The impact of COVID-19 on private healthcare service utilisation: time series analysis in the capital region of Finland during 2020–2022

Background: Most current studies on COVID-19 have focused on the first weeks or months of the pandemic or have addressed specific diseases. The long-term effects of the COVID-19 pandemic on healthcare services are insufficiently understood. We analysed the long-term effects of the COVID-19 pandemic on outpatient healthcare utilisation in the full spectrum of diseases in Uusimaa, the capital region of Finland. Methods: Our data included 632 466 individual patients between the ages of 18 and 65 and 6 521 394 visits to healthcare personnel from electronic health records. We fitted an autoregressive integrated moving average (ARIMA) model to pre-2020 data to predict the expected quantity of diagnoses for the period between 1 January 2020 and 16 June 2022. Expected and observed quantities of diagnoses were compared. Results: The overall quantity of diagnoses declined by one-fourth immediately following the onset of the pandemic and remained around 15% below predicted values for most of the pandemic. After the pandemic-related restrictions were lifted, the total diagnostic activity started to recover to pre-pandemic projection levels. However, this recovery has been mainly driven by upper respiratory system-related activity. The number of diagnoses for many diagnosis groups has remained below the predicted levels, even after the concurrent removal of mobility restrictions and increased coverage of vaccinations in this population. Conclusions: The pandemic resulted in an overall reduction in outpatient healthcare utilisation which persisted for 30 months. While the overall diagnostic activity has eventually recovered to predicted levels, many classes of diagnoses display reduced levels in the study population over the follow-up period. Some diseases that may have long-lasting effects when left untreated seem to remain underdiagnosed, potentially increasing pressure on the healthcare system in the future.