Piloting of a surveillance system for acute respiratory diseases: COVID-19 monitoring using Sick Leave Certificates

Abstract

Introduction

With the end of the COVID-19 pandemic and the decreasing significance of official reporting figures, the Lower Saxony State Health Office developed and tested a new indicator: the "7-day sick leave incidence". Unlike previous surveillance indicators, it is intended for syndromic surveillance of COVID-19. This article explains the methodological development as well as its benefits, possible applications, and limitations.

Methods

The indicator is based on the weekly number of sick leaves due to COVID-19 per 100,000 health insurance members entitled to sickness benefits (KGbM) of the AOK Lower Saxony (AOKN). The development of the indicator involved differentiating between initial and follow-up sick leaves, investigating fluctuations in the number of KGbM, analysing the doctors’ assignments of ICD Codes U07.1! and U07.2!, and ensuring the timely availability of sick leave data.

Results

Initial and follow-up sick leaves were distinguished using a temporal algorithm. In 2022 and 2023, on average, 83.0% (s=5.4%) and 88.9% (s=2.3%) of all initial COVID-19-related sick notes were submitted on time by the end of the respective calendar week. Four out of 5 initial sick notes contained the doctors’ ICD code U07.1! (lab-confirmed COVID-19). The number of KGbM proved to be stable (M=1.218.202, s=11.003). When comparing the new “7-day sick leave incidence” with the officially used “7-day incidence rates” during pandemic, trends were highly similar in 2022 (r=0.89), but diverged significantly in 2023 (r=0.26) due to declining diagnostic activities for the “7-day incidence rates”.

Conclusion

The new 7-day-sick-leave incidence is a good representation of the post-pandemic COVID-19 infection dynamics. The indicator uses routine data and is easy to establish. Limitations relate to possible changes in diagnostic procedures, doctors’ coding behaviors and changing demands for sick leave.

Keywords

Covid-19 - Epidemiology - Syndromic surveillance - Secondary data analysis - Public reporting of healthcare data

Background

With the transition of the COVID-19 pandemic into an endemic phase, control measures were relaxed across Germany. Similarly, isolation and testing requirements were gradually lifted in Lower Saxony (see info box): Targeted and mandatory testing was discontinued, testing centers closed, and once easily accessible antigen tests expired or were discarded. This had significant implications for the reported infection numbers.

Info box: Legal Frameworks Impacting the 7-Day Reporting Incidence [21] [22].

01 February 2020	The regulation mandating the reporting of suspected cases, confirmed cases, and deaths related to COVID-19 was enacted (integrated into IfSG on 23 May 2020).
01 March 2020	The RKI and other authorities began publishing the 7-day incidence for PCR-confirmed COVID-19 cases reported under the IfSG.
14 May 2020	The nationwide Coronavirus Testing Regulation granted citizens in risk areas free point-of-care antigen rapid tests.
08 March 2021	Antigen rapid tests were incorporated into the national testing strategy, and mandatory confirmatory PCR testing after a positive rapid test was gradually phased out (varied by federal state).
01 July 2022	A citizen co-payment of 3 euros was introduced for point-of-care antigen rapid tests. Selected groups continued to receive free tests.
25 November 2022	Point-of-care antigen tests for the general population (with co-payment) were discontinued. The groups eligible for free tests were further restricted.
16 January 2023	The nationwide Coronavirus Testing Regulation was amended. Tests for proof of infection-free status and for the termination of isolation were no longer offered free of charge.
31 January 2023	The Lower Saxony Isolation Regulation expired.
01 March 2023	The Lower Saxony Coronavirus Regulation was repealed. Test certificates were no longer required when entering medical facilities.

Until then, the so-called “7-day incidence rate” had been the key indicator for the implementation or cessation of control measures. Between March 2020 and April 2023, this indicator was published by the Robert Koch-Institute (RKI) and other authorities, based on SARS-CoV-2 cases reported to all local health authorities under the Infection Protection Act (IfSG). For inclusion, cases had to meet the reference definition established by the RKI, meaning they had to be confirmed in the laboratory either through nucleic acid amplification (e. g., polymerase chain reaction [PCR]) or pathogen isolation [1]. Depending on the phase of the pandemic, it is estimated that this indicator reflected 20% to 75% of all infections [2]. Since the indicator only accounts for infections reported under the IfSG, it will be referred to hereafter as the “7-day reporting incidence.” With fewer legally mandated testing occasions, the number of PCR tests performed decreased, leading to fewer COVID-19 cases reported to public health authorities in 2023. The change in the number of reported cases necessitated novel surveillance methods better suited to the transition into the endemic phase.

