Burnout was assessed with the Maslach Burnout Inventory (MBI), a widely-used measure of three components of burnout: emotional exhaustion, depersonalization, and personal accomplishment. The survey contains 22 statements of job-related feelings and staff were asked to read each statement and decide if they ever felt that way about their job. The item scores were averaged. Scale: 0 (Never), 1 (A few times a year or less), 2 (Once a month or less), 3 (A few times a month), 4 (Once a week), 5 (A few times a week), 6 (Every Day).

Maslach C, Jackson, S. E., Leiter, M. P. Maslach Burnout Inventory Manual. 3 ed. Palo Alto, California: Consulting Psychologists Press; 1996.

The number analyzed at each time point differs from the overall number enrolled due to participant drop out and/or missing data.

Burnout was assessed with the Maslach Burnout Inventory (MBI), a widely-used measure of three components of burnout: emotional exhaustion, depersonalization, and personal accomplishment. The survey contains 22 statements of job-related feelings and staff were ased to read each statement and decide if they ever felt that way about their job. The item scores were averaged. Scale: 0 (Never), 1 (A few times a year or less), 2 (Once a month or less), 3 (A few times a month), 4 (Once a week), 5 (A few times a week), 6 (Every Day).

Maslach C, Jackson, S. E., Leiter, M. P. Maslach Burnout Inventory Manual. 3 ed. Palo Alto, California: Consulting Psychologists Press; 1996.

The number analyzed at each time point differs from the overall number due to participant drop out and/or missing data.

Burnout was assessed with the Maslach Burnout Inventory (MBI), a widely-used measure of three components of burnout: emotional exhaustion, depersonalization, and personal accomplishment. The survey contains 22 statements of job-related feelings and staff were ased to read each statement and decide if they ever felt that way about their job. The item scores were averaged. Scale: 0 (Never), 1 (A few times a year or less), 2 (Once a month or less), 3 (A few times a month), 4 (Once a week), 5 (A few times a week), 6 (Every Day).

Maslach C, Jackson, S. E., Leiter, M. P. Maslach Burnout Inventory Manual. 3 ed. Palo Alto, California: Consulting Psychologists Press; 1996.

The number analyzed at each time point differs from the overall number enrolled due to participant drop out and/or missing data.

Job satisfaction was assessed with one item from the Job Diagnostics Survey: Overall, I am satisfied with my job. Scale: 1 (Strongly Disagree) to 7 (Strongly Agree)

Hackman JR, Oldham GR. The Job Diagnostic Survey: An Instrument for the Diagnosis of Jobs and the Evaluation of Job Redesign Projects. 1974.

The number analyzed at each time point differs from the overall number enrolled due to participant drop out and/or missing data.

This is the first of two questions in which staff were asked about turnover intentions. Staff were asked, "How often have you seriously considered leaving your job in the past six months?" Scale: 1 (Never), 2 (Once every few months), 3 (Once a month), 4 (several times a month), 5 (Once a week), 6 (Several times a week)

The number analyzed at each time point differs from the overall number enrolled due to participant drop out and/or missing data.

This is the second of two questions in which staff were asked about turnover intentions. Staff were asked, "How likely are you to leave your job in the next six months?" Scale: 1 (Not likely at all), 2 (Not very likely), 3 (Somewhat likely), 4 (Very likely)

The number analyzed at each time point differs from the overall number enrolled due to participant drop out and/or missing data.

Work-Life Balance was assessed with a six-item measure adapted from an 18-item measure developed by Carlson et al. The measure assesses three types (time-, strain-, and behavior-based) and two directions (work conflict with family and family conflict with work) of balance. The outcome described here is work conflict with family. The measure consists of a series of statements regarding one's work and family situation, to which participants are asked to indicate their level of agreement or disagreement on a 5-point Likert-type scale: 1 (Strongly disagree) to 5 (Strongly agree). The item scores were averaged.

Carlson DS, Kacmar KM, Williams LJ. Construction and initial validation of a multidimensional measure of work-family conflict. Journal of Vocational Behavior. 2000;56(2):249-276.

