Original article
Healthcare-Associated Infections and the Use of Antibiotics in German Hospitals
Results of the point prevalence survey of 2022 and comparison with earlier findings
; ; ; ; ; ; ; ;
Background: A national point prevalence survey (PPS) of healthcare-associated infections (HAI) and antibiotic use (AU) was carried out in Germany in 2022 in the framework of the European PPS conducted by the European Centre for Disease Prevention and Control (ECDC). The objective was to determine the prevalence of HAI and AU in German hospitals and to compare the obtained values with those of the most recent previous PPS, which was carried out in 2016.
Methods: The German National Reference Center for the Surveillance of Nosocomial Infections was entrusted with the organization of the PPS of 2022. As recommended by the ECDC, each hospital in a representative sample of 50 hospitals was invited to participate, and all other interested hospitals in Germany were also able to participate if desired. The data were collected by specially trained hospital staff in May, June, and July 2022. The definitions and methods put forth by the ECDC were used.
Results: Data from 66 586 patients in 252 hospitals were included. The prevalence of HAI in all participating hospitals was 4.9%, and that of AU was 26.9%. The HAI and AU prevalences were essentially unchanged in comparison to 2016. The most common types of HAI were surgical site infection (23.5%), lower respiratory tract infection (21.6%), and urinary tract infection (19.0%).
Conclusion: HAI were just as frequent in 2022 as in 2016, affecting approximately one in twenty hospitalized patients on any given day.
Healthcare-associated infections (HAI) are linked to poorer treatment outcomes and increased morbidity and mortality (1, 2). Point prevalence surveys (PPS) are an effective, low-effort means of obtaining an overview of HAI frequency. In addition to their benefits for individual hospitals, PPS are particularly valuable at the national and international levels for estimating the current status and resulting disease burden (3, 4).
A national PPS on HAI and antibiotic use (AU) was first carried out in Germany in 1994 (5). In 2011, Germany followed the call of the European Center for Disease Prevention and Control (ECDC) and participated in the 2011/2012 European PPS on HAI and AU (6). The ECDC’s PPS was repeated in 2016/2017 (7). Thus, German participation in 2016 represented the third national PPS on HAI and AU (8). Once again on the initiative of the ECDC, this time for the 2022/2023 European PPS, a PPS on HAI and AU was conducted in Germany in 2022. As with the previous point prevalence surveys, the 2022 PPS was organized by the German National Reference Center for Surveillance of Nosocomial Infections (Nationales Referenzzentrum für Surveillance von nosokomialen Infektionen, NRZ).
The objective of the PPS was to determine the frequency of HAI and AU, expressed as the prevalence of patients with HAI and AU in German hospitals, as well as the most common HAI and antibiotics. Additional aims included determining the current status with regard to infection prevention and control structure and process parameters and, by repeating the survey, analyzing developments over time.
Methods
The ECDC has published a standardized protocol for participating countries (9). In addition to methodological specifications, including the definitions of the HAI (pages 51–75 of the abovementioned protocol), the PPS protocol contained the survey forms with the variables to be documented. The PPS protocol was translated into German by the NRZ and made available to download for free from the NRZ website (www.nrz-hygiene.de/das-nrz) (10).
The ECDC asked participating countries to define a representative sample of hospitals (referred to as the ECDC sample). For Germany, the number of hospitals was set at 50 following consultation between the ECDC and the NRZ. The ECDC sample was determined by the NRZ using the 2019 National Hospital Register (11). Detailed information on the determination of the ECDC sample can be found in the eSupplement of this article. Interested hospitals were trained in the methodology and definitions of the PPS by the NRZ in a total of five free online training courses in March and April 2022. Participation in the PPS was voluntary in all cases, free of charge for the hospitals, and not remunerated by the NRZ.
