==== Front BMC Pediatr BMC Pediatr BMC Pediatrics 1471-2431 BioMed Central London 4152 10.1186/s12887-023-04152-5 Research Health service utilisation for acute respiratory infections in infants graduating from the neonatal intensive care unit: a population-based cohort study Stevenson Paul G. 1 Cooper Matthew N. 1 Billingham Wesley 1 de Klerk Nicholas 1 Simpson Shannon J. 123 Strunk Tobias 45 Moore Hannah C. hannah.moore@telethonkids.org.au 56 1 grid.1012.2 0000 0004 1936 7910 Telethon Kids Institute, The University of Western Australia, Perth, WA Australia 2 grid.414659.b 0000 0000 8828 1230 Wal-yan Respiratory Centre, Telethon Kids Institute, Perth, WA Australia 3 grid.1032.0 0000 0004 0375 4078 School of Allied Health, Curtin University, Perth, WA Australia 4 Neonatal Directorate, Child and Adolescent Health Service, Perth, WA Australia 5 grid.414659.b 0000 0000 8828 1230 Wesfarmers Centre of Vaccines and Infectious Diseases, Telethon Kids Institute, PO Box 855, West Perth, WA 6872 Australia 6 grid.1032.0 0000 0004 0375 4078 School of Population Health, Curtin University, Perth, WA Australia 1 7 2023 1 7 2023 2023 23 33516 1 2023 24 6 2023 © The Author(s) 2023 https://creativecommons.org/licenses/by/4.0/ 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. Background Despite advances in neonatal intensive care, babies admitted to Neonatal Intensive Care Units (NICU) suffer from adverse outcomes. We aim to describe the longer-term respiratory infectious morbidity of infants discharged from NICU using state-wide population-based linked data in Western Australia. Study design We used probabilistically linked population-based administrative data to analyse respiratory infection morbidity in a cohort of 23,784 infants admitted to the sole tertiary NICU, born 2002–2013 with follow up to 2015. We analysed incidence rates of secondary care episodes (emergency department presentations and hospitalisations) by acute respiratory infection (ARI) diagnosis, age, gestational age and presence of chronic lung disease (CLD). Poisson regression was used to investigate the differences in rates of ARI hospital admission between gestational age groups and those with CLD, after adjusting for age at hospital admission. Results From 177,367 child-years at risk (i.e., time that a child could experience an ARI outcome), the overall ARI hospitalisation rate for infants and children aged 0–8 years was 71.4/1000 (95% confidence interval, CI: 70.1, 72.6), with the highest rates in infants aged 0–5 months (242.9/1000). For ARI presentations to emergency departments, equivalent rates were 114/1000 (95% CI: 112.4, 115.5) and 337.6/1000, respectively. Bronchiolitis was the most common diagnosis among both types of secondary care, followed by upper respiratory tract infections. Extremely preterm infants (< 28 weeks gestation at birth) were 6.5 (95% CI: 6.0, 7.0) times more likely and those with CLD were 5.0 (95% CI: 4.7, 5.4) times more likely to be subsequently admitted for ARI than those in NICU who were not preterm or had CLD after adjusting for age at hospital admission. Conclusions There is an ongoing burden of ARI in children who graduate from the NICU, especially those born extremely preterm, that persists into early childhood. Early life interventions to prevent respiratory infections in these children and understanding the lifelong impact of early ARI on later lung health are urgent priorities. Supplementary Information The online version contains supplementary material available at 10.1186/s12887-023-04152-5. Keywords Acute respiratory infection Neonatal intensive care unit Hospital morbidity Record linkage Infant Child Preterm Telethon Perth Children’s Hospital Research Fundhttp://dx.doi.org/10.13039/501100020265 Stan Perron Charitable Foundation issue-copyright-statement© BioMed Central Ltd., part of Springer Nature 2023 ==== Body pmcBackground Improvements in obstetric care and neonatal practice have led to significant reductions of neonatal mortality in high-income settings [1–3]. However, over the past decade, studies utilising large population cohorts have reported variable admission rates to Neonatal Intensive Care Units (NICUs) including increasing rates [4] and decreasing rates [5]. Furthermore, preterm birth, one of the more common reasons for NICU admission, remains a substantial health problem on a global level [6] and the prevalence of chronic lung disease of prematurity (CLD) has not changed [7]. Globally, acute respiratory infections (ARI) are a leading contributor to childhood morbidity. The leading viral pathogens associated with ARI in high-income settings are respiratory syncytial virus (RSV), parainfluenza virus, influenza virus and rhinovirus [8–10]. In 2019, RSV-alone was estimated globally to cause 3.6 million hospital admissions for ARI in children aged less than 5 years [11]. Preterm infants, and/or those of low birthweight (< 2500 g) as well as children with CLD are at greater risk of severe morbidity due to viral ARIs [12, 13]. Given these characteristics, collectively, NICU graduates are a particular high-risk group for ARI morbidity and subsequent health service utilisation. Prevention measures, such as immunoprophylaxis with the RSV monoclonal antibody, palivizumab, are recommended to select groups of infants, including those born extremely preterm within the NICU to reduce severe infections caused by RSV, [14, 15] but we have reported low use in our setting [16]. Lung function trajectories are impaired in survivors of very preterm birth, [17] increasing the susceptibility to viral ARIs. However, the longer-term patterns of morbidity in all NICU graduates including those who are not preterm, remains unknown, particularly for respiratory infections occurring beyond infancy. With near-to-market RSV prevention measures aimed at all infants, and not only those preterm, [18] quantifying the long-term burden from ARI in NICU graduates would be useful