Epidemiology and surveillance of influenza, RSV and SARS-CoV-2 in children admitted with severe acute respiratory infection in West bengal, India from 2022 to 2023

Abstract

Background

Evaluating the burden of respiratory syncytial virus (RSV) and influenza among young children in LMICs is crucial to inform implementation policies, given the importance of maternal influenza and RSV vaccination, which may not yet be widely available.

Methods

This study established a one-year surveillance of severe acute respiratory infection (SARI) from June 2022–2023 in hospitalized children 1–24 months from rural West Bengal India. We tested nasopharyngeal swabs collected from children admitted with SARI using multiplex real-time PCR for influenza, RSV, SARS-CoV-2, with a subset (N = 81) tested for additional respiratory pathogens and analyzed clinical features, factors influencing infections, and hospitalization duration.

Results

Of 1842 children admitted with SARI, 77% (1419) were between 1 and 24 months. Of 191 sampled, 21 required intensive care, and 3 died. The majority of mothers (83.7%) were vaccinated against COVID-19, but none against influenza, pertussis, or RSV. Viruses were detected in 44% (84/191), with RSV being the most common 60/190 (31.6%), followed by influenza 12/190 (6.3%), and SARS-CoV-2 2/191 (1%). Influenza subtypes included influenza A/H3 (6/16), A/H1N1pdm (5/16), Influenza B (4/16), and Influenza C (1/16). RSV peaked during autumn, influenza during winter and monsoon. Influenza was more common in infants < 6 months (13.4%, p = 0.03). RSV affected both infants under 6 months and over similarly (34% vs. 29.6%, p = 0.5). Infants < 6 months frequently required oxygen support (p = 0.02), though ICU admissions were similar (p = 0.98). RSV was associated with 19% of ICU admissions and influenza with 14%. Additional pathogens included Haemophilus influenzae (23.45%), Streptococcus pneumoniae (22%), rhinovirus (13.6%), parainfluenza virus group (6.1%), Staphylococcus aureus (8.6%), Moraxella catarrhalis (5%), bocavirus (3.7%), adenovirus (3.7%), Chlamydia pneumoniae (1%), and Bordetella (1%). Viral-bacterial co-detection occurred in 34%, especially in infants < 6 months. Children with RSV had increased risk of having S. pneumoniae [Odds Ratio OR 6.2, 95% CI 1.8–21.3]. Rhinovirus cases were associated with ICU admission, mechanical ventilation, and longer length of stay, regardless of age.

Conclusion

RSV and influenza were the key contributors to SARI in children under-2. Findings highlight the need for diagnostics to guide vaccination, reduce antibiotic use, and improve indoor air quality for alleviating the SARI burden in rural settings.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12879-025-11421-4.

Introduction

Lower respiratory tract infections (LRI) are a leading cause of paediatric morbidity and mortality in low- and middle-income countries (LMIC) [1, 2]. LRIs also pose a substantial economic burden, from hospitalizations in low-income settings [3, 4]. In 2015, pneumonia was the second leading cause of mortality among children under-5 in India, associated with 0.191 [0.168–0.219] million deaths (15.9%) [5].

Respiratory syncytial virus (RSV) and influenza virus are the most common viral cause of LRIs in children under-5, especially in infants and young children [6, 7]. RSV infections are prevalent in children under-2 with the highest severity in infants under 6 months, those born prematurely, and those with comorbidities [6, 8–10]. Approximately 60–70% of children are infected by RSV in the first year of life, and by age two nearly all have an episode of RSV [11, 12]. In 2019, there were 3.6 million RSV-associated hospitalizations among children 0–60 months worldwide. Almost one-third (36%) of RSV associated hospitalizations, and half (51%) the mortality was in infants 0–6 months [9]. Further, the majority (97%) of deaths occurred in LMICs. Newer strategies to prevent severe RSV disease in infants under 6 months are now available. These include maternal immunization during pregnancy using RSV vaccines (Abrysvo) and passive immunization of infants with the long-acting prophylactic monoclonal antibody (mAb), Nirsevimab [13, 14]. These interventions have been implemented in several high-income countries, while access to these in LMICs may require additional regional evidence-based studies.

Influenza was associated with 10.1 million LRI episodes worldwide in 2018 among children under-5 [15]. Infants less than 6 months of age are particularly vulnerable to severe influenza, resulting in one-third (36%) of total influenza-related deaths in < 6-months-age-group, with 82% occurring in low-income (LICs) and LMICs [15].

Alongside RSV and influenza, other respiratory viruses such as rhinovirus, human metapneumovirus, parainfluenza virus, and adenovirus are important causes of acute respiratory infections in young children [16, 17]. Respiratory viral infections often predispose individuals to secondary bacterial infections commonly caused by Streptococcus pneumoniae, Haemophilus influenzae type B, and Staphylococcus aureus. [18, 19] Streptococcus pneumoniae (pneumococcus) contributes to the greatest proportion of LRIs and deaths followed by S. aureus [2, 6, 20]. S. aureus is a notable cause of complicated pneumonia and poor clinical outcomes [2, 20]. However, the presence of bacterial pathogens within the nasopharynx does not correlate with invasive disease due to high asymptomatic carriage among children [21].

Respiratory viral circulation is influenced by season and public health measures. The COVID-19 pandemic caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) was the main source of LRIs in 2020–2021, that profoundly disrupted the infection landscape of various respiratory pathogens. The non-pharmaceutical interventions implemented to mitigate the spread of SARS-CoV-2, led to a substantial decline in global LRIs and mortality from RSV and influenza [2, 22]. Post-pandemic when the mitigation measures were eased, the prevalence of influenza and RSV increased in 2022 [23, 24].

Public health strategies to reduce LRIs need rigorous studies that delineate the etiology and epidemiology of LRIs. Limited studies exist on the burden of hospitalized influenza and RSV among rural communities in LMICs, where most children live. In India the National Family Health Survey (NFHS-5) 2019-20, reported a mortality rate of 42 deaths per 1000 live births in children under-five, with West Bengal having 25 per 1000 [25]. Epidemiological studies of severe acute respiratory infections (SARI) have been studied in different parts of India [26–30]; however, data from rural communities in eastern India are limited, the majority being from northern India [31, 32]. Evaluating the disease burden in children under-two in rural areas is important, especially now with the availability of maternal RSV vaccines and mAbs [33]. While maternal influenza vaccines have been recommended by the Indian Government, they are yet to be included in the universal immunization program (UIP) [34].

Severe acute respiratory infections (SARI) are characterized by the sudden onset of fever, cough, or difficult breathing requiring hospitalization [35]. Among young children, SARI is a critical condition that necessitates clinical support and laboratory testing to identify the causal pathogens. Molecular diagnostic techniques, such as multiplex real-time PCR allow the simultaneous detection of multiple pathogens [36].

Systemic surveillance of SARI is required to assess the burden of influenza and RSV associated LRIs, identify influenza types, differentiate viral and bacterial infections, prioritize at-risk groups, and understand disease seasonality and severity. This information is needed for guiding treatment and vaccination policies to reduce the SARI burden in infants and young children. The objective of this study was to assess the burden of influenza, RSV, and SARS-CoV-2, their seasonality, coinfection patterns, association between pathogens, and infection characteristics with respect to age.

Methods

Study design

This multicentric, prospective, active hospital-based study aimed to understand the clinical and epidemiological features of influenza, RSV, and SARS-CoV-2 in children 1–24 months, hospitalized with SARI in the rural West Midnapore district of West Bengal, India, from 2022 to 2023.

Study settings

This study was conducted in the paediatric wards and paediatric intensive care unit (PICUs) of the Department of Paediatrics and Neonatology of Midnapore Medical College & Hospital (MMCH) and the Kharagpur Sub Divisional Hospital (KSDH) of West Midnapore. For SARI, patients typically first present to public or private facilities in the region. MMCH is the only public tertiary hospital of the district, located in the city of Midnapore, with two paediatric indoor units with 67 bed capacity, a PICU, and a special new-born care unit. MMCH receives referred patients from across the district and neighbouring districts, where paediatric wards are often full, and beds are frequently shared. KSDH is a secondary hospital in Kharagpur city with one paediatric indoor unit, and a 36-bed occupancy. MMCH and KSDH cater to approximately 191,430 children under-2, the data shared by the office of Chief Medical Officer of Health, W Midnapore.

