Using Peer Mentors to Deliver Depression Care

Recruitment of peer mentors. We will recruit peer mentors from a community organization who are in recovery and are recommended by the organization and have had basic training in supportive listening and experience working with persons in the community in need of mental health support. We will recruit 10 peer mentors, striving for a balance of men and women, and accounting for attrition, so that at least 5 peer mentors will participate in the project. The study doctor will meet and interview interested peers to screen for eligibility criteria, motivation and aptitude. Persons who do not meet criteria for continued participation in the study will be referred to other volunteer opportunities when appropriate.

Recruitment of patients. Twenty patient participants will be recruited through the internal medicine and psychiatry clinics at Johns Hopkins Bayview Medical Center as well as the community. A research assistant will screen patients for depressive symptoms with the Patient Health Questionnaire-9 (PHQ-9). Patients meeting eligibility criteria will give written informed consent. Eligible patients will be scheduled for another appointment to meet with the study doctor in a private room at offices in the clinical campus.

Training of peer mentors. The PI and research assistant will conduct 5 sessions totaling 20 hours of training for the peer mentors. Training will refine the peer mentor's social support behaviors, which include the use of effective communication skills, rapport building, active listening, and sharing of experiential knowledge, and expression of empathy. Issues of cultural interviewing, cultural competence, patient confidentiality, DSM-IV diagnosis, differentiation of peer and professional roles and patient safety will be addressed. Knowledge regarding how mental health and illness presents itself in old age and the specific developmental problems older adults face will be provided. Training will include review of written materials, lectures, discussion of case studies and role play. Upon completion of the training, peer mentors will be assessed for mental health knowledge and interpersonal skills and those who pass muster will be matched with patients.

The peer mentor role: Activities of the peer mentor with the patient will be structured by the following principles which are derived from important elements of both peer support and psychotherapy and our preliminary work:

Principles to structure peer mentor interactions with patients

1. Establish a working relationship
2. Establish goals of meetings with patient
3. Provide active listening, empathy, reciprocity, mutual respect, provision of emotional support Provide support and encourage work toward patient-identified goals to relieve depression
4. Work on patient-identified goals by engaging multiple techniques as appropriate: sharing appropriate coping strategies eg. positive thinking, experiential knowledge, problem solving, consideration of different perspectives.
5. Provide psychoeducation. Discuss the kinds of depression services available, medications, psychotherapy Acknowledge socioeconomic contributors to depression
6. Provide information on community resources
7. Connect to professional clinical services if appropriate The psychiatrist role: Peer mentors will meet with the candidate who is a psychiatrist on a weekly basis. The psychiatrist will provide: (1) clinical leadership and structure by reinforcing the principles of the peer-patient meetings; and (2) supervision and coaching which will involve encouragement, continuing education, training and skill enhancement for the peer mentor. The psychiatrist will guide the peer mentor to ensure that a positive social experience is being provided and to manage the patient in a way that ensures the realization of the patient's goal. The peer mentor will share impressions of patient progress and receive any guidance needed from the psychiatrist.

Study procedures. After patient consent is obtained, the psychiatrist will meet with the patient for a clinical interview so that the psychiatrist can form an impression of the patient and provide relevant guidance to the peer mentor. The patient will then be matched with a peer mentor. The peer mentor will contact the patient by telephone to schedule a meeting and meet with the peer mentor at his home or in the clinic. The peer mentor will meet the patient weekly for 8 one-hour sessions. The peer mentors will meet individually on a weekly basis with the psychiatrist during the study period in a collaborative and supervisory effort regarding the care of the patient Data Analysis. Analysis will involve both quantitative and qualitative methods. We will identify measured factors related to engagement and depression response and remission so that we can understand how the delivery model may need to be tailored or modified. This analysis will consider depression response and remission and engagement as dependent variables, with other variables as predictors or effect modifiers. We will also carry out debriefing interviews with peers and patients in order to look for themes that will call attention to factors that are most salient to patients and peers.

Quantitative analysis. Influential variables will be identified in the statistical models according to association with the dependent variables, namely 1) engagement in depression care defined as two or more sessions attended with the peer mentor, and 2) decrease in depressive symptoms which will be treated as a continuous variable.

Qualitative analysis. We will collect qualitative data to obtain the perspective of peer mentors and patients themselves regarding study processes and mediators from their point of view. Two semi-structured interviews with patients and peer mentors will be conducted by the PI at 4 weeks and upon completion of the study. The purpose will be to obtain feedback regarding positive and negative aspects of the peer delivery model which will provide perspectives different from the expert and valuable for that reason.

ANALYSIS OF PROCESS We will analyze the process of the peer mentor-patient interactions by 1) conducting an interaction analysis which will provide quantitative associations between communication elements and outcomes to determine which elements are salient to the effectiveness of the peer delivery model, and 2) conducting a qualitative analysis of the voice recordings to assess how the peer mentors implement the critical elements and how patients respond to them. The analyses of process will serve to determine the content of the training for peer mentors, the activities peers will perform with patients and manual development.

The Roter Interaction Analysis System (RIAS) is a widely used system for interaction assessment in medical contexts. The RIAS has demonstrated predictive validity to patient satisfaction, patient recall, and improvements in levels of emotional distress. Combined with quantitative process and outcome measures of care, statistical analysis can establish associations between specific communicative thematic categories and patient outcomes.

Data Analysis. We will evaluate the specific interactions between peer and patient employing the RIAS to investigate the specific interactions that are associated with engagement and depression response and remission so that we can refine the training required for peers and develop measures tapping effective interaction strategies. We will also carry out open-ended interviews with patients that complement the RIAS analysis to assess how the peer mentors implement the critical elements and how patients respond.

Quantitative analysis. The voice recordings to be analyzed with the RIAS in Phase 2 are obtained from two sources: (1) 28 voice recordings of peer mentor-patient interactions were obtained in our preliminary work and (2) 160 voice recordings of peer mentor-patient interactions will be obtained in the pilot study proposed in this application. The PI will work with Dr. Roter's laboratory for consultation where coding, reliability testing, and data entry into an SPSS file will occur. We will draw upon our conceptual model as well as incorporate the results of the quantitative and qualitative analysis regarding mediators to construct peer mentor and patient communication profiles which will then be linked to outcomes. Those elements which are associated with outcomes should be the focus of peer mentor training and be integrated into the manual. We will use linear regression to assess the presence, strength, and statistical associations between RIAS categories and outcomes.

Qualitative analysis. The goal of qualitative analysis of the voice recordings will be to enrich the inferences we make using RIAS data to capture the nuances of what is occurring in the peer mentor-patient relationship. The PI and the research assistant will listen to the voice recordings of peer mentor-patient interactions at weeks 1, 4, and 8. They will mark and transcribe sections in the voice recordings that show how interaction elements such as affective exchange take place between the peer mentor and patient, and those sections that provide examples of variations in the methods the peer mentors use and the responses of the patients. The transcriptions will be coded and analyzed for themes. This analysis will contextualize the results of the RIAS analysis and provide knowledge about peer mentor communication and behavior.

Manual development: Knowledge obtained from the study will be developed into a manual which will form the basis for a larger peer mentor study.