to evaluate post-licensure effectiveness. Our aim in this study is to describe health service utilisation patterns for ARI in a population cohort of all NICU graduates recorded on an administrative NICU database in terms of incidence rates of subsequent emergency department (ED) presentations and hospitalisations in the first 8 years of life. We describe ED presentations and hospitalisations by age group and ARI diagnosis, and further investigate ARI hospitalisations by selected at-risk groups within the NICU. Methods Study setting Western Australia (WA) is the largest state in Australia, covering approximately 2.5 million square kilometres, and in 2016 had an estimated population of 2.5 million [19]. In 2013, 2.6% of all registered live births in Australia had a NICU admission; since 2004, this proportion has varied from 2.2% to 2.6% [1]. In WA there is only one tertiary neonatal directorate, with one NICU located at Princess Margaret Hospital for Children (now Perth Children’s Hospital) and the other at the only perinatal centre, King Edward Memorial Hospital for Women. Study population and data sources We conducted a retrospective record linkage cohort study. Details of this study and description of the use and effectiveness of palivizumab are described elsewhere [16, 20]. Briefly, the cohort was defined as infants born in WA between 1 January 2002 and 31 December 2013 inclusive, and admitted to the NICU at either Princess Margaret Hospital for Children or King Edward Memorial Hospital after birth. Infants who died before they were discharged from NICU or hospital were excluded; hence our cohort was defined as NICU graduates. The WA NICU database was used to identify the study cohort which has full capture of NICU activity with no missing data on variables required for this analysis. Outcome data came from three databases with varying end dates depending on the time of data extraction prior to linkage. Mortality data was available from the Death Registry, with all deaths in the cohort, between January 2002 and December 2015, being sourced. Hospitalisations were identified from the Hospital Morbidity Data Collection (HMDC) between January 2002 and June 2015 and emergency department presentations were identified from the Emergency Department Data Collection (EDDC) between January 2002 and October 2015. The HMDC includes all hospital admissions and separations from public and private free-standing hospitals across WA and the EDDC contains information for emergency department activity in public and private hospitals across WA. Records from these datasets were probabilistically linked by the Western Australia Department of Health using a set of unique person identifiers and according to established best practice protocols for linking administrative data [21]. Disease classification Each hospitalisation in the linked dataset has a primary diagnosis code, and up to 21 additional diagnosis codes, classified using the International Statistical Classification of Diseases and Related Health Problems 10th Revision, Australian modification (ICD-10-AM). Classification rules used to identify ARI diagnoses in the HMDC and EDDC were based on our prior research in this area [22, 23] and are listed in Supplementary Table 1. Each of primary and additional diagnosis codes were used in the hospital data to classify ARI. ED presentations are either coded with a single ICD-10-AM diagnosis, a symptom code, and/or a field where free text can be entered (see Supplementary Table 1 for search terms). Disease classifications for ARI were made using regular expressions by mapping: the ICD-10-AM codes against the diagnosis field(s), symptom codes (ED presentations only), and free text fields (ED presentations only). As in our previous analyses, symptom codes and free text fields in ED are only used if the ICD code is missing [23]. Episodes of infection Dates of admission and separation were available for the NICU and HMDC dataset and date of presentation was available for the EDDC dataset. As per our previous research using these datasets [24] any hospitalisation (or NICU stay) within 14 days of a previous hospitalisation were collapsed into a single ‘episode of infection’ (episode of care). Each morbidity record (HMDC or EDDC) was assessed separately and given a unique diagnosis classification. Given a single episode of infection may have multiple records from varied sources with numerous diagnosis classifications, the “overall ARI diagnosis” was determined by a hierarchical diagnosis that ranked diseases in order of clinical severity: whooping cough/pertussis, pneumonia, bronchiolitis, influenza, unspecified acute lower respiratory infection, bronchitis, URTI, and other ARI unrelated diagnosis. This hierarchal diagnosis approach follows our previously published work [22, 25]. Time-at-Risk Time-at-risk for each child during the study commenced 24 h following separation from their birth-related NICU episode of care and ended upon either their death or the end of the study period (December 31, 2015), whichever occurred first. Children were not classified as ‘at risk of infection’ if there were already in hospital. Statistical analysis ARI incidence rates were calculated using the number of distinct episodes of care for each overall ARI diagnosis and the total time at risk and reported per 1,000 child-years. ARI incidence rates (for each diagnosis classification) were calculated (as above) separately by age in pre-specified age groups (0–5 months, 6–11 months, 12–23 months, 2–3 years, and 4–8 years) and year of hospitalisation (2002 to 2015). ARI hospitalisation rates (noted as hospital admission rates) were further examined between chosen at-risk groups which included gestational age and presence of CLD. Gestational age at birth was categorised into extremely preterm (< 28 weeks gestation), very preterm (28–32 weeks), moderate to late preterm (33–36 weeks) and term (≥ 37 weeks). CLD was