Laboratory testing was conducted in the ICMR-Viral Research Diagnostic Laboratory (VRDL), at MMCH. VRDL comes under the aegis of National Institute of Virology, the National Influenza Centre. No routine influenza and RSV surveillance existed in the two hospitals prior to this study; however, during the COVID-19 pandemic, respiratory specimens were collected and tested for SARS-CoV-2 at VRDL, MMCH.

Surveillance methodology

From June 26, 2022 to July 1, 2023, systematic hospital-based active SARI surveillance was conducted in the paediatric wards of MMCH (June 26, 2022 to July 1, 2023) and KSDH (November 13, 2022 to July 1, 2023). We followed the modified World Health Organization SARI case definition for influenza and RSV surveillance defined as an illness with a measured or history of fever of ≥ 38 ˚C (100.4 ˚F), and cough, or shortness of breath with illness onset within the last 10 days, and requiring hospitalization [35]. Unlike the old WHO 2011 definition, which used “AND,” for shortness of breath, this modified version used “or”. Tactile temperatures were reported as thermometers were not available at home.

The inclusion criteria were children 1–24 months of age, fulfilling the SARI case definition, hospitalized in the last 24 h, and whose parents provided consent for participation. Inpatient SARI patients transferred to the PICU were enrolled from the PICU. Written informed consent was obtained from the guardians or caregivers of the children, and a standardized data collection form was used to collect details of illness, socio-demographics, vaccination history of the mothers, clinical course, and laboratory data. Nasal or nasopharyngeal swabs were collected by a trained laboratory technician.

The sampling strategy aimed to achieve a total of five random samples per week across 52 weeks. Sampling occurred twice weekly (Tuesdays and Fridays), excluding weekends and holidays at MMCH from June 26 (influenza week 26) to November 11, 2022 (influenza week 46), except during the July COVID-19 wave (once weekly). From November 13, 2022 onward, sampling occurred weekly at each site (MMCH: Monday/Tuesday; KSDH: Monday/Wednesday). All eligible cases were enrolled, with an average of 2 cases per visit.

Respiratory pathogen testing

Viral transport medium vials (Himedia, MS2760A) containing swabs were stored in cold ice packs in a cool box, and transported to VRDL, MMCH at 4 °C within one hour, for storage at − 80 °C. Samples were tested at VRDL by multiplex real-time polymerase chain reaction, for influenza, RSV, and SARS-CoV-2. RNA was manually extracted (HiPurA Viral RNA purification kit, Himedia) and tested for the influenza panel [A, B, A/H1N1pdm(2009) and A/H3] and RSV [A and B] using the Allplex Respiratory Panel 1 Kit, (Seegene, Seoul, Republic of Korea). SARS-CoV-2 was tested using TRUPCR SARS-CoV-2 kit, (3B Blackbio, India). A confirmed case was defined as a subject meeting the SARI case definition with a laboratory-confirmed positive result, based on the kit’s cycle threshold (Ct) cut-off of the respective kits.

A representative convenient subset of samples was tested via the TRUPCR Respiratory Pathogen Panel Kit (3B Blackbio, India) for qualitative detection of 14 bacteria and 17 respiratory viruses. The pathogens tested in this kit are described elsewhere [37]. Sampling was initially planned to be representative of each week to capture viral seasonality. However, due to funding constraints, samples collected between March-June 2023 could not be tested using the respiratory pathogen panel. RNA/DNA extraction was performed using the Total Viral Nucleic Acid Extraction Kit (3B Blackbio), and amplification was performed on CFX96 Real-time PCR detection system (Bio-Rad, USA) following the manufacturer’s instructions.

Data management and analysis

All patient data was stored and managed using CDC Epi info™ by trained staff. Access to data was restricted to the investigator and the study staff responsible for data management. Regular backup was made by exporting the data to external hard drives. For each ward, we documented the weekly total number of inpatient admissions, new SARI admissions, and SARI-related deaths using hospital records. Seasons were categorized according to the Indian Meteorological Department into summer, monsoon, postmonsoon/autumn and winter [38]. The socioeconomic standing of families was measured using the Modified Kuppuswamy Socioeconomic Scale 2021 which is based on the monthly family income, education, and occupation of the head of family [39]. The geographic location of study participants were mapped using Quantum GIS software. The nutritional status was measured by weight-for-age-z scores (WAZ). Categorical variables are expressed as frequencies (%), whereas continuous variables are presented as medians with interquartile range (IQR). Comparisons between categorical variables are made using chi-square (χ²) test and Wilcoxon rank-sum test for continuous variables. To analyse the risk factors for duration of hospitalization, linear regression with natural logarithm of time(s) was employed to calculate coefficients and 95% confidence intervals (CI). To analyse the risk factors for the detection of pathogens we used unadjusted odds ratios (OR) and 95% CI. A p value of < 0.05 was considered statistically significant. Statistical analyses were performed using STATA/BE 17.0 (STATA Corp, Texas, USA).

Results

SARI presentations

We conducted a 1-year surveillance of inpatient SARI in the paediatric wards of MMCH (from June 26, 2022 to July 1, 2023) and KSDH (from November 13, 2022 to July 1, 2023), totalling 91 sentinel site visits- 63 to MMCH and 28 to KSDH. During 15 of these visits, no eligible SARI cases were found. Surveillance activity stopped during the week of July 10, 2022 due to the COVID-19 outbreak.

In one-year, 8649 children < 15 years were admitted to MMCH (N = 6934) and KSDH (N = 1715), see Fig. 1. Among these, 1842 fulfilled the SARI case definition. The majority of SARI cases were in children 0–2-years 1419 (77%); followed by > 2–5-year-olds 217 (11.78%), and > 5–14-year-olds 206 (11.2%), p = < 0.001. The cumulative proportion of SARI age-wise was as follows:

Fig. 1.

SARI surveillance and enrolment of children under 2 years of age

• 0–2-year-olds: 1419/3654 [38.8% (95% CI:37.2–40.4%)]

• > 2–5-year-olds: 217/1840 [11.8% (95% CI:10.3–13.3%)]

• > 5–14-year-olds: 206/3155 [6.5% (95% CI:5.7–7.45%)]

Figure 2 illustrates the weekly SARI admissions by age group, highlighting the majority of cases occurred in the 0–2-year age group.

Fig. 2.

Weekly SARI admissions by age group from June 26, 2022 to July 1, 2023

Enrolment

We followed a serial cross-sectional sampling method for inpatient enrolment. One hundred ninety-one children 1–24 months with SARI were enrolled (163 from MMCH, 28 from KSDH), accounting for nearly 14% of SARI cases in children under-2. The highest enrolment occurred from October-November, 2022 (n = 51) and March-April, 2023 (n = 50). The majority (83.7%) came from West Midnapore and some (31/191) from adjoining districts. Figure 3 illustrates the geographic location of study participants.

Fig. 3.

Geographic location of study participants (N = 191). This study was done in District West Midnapore (yellow) of state West Bengal in India. Light green dots represent the location of individual study participants and red triangles the study sites

Individual and family characteristics

The individual and sociodemographic characteristics of the enrolled participants are summarized in online supplementary Table S1. Of 191, 47 were < 3 months-old, 35 were 3–5 months-old, 67 were 6–11 months-old, 41 were 12–23 months-old and one was 24 months-old. Males outnumbered females (112:79). Around 43% (81/189) were found underweight (< − 2 WAZ). They came from families having a median of 5 members (IQ4: 4, 7), predominantly of Hindu faith (72.7%), and remaining Muslims (27.3%). Two (1%) children had household tuberculosis exposure.