For data collection, the participating countries were able to choose between two approaches: the patient-based method (full protocol) and the ward-based method (light protocol). Under the ward-based method, in contrast to the patient-based method, demographic data and risk factors did not need to be documented for all patients, but only for those with active HAI or AU. For the German 2022 PPS, the NRZ chose the ward-based method, as in the 2011 and 2016 PPS, in order to keep documentation requirements as low as possible for participating hospitals. One ward of a hospital needed to be completely recorded on 1 day. Only wards with in-patient beds were included in the survey. All patients on a ward who were present on the ward at 8 a.m. on the day of the survey as well as at the time of the survey were included and evaluated for the presence of HAI or antibiotic prescriptions based on a review of patient records. Findings that were pending at the time of the survey were not retrospectively investigated. The recommendation to hospitals was to include all in-patient wards and to record the data in a stepwise approach within a maximum of 3 weeks.
The data were collected by specially trained hospital staff in May, June, and July 2022. Participating hospitals submitted their data to the NRZ via an online portal (PPS portal, https://webkess.charite.de/PPS-II). Once data collection was completed, the NRZ validated the data set and gave participants the opportunity to correct their data. The data set as of 19 September 2022 was defined as the definitive data set. Hospitals for which methodological errors that could not be corrected were identified during data validation or which included less than 50% of all hospital beds in the survey were excluded from all analyses. Data collection was carried out in accordance with the legal obligation for surveillance set out in the German Infection Protection Act (Infektionsschutzgesetz) (12). Therefore, ethics committee approval and informed consent were not required.
Three primary endpoints were considered in the data analysis:
- The prevalence of patients with an HAI, irrespective of whether they acquired the HAI in the hospital in which they were observed or in another hospital (HAI prevalence)
- The prevalence of patients with an HAI who had acquired the infection in the hospital in which they were observed (in-house HAI prevalence)
- The prevalence of patients who received at least one antibiotic on the day of the survey (AU prevalence).
The data of the 2022 PPS were descriptively analyzed for the group of all participating hospitals as well as for the ECDC sample. In addition, a comparison of the 2022 PPS data with the 2016 PPS data was carried out for the two abovementioned groups as well as for the core group, i.e., the hospitals that participated in both the 2022 PPS and the 2016 PPS. Infections acquired in long-term care facilities, which also needed to be recorded in 2022 according to ECDC protocol specifications, were not considered and were excluded from all analyses. For HAI and AU prevalences, 95% confidence intervals were calculated using the statistical software R (13, 14). Differences in prevalence were analyzed using the Chi-square test. For structural and process parameters, the Wilcoxon test was used. The calculated p-values are descriptive in nature.
Results
A total of 262 hospitals took part in the 2022 PPS. The datasets from 10 hospitals needed to be excluded due to errors in data collection. Thus, the final dataset of the 2022 PPS included 252 hospitals with 66,586 observed patients. In total, 29 (11.5%) of the hospitals stated that they were tertiary care hospitals, seven of which were university hospitals. The ECDC sample comprised data from 50 hospitals and 8857 patients. The core group comprised 93 hospitals and 31,132 patients.
Table 1 shows selected structural and process parameters for the group of all participating hospitals, the ECDC sample, and the core group. The median number of beds was considerably lower in the ECDC sample compared to the other two groups. The median length of hospital stay declined in all groups over time. The use of alcohol-based hand sanitizer, measured in milliliters per patient day, and the number of blood cultures per 1000 patient days markedly increased in all groups. The number of stool tests for Clostridioides difficile infections decreased over time in the all-hospitals group and in the core group. The number of infection control personnel (infection control nurses and physicians) improved in all groups over time.
Table 2 shows the three primary endpoints of the survey (HAI prevalence, in-house HAI prevalence, AU prevalence). In the group of all hospitals, HAI prevalence was 4.9% and thus higher than in the 2016 PPS 2016 (4.6%). In the ECDC sample, HAI prevalence was 3.9% and thus also higher than in the 2016 PPS (3.6%). The AU prevalence in the group of all hospitals as well as in the ECDC sample was higher compared to that in the 2016 PPS. It was 26.9% in the all-hospitals groups. In the core group, HAI prevalence and AU prevalence for the 2022 PPS and the 2016 PPS were virtually identical. The results presented below always refer to the group of all participating hospitals.
