Conclusion contradicts data

This qualitative study of 28 recently circumcised Kenyan men, and another 18 awaiting circumcision, appeared in Health Education Research last week (click here).

As the authors admit, the design was inappropriate, and the samples too small and unrepresentative, to measure the rate of anything so, correctly for a qualitative study, they reported no numbers. Instead their aim was to tease out the existence of beliefs and behaviours which might have been missed in larger surveys, using in depth interviews and representative quotations. They were interested in whether men realised that circumcision only provided partial protection against HIV infection, and whether they were likely to increase sexually risky behaviour as a result.

For the first question it turned out that all respondents knew that circumcision provided only partial protection against HIV transmission. The authors concluded accurately that “Participants demonstrated good understanding of partial protection”.  

But when they turned to…