defined using an indicator variable on the NICU database (for oxygen requirement after 36 weeks of postmenstrual age) which is the standardised definition used in NICUs within the Australian and New Zealand Neonatal Network [1] as well as all ICD-10 AM diagnosis codes in the hospitalisation dataset (P27.1). These groups were chosen as they represented criterion groups for palivizumab recommendations under extended guidelines from 2010 in the King Edward Memorial Hospital NICU [20]. Poisson regression was used to investigate the differences in rates of ARI hospital admission between gestational age groups and those with CLD, after adjusting for age at hospital admission. These models included an offset term defined as the natural log of the adjusted time at risk. Results are reported as incidence rate ratios (IRR) with 95% confidence intervals. As per data custodian requirements, individual cell sizes of less than 5 in the results tables have been suppressed. Ethics approvals were granted by the Western Australian Department of Health and Child Adolescent Health Service (RGS2503). Data access was approved by the Western Australian Data Linkage Branch and relevant data custodians. All data cleaning and analysis was completed with R version 3.4.4 with RStudio. Results Our cohort consisted of 24,090 children born in WA between Jan 1, 2002 and Dec 12, 2013 who were admitted to NICU during their birth episode. Of these, 306 infants died before discharge. Subsequently, our analytical cohort consisted of 23,784 NICU graduates providing an overall time-at-risk of 177,367 child-years (with follow-up to 2015). Of this cohort, 12,657 (53.2%) were born preterm (< 37 weeks gestation at birth) with 3,606 (15.2% of the total cohort) being very preterm (28–32 weeks) and 1,049 (4.4% of the total cohort) being extremely preterm (< 28 weeks gestation). Of those extremely preterm, 688 (65.6%) were classified as having CLD and 230 (6.4%) of those very preterm were classified as having CLD. The presence of CLD was low in those who were moderate to late preterm (33–36 weeks gestation, 0.2%) or term (0.1%). Other general cohort characteristics are presented elsewhere [16, 20]. The length of NICU-associated stay from birth to first hospital discharge ranged from 2 to 1,410 days (median 27.5 days; inter-quartile range [IQR] 39 days). From 2002 to 2015, the NICU graduate cohort had 47,443 subsequent hospital admissions and 123,900 subsequent ED presentations, across 57,515 episodes of care. Per individual, the median number of hospital admission episodes of care per NICU graduate was 2 (range 1 to 38; IQR 2 days) and the median number of emergency department presentations was 4 (range 1 to 212; IQR 6 days). ARI hospital admission rates for NICU graduates between 2002 and 2015 are provided in Table 1. The overall hospitalisation rate for infants aged up to 8 years was 71.4/1000 child-years. The highest rates overall were seen in children in the first 6 months of life (242.9/1000) which then declined with age. Rates were highest for bronchiolitis, especially in younger children with the rate of 139.4/1000 in those aged 0–5 months and 95.6/1000 in those aged 6–11 months (Table 1). Rates were also high for URTI, especially in the second year of life (74.2/1000 for those aged 12–23 months). Annual rates of ARI hospital admission between 2002 and 2015 are shown in Fig. 1. ARI hospital admissions declined over this interval, more markedly in infants aged less than 12 months where, for example, rates were 165/1000 (95% CI: 114/1000 to 230/1000) in 2002 for those aged 6–11 months which declined to 107/1000 (95% CI: 87.8/1000 to 130/1000) in 2014; Fig. 1).Table 1 Number and incidence rates of hospital admissions by age and diagnosis in infants attending NICU, 2002–2015 Hierarchical Diagnosis 0–5 months 6–11 months 12–23 months 2–3 years 4–8 years Overall n Ratea (95%CI) n Ratea (95%CI) n Ratea (95%CI) n Ratea (95%CI) n Ratea (95%CI) n Ratea (95%CI) Whooping Cough 49 4.6 (3.4, 6.0) 11 0.9 (0.5, 1.7) 6 0.3 (0.1, 0.6) 8 0.2 (0.1, 0.4)  < 5 0.0 (0.0, 0.1) 75 0.4 (0.3, 0.5) Pneumonia 279 26.0 (23.1, 29.3) 221 18.7 (16.3, 21.3) 418 17.6 (16.0, 19.4) 415 9.8 (8.9, 10.8) 190 3.3 (2.8, 3.8) 1523 8.6 (8.2, 9.0) Bronchiolitis 1493 139.4 (132.4, 146.6) 1132 95.6 (90.1, 101.4) 614 25.9 (23.9, 28.0) 89 2.1 (1.7, 2.6) 12 0.2 (0.1, 0.4) 3340 18.8 (18.2, 19.5) Influenza 53 4.9 (3.7, 6.5) 49 4.1 (3.1, 5.5) 51 2.2 (1.6, 2.8) 61 1.4 (1.1, 1.9) 25 0.4 (0.3, 0.6) 239 1.3 (1.2, 1.5) Unspecified ALRI 97 9.1 (7.3, 11.0) 136 11.5 (9.6, 13.6) 352 14.8 (13.3, 16.5) 340 8.0 (7.2, 8.9) 139 2.4 (2.0, 2.8) 1064 6.0 (5.6, 6.4) Bronchitis 10 0.9 (0.4, 1.7) 19 1.6 (1.0, 2.5) 33 1.4 (1.0, 2.0) 22 0.5 (0.3, 0.8) 7 0.1 (0.0, 0.2) 91 0.5 (0.4, 0.6) URTI 621 58.0 (53.5, 62.7) 738 62.3 (57.9, 67.0) 1760 74.2 (70.8, 77.8) 2047 48.3 (46.3, 50.5) 1158 19.9 (18.8, 21.1) 6324 35.7 (34.8, 36.5) Overall ARI 2602 242.9 (233.6, 252.4) 2306 194.8 (186.9, 202.9) 3234 136.4 (131.7, 141.1) 2982 70.4 (67.9, 73.0) 1532 26.3 (25.0, 27.7) 12,656 71.4 (70.1, 72.6) Abbreviations: ALRI acute lower respiratory infection, ARI acute respiratory infections, n number, NICU neonatal intensive care unit, URTI upper respiratory tract infection aRate per 1000 child-years Fig. 1 Annual rates of ARI hospital admissions from 2002 to 2015 by age at admission; 95% confidence intervals are indicated by error bars The frequency of ED presentations in NICU graduates from 2002–2015 by overall ARI diagnosis classification and age, was similar to that observed for hospital admissions (Table 2). Overall, the ARI ED presentation rate was 114.0/1000 and was highest in those aged 0–5 months (337.6/1000) and 6–11 months (366.0/1000) and then declined with age. Bronchiolitis and URTI were the most common diagnoses in infants aged less than 12 months and URTI remained the most common diagnosis in children beyond 2 years of age (e.g., 92.7/1000 in those aged 2–3 years and 31.1/1000 