In Reply Drs Thombs and Ziegelstein note 3 differences between our study registration in May 2010 and the protocol published in BioMed Central (BMC) Health Services Research1 in April 2013 and the JAMA article. First, the follow-up interval was omitted from the primary composite process outcome due to lack of evidence for any particular follow-up interval and lack of specificity for the reason for follow-up. Subsequent reanalysis including follow-up interval in the process outcome did not materially affect the results (the results are available from the authors).
Kravitz RL, Jerant A, Tancredi DJ. Patient Engagement Programs and Treatment of Depression—Reply. JAMA. 2014;311(11):1159-1160. doi:10.1001/jama.2014.579