Risk of infection following a visit to the emergency department: a cohort study

Caroline Quach; Margaret McArthur; Allison McGeer; Lynne Li; Andrew Simor; Marc Dionne; Edith Lévesque; Lucie Tremblay

doi:10.1503/cmaj.110372

Submit a Response to This Article

Compose Response

Title *

References

Author Information

Contributors

First Name and Middle Initial *

First or given name, e.g. 'Peter'.

Last Name *

Your last, or family, name, e.g. 'MacMoody'.

Degrees

Email Address *

Your email address, e.g. higgs-boson@gmail.com

Occupation *

Your role and/or occupation, e.g. 'Orthopedic Surgeon'. Minimum 7 characters.

Affiliation *

Your organization or institution (if applicable), e.g. 'Royal Free Hospital'. Minimum 12 characters.

Address

Your organization, institution's or residential address.

Statement of Competing Interests

Competing interests? *

Yes

Please describe the competing interests

Vertical Tabs

Jump to comment:

Re:Inferring Causality from an Association Between Emergency Department Visits and Subsequent Infections
Caroline Quach

Posted on: 09 February 2012
Inferring Causality from an Association Between Emergency Department Visits and Subsequent Infections
Timothy F. Platts-Mills

Posted on: 01 February 2012

Posted on: (9 February 2012)
Page navigation anchor for Re:Inferring Causality from an Association Between Emergency Department Visits and Subsequent Infections
Re:Inferring Causality from an Association Between Emergency Department Visits and Subsequent Infections
Caroline Quach, Infection Control Physician
We wish to thank Dr. Platts-Mills for his comment, which highlights the methodological difficulties associated with the question asked, particularly in the absence of microbiological confirmation, and we agree - as stated in our conclusion that "confirmation of these results with studies of specific types of infection with laboratory testing is required"(1). Aiming at decreasing this potential bias, we only included lo...
Show More

We wish to thank Dr. Platts-Mills for his comment, which highlights the methodological difficulties associated with the question asked, particularly in the absence of microbiological confirmation, and we agree - as stated in our conclusion that "confirmation of these results with studies of specific types of infection with laboratory testing is required"(1). Aiming at decreasing this potential bias, we only included long-term care facility residents with a visit to the Emergency Department (ED) for a diagnosis other than a respiratory or a gastrointestinal infection, using the ED discharge diagnosis (not the chief complaint at presentation), and screening ED charts and long-term care facilities charts to remove patients with symptoms of these illnesses prior to or during their ED visit. Although it is thought that long-term care facility residents may be more likely to have atypical presentations of infections, it was shown by Berman and colleagues (3) that "in most elderly patients who develop infection, there remain clear clinical pointers to the diagnosis. The symptoms may be absent or unreliable but the physical signs remain". Further, in our study population, the onset of symptoms started on average 4.1 days after the resident's returned from the ED (median: 3.5), which is the typical incubation period for respiratory and gastrointestinal infections following an exposure (4).
Moreover, if residents exposed to the ED were incubating a respiratory or gastrointestinal infection at the time of ED exposure, we would have expected that transmission would also have occurred on the resident's unit in the long-term care facility. Therefore, we used as non- exposed, 2 randomly chosen residents matched on the unit or ward in the same long-term care facility and on the index date (upon the exposed resident's return from the ED), expecting that if the exposed resident had been exposed in his/her long-term care facility before presenting to the ED, the randomly chosen non-exposed residents would have the same chance of exposure - which would have decreased the strength of the association found.
Although the objective of our study was not to infer causality, the association found is biologically plausible and is in keeping with the increased risk described by Troko et al (5) associated with other densely populated areas where symptomatic people may gather (i.e. bus and trams in the UK). Troko et al. reported that recent bus or tram use within five days of symptom onset was associated with an almost six-fold increased risk of consulting for acute respiratory infections (adjusted OR = 5.94 95% CI 1.33-26.5).
We agree with Dr. Platts-Mills that EDs - and hospitals - are important sites of care for long term care facility residents. This does not, however, mean that we can assume that care in these sites is free of adverse events, and cannot be improved. We hope that the results of our study stimulate other investigators to confirm or refute our findings, and support staff in EDs as they work to ensure the safest possible environment for all the patients in their care.
Sincerely,
Caroline Quach, M.D. M.Sc. - Division of Infectious Diseases, Department of Pediatrics and Medical Microbiology - The Montreal Children's Hospital, McGill University Allison McGeer, M.D. M.Sc. - Department of Microbiology, Mount Sinai Hospital, University of Toronto Andrew Simor, M.D. - Department of Microbiology, Sunnybrook Health Sciences Center, University of Toronto
References
1. Quach C, McArthur M, McGeer A, et al. Risk of infection following a visit to the emergency department: a cohort study. CMAJ 2012.
2. Fleischauer AT, Silk BJ, Schumacher M, et al. The validity of chief complaint and discharge diagnosis in emergency department-based syndromic surveillance. Acad Emerg Med 2004;11:1262-7.
3. Berman P, Hogan DB, Fox RA. The Atypical Presentation of Infection in Old Age. Age and Ageing 1987; 16: 201-7.
4. Pickering LK, editor. Red Book: 2009 Report of the Committee on Infectious Diseases 28th edition, Elk Grove Village IL: American Academy of Pediatrics, 2009: 984 pages.
5. Troko J, Myles P, Gibson J. et al. Is Public Transport a Risk Factor for Acute Respiratory Infection? BMC Infect Dis 2011; 11: 16.
Conflict of Interest:
None declared
Show Less