in those aged 4–8 years; Table 2). Annual rates of ED presentations for ARI between 2002 and 2015 are shown in Fig. 2. Unlike hospital admissions, overall ED presentation rates did not change over time, except for in those aged 0–5 months, where after rates peaked at 258/1000 (95% CI: 226/1000 to 294/1000) in 2008, rates declined to 85/1000 (95% CI: 59/1000 to 124/1000) in 2014 (Fig. 2).Table 2 Number and incidence rates of emergency department presentations by age and diagnosis in infants attending NICU, 2002–2015 Hierarchical Diagnosis 0–5 months 6–11 months 12–23 months 2–3 years 4–8 years Overall n Ratea (95%CI) n Ratea (95%CI) n Ratea (95%CI) n Ratea (95%CI) n Ratea (95%CI) n Ratea (95%CI) Whooping Cough 46 4.3 (3.1, 5.7) 15 1.3 (0.7, 2.1) 12 0.5 (0.3, 0.9) 14 0.3 (0.2, 0.6) 10 0.2 (0.1, 0.3) 97 0.5 (0.4, 0.7) Pneumonia 119 11.1 (9.2, 13.3) 170 14.4 (12.3, 16.7) 456 19.2 (17.5, 21.1) 484 11.4 (10.4, 12.5) 218 3.7 (3.3, 4.3) 1447 8.2 (7.7, 8.6) Bronchiolitis 1933 180.4 (172.5, 188.7) 1937 163.6 (156.4, 171.1) 847 35.7 (33.3, 38.2) 116 2.7 (2.3, 3.3)  < 5 0.0 (0.0, 0.1) 4835 27.3 (26.5, 28.0) Influenza 11 1.0 (0.5, 1.8) 20 1.7 (1.0, 2.6) 37 1.6 (1.1, 2.2) 47 1.1 (0.8, 1.5) 32 0.6 (0.4, 0.8) 147 0.8 (0.7, 1.0) Unspecified ALRI 61 5.7 (4.4, 7.3) 70 5.9 (4.6, 7.5) 156 6.6 (5.6, 7.7) 156 3.7 (3.1, 4.3) 69 1.2 (0.9, 1.5) 512 2.9 (2.6, 3.1) Bronchitis 19 1.8 (1.1, 2.8) 23 1.9 (1.2, 2.9) 46 1.9 (1.4, 2.6) 40 0.9 (0.7, 1.3) 13 0.2 (0.1, 0.4) 141 0.8 (0.7, 0.9) URTI 1428 133.3 (126.5, 140.4) 2097 177.2 (169.7, 184.9) 3775 159.2 (154.1, 164.3) 3927 92.7 (89.8, 95.7) 1807 31.1 (29.7, 32.5) 13,034 73.5 (72.2, 74.8) Overall ARI 3617 337.6 (326.7, 348.8) 4332 366.0 (355.2, 377.1) 5329 224.7 (218.7, 230.8) 4784 113.0 (109.8, 116.2) 2151 37.0 (35.4, 38.6) 20,213 114.0 (112.4, 115.5) Abbreviations: ALRI acute lower respiratory infections, ARI acute respiratory infection, n number, NICU neonatal intensive care unit, URTI upper respiratory tract infection aRate per 1000 child-years Fig. 2 Annual rates of ARI emergency department presentations from 2002 to 2015 by age at presentation; 95% confidence intervals are indicated by error bars. NOTE: Annual presentation rates for whooping cough not shown due to small numbers We assessed ARI hospitalisation rates within gestational age groups and by CLD status for those who were classified as extremely or very preterm (Table 3). ARI admission rates were highest in those born extremely preterm (< 28 weeks gestation at birth) with CLD and these higher rates persisted into early childhood. The ARI rate for 0–5 months for those extremely preterm with CLD was 1149.6/1000 compared with 181.9/1000 for those born at term. The group with the next highest ARI rates across all ages were those born very preterm (28–32 weeks) with CLD (e.g., 731.3/1000 for those 0–5 months) and this was higher than those extremely preterm with no CLD (e.g., 581.8/1000 for those 0–5 months, Table 3). ARI rates declined with age across all groups, however the discrepancy in rates between preterm groups with and without CLD persisted into early childhood. For example, at age 4–8 years, ARI rates were 60.7/1000 in those born extremely preterm with CLD and 44.9/1000 in those very preterm with CLD which was still 2.6 and 1.9 times higher, respectively, than ARI rates at age 4–8 years in those born at term (23.6/1000; Table 3). We conducted Poisson regression analysis to formally compare ARI hospitalisation rates between age at admission, and levels of preterm birth shown in Table 4 and presence of CLD in Table 5. After adjusting for age at subsequent hospital admission, on average, children born extremely preterm experienced ARI-related subsequent hospital admissions 6.51 times higher (95% CI: 6.01–7.04) than children born at term (Table 4). The largest discrepancy was seen with bronchiolitis. After adjusting for age at hospitalisation, on average, the IRR for a child of any gestational age with CLD being hospitalised with ARI, compared to a child without CLD, was 5.04 (95% CI: 4.69 to 5.41) with the largest discrepancy seen with bronchiolitis and unspecified ALRI (Table 5).Table 3 Number and incidence rates of ARI hospital admission by gestational age and chronic lung disease at different ages of admission Group 0–5 months 6–11 months 12–23 months 2–3 years 4–8 years Overall n Ratea (95%CI) n Ratea (95%CI) n Ratea (95%CI) n Ratea (95%CI) n Ratea (95%CI) n Ratea (95%CI)  < 28w + CLD 157 1149.6 (976.8, 1344.2) 227 686.7 (600.3,782.1) 311 459.5 (409.8, 513.5) 265 215.4 (190.2, 242.9) 102 60.7 (49.5, 73.7) 1062 213.4 (200.7, 226.6)  < 28w no CLD 61 581.8 (445.1, 747.4) 79 441.9 (346.9, 550.8) 85 236.8 (189.2, 292.9) 68 102.5 (79.6, 129.9) 32 31.9 (21.8, 45.0) 325 119.0 (106.4, 132.7) 28–32w + CLD 49 731.3 (541.0, 966.8) 57 523.0 (396.1, 677.6) 59 270.0 (205.5, 348.3) 65 170.3 (131.4, 217.0) 24 44.9 (28.8, 66.9) 254 152.4 (134.2, 172.3) 28–32w no CLD 372 433.4 (390.5, 479.8) 317 288.8 (257.8, 322.4) 400 181.9 (164.5, 200.6) 344 85.7 (76.9, 95.2) 185 32.5 (28.0, 37.5) 1618 97.1 (92.4, 101.9) 33–36w 997 235.3 (221.0, 250.4) 796 174.2 (162.3, 186.7) 1168 127.6 (120.4, 135.1) 1048 63.8 (60.0, 67.8) 569 24.7 (22.7, 26.9) 4578 66.3 (64.4, 68.2)  ≥ 37w 966 181.9 (170.6, 193.8) 830 149.6 (139.6, 160.1) 1211 109.0 (103.0, 115.3) 1192 60.7 (57.3, 64.2) 620 23.6 (21.8, 25.6) 4819 58.6 (57.0, 60.3) Abbreviations: CLD, chronic lung disease; n, number; NICU, neonatal intensive care unit; w, weeks gestation aRate per 1000 child-years Table 4 Incidence rate ratios by admission and gestational age from Poisson regression by ARI diagnosis Risk Group Incidence rate ratio (95% confidence interval) Overall Whooping Cough Pneumonia Bronchiolitis Influenza Unspecified ALRI Bronchitis Age at admission  0–5 months Reference Reference Reference Reference Reference Reference Reference  6–11 months 0.670 *** (0.627, 0.716) 0.192 *** (0.094, 0.355) 0.672 *** (0.562, 0.801) 0.635 *** (0.588, 