The number analyzed at each time point differs from the overall number enrolled due to participant drop out and/or missing data.

Work-Life Balance was assessed with a six-item measure adapted from an 18-item measure developed by Carlson et al. The measure assesses three types (time-, strain-, and behavior-based) and two directions (work conflict with family and family conflict with work) of balance. The outcome described here is family conflict with work. The measure consists of a series of statements regarding one's work and family situation, to which participants are asked to indicate their level of agreement or disagreement on a 5-point Likert-type scale: 1 (Strongly disagree) to 5 (Strongly agree). The item scores were averaged.

Carlson DS, Kacmar KM, Williams LJ. Construction and initial validation of a multidimensional measure of work-family conflict. Journal of Vocational Behavior. 2000;56(2):249-276.

The number analyzed at each time point differs from the overall number enrolled due to participant drop out and/or missing data.

The Emotional Labor Scale includes 14 questions regarding the relationship between emotions and interactions with clients. Surface Acting is a subset of these questions (e.g., I put on an act in order to deal with clients in an appropriate way). The item scores were averaged. Scale: 1 (strongly disagree) to 5 (strongly agree)

The number analyzed at each time point differs from the overall number enrolled due to participant drop out and/or missing data.

The Emotional Labor Scale includes 14 questions regarding the relationship between emotions and interactions with clients. Deep Acting is a subset of these questions (e.g., I try to actually experience the emotions that I must show to clients). The item scores were averaged. Scale: 1 (strongly disagree) to 5 (strongly agree)

The number analyzed at each time point differs from the overall number enrolled due to participant drop out and/or missing data.

The Emotional Labor Scale includes 14 questions regarding the relationship between emotions and interactions with clients. Genuine Emotions is a subset of these questions (e.g., The emotions that I express to clients are genuine). The item scores were averaged. Scale: 1 (strongly disagree) to 5 (strongly agree)

The number analyzed at each time point differs from the overall number enrolled due to participant drop out and/or missing data.

Staff were asked, "How important is it for you to reduce your work-related stress right now?" This single item score was averaged. Scale: 1 (not at all important) to 10 (extremely important)

The number analyzed at each time point differs from the overall number enrolled due to participant drop out and/or missing data.

Staff were asked, "How confident are you that you can reduce your work-related stress in your life?" Scale: 1 (not at all confident) to 10 (extremely confident)

The number analyzed at each time point differs from the overall number enrolled due to participant drop out and/or missing data.

Staff were asked, "How important is it for you to consistently interact with consumers/clients in a relaxed, non-judgmental way?" Scale: 1 (not at all important) to 10 (extremely important)

The number analyzed at each time point differs from the overall number enrolled due to participant drop out and/or missing data.

Staff were asked, "How confident are you that you can consistently interact with consumers/clients in a relaxed, non-judgmental way?" Scale: 1 (not at all confident) to 10 (extremely confident)

The number analyzed at each time point differs from the overall number enrolled due to participant drop out and/or missing data.

Perceived Quality of Care was assessed with a 31 item scale developed with one of the mental health agencies as part of this project. Person Centered Care was measured with a subset of questions from this scale (e.g., I felt like I was able to really show compassion to a client.). The item scores were averaged. Scale: 0 (never) to 5 (always)

The number analyzed at each time point differs from the overall number enrolled due to participant drop out and/or missing data.

Perceived Quality of Care was assessed with a 31 item scale developed with one of the mental health agencies as part of this project. Discordant Care was measured with a subset of questions from this scale (e.g., I had conflicts with clients.). The item scores were averaged. Scale: 0 (never) to 5 (always)

The number analyzed at each time point differs from the overall number enrolled due to participant drop out and/or missing data.

Perceived Quality of Care was assessed with a 31 item scale developed with one of the mental health agencies participating in this project and then refined to 22 items through data collected and analyzed in this study. Items were related to person or client centered care, work conscientiousness, errors, interactions with clients, and how stress affects client interactions or outcomes. The item scores were averaged. Scale: 0 (never) to 5 (always)

The number analyzed at each time point differs from the overall number enrolled due to participant drop out and/or missing data.