The majority had exposure to indoor air pollution due to the type of fuel used for cooking: 72% relied on solid fuel (wood, animal dung cakes), 18% used both liquefied petroleum gas (LPG) and solid fuel, while only 10% used LPG.

Nearly 5% of mothers and 3% of fathers could not read or write. Mothers had a higher level of education than fathers [grade 10 [8, 12] vs. grade 4 [3, 5], respectively. Most mothers were housewives (93%) and fathers were doing elementary jobs (45%). The majority came from upper-lower socioeconomic households having a median monthly family income of INR 12,560.

Medical and vaccination history of children and their mothers

More than one-third of the children were born via caesarean (76/191, 40%), few were preterm (< 37 gestation weeks) (31/191, 16%), and 29% (52/179) had low birth weight (< 2500 g) (online Supplemental Table S2). A few (7/191) had underlying conditions, such as macrocephaly (n = 2), cardiovascular disease (n = 2), asthma (n = 1), neurological disorders (n = 1) and comorbidity of asthma with cardiovascular disease (n = 1). Some (16/191, 8%) had received vaccines by paying out of pocket, outside of the UIP, suggesting their readiness for vaccines not included in the UIP. The majority of mothers (83.7%) were vaccinated against COVID-19, but none against influenza, RSV, pertussis, or hepatitis B.

Illness details

Participants had been ill for a median of 4 days (95% CI: 3, 6) at admission, see Table 1. Over one-third (38.7%) were referred from another health care facility. Children under 6 months were more frequently referred than children ≥ 6 months-old, (50% vs. 30%, p = 0.02).

Table 1.

Illness presentation, severity and outcome of children 1–24 months of age

	Total (N = 191)	1–5 month (N = 82) A	6–11 month (N = 67) B	12–24 month (N = 42) C	P value* A vs. B + C
Reported symptoms
Fever	167 (87)	65 (79.3)	63 (94)	39 (92.9)	0.003
Cough	164 (86)	71 (86.6)	58 (86.6)	35 (83.3)	0.8
Rhinorrhea	46 (24)	16 (19.5)	21 (31.3)	9 (21.4)	0.2
Nasal congestion	86 (45)	39 (47.6)	34 (50.7)	13 (31)	0.54
Sore throat	6 (3.1)	2 (2.4)	4 (6)	0	0.63
Difficulty breathing	143 (75)	66 (80.5)	48 (71.6)	29 (69)	0.12
Sneezing	69 (36)	37 (45.1)	25 (37.3)	7 (16.7)	0.025
Wheezing	96 (50.3)	40 (48.8)	39 (58.2)	15 (35.7)	0.917
Difficulty waking up	29 (15.2)	10 (12.2)	9 (13.4)	11 (26.2)	0.247
Redness of eyes	4 (2)	1 (1.2)	3 (4.5)	0	0.46
Poor feeding	72 (37.7)	31 (37.8)	25 (37.3)	16 (38.1)	0.98
Decrease in urine output	10 (5.2)	6 (7.3)	2 (3)	2 (4.8)	0.26
Decreased activity/fatigue	3 (1.6)	0	1 (1.5)	2 (4.8)	0.13
Vomiting or diarrhoea	23 (12)	9 (11)	8 (11.9)	6 (14.3)	0.69
Cyanosis	4 (2)	2 (2.4)	2 (3)	0	0.77
Sickness duration on admission	4 (3, 6)	4.5 (3, 6)	4 (3, 7)	3 (3, 6)	0.09
Referral case	74 (38.7)	41 (50)	23 (34.3)	10 (23.8)	0.022
Physical examination
Temperature at enrolment (n = 187)		80	67	40	0.43**
≥ 100.4 °F	11 (6)	2 (2.5)	4 (6)	5 (12.5)
99.5–100.3 °F	10 (5.3)	4 (5)	3 (4.4)	3 (7.5)
98.6–99.4 °F	24 (12.8)	14 (17.5)	4 (6)	6 (15)
Afebrile	142 (76)	60 (75)	56 (83.6)	26 (65)
Oxygen saturation (n = 64)		31	22	11	0.33**
94–100	38 (59.4)	20 (64.5)	14 (63.6)	4 (36.4)
90–93	13 (20)	6 (19.3)	3 (13.6)	4 (36.4)
< 90	13 (20)	5 (16)	5 (22.7)	3 (27.3)
Pulse rate/minute (n = 147)		70	51	26	0.36**
40–100	25 (17)	13 (18.6)	6 (11.76)	6 (23)
101–160	108 (73.5)	48 (68.57)	43 (84.3)	17 (65.4)
161–200	14 (9.5)	9 (12.8)	2 (3.9)	3 (11.5)
Respiratory rate/min (n = 139)		65	50	24	0.047**
20–39	15 (10.8)	4 (6)	7 (14)	4 (16.6)
40–49	31 (22.3)	12 (18.46)	14 (28)	5 (21)
50–99	86 (62)	45 (69.23)	28 (56)	13 (54)
≥ 100	7 (5)	4 (6)	1 (2)	2 (8.3)
Disease severity
Nebulization requirement (n = 189)	174 (92)	77 (95)	62 (94)	35 (83)	0.187
Intravenous fluid requirement (n = 190)	92 (48.4)	44 (54.3)	33 (49.2)	15 (35.7)	0.16
Mechanical ventilation requirement (n = 190)	5 (2.6)	2 (2.4)	2 (3)	1 (2.4)	0.9
Oxygen requirement (n = 190)	69 (36.3)	37 (45.6)	18 (27)	14 (33.3)	0.021
ICU requirement (n = 190)	21 (11)	9 (11.1)	7 (10.4)	5 (11.9)	0.98
Antibiotic prescription	186 (97.4)	81 (98.8)	63 (94)	42 (100)	0.3
Similar sickness within household	25 (13)	16 (19.5)	7 (10.4)	2 (4.7)	0.022
Duration of hospitalization	4 (3, 6)	4 (3, 6)	5 (3, 8)	3 (2, 5)	0.62**
0–3 days	82 (42)	35 (42.7)	21 (31.3)	26 (61.9)
4–6 days	64 (33.5)	32 (39)	24 (35.8)	8 (19)
7–10 days	29 (15)	12 (14.6)	11 (16.4)	6 (14.3)
> 10 days	16 (8.4)	3 (3.6)	11 (16.4)	2 (4.76)
Hospitalization outcome					0.13
Recovered and discharged	178 (93)	74 (90.2)	64 (95.5)	40 (95.2)
Referred	2 (1)	1 (1.2)	1 (1.5)	0
Left Against Medical Advice (LAMA)	8 (4)	6 (7.3)	1 (1.5)	1 (2.4)
Died	3 (1.57)	1 (1.2)	1 (1.5)	1 (2.4)

*p values for categorical variables were calculated by chi2 (χ²) test; **p values for continuous variables were calculated by Wilcoxon rank-sum test; values in bold indicate statistical significance

Some children (25/191) had household exposures, with siblings (40%) and mothers (28%) being the most commonly affected. Notably, household exposure was significantly more prevalent among families of children aged < 6 months than those aged 6-24-months (16/82 vs. 9/109, p = 0.022).

Participants presented with a median of 5 symptoms (IQR: 3–6) (online Supplemental Table S2). At enrolment, fever (temperature ≥ 100.4 °F) was found in only 11/191 (5.7%), especially among older children. History of fever was more commonly reported in children 6–24 months than children < 6 months (102/109 vs. 65/82, p = 0.003). In contrast, sneezing was more prevalent in children < 6 months-old than in ≥ 6 months-old (37/82 vs. 32/109, p = 0.025).

Cough and difficulty breathing were reported in 86% (164/191) and 75% (143/191) of the children, respectively. Among preterm children, the majority (25/31) presented with difficult breathing. Fast breathing as per age was noted in 61.8% (86/139), and wheezing in 50% (96/191). More than one-third had poor feeding (72/191), and 2% had cyanosis.