The adaptation of surveillance methods was in line with the recommendations of the European Centre for Disease Prevention and Control (ECDC), which, following discussions with eight European countries, emphasized the importance of managing severe COVID-19 cases and protecting vulnerable populations [3]. Additionally, the ECDC called for the establishment of sustainable surveillance systems to monitor infections in the general population, incorporating syndromic disease burden and integrating with broader acute respiratory infections (ARI) surveillance efforts [4].

The Lower Saxony State Health Office (NLGA) took the ECDC's recommendations as an opportunity to develop a new, innovative surveillance system. The goal was to reliably and promptly reflect the syndromic infection dynamics in both space and time [5]. The Pandemic Check in Lower Saxony (PanCHECK-iN) study was launched on 30 September 2022 in cooperation with a statutory health insurance fund with a large membership base (“Die Gesundheitskasse für Niedersachsen” [AOKN]). The new system uses sick notes (SNs) that are promptly available to the health insurer and can then be used as secondary data for PanCHECK-iN. Additionally, we examined whether SN data could be utilized not only for COVID-19 but also for the general surveillance of ARIs.

This paper aims to provide readers with a framework for establishing a surveillance system based on routine SN data. To this end, the methodological development steps, along with an assessment of feasibility, added value, as well as opportunities and limitations, are presented. The examinations conducted to establish the data basis and the resulting outcomes are presented in detail. This includes:

• the inclusion and exclusion criteria for SNs routinely submitted to the AOKN,

• the selection of an indicator to define an acute COVID-19 illness, and

• the sickness benefit-eligible members (SBMs) as a reference for the 7-day SN incidence.

Based on this, the SN surveillance indicator was compared with the 7-day reporting incidence in a time series.

Methods

The data protection concept of the NLGA-AOKN cooperation project PanCHECK-iN was reviewed by the Ministry of Social Affairs, Labour, Health, and Equality on 30 September 2022, based on § 75 SGB X (Transmission of social data for research and planning purposes). Due to the high level of data aggregation, no concerns were raised regarding the analysis of the SN data, and weekly data transmission was approved. The SN data did not allow re-identification or tracing back to an individual, so consent from the insured persons or ethics approval were not required.

Data Delivery and Establishment of a Data Basis

Each month, the AOKN transmits the number of SBMs insured as of a specific reporting date. SBMs refers to all individuals insured with the AOKN who are entitled to sickness benefits, meaning they would receive sickness benefits from the AOKN in the case of prolonged illness. This group primarily includes employees and recipients of unemployment benefits. Furthermore, self-employed individuals can opt for membership with sickness benefit entitlement. Since the number of SBMs per month can only be provided retrospectively, analyses were always conducted with SBMs data from the 1^st of the preceding month.

Every week, the AOKN collects all SNs received by Friday and processes them by the following Monday. On Tuesday morning, the anonymised and aggregated data are delivered to the NLGA in an Excel file. The AOKN uses specific information from the medical SNs of its SBMs and processes them as follows:

• The registered addresses of AOKN SBMs on sick leave are aggregated at the district or independent city level. Only individuals officially registered in Lower Saxony are considered.

• The exact dates of the initial SNs are aggregated into the respective weeks.

• The ICD-10 codes (10^th version of the International Statistical Classification of Diseases and Related Health Problems [6]) documented on the SNs were categorised into four diagnostic groups. The first diagnostic group comprises ARIs and was assembled based on the SEED^ARE project of the RKI’s Influenza Working Group [7]. The ARI group encompasses the codes J00–J06 (acute infections of the upper respiratory tract), J09–J11 (influenza), J12–J18 (pneumonia), and J20–J22+J44.0 (other acute infections of the lower respiratory tract) [8]. The second diagnostic group contains SNs with the ICD-10 codes U07.1! (COVID-19, virus detected) or U07.2! (COVID-19, virus not detected), and the third group includes only those with U07.1!. The fourth group includes all SNs, irrespective of ICD-10 code.

The ICD-10 codes U07.1! and U07.2! were officially introduced as supplementary codes in 2020. They are used to document acute COVID-19 cases and must be recorded alongside at least one other ICD-10 code related to the illness (e. g. J06.9, acute upper respiratory infection, unspecified).

