Hostname: page-component-78c5997874-8bhkd Total loading time: 0 Render date: 2024-11-07T16:33:56.583Z Has data issue: false hasContentIssue false

Limitations of observational studies with known confounders

Published online by Cambridge University Press:  02 January 2018

Nuwan Galappathie
Affiliation:
Fromeside, Blackberry Hill, Bristol BS16 1ED, email: [email protected]
Ragini Heeramun
Affiliation:
Fromeside
Rights & Permissions [Opens in a new window]

Abstract

Type
The columns
Creative Commons
Creative Common License - CCCreative Common License - BY
This is an Open Access article, distributed under the terms of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted re-use, distribution, and reproduction in any medium, provided the original work is properly cited.
Copyright
Copyright © Royal College of Psychiatrists, 2008

The recent article by Dibben et al (Reference Dibben, Saeed and Stagias2008) provides positive evidence that crisis resolution and home treatment teams for older people reduce admission rates to hospital by 31%. Their findings are based upon a 6-month observational study conducted before and after the introduction of the new service. We believe the results are unreliable and should be treated with caution.

The analysis of baseline characteristics shows no significant difference between the patient groups; however, it is misleading as external factors are likely to confound the results in this case.

The authors confirm that during the study period a dementia care ward was closed, as were two day-hospitals. These changes are likely to represent known confounding variables. Closure of the dementia ward will invariably lead to reduced admission rates as a direct result of reduced capacity to admit service users. The day hospital closure may also serve to raise the clinical threshold required for admission given the reduction in day attendance capacity.

No data is provided regarding in-patient bed capacity and how this was reduced following bed closures, thus readers are unable to estimate the magnitude of the potential impact upon results. In addition, no data is provided for the potential known confounding variable of out-of-area admissions, which can occur when a hospital reaches full occupancy.

Clearly, the gold standard for study design would have been a randomised controlled trial. This is often not possible given feasibility implications. Ideally, the authors should have used a control group of service users who were not treated by the new service. This should then have been followed by appropriate use of multivariate regression analysis to account for the effects of the known confounding variables. The authors rely on past randomised research to support their findings (Reference Johnson, Nolan and PillingJohnson et al, 2005; Reference Joy, Adams and RiceJoy et al, 2006). The participants of the cited studies were not solely older adults and the results are therefore not directly generalisable to their sample. We accept their research is novel and designed with service limitations in mind, but we believe they offer only a basic understanding of their subject.

References

Dibben, C., Saeed, H., Stagias, K., et al (2008) Crisis resolution and home treatment teams for older people with mental illness. Psychiatric Bulletin, 32, 268270.CrossRefGoogle Scholar
Johnson, S., Nolan, F., Pilling, S., et al (2005) Randomised controlled trial of acute mental health care by a crisis resolution team: the north Islington crisis study. BMJ, 331, 599.CrossRefGoogle Scholar
Joy, C. B., Adams, C. E. & Rice, K. (2006) Crisis intervention for people with severe mental illnesses. Cochrane Database of Systematic Reviews, 18, CD001087.Google Scholar
Submit a response

eLetters

No eLetters have been published for this article.