661 Identifying children ‘missing’ during UK COVID-19 lockdown: A retrospective cohort study (2015–2020) of Oxfordshire ED attendances and inpatient diagnoses

2021 
BackgroundOn 23rd March 2020, the UK ‘stay at home’ order heralded the first national lockdown, lasting 7-weeks. Dramatic reductions of paediatric attendances/admissions were noted nationally and internationally, with concerns surrounding diagnosis which may be missing from secondary care.We conducted a systematic review on 9th January 2021, searching PubMed for ‘COVID’, AND ‘lockdown’, AND ‘paediatric’ OR ‘children’, AND ‘attendance’ OR ‘admission’, in all languages using MESH terms. This identified n=41 primary articles exploring paediatric healthcare usage during a lockdown period. These covered Europe, Brazil, USA, Canada, Iran, and India. 44% examined ED settings only;49% reported on admissions for a single disease or specialty;only 2 examined all inpatients.No previous study has explored the changing patterns of diagnoses amongst all paediatric inpatients during a lockdown period, compared to historical years.ObjectivesTo establish the changing patterns of ED attendance and inpatient diagnoses across two Oxfordshire hospitals during the first COVID-19 lockdown in 2020, compared to five historical years (2015–2019).MethodsWe retrospectively reviewed anonymised electronic records for all ED attendances and inpatients aged 0–15 years, across two Oxfordshire hospitals providing secondary and tertiary care services. Discharge ICD-10 coding were analysed to identify significant differences in lockdown inpatient diagnoses, compared to a matched 7-weeks in 2015–2019 (Mann-Whitney U test, admissions-per-week).ResultsDuring the first 2020 lockdown period, 2,843 diagnoses were associated with 1,416 admissions (mean 4.81 diagnoses/patient), compared with 12,458 admissions and 19,946 diagnoses across matched dates 2015–2019 (2.97 diagnoses/patient). Lockdown ED attendances (n=4030) and admissions (n=1416) were reduced by 56.8% and 59.4%, respectively, compared to 2015–2019 (mean n=7446.8 and n=2491.6, respectively). Proportions of admissions from ED and patients’ subsequent length of stay were similar across all years. Reductions in hospital admissions were highest in 1–5-year-old (age <1 =48.4%, 1–5 =67.2%, 6–10 =53.3% and 11–15 years =48.9%).We categorised diagnostic codes significantly reduced during lockdown (’missing’) compared to 2015–2019: 80% were infectious diseases or their sequelae;non-specific pains/aches/malaise (11%) and accidental injury/poisonings (9%) accounted for the remaining 20%. Categories with increased diagnoses (24% of lockdown diagnoses) were ‘related to pandemic screening’, ‘incidental finding/co-morbidity’ and ‘other diagnoses’. We also found significantly greater numbers of neoplasms (benign and malignant) diagnosed during lockdown (p=0.0123).ConclusionsPandemic measures and messaging are altering paediatric disease presentation. Our study confirms large reductions in paediatric ED attendances and inpatient admissions during the first national lockdown, raising concerns of vulnerable children ‘lost’ to secondary care.Our assessment of ‘missing’ paediatric diagnoses uses internationally comparable ICD-10 codes. We therefore postulate that the 80% of infection-related diagnoses ‘missing’ during the lockdown period are driven by a combination of stringent infection-control measures, parents/carers management of mild/self-limiting disease at home, and/or increased anxiety surrounding hospital attendance. As 20% were non-specific or accidental injuries, we remain concerned about significant disease with late presentations or patients with safeguarding concerns who may not be brought to hospital, amongst these patients.Prospective studies are necessary to establish whether parents/carers are adequately supported, have adequate contact with health professionals and feel empowered to use referral pathways for hospital review.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    0
    References
    0
    Citations
    NaN
    KQI
    []