As part of the 2022 PPS, a total of 3467 HAI were documented in 3292 patients. Surgical site infections (n = 816), lower respiratory tract infections (n = 749), and urinary tract infections (n = 659), each accounting for approximately 20% of all HAI, were the most common types of HAI, as was already the case in 2016. These were followed by primary bloodstream infections (n = 232) with a share of 6.7%. Naturally, SARS-CoV-2 infections (n = 167) represented a new type of infection and accounted for 4.8%. Table 3 shows the five most common HAI in the 2022 PPS compared to the 2016 PPS. Fewer Clostridioides difficile infections were documented in the 2022 PPS compared to 2016 (prevalence of 0.2% vs. 0.5%, share of 3.9% vs. 10.0%; p < 0.001), meaning that in contrast to 2016, Clostridioides difficile infections were not among the five most common HAI.
In 57.2% (1982 of 3467) of HAI, at least one causal pathogen was detected, approximately corresponding to the share in the 2016 PPS (58.5%). A total of 2564 pathogens were documented. The most common pathogens included Escherichia coli (n = 377, with a 14.7% share), Staphylococcus aureus (n = 342, 13.3%), and Enterococcus faecalis (n = 170, 6.6%).
In the 2022 PPS, 22 422 antibiotic prescriptions were recorded in 17 883 patients. The commonest antibiotic groups included penicillins plus beta-lactamase inhibitors (n = 7452, share of 33.2%), third-generation cephalosporins (n = 2140, 9.5%), and second-generation cephalosporins (n = 2036, 9.1%). Table 4 lists the 10 most common antibiotic groups in the 2022 PPS compared to the 2016 PPS. With a share of 23.2%, penicillins plus beta-lactamase inhibitors had already been the most frequently used antibiotic group back in 2016. The AU of this group has nevertheless risen further compared to 2016. The use of fluoroquinolones declined to a third (1.3 vs. 3.9 prescriptions per 100 patients).
Most AU was to treat community-acquired (n = 12 476, share of 55.6%) and hospital-acquired (n = 4367, 19.5%) infections. Compared to 2016, more instances of therapeutic AU were documented in the 2022 PPS. The prevalence of single-dose antibiotic surgical prophylaxis was higher in 2022 than in 2016. Correspondingly, the prevalence of antibiotic surgical prophylaxis over several days was lower. Further details on the indications for AU are provided in Table 5.
Discussion
As in previous PPS, German hospitals showed a high level of willingness to participate in prevalence surveys. Compared to the 2016 PPS, the number of participants even increased by 34 hospitals (252 vs. 218), which is particularly remarkable given the major challenges still posed by the COVID-19 pandemic in spring 2022. Although the median number of beds was very similar across all hospitals in the 2022 and the 2016 PPS (300 vs. 305), only slightly more patients were included in 2022 (66 586 vs. 64 412). This suggests lower occupancy of hospital beds at the time of the survey, which may have been directly or indirectly (staff shortages) related to the pandemic.
HAI prevalence was around 5% and 4%, respectively, in the ECDC sample. This means that on any given day, approximately one in every 20 hospitalized patients in Germany has an active HAI. For AU, the figure is around one in every four patients. The prevalence rates of HAI and AU were higher in the group of all hospitals in 2022 compared to 2016. However, considering the scale of the prevalence rates and the results in the core group, the overall picture tended more towards one of stability with regard to the primary survey endpoints. If one additionally looks at the results of the 2011 PPS, one sees that the prevalence rates of HAI and AU have remained remarkably stable over a period of more than 10 years. For example, the HAI prevalence in 2011 was around 5% and the AU prevalence around 26%, which corresponds to the level in the 2022 PPS (6, 15). The question of whether this represents a success or a failure cannot be answered unequivocally, particularly for HAI, on the basis of the available data. The lack of decline despite improved numbers of infection control staff over time and the increased use of alcohol-based handrub could be interpreted as negative. The infection prevention and control measures that were intensified over the course of the pandemic were aimed in particular at preventing the transmission of viral pathogens for respiratory infections. Therefore, for the majority of HAI documented in the PPS and which are usually caused by endogenous bacterial pathogens, we did not expect any effect. On the other hand, one can speculate that the lower bed occupancy rate described above was attributable to the faster discharge of uncomplicated cases as part of the pandemic-related measures and that, as a result, the patients that remained had more severe underlying diseases on average. If one works on this assumption, the stable values over time would be seen in more of a positive light.