The weekly data transfer from the AOKN adheres to data protection regulations and includes an Excel file with four data sheets, each corresponding to a diagnostic group (as described above). These sheets display the number of initial SNs (no follow-up SNs) received by AOKN across the 45 districts and independent cities of Lower Saxony, starting from week 29/2022.

Study of Routine Data for Building the Analytical Dataset

To ensure a robust data analysis of the weekly incoming COVID-19 SNs, various aspects were considered, following the standards outlined in “Best Practices for Secondary Data Analysis” [9]. This included the handling of initial and follow-up SNs, late transmissions of SNs, fluctuations in the number of SBMs (i. e., the denominator), and the selection of the target variables U07.1! and U07.2!. The investigations were conducted in close collaboration with the data holder, AOKN, and are described as follows.

Managing Initial and Follow-up SNs

When transmitting a SN to a statutory health insurance fund, the doctor must indicate whether it is an initial or a follow-up SN. Additionally, the date of the onset of sick leave, the date of the assessment, and the expected end date of the sick leave must be provided. For the analysis, only new cases of ARIs or COVID-19 were considered. The AOKN’s data routines do not allow for the direct distinction between initial and follow-up SNs. To select only initial SNs for the analysis, the fields for “doctor’s assessment date” and “start date of sickness leave” were cross-referenced. For initial SNs, the assessment date and the start date of sickness leave are typically the same day or only slightly offset. For follow-up SNs, doctors update the date of assessment but not the start date of the sick leave. SNs are rarely backdated, and then typically for no more than 3 days. [10]. A court ruling has actually limited retroactive SNs to a maximum of just 2 days [11]. Accordingly, an SN was considered initial if the period between the start of the sick leave and the doctor's assessment date was 0 to 2 days. If the period was longer, it was classified as a follow-up.

[Fig. 1] illustrates how the AOKN selects and assigns incoming SNs for PanCHECK-iN, using the example of week 45/2022.

Fig. 1 Data selection process at AOKN for week 45/2022.

Availability of initial SNs: Managing late transmissions

The AOKN provided the NLGA with all SNs from the previous week. Evaluating delays in the transmission of SNs from primary care to AOKN was crucial for the SN surveillance system, as these delays would inhibit their inclusion in the weekly analysis. For exploration purposes, late transmissions within a period of up to 4 weeks were considered. The percentage of initial SNs transmitted immediately, one week late, and up to four weeks late was determined. The calculations were conducted separately for the years 2022 and 2023, as starting January 1, 2023, the electronic transmission of SNs became mandatory. Due to the legal change, paper-based SNs were no longer permitted. For 2022, data was available starting from week 29.

The results section provides the average proportion of late transmissions per year. Additionally, time series of 7-day SN incidences, with and without the late transmissions from one week, are shown ([Fig. 2]).

Fig. 2 Comparison of the 7-day SN incidence with and without late transmissions in 2022 and 2023.

Managing Fluctuations in the Number of SBMs

To assess whether a constant number of SBMs insured with the AOKN could be assumed, the average number of members over 16 months, including the standard deviation, was calculated.

Managing ICD-10 Codes on SNs

For the development of a COVID-19 surveillance, the number of SNs with the ICD-10 codes U07.1! and U07.2! was of primary interest. The coding recommendations of the National Association of Statutory Health Insurance Physicians stipulate that U07.1! is for COVID-19 cases for whom SARS-CoV-2 has been confirmed by a laboratory test [12]. According to the Federal Institute for Drugs and Medical Devices, the use of U07.1! requires direct confirmation of the virus [6], which includes antigen rapid tests. U07.2! is used for COVID-19 cases that were not confirmed by a laboratory test but were clinically and also epidemiologically confirmed. For the surveillance system based on SNs, it was necessary to decide whether both laboratory-confirmed and clinically-epidemiologically confirmed COVID-19 cases should be included.

For an initial overview, the 7-day SN incidence rates for U07.1! and for U07.1!+U07.2! were compared. Time series were created for the years 2022 and 2023, with data for 2022 available from week 29. For a more precise assessment, the average proportion of SNs with U07.1! relative to all SNs with U07.1!+U07.2! was calculated for each week. A high proportion would suggest that COVID-19 diagnoses were predominantly confirmed by test results. To assess the extent to which COVID-19 diagnoses were consistently based on test results, the calculations were carried out separately for the years 2022 and 2023.