0.686) 0.799 (0.540, 1.179) 1.200 (0.926, 1.561) 1.741 (0.827, 3.900)  12–23 months 0.314 *** (0.293, 0.336) 0.061 *** (0.024, 0.133) 0.633 *** (0.544, 0.737) 0.172 *** (0.156, 0.188) 0.414 *** (0.282, 0.610) 1.547 *** (1.241, 1.948) 1.347 (0.684, 2.897)  2–3 years 0.107 *** (0.099, 0.116) 0.039 *** (0.017, 0.077) 0.350 *** (0.301, 0.408) 0.014 *** (0.011, 0.017) 0.277 *** (0.191, 0.401) 0.834 (0.669, 1.051) 0.501 (0.242, 1.112)  4–8 years 0.036 *** (0.032, 0.040) 0.008 *** (0.000, 0.037) 0.116 *** (0.096, 0.139) 0.001 *** (0.001, 0.002) 0.082 *** (0.050, 0.131) 0.247 *** (0.191, 0.321) 0.130 *** (0.047, 0.339) Gestational age   < 28w 6.510 *** (6.013, 7.041) 5.246 *** (2.080, 11.599) 4.999 *** (4.272, 5.831) 6.783 *** (6.067, 7.572) 3.922 *** (2.492, 5.977) 4.082 *** (3.389, 4.892) 4.610 ** (1.683, 10.809)  28–32w 2.637*** (2.449, 2.837) 1.944 (0.817, 4.147) 2.106*** (1.805, 2.449) 3.296*** (2.983, 3.641) 2.290*** (1.555, 3.316) 1.521*** (1.253, 1.835) 2.711*** (1.367, 5.173)  33–36w 1.351*** (1.273, 1.434) 1.684* (1.009, 2.850 1.169* (1.036, 1.319) 1.563*** (1.438, 1.700) 1.299 (0.963, 1.753) 0.960 (0.832, 1.106) 2.366*** (1.462, 3.930)   ≥ 37w Reference Reference Reference Reference Reference Reference Reference Abbreviations: ALRI acute lower respiratory infection, ARI acute respiratory infection, w weeks gestation ***p < 0.001; ** p < 0.01; * p < 0.05 Table 5 Incidence rate ratios by admission age and CLD from Poisson regression by ARI diagnosis Risk Group Incidence rate ratio (95% confidence interval) Overall Whooping Cough Pneumonia Bronchiolitis Influenza Unspecified ALRI Bronchitis Age at admission  0–5 months Reference Reference Reference Reference Reference Reference Reference  6–11 months 0.670 *** (0.627, 0.716) 0.190 *** (0.094, 0.352) 0.675 *** (0.565, 0.805) 0.641 *** (0.593, 0.693) 0.800 (0.541, 1.180) 1.185 (0.914, 1.541) 1.636 (0.776, 3.667)  12–23 months 0.314 *** (0.293, 0.336) 0.052 *** (0.020, 0.111) 0.636 *** (0.546, 0.740) 0.173 *** (0.157, 0.190) 0.415 *** (0.282, 0.610) 1.526 *** (1.224, 1.921) 1.416 (0.723, 3.032)  2–3 years 0.111 *** (0.103, 0.120) 0.038 *** (0.017, 0.077) 0.353 *** (0.303, 0.411) 0.014 *** (0.011, 0.017) 0.278 *** (0.192, 0.402) 0.825 (0.661, 1.039) 0.528 (0.256, 1.167)  4–8 years 0.034 *** (0.031, 0.038) 0.004 *** (0.000, 0.019) 0.118 *** (0.098, 0.141) 0.001 *** (0.001, 0.002) 0.083 *** (0.051, 0.132) 0.246 *** (0.190, 0.319) 0.142 *** (0.052, 0.371) Chronic Lung Disease  No CLD Reference Reference Reference Reference Reference Reference Reference  CLD 5.040 *** (4.694, 5.405) 5.112 *** (2.364, 9.778) 4.629 *** (4.004, 5.326) 5.123 *** (4.635, 5.650) 3.662 *** (2.414, 5.340) 5.178 *** (4.396, 6.066) 3.928 *** (1.968, 7.097) Abbreviations: ALRI acute lower respiratory infection, ARI acute respiratory infection, CLD chronic lung disease ***p < 0.001; ** p < 0.01; * p < 0.05 Discussion From a population cohort of infants being discharged from NICU, we have successfully described the health service utilisation and the resulting ongoing burden of ARI with overall rates of ARI in children up to age 8 years of 71/1000 child-years for hospital admission and 114/1000 child-years for ED presentations. Although ARI secondary care rates in NICU graduates were highest in the first year of life, the burden of ARI persisted into early childhood, even in those who were not in our pre-determined at-risk groups such as those born preterm, with or without CLD. However, within the NICU cohort, extreme preterm infants or those with CLD were 5–6 times more likely to be subsequently admitted to hospital for an ARI compared to term-born NICU graduates or those without CLD. A previous population-based birth cohort of 337,909 births in WA between 2001 and 2012 reported ARI rates in infants aged less than 12 months at 43.7/1000 [26]. Here, we report rates of ARI in NICU graduates are approximately 4.5–5.5 times higher, representing a significant disproportionate burden. These high rates of ARI hospitalisation and ED presentation will likely represent a significant health service cost and associated societal and economic burden that warrants attention, especially as rates of preterm birth and CLD are not declining. Bronchiolitis had the highest rate of all ARI disease categories, a finding consistent with numerous studies. Respiratory syncytial virus (RSV) is the major pathogen associated with bronchiolitis and pneumonia in children [9, 27] and there is renewed interest in global RSV prevention through maternal vaccination strategies and single dose longer-acting monoclonal antibodies; the latter of which has been shown in clinical trials to reduce medically-attended RSV ARI in preterm infants by 70% [28]. The licensure and introduction of such promising therapeutics is therefore likely to be a benefit in reducing the ARI burden in NICU graduates and associated studies addressing awareness and acceptability of future RSV prevention measures studies are now needed. Single-dose long-acting monoclonal antibodies are close to market for RSV, [18] and while they are expected to reduce ARI rates in young infants, their impact on long-term ARI morbidity is unknown. Our data presented here provide robust baseline data that can be compared in post-licensure effectiveness studies once such long-acting monoclonal antibodies are available in our population. Similar to previous total population birth cohort analyses, [29] we report an ongoing burden of ARI in those born preterm and with CLD up to age 8 years in our NICU graduating cohort. There is now a large body of evidence supporting the detrimental role of early life ARI on poor respiratory health outcomes through life, including low lung function, [30] the early origins of chronic obstructive pulmonary disorder (COPD) [31] and asthma diagnosis. Indeed, a popular paradigm postulates that recurrent respiratory viral infections at critical time periods of immune and lung development in infancy and