Around 20% (13/64) of children, were hypoxemic (SpO2 < 90%) [42]. The majority (91%) needed nebulization, close to half required intravenous fluid administration (48%), and more than one-third needed oxygen support (36%). Some children required intensive care [21/191 (11%)], with a small proportion requiring mechanical ventilation (2.6%). Compared with children ≥ 6 months, children < 6 months-old required oxygen support more frequently (29.6% vs. 45.6%, p = 0.02). Antibiotic prescriptions were high (97.4%) among all hospitalizations, regardless of age. Antibiotics involved a wide range including penicillins (amoxicillin), second and third generation cephalosporins (cefuroxime, ceftriaxone, cefotaxime, cefixime), aminoglycosides (amikacin), glycopeptides (vancomycin), lincosamides (clindamycin), carbapenem (meropenem) and macrolide (clarithromycin, azithromycin) class of drugs. Antibiotic exposure was associated with nebulization need (p = 0.002), but not with oxygen or ICU requirement.

Laboratory findings according to age

Viruses were detected in 84/191 (44%), either as singly (25.6%) or with other viruses (3.66%). The majority of viral detections occurred in infants aged < 12 months. RSV A/B was the major viral cause 60/190 (31.6%), followed by influenza 12/190 (6.3%), and SARS-CoV-2 2/191 (1%), shown in Table 2. In 12 children, 16 influenza types or subtypes were detected. Among the influenza cases, the subtypes included IA/H3 (6/16), IA/H1N1pdm2009 (5/16), IB (4/16) and IC (1/16). Among the RSV-positive samples, twenty-four were subtyped, and all were RSV B.

Table 2.

Laboratory findings from nasopharyngeal samples collected from children with severe acute respiratory infection (SARI)

Pathogen	Overall	< 6 month n (%) A	6–11 month n (%) B	12–24 month n (%) C	p* A & B	p A & C	p B & C	p A & (B + C)
N	191	82	67	42
N tested for influenza and RSV	190	82 (100)	66 (98.5)	42 (100)
Influenza	12 (6.3)	8 (9.7)	3 (4.5)	1 (2.4)	0.23	0.13	0.56	0.09
RSV A/B	60 (31.6)	28 (34)	22 (33.3)	10 (23.8)	0.91	0.237	0.29	0.5
SARS-CoV-2 (N = 191)	2 (1)	1 (1)	0	1 (3)	–	0.63	–	0.84
Influenza subtypes	16
Influenza B	4 (2.1)	2 (2.4)	2 (3)	0	0.84	–	–	0.78
Influenza A/H1N1pdm	5 (2.6)	3 (3.6)	1 (1.5)	1 (2.4)	0.42	0.7	0.74	0.44
Influenza A/H3	6 (3.1)	5 (6)	1 (1.5)	0	0.16	–	–	0.04
Influenza C	1 (0.5)	1 (1.2)	0	0	–	–	–	–
N tested for respiratory panel	81 (42.4)	35	30	16
Rhinovirus	11 (13.6)	6 (17)	5 (16.6)	0	0.96	–	–	0.41
Bocavirus	3 (3.7)	0	1 (3.3)	2 (12.5)	–	–	0.23	–
Adenovirus	3 (3.7)	1 (2.8)	1 (3.3)	1 (6.25)	0.91	0.56	0.64	0.72
Parainfluenza virus 1	2 (2.47)	1 (2.8)	0	1 (6.25)	–	0.56	–	0.84
Parainfluenza virus 3	2 (2.47)	1 (2.8)	1 (3.3)	0	0.91	–	–	0.84
Parainfluenza virus 2	1 (1.2)	0	1 (3.3)	0	–	–	–	–
Enterovirus	1 (1.2)	0	1 (3.3)	0	–	–	–	–
Bacteria
Chlamydia pneumoniae	1 (1.2)	1 (2.8)	0	0	–	–	–
Bordetella spp	1 (1.2)	1 (2.8)	0	0	–	–	–
Haemophilus influenzae**	19 (23.45)	9 (25.7)	5 (16.6)	5 (31.25)	0.37	0.68	0.25	0.67
Streptococcus pneumoniae**	17 (22)	9 (25.7)	7 (23.3)	1 (6.25)	0.82	0.1	0.14	0.36
Staphylococcus aureus**	7 (8.6)	7 (20)	0	0	–	–	–
Moraxella catarrhalis**	4 (5)	3 (8.6)	1 (3.3)	0	0.38	–	–	0.18
Acinetobacter baumannii**	1 (1.2)	1 (2.8)	0	0	–	–	–

*p values are calculated by chi2 (χ²) test; ** colonizing bacteria; The pathogens represent single detection or with other pathogens; Mycoplasma pneumoniae, human metapneumovirus, Klebsiella pneumoniae, Streptococcus agalactiae, Pseudomonas aeruginosa, Legionella pneumophila, human parechovirus and human coronavirus (alpha and beta) were not detected in the 81 cases tested; values in bold indicate statistical significance

A representative subset of samples (81/191, 42%) were tested for other respiratory pathogens. Bacteria were detected in 36/81 (44.4%) of samples, the majority in infants < 6 months. These included colonizing bacteria including Haemophilus influenzae (19/81, 23.45%) Streptococcus pneumoniae (17/81, 22%), Staphylococcus aureus (7/81, 8.6%), Moraxella catarrhalis (4/81, 5%), and Acinetobacter baumannii (1/81, 1.2%). Bordetella (1/81, 1.2%) and Chlamydia pneumoniae (1/81, 1.2%) were also found. Other respiratory viruses detected were rhinovirus (11/81, 13.6%), bocavirus, adenovirus (3/81 each, 3.7%), parainfluenza virus 1 (2/81, 2.5%), parainfluenza virus 3 (2/81, 2.5%), parainfluenza virus 2 (1/81, 1.2%), and enterovirus (1/81, 1.2%).

Influenza admissions were more common in children < 6 months (11/82, 13.4%) than those 6-24-months (5/108, 4.6%), p = 0.03, with median age 3.5 (IQR: 2–6.5) months. RSV was prevalent in both age groups [28/82 (34%) vs. 32/108 (29.6%), p = 0.5] with median age 6 (IQR: 3–9) months. SARS-CoV-2 was detected in a 4-month-old child and a 13-month-old child. The median age of rhinovirus admission was 5 (IQR: 3–8) months. Interestingly, S. aureus, Bordetella, C. pneumoniae, and A. baumannii were found only in < 6-month-olds, suggesting an increased vulnerability of this age group.

Online Supplementary Table S3 describes the single and co-detections. Viral-bacterial co-detections were found in 28/81 (34%) samples, frequently in infants under 6 months (17/28). Among the bacteria, S. pneumoniae (n = 14) and H. influenzae (n = 13) were frequently co-detected with other pathogens. RSV was the most common virus found with bacteria. Children with RSV were at 6 times increased risk of having S. pneumoniae [OR 6.2 (95% CI: 1.8–21.3); p = 0.004]. In 35 RSV infections recorded through the panel, S. pneumoniae (9, 25.7%), Haemophilus influenzae (6, 17%) or both (4, 11.4%) were frequently co-detected. Other bacteria detected with RSV were S. aureus (3, 8.6%), S. aureus + H. influenzae (1, 2.8%), S. aureus + S. pneumoniae (1, 2.8%), M. catarrhalis + H. influenzae + A. baumannii (1, 2.8%), M. catarrhalis + C. pneumoniae + Rhinovirus (1, 2.8%) and M. catarrhalis + S. aureus + Rhinovirus (1, 2.8%). Presence of S. pneumoniae in children with RSV increased the risk of hypoxaemia [OR 10.6 (95% CI 1.6–72); p = 0.015], but not the requirement of oxygen, mechanical ventilation or ICU.