Validation Against the 7-day Reporting Incidence

Based on the above considerations, the 7-day SN incidence was calculated and compared with the 7-day reporting incidence in a time series chart. For the sake of timely reporting and at the expense of completeness, only the data available for each week were used, excluding any late transmissions from the analysis. Furthermore, the average correlations between the 7-day SN incidence and the 7-day reporting incidence for the years 2022 and 2023 were calculated.

Results

Assessing Fluctuations in the Number of SBMs

The number of SBMs insured with the AOKN remained stable throughout the 16-month observation period. On average, there were 1,218,202 members (M) with a minimal standard deviation of 11,003 (s).

Assessing the Timely Availability of COVID-19 SNs

In 2022, an average of 83.0% (s=5.4%) of all SNs with the ICD-10 codes U07.1! or U07.2! were transmitted to the AOKN during the week of their issuance. In 2023, the average was 88.9% (s=2.3%), allowing a larger proportion of all SNs to be included in the weekly analyses. The weekly fluctuations were minimal, as indicated by the standard deviation. If SN transmissions from doctors to the AOKN, delayed by up to 7 days, were included, almost all SNs would be accounted for in the analysis. Under these conditions, the average proportion would be 97.7% (s=0.8%) in 2022 and 98.7% (s=1.3%) in 2023.

[Fig. 2] compares the 7-day SN incidences with and without late submissions from one week for the years 2022 and 2023 in a time series chart. For both the ICD-10 code group U07.1!+U07.2! and U07.1! alone, accounting for delayed transmissions does not lead to any significant differences in the trend. Furthermore, there is no evidence of a systematic delay in the transmission of SNs with laboratory confirmation for COVID-19 compared to those confirmed solely through clinical-epidemiological assessment.

Assessing ICD-10 Codes U07.1! and U07.2! on SNs

In 2022, 75.7% of all COVID-19-related SNs (i. e., U07.1!+U07.2!) were coded with U07.1! (COVID-19, laboratory-confirmed). This proportion increased to 83.4% in 2023. Across the entire observation period, U07.1! was documented in 81.3% (s=4.3%) of COVID-19-related SNs.

Comparison of the 7-day SN Incidence and the 7-day Reporting Incidence

In a time series analysis for 2022 and 2023, the 7-day reporting incidences (cut-off on Friday) were compared with the AOKN data for the same week ([Fig. 3]). For both incidences, only the currently available data were used, excluding any late transmissions. The 7-day SN incidence was calculated using the number of SNs with U07.1!+U07.2! in the numerator and the number of SBMs from the previous month in the denominator.

Fig. 3 Comparison of the 7-day SN and 7-day reporting incidence in 2022 and 2023.

In 2022, the values and trends of the 7-day SN incidence and 7-day reporting incidence were strongly correlated (r=0.89). However, from the beginning of 2023, the trends diverged significantly (r=0.26). From late summer 2023 (particularly from week 42 onwards), the 7-day SN incidence showed a new wave of infections much more clearly than the 7-day reporting incidence.

Discussion

The examination of AOKN SN routine data has led to the following decisions for the SN surveillance indicator:

• Only SNs that were transmitted to the AOKN from Saturday to Friday of the week in which they were issued were included. Any late transmissions were therefore not considered.

• An acute COVID-19 disease was defined as the ICD-10 diagnostic code U07.1! or U07.2!.

• The incidence was calculated per 100,000 SBMs.

The evaluation of the late transmissions led to the decision to conduct the trend analyses using only the current COVID-19 SN data. Particularly since 1 January 2023, when the electronic transmission of SNs became mandatory, a consistently high rate of timely SN transmissions from medical practices to the AOKN has been observed. The analysis of the ICD-10 codes revealed a high proportion of SNs with laboratory confirmation of a COVID-19 diagnosis (U07.1!) among all COVID-19 SNs (U07.1!+U07.2!). In 2023, this proportion increased further, with 4 out of 5 COVID-19 SNs being based on laboratory confirmation. As previously mentioned, COVID-19 codes may only be used in addition to an ARI code on SN forms. We therefore assumed that doctors would only undertake the extra documentation effort for U07.2! (COVID-19, not laboratory-confirmed) if they had a strong suspicion of a COVID-19 infection. The number of SBMs remained constant throughout the entire observation period. Nevertheless, to ensure better comparability with other surveillance systems, we decided to calculate and report the incidence per 100,000 SBMs.