childhood, coupled with allergic sensitisation, are associated with the development of asthma [32]. It is likely that the negative impacts of early life ARI are further amplified for survivors of preterm birth, who have immature lungs and other adverse early life exposures such as invasive respiratory support in the NICU. While no studies have explicitly looked at the link between early life ARI and later lung health outcomes in survivors of preterm birth, recent studies show that the respiratory health burden after preterm birth is high. Survivors of preterm birth are up to five times more likely to develop childhood wheeze or “asthma” than their term counterparts, [33] and have reduced lung function during childhood [17] that is progressively diverging further from the “normal” trajectory over time [34, 35]. The mechanisms underlying increased risk of more frequent and/or severe ARI in the years following preterm birth are currently unclear. However, emerging data suggest a multifactorial foundation. Preterm infants have an immature cellular innate and adaptive immune system at birth, as well as lower immunoglobulin levels since maternal transfer normally occurs in the third trimester of pregnancy [36, 37]. Further, the microbiome helps shape the innate immune system and contributes to effective barrier function, but infants born preterm have low microbial diversity and altered dominant microbial species [38]. Altered microbiota composition persists to young-adulthood and is correlated with reduced lung function in this population [39]. Our findings of increased severe ARI through childhood in preterm babies from NICU and emerging evidence from these research areas, suggest that those born preterm have reduced ability to detect and effectively eliminate respiratory pathogens during early life and beyond; highlighting the importance of further work in this area. However, as 47% of our NICU cohort were born at term, the ongoing burden of ARI is not just limited to preterm birth and measures to prevent respiratory pathogens in early life need to be applicable to all infants. The main strength of our study is in the availability and use of population-based datasets, with complete perinatal and demographic information to form the study cohort. This allowed us to document age-specific ARI incidence rates up to age 8 years, where other studies have only assessed frequency of re-hospitalisation following NICU discharge up to age 2 or 3 years [40, 41]. The higher frequency reported in these studies, along with our findings here, suggests this is indicative of a true higher disease burden in NICU graduates, and not due to health seeking practices in our jurisdiction. Our observational study is not without limitations. First, there are inherent difficulties with using ICD codes to distinguish different respiratory infection presentations, as highlighted in our previous analyses of administrative health data [9]. However, we have used validated sets of diagnosis codes used in our prior research to classify each ARI diagnosis. While these outcomes represent clinical diagnoses and not virus-specific outcomes, we feel these are still useful as baseline data for future immunisation strategies against RSV, considering that in our population, the most common aetiological agent for bronchiolitis and pneumonia hospitalisations is RSV. Second, we chose a priori to examine age-specific ARI rates within two key risk groups in NICU (children born preterm and those with CLD) based on extended guidelines for palivizumab recommendations in our jurisdiction to provide further background for assessing palivizumab effectiveness [20]. However, we acknowledge there are other risk groups that necessitate admission to NICU. Finally, there are other socio-demographic factors and population subgroups that we did not assess ARI rates for in the cohort of NICU graduates including Aboriginal and/or Torres Strait Islander children and children from lower socio-economic backgrounds. Conclusions Our study highlights the on-going impact of NICU admittance, as well as preterm birth and its complications (CLD) on respiratory health, and well-designed studies examining the lifelong impact of early ARI on later lung health are urgently needed in this population. Minimising early life respiratory infections in NICU graduates through parental education and awareness on the importance of on-time current infant vaccines targeting respiratory infections and future RSV monoclonal antibodies as well as non-pharmaceutical prevention measures such as increased hand hygiene, social distancing from others when symptomatic, should also be an urgent priority in this high-risk population. Supplementary Information Additional file 1: Supplementary Table 1. Diagnosis, symptom and discharge codes for acute respiratory infection in hospitalisation and emergency department data. Abbreviations ARI Acute respiratory infection CLD Chronic lung disease ED Emergency department EDDC Emergency department data collection HMDC Hospital morbidity data collection ICD International Classification of Diseases IRR Incidence rate ratio NICU Neonatal intensive care unit RSV Respiratory syncytial virus URTI Upper respiratory tract infection WA Western Australia Acknowledgements We would like to like acknowledge Dr Anthony Keil, Professor Tom Snelling and Professor Peter Richmond as co-investigators on this project. We would like to thank the staff within the data services at the Western Australian Department of Health and data custodians of the databases used in this study: Neonatal Intensive Care Unit Database, Death Register, Hospital Morbidity Data Collection and Emergency Department Data Collection that was used in this analysis. Authors’ contributions HCM, NdK, TS conceptualised the study and acquired funding. PGS led the data cleaning and analysis with MNC, wrote the first draft of the manuscript and prepared Figs. 1 and 2. WB