defined using the World Health Organization criteria between 2013 and 2021 were excluded. Trained data clerks extract deidentified individual patient data from routine hospital records into a REDCap database upon discharge [22, 24]. For this analysis, data retrieved comprised: the patient’s unique identifier, age, sex, hospital unique identifier, prior pneumonia admission (we could not compute the time from the first admission given the nature of the data), known comorbidities or risk factors such as malnutrition, anaemia clinically assessed as pallor, chronic neurological disorder, asthma, cardiac disease, renal disease, HIV, TB and sickle cell disease. We also extracted clinical data comprising management instituted during hospitalisation such as routine laboratory investigations e.g. full blood count; plain chest radiography; supportive treatment e.g. oxygen supplementation; antimicrobials treatment (first line, second line, third line antibiotics, coverage for atypical pathogens, antitubercular and antifungal medications); inhaled or systemic corticosteroids; and outcomes including discharge, referral or death. It should be noted that the structured forms used in CIN to document diagnostics and treatments only record affirmative entries (i.e., “Yes,” “Ordered,” or “Prescribed”). As a result, we inferred that entries which were not recorded on the forms were negative.

Statistical analysis

Data were exported to STATA (StataCorp. 2019. Stata Statistical Software: Release 16.1 College Station, TX: StataCorp LLC.) for statistical analysis. Descriptive statistics were summarized as frequencies with percentages for categorical data, means with standard deviations for continuous data that were normal and medians with interquartile range for continuous data that were skewed. We estimated the proportion of severe pneumonia re-admissions among children aged 2 months to 14 years hospitalised within the CIN.

A multivariable mixed-effects logistic model was fitted to determine factors associated with re-admission among patients with severe pneumonia in this clinical cohort and we reported adjusted odds ratios. CIN data are hierarchical because admission episodes are nested within 20 hospitals, thus we used a random intercept logistic regression model to account for clustering [25]. In the formula that follows, Yij denotes pneumonia readmission for the ith child admitted in the jth hospital (Yij = 1 denotes pneumonia readmission, while Yij = 0 denotes no pneumonia readmission); X1ij, through Xkij denote the k explanatory variables measured on the ith child admitted in the jth hospital (e.g., child’s age); and α 0j denotes the unobserved cluster effect. Our model had no hospital level variables. It incorporates cluster-specific random effects to account for the within‐cluster correlation of patient outcomes.

$$\mathrm{logit}(\Pr({\mathrm Y}_{ij}\;=\;1))\;=\;{\mathrm\alpha}_0\:+\:{\mathrm\alpha}_{\mathrm{oj}}\:+\:{\mathrm\alpha}_1X_{1ij}\;+\;\cdots\;+\;{\mathrm\alpha}_{\mathrm k}X_{kij}\;\mathrm{where}\;{\mathrm\alpha}_{0\mathrm j}\sim\mathrm N(0,\;T^2)$$

The assumption is made that the random effects are independent of the model covariates (X). Explanatory variables included age, sex, weight-for-age z- score (WAZ), vaccination status, previous wheeze and comorbidities including anaemia, HIV, TB, and neurological disorders. The Stata command “or: xtmelogit” was used to fit the random intercept model.

To determine factors associated with mortality among children with severe pneumonia readmissions, we fitted a multivariable mixed-effects Cox proportional hazard model (shared gamma frailty model) and reported adjusted hazard ratios with corresponding 95% confidence intervals. In the formula that follows, αj denotes the random effect associated with the j-th hospital. The shared frailty term has a multiplicative effect on the baseline hazard function: hi (t ) = h0(t ) exp(αj ) exp(Xj β) [26]. Explanatory variables included age, sex, weight-for-age z- score (WAZ), vaccination status, previous wheeze and comorbidities including anaemia, HIV, TB, and neurological disorders. We generated Cox proportional hazard regression plots for survival by type of pneumonia admission (first episode pneumonia and pneumonia re-hospitalisation); by age group (under 12 months, 12–59 months and over 5’s), gender, WAZ category (<-3 SD, -2 to -3 SD, >-2SD), vaccination and anaemia status. The Stata command “stcox” was used to fit the shared frailty Cox proportional hazard model. Censoring occurred at the time of death, discharge or referral, whichever came first.

For all models, explanatory variables were selected a priori based on existing literature, clinical relevance regarding effect on outcome, ease of identification for risk stratification, and biological plausibility. P-values were rounded to three decimal places and were deemed statistically significant if they were < 0.05 (2-sided). To handle missing data, we performed multiple imputation by chained equations generating 10 imputed datasets under the assumption that data were missing at random (MAR) [27].

Results

Out of 49,872 children admitted with pneumonia during the study period, 29,269 (59%) were excluded from the analysis. Exclusions were 27,977 children who did not have severe pneumonia, and 1,292 children aged < 2 months or > 14 years. Figure 1 shows the study population inclusion process. The main variables with missing values were gender, which was missing for 127 children (0.6%), wheeze which was missing for 545 children (2.7%), and vaccination status which was missing for 13,468 children (65%).

Fig. 1
figure 1

STROBE Flow Chart. Flow diagram of eligible study participants

Full size image

Proportion of children re-admitted with severe pneumonia

Of the population included in the analysis comprising 20,603 children aged 2 months to 14 years hospitalised with severe pneumonia over the study period, 2,274 were readmitted, giving an estimated prevalence of 11.0% (95% CI 10.6–11.5) for readmissions among children with severe pneumonia.

Characteristics of study participants

The median age of the study population was 12 months (IQR 6.0–24.0), 9,907/20,603 (48%) of children were under 12 months of age, 9,287/20,603 (45%) of the children were aged 12–59 months and 1,409/20,603 (6.8%) were aged 5–14 years. The majority of children were reported to be up to date with their vaccination status 6,950/7,135 (97%). However, most were severely or moderately underweight 17,018/20,603 (83%). Compared to children admitted for the first time with severe pneumonia, children re-admitted with severe pneumonia were slightly older (median age 14.4 months vs. 12 months), severely or moderately underweight (73.1% vs. 83.8%), were diagnosed with TB (1.8% vs. 0.7%), and neurological disorders comprising cerebral palsy, convulsive disorder or epilepsy (0.4% vs. 0.1%) more frequently (Table 2).