The distribution of the most common HAI has changed over time. Primary bloodstream infections occurred more frequently in 2022 than in 2016. However, in addition to an actual increase in bloodstream infections, this rise may also reflect an improvement in diagnosis, as evidenced by a significant increase in the frequency with which blood cultures are taken. Clostridioides difficile infections significantly declined from 2016 to 2022. This can possibly be attributed to successes in antibiotic stewardship (16), the reduction in fluoroquinolone use (17), improved hygiene measures (18), or the decline in virulent strains (19). The frequency of stool tests for Clostridioides difficile also fell between 2016 and 2022, but only slightly, meaning that this does not explain the observed sharp decline in the prevalence of Clostridioides difficile infections.
With regard to AU, it should be noted that over 50% of antibiotic prescriptions were accounted for by penicillins plus beta-lactamase inhibitors, third-generation cephalosporins, carbapenems, and fluoroquinolones, and thus, extended-spectrum antibiotics. The high proportion of extended-spectrum antibiotics was already observed in the 2016 PPS and tends to be above the European average (20, 21). Together with the continuing high proportion of prolonged surgical prophylaxis, these results underline the need to further strengthen antimicrobial stewardship programs in German hospitals.
The study has a number of relevant limitations. The design of the PPS causes a bias towards longer hospital stays and thus towards HAI and AU, which are associated with longer hospital stays. One must assume heterogeneity between the local PPS detection teams in terms of the sensitivity and specificity of HAI identification (22). This issue was addressed by making it mandatory for hospitals to participate in an introductory event. Since the ward-based version of the PPS protocol was implemented in Germany, it is not possible to make any statements about the characteristics of patients without HAI and AU, which hampers the analysis of risk factors. Moreover, it must be borne in mind that the ECDC sample was determined based solely on the criterion of number of beds. Other factors, such as focus of care, were not taken into account.
The European data from the 2022/2023 ECDC PPS are not yet available. If one takes the most recent previous European PPS conducted in 2016/2017 as a reference—in which an HAI prevalence of 6.5% and an AU prevalence of 30.5% was calculated for all participating countries (7, 21)—Germany is among the European countries with low prevalence rates. However, any international comparison should be interpreted with caution, since there are considerable differences between the respective national healthcare systems and the number of inpatient treatment cases per 100,000 inhabitants, which in turn have a significant impact on national HAI disease burden.
In summary, it is evident from the data obtained in the 2022 PPS and the comparison with previous surveys that the frequency of HAI has remained largely unchanged over time. This illustrates an ongoing need to record HAI using additional methods (for example, as part of the German hospital infection surveillance system [KISS]) in order to identify or investigate individual risk factors and prevention approaches.
Conflict of interest statement
This project was financially supported by the German Federal Ministry of Health (Bundesministerium für Gesundheit) (Grant No. ZMI5–2522PAT004). The grant was awarded to the Institute of Hygiene and Environmental Medicine (Institut für Hygiene und Umweltmedizin), Charité Berlin, and not to an individual person. The authors declare that no conflict of interests exists.
Manuscript received on 25 October 2023, revised version accepted on 8 February 2024.
Translated from the original German by Christine Rye.