assisted in secondary data analysis. SJS provided expert opinion on the manuscript and interpretation of the data. HCM initially cleaned the data, had supervision over data analysis along with NdK and edited drafts of the manuscript. All authors contributed to the editing of the manuscript, interpretation of results and approved the final version for submission. Funding Funding for this study was provided by the Telethon-Perth Children’s Hospital Research Fund. HCM is funded by a Stan Perron Charitable Foundation Fellowship and further supported by the Future Health Research and Innovation Fund through the WA Near-miss Awards program. Availability of data and materials This study used individual record data. Data are not available to the public but can be required from the Western Australian Department of Health. Code for data analyses can be made available by request from HCM. Declarations Ethics approval and consent to participate Ethics approvals were granted by the Western Australian Department of Health and Child and Adolescent Health Service (RGS#2503). Data access was approved by Data Services at the Western Australian Department of Health and relevant data custodians. Consent to participate is not relevant for this project as the study design is focused on analysis of administrative pre-collected data that are held in database collections. There was no contact with participants in this study. The need for informed consent was waived by the Department of Health WA Human Research Ethics Committee and the Child and Adolescent Health Service Human Research Ethics Committee (RGS#2503). All methods have been performed in accordance with the Declaration of Helsinki. Consent for publication Not applicable. Competing interests HCM has received institutional honoraria for participation in Merk Sharpe and Dohme (Australia) Expert Input Forums on RSV epidemiology and is in receipt of a Merck Investigator Studies Program; none of these are related to the work reported in this manuscript. Other authors declare that they have no competing interests. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. ==== Refs References 1. Chow SSW lMR, Hossain S., Haslam R, Lui K. Report of the Australian and New Zealand Neonatal Network 2013. Sydney; 2015.  https://npesu.unsw.edu.au/sites/default/files/npesu/surveillances/Report%20of%20the%20Australian%20and%20New%20Zealand%20Neonatal%20Network%202013.pdf. 2. Quinn M Gephart S Evidence for Implementation Strategies to Provide Palliative Care in the Neonatal Intensive Care Unit Adv Neonatal Care 2016 16 6 430 438 10.1097/ANC.0000000000000354 27775989 3. Kotecha SJ Dunstan FD Kotecha S Long term respiratory outcomes of late preterm-born infants Semin Fetal Neonatal Med 2012 17 2 77 81 10.1016/j.siny.2012.01.004 22277112 4. Harrison W Goodman D Epidemiologic trends in Neonatal Intensive Care, 2007–2012 JAMA Pediatr 2015 169 9 855 862 10.1001/jamapediatrics.2015.1305 26214387 5. Braun D Braun E Chiu V Burgos AE Gupta M Volodarskiy M Trends in Neonatal Intensive Care Unit utilization in a large Integrated Health Care System JAMA Netw Open 2020 3 6 e205239 10.1001/jamanetworkopen.2020.5239 32556257 6. Blencowe H Cousens S Oestergaard MZ Chou D Moller AB Narwal R National, regional, and worldwide estimates of preterm birth rates in the year 2010 with time trends since 1990 for selected countries: a systematic analysis and implications Lancet 2012 379 9832 2162 2172 10.1016/S0140-6736(12)60820-4 22682464 7. Owen LS Manley BJ Davis PG Doyle LW The evolution of modern respiratory care for preterm infants Lancet 2017 389 10079 1649 1659 10.1016/S0140-6736(17)30312-4 28443559 8. Moore HC Lim FJ Fathima P Barnes R Smith DW de Klerk N Assessing the burden of laboratory-confirmed respiratory syncytial virus infection in a population cohort of Australian children through record linkage J Infect Dis 2020 222 1 92 101 10.1093/infdis/jiaa058 32031631 9. Lim FJ Blyth CC Fathima P de Klerk N Moore HC Record linkage study of the pathogen-specific burden of respiratory viruses in children Influenza Other Respir Viruses 2017 11 6 502 510 10.1111/irv.12508 28991397 10. Taylor S Lopez P Weckx L Borja-Tabora C Ulloa-Gutierrez R Lazcano-Ponce E Respiratory viruses and influenza-like illness: Epidemiology and outcomes in children aged 6 months to 10 years in a multi-country population sample J Infect 2017 74 1 29 41 10.1016/j.jinf.2016.09.003 27667752 11. Li Y Wang X Blau DM Caballero MT Feikin DR Gill CJ Global, regional, and national disease burden estimates of acute lower respiratory infections due to respiratory syncytial virus in children younger than 5 years in 2019: a systematic analysis Lancet 2022 399 10340 2047 2064 10.1016/S0140-6736(22)00478-0 35598608 12. Homaira N Rawlinson W Snelling TL Jaffe A Effectiveness of palivizumab in preventing RSV hospitalization in high risk children: A real-world perspective Int J Pediatr 2014 2014 571609 10.1155/2014/571609 25548575 13. Burgner DP Doherty D Humphreys J Currie A Simmer K Charles A Maternal chorioamnionitis and postneonatal respiratory tract infection in ex-preterm infants J Pediatr 2017 184 62 7.e2 10.1016/j.jpeds.2017.01.037 28233549 14. The IMpact-RSV Study Group Palivizumab, a humanized respiratory syncytial virus monoclonal antibody, reduces hospitalization from respiratory syncytial virus infection in high-risk infants Pediatrics 1998 102 3 531 537 10.1542/peds.102.3.531 15. Feltes TF Cabalka AK Meissner HC Piazza FM Carlin DA Top FH Jr Palivizumab prophylaxis reduces hospitalization due to respiratory syncytial virus in young children with hemodynamically significant congenital heart disease J Pediatr 2003 143 4 532 540 10.1067/S0022-3476(03)00454-2 14571236 16. Xu R Fathima P Strunk T de Klerk N Snelling TL Richmond PC RSV prophylaxis use in high-risk infants in Western Australia, 2002–2013: a record linkage cohort study BMC Pediatr 2020 20 1 490 10.1186/s12887-020-02390-5 33092566 17. Simpson SJ Logie KM O'Dea CA Banton GL Murray C Wilson AC Altered lung structure and function in mid-childhood survivors of very preterm birth Thorax 2017 72 8 702 711 10.1136/thoraxjnl-2016-208985 28119488 18. Mazur NI Terstappen J Baral R Bardají A Beutels P Buchholz UJ Cohen C Crowe JE Cutland CL Eckert L Feikin D Respiratory syncytial virus prevention within reach: the vaccine and monoclonal antibody landscape The Lancet Infectious Diseases. 