Table 2 Admission characteristics of study participants
Full size table

Factors associated with severe pneumonia re-admissions

Age (12–59 months vs. 2–11 months: adjusted odds ratio (aOR) 1.70, 95% 1.54 to 1.87), over 5 years vs. 2–11 months: aOR 1.85, 95% 1.55 to 2.22), severe malnutrition (WAZ <-3SD vs. WAZ> -2SD) (aOR 2.05, 95%1.84 to 2.29), moderate malnutrition (WAZ <-3SD vs. WAZ> -2SD) (aOR 1.37, 95%1.20 to 1.57), wheeze (aOR 1.17, 95% 1.03 to 1.33), and neurological disorder (aOR 4.42, 95% 1.70 to 11.48) were independent risk factors for re-admission adjusting for gender, vaccination status, any wheeze, anaemia, HIV disease and TB (Table 3). Older age was an independent risk factor for readmission in most one-year age windows when compared to age 2–11 months (Supplementary Table 1). Independent risk factors for re-admission among children aged 2–11 months were severe malnutrition, moderate malnutrition, and vaccination; and among 12–59 months were age 24–35 months, severe malnutrition, moderate malnutrition, and neurological disorder (Supplementary Tables 2, 3 and 4).

Table 3 Risk factors for readmission among study participants
Full size table

Investigations and treatment provided to study participants

Pulse oximetry was documented in 13,245/20,603 (64.3%) children. Children re-hospitalised with severe pneumonia compared to children admitted for the first time with severe pneumonia had less frequent documentation of pulse oximetry (1,305/2,274 [57.4%] vs. 11,940/18,329 [65.1%], p < 0.001). Oxygen prescription frequencies for children with severe pneumonia and documented saturations < 90% (hypoxia) in the overall group, first hospitalisation and readmission groups were 5,839/7,717 (75.7%), 5,216/6,946 (75.1%) and 623/771 (80.8%) respectively (Fig. 2). Compared to children admitted for the first time with severe pneumonia, children re-admitted with severe pneumonia had chest radiography performed more frequently (540/2,274 [23.7%] vs. 3,102/18,329 [16.9%], p < 0.001); and more prescriptions for second line antibiotics, specifically ceftriaxone (808/2,274 [35.5%] vs. 5,538/18,329 [30.2%], p < 0.001), TB treatment (69/2,274 [3.0%] vs. 298/18,329 [1.6%], p < 0.001), salbutamol (530/2,274 [23.3%] vs. 3,707/18,329 [20.2%], p = 0.003) and prednisone (157/2,274 [6.9%] vs. 764/18,329 [4.2%], p < 0.001) (Table 4).

Fig. 2
figure 2

Supplemental oxygen cascade among study participants

Percentages derive on denominators from the preceding bars

Full size image
Table 4 Diagnostics and treatments used among study participants
Full size table

Outcomes of study participants

The median length of hospital stay was 4 days IQR [2,3,4,5,6,7] for children hospitalised with a severe pneumonia for the first time; and 5 days IQR [3,4,5,6,7,8] for children readmitted with severe pneumonia. We noted that children readmitted with severe pneumonia were hospitalised for longer than seven days more frequently than those admitted for the first time with severe pneumonia (34% vs. 26%, p < 0.001). Among 18,329 children admitted with severe pneumonia for the first time, 2,354 (12.8% [12.4–13.3]) died during their hospital stay, while 269 (11.8% [10.6–13.2]) of 2,274 children readmitted with severe pneumonia died during their hospital stay (Table 5).

Table 5 Outcomes of study participants
Full size table

Risk factors associated with mortality

Age 12–59 months (adjusted HR (aHR) 0.62, 95% CI 0.57 to 0.67), male gender (aHR 0.81, 95% CI 0.75 to 0.88), severe malnutrition (aHR 1.87 95% CI 1.71 to 2.05), moderate malnutrition (aHR 1.46, 95% CI 1.31 to 1.63), complete vaccination (aHR 0.74, 95% CI 0.60 to 0.91), wheeze (aHR 0.87, 95% CI 0.78 to 0.98) and clinical pallor (aHR 2.14, 95% CI 1.89 to 2.43) were independently associated with mortality amongst children admitted with severe pneumonia when adjusted for pneumonia readmission, HIV, TB and a neurological disorder (Table 6; Fig. 3).

Table 6 Factors associated with mortality among children admitted with severe pneumonia in Kenyan hospitals within the Clinical Information Network (CIN)
Full size table
Fig. 3
figure 3

Cox proportional hazard plots for mortality among study participants

Full size image

Discussion

Our analysis revealed a high proportion of children aged 2 months to 14 years with severe pneumonia in Kenyan secondary-level health facilities had previously been admitted for care. Risk factors for re-hospitalisation for severe pneumonia are age > 12months, moderate/severe undernutrition, wheeze, and neurological disorder. The proportion of children getting plain chest radiography, second line antibiotics, TB medication, salbutamol and prednisone was modestly higher among those re-admitted compared to those hospitalised for the first time. Mortality among children hospitalised with severe pneumonia is high, but not significantly different among those readmitted compared to first admissions. Children who are < 12 months, over 5 years, female, with moderate/severe undernutrition clinical pallor or no wheeze are at a higher risk of dying.