Corresponding author
PD Dr. med. Seven Johannes Sam Aghdassi
Charité – Universitätsmedizin Berlin, corporate member of Freie Universität Berlin and Humboldt-Universität zu Berlin
Institut für Hygiene und Umweltmedizin
Hindenburgdamm 27, 12203 Berlin, Germany
seven-johannes-sam.aghdassi@charite.de
Cite this as:
Aghdassi SJS, Hansen S, Peña Diaz LA, Gropmann A, Saydan S, Geffers C, Gastmeier P, Piening B, Behnke M: Healthcare-associated infections and the use of antibiotics in German hospitals—results of the point prevalence survey of 2022 and comparison with earlier findings. Dtsch Arztebl Int 2024; 121: 277–83. DOI: 10.3238/arztebl.m2024.0033
Charité – Universitätsmedizin Berlin, corporate member of Freie Universität Berlin and Humboldt-Universität zu Berlin, Institute of Hygiene and Environmental Medicine, Berlin, Germany: PD Dr. med. Seven Johannes Sam Aghdassi; PD Dr. med. Sonja Hansen; Luis Alberto Peña Diaz, M.Sc.; Alexander Gropmann; Selin Saydan, M.Sc.; Prof. Dr. med. Christine Geffers; Prof. Dr. med. Petra Gastmeier; Dr. med. Brar Piening; Dr. rer. medic. Michael Behnke
German National Reference Centre for Surveillance of Nosocomial Infections , Berlin, Germany: PD Dr. med. Seven Johannes Sam Aghdassi; PD Dr. med. Sonja Hansen; Luis Alberto Peña Diaz, M.Sc.; Alexander Gropmann; Selin Saydan, M.Sc.; Prof. Dr. med. Christine Geffers; Prof. Dr. med. Petra Gastmeier; Dr. med. Brar Piening; Dr. rer. medic. Michael Behnke
Berlin Institute of Health at Charité – Universitätsmedizin Berlin, BIH Biomedical Innovation Academy, BIH Charité Digital Clinician Scientist Program, Berlin, Germany: PD Dr. med. Seven Johannes Sam Aghdassi
| 1. | Rosenthal VD, Yin R, Lu Y, et al.: The impact of healthcare-associated infections on mortality in ICU: a prospective study in Asia, Africa, Eastern Europe, Latin America, and the Middle East. Am J Infect Control 2023; 51: 675–82 CrossRef MEDLINE |
| 2. | Cai Y, Lo JJ, Venkatachalam I, et al.: The impact of healthcare associated infections on mortality and length of stay in Singapore—a time-varying analysis. Infect Control Hosp Epidemiol 2020; 41: 1315–20 CrossRef MEDLINE |
| 3. | Cassini A, Plachouras D, Eckmanns T, et al.: Burden of six healthcare-associated infections on European population health: estimating incidence-based disability-adjusted life years through a population prevalence-based modelling study. PLoS Med 2016; 13: e1002150 CrossRef MEDLINE |
| 4. | Zacher B, Haller S, Willrich N, et al.: Application of a new methodology and R package reveals a high burden of healthcare-associated infections (HAI) in Germany compared to the average in the European Union/European Economic Area, 2011 to 2012. Euro Surveill 2019; 24: 1900135 CrossRef MEDLINE PubMed Central CrossRef MEDLINE |
| 5. | Ruden H, Gastmeier P, Daschner F, Schumacher M: Nosokomiale Infektionen in Deutschland: Epidemiologie in den alten und neuen Bundesländern. Dtsch Med Wochenschr 1996; 121: 1281–7. |
| 6. | Behnke M, Hansen S, Leistner R, et al.: Nosokomiale Infektionen und Antibiotika-Anwendung. Zweite nationale Prävalenzstudie in Deutschland. Dtsch Arztebl Int 2013; 110: 627–33. |
| 7. | Suetens C, Latour K, Kärki T, et al.: Prevalence of healthcare-associated infections, estimated incidence and composite antimicrobial resistance index in acute care hospitals and long-term care facilities: results from two European point prevalence surveys, 2016 to 2017. Euro Surveill 2018; 23: 1800516 CrossRef MEDLINE PubMed Central |
| 8. | Behnke M, Aghdassi SJ, Hansen S, Peña Diaz LA, Gastmeier P, Piening B: The prevalence of nosocomial infection and antibiotic use in German hospitals. Dtsch Arztebl Int 2017; 114: 851–7 VOLLTEXT |
