Machine learning forecasting for COVID-19 pandemic-associated effects on paediatric respiratory infections.

OBJECTIVE: The COVID-19 pandemic and subsequent government restrictions have had a major impact on healthcare services and disease transmission, particularly those associated with acute respiratory infection. This study examined non-identifiable routine electronic patient record data from a specialist children's hospital in England, UK, examining the effect of pandemic mitigation measures on seasonal respiratory infection rates compared with forecasts based on open-source, transferable machine learning models. METHODS: We performed a retrospective longitudinal study of respiratory disorder diagnoses between January 2010 and February 2022. All diagnoses were extracted from routine healthcare activity data and diagnosis rates were calculated for several diagnosis groups. To study changes in diagnoses, seasonal forecast models were fit to prerestriction period data and extrapolated. RESULTS: Based on 144 704 diagnoses from 31 002 patients, all but two diagnosis groups saw a marked reduction in diagnosis rates during restrictions. We observed 91%, 89%, 72% and 63% reductions in peak diagnoses of 'respiratory syncytial virus', 'influenza', 'acute nasopharyngitis' and 'acute bronchiolitis', respectively. The machine learning predictive model calculated that total diagnoses were reduced by up to 73% (z-score: -26) versus expected during restrictions and increased by up to 27% (z-score: 8) postrestrictions. CONCLUSIONS: We demonstrate the association between COVID-19 related restrictions and significant reductions in paediatric seasonal respiratory infections. Moreover, while many infection rates have returned to expected levels postrestrictions, others remain supressed or followed atypical winter trends. This study further demonstrates the applicability and efficacy of routine electronic record data and cross-domain time-series forecasting to model, monitor, analyse and address clinically important issues.

An exploration of 11 years of seasonality in diagnoses at GOSH

BackgroundWith the extensive impact of the COVID-19 pandemic and subsequent government interventions on the development, diagnosis and treatment of illnesses, building an understanding of ‘typical’ diagnosis trends at GOSH is critical for predicting future demands and potential clinical challenges. Seasonality analysis is an effective method with which one can explore, model and predict the occurrence of events over time when – as with many common diagnoses at GOSH – they generally exhibit a periodic trend over the year.MethodsTo investigate diagnosis seasonality at GOSH, we have extracted all diagnoses recorded in the Legacy and Epic systems, since the year 2010. We have developed an analytics pipeline that uses these data to compute historical rates for any given diagnosis, or group of diagnoses. Based on these diagnosis rates, our pipeline applies a widely used regressive, multiplicative, seasonal decomposition model with integrated model evaluation.ResultsFor the analysis, a total of 3,480,887 diagnosis events were considered across 29,529 patients between receiving a diagnosis between 1stJanuary 2010 and 30th September 2021. This exploration presents data on many of the common diagnoses at GOSH that exhibit a clear seasonal trend in combination with a statistically significant deviation from that trend since March 2020, likely due to the pandemic. In addition, we illustrate how the available data and model allow us to predict the diagnostic shortfall during the same period.

Impact of the COVID-19 pandemic on radiology appointments in a tertiary children's hospital: a retrospective study.

In this retrospective observational study, we evaluated the impact of the COVID-19 pandemic in London on paediatric radiology activity, as a surrogate of overall hospital activity. We showed a large reduction in overall outpatient imaging activity: 49 250 records occurred in the 371 days post COVID-19 period compared with an expected 67 806 records pre COVID-19 period, representing 18 556 'missed' records. Governmental restrictions were associated with reductions in activity, with the largest reduction in activity during tiers 3 and 4 restrictions. Rescheduling such missed outpatients' appointments represents considerable resource planning and the associated clinical impact on paediatric healthcare remains to be determined.