Publications on severe pneumonia readmission in children in LMICs are limited. Conversely, financial incentives and penalties imposed by hospital readmission reduction programs seem to drive such studies in high income countries (HICs), more so among adults [28]. Re-admissions can be classified as early (\(\le\)30 days) or late (\(\ge\)31 days) [29]. Re-admissions that occur within 30-days often correspond to those that are potentially avoidable, most of which are direct or indirect complications of underlying comorbidities [30]. These are the readmissions health facilities and insurance companies are keen on reducing, and which hospitals in HICs are subject to penalties if unacceptable thresholds are reached [31]. Large longitudinal studies of various tertiary referral hospitals in the United States of America (USA), reported a 30-day lower respiratory infection specific re-admission rate and paediatric pneumonia specific re-admission rate of 2.7% [32] and 3.3% [5] respectively, almost four times lower than in our study. A more recent analysis conducted in the USA using the 2018 Nationwide Readmissions Database reported a 30-day severe pneumonia specific re-admission rate among children of 8.7%, while the re-admission rate in children with non-severe pneumonia was 4.7% [33]. A severe pneumonia readmission prevalence of 11% is higher than reports in the literature (2.7–8.7%) and is likely dependent on the context. In high income contexts such as the USA, patient level risk factors such as anaemia and malnutrition are much lower than those reported in our analysis. In addition, children in highly resourced settings may not experience system level barriers to health care such as those that may be more common in the African setting. Children in HICs have easier access to intensive outpatient facilities that are well equipped, sufficiently staffed and with adequate referral channels to tertiary/quarternary hospitals, where children with chronic conditions are likely to receive highly specialised treatment. The lower estimates found in the USA may also be attributed to pneumonia definitions, and the fact that these were strictly 30-day outcomes. All the children in our analysis had severe pneumonia based on WHO guidelines. Furthermore, we could only ascertain that they had a prior hospital admission, but we could neither determine the reason for the initial admission nor compute the time from the first admission given the nature of the data. It is imperative to obtain this time variable in future studies to report on comparable data regarding early and late pneumonia readmissions.

Hospital re-admission may signal failure in the health system performance. However, its interpretation may be complicated by observed and unobserved differences in the clinical risk of patient populations and their survival rates. For example, health facilities with low childhood mortality may have a higher unobserved volume of sick children at risk of re-admission [34]. In our study, mortality among children hospitalised with severe pneumonia was high, but not significantly different among those readmitted compared to first admissions. Indeed, hospital re-admissions are a more complex phenomenon than mortality, in that they arise from an “interplay among patient-, hospital-, community-, and environmental-level factors” [35]. Comorbidities including neurological disorders and malnutrition, were risk factors for re-hospitalisation for pneumonia in our study, findings similarly reported in the literature [5, 32]. In a cross-sectional study conducted at a Canadian tertiary hospital including almost 3,000 children, Owayed et al. reported that most children hospitalised with recurrent pneumonia have a predisposing factor, the most common being oropharyngeal incoordination [36]. This is a plausible explanation for the four-fold increased risk with severe pneumonia re-hospitalisation observed in children with a neurological disorder in our study. It calls for a high index of suspicion among clinicians from the first point of patient contact and consideration of medical or surgical interventions early. Immunosuppressive conditions have also been identified as a risk factor for pneumonia readmission in children [33]. Immunosuppression can be due to moderate/severe malnutrition which was significantly associated with readmission in this study. It is interesting that a higher proportion of children admitted for the first time with severe pneumonia were severely/moderately underweight compared to those readmitted. Notably, a higher proportion of children who were admitted for the first time with severe pneumonia were also younger, and perhaps more vulnerable to malnutrition. Moderate or severe malnutrition and neurological disorders in children with severe pneumonia should signal a need for post-admission follow up and potential re-admission. Optimizing nutrition may avert re-admission with severe pneumonia. Going forward, it will be important to conduct studies that objectively assess possible aspiration/reflux in children with neurological disorders readmitted with pneumonia in these settings perhaps using simple questionnaires at district-level health facilities [37] or/and more elaborate tests such as videofluoroscopic swallowing studies to assess swallowing [38] and scintigraphy, pH-metry and other studies to quantify gastro-oesophageal reflux in referral centres [39]. Establishing the multi-disciplinary capacity necessary to improve care for children with neurologic disorders, specifically training physiotherapists and speech therapists is an urgent requirement.

In this analysis, HIV was not associated with severe pneumonia readmission; perhaps due to timely initiation of antiretroviral therapy within prevention of mother-to-child transmission services [40]. Patient data on other conditions that present with recurrent pneumonia such as primary immunodeficiency, cystic fibrosis and primary ciliary dyskinesia that can lead to bronchiectasis [41] were not available. Unlike our study, age less than 1 year [5, 32] and male gender [32] have been identified as risk factors for re-hospitalisation for pneumonia. This could reflect differences in our study population or no actual gender risk for pneumonia rehospitalisation. There may be multiple factors that contribute to having a higher odds of readmission in older age groups compared with the 2–11-month age group. First, older children have more opportunity to be readmitted simply due to sequencing of events since readmission requires an initial admission at a younger age. Second, children may have respiratory related conditions that manifest later such as asthma, or pneumonia-related conditions/complications that develop later such as bronchiectasis. In both these cases, the exacerbations may become more apparent at an older age. Our additional analysis comparing narrower age windows (one-year intervals) was consistent with older age being an independent risk factor. Prior respiratory syncytial virus lower respiratory tract infection [42], longer index hospitalisations [5, 33], complicated pneumonia [5], and hospital case volume and teaching status [33] were other risk factors for re-hospitalisation for pneumonia identified in the literature but were not measured in this study. These are important variables that need to be examined in future studies.