| 9. | European Centre for Disease Prevention and Control: Point prevalence survey of healthcare-associated infections and antimicrobial use in European acute care hospitals—protocol version 6.1. www.ecdc.europa.eu/sites/default/files/documents/antimicrobial-use-healthcare-associated-infections-point-prevalence-survey-version6-1.pdf (last accessed on 24 October 2023). |
| 10. | Nationales Referenzzentrum für Surveillance von nosokomialen Infektionen: Europäische Prävalenzerhebung zum Vorkommen von nosokomialen Infektionen und zur Anwendung von Antibiotika. Protokoll für teilnehmende Krankenhäuser. Deutsche Version 6.2 (basierend auf ECDC Protokoll-Version 6.0 sowie Materialien des „Webinar for information on the methods for the third ECDC PPS“ am 14. März 2022). Stand 08.04.2022. www.nrz-hygiene.de/files/Projekte/PPS%202022/EUPPS2022DE_Protokoll_Version_6.2.pdf (last accessed on 24 October 2023). |
| 11. | Statistische Ämter des Bundes und der Länder: Verzeichnis der Krankenhäuser und Vorsorge- oder Rehabilitationseinrichtungen in Deutschland 2019. https://www.statistischebibliothek.de/mir/receive/DEHeft_mods_00149046 (last accessed on 24 October 2023). |
| 12. | Infektionsschutzgesetz vom 20. Juli 2000 (BGBl. I S. 1045), das zuletzt durch Artikel 2 des Gesetzes vom 17. Juli 2023 (BGBl. 2023 I Nr. 190) geändert worden ist. www.gesetze-im-internet.de/ifsg/BJNR104510000.html (last accessed on 24 October 2023). |
| 13. | R Core Team (2023). R version 4.0.3: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. www.R-project.org/. |
| 14. | Stevenson M, Sergeant E, Nunes T, et al.: epiR: Tools for the Analysis of Epidemiological Data. R package version 2.0.58, 2023. www.CRAN.R-project.org/package=epiR. |
| 15. | Hansen S, Sohr D, Piening B, et al.: Antibiotic usage in German hospitals: results of the second national prevalence study. J Antimicrob Chemother 2013; 68: 2934–9 CrossRef MEDLINE |
| 16. | Aldeyab MA, Kearney MP, Scott MG, et al.: An evaluation of the impact of antibiotic stewardship on reducing the use of high-risk antibiotics and its effect on the incidence of Clostridium difficile infection in hospital settings. J Antimicrob Chemother 2012; 67: 2988–96 CrossRef MEDLINE |
| 17. | Deshpande A, Pant C, Jain A, Fraser TG, Rolston DD: Do fluoroquinolones predispose patients to Clostridium difficile associated disease? A review of the evidence. Curr Med Res Opin 2008; 24: 329–33 CrossRef MEDLINE |
| 18. | Vonberg RP, Kuijper EJ, Wilcox MH, et al.: Infection control measures to limit the spread of Clostridium difficile. Clin Microbiol Infect 2008; 14 (Suppl 5): 2–20 CrossRef MEDLINE |
| 19. | Vehreschild MJGT, Schreiber S, von Muller L, et al.: Trends in the epidemiology of Clostridioides difficile infection in Germany. Infection 2023; 51: 1695–1702 CrossRef MEDLINE PubMed Central |
| 20. | Aghdassi SJS, Gastmeier P, Piening BC, et al.: Antimicrobial usage in German acute care hospitals: results of the third national point prevalence survey and comparison with previous national point prevalence surveys. J Antimicrob Chemother 2018; 73: 1077–83 CrossRef MEDLINE |
| 21. | Plachouras D, Karki T, Hansen S, et al.: Antimicrobial use in European acute care hospitals: results from the second point prevalence survey (PPS) of healthcare-associated infections and antimicrobial use, 2016 to 2017. Euro Surveill 2018; 23: 1800393 CrossRef MEDLINE PubMed Central |
| 22. | Schröder C, Behnke M, Gastmeier P, Schwab F, Geffers C: Case vignettes to evaluate the accuracy of identifying healthcare-associated infections by surveillance persons. J Hosp Infect 2015; 90: 322–6 CrossRef MEDLINE |
| e1. | Statistische Ämter des Bundes und der Länder: Verzeichnis der Krankenhäuser und Vorsorge- oder Rehabilitationseinrichtungen in Deutschland 2019. https://www.statistischebibliothek.de/mir/receive/DEHeft_mods_00149046 (last accessed on 24 October 2023). |
