Thank you for the opportunity of reviewing this excellent paper by a well-established research group, tracking the burden and geographic variability of respiratory viral carriage across 11 sites in Kenya.

Major comments: Collection over one calendar year is limiting considering the seasonal variation in carriage and infection rates related to circulating respiratory viruses. Temporal and regional patterns of respiratory virus activity cannot be adequately appreciated in a 12 month analysis. This limitation has been highlighted by the authors, and should serve as a strong motivation for ongoing, multi-year surveillance of virus activity across Kenya.

What justification was applied in applying test positivity only to PCR reactions that had Ct values of < 35? This should be referenced in the Methods section.

Relatively sparse representation of sites from the northeast of Kenya is a concern, and should be addressed in future respiratory virus surveillance activities by this research group.

Mention is made of informed consent. Was assent applied to children old enough to give assent? If so, this should be mentioned in the manuscript.

Temporal shifts in RSV serotype predominance cannot be adequately interpreted from the data presented in this manuscript, as a large proportion of RSV detections were of the combined “RSVA/B” category. If data are available to appreciate the RSV serotype in all tested specimens, then it would be appropriate to present a dichotomised “RSVA” and “RSVB” analysis; however, if these data are not available for all sites, I would recommend that the authors merely present a consolidated “RSV” summary prevalence for each site.

A formal analysis of the test positivity rates for RSV in inpatient compared to outpatient settings is not presented in the Results section, yet is prominently mentioned in the Abstract, Discussion and Conclusion sections of the manuscript. It is important to present this analysis in results, in order to justify the authors’ assertion that RSV vaccine delivery would be expected to impact favourably on hospitalisation rates, once such a vaccine becomes available.

Minor comments: Abstract: In the Conclusions, mention is made of the fact that respiratory syncytial virus (RSV) was more prevalent in inpatients. This finding is not presented in the Abstract Results, and should be presented there. Additionally, consider refining the sentence in the Conclusions to indicate that RSV was more prevalent in infants. Consider using: “Higher RSV positivity in inpatients, and in infants, strengthens the case for RSV vaccination.”

Presentation of RSVA/B, RSVA and RSVB bars in Figure 3 muddies the interpretation of RSV prevalence at each site somewhat. Suggest collapse these into one “RSV” bar for presentation in the main paper, and move the serotype specific RSV bar charts to Supplementary Materials.

Presentation of HCoVs, HCoV-OC43, HCoV-NL63 and HCoV-229E bars in Figure 3 muddies the interpretation of HCoV prevalence at each site somewhat. Suggest collapse these into one “HCoV” bar for presentation in the main paper, and move the serotype specific HCoV bar charts to Supplementary Materials.

Include “ADV, adenovirus” and “RV, human rhinovirus” in the legend to Figure 3.

Remove “HRV, parainfluenza virus types 1-4; OC43, human coronavirus OC43; NL63, human coronavirus NL63; E229, human coronavirus E229” from the legend in Figure 3.

Remove “RSVA, RSV group A; RSVB, RSV group B” from the legend in Figure 4, and replace with “RSV”.

Results section, second last paragraph: replace “didn’t” with “did not”.

In the Discussion, second paragraph on page 13, suggest reword to “Taken together the findings show that it is important to have robust, ongoing, year-round surveillance in different parts of the country as there may be important sub-national variations in circulation patterns…

Major comments: Collection over one calendar year is limiting considering the seasonal variation in carriage and infection rates related to circulating respiratory viruses. Temporal and regional patterns of respiratory virus activity cannot be adequately appreciated in a 12 month analysis. This limitation has been highlighted by the authors, and should serve as a strong motivation for ongoing, multi-year surveillance of virus activity across Kenya.

What justification was applied in applying test positivity only to PCR reactions that had Ct values of < 35? This should be referenced in the Methods section.

Relatively sparse representation of sites from the northeast of Kenya is a concern, and should be addressed in future respiratory virus surveillance activities by this research group.