Many of the children readmitted with severe pneumonia in our analysis were likely recurrent pneumonia cases based on a few plausible reasons. First, we could ascertain that they had at least two hospitalizations. Second, readmission was significantly associated with wheeze [43], malnutrition and neurological conditions [5, 32, 43], associated comorbidities that have been reported in the literature. Lastly, their median duration of hospitalisation was 5 days, with 99% of the children being hospitalized for < 30 days and 85% being discharged, signifying interval clinical improvement. This group of children with recurrent pneumonia is likely to be heterogenous [8]. Wheeze as a sign may be a proxy for asthma, as we note its positive association with readmission and negative association with mortality, suggesting a relatively less severe clinical phenotype. Baseer et al. estimated that 11.4% of children admitted with pneumonia during a one-year study in Egypt had recurrent pneumonia, a proportion similar to our study [43]. We further postulate that there may have been a smaller proportion of children with persistent pneumonia in both the first admission and readmission groups evidenced by a long hospital stay of ≥ 30 days serving as a proxy for persistent symptoms. In our analysis, almost 20% of children had a chest radiograph, with a significantly higher frequency of requests among children who were readmitted compared to those admitted with pneumonia for the first time. Data on interval assessment of chest radiographs was not available. It may be useful in future studies to examine chest radiography findings and ascertain if children readmitted with severe pneumonia have recurrent pneumonia/persistent pneumonia based on interval clearance or persistence of opacities respectively. Researchers in the future should aim to characterize children with recurrent/persistent pneumonia better as timely interventions could potentially reverse, prevent or better manage chronic lung disease [8, 44].

Similar to our findings, other studies conducted in comparable epidemiological and geographical contexts, specifically Kilifi at the Kenyan coast, Bangladesh and India, have found malnutrition [45,46,47] and incomplete vaccination [48] to be associated with mortality in children with severe pneumonia. We found clinically assessed pallor to be associated with mortality in children with severe pneumonia [47, 49], consistent with previous findings among children with non-severe pneumonia [6]. This association in children with pneumonia may be indicative of a reduced threshold for respiratory decompensation in this vulnerable population. Timely identification and management of anaemia is needed to prevent death in children hospitalised with severe pneumonia. Previous studies have described the association between higher mortality and infants with severe pneumonia [50], which we also found in our analysis. Interestingly, mortality in children over 5 years was similar to that observed among infants. Mortality among children aged 5–14 years has received little attention [51], with those in sub-Saharan Africa faring worse [52]. Older children may have attendant complex comorbidities that put them at risk for death. Evidence-based treatment algorithms for this neglected population are needed.

Approximately two thirds of children hospitalised with severe pneumonia in our analysis had documented pulse oximetry, an essential vital sign [53]. In an analysis of CIN hospitals between 2014 and 2020, Tuti et al. found that pulse oximetry was documented in only 49% of admission cases, with increasing but variable adoption in different hospitals [54]. Barriers to pulse oximetry adoption reported in the literature include lack of equipment, poor maintenance, inadequate local leadership, clinical training and pulse oximetry promotion in respective health facilities [54, 55]. Although our data suggests improvement in documented pulse oximetry use, 64% is sub-optimal. In addition to these barriers, it is possible that pulse oximetry was measured but not documented. Surprisingly, pulse oximetry was less frequently documented in children re-admitted with severe pneumonia compared to those admitted for the first time (57% vs. 65%) and could have underestimated hypoxia in this group of children. These crude associations are subject to confounding. Children who were readmitted may have been so severely ill that pulse oximetry was skipped or conversely, they may have displayed clinical characteristics that were less likely to prompt pulse oximetry assessment. Overall, oxygen was ordered for most of the children documented to be hypoxic (76%), with higher oxygen prescription frequencies among hypoxic children who were readmitted compared to those hospitalised for the first time (81% vs. 75%). This differentially lower documentation of pulse oximetry and higher oxygen prescription among hypoxic children readmitted with severe pneumonia compared to those admitted for the first time is interesting and may point to other factors that may influence management, including healthcare worker behavior regarding disease severity perception. Interrogating the supplemental oxygen cascade comprehensively and prospectively from measurement to appropriate action may be required. In addition to assessing documentation, we may need to use a robust mixed methods approach [55,56,57] that is contemporary, multidimensional and iterative. Important considerations include: (i) observing and enquiring about healthcare worker behavior; (ii) setting national/global targets for supplemental oxygen cascade documentation in admission records e.g. 95% for pulse oximetry documentation and 95% for correct oxygen supplementation in hypoxic patients; and (iii) providing timely solutions to barriers identified in the provision of pulse oximetry and appropriate oxygen supplementation.

In this study, referral of patients with severe pneumonia who had been readmitted was quite low at 3%, similar to those with a first hospitalisation with pneumonia. It is difficult to ascertain whether clinicians are reluctant to refer patients; or whether patients would honour the referral instructions. Exploring referral systems in future studies may prompt changes in patient pathways for children with severe pneumonia who have been re-hospitalised. This may inform the health facilities on the areas that could be improved, which could include increasing the number of specialists and sub-specialists in the counties through models such as the African Paediatric Fellowship Program [58]. Remote consultation [59], among other solutions that adopt a contextually appropriate reasoned diagnostic approach for children with recurrent/persistent pneumonia may be considered [8].

To the best of our knowledge, this is the first published study to report on severe pneumonia rehospitalisation in children in sub-Saharan Africa. The large sample size across multiple settings provides a contemporary representation of paediatric severe pneumonia re-admission in a high burden context, offering a foundation for research and interventions averting or mitigating the clinical sequelae of pneumonia. We used the WHO 2013 guidelines to classify severe pneumonia for all the children included in this analysis, therefore changes in national guidance did not affect our study findings. Similarly, antibiotic recommendations may not have affected our findings substantially since the population studied was hospitalized, with majority receiving intravenous antibiotics and only 6% receiving amoxicillin. The study is limited by its retrospective nature and use of routinely collected data, which may result in missing data on variables of interest, such as the interval between first admission and subsequent readmissions, challenges linking records for patients with multiple episodes of readmission, specialised modalities of testing employed (e.g., videofluoroscopy, echocardiography), relevant complications (e.g., pleural effusions, abscess, air leaks), and underestimating diagnostics and treatments received. Our results represent a conservative estimate of diagnostics and treatments available. Although the proportion of missingness of vaccination status is 65%, the proportion of missingness of other auxiliary variables in our multivariable models was minimal. Our multiple imputation approach employing 10 imputations in a large data set of > 20,000 participants under the assumption of MAR is robust in providing unbiased estimates with improved efficiency when compared to complete case analysis. There is limited evidence regarding missing data thresholds for which multiple imputation should not be performed [60, 61]. Previous analyses have examined the validity of the assumption of MAR using CIN data [62]. There is potential for misclassification of the vaccination status. If there was a systematic bias towards underreporting or overreporting vaccination status being up to date, then our analysis may have underestimated or overestimated the association respectively. It was not possible to reach patients and prospectively verify vaccination status using the mother child booklet. Nonetheless our results are consistent with what would be plausible and documented elsewhere in the literature.

