Report reviewer 1 Title : Reproductive health issues in rural Western Kenya

no need to define MUAC if you don't report it in the abstract Response: We agree and removed MUAC from the abstract. The results section of the abstract should specify "symptoms" of zoster and thrush, unless these were more definitive diagnoses? Response: We inserted “symptoms of” as suggested. Methods, second para: do you mean "loss to follow-up" or "non-response"? Response: We mean indeed non-response here and changed it accordingly. Thanks. Discussion ~ it would be nice to pick up the thread of MDGs in the discussion (since it was introduced earlier). Specifically, what are the implications of these findings for interventions to achieve the MDGs? Response: We thank the reviewer for bringing this up, and added a paragraph on this as follows (at the end of the discussion section, before conclusion): “It is obvious there are many challenges in this area to achieve the MDGs of reduction of child mortality, improvement of maternal health and combating HIV/AIDS, malaria and other diseases. Programs are underway to decrease HIVtransmission, and improve the care for HIV-infected persons. In the antenatal clinics, programs have been implemented to improve maternal health and decrease mother-to-child transmission of HIV. ITNs are promoted and used. Family planning programs are present in the area. It will be important to continue evaluating if the existing programs are optimally functioning, and if the target group is optimally using them; in addition, if programs are not used, it is important to evaluate what barriers exist, and how programs can be changed to achieve their goals.” ** Major compulsory revisions ** Results: the results current read as a simple description of proportions/means, etc. While these descriptive statistics are valuable, the authors could greatly enhance the paper by providing a more detailed analysis of the demographic characteristics associated with different reproductive health states. Multivariate analysis is not necessary (but would be nice), but the crude associations between different repro health measures and participant age, education, SES, etc, would be very interesting. This could be reflected in the tables, also. Response: Although we had added quite some information with regards to the above in the text, we now changed the tables as well. We split table 1 in table 1 and table 4, whereby table 1 contains the basic characteristics and table 4 contains the different health issues stratified by age. We also examined other characteristics (level of education, marital status, and SES), but concluded that age was the most interesting variable, whereas the others were not associated with any of the health indicators in table 4. There is co-variance between age and gravidity, and for this reason factors associated with age were likely to be associated with gravidity. For some health indicators, the association with gravidity is covered in figure 1. We added the information of (previous) table 2 to the text, and replaced this table with a table showing the univariate and multivariate analysis of factors associated with the use of modern contraceptives (definition and method added to methods section). We looked at factors associated with outcomes in table 3, but did not find an association between marital status, SES or education level and these outcomes, and included this in the text discussing this table. The stratification in table 1 according to the availability of SES data is not particularly helpful, and could be mentioned briefly in the text. instead, why not provide a breakdown by participant age groups? Response: See above: We agree with the reviewer and removed the SES stratification from table 1. Table 2 is interesting, but could also be analysed by participant age/education levels, to give a better sense of the patterning of contraceptive use in the population Response: We changed the table as suggested. In figure 1, it would be useful to provide the N's for each gravidity category Response: We added the N’s for each gravidity category. Report reviewer 2: Discretionary revisions 1.Abstract -End of last sentence: nets....and (high) prevalence HIV -My initial thought was, what is Gem, maybe state it is an area/district. Response: We removed Gem and changed it to a rural area in western Kenya. The results presented within the abstract would flow better and be easier to read if they were laid out more methodically i.e ITN use, malaria, anaemia, helminths, underweight, contraceptive methods, conception. Response: We changed the order of the abstract as suggested. HIV rate is stated to be high, but what was the prevalence? Response: We added the prevalence as described for Nyanza Province in the Demographic and Health Survey in 2003 to the abstract and the method section. 1 Background. promotion of gender equality... form a VITAL role, not simply an important role. Response: We changed this as suggested. 2 Last paragraph is difficult to follow and requires some clarification; 17% women delivered with skilled attendant and (instead of whereas) 18% on their own. A subsequent survey here..... (where, when). This needs clarification. Response: We changed as suggested, and described timing and place of the surveys. 3 Methods. Given the ITN 5 yr trial,a long with the follow up is it really necessary to state that malaria is holoendemic? Response: We agree that this sentence is superfluous and have removed it. -Last sentence re total fertiltiy rate: was or is 5.3%? Response: We added “calculated for 2002” to the sentence, to make clear to what year this fertility rate applies. -Procedures. To me there needs to be some clarification here. Were the lab technicians/clinical officers trained up to follow study guidelines, were the same practtitioners used throughout the study. One clinicians findings may differ from that of another, were there specific procedures put in place? Response: the lab technicians and clinical officers (3) were trained to follow the study guidelines, and the same technicians and clinical officers were used throughout the study. All forms were checked before data-entry and inconsistencies were checked by the investigators with the study staff responsible for the forms.