Different influenza types and subtypes were co-detected in 3 children. Influenza was found with S. aureus, parainfluenza virus 1, or rhinovirus in one individual, each. Influenza and RSV were not co-detected, suggesting their negative association. SARS-CoV-2 and RSV were co-detected in one child. Rhinovirus was detected in 11/81 children out of which it was found singly in 1 (9%), with other viruses in 3 (27%), with bacteria in 3 (27%), and with other viruses and bacteria in 4 (36%). Children with rhinovirus were at 25 times increased risk of having Moraxella catarrhalis [OR 25.8 (95% CI: 2.4–279); p = 0.007]. Rhinovirus and bocavirus were common co-infecting viruses with RSV. Antibiotic prescriptions were common in viral-bacterial co-detections (98%) (p = 0.004).

Figures 4 and 5 depict the monthly activity and seasonality of influenza, RSV and SARS-CoV-2, respectively. Influenza was notably higher during June 2022 followed by a small peak from August-September 2022 (monsoon season), which rose again from January-February 2023 (winter season). RSV positivity peaked during October-November 2022 (autumn season) and a smaller peak from April-May 2023 (summer). SARS-CoV-2 activity was high during July 2022, coinciding with the COVID-19 wave.

Fig. 4.

Viral circulation patterns from June 26, 2022 to July 1, 2023 among children with SARI. A Monthly activity of influenza (blue), B RSV A/B (red), C SARS-CoV-2 (green)

Fig. 5.

Seasonality of influenza, RSV A/B and SARS-CoV-2 from June 26, 2022 to July 1, 2023. The bars represent percent of samples positive for SARS-Cov-2 (blue), influenza (green), RSV A/B (grey) in each season.

Illness details by pathogens

Illness details by pathogens and demographics are detailed in Table 3 and the risk factors for influenza, RSV, SARS-CoV-2 and rhinovirus are summarized in online supplemental S4. Approximately (10/60) 16.6% of RSV-positive children were preterm. The most common symptoms among children with RSV were cough (93.3%), fever (83.3%), difficulty breathing (81.6%) and wheezing (58.3%). As compared to RSV negative children, children with RSV had higher cough, were feeding poorly and had decreased activity (p = 0.043, p = 0.019, p = 0.01). 65% of children with hypoxemia were RSV positive. RSV infection significantly increased during autumn (p = < 0.001) and was associated with nebulization requirement (p = 0.029). RSV was found more in households that solely used solid fuel for cooking (wood or animal dung) (44/60) vs. LPG (5/60), p = < 0.001. Antibiotics were used in the majority (58/60) of RSV infections with or without bacteria.

Table 3.

Illness details based of RSV, influenza and rhinovirus total, single or mixed infections

Item	Total	Total RSV	RSV single infection	RSV mixed infection	p	Total Influenza	Influenza single or co-infection*	Influenza mixed infection	p	Total Rhinovirus	Rhinovirus single infection	Rhinovirus mixed infection	p**
N	191	60	36	24		12	9	3		11	1	10
Age
1–5 months	82 (43)	28 (46.6)	14 (39)	14 (58.3)	0.18	8 (66.6)	6 (66.6)	2 (66.6)	1	6 (54.5)	0	6 (60)	0.45
6–11 months	67 (35.1)	22 (36.7)	16 (44.4)	6 (25)	0.17	3 (25)	3 (33.3)	0		5 (45.5)	1 (100)	4 (40)
12–24 months	42 (22)	10 (16.7)	6 (16.6)	4 (16.6)	1	1 (8.3)	0	1 (33.3)		0	0	0
Gender
Male	112 (58.6)	37 (61.6)	21 (58)	16 (66.6)	0.6	6 (50)	4 (44.4)	2 (66.6)	0.55	5 (45)	0	5 (50)	1
Female	79 (41.4)	23 (38.3)	15 (41.6)	8 (33.3)		6 (50)	5 (55.5)	1 (33.3)		6 (54.5)	1 (100)	5 (50)
Religion
Hindu	139 (73)	43 (71.6)	26 (72)	17 (71)	1	7 (58.3)	5 (55.5)	2 (66.6)	1	10 (91)	1 (100)	9 (90)	1
Muslim	52 (27)	17 (28.3)	10 (27.7)	7 (29.2)		5 (41.6)	4 (44.4)	1 (33.3)		1 (9)	0	1 (10)
Overcrowding at home
< 6	107 (56)	39 (65)	20 (55.5)	19 (79.2)	0.09	8 (66.6)	6 (66.6)	2 (66.6)	11	5 (45)	0	5 (50)	1
6 and more	84 (44)	21 (35)	16 (44.4)	5 (21)		4 (33.3)	3 (33.3)	1 (33.3)		6 (54.5)	1 (100)	5 (50)
Birth characteristics
Premature birth	31 (16.2)	10 (16.6)	6 (16.6)	4 (16.6)	1	2 (16.6)	2 (22.2)	0	1	1 (9)	0	1 (10)	1
Caesarean delivery	76 (39.8)	21 (35)	14 (39)	7 (29.2)	0.58	8 (66.6)	7 (77.7)	1 (33.3)	0.13	1 (9)	0	1 (10)	1
Low birth weight (N = 179)	52 (29)	17 (31)	12 (33)	5 (21)	0.38	3 (25)	3 (33.3)	0	0.5	3 (27)	0	3 (30)	1
Underlying condition	7 (3.6)	1 (1.6)	0	1 (4.2)	0.4	0	0	0		0	0	0
Referred case	74 (38.7)	17 (28.3)	10 (27.7)	7 (29.2)	1	6 (50)	4 (44.4)	2 (66.6)	0.55	6 (54.5)	1 (100)	5 (50)	1
Similar sickness within household	25 (13)	3 (5)	2 (5.5)	1 (4.2)	1	3 (25)	2 (22.2)	1 (33.3)	0.49	1 (9)	0	1 (10)	1
Clinical Symptoms
Reported fever	167 (87.4)	50 (83.3)	32 (89)	18 (75)	0.18	11 (91.6)	8 (89)	3 (100)	1	11 (100)	1 (100)	10 (100)	1
Cough	164 (86)	56 (93.3)	33 (91.6)	23 (96)	0.64	10 (83.3)	8 (89)	2 (66.6)	0.35	10 (91)	0	10 (100)	0.09
Rhinorrhoea	46 (24)	14 (23.3)	7 (19.4)	7 (29.2)	0.53	1 (8.3)	1 (11)	0	0.007	2 (18)	0	2 (20)	1
Nasal congestion	86 (45)	29 (48.3)	15 (41.6)	14 (58.3)	0.3	6 (50)	5 (55.5)	1 (33.3)	1	8 (73)	1 (100)	7 (70)	1
Sore throat	6 (3)	3 (5)	1 (2.7)	2 (8.3)	0.56	0	0	0	-	2 (18)	0	2 (20)	1
Difficult breathing	143 (74.9)	49 (81.6)	29 (80.5)	20 (83.3)	1	10 (83.3)	8 (89)	2 (66.6)	0.35	11 (100)	1 (100)	10 (100)	1
Sneezing	69 (36)	26 (43.3)	14 (39)	12 (50)	0.43	3 (25)	2 (22.2)	1 (33.3)	0.49	6 (54.5)	0	6 (60)	0.45
Wheezing	94 (49.2)	35 (58.3)	18 (50)	17 (71)	0.18	7 (58.3)	5 (55.5)	1 (33.3)	0.55	7 (63.6)	0	7 (70)	0.36
Difficulty waking up	30 (15.7)	9 (15)	3 (8.3)	6 (25)	0.14	1 (8.3)	1 (11)	0	1	3 (27.3)	0	3 (30)	1
Redness of eyes	4 (2)	1 (1.6)	1 (2.7)	0	1	0	0	0	-	0	0	0	-
Poor feeding	72 (37.7)	30 (50)	14 (39)	16 (66.6)	0.6	3 (25)	3 (33.3)	0	0.5	7 (63.6)	1 (100)	6 (60)	1
Decrease in urine output	10 (5.2)	4 (6.6)	0	4 (16.6)	0.02	2 (16.6)	2 (22.2)	0	0.53	1 (9)	0	1 (10)	1
Decreased activity	3 (1.6)	3 (5)	1 (2.7)	2 (8.3)	0.56	0	0	0	-	1 (9)	1 (100)	1 (10)	0.18
Vomiting/diarrhoea	23 (12)	4 (6.6)	0	4 (16.6)	0.02	1 (8.3)	1 (11)	0	1	5 (45)	1 (100)	4 (40)	0.45
Cyanosis	4 (2)	0	0	0		1 (8.3)	1 (11)	0	1	0	0	0	-
Disease severity
Nebulization	174 (91)	59 (98.3)	36 (100)	23 (96)	0.15	12 (100)	9 (100)	3 (100)	1	8 (73)	0	8 (80)	0.27
Intravenous fluid	92 (48)	29 (48.3)	15 (42)	14 (58.3)	0.3	7 (58.3)	5 (55.5)	2 (66.6)	1	7 (63.6)	0	7 (70)	0.36
Mechanical ventilation	5 (2.6)	2 (3.3)	2 (4.76)	0	0.51	1 (8.3)	0	1 (33.3)	1	2 (18)	0	2 (20)	1
Oxygen requirement	69 (36)	20 (33.3)	10 (27.7)	10 (42)	0.28	7 (58.3)	5 (55.5)	2 (66.6)	1	4 (36.4)	0	4 (40)	1
ICU requirement	21 (11)	4 (6.6)	3 (8.3)	1 (4.2)	0.64	3 (25)	2 (22.2)	1 (33.3)	0.48	3 (27.3)	0	3 (30)	1
Antibiotic prescription	186 (97.4)	58 (96.6)	36 (100)	22 (91.6)	0.15	12 (100)	9 (100)	3 (100)	1	10 (91)	1 (100)	9 (90)	1
Duration of hospitalization**
0–3 days	82 (42)	26 (43)	16 (44.4)	9 (37.5)	0.8	4 (33)	4 (44.4)	0	0.5	2 (18)	0	2 (20)	1
4–6 days	64 (33.5)	21 (35)	12 (33.3)	9 (37.5)	0.8	6 (50)	4 (44.4)	2 (66.6)	1	3 (27.3)	0	3 (30)	1
7–10 days	29 (15)	9 (15)	5 (14)	4 (16.6)	1	2 (16.6)	1 (11)	1 (33.3)	0.35	4 (36.4)	0	4 (40)	1
> 10 days	16 (8.4)	4 (6.6)	3 (8.3)	1 (4.2)	0.64	0	0	0	-	1 (9)	1 (100)	1 (10)	0.18