Conclusion

The proportion of children re-hospitalised with severe pneumonia and their in-hospital mortality is high. Although in-hospital mortality is similar in the re-admission group and first-time pneumonia admission group, 12.8% and 11.8% respectively are substantial proportions of death in children. Age, nutritional status, wheeze and neurological disorder are risk factors of severe pneumonia recurrence; additionally, gender, incomplete vaccination and clinical pallor, are associated with increased risk of mortality. Optimal care for severe pneumonia will require carefully designed studies of clinical algorithms for the screening, case management, and follow up of children, prioritising those with underlying risk factors.

Availability of data and materials

The data utilized in this work was made available to the research team by the participating hospitals and the Ministry of Health, and thus we are not the primary data owners; our use for these routine hospital data is approved as part of a specific ethical review process. Further access to the data can be sought through a request to KEMRI Wellcome Trust Research Programmes’s Data Governance Committee through email: dgc@kemri-wellcome.org.

Abbreviations

aHR:

Adjusted Hazard Ratio

aOR:

Adjusted Odds Ratio

CIN:

Clinical Information Network

KEMRI:

Kenya Medical Research Institute

LMIC:

Low-and-middle-income country

TB:

Tuberculosis

WAZ:

Weight-for-Age Z-score

WHO:

World Health Organization

References

  1. Gray DM, Turkovic L, Willemse L, Visagie A, Vanker A, Stein DJ, et al. Lung function in African infants in the Drakenstein child health study. Impact of lower respiratory tract illness. Am J Respir Crit Care Med. 2017;195(2):212–20.

    Article  PubMed  PubMed Central  Google Scholar 

  2. Gray D, Willemse L, Visagie A, Czovek D, Nduru P, Vanker A, et al. Determinants of early-life lung function in African infants. Thorax. 2017;72(5):445–50.

    Article  PubMed  Google Scholar 

  3. Hadning I, Andayani TM, Endarti D, Triasih R. Health-related quality of life among children with pneumonia in Indonesia using the EuroQoL descriptive system value set for Indonesia. Value Health Reg Issues. 2021;24:12–6.

    Article  PubMed  Google Scholar 

  4. Jackson P, Muyanja SZ, Siddharthan T. Health equity and respiratory diseases in low- and Middle-Income Countries. Clin Chest Med. 2023;44(3):623–34.

    Article  PubMed  Google Scholar 

  5. Neuman MI, Hall M, Gay JC, Blaschke AJ, Williams DJ, Parikh K, et al. Readmissions among children previously hospitalized with pneumonia. Pediatrics. 2014;134(1):100–9.

    Article  PubMed  PubMed Central  Google Scholar 

  6. Macpherson L, Ogero M, Akech S, Aluvaala J, Gathara D, Irimu G, et al. Risk factors for death among children aged 5–14 years hospitalised with pneumonia: a retrospective cohort study in Kenya. BMJ Glob Health. 2019;4(5): e001715.

    Article  PubMed  PubMed Central  Google Scholar 

  7. Ciftci E, Gunes M, Koksal Y, Ince E, Dogru U. Underlying causes of recurrent pneumonia in Turkish children in a university hospital. J Trop Pediatr. 2003;49(4):212–5.

    Article  PubMed  Google Scholar 

  8. Montella S, Corcione A, Santamaria F. Recurrent Pneumonia in Children: A Reasoned Diagnostic Approach and a Single Centre Experience. Int J Mol Sci. 2017;18(2):296. https://doi.org/10.3390/ijms18020296.

  9. Wald ER. Recurrent and nonresolving pneumonia in children. Semin Respir Infect. 1993;8(1):46–58.

    CAS  PubMed  Google Scholar 

  10. Verwey C, Gray DM, Dangor Z, Ferrand RA, Ayuk AC, Marangu D, et al. Bronchiectasis in African children: challenges and barriers to care. Front Pead. 2022;10:954608.

    Article  Google Scholar 

  11. Nathan N, Griese M, Michel K, Carlens J, Gilbert C, Emiralioglu N, et al. Diagnostic workup of childhood interstitial lung disease. Eur Respir Rev. 2023;32:167.

    Article  Google Scholar 

  12. Erjaee A, Bagherpour M, Van Rooyen C, Van den Berg S, Kinnear CJ, Green RJ, et al. Primary immunodeficiency in Africa - a review. S Afr Med J. 2019;109(8b):3–11.

    Article  CAS  PubMed  Google Scholar 

  13. Van Rensburg J, Alessandrini M, Stewart C, Pepper MS. Cystic fibrosis in South Africa: a changing diagnostic paradigm. S Afr Med J = Suid-Afrikaanse Tydskrif vir Geneeskunde. 2018;108(8):624–8.

    PubMed  Google Scholar 

  14. Fassad MR, Shoman WI, Morsy H, Patel MP, Radwan N, Jenkins L, et al. Clinical and genetic spectrum in 33 Egyptian families with suspected primary ciliary dyskinesia. Clin Genet. 2020;97(3):509–15.

    Article  CAS  PubMed  Google Scholar 

  15. Ardura-Garcia C, Kuehni CE. Reducing childhood respiratory morbidity and mortality in low and middle income countries: a current challenge. Eur Respir J. 2019;54(1):1900987.

    Article  PubMed  Google Scholar 

  16. Nguyen PTK, Tran HT, Fitzgerald DA, Tran TS, Graham SM, Marais BJ. Characterisation of children hospitalised with pneumonia in central Vietnam: a prospective study. Eur Respir J. 2019;54(1):1802256.

