Authors’ response

We thank Nair and colleagues for their comments on our study on HIV prevention programmes and new interventions among truckers in India1. Our responses to the issues raised by the authors are as follows:

1. Material & Methods, paragraph 21 - It was not stated in the paper that the sample size was based on a previous study that the authors have referred to. The statement is as follows: “Assuming HIV prevalence of 6% from the published literature in this population”1.

2. Material & Methods, paragraph 31 – The sampling interval was chosen based on the number of truckers available and the sample required for the study from a particular halt-point. Describing the sampling interval as every 3^rd or 5^th truck was to convey to the readers the approach used. It does not imply that this interval was used exactly across all halt-points.

3. Material & Methods, paragraph 3-

   1. Selection of one trucker from each truck – The following details are provided in the paper: “As trucks were parked in an organized fashion in most of these halt points, systematic listing of the trucks parked (from entry to exit point) was undertaken to document the vehicle number, whether long-haul truck or not, number of truckers per truck with their age and language spoken by them. This information was used to identify the eligible truckers for the study. A potential respondent was chosen to participate in the study from every 3^rd or 5^th truck depending on the sample size required from each halt point with the first respondent chosen randomly from the sampling interval1”. Therefore, if more than one eligible truckers were listed for a truck, the sampling strategy did not allow for the other to get selected, and the selection was random.

   2. Exclusion of certain categories: N3 category trucks were excluded because the proportion of these trucks was negligible. We included languages that were most commonly spoken by the truckers who passed through the selected halt-points.

   3. Generalizability of results: We have not claimed that the study findings can be generalized to the whole of south India. The statement in the paper is as follows: “The results of this study may not be generalizable to truckers all over India, but the relatively large sample size of the study conducted on truckers passing through in a large city in southern India suggest that the findings are likely to be applicable to a large proportion of truckers either living or passing through southern India”1.

   4. Social desirability is more likely to result in over-reporting of contact with HIV programmes than under-reporting of contact.

4. Results, Multivariate analysis (Tables II and III) - The results of multivariate analysis are reported as “odds ratio” under the columns with headers - “adjusted odds for ever contact”; “adjusted odds for undergone HIV test”; and “odds of willingness to undergo circumcision”1.

5. Results, Table III - In this Table, the categories are clearly stated: working for <5 years and >5 years; age <30 years and >30 years.

6. We have provided characteristics for willingness for circumcision and telephone counselling, as these are of immediate interest. It is not possible to get accurate data on income in most surveys. For example, national surveys such as NFHS and DLHS use assets index to arrive at socio-economic status of a household.

7. Discussion, paragraph 7 - This appears to be a typographical error. The text should have read as “female clients” and not “FSW clients”. Reference 23 describes the impact of introducing the rapid oral fluid HIV testing of female clients at a tertiary care hospital in rural India. Reference 24 is an editorial in which acceptability of rapid oral fluid HIV testing is described in different settings in developing world, of which three are from India.