Acute lower respiratory tract infections accounted for 56.2% of hospitalized COVID-19 cases in Germany during the first three waves

The official number of hospitalized COVID-19 cases based on the World Health Organization definition (a person with confirmed detection of SARS-CoV-2) is used in public health reporting and may have major implications on public health measures. But the definition may not adequately describe whether patients were admitted for COVID-19 or if the principal diagnosis was another disease and the patient acquired COVID-19 in the hospital or was tested positive for SARS-CoV-2 during the hospital stay without any respiratory tract symptoms. That is why we analysed the remuneration data of the national health insurances during the first three COVID-19 waves in Germany (January 2020―May 2021) provided by the Institute for the Hospital Remuneration System.1 Any patient admitted and treated in a hospital, irrespective of the duration of stay, was regarded as a hospitalized COVID-19 case when the following criterion was fulfilled: a laboratory-confirmed detection of SARS-CoV-2 using polymerase chain reaction (PCR) or antigen test, irrespective of the type and presence of symptoms. COVID-19 was assumed whenever ICD-10 (International Classification of Diseases, 10th Revision) codes for an acute lower respiratory tract infection were found (J09-J22)2 in combination with detection of SARS-CoV-2 (U07.1).3 A total of 206 179 cases were counted, representing 56.2% of all 366 964 hospitalized COVID-19 cases in the first three waves in Germany (Table 1). The proportion was higher in the subgroup of 55 550 cases with and intensive care unit (ICU) stay, at 63.7%. A total of 1715 cases with an acute lower respiratory tract infection (0.8%) and a confirmed SARS-CoV-2 co-infection had an infection with bacterial species such H. influenzae or other viruses such as respiratory syncytial virus (RSV) as the principial diagnosis.4 An upper respiratory tract infection was described in 6119 COVID-19 cases (1.7%), with only 201 of them being transiently on an ICU; 154 666 persons were treated in hospitals with other principal diagnoses, despite being COVID-19 cases (42.1%). Overview of principal diagnoses of hospitalized COVID-19 cases in Germany between January 2020 and May 2021 (first three waves); source: InEK Datenportal1 ICU, intensive care unit; RSV, respiratory syncytial virus. Overview of principal diagnoses of hospitalized COVID-19 cases in Germany between January 2020 and May 2021 (first three waves); source: InEK Datenportal1 ICU, intensive care unit; RSV, respiratory syncytial virus. The high number of other principal diagnoses may partly be explained by an extrapulmonary and atypical clinical presentation,5 a secondary diagnosis of symptomatic hospital-acquired COVID-19 or by a positive PCR test in the absence of any COVID-19 symptoms. Misclassifications such as errors in reporting may also be an explanation. These data indicate that only a proportion of ‘COVID-19 cases’ in hospitals were primarily treated for COVID-19 presenting as an acute lower respiratory tract infection. The primary diagnosis is typically evaluated at the end of hospitalization and should describe the disease which was primarily responsible for causing the hospitalization. Some uncertainty may remain in cases with two or more relevant diagnoses during admission, so that the final assignment is usually based on the disease that required most resources for diagnosis and treatment. Overall, the official reporting of hospitalized COVID-19 cases seems to overestimate the real number of acute COVID-19 in hospitals during the first three waves. This finding may have implications for public health measures. Ethics approval was not needed because the evaluation involved only anonymous remuneration data from a public source. The data underlying this article are available in InEK DatenBrowser at [https://datenbrowser.inek.org/]. Use of the data is restricted to non-commercial purposes. Both authors had the conceptual idea. K.M. analysed and interpreted the data, G.K. wrote the manuscript and searched the literature, both authors revised and edited the manuscript. Both authors contributed equally. The authors have no conflict of interest with relevance to the content of the article. The views expressed here are those of the authors and do not necessarily reflect those of the university.