    Article  CAS  PubMed  Google Scholar 

  17. Ardura-Garcia C, Stolbrink M, Zaidi S, Cooper PJ, Blakey JD. Predictors of repeated acute hospital attendance for asthma in children: a systematic review and meta-analysis. Pediatr Pulmonol. 2018;53(9):1179–92.

    Article  PubMed  PubMed Central  Google Scholar 

  18. WHO Pocket book of hospital care for children: guidelines for the management of common illnesses with limited resources. 2nd edn. Geneva: World Health Organization; 2013.

  19. von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP. The strengthening the reporting of observational studies in epidemiology (STROBE) statement: guidelines for reporting observational studies. PLoS Med. 2007;4(10): e296.

    Article  Google Scholar 

  20. English M, Irimu G, Agweyu A, Gathara D, Oliwa J, Ayieko P, et al. Building Learning health systems to accelerate research and improve outcomes of clinical care in low- and middle-income countries. PLoS Med. 2016;13(4): e1001991.

    Article  PubMed  PubMed Central  Google Scholar 

  21. English M, Irimu G, Akech S, Aluvaala J, Ogero M, Isaaka L, et al. Employing learning health system principles to advance research on severe neonatal and paediatric illness in Kenya. BMJ Glob. BMJ Glob Health. 2021;6(3):e005300.

    Article  PubMed  PubMed Central  Google Scholar 

  22. Ayieko P, Ogero M, Makone B, Julius T, Mbevi G, Nyachiro W, et al. Characteristics of admissions and variations in the use of basic investigations, treatments and outcomes in Kenyan hospitals within a new clinical information network. Arch Dis Child. 2016;101(3):223–9.

    Article  PubMed  Google Scholar 

  23. Ministry of Health K. Basic Paediatric Protocols. 2016. 4th Edition: http://guidelines.health.go.ke:8000/media/Basic_Paediatric_Protocols_2016.pdf.

  24. Research Electronic Data. Capture (REDCap). Available from: https://www.project-redcap.org/.

  25. Austin PC, Merlo J. Intermediate and advanced topics in multilevel logistic regression analysis. Stat Med. 2017;36(20):3257–77.

    Article  PubMed  PubMed Central  Google Scholar 

  26. Austin PC. A tutorial on multilevel survival analysis: methods, models and applications. Int Stat Rev. 2017;85(2):185–203.

    Article  PubMed  PubMed Central  Google Scholar 

  27. Azur MJ, Stuart EA, Frangakis C, Leaf PJ. Multiple imputation by chained equations: what is it and how does it work? Int J Methods Psychiatr Res. 2011;20(1):40–9.

    Article  PubMed  PubMed Central  Google Scholar 

  28. Wallace SS, Quinonez RA. Solving the readmissions puzzle: how do variability and preventability fit? Pediatrics. 2017;140(2):e20171681.

    Article  PubMed  Google Scholar 

  29. Prescott HC, Sjoding MW, Iwashyna TJ. Diagnoses of early and late readmissions after hospitalization for pneumonia. A systematic review. Annals Am Thorac Soc. 2014;11(7):1091–100.

    Article  Google Scholar 

  30. Donzé J, Lipsitz S, Bates DW, Schnipper JL. Causes and patterns of readmissions in patients with common comorbidities: retrospective cohort study. BMJ. 2013;347: f7171.

    Article  PubMed  PubMed Central  Google Scholar 

  31. Medicare To Penalize 2,217 Hospitals For Excess Readmissions. KFF Health News. Updated on Oct. 12 with revised data from Medicare. https://kffhealthnews.org/news/medicare-hospitals-readmissions-penalties/.

  32. Nakamura MM, Zaslavsky AM, Toomey SL, Petty CR, Bryant MC, Geanacopoulos AT, et al. Pediatric readmissions after hospitalizations for lower respiratory infections. Pediatrics. 2017;140(2):e20160938.

    Article  PubMed  Google Scholar 

  33. Lewis MO, Tran PT, Huang Y, Desai RA, Shen Y, Brown JD. Disease severity and risk factors of 30-day hospital readmission in pediatric hospitalizations for pneumonia. J Clin Med. 2022;11(5):1185.

    Article  PubMed  PubMed Central  Google Scholar 

  34. Laudicella M, Li Donni P, Smith PC. Hospital readmission rates: signal of failure or success? J Health Econ. 2013;32(5):909–21.

    Article  PubMed  Google Scholar 

  35. Weinreich M, Nguyen OK, Wang D, Mayo H, Mortensen EM, Halm EA, et al. Predicting the risk of readmission in pneumonia. A systematic review of model performance. Annals Am Thorac Soc. 2016;13(9):1607–14.

    Article  Google Scholar 

  36. Owayed AF, Campbell DM, Wang EE. Underlying causes of recurrent pneumonia in children. Arch Pediatr Adolesc Med. 2000;154(2):190–4.

    Article  CAS  PubMed  Google Scholar 

  37. Duncan DR, DiFilippo C, Kane M, Lurie M, McSweeney ME, Rosen RL. Overlapping symptoms of gastroesophageal reflux and aspiration highlight the limitations of validated questionnaires. J Pediatr Gastroenterol Nutr. 2021;72(3):372–7.

    Article  PubMed  PubMed Central  Google Scholar 

  38. Martin-Harris B, Canon CL, Bonilha HS, Murray J, Davidson K, Lefton-Greif MA. Best practices in modified barium swallow studies. Am J Speech Lang Pathol. 2020;29(2s):1078–93.

    Article  PubMed  PubMed Central  Google Scholar 

  39. Rosen R. Novel advances in the evaluation and treatment of children with symptoms of gastroesophageal reflux disease. Front Pead. 2022;10: 849105.

    Article  Google Scholar 

  40. Mwau M, Bwana P, Kithinji L, Ogollah F, Ochieng S, Akinyi C, et al. Mother-to-child transmission of HIV in Kenya: a cross-sectional analysis of the national database over nine years. PLoS ONE. 2017;12(8): e0183860.