Note: Total infections is the child having both single and mixed infections; Single infections are those where the child had only virus infection; Mixed infection are those where along with the virus there is a co-detection of another virus or bacteria or both; *Influenza single infection or co-infection of influenza subtypes in a child; **p values comparing single infection with mixed infection via the Fisher’s exact test

Influenza infections significantly increased during the monsoon season (p = 0.038). Nearly half of the children with influenza were referred. Those born via caesarean section had increased influenza positivity, p = 0.046. The most common symptoms of influenza positive children were fever (91.6%), cough (83.3%), difficulty breathing (83.3%) and wheezing (58.3%). Influenza admissions were associated with a trend of greater requirements for oxygen (58.3%) and ICU (25%) admission than influenza-negative (34.8%, 10%; p = 0.1, p = 0.11, respectively). Antibiotics were prescribed in all the influenza and SARS-CoV-2 patients.

The most common symptoms among children with rhinovirus was fever (100%), difficulty breathing (100%), cough (91%) and nasal congestion (73%). Rhinovirus-positive children presented with more vomiting or diarrhoea and sore throat than rhinovirus negative children (p = 0.002, p = 0.029). Rhinovirus admissions were significantly associated with greater requirement of ICU (27.3%) and mechanical ventilation (18%) than rhinovirus negative [5.7%, (p = 0.01), 1.4% (p = 0.004), respectively], irrespective of their age. Antibiotics were prescribed in 10/11 of rhinovirus patients. S. pneumoniae and H. influenzae were commonly detected in households that used solid fuel (13/17, 12/19) or a mix of solid and clean fuel (4/17, 6/19).

PICU cases

In total, twenty-one cases were enrolled from the PICU, of whom the majority (9/21, 43%) were < 6 months, followed by 6–11 months-old (7/21, 33%) and > 12 months-old (5/21, 24%). Of those in ICU, 62% were males, and used solid fuel (13/21) or a mix (5/21). Four (19%) were RSV positive, two of whom were < 6 months-old and two 6-11-months-old. Two (9.5%) had influenza A/H1N1 and one (4.7%) influenza C, all between 1 and 6 months. Seven ICU samples were tested for other pathogens, where we identified rhinovirus (n = 3), H. influenzae (n = 3), S. pneumoniae (n = 2), M. catarrhalis (n = 1) and adenovirus (n = 1). Antibiotics were prescribed in all the ICU cases.

Hospitalization outcome

The majority of children (178/191) recovered. 42% were discharged within 72 h. Two were referred to another facility, of which a one-month-old infant had RSV + S. pneumoniae and another 8-month-old had H. influenzae. Few (8/191) left against medical advice, of which two had RSV, two had influenza (H1N1/H3), one had RSV + H. influenzae and one had RSV + H. influenzae + Bocavirus.

There were three deaths, two males and one female, all infants-aged 3 months, 9 months and 12 months. Of the three, sample from 12-month-old male child was tested with the respiratory panel and found positive with H. influenzae. None of the deaths were related with influenza or RSV or SARS-CoV-2.

Duration of hospitalization

The median duration of hospitalization was 4[3-6] days. Children 6–11-months old were admitted for a median of 5[3-8] days, < 6 months old for 4 [3-6] days, and 12–24 months old for 3 [2-5] days. Children in ICU stayed longer than those in wards 9 [6-14] vs. 4[2-6] days, p = < 0.001.

Table 4 describes the factors that influence the duration of hospitalization. Referral from another health care facility increased the duration [Coeff 0.19, 95% CI:0.02-0.36, p = 0.03), and so did symptoms of cyanosis [Coeff 0.62 (0.04–1.2), p = 0.036]. Children requiring intravenous fluids [Coeff 0.33 (0.17–0.5), p = < 0.001], ICU [Coeff 0.72 (0.47–0.96), p = < 0.001] or oxygen [Coeff 0.2 (0.03–0.37), p = 0.02] had longer stay. Among the pathogens, only rhinovirus was associated with increased duration of hospitalization [Coeff 0.4 (0.048–0.76), p = 0.026]. Association between antibiotic use and length of stay was not significant (p = 0.5), though saw a negative trend.

Table 4.

