On 2016 Nov 12, Tom Yates commented:

Pretorius and colleagues are to be congratulated for undertaking perhaps the largest case control study to date exploring the viral aetiology of severe acute respiratory illness (SARI) (Pretorius MA, 2016). The study is remarkable for two reasons other than its size.

First, large numbers of HIV positive cases and controls were enrolled. Second, unlike other studies of acute respiratory tract infection or pneumonia undertaken in Sub Saharan Africa (Hammitt LL, 2012, Bénet T, 2015, Breiman RF, 2015), human rhinovirus was isolated more frequently in cases than in controls. The investigators estimated that the population fraction of SARI attributable to rhinovirus (an entity they call ‘adjusted prevalence’) was approximately twenty percent.

One potential reason for an association between rhinovirus infection and SARI being observed in this study but not in other studies would be if rhinovirus were only a cause of SARI in the setting of HIV-related immunocompromise. It would therefore be very interesting to see these data presented separately by HIV status.

Tom A. Yates ^a ^b ^*

Patrick K. Munywoki ^a ^c

D. James Nokes ^a ^d

a) Virus Epidemiology and Control Research Group, Epidemiology and Demography Department, KEMRI-Wellcome Trust Research Programme, Kilifi, Kenya b) Institute for Global Health, University College London, London, UK c) School of Health and Human Sciences, Pwani University, Kilifi, Kenya d) School of Life Sciences and WIDER, University of Warwick, Coventry, UK

* Dr Tom Yates, t.yates@ucl.ac.uk

On 2016 Nov 12, Tom Yates commented:

Pretorius and colleagues are to be congratulated for undertaking perhaps the largest case control study to date exploring the viral aetiology of severe acute respiratory illness (SARI) (Pretorius MA, 2016). The study is remarkable for two reasons other than its size.

First, large numbers of HIV positive cases and controls were enrolled. Second, unlike other studies of acute respiratory tract infection or pneumonia undertaken in Sub Saharan Africa (Hammitt LL, 2012, Bénet T, 2015, Breiman RF, 2015), human rhinovirus was isolated more frequently in cases than in controls. The investigators estimated that the population fraction of SARI attributable to rhinovirus (an entity they call ‘adjusted prevalence’) was approximately twenty percent.

One potential reason for an association between rhinovirus infection and SARI being observed in this study but not in other studies would be if rhinovirus were only a cause of SARI in the setting of HIV-related immunocompromise. It would therefore be very interesting to see these data presented separately by HIV status.

Tom A. Yates ^a ^b ^*

Patrick K. Munywoki ^a ^c

D. James Nokes ^a ^d

a) Virus Epidemiology and Control Research Group, Epidemiology and Demography Department, KEMRI-Wellcome Trust Research Programme, Kilifi, Kenya b) Institute for Global Health, University College London, London, UK c) School of Health and Human Sciences, Pwani University, Kilifi, Kenya d) School of Life Sciences and WIDER, University of Warwick, Coventry, UK

* Dr Tom Yates, t.yates@ucl.ac.uk