Figure. Association (odds ratio [95% confidence interval]) between prevalence of patients with infectious diseases in the hospital ward and staff sickness absence by the level of patient overcrowding. Models are adjusted for employee age, sex, specialty, occupation, chronic disease, type of employment, and patient characteristics at the ward level (mean age, number of patients, mean number of invasive devices in patients, and prevalence of operated patients). CI indicates confidence interval; ID, infectious diseases; and Ref, reference. *Prevalence of sickness absence onset among employees.
Virtanen M, Terho K, Oksanen T, Kurvinen T, Pentti J, Routamaa M, Vartti A, Peltonen R, Vahtera J, Kivimäki M. Patients With Infectious Diseases, Overcrowding, and Health in Hospital Staff. Arch Intern Med. 2011;171(14):1296-1298. doi:10.1001/archinternmed.2011.313
Author Affiliations: Finnish Institute of Occupational Health, Helsinki, Finland (Drs Virtanen, Oksanen, Vahtera, and Kivimäki and Ms Pentti); Department of Hospital Hygiene and Infection Control, Turku University Hospital, Turku, Finland (Mss Terho, Kurvinen, and Routamaa and Dr Peltonen); Department of Strategic and Defence Studies, National Defence University, Helsinki (Ms Vartti); Department of Public Health, University of Turku, Turku (Dr Vahtera); and Department of Behavioral Sciences, University of Helsinki, Helsinki, and Department of Epidemiology and Public Health, University College London Medical School, London, England (Dr Kivimäki).
The evidence on whether treating patients with infectious diseases increases the risk of ill health among hospital staff is limited to specific infectious agents, such as methicillin-resistant Staphylococcus aureus (MRSA).1- 7 However, any infectious agent that can be transmitted by airborne transmission or during treatment contact can be acquired at the workplace. We therefore examined whether the overall prevalence of infectious diseases among patients predicts ill health in hospital staff treating them, as indicated by increased absence from work because of sickness and antibiotic medication use. Furthermore, as patient overcrowding has been suggested to increase the transmission of infectious diseases within hospitals, we also studied whether there is an association between patient overcrowding and prevalence of infectious diseases and whether the association between the prevalence of patients with infectious diseases and ill health in hospital staff is dependent on ward overcrowding.
Study participants comprised 993 physicians and nurses (mean age, 42.4 years; 93.7% female; 84.7% registered nurses) in 54 somatic disease hospital wards in 5 acute care hospitals in Finland. The assessment methods used have been described previously.8,9 Briefly, ward-level prevalence of infectious diseases (hospital and community acquired) and other patient characteristics were assessed from case records of the 1102 patients in these wards.8 Ward overcrowding was determined using routinely collected monthly figures on bed occupancy for each ward.9 These ward-level data were linked to individual records on the employee sickness absence and antibiotic medication use (purchases of medicine with the World Health Organization Anatomical Therapeutic Chemical Classification code J01) during the subsequent 150 days. The records were obtained from employers' and nationwide health registers.9
Binary logistic regression analysis with the SAS multilevel GLIMMIX procedure was used to study the associations of ward-level exposure to infectious diseases with individual-level employee sickness absence (yes/no) and recorded antibiotic use (yes/no). The models were adjusted for employee characteristics (sex, age, occupation, type of employment, and chronic disease) and ward-level characteristics (ward specialty, mean age of patients, number of patients, mean number of invasive devices in patients, prevalence of operated patients, and patient overcrowding). To examine whether the associations were dependent on the level of patient overcrowding at the ward, the interaction term “overcrowding × exposure to infectious diseases” was entered into the model after entering the main effects of overcrowding and exposure to infectious diseases.
Of the 54 wards, 12 (22%) were overcrowded, as indicated by bed occupancy of more than 85%9 during the study month. The mean overall patient infection prevalence was 25.1% vs 41.7% (P = .02) in nonovercrowded and overcrowded wards (hospital-acquired infection prevalence, 6.9% vs 14.0%; P = .03) (P < .05 for all, before and after adjustment for mean number of invasive devices in patients).
A total of 468 employees (47.1%) had at least 1 sickness absence spell, and 118 (11.9%) made a purchase of antibiotics during the follow-up. After adjustment for participants' age, sex, and specialty, ward overcrowding was associated with a 1.77-fold (95% confidence interval [CI], 1.13-2.77) odds of sickness absence but was not associated with antibiotic medication use. In addition, the high overall rate (highest tertile, ≥40.0%, vs lowest tertile, 0%-22.3%) of patients with infectious diseases was associated with employee antibiotic use (odds ratio [OR], 2.80; 95% CI, 1.11-7.07) after adjustment for all covariates.
We found a significant interaction between overcrowding and overall rate of infectious diseases in predicting sickness absence among staff (P value for interaction, .004) (Figure). In wards that were not overcrowded, there was no association between the prevalence of patients with infectious diseases and employee sickness absence. In contrast, in overcrowded wards, a high overall prevalence of infectious diseases (ie, above median, 27.5%) was related to a 2.39-fold (95% CI, 1.09-5.26) odds of sickness absence among the staff. A significant interaction between ward overcrowding and overall prevalence of infectious diseases in predicting sickness absence among the staff was also found when using a tertile-categorized infectious disease variable (P = .04).
The OR for the association between infectious diseases among patients and employees' antibiotic medication use was 1.64 (95% CI, 0.70-3.80) in nonovercrowded wards and 2.23 (95% CI, 0.53-9.36) in overcrowded wards (P value for interaction, .41). More detailed findings are available from the corresponding author on request.
Infectious diseases are currently an important public health issue and a challenge for hospitals, which additionally may have problems with patient overcrowding. Our data suggest that a high prevalence of patients with infectious diseases may adversely affect the health of staff, especially when the ward is overcrowded. Ward overcrowding was also associated with higher prevalence of infectious diseases among patients. In light of these findings, minimizing overcrowding in hospital wards, especially in those with high rates of patients with infectious diseases, might not only be beneficial to the patients but also could be regarded as a target to prioritize in the promotion of health among hospital staff.
Correspondence: Dr Virtanen, Finnish Institute of Occupational Health, Topeliuksenkatu 41 aA, FIN-00250 Helsinki, Finland (email@example.com).
Author Contributions: Drs Virtanen, Vahtera, and Kivimäki and Ms Pentti had full access to the data and take responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Virtanen, Terho, Oksanen, Kurvinen, Routamaa, Peltonen, and Kivimäki. Acquisition of data: Virtanen, Terho, Kurvinen, Routamaa, and Peltonen. Analysis and interpretation of data: Virtanen, Terho, Oksanen, Kurvinen, Pentti, Routamaa, Vartti, Peltonen, Vahtera, and Kivimäki. Drafting of the manuscript: Virtanen and Kivimäki. Critical revision of the manuscript for important intellectual content: Terho, Oksanen, Kurvinen, Pentti, Routamaa, Vartti, Peltonen, Vahtera, and Kivimäki. Statistical analysis: Virtanen and Pentti. Obtained funding: Vahtera and Kivimäki. Study supervision: Peltonen, Vahtera, and Kivimäki.
Financial Disclosure: None reported.
Funding/Support: The study was supported by grants 124322, 124271, 123621, 133535, and 129262 from the Academy of Finland; grant R01 HL036310 from the National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, Maryland; and the BUPA Foundation, United Kingdom.
Role of the Sponsors: The funders played no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; and preparation, review, or approval of the manuscript.
Disclaimer: The views expressed in this article are those of the authors and not necessarily of the funding bodies.