    Article  PubMed  PubMed Central  Google Scholar 

  41. Bush A, Floto RA. Pathophysiology, causes and genetics of paediatric and adult bronchiectasis. Respirology. 2019;24(11):1053–62.

    Article  PubMed  Google Scholar 

  42. Munywoki PK, Ohuma EO, Ngama M, Bauni E, Scott JA, Nokes DJ. Severe lower respiratory tract infection in early infancy and pneumonia hospitalizations among children, Kenya. Emerg Infect Dis. 2013;19(2):223–9.

    Article  PubMed  PubMed Central  Google Scholar 

  43. Abdel Baseer KA, Sakhr H. Clinical profile and risk factors of recurrent pneumonia in children at Qena Governorate, Egypt. Int J Clin Pract. 2021;75(4):e13695.

    Article  CAS  PubMed  Google Scholar 

  44. Patria F, Longhi B, Tagliabue C, Tenconi R, Ballista P, Ricciardi G, et al. Clinical profile of recurrent community-acquired pneumonia in children. BMC Pulm Med. 2013;13: 60.

    Article  PubMed  PubMed Central  Google Scholar 

  45. Chisti MJ, Graham SM, Duke T, Ahmed T, Faruque AS, Ashraf H, et al. Post-discharge mortality in children with severe malnutrition and pneumonia in Bangladesh. PLoS ONE. 2014;9(9): e107663.

    Article  PubMed  PubMed Central  Google Scholar 

  46. Ngari MM, Fegan G, Mwangome MK, Ngama MJ, Mturi N, Scott JAG, et al. Mortality after Inpatient treatment for severe pneumonia in children: a cohort study. Paediatr Perinat Epidemiol. 2017;31(3):233–42.

    Article  PubMed  PubMed Central  Google Scholar 

  47. Kapoor A, Awasthi S, Kumar Yadav K. Predicting mortality and use of RISC scoring system in hospitalized under-five children due to WHO defined severe community acquired Pneumonia. J Trop Pediatr. 2022;68(4):fmac050.

    Article  PubMed  Google Scholar 

  48. Kasundriya SK, Dhaneria M, Mathur A, Pathak A. Incidence and risk factors for severe pneumonia in children hospitalized with pneumonia in Ujjain, India. International journal of environmental research and public health. 2020;17(13):4637.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  49. Tiewsoh K, Lodha R, Pandey RM, Broor S, Kalaivani M, Kabra SK. Factors determining the outcome of children hospitalized with severe pneumonia. BMC Pediatr. 2009;9: 15.

    Article  PubMed  PubMed Central  Google Scholar 

  50. Sonego M, Pellegrin MC, Becker G, Lazzerini M. Risk factors for mortality from acute lower respiratory infections (ALRI) in children under five years of age in low and middle-income countries: a systematic review and meta-analysis of observational studies. PLoS ONE. 2015;10(1): e0116380.

    Article  PubMed  PubMed Central  Google Scholar 

  51. Patton GC, Azzopardi P. Missing in the middle: measuring a million deaths annually in children aged 5–14 years. Lancet Global Health. 2018;6(10):e1048-1049.

    Article  PubMed  Google Scholar 

  52. Masquelier B, Hug L, Sharrow D, You D, Hogan D, Hill K, et al. Global, regional, and national mortality trends in older children and young adolescents (5–14 years) from 1990 to 2016: an analysis of empirical data. Lancet Global Health. 2018;6(10):e1087-1099.

    Article  PubMed  Google Scholar 

  53. Zar HJ, McCollum ED. Pulse oximetry to detect paediatric hypoxaemia-the fifth vital sign. Lancet Global Health. 2023;11(11):e1684-1685.

    Article  CAS  PubMed  Google Scholar 

  54. Tuti T, Aluvaala J, Akech S, Agweyu A, Irimu G, English M. Pulse oximetry adoption and oxygen orders at paediatric admission over 7 years in Kenya: a multihospital retrospective cohort study. BMJ Open. 2021;11(9): e050995.

    Article  PubMed  PubMed Central  Google Scholar 

  55. Enoch AJ, English M, McGivern G, Shepperd S. Variability in the use of pulse oximeters with children in Kenyan hospitals: a mixed-methods analysis. PLoS Med. 2019;16(12): e1002987.

    Article  PubMed  PubMed Central  Google Scholar 

  56. Hedible GB, Louart S, Neboua D, Catala L, Anago G, Sawadogo AG, et al. Evaluation of the routine implementation of pulse oximeters into integrated management of childhood illness (IMCI) guidelines at primary health care level in West Africa: the AIRE mixed-methods research protocol. BMC Health Serv Res. 2022;22(1):1579.

    Article  PubMed  PubMed Central  Google Scholar 

  57. Graham HR, Olojede OE, Bakare AAA, McCollum ED, Iuliano A, Isah A, et al. Pulse oximetry and oxygen services for the care of children with pneumonia attending frontline health facilities in Lagos, Nigeria (INSPIRING-Lagos): study protocol for a mixed-methods evaluation. BMJ Open. 2022;12(5): e058901.

    Article  PubMed  Google Scholar 

  58. Zar HJ, Vanker A, Gray D, Zampoli M. The African pediatric fellowship training program in pediatric pulmonology: a model for growing African capacity in child lung health. Annals Am Thorac Soc. 2017;14(4):500–4.

    Article  Google Scholar 

  59. Ambrosino N, Makhabah DN, Sutanto YS. Tele-medicine in respiratory diseases. Multidisciplinary Respiratory Med. 2017;12:9.

    Article  Google Scholar 

  60. Madley-Dowd P, Hughes R, Tilling K, Heron J. The proportion of missing data should not be used to guide decisions on multiple imputation. J Clin Epidemiol. 2019;110:63–73.

    Article  PubMed  PubMed Central  Google Scholar 

  61. Lee JH, Huber JC. Jr. Evaluation of multiple imputation with large proportions of missing data: how much is too much? Iran J Public Health. 2021;50(7):1372–80.