Predictors of length of stay using linear regression

Independent variables	Coefficient	95% CI	p value
Age (1–5 m, 6–11 m, 12–24 m)	−0.02	− 0.13–0.08	0.61
Gender	− 0.02	− 0.19–0.14	0.77
Religion (Hindu, Islam)	− 0.07	− 0.26–0.11	0.42
Child referred from another hospital	0.19	0.02–0.36	0.03
Duration from symptom onset	0.02	0.006–0.04	0.01
Sickness at home	− 0.017	− 0.26–0.23	0.89
Premature birth	− 0.015	− 0.24–0.21	0.89
Caesarean delivery	− 0.06	− 0.23–0.11	0.48
Number of children at home	0.014	− 0.09–0.12	0.798
Overcrowding (> 6 at home)	− 0.02	− 0.19–0.14	0.81
Fuel for cooking (LPG vs. solid fuel)	0.14	− 0.14–0.43	0.33
Father’s education	0.05	− 0.005-0.1	0.076
Low birth weight	− 0.045	− 0.24–0.15	0.647
Chronic medical condition	0.44	− 0.0–0.09	0.05
Received UIP vaccines	0.14	− 0.12-0.4	0.3
Outside UIP vaccines	− 0.03	− 0.33–0.27	0.83
Tuberculosis at home	− 0.06	− 0.88–0.76	0.87
Summer season	− 0.12	− 0.31–0.05	0.17
Monsoon season	0.08	− 0.126–0.3	0.427
Autumn season	0.07	−0.12–0.26	0.475
Winter season	− 0.001	− 0.2–0.2	0.988
Fever	− 0.08	− 0.33–0.17	0.52
Cough	0.11	− 0.13–0.35	0.37
Rhinorrhea	− 0.06	− 0.25–0.14	0.55
Nasal congestion	0.12	− 0.05–0.28	0.173
Sneezing	0.02	− 0.1-0.2	0.82
Sore throat	0.02	− 0.46-0.5	0.93
Difficult breathing	0.12	− 0.07–0.31	0.21
Wheezing	0.06	− 0.1-0.23	0.47
Difficult waking up	0.09	− 0.13–0.33	0.4
Red eye	0.27	− 0.3–0.86	0.35
Decreased appetite	0.06	− 0.11–0.23	0.48
Low urine output	− 0.28	− 0.65–0.09	0.14
Decreased movement	0.19	− 0.47–0.87	0.56
Vomiting/diarrhea	− 0.18	− 0.44-0.07	0.156
Cyanosis	0.62	0.04–1.2	0.036
Antibiotic prescription	−  0.18	−0.7–0.34	0.5
Nebulization	0.21	− 0.08–0.52	0.16
Intravenous fluid administration	0.33	0.17–0.5	< 0.001
Mechanical ventilation	0.25	−0. 27–0.77	0.35
Oxygen requirement	0.2	0.03–0.37	0.02
ICU requirement	0.72	0.47–0.96	< 0.001
Single viral detection	− 0.06	−0.25–0.13	0.52
Single bacterial detection	0. 28	−0.19-0.76	0.237
Viral-viral co-detection	− 0.12	−0.56–0.33	0.6
Bacterial-bacterial co-detection	0.7	−0.11–1.5	0.09
Viral-bacterial co-detection	0.16	− 0.08–0.41	0.19
SARS-CoV-2	− 0.42	− 1.24–0.4	0.32
RSV A/B	−0.04	− 0.22–0.14	0.67
Influenza	− 0.15	− 0.–0.2	0.4
Staphylococcus aureus	− 0.13	− 0.57–0.32	0.57
Streptococcus pneumoniae	0.276	− 0.025-0.57	0.07
Bordetella spp	0.22	− 0.91–1.35	0.701
Chlamydia pneumoniae	0.22	− 0.91–1.35	0.701
Acinetobacter baumannii	− 0.12	− 1.26–1.01	0.83
Haemophilus influenzae	0.14	− 0.16–0.43	0.36
Parainfluenza virus 2	0.47	− 0.66-1.6	0.407
Parainfluenza virus 3	− 0.21	− 1.02–0.59	0.6
Enterovirus	− 0.35	−1.48–0.78	0.54
Rhinovirus	0.4	0.048–0.76	0.026
Bocavirus	− 0.17	− 0.84–0.48	0.6
Adenovirus	0.16	− 0.5–0.82	0.63

Coefficient: Regression Coefficient; Values in bold are statistically significant

Discussion

Robust surveillance studies are needed to guide interventions and vaccination strategies, particularly in LMICs. This is one of the first studies in a rural district of Eastern India that conducted a one-year surveillance of SARI in two hospitals. Conducted post-COVID-19 pandemic the study aimed to understand the epidemiology of SARI in children under 2 years. Our findings add to the global evidence that viruses, especially RSV, are a significant cause of SARI hospitalizations in young children. Additionally, we demonstrated that the majority of influenza-related hospitalizations occurred in infants < 6 months of age, who also faced frequent viral-bacterial co-detections, and faced an increased risk of severe LRIs requiring referrals, oxygen support, and ICU admission compared with children aged ≥ 6–24 months.

The majority (77%) of SARI hospitalizations were in children 0–2 years, which is consistent with findings from pan-Indian studies, identifying children under one year as the most vulnerable age group for severe infections [43, 44]. This heightened vulnerability is attributed to immature immune systems and increased exposure to pathogens due to close household contact.

Viruses constitute a major cause of SARI in children, accounting for 44% of etiologies identified in our study. In agreement a study from Western India reported viral presence in 46.2% of children with SARI [27], whereas studies from other regions of India and neighbouring countries reported a range of 40–98.6% [12, 27, 29, 32, 43, 45–47]. The symptoms of tachypnoea (61.8%), wheeze (50%), cough (86%), and difficulty breathing (75%), were also suggestive of viral bronchiolitis [45, 48]. Our study observed a high rate of antibiotic use in managing SARI among young children and the significant role of viruses in SARI—often overlooked due to the lack of routine viral diagnostics in the public health system.

RSV and influenza emerged as the major viral cause of SARI, aligning with several pre-pandemic studies from India, that identified these viruses as the most common causes of paediatric LRIs [26, 27, 43]. While previous research focused primarily on urban areas, our study emphasized that RSV and influenza also contribute significantly to paediatric LRIs in rural Eastern India, corroborating findings from rural Northern India [31, 32]. During the study period, another study from Uttar Pradesh reported hospitalizations in children under-5 from RSV (12.6%), adenovirus (7.8%), parainfluenza virus (6.3%), influenza virus (3.01%) and SARS-CoV-2 (1.51%) [30]. While their RSV and influenza rates were lower than our study, the SARS-CoV-2 rates (1%) were similarly low, which could be due to high maternal COVID-19 immunization coverage. The high rates of RSV seen in our study post-pandemic are in accordance with global reports of a RSV surge in late 2022 [49–51]. A higher need of respiratory support was found in children from RSV during the 2022–2023 post-pandemic season in the US [51]. This is possibly due to low population immunity due to reduced exposure to viruses during the pandemic from non-pharmaceutical interventions, viral interference, changes in health seeking behaviors, and health system factors [52, 53].

Severe RSV infections involve increased host inflammatory response resulting in airway narrowing and bronchiolitis among infants [11, 27]. The majority of RSV admissions in our study required nebulization, supporting bronchiolitis as a common presentation. Further, we noted 65% of hypoxemic children (SpO2 < 90%) were RSV positive, suggesting reduced levels of oxygen absorption in the bloodstream from the infected lungs, as manifested by difficulty breathing in 81%. Notably, 16.6% of RSV admissions involved those born prematurely, aligning with a meta-analyses that reported preterm birth as a risk factor for RSV-associated LRI [47, 54].

The majority of influenza admissions were in infants under 6 months, supporting the global evidence of their increased risk for severe disease and hospitalizations [15, 55, 56]. This study builds the evidence base for maternal influenza immunization to protect this high-risk group against influenza hospitalizations. We detected one case of influenza C, consistent with previous reports from India [57].

Influenza peaks were observed in June and August, consistent with national surveillance data during the same period [44]. Influenza activity typically peaks during the monsoon (July-September) in West Bengal and northern India [58, 59]. In early 2023, influenza activity increased aligning with the increase across India [60]. RSV infections peaked during autumn, consistent with seasonal patterns in the Northern Hemisphere [61]. RSV B was predominant during 2022–2023 in Bulgaria, similar to our observations [49].

A significant association was found between solid household fuel use and RSV, and bacterial colonization (S. pneumoniae and H. influenzae). Exposure to household air pollution has been widely linked to increased RSV and respiratory infection risk [62, 63], supporting findings from other Indian studies [64]. In fact, a 10 year observational study from China reported air pollution predictors contributed more to severe pneumonia in children than respiratory viruses [65].

Rhinovirus was detected particularly in children requiring ICU admission and mechanical ventilation, highlighting its severity, as reported in other countries [17, 45]. Accumulating evidence supports the contribution of rhinoviruses in pneumonia individually or in combination with other pathogens [66, 67]. We found the majority (91%) of rhinovirus cases in synergy with other pathogens. Viral infections superimposed by bacterial infections often lead to high morbidity and mortality [19]. Viral-bacterial co-detections were common (34%), particularly in infants under six months. Although, children carry bacteria in their nasopharynx asymptomatically, the presence of pneumococcus and/or H. influenzae in ICU cases, referrals, and fatalities underscores their role in severe disease pathogenesis. Mechanistically, viral infections facilitate bacterial adherence to the nasopharyngeal epithelium and increase mucus production [68].

