Exploring changes in non-COVID-19 hospitalization rates before and during COVID-19

In a recent study posted to the medRxiv* preprint server, researchers assessed the trends in non-coronavirus disease 2019 (COVID-19) hospitalizations before and during the COVID-19 pandemic in the United States (US) between 2017 and 2021.

Study: Trends in non-COVID-19 hospitalizations prior to and during the COVID-19 pandemic period, United States, 2017 – 2021. Image Credit: Gorodenkoff/Shutterstock


The shift in healthcare utilization with lockdowns and non-pharmaceutical interventions (NPI) such as face mask use during COVID-19, and aversion to approaching medical care due to COVID-19 concerns, influenced viral transmissions.

The count of respiratory syncytial virus (RSV) and influenza virus infections sharply declined with a concomitant increase in mental health conditions.

The analyses of trends in non-COVID-19 hospitalizations during COVID-19 mainly focused on particular health disorder subsets. Thus, comprehensive analyses of diagnosis-specific patterns compared to the overall hospitalizations before and during COVID-19 are required. Such studies would yield a comprehensive estimation of the COVID-19 burden and improve hospital preparedness during subsequent COVID-19 waves.

About the study

In the present study, researchers leveraged the national COVID-19 Research Database (C19RDB) health insurance/billing claims for estimating the trends in incidences of disease-specific hospitalizations between 1 January 2017 and 30 June 2021 through multiple COVID-19 waves.

The billing database included data on healthcare settings (outpatient, inpatient), primary and secondary diagnostic codes, service dates, billing sites procedural terms codes, sex, and age. However, only primary diagnostic codes of inpatient records were used for the analysis.

Primary diagnostic codes of the patients were grouped by International Classification of Diseases-10 Clinical Modifications (ICD-10-CM, 2016 edition), subchapter, and period (month, year). Cases were excluded if their ICD-10 codes were U07. 1 or J12.82 [severe acute respiratory syndrome (SARS)-associated coronavirus pneumonia]B97.2 (coronavirus as the causative agent classified elsewhere), or M35.81 (COVID-19-associated multisystem inflammatory syndrome). Three chapters on external morbidity, codes for approaching medical services, and special purposed codes; and 47 subchapters with <5000 diagnoses between 2017 to 2019 were excluded from the analysis. As a result, 19 chapters and 189 subchapters were considered and four clusters A, B, C, and D of diagnoses were formed.

Monthly incidence rate ratios (IRRs) of hospitalizations were determined and fitted to the ICD-10 diagnostic codes per chapter/subchapter in a Poisson regression model to evaluate the incidence trends in pre-pandemic times (1 January 2017 to 31 December 2019) and during the pandemic.

The 2020 and 2021 offsets were predicted by fitting seasonal and linear trends and the monthly count of individuals before March 2020 into a Poisson regression model, which was extrapolated to the period between March 2020 and June 2021. Estimation intervals for IRRs were calculated by Monte Carlo resampling-based simulations. Diagnoses with shared temporal IRR trends were grouped using Ward’s minimum variance-based hierarchical clustering algorithms.


In the period from January 2017 to December 2019, the incidence of hospitalizations per month was about over 3,00,000. In 2020, the incidence counts were 263,000 in March, further reduced to about 150,000 in April 2020, and rebounded slightly in June to over 2,12,000.

All chapters showed increased hospitalizations in January 2020 (IRR between 0.96 and 1.2) and February 2020, which decreased in March of the same year. Significant decreases in incidences of eye and ear infections were noted in April 2020, while minor changes were observed for mental health, prenatal, natal, and postnatal chapters. Pulmonary hospitalizations were lower than predicted post-April 2020, with the most remarkable decrease in January 2021 (IRR 0.3).

Cluster A comprised: pneumonia & influenza (P&I) (J09-J18), infections of the middle ear (H65-H75), intestinal communicable disorders (A00-A09), conjunctivitis (H10-H11), viral dermatological and mucous membrane infections ( B00-B09), biomechanical injury (M99), acute lower respiratory infections (LRI), and acute upper respiratory infections (URI). The subchapters showed a steep decline in hospitalizations in April 2020 that which remained below expected rates between March 2020 and June 2021 (IRR 0.5).

The P&I incidences during January and February 2021 were 89%, lower than expected with only 9,427 diagnoses between October 2020 and March 2021 in comparison to about 61,000 in the corresponding months in 2019. The subchapters demonstrated the greatest decrease during April 2020 and January 2021. The lowest IRRs (0.2) for biomechanical injuries were reported during November 2020. Notably, only the incidences of intestinal infectious disorders and acute respiratory infections attained pre-pandemic IRRs (~1.0) during May-June 2021.

Cluster B comprised: intellectual disorders (F70-F79), sexually transmitted infections (STIs) (A50-A64), psychiatric disorders (F50-F59), coagulation dysfunction (D65-D69), and pregnancy-associated codes [delivery encounters (O80-O82), abortions (O00-O08), and infantile gestational disorders (P05-P08)]. In contrast to cluster A, cluster B had negligible decreases in hospitalization incidences between March and May 2020 with IRRs below expected levels. However, the IRRs rebounded and surpassed (IRR 1.2) pre-pandemic levels from March 2020 to June 2021.

Cluster C comprised: appendix disorders (K35-K38), hypertension (I10-I16), malnourishment (E40-E46), neoplasms, and diabetes (E08-E13). The hospitalizations reduced during March, April, May of 2020, and January of 2021, but rebounded to near pre-pandemic levels subsequently (IRR 0.95) between March 2020 and June 2021. The subchapters demonstrated IRRs <1 during 2020 that gradually increased in 2021. Lens disorders showed the greatest fall in incidence (IRR 0.1) in April 2020 but rapidly attained pre-pandemic levels before July 2020.

Cluster D comprised: cardiovascular disorders (I26-I28), injuries (shoulder, thorax, neck, hip, knee, hand-foot, head) (S00-S99), ischemic heart disorders (I20- I25), chronic LRI (J40- J47), skin neoplasms (C43-C44) and chronic rheumatic fever (I05-I09). The hospitalizations decreased between March and May 2020, which remained below pre-pandemic rates from March to June 2021 (IRR 0.8).


The study findings showed a decrease in non-COVID-19-related primary hospitalization incidences in March 2020. The hospitalization IRRs for neoplasms, diabetes, and reproductive hypertension regained pre-pandemic rates in late 2020 and early 2021. However, for diagnoses such as pulmonary infections, IRRs did not rebound. The estimates offer a novel insight into the non-COVID-19 infection trends during the COVID-19 pandemic and could aid in interpreting reported trends in hospitalization and surveillance data.

*Important notice

medRxiv publishes preliminary scientific reports that are not peer-reviewed and, therefore should not be regarded as conclusive, guide clinical practice/health-related behavior, or treated as established information.


Leave a Reply

Your email address will not be published.