    PubMed  PubMed Central  Google Scholar 

  62. Agweyu A, Lilford RJ, English M. Appropriateness of clinical severity classification of new WHO childhood pneumonia guidance: a multi-hospital, retrospective, cohort study. Lancet Global Health. 2018;6(1):e74-83.

    Article  PubMed  Google Scholar 

Download references

Acknowledgements

The authors would like to thank the Ministry of Health who gave permission for this work to be developed and have supported the implementation of the Clinical Information Network (CIN) together with the County Health Executives, Hospital Management Teams, the Kenya Paediatric Association, the Kenya Ministry of Health, and the University of Nairobi for promoting the aims of the CIN. Specifically, teams based in: Busia County Hospital (Emma Sarah Namulala, Yuvane Maiyo); Bungoma County Hospital (Dickens Lubanga, Felistus Makokha); Homabay County Hospital (Meshack Liru, Edith Ogada); Vihiga County Hospital (Vitalis Juma); JOOTRH (Josephine Ojigo, Maureen Muchela); Kakamega CGTRH (Boniface Nyumbile, Roselyn Malangachi, Ijusa Midecha); Kitale County Hospital (Rachel Inginia, Paul J W Njanwe, Mwende Mutunga); Kiambu County Hospital (Grace Ochieng, Lydia Thuranira and Catherine Murianki); Mama Lucy Kibaki Hospital (Cecilia Mutiso, Celia Muturi, Elizabeth Atieno Jowi); Machakos County Hospital (Charles Nzioki, Penina Musyoka); Kisumu County Hospital (Achieng Adem, Sharon Ocharo); Embu County Hospital (Esther Mukami Njiru, Mwangi Ngina, Patrick Lydia); Naivasha County Hospitals (Julie Barasa); Thika Level 5 Hospital (Bernadette Lusweti, Maureen Njoroge, Sylvia Mwathi, Patrick Mburugu); Nakuru County Hospital (Angeline Ithondeka, Linda Ombito, Alice Nkirote Nyaribari, Elizabeth Kibaru); Kerugoya County Hospital (Peninah Muthoni Mwangi); Mbagathi County Hospital (Christine Manyasi, David Kimutai, Rukia Aden, Zanuba Mohammed); Nyeri County Hospital (Wagura Mwangi, Agnes Mithamo); Pumwani Maternity Hospital (Catherine Mutinda, Beth Maina, Rashid Musa, Caren Emadau, Maureen Muirithi, Esther Muthoni Ogolla); Migori County Hospital (Joyce Akuka); Malindi Sub-County Hospital (Yusuf Rasheed); Kisii County Hospital (Mourine Ikol); and Karatina Sub-County Hospital (Stella Mbuga).

The authors also acknowledge the other members of the CIN leadership: Jalemba Aluvaala, Anthony Etyang, Samuel Akech, Muthoni Ogola, Jacquie Oliwa, Michuki Maina, David Gathara, Fredrick Were, Grace Irimu, and Mike English, as well as technical and administrative assistance provided by Metrine Saisi, Elizabeth Muthini, James Wafula, Cynthia Khazenzi, Joyce Kigo, George Mbevi, Henry Gathuri, Wycliffe Nyachiro, and Frankline Oketch. This work is published with the permission of the Director of KEMRI CGMRC.

Funding

The research reported is funded by a Joint Global Health Trials Award (MR/R006083/1) from the Department of Health and Social Care (DHSC), the Foreign, Commonwealth & Development Office (FCDO), the Medical Research Council (MRC) and the Wellcome Trust. This UK funded award is part of the EDCTP2 programme supported by the European Union.

Author information

Authors and Affiliations

  1. Paediatric Pulmonology Division, Department of Paediatrics and Child Health, University of Nairobi, Nairobi, Kenya

    Diana Marangu-Boore

  2. Epidemiology and Demography Department, KEMRI-Wellcome Trust Research Programme, Nairobi, Kenya

    Paul Mwaniki, Lynda Isaaka, Teresiah Njoroge, Livingstone Mumelo, Dennis Kimego, Conrad Wanyama & Ambrose Agweyu

  3. Department of Health, Kisumu County, Kenya

    Achieng Adem

  4. Department of Health, Nairobi County, Kenya

    Elizabeth Jowi

  5. Department of Health, Nakuru County, Kenya

    Angeline Ithondeka

  6. Department of Infectious Disease Epidemiology, London School of Hygiene and Tropical Medicine, London, Great Britain

    Ambrose Agweyu

Authors
  1. Diana Marangu-Boore
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Paul Mwaniki
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Lynda Isaaka
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Teresiah Njoroge
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Livingstone Mumelo
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Dennis Kimego
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Achieng Adem
    View author publications

    You can also search for this author in PubMed Google Scholar

  8. Elizabeth Jowi
    View author publications

    You can also search for this author in PubMed Google Scholar

  9. Angeline Ithondeka
    View author publications

    You can also search for this author in PubMed Google Scholar

  10. Conrad Wanyama
    View author publications

    You can also search for this author in PubMed Google Scholar

  11. Ambrose Agweyu
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

DM-B and AA conceived the study and DM-B performed the analysis with support from AA. DM drafted the initial manuscript with support from AA, PM, LI, TN, LM, DK, AA, EJ, AI, and CW. All authors read and approved the final version of the manuscript.

Corresponding author

Correspondence to Diana Marangu-Boore.

Ethics declarations

Ethics approval and consent to participate

The Kenya Medical Research Institute (KEMRI) Scientific and Ethics Review Unit approved the collection of the deidentified data analysed in this study. The CIN is run in collaboration with the Ministry of Health and participating hospitals with the aim to improve the quality of routine paediatric hospital data for use in audits, observational and interventional research. Individual consent for access to deidentified patient data was not required.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Marangu-Boore, D., Mwaniki, P., Isaaka, L. et al. Characteristics of children readmitted with severe pneumonia in Kenyan hospitals. BMC Public Health 24, 1324 (2024). https://doi.org/10.1186/s12889-024-18651-2

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s12889-024-18651-2

Keywords

BMC Public Health

ISSN: 1471-2458

Contact us