Mention is made of informed consent. Was assent applied to children old enough to give assent? If so, this should be mentioned in the manuscript.

Temporal shifts in RSV serotype predominance cannot be adequately interpreted from the data presented in this manuscript, as a large proportion of RSV detections were of the combined “RSVA/B” category. If data are available to appreciate the RSV serotype in all tested specimens, then it would be appropriate to present a dichotomised “RSVA” and “RSVB” analysis; however, if these data are not available for all sites, I would recommend that the authors merely present a consolidated “RSV” summary prevalence for each site.

A formal analysis of the test positivity rates for RSV in inpatient compared to outpatient settings is not presented in the Results section, yet is prominently mentioned in the Abstract, Discussion and Conclusion sections of the manuscript. It is important to present this analysis in results, in order to justify the authors’ assertion that RSV vaccine delivery would be expected to impact favourably on hospitalisation rates, once such a vaccine becomes available.

Minor comments: Abstract: In the Conclusions, mention is made of the fact that respiratory syncytial virus (RSV) was more prevalent in inpatients. This finding is not presented in the Abstract Results, and should be presented there. Additionally, consider refining the sentence in the Conclusions to indicate that RSV was more prevalent in infants. Consider using: “Higher RSV positivity in inpatients, and in infants, strengthens the case for RSV vaccination.”

Presentation of RSVA/B, RSVA and RSVB bars in Figure 3 muddies the interpretation of RSV prevalence at each site somewhat. Suggest collapse these into one “RSV” bar for presentation in the main paper, and move the serotype specific RSV bar charts to Supplementary Materials.

Presentation of HCoVs, HCoV-OC43, HCoV-NL63 and HCoV-229E bars in Figure 3 muddies the interpretation of HCoV prevalence at each site somewhat. Suggest collapse these into one “HCoV” bar for presentation in the main paper, and move the serotype specific HCoV bar charts to Supplementary Materials.

Include “ADV, adenovirus” and “RV, human rhinovirus” in the legend to Figure 3.

Remove “HRV, parainfluenza virus types 1-4; OC43, human coronavirus OC43; NL63, human coronavirus NL63; E229, human coronavirus E229” from the legend in Figure 3.

Remove “RSVA, RSV group A; RSVB, RSV group B” from the legend in Figure 4, and replace with “RSV”.

Results section, second last paragraph: replace “didn’t” with “did not”.

In the Discussion, second paragraph on page 13, suggest reword to “Taken together the findings show that it is important to have robust, ongoing, year-round surveillance in different parts of the country as there may be important sub-national variations in circulation patterns…”

Thank you for the opportunity of reviewing the revised version of this manuscript. The authors highlight patterns of respiratory virus detection across 11 sites in Kenya during the 2014 calendar year. While considerable regional differences in virus detection occurred, a consistent finding was that of respiratory syncytial virus (RSV) detection in the youngest age groups, and in hospitalised patients, and adenovirus detection in children 12-23 months of age. The authors motivate for establishment of a national surveillance platform to track trends in respiratory virus isolation using standardised case definitions. Such a platform would be important in the lead up to introduction of candidate vaccines to prevent respiratory viral hospitalisations.

Major comments: The fact that RSV was more prevalent in hospitalised participants is not highlighted in the Results and Discussion sections. It would be useful to include a Table with viral detection prevalence stratified by inpatient versus outpatient status.

Death as an outcome is not described in this study. How many of the participants had an outcome of death? It would be useful to include a Table with viral detection prevalence stratified by survival status.

How did viral co-infection impact on hospitalisation and outcome status? It would be useful to include this in the suggested tables mentioned above. Minor comments: In the Abstract, mention in the results section that RSV was more prevalent in inpatients compared to outpatients.

In the first paragraph of the Introduction, suggest omit the word “recent” in the sentence which highlights the introduction of bacterial conjugate vaccines.