Competing Interests: None declared.
Posted on: (1 February 2012)
Page navigation anchor for Inferring Causality from an Association Between Emergency Department Visits and Subsequent Infections
Inferring Causality from an Association Between Emergency Department Visits and Subsequent Infections
Timothy F. Platts-Mills, Assistant Professor
Other Contributors:
We read with interest the article by Quach and colleagues, in which an association was identified between emergency department (ED) visits and subsequent infections for elderly residents of long term care facilities.1 We, however, disagree with the authors' interpretation that the findings suggest a causal link between emergency department visits and subsequent infections. What is more likely is that infections in long te...
Show More

We read with interest the article by Quach and colleagues, in which an association was identified between emergency department (ED) visits and subsequent infections for elderly residents of long term care facilities.1 We, however, disagree with the authors' interpretation that the findings suggest a causal link between emergency department visits and subsequent infections. What is more likely is that infections in long term care facility patients caused ED visits for non-specific symptoms, that the emergency provider did not always make the diagnosis, and that the diagnosis was made in the subsequent week.
An essential feature of a cohort study is that patients must be disease free at the time the exposure is assessed. In this study the exposure is the ED visit; so, excluding infection as the cause of the ED visit is paramount. The authors' method of excluding infections at the time of the ED visit was to examine the reason for the ED visit. This approach is problematic because reasons for visit often match poorly with the clinical diagnosis.2 For example, a complaint of 'mobility impairment' in a nursing home patient does not exclude infection as an etiology. That the ED visit came before the diagnosis of infection in no way establishes the direction of causality, a problem described by Bradford-Hill in 1965 as a problem of temporality.3
Emergency departments are an important site of care for nursing home residents, and inferring with insufficient evidence that they are harmful does a disservice to patients who need such care.4 In our opinion, concluding that people get sick after visiting an ED when the sickness may have predated the ED visit is analogous to concluding that calling a domestic violence hotline causes personal injury if (as would no doubt be true) persons who'd called the hotline had more violent episodes in the week after the call than persons who did not call the hotline. We do not dispute that any public place can lead to transmission of infection, including potentially hospital emergency departments, but in our opinion the evidence presented by Quach and colleagues and the methods used to gather that evidence are insufficient to support this conclusion.
References
1. Quach C, McArthur M, McGeer A, et al. Risk of infection following a visit to the emergency department: a cohort study. Cmaj 2012.
2. Fleischauer AT, Silk BJ, Schumacher M, et al. The validity of chief complaint and discharge diagnosis in emergency department-based syndromic surveillance. Acad Emerg Med 2004;11:1262-7.
3. Bradford Hill A. The environment and disease: association or causation? Proceedings of the Royan Society of Medicine 2965;58:295-300.
4. Wang HE, Shah MN, Allman RM, Kilgore M. Emergency department visits by nursing home residents in the United States. J Am Geriatr Soc 2011;59:1864 -72.
Conflict of Interest:
None declared
Show Less

Competing Interests: None declared.