Table 5.

Methodological Challenges Experienced During the Study and Solutions

PROBLEM or CHALLENGE	SOLUTION
Subject burden.	Decrease frequency of administration of Short Physical Performance Battery.
10/18/11 Unclear randomization procedure.	Randomization procedure will include stratification history of major depression or an anxiety disorder. The data manager will maintain the randomization list. Once consent is signed, study staff will contact data manager for randomization assignment. Because of the sequenced design, will need to maintain two randomization lists.
Timing of booster sessions	Booster sessions will be administered at 3 and 9 months. Emergency booster sessions are an option and will be documented in the database.
2–10–12 First participant consented and randomized.
Clinical and research management of participants who become syndromal	If a subject develops MDD or anxiety disorder they will exit the intervention but continue in follow-up.
Timing of initiation of follow-up phase	For participants in EUC, theT2 date will be projected for 12 weeks after randomization. For participants receiving interventions, the T2 date may vary and will occur within 1–2 weeks following end of the intervention.
Unclear description about depression symptoms and treatment history at entry into study	Consistent with entry criteria, we will not enroll participants taking antidepressants. Potential participants must endorse either item 1 or 2 on the PHQ9 indicating emotional distress. Without these symptoms endorsed, they may not be at elevated risk for MDD or anxiety disorder.
Assessment of outcomes of interest	The PRIME MD will be done at all assessment timepoints regardless of PHQ & GAD scores to fully evaluate for the onset of a mood or anxiety disorder.
Capture of serious adverse event data. Reminder about anxiety disorders as outcome of interest. Missing items.	1.Added a 2 item form to capture hospitalizations that occur during the study. 1.In addition to MDD, the onset of an anxiety disorder during study participation is considered an end point of the study. This would include any anxiety disorder that would have excluded someone from initial study participation (e.g, generalized anxiety disorder, panic disorder, post traumatic stress disorder). Participants who meet criteria for social anxiety disorder, specific phobias, and anxiety disorder NOS are allowed in the study and these conditions, if they develop during follow-up, would not be considered an endpoint. 1.When administering the WOMAC, there should not be any missing items. The items should be scored “as if” the person would do the task. For example, if someone only takes showers and the question is “do you have difficulty getting in/out of bathtub”?: ask how much difficulty they would have if they had to get in/out of tub and score accordingly.
Measuring participant engagement	Added a homework Effort/Participation form to capture participant engagement in doing homework between sessions.
Low rates of serial blood draws for biomarkers	To increase rates of blood draws for biomarkers, we will pay for parking and/or a cab to get the blood drawn at the on-site clinic by a trained phlebotomist.
Retention during follow-up	To improve retention during the year of follow-up, we will schedule the next assessment before a participant leaves. We will also send a reminder letter 2–3 weeks before each assessment timepoint and call the subject 2 days before the appointment.
Collection of nutritional supplement data.	All over the counter nutritional supplements will be included on the medication list and entered into the database.
Management of participants who start an antidepressant during participation.	A subject started on an antidepressant would continue in the study as a regular participant, and they will be eligible for boosters and for scheduled assessments. The data base will reflect exposure to the antidepressant, dose, and medical indication for its having been prescribed.
Emerging concerns about participant burden	When administering the SCID at baseline, there is no need to administer the somatoform disorders and eating disorders sections.
Management of drop-outs.	If a subject withdraws from the intervention phases of the study after randomization because they do not like the randomized assignment, we will attempt to complete all assessments over the year of follow-up. If the subject meets criteria for the second randomization and once again refuses the assignment, the plan will still be to complete all assessment time points until the end of the study.
Window of time allowed for collection of follow-up data.	If a missed visit is more than 6 weeks outside of an assessment time point, we will consider it a missed visit. At the next scheduled assessment, we will just do the assessments required at the current time point. If the final (T6) assessment is difficult to obtain, we can collect this data up to 3 months after the due date to minimize missing data.
Assurance that entry criteria have been met prior to randomization	To assure participants meet all inclusion and exclusion criteria, the multidisciplinary clinical case review must always be completed prior to randomization.
Further evaluation of alcohol misuse	To further evaluate alcohol use, we will screen for alcohol with the Alcohol Use Disorders Identification Test (AUDIT- C)40 will be done on the day of SCID for all participants and this information discussed at the multidisciplinary clinical case review.
Management of delay between screening and baseline.	If more than 4 weeks lapse between screening and baseline, we must repeat the PHQ9 and Prime MD to assure continued study eligibility.
Management of participants randomized but who never received any intervention.	Participants who have been randomized but never started their assigned intervention will be considered to be “follow-up only” and every attempt will be made to collect assessments for the following 12 months.