[Table, Appendix E: Evidence Table]. - Integration of Mental Health/Substance Abuse and Primary Care

Project Name or 1^st Author, Year, Study Design	Study Aim	Study Period	Patient Population	Settings	Outcomes Measured
Depression Disorders
Fortney, 2007 ^1, ² RCT Randomized by matched site	Assess telemedicine - based collaborative care vs. usual care to improve depression care at small clinics without on-site psychiatrists.	Recruitment 2003. Study period 12 months.	Current VA patients diagnosed with depression. 92% male, 75% white, mean age 59. Control N=218 Intervention N=177	7 rural VA community-based outpatient clinics with no on-site psychiatry or psychology in AK, MS, LA.	Depression symptoms, remission, treatment response, adherence. Physical, mental quality of life, wellbeing, and patient satisfaction. Model fidelity.

PRISM-E (for depression) ³ ^– ⁵ RCT Randomized by patient	Assess integrated vs. enhanced referral care for improving depression outcomes in elderly patients.	Recruitment March 2000 to March 2002. Study period 6 months	Elderly primary care patients: 31% female, 55% non-white, mean age 74 Integrated N=758 Referral N=773	10 practices with 34 urban, suburban, and rural clinics. 5 VA, 3 community health, 2 hospital networks in the Northeast, Miami, and Chicago	Depression symptoms, remission, MH QoL. Program use.

Geron, 2006 ⁶ RCT	Assess social worker care manager vs. usual care for depressed home-dwelling frail elderly	Study period 12 months. Recruitment period not completed.	Current patients over 65 years with 2 or more chronic medical conditions, ER visit or hospital admission in past 6 months	An MCO urban primary care clinic.	Depression symptoms, satisfaction, QoL, adverse health outcomes, physical function, utilization, cost

Grypma, 2006 ⁷ Cohort	Assess adapted version of IMPACT post trial vs. usual care on depression care for adults.	Study period 12 months. IMPACT study period 1999-2001. Post-trial data from 2002-2004.	Current adult patients. 8.4% male, average age 63, 63% above 60 years RCT controls N=116 Post-study intervention N=95	2 Kaiser Permanente practices in San Diego area	Depression symptoms, utilization

IMPACT ⁸ ^– ¹² RCT Randomized by patient	Assess collaborative care vs. usual care on depression care for elderly.	Recruitment July 1999 to August 2000. Intervention 12 months. Study period 2 years.	Current patients 60+ years old with depression. 65% female, 77% white, Control N=895 Intervention N=906	7 national sites in Indiana, Texas, Washington, and California. Rural and urban. Group and academic practices, and VA.	Depression symptoms, treatment response, remission, patient self-efficacy, function and QoL, satisfaction, antidepression medication use, treatment utilization

Clarke, 2005 ¹³ RCT Randomized by patient	Assess collaborative care with CBT vs. usual care for depressed HMO pediatric primary care patients.	Recruitment March 2000 to November 2001. Study period 1 year.	Pediatric patients age 12–18 years old in a current major depression episode. Average age 15, 77% female, 14% non-white. Control N=75 Intervention N=77	HMO pediatric clinic in Portland, OR, part of Kaiser Permanente	Depression symptoms, relapse, QoL, satisfaction, utilization

PROSPECT ¹⁴ ^– ¹⁶ RCT Randomization by matched sites	Assess guideline based depression recognition and treatment program vs. usual care for elderly patients to prevent and reduce suicidal behavior	Recruitment May 1999 to August 2001. Study period 2 years	English speaking patients over 60 years with major depression. 31% above age 75, 72% female, 32% non-white. Control N=278 Intervention N=320	18 clinics in New York, Pennsylvania, and Pittsburgh. Group, university affiliated, and solo practices in urban, suburban, and rural locations.	Depression symptoms, treatment response, and remission, utilization

Pathways ^17, ¹⁸ RCT Randomized by patient	Assess collaborative care vs. usual care for adult diabetes patients with depression	Recruitment April 2001 to May 2002. Intervention 12 months. Study period 2 years	English speaking adult diabetes patients with major depression. Average age 58, 65% female, 81% white. Control N=165 Intervention N-165	9 HMO clinics within 40 mile radius of Seattle.	Depression symptoms, diabetes outcomes and self-care, functional and QoL, adherence and utilization, cost-effectiveness.

RESPECT-D ^19, ²⁰ RCT	Assess evidence-based model of depression management vs. usual care for adult patients with depression	Recruitment February 2002 to February 2003. Patient study period 6 months.	English speaking patients 18 years or older starting treatment for major depression. Average age 42, 80% female, 17% non-white. Control N=181 Intervention N=224	3 medical groups and 2 health plans across U.S., each with at least 10 PC practices and established QI programs. 60 practices, matched and randomized.	Depression symptoms, treatment response, remission, utilization

Simon, 2004 ²¹ RCT Randomized by patient	Assess telephone care management and telephone care management plus psychotherapy vs. usual care for adult patients with depression.	Recruitment November 2000 to May 2002. Study period 6 months.	Adult patients beginning antidepressant treatment. Average age 44, 74% female, 79% white Control N=195 Telephone care N=207 Telephone care + psychotherapy N=198	7 urban and suburban HMO clinics in Washington State.	Depression symptoms, remission, adequate pharmacotherapy.

Adler, 2004 ^22, ²³ RCT Randomized by patient	Assess pharmacist adherence management vs. usual care for adult patients with depression.	Study period 6 months. Recruitment not reported.	English speaking adults with major depression. Average age 42, 72% female, 72% white. Control N=265 Intervention N=268	9 group practice clinics in Boston area, with 5 clinics at an academic medical center.	Depression symptoms, antidepressant utilization and adherence.

Finley, 2004 ²⁴ RCT Randomized by patient	Assess collaborative care with pharmacist care manager vs. usual care for adults with depression.	Study period 6 months. Recruitment not reported	Adult patients beginning antidepressant treatment. Average age 54, 85% female. Control N=50 Intervention N=75	HMO clinic in San Rafael, CA.	Depression symptoms, treatment response, remission, change in disability, adherence and utilization, cost.

Swindle, 2003 ²⁵ RCT Randomized by patient	Assess collaborative care with MH clinical nurse care manager vs. usual care for veterans with depression	Study period 12 months. Recruitment not reported.	Community dwelling adult patients with depression. 97% male, 85% white. Control N=134 Treatment N=134	2 Indianapolis VA clinics, randomized by site.	Depression symptoms, utilization, cost.

Partners in Care ²⁶ ^– ³⁰ RCT Randomized by site	Assess quality improvements in medication management and therapy vs. usual care for adults with depression	Intervention 6 months. Study period 2 years. Recruitment not reported.	English or Spanish speaking adult patients with depression. Average age 44, 71% female, 30% Hispanic. Control N=430 QI Meds N=405 QI Therapy N=464	6 MCOs representing geographically diverse regions in U.S., with 46 clinics.	Depression symptoms, QoL, employment, utilization, overall poor outcome (constructed measure)

Datto, 2003 ³¹ RCT Randomized across sites	Assess telephone-based depression management for acute phase depression vs. usual care for adult patients.	Study period 16 weeks. Recruitment not reported.	Patients with depression. Average age 48, 61% female, 80% white. Control N=31 Intervention N=30	35 urban and suburban clinics in Pennsylvania.	Depression symptoms, QoL, clinician and patient adherence.

Hedrick, 2003 ³² RCT Randomized across sites	Assess collaborative care vs. usual consult-liaison care for VA patients with depression.	Study period 9 months. Recruitment January 1998 to March 1999.	Current patient with major depression, dysthymia, or both. Average age 57, 95% male, 80% white. Control N=186 Intervention N=168	4 clinics in Seattle division of VA-Puget Sound.	Depression symptoms, treatment response, remission, QoL, medication utilization.

Katon, 1995 ³³ RCT Randomized by patient	Assess collaborative care vs. usual care for adult patients with depression.	Study period 12 months. Intervention period up to 9 months. Recruitment not reported.	English speaking, current adult patients beginning antidepressants. Average age 48, 76% female. Control N=109 Intervention N=108	Northgate Medical Center, Group Health Cooperative HMO in western Washington state, a family physician clinic.	Depression symptoms, disability, medication adherence, satisfaction, utilization

Katon, 1999 ³⁴ RCT Randomized by patient	Assess stepped collaborative care vs. usual care for adult patients with depression.	Study period 6 months. Recruitment not reported.	English speaking, current adult patients beginning antidepressants. Average age 47, 75% female, 80% white. Control N=114 Intervention N=114	4 Group Health Cooperative HMO clinics in Seattle area.	Depression symptoms, disability, medication adherence, costs

Katon, 1996 ³⁵ RCT Randomized by patient	Assess collaborative care vs. usual care for adult patients with depression.	Study period 6 months. Recruitment not reported.	English speaking, current adult patients beginning antidepressants. Average age 46, 74% female, 87% white. Control N=76 Intervention N=77	Northgate Medical Center, Group Health Cooperative HMO in western Washington state, a family physician clinic.	Depression symptoms, disability, medication adherence, costs

Katon, 2001 ^36, ³⁷ RCT Randomized by patient	Assess collaborative care vs. usual care for adult patients at risk for depression relapse	Study period 12 months. Recruitment not reported.	English speaking, current adult patients beginning antidepressants. Average age 46, 73% female, 90% white. Control N=192 Intervention N=194	4 Group Health Cooperative HMO clinics in Seattle area.	Depression symptoms, depression relapse, medication adherence

Capoccia 2004 ^38, ³⁹ RCT Randomized by patient	Assess pharmacist based collaborative care vs. usual care for adults with depression	Recruitment from November 1999 to March 2001. Study period 12 months.	English speaking current adult patients beginning antidepressants. Average age 39, 57% female, 22% non-white. Control N=33 Intervention N=41	Academic family practice clinic in Seattle.	Depression symptoms, QoL, medication adherence, utilization, cost.

Tutty, 2000 ⁴⁰ Cohort	Assess telephone counseling and medication monitoring for adult patients with depression.	Study period 6 months. Recruitment not reported.	Adult patients beginning antidepressants. Average age 47, 69% female. Control N=94 Intervention N=28	One Group Health Cooperative clinic in Olympia.	Depression symptoms, treatment response, remission, adequate dosage.

Hunkeler 2000 ⁴¹ RCT Randomized by site	Assess nurse telehealth care vs. usual care for adults with depression.	Study period 6 months. Recruitment not reported.	English speaking adults with SSRI prescription for depression. About 70% female, 37% non-white. Control N=123 Intervention N=179	2 Kaiser Permanente clinics in northern CA.	Depression symptoms, treatment response, QoL, adherence.

QuEST ⁴² ^– ⁴⁴ RCT Randomized by matched site	Assess guideline based depression treatment program vs. usual care for adult patients with depression.	Recruitment from April 1996 to September 1997. Study period 2 years.	English reading current adult patients with depression. Average age 43, 84% female, 16% non-white. Control N=240 Intervention N=239	12 practices across U.S. Urban and rural.	Depression symptoms, QoL, guideline concordant care.

Simon, 2000 ⁴⁵ RCT Randomized by patients	Assess feedback only or feedback plus care management vs. usual care for adult patients with depression.	Study period 6 months. Intervention period 4 months. Recruitment period not reported.	Current adult patients newly prescribed antidepressants. Average age 46 years, 72% female. Control N=196 Feedback only N=221 FB and care mgmt N=196	5 HMO primary care clinics in Washington state	Depression symptoms, treatment response, remission, adequate dosage, cost

Hilty, 2007 ⁴⁶ RCT Randomized by patient	Assess usual care depression management with telepsychiatric and PCP training vs. usual care depression management for adult patients with depression.	Study period 1 year. 2 year recruitment, period not reported.	English speaking current adult patients with depression willing to take antidepressants. Median age 46, 80% female, 10% non-white. Control N=41 Intervention N=52	8 rural primary care clinics, average 140 miles from UC Davis Medical Center.	Depression symptoms, functioning and QoL, satisfaction.

Katzelnick, 2000 ⁴⁷ RCT Randomization across sites by physician practices	Assess depression management vs. usual care for high utilizers with depression, not in active treatment	Study period 12 months. Recruitment period not reported.	Current adult patients above 85^th percentile in utilization for previous 2 years. Average age 45, 77% female, 83% white Control N=189 Intervention N=218	3 HMOs in the Midwest, Northwest and New England regions, 163 primary care practices.	Depression symptoms, treatment response, remission, functioning and QoL, utilization

Anxiety Disorders
Roy-Byrne, 2001 ⁴⁸ RCT Randomized by patient	Assess collaborative care vs. usual care for adult patients with panic disorder.	Study period 12 months. Recruitment not reported.	English speaking adult patients with at least one panic attack in last month. Average age 41, 57% female, 67% white. Control N=58 Intervention N=57	3 urban and suburban group practice clinics in Seattle area, 2 are university associated.	Panic, anxiety, and depression symptoms, treatment response, remission, QoL, appropriate medication and dosage, adherence.

CCAP ^49, ⁵⁰ RCT Randomized by patient, stratified within site	Assess collaborative care vs. usual care for adults with panic disorder.	Recruitment March 2000 to March 2002. Study period 1 year.	English speaking adult patients with at least one panic attack within last week. Average age 41, 