2023 23 1 e2 1 10.1016/S1473-3099(22)00291-2 35952703 19. Australian Bureau of Statistics. 2016 Census QuickStats 2018 [11 July 2019]. Available from: https://quickstats.censusdata.abs.gov.au/census_services/getproduct/census/2016/quickstat/036. Accessed 11 Jul 2019. 20. Moore HC de Klerk N Richmond PC Fathima P Xu R Keil AD Effectiveness of palivizumab against respiratory syncytial virus: Cohort and case series analysis J Pediatr 2019 214 121 7.e1 10.1016/j.jpeds.2019.06.058 31378522 21. Shay DK Holman RC Newman RD Liu LL Stout JW Anderson LJ Bronchiolitis-associated hospitalizations among US children, 1980–1996 JAMA 1999 282 15 1440 1446 10.1001/jama.282.15.1440 10535434 22. Moore HC de Klerk N Jacoby P Richmond P Lehmann D Can linked emergency department data help assess the out-of-hospital burden of acute lower respiratory infections? A population-based cohort study BMC Public Health 2012 12 703 10.1186/1471-2458-12-703 22928805 23. Barnes R Blyth CC de Klerk N Lee WH Borland ML Richmond P Geographical disparities in emergency department presentations for acute respiratory infections and risk factors for presenting: a population-based cohort study of Western Australian children BMJ Open 2019 9 2 e025360 10.1136/bmjopen-2018-025360 30804033 24. Moore HC de Klerk N Richmond P Lehmann D A retrospective population-based cohort study identifying target areas for prevention of acute lower respiratory infections in children BMC Public Health 2010 10 757 10.1186/1471-2458-10-757 21138593 25. Moore HC Lehmann D de Klerk N Jacoby P Richmond PC Reduction in disparity for pneumonia hospitalisations between Australian Indigenous and non-Indigenous children J Epidemiol Community Health 2012 66 6 489 494 10.1136/jech.2010.122762 21258115 26. Moore HC de Klerk N Blyth CC Gilbert R Fathima P Zylbersztejn A Temporal trends and socioeconomic differences in acute respiratory infection hospitalisations in children: an intercountry comparison of birth cohort studies in Western Australia, England and Scotland BMJ Open 2019 9 5 e028710 10.1136/bmjopen-2018-028710 31110110 27. Bhuiyan MU Snelling TL West R Lang J Rahman T Granland C The contribution of viruses and bacteria to community-acquired pneumonia in vaccinated children: a case-control study Thorax 2019 74 3 261 269 10.1136/thoraxjnl-2018-212096 30337417 28. Griffin MP Yuan Y Takas T Domachowske JB Madhi SA Manzoni P Single-dose nirsevimab for prevention of RSV in preterm infants N Engl J Med 2020 383 5 415 425 10.1056/NEJMoa1913556 32726528 29. Miller JE Hammond GC Strunk T Moore HC Leonard H Carter KW Association of gestational age and growth measures at birth with infection-related admissions to hospital throughout childhood: a population-based, data-linkage study from Western Australia Lancet Infect Dis 2016 16 8 952 961 10.1016/S1473-3099(16)00150-X 27052469 30. van Meel ER den Dekker HT Elbert NJ Jansen PW Moll HA Reiss IK A population-based prospective cohort study examining the influence of early-life respiratory tract infections on school-age lung function and asthma Thorax 2018 73 2 167 173 10.1136/thoraxjnl-2017-210149 29101282 31. Martinez FD Early-life origins of chronic obstructive pulmonary disease N Engl J Med 2016 375 9 871 878 10.1056/NEJMra1603287 27579637 32. Holt PG Sly PD Viral infections and atopy in asthma pathogenesis: new rationales for asthma prevention and treatment Nat Med 2012 18 5 726 735 10.1038/nm.2768 22561836 33. Been JV Lugtenberg MJ Smets E van Schayck CP Kramer BW Mommers M Preterm birth and childhood wheezing disorders: a systematic review and meta-analysis PLoS Med 2014 11 1 e1001596 10.1371/journal.pmed.1001596 24492409 34. Simpson SJ Turkovic L Wilson AC Verheggen M Logie KM Pillow JJ Lung function trajectories throughout childhood in survivors of very preterm birth: a longitudinal cohort study Lancet Child Adolesc Health 2018 2 5 350 359 10.1016/S2352-4642(18)30064-6 30169268 35. Doyle LW Adams AM Robertson C Ranganathan S Davis NM Lee KJ Increasing airway obstruction from 8 to 18 years in extremely preterm/low-birthweight survivors born in the surfactant era Thorax 2017 72 8 712 719 10.1136/thoraxjnl-2016-208524 27601432 36. Collins A Weitkamp JH Wynn JL Why are preterm newborns at increased risk of infection? Arch Dis Child Fetal Neonatal Ed 2018 103 4 F391 F394 10.1136/archdischild-2017-313595 29382648 37. Townsi N Laing IA Hall GL Simpson SJ The impact of respiratory viruses on lung health after preterm birth Eur Clin Respir J 2018 5 1 1487214 10.1080/20018525.2018.1487214 30128088 38. Dahl C Stigum H Valeur J Iszatt N Lenters V Peddada S Preterm infants have distinct microbiomes not explained by mode of delivery, breastfeeding duration or antibiotic exposure Int J Epidemiol 2018 47 5 1658 1669 10.1093/ije/dyy064 29688458 39. Rofael SAD McHugh TD Troughton R Beckmann J Spratt D Marlow N Airway microbiome in adult survivors of extremely preterm birth: the EPICure study Eur Respir J. 2019 53 1 1801225 10.1183/13993003.01225-2018 30464016 40. Do CHT Børresen ML Pedersen FK Geskus RB Kruse AY Rates of rehospitalisation in the first 2 years among preterm infants discharged from the NICU of a tertiary children hospital in Vietnam: a follow-up study BMJ Open 2020 10 10 e036484 10.1136/bmjopen-2019-036484 33020086 41. Hong T Bolisetty S Bajuk B Abdel-Latif M Oei J Jaffe A A population study of respiratory rehospitalisation in very preterm infants in the first 3 years of life J Paediatr Child Health 2016 52 7 715 721 10.1111/jpc.13205 27203818