Electronic Letters to:

PAPERS:
STEFAN PRIEBE, GEMMA JONES, ROSEMARIE McCABE, JANE BRISCOE, DONNA WRIGHT, MICHELLE SLEED, and JENNIFER BEECHAM
Effectiveness and costs of acute day hospital treatment compared with conventional in-patient care: Randomised controlled trial
The British Journal of Psychiatry 2006; 188: 243-249 [Abstract] [Full text] [PDF]
*eLetters: Submit a response to this article

Electronic letters published:

[Read eLetter] Statistical significance and clinical significance
Mohammed J Abbas, On behalf of the journal club in Chesterfield   (21 June 2006)

Statistical significance and clinical significance 21 June 2006
Top
Mohammed J Abbas,
Specialist registrar in psychiatry
Sheffield Care Trust,
On behalf of the journal club in Chesterfield

Send letter to journal:
Re: Statistical significance and clinical significance

mohdgum{at}hotmail.com Mohammed J Abbas, et al.

Priebe et al (2006) conclude that “day hospital treatment for voluntary psychiatric patients in an inner-city area appears more effective in terms of reducing psychopathology in the short term and generates greater patient satisfaction than conventional in-patient care”. This paper was presented in the journal club as part of the postgraduate teaching program of North Derbyshire Mental Health Services NHS Trust in Chesterfield. Among other points raised during our discussion, we would like to point out two. The conclusion about patient satisfaction was based on a “significantly” higher mean CAT (Client’s Assessment of Treatment Scale) scores in the day hospital group at discharge (P=0.004) and 3 months after discharge (P=0.005). However, in both cases no confidence intervals of the mean differences were presented making it difficult to judge the generalisability of these findings.

When the confidence interval was presented for the other “statistically significant finding” (reduction in psychopathology at discharge using Brief Psychiatric Rating Scale as an outcome measure), it was very close to zero (lower limit=0.03) indicating a very small difference and limiting its generalisability. We think that this small size of difference should have been mentioned in the conclusions and limitations.

A statistically significant difference, especially when small, does not always indicate a clinical significance. We welcome the publication of papers which report negative or very small positive result on the assumption that the result and its size are reflected clearly in the conclusions.

References

Priebe, S., Jones, G., McCabe, R., Briscoe, J., Wright, D., Sleed, M., and Beecham, J. (2006) Effectiveness and costs of acute day hospital treatment compared with conventional in-patient care: Randomised controlled trial. British Journal of Psychiatry 188,243-249.