The 19 item Perceptions of Supervisory Support Scale was used to gather information on staff's experience of interactions with their supervisors (e.g., How often did you think supervision improved your relationship with your supervisor?). The item scores were averaged. Scale: 1 (never) to 6 (always)

The number analyzed at each time point differs from the overall number enrolled due to participant drop out and/or missing data.

Number of staff participants who separated from their respective agency before their anticipated study completion date. The mental health agencies provided separation dates, if applicable, for staff study participants.

Hope was assessed with clients using the 12-item Adult State Hope Scale (e.g., I can think of many ways to get the things in life that are most important to me.). The item scores were averaged. Scale: 1 (Definitely False) to 8 (Definitely True)

Snyder CR, Sympson SC, Ybasco FC, Borders TF, Babyak MA, Higgins RL. Development and validation of the State Hope Scale. Journal of Personality and Social Psychology. 1996;70(2):321 - 335.

The number analyzed at each time point differs from the overall number enrolled due to participant drop out and/or missing data.

Medication adherence (for clients who are prescribed medications for their mental health) was rated with a subset of 4 items from the MARS, a 10-item scale assessing attitudes toward medication (e.g., Do you ever forget to take your medication? Are you careless at times about taking your medicine?). The item scores were summed and averaged (range: 0-4). Scale: 0 (No) to 1 (Yes)

Thompson K, Kulkarni J, Sergejew AA. Reliability and validity of a new Medication Adherence Rating Scale (MARS) for the psychoses. Schizophrenia Research. May 5 2000;42(3):241-247.

The number analyzed at each time point differs from the overall number enrolled due to participant drop out, clients reporting they are not prescribed medications for their mental health, and/or missing data.

Medication attitudes (for clients who are prescribed medications for their mental health) was rated with the MARS, a 10-item scale assessing attitudes toward medication (e.g., It is unnatural for my mind and body to be controlled by medication.). The items scores were summed and averaged (range: 0-10). Scale: 0 (No) to 1 (Yes)

Thompson K, Kulkarni J, Sergejew AA. Reliability and validity of a new Medication Adherence Rating Scale (MARS) for the psychoses. Schizophrenia Research. May 5 2000;42(3):241-247.

The number analyzed at each time point differs from the overall number enrolled due to participant drop out, clients who reported they are not prescribed medications for their mental health, and/or missing data.

Perceived autonomy support was assessed with this 15-item scale (e.g., I am able to be open with [name] at our meetings.). Clients were prompted to report on the specific clinician from whose caseload they were randomly selected. The item scores were averaged. Scale: 1 (Strongly Disagree) to 7 (Strongly Agree)

Williams GC, McGregor HA, King D, Nelson CC, Glasgow RE. Variation in perceived competence, glycemic control, and patient satisfaction: relationship to autonomy support from physicians. Patient Education & Counseling. Apr 2005;57(1):39-45.

The number analyzed at each time point differs from the overall number enrolled due to participant drop out and/or missing data. Some clients did not recall the clinician being asked about and/or discontinued treatment with that clinician during their study participation.

Perceived relatedness was assessed with this short form of the patient version of the WAI and is 12 items (e.g., We agree on what is important for me to work on.). Clients were prompted to report on the specific clinician from whose caseload they were randomly selected. The item scores were averaged. Scale: 1 (Never) to 7 (Always)

Tracey TJ, Kokotovic AM. Factor structure of the Working Alliance Inventory. Psychological Assessment: A Journal of Consulting and Clinical Psychology. 1989;1(3):207.

The number analyzed at each time point differs from the overall number enrolled due to participant drop out and/or missing data. Some clients did not recall the clinician being asked about and/or discontinued treatment with that clinician during their study participation.

Perceived relatedness was assessed with this short form of the patient version of the WAI and has 12 items in total. This outcome is for the tasks subscale. Clients were prompted to report on the specific clinician from whose caseload they were randomly selected. The item scores were summed and averaged (range: 4-28). Scale: 1 (Never) to 7 (Always)

Tracey TJ, Kokotovic AM. Factor structure of the Working Alliance Inventory. Psychological Assessment: A Journal of Consulting and Clinical Psychology. 1989;1(3):207.

The number analyzed at each time point differs from the overall number enrolled due to participant drop out and/or missing data. Some clients did not recall the clinician being asked about and/or discontinued treatment with that clinician during their study participation.

Perceived relatedness was assessed with this short form of the patient version of the WAI and has 12 items in total. This outcome is for the goals subscale. Clients were prompted to report on the specific clinician from whose caseload they were randomly selected. The items scores were summed and averaged (range: 4-28). Scale: 1 (Never) to 7 (Always)

Tracey TJ, Kokotovic AM. Factor structure of the Working Alliance Inventory. Psychological Assessment: A Journal of Consulting and Clinical Psychology. 1989;1(3):207.

The number analyzed at each time point differs from the overall number enrolled due to participant drop out and/or missing data. Some clients did not recall the clinician being asked about and/or discontinued treatment with that clinician during their study participation.

Perceived relatedness was assessed with this short form of the patient version of the WAI and has 12 items in total. This outcome is for the bonds subscale. Clients were prompted to report on the specific clinician from whose caseload they were randomly selected. The items scores were summed and averaged (range: 4-28). Scale: 1 (Never) to 7 (Always)

Tracey TJ, Kokotovic AM. Factor structure of the Working Alliance Inventory. Psychological Assessment: A Journal of Consulting and Clinical Psychology. 1989;1(3):207.

The number analyzed at each time point differs from the overall number enrolled due to participant drop out and/or missing data. Some clients did not recall the clinician being asked about and/or discontinued treatment with that clinician during their study participation.

Competence related to mental health management was assessed with the 13-item Patient Activation Measure-Mental Health (PAM-MH) (e.g., I know what each of my prescribed mental health medications does.). Each question was answered on a 4-point Likert-type scale: 1 (Strongly Disagree) to 4 (Strongly Agree). Higher scores=greater activation.

Hibbard JH, Mahoney ER, Stockard J, Tusler M. Development and testing of a short form of the patient activation measure. Health Services Research. Dec 2005;40(6 Pt 1):1918-1930.

The number analyzed at each time point differs from the overall number enrolled due to participant drop out and/or missing data.

Physical and mental health functioning was assessed with the Short Form 12-Item Health Survey (SF-12). The SF-12 is a health-related quality of life measure, derived from the 36-item Medical Outcomes Study survey and containing items yielding a Mental Health Component Score and a Physical Health Component Score. Higher composite scores indicate higher health-related quality of life. Items are weighted and then transformed into norm-based scores (range: 0-100).

Ware JE, Jr. , Kosinski M, Keller SD. A 12-item short-form health survey: Construction of scales and preliminary tests of reliability and validity. Medical Care. 1996;34(3):220 -233.

The number analyzed at each time point differs from the overall number enrolled due to participant drop out and/or missing data.

Physical and mental health functioning was assessed with the Short Form 12-Item Health Survey (SF-12). The SF-12 is a health-related quality of life measure, derived from the 36-item Medical Outcomes Study survey and containing items yielding a Mental Health Component Score and a Physical Health Component Score. Higher composite scores indicate higher health-related quality of life. Items are weighted and then transformed into norm-based scores (range: 0-100).

Ware JE, Jr. , Kosinski M, Keller SD. A 12-item short-form health survey: Construction of scales and preliminary tests of reliability and validity. Medical Care. 1996;34(3):220 -233.

The number analyzed at each time point differs from the overall number enrolled due to participant drop out and/or missing data.

The PHQ-9 is a brief, self-report assessment. It provides a summed total score that indicates likelihood of major depressive disorder. Scores ≥10 are considered a positive screen (sensitivity 88%, specificity 88%) and also effectively measures response to treatment (<5 indicate remission, of 5-9 indicate partial response, and ≥10 indicates no response). Item scores are summed and averaged (range: 0-27). Scale: 0 (Not at all), 1 (Several days), 2 (More than half the days), 4 (Nearly every day). When problems are identified, the difficulty of those problems are rated on 4 point scale (Not difficult at all to Extremely difficult).

Kroenke K, Spitzer RL, Williams JB. The PHQ-9: validity of a brief depression severity measure. Journal of General Internal Medicine. Sep 2001;16(9):606-613.

American Psychiatric Association. Diagnostic and statistical manual of mental disorders - Text Revision (4th ed.). Washington, DC: American Psychiatric Association; 2000.

The number analyzed at each time point differs from the overall number enrolled due to participant drop out and/or missing data.

Anxiety was assessed with the 7-item Generalized Anxiety Disorder (GAD-7). It can be scored continuously on a 0-21 severity scale and cutpoints have been established for estimating the probability of the 4 most common and clinically relevant anxiety disorders - generalized anxiety disorder, panic disorder, post-traumatic stress disorder, and social anxiety disorder. Scale: 0 (not at all), 1 (several days), 2 (more than half the days), 3 (nearly every day)

Spitzer RL, Kroenke K, Williams JB, Lowe B. A brief measure for assessing generalized anxiety disorder: the GAD-7. Archives of Internal Medicine. May 22 2006;166(10):1092-1097.

Kroenke K, Spitzer RL, Williams JB, Monahan PO, Lowe B. Anxiety disorders in primary care: prevalence, impairment, comorbidity, and detection. Ann Intern Med. Mar 6 2007;146(5):317-325.

The number analyzed at each time point differs from the overall number enrolled due to participant drop out and/or missing data.

Engagement was assessed with patient satisfaction using the Client Satisfaction Questionnaire, an 8-item satisfaction checklist (e.g., How would you rate the quality of service you have received? and, If a friend were in need of similar help, would you recommend [name of agency] to him or her?). The item scores were averaged. Scale: 1 to 4 with response text dependent upon the question (e.g., 1-Poor to 4-Excellent, 1-No, definitely not to 4-Yes, definitely, or 1-Quite dissatisfied to 4-Very satisfied).

The number analyzed at each time point differs from the overall number enrolled due to participant drop out and/or missing data.

Perceived Quality of Care was assessed with a 31 item scale (e.g., Staff spent extra time with me when I needed them.). This scale for clients was adapted from the one developed for staff. Person Centered Care was measured with a subset of questions from this scale. The item scores were averaged. Scale: 0 (never) to 5 (always)

The number analyzed at each time point differs from the overall number enrolled due to participant drop out and/or missing data.

Perceived Quality of Care was assessed with a 31 item scale (e.g., Staff spent extra time with me when I needed them.). This scale for clients was adapted from the one developed for staff. Negative Interactions were measured with a subset of questions from this scale. Item scores were averaged. Scale: 0 (never) to 5 (always)

The number analyzed at each time point differs from the overall number enrolled due to participant drop out and/or missing data.

Perceived Quality of Care was assessed with a 31 item scale (e.g., Staff spent extra time with me when I needed them.). This scale for clients was adapted from the one developed for staff. Inattentive care was measured with a subset of questions from this scale. Item scores were averaged. Scale: 0 (never) to 5 (always)

The number analyzed at each time point differs from the overall number enrolled due to participant drop out and/or missing data.

Perceived Quality of Care was assessed with a 31 item scale (e.g., Staff spent extra time with me when I needed them.) and then refined to 22 items through data collected and analyzed in this study. This scale for clients was adapted from the one developed for staff. Item scores were averaged. Scale: 0 (never) to 5 (always)

The number analyzed at each time point differs from the overall number enrolled due to participant drop out and/or missing data.

Patient engagement was assessed by the proportion of missed appointments (when the client cancelled or did not show for a scheduled appointment divided by the total scheduled). This data was retrieved from client medical records at the agencies. Data from 3 time periods were analyzed (6 months prior to baseline through baseline, baseline to 6 months, and 6 months to 12 months). The below table illustrates the missed appointments for each time period.

The number analyzed at each time point differs from the overall number enrolled due to missing data (medical records not available) and clients discontinuing treatment at their respective agency.