The 7-day SN incidence and the 7-day reporting incidence were compared starting from July 2022. At that time, the COVID-19 pandemic was still ongoing, and the 7-day reporting incidence indicated an infection wave during the summer. During 2022, the 7-day SN incidence and the 7-day reporting incidence initially showed a strong correlation, with both indicators displaying almost identical trends ([Fig. 3]). However, a larger infection wave in spring 2023 is, at first glance, only noticeable with the 7-day SN incidence [13].

Feasibility, Limitations, and Added Value

Secondary data are rarely used for surveillance purposes. Other ARI surveillance systems that complement the IfSG reporting data typically rely on voluntary participation (e. g., from practices, laboratories, and hospitals). An advantage of the SN surveillance lies in the use of routine data. This eliminates the need for resource-intensive collection of new data. PanCHECK-iN also addresses the call to make scientific use of secondary data in Germany and to derive public health measures from it [14]. The presented SN surveillance is relatively low-maintenance in terms of data provision, transmission, and analysis. After a data protection review and the initial implementation of data processing and plausibility checks, the compilation and transmission of the aggregated data is largely automated.

More than 3 million individuals insured with the AOKN make up approximately 37% of Lower Saxony’s population, forming a substantial sample for analyses [15]. Overall, the AOKN population is representative of the general population of both Germany and Lower Saxony in terms of age and gender distribution. Only minor deviations are observed with regard to professional qualifications [16] [17]. A limitation to consider is that SBMs represent only a subset of AOKN-insured individuals. SBMs make up 41.3% of all AOKN-insured individuals, which corresponds to about 15% of the total population of Lower Saxony. Groups such as children, family-insured individuals, and pensioners, who typically do not experience income loss due to illness, are not covered by the SN surveillance. Based on prior experiences with the COVID-19 pandemic, it cannot be assumed that the spread of infections is confined to specific age groups. This assumption is supported by the time series analyses presented here. These analyses show that, up until the end of 2022, the 7-day SN incidence followed a trend and trajectory similar to the 7-day reporting incidence, which includes all age groups and the general population ([Fig. 3]). It can thus be assumed that the 7-day SN incidence for COVID-19 is transferable to the general population of Lower Saxony. The extent to which conclusions can be drawn about the general population for other SARS-CoV-2 variants or ARIs should be examined when developing SN surveillance systems.

The RKI established a Sentinel for the electronic recording of medical diagnosis codes for acute respiratory infections (SEED^ARE) in 2007. Like our SN surveillance, SEED^ARE also utilizes secondary data originally intended for statutory health insurance funds. Participation of primary care practices is voluntary and requires the setup of an electronic interface with the RKI. The RKI provides this interface only for selected electronic medical record systems. Through this interface, the RKI can directly access patient data required for the ARI surveillance, while adhering to strict data protection regulations. Additionally, the incidence of primary care visits due to ARI needs to be estimated indirectly by extrapolating the sentinel sample of physicians to the total number of primary care physicians caring for the total population. On the other hand, SEED^ARE has access to more comprehensive data from primary care, allowing for more detailed analyses of ARI. For instance, stratification by sex and age groups is possible [18]. The sentinel also covers all individuals with statutory health insurance.

It is important to regularly review the influence of external factors on both SN and IfSG data. For example, physician coding behaviour may change due to factors such as reimbursement regulations, access to antigen tests, or the relevance of differentiated ARI diagnoses. When interpreting the SN data, it is important to consider the obligation for the electronic transmission of SNs that began in early 2023, which led to higher data completeness. The option of issuing SNs by phone, which was introduced during the COVID-19 pandemic and has been permanently available since December 2023, must also be considered. SN data are unstable during holiday periods. Closed medical practices, reduced patient consultations, and resulting lower diagnostic activity contribute to this instability. This underreporting is visible in the time series shown for Christmas and Easter. Typically, weeks with one or more public holidays show a significant but purely temporary decline in SNs. This is likely due to the fact that many collective bargaining and employment contracts require an SN only after a few days, with holidays often not being considered and thus not triggering SNs.

Besides capturing current COVID-19 cases, an additional benefit of PanCHECK-iN is the ability to report SN rates due to ARIs in general, regardless of the specific cause. Retrospective analyses of SN data by several health insurance funds have shown that the number of sick days in 2023 due to common colds significantly exceeded pre-pandemic levels. This increase in sick days has led to economic losses [19]. An ARI indicator that reflects such developments on a weekly basis could provide the basis for targeted recommendations, such as working remotely.