To our knowledge this is the first study to describe the clinical and epidemiological characteristics of influenza, RSV and SARS-CoV-2 associated SARI among young children from a rural district in West Bengal. The study was conducted in secondary and tertiary hospitals to capture hospitalization in both tiers of the health system. MMCH is the only public tertiary centre in this area and adjoining area, so the hospitalizations may be representative. The year-long surveillance allowed us to capture the prevalence, high-risk groups, virus seasonality in accordance with the WHO Global Influenza Surveillance and Response System and the WHO Global RSV Surveillance System. Respiratory panel testing enhanced the detection of multiple pathogens and provided insights into viral-bacterial co-detections. Additionally, the study incorporated birth history and sociodemographic data, enabling analysis of environmental contributors such as household fuel use. This study provided insights into viral circulation patterns post pandemic, and their relative proportions in SARIs after the lifting of non-pharmaceutical measures and widespread COVID-19 immunization in adults. Additionally, the study underscores the widespread usage of antibiotics in indoor wards and ICUs which may contribute to the economic burden and the prevalence of antibiotic resistance in India.

Our study has several limitations. First, the hospitalization data from only two facilities may not fully represent the broader geographic population, as health-seeking behaviors in the region remain unknown. This limitation and lack of reliable population data hinders population-level RSV and influenza incidence estimates. Second, the small sample size, due to the strict inclusion criteria of enrolment within 24 h of hospitalization, reduced representativeness. Third, we used upper respiratory tract specimens, as lower respiratory tract specimens were difficult to collect from young children, which may hinder interpretation of bacterial causes, as nasopharyngeal detection could indicate carriage, infection, or convalescent shedding [69]. Fourth, the absence of a control group and data on the time course and density of bacteria limits us to assess asymptomatic carriage, acquisition and attribute causality to LRI. Fifth, reliance on inpatients may underestimate the burden of milder SARI cases that do not require hospitalization. Sixth, missing data points for pulse rate, respiratory rate, and SpO2 further constrained analysis. Seventh, our study did not capture neonatal ICU admissions, a high-risk group for RSV. Eighth, seasonal variability in respiratory panel testing prevented assessment of other respiratory viruses’ seasonality and only one death and seven PICU samples were tested with respiratory panel limiting us to draw correlations. Finally, widespread antibiotic use may have influenced bacterial detection and masked the true burden.

Future studies are needed to understand RSV and influenza genotypes, and strain characterization across multiple seasons, regions and settings to improve the generalizability of findings. Future research is needed on the interrelationships of RSV and other viruses with bacteria especially involving pneumococcus, H. influenzae and S. aureus. Additionally, further investigations are needed to explore the impact of environmental factors, such as indoor air pollution, on respiratory disease severity.

Given the substantial burden of RSV and influenza in young children, access to maternal RSV, nirsevimab, and influenza vaccines would be crucial for reducing the burden of SARI in young children [70, 71]. Maternal immunization has been shown to confer passive immunity to infants, significantly reducing hospitalizations and severe outcomes [13, 72]. Expanding access to these vaccines in India, particularly in rural settings, could greatly mitigate SARI-related morbidity, hospitalizations and mortality. Our study underscores the urgent need for continuous integrated surveillance of influenza and RSV and targeted vaccination programs to protect vulnerable paediatric populations from RSV and influenza-related complications.

Conclusion

In conclusion, this study highlights the significant burden of SARI among young children, with RSV and influenza being the predominant pathogens. Further, the study underscores the higher influenza hospitalizations rates in infants < 6 months and the need for laboratory diagnosis to reduce antibiotic use. These findings also underscore the need of maternal RSV and influenza vaccination strategies, as well as improving household air quality to decrease hospitalizations and deaths from SARI -particularly in infants in low- and middle-income countries. Further, the study underscores the importance of surveillance studies to guide implementation strategies for maternal vaccine(s) and nirsevimab in LMICs to make an impact in the regions where the morbidity and mortality burden of LRI is greatest.

Abbreviations

LRI

Lower Respiratory tract Infections

LMIC

Low- and middle-income country

RSV

Respiratory Syncytial Virus

LIC

Low Income Country

mAb

Monoclonal antibody

COVID-19

Coronavirus disease 2019

SARS-CoV-2

Severe Acute Respiratory Syndrome Coronavirus 2

SARI

Severe Acute Respiratory Infection

UIP

Universal Immunization Program

PICU

Paediatric Intensive Care Unit

MMCH

Midnapore Medical College & Hospital

KSDH

Kharagpur Sub Divisional Hospital

VRDL

Viral Research Diagnostic Laboratory

WAZ

Weight-for-age-z score

IQR

Interquartile range

Coeff

Regression Coefficient

95% CI

95% Confidence Interval

LPG

Liquefied Petroleum Gas

SpO2

Saturation of peripheral Oxygen (oxygen saturation)

Tila Khan

Tila Khan is a DBT/Wellcome Trust India Alliance Fellow in the School of Medical Science & Technology at Indian Institute of Technology – Kharagpur, Kharagpur, West Bengal, India. She is a member of the Indian Public Health Association, Lancet Citizens’ Commission on Reimagining India’s Health System and International Society for Infectious Diseases. Dr Khan obtained her undergraduate degree in veterinary medicine in 2005 from Govind Ballabh Pant University of Agriculture & Technology and a Ph.D. in 2012 in Molecular Virology and Vaccinology at the Virginia Tech, University, USA. Her post PhD positions have largely been in clinical and public health research. Her research interests are molecular epidemiology, respiratory infections, virology, vaccinology, vaccine access, vaccine policy and evidence-based research. She received the Robert Austrian Research Award in pneumococcal vaccinology at Scotland in June 2016. She is a recipient of the prestigious DBT/Wellcome Trust India Alliance Early Career Fellowship 2020 in Clinical and Public Health Research on building the evidence base for maternal influenza and RSV immunization to protect infants from lower respiratory tract infections. In June 2023, she was a visiting scholar in the International Vaccine Access Center at the John Hopkins Bloomberg School of Public Health in Baltimore. She has authored several international peer-reviewed publications.

Author contributions

T.K., S.D.B., P.S. and T.G. conceived the study. T.K. prepared the data collection and informed consent forms with inputs from S.D.B. and R.S.D. R.S.D. and R.M.M. screened children for SARI, collected written informed consents, and data from the guardians of children and wards, under the supervision of T.G. and A.M. S.H. performed collection, storage, transport and aliquot of NP swabs. A.J. performed RT-PCR tests with S.H. under the supervision of P.S. R.S.D. maintained data in EpiInfo. S.H. prepared the maps. T.K. analyzed and interpreted the data, which was reviewed by all authors. T.K. prepared the first draft with critical inputs from S.D.B. and all authors. All authors read and approved the final manuscript. All authors have agreed both to be personally accountable for the author’s own contributions and to ensure that questions related to the accuracy or integrity of any part of the work, even ones in which the author was not personally involved, are appropriately investigated, resolved, and the resolution documented in the literature.

Funding

This study was funded by the DBT/Wellcome Trust India Alliance (IA/CPHE/19/1/504599) awarded to TK. The funder had no role in the conceptualization, design, data collection, analysis, decision to publish, or preparation of the manuscript.

Data availability

Data is provided within the manuscript or supplementary information files.

Declarations

Ethics approval and consent to participate

This study was performed in accordance with the principles of the Declaration of Helsinki. The study was approved by the institutional ethics committees of Midnapore Medical College & Hospital ((#IEC/2021/02) and the Indian Institute of Technology Kharagpur ((#IIT/SRIC/DeanSRIC/2021). Approval was obtained from the office of Chief Medical Officer of Health of District West Midnapore ((#1655) and the Department of Health and Family Welfare, West Bengal (#SS(ME)/Spl./184/2021). Written informed consents were obtained from the guardians or caregivers of the children.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

All manuscripts must contain the following sections under the heading ‘Declarations’: