Reader Comments

Post a new comment on this article

The vaccine efficacy estimates are biased

Posted by CHvanWerkhoven on 13 Feb 2017 at 09:50 GMT

The controls in this study were "patients with unplanned hospital admission due to causes other than pneumonia or acute respiratory disease." These controls are in no way representative of patients at risk of being "hospitalized for at least 24h with a diagnosis of CAP." This is a form of selection bias. Therefore, the PPSV23 vaccine efficacy estimate in this study is invalid.

No competing interests declared.