The following is the abstract of a paper:
Objective To compare standard management of keeping wounds dry and covered with allowing wounds to be uncovered and wet in the first 48 hours after minor skin excision.
Design Prospective, randomised controlled, multicentre trial testing for equivalence of infection rates.
Setting Primary care in regional centre, Queensland, Australia.
Participants 857 patients randomised to either keep their wound dry and covered (n = 442) or remove the dressing and wet the wound (n = 415).
Results The incidence of infection in the intervention group (8.4%) was not inferior to the incidence in the control group (8.9%) (P < 0.05). The one sided 95% confidence interval for the difference of infection rates was infinity to 0.028.
Conclusion These results indicate that wounds can be uncovered and allowed to get wet in the first 48 hours after minor skin excision without increasing the incidence of infection.
(Heal et al., 2006)
The Results section of the paper contined the following table:
| Variable | Wet (intervention) (n=450) | Dry (control) (n=420) | P value |
|---|---|---|---|
| Mean (SD) age (years) | 55.9 (16.6) | 56.5 (16.2) | 0.58 |
| Male patients | 249 (55) | 208 (50) | 0.08 |
| Mean (SD) days to removal of sutures | 8.6 (2.2) | 8.6 (2.2) | 1 |
| Presence of diabetes | 9 (2) | 14 (3) | 0.2 |
| History of other medical condition* | 8 (2) | 10 (2) | 0.5 |
| Treated with 1% lignocaine adrenaline | 435 (97) | 411 (98) | 0.3 |
| Excision of skin cancer | 294 (65) | 289 (69) | 0.3 |
| Excision at lower limb | 112 (25) | 106 (25) | 1 |
| * Chronic obstructive pulmonary disease (n=8), anaemia (1), "aspirin" (2), "steroids" (3), "warfarin" (2), ischaemic heart disease (1), and peripheral vascular disease (1). | |||
What null hypothesis are these tests testing?
Is this a null hypothesis which should be tested in a randomised trial? What would it mean if one of these tests was significant?
In the Methods section of the paper, we read:
"We calculated sample size on the basis of a pilot study done in February to June 2004 and involving 543 patients, which showed an overall infection rate of 5.7%. On the basis of a projected infection rate of 5%, we decided that an increase in incidence of infection of 5% would be clinically significant."
Hence the view of these authors was that there would be no important difference between the treatments if the proportion experiencing infection rose from 5% for covered sutures to 10% for uncovered sutures. They intended to cary out "a one sided equivalence test of proportions".
These authors wanted to know whether allowing wounds to get wet would be detrimental to the patient, in particular whether it would increase the risk of infection by more than 5 percentage points. In the Results section, Heal et al. (2006) report that:
"The intervention group had an infection rate of 8.4% compared with 8.9% in the control group. The one sided 95% confidence interval of the difference of the two proportions was infinity to 0.028, so the non-inferiority side was lower than 0.05, the maximum allowable difference. We therefore concluded that the intervention group was not inferior to the control group with respect to the resulting infection rates (P < 0.05)."
What is a one-tailed or one-sided test?
What null hypothesis and alternative hypotheses were being tested here? How would would a significant result be interpreted?
What argument might be made for a one-tailed test here? Do you think it is sound?
How could we interpret the confidence interval infinity to 0.028? Why must it be wrong?
What alternative analysis could these authors have done?
Heal C, Buettner P, Raasch R, Browning S, Graham D, Bidgood R, Campbell M, Cruikshank R. (2006) Can sutures get wet? Prospective randomised controlled trial of wound management in general practice. British Medical Journal 332, 1053-1056.
To Clinical Biostatistics index.
To Martin Bland's M.Sc. index.
This page maintained by Martin Bland.
Last updated: 31 July, 2006.