Surveillance often involves assessing the interplay of various indicators. In the best case, surveillance systems complement each other and respond differently to the same influencing factors, as demonstrated here with the end out of the COVID-19 testing regulation. In addition to the SN data, the NLGA also holds sentinel data on ARI-related sickness absences in childcare settings. The surveillance is supplemented by the virological examination of throat swabs from patients with an ARI from selected medical practices and hospitals. The weekly ARI reporting from the NLGA is now enhanced by the AU data from AOKN, which provide insights not only into the COVID-19 but also the general ARI infection trends [20].

Conclusion

The 7-day SN incidence has proven to be an easily implementable surveillance indicator for syndromic ARI, particularly for COVID-19. Even in the post-pandemic period, the 7-day SN incidence continues to provide a reliable estimate of infection rates for COVID-19. Moreover, this surveillance system is easy to maintain, as current secondary data is transmitted automatically. The significance of anonymised, aggregated health insurance data has already been demonstrated in other medical fields and is now also apparent for the surveillance of infectious diseases. At the same time, the SN surveillance must be continuously interpreted in light of possible changes in coding and testing practices.

In Lower Saxony, the SN surveillance has proven effective, and the weekly ARI report now includes a specific COVID-19 indicator, as well as an overarching ARI indicator. Together with the existing indicators, this enables a more accurate and reliable assessment of the post-pandemic infection situation.

This article is part of the DNVF Special Issue “Health Care Research and Implementation”

Conflict of Interest

The authors declare that they have no conflict of interest.

• References

• 1 Robert-Koch-Institut (2020) Coronavirus-Krankheit-2019 (COVID-19) (SARS-CoV-2). Available at https://www.rki.de/DE/Content/InfAZ/N/Neuartiges_Coronavirus/Falldefinition.pdf Accessed: 19 June 2024
• 2 Oberste M, Pusch L-M, Roth R. et al. Results of the Cologne Corona surveillance (CoCoS) study – a prospective population-based cohort study: incidence data and potential underestimation of new SARS-CoV-2 adult infections by health authorities. BMC Public Health 2022; 22: 1379
• 3 European Centre for Disease Prevention and Control. Transitioning beyond the acute phase of the COVID-19 pandemic. Approaches and tools used by a sample of EU countries in the transition and de-escalation phase – interim report. 2022 Available at https://www.ecdc.europa.eu/en/publications-data/transitioning-beyond-acute-phase-COVID-19-pandemic-interim-report Accessed: 19 June 2024
• 4 European Centre for Disease Prevention and Control. COVID-19 surveillance guidance. October 2021. Transition from COVID-19 emergency surveillance to routine surveillance of respiratory pathogens. 2021 Available at https://www.ecdc.europa.eu/sites/default/files/documents/COVID-19-surveillance-guidance.pdf Accessed: 29 May 2024
• 5 Dreier M, Kramer S, Stark K. Epidemiologische Methoden zur Gewinnung verlässlicher Daten. In: Schwartz FW, Walter U, Siegrist J et al. (eds) Public Health. Elsevier; 2012: 409-449
• 6 Bundesinstitut für Arzneimittel und Medizinprodukte (2021) ICD-10-GM Kodierempfehlung zu Fallkonstellationen im Zusammenhang mit dem Infektionsgeschehen SARS-CoV-2/COVID-19. Available at https://www.bfarm.de/SharedDocs/Downloads/DE/Kodiersysteme/Kodierfragen/Kodierempfehlung-COVID-19.html Accessed: 15 May 2024
• 7 Goerlitz L, Tolksdorf K, Buchholz U. et al. Überwachung von COVID-19 durch Erweiterung der etablierten Surveillance für Atemwegsinfektionen (Monitoring of COVID-19 by extending existing surveillance for acute respiratory infections). Bundesgesundheitsblatt Gesundheitsforschung Gesundheitsschutz 2021; 64: 395-402
• 8 Zanuzdana A, Köpke K, Haas W. Influenza und ambulant erworbene Pneumonie in hausärztlich tätigen Arztpraxen in Deutschland (Influenza and community-acquired pneumonia in German primary care). Bundesgesundheitsblatt Gesundheitsforschung Gesundheitsschutz 2016; 59: 1492-1502
• 9 Arbeitsgruppe Erhebung und Nutzung von Sekundärdaten (2012) Gute Praxis Sekundärdatenanalyse (GPS). Leitlinien und Empfehlungen. Available at https://www.dgepi.de/assets/Leitlinien-und-Empfehlungen/GPS_revision2-final_august2014.pdf Accessed: 19 June 2024
• 10 Gemeinsamer Bundesausschuss (2024) Richtlinie des Gemeinsamen Bundesausschusses über die Beurteilung der Arbeitsunfähigkeit und die Maßnahmen zur stufenweisen Wiedereingliederung nach §92 Absatz 1 Satz 2 Nummer 7 SGB V. Arbeitsunfähigkeits-Richtlinie. Available at https://www.g-ba.de/downloads/62-492-3374/AU-RL_2023-12-07_iK-2024-02-21.pdf Accessed: 19 June 2024
• 11 Berner B (2013) Rückwirkende Ausstellung einer Arbeitsunfähigkeitsbescheinigung. Rechtsreport. Deutsches Ärzteblatt 110:A152
• 12 Kassenärztliche Bundesvereinigung (2021) Coronavirus: Empfehlungen zum Kodieren mit Kodierbeispielen. Available at https://www.kbv.de/media/sp/PraxisInfo_Coronavirus_Kodieren.pdf Accessed: 15 May 2024
• 13 Junius-Walker U, Overesch I, Wollenweber M. et al. COVID-19 surveillance based on sick leave certification rates. Dtsch Arztebl Int 2023; 120: 765-766
• 14 Schmitt J, Bierbaum T, Geraedts M. et al. Das Gesundheitsdatennutzungsgesetz – Potenzial für eine bessere Forschung und Gesundheitsversorgung. Gesundheitswesen 2023; 85: 215-222
• 15 AOK Niedersachsen (2022) Geschäftsbericht 2022. Available at https://www.aok.de/mk/niedersachsen/geschaeftsbericht22/ Accessed: 15 May 2024
• 16 Epping J, Geyer S, Eberhard S. et al. Völlig unterschiedlich oder doch recht ähnlich? Die soziodemografische Struktur der AOK Niedersachsen im Vergleich zur niedersächsischen und bundesweiten Allgemein- und Erwerbsbevölkerung (Completely different or quite similar? The sociodemographic structure of the AOK Lower Saxony in comparison to the general and working population in Lower Saxony and the Federal Republic of Germany). Gesundheitswesen 2021; 83: S77-S86
• 17 Jaunzeme J, Eberhard S, Geyer S. Wie „repräsentativ“ sind GKV-Daten? Demografische und soziale Unterschiede und Ähnlichkeiten zwischen einer GKV-Versichertenpopulation, der Bevölkerung Niedersachsens sowie der Bundesrepublik am Beispiel der AOK Niedersachsen (How „representative“ are SHI data? Demographic and social differences and similarities using the example of the AOK in Lower Saxony). Bundesgesundheitsblatt Gesundheitsforschung Gesundheitsschutz 2013; 56: 447-454
• 18 Robert-Koch-Institut (2023) SEEDARE – Dokumentation der Schnittstelle zwischen Arztinformationssystem und RKI. Available at https://www.rki.de/DE/Content/Institut/OrgEinheiten/Abt3/FG36/SEED.html Accessed: 15 May 2024
• 19 Tagesschau (2024) Drückte Krankenstand Deutschland in Rezession? Available at https://www.tagesschau.de/wirtschaft/konjunktur/rekord-krankenstand-rezession-100.html Accessed: 15 May 2024
• 20 Niedersächsisches Landesgesundheitsamt (2023) Wochen- und Saisonberichte der ARE-Surveillance. Available at https://www.nlga.niedersachsen.de/are/archiv-203082.html Accessed: 15 May 2024
• 21 Bundesministerium für Gesundheit (2023) Chronik zum Coronavirus SARS-CoV-2. Available at https://www.bundesgesundheitsministerium.de/coronavirus/chronik-coronavirus Accessed: 15 May 2024
• 22 Niedersächsische Staatskanzlei (2023) (Ehemalige) Corona-Vorschriften. Available at https://www.niedersachsen.de/Coronavirus/vorschriften-der-landesregierung-185856.html Accessed: 29 May 2024

Publication History

Received: 16 August 2024

Accepted: 05 December 2024

Accepted Manuscript online:
06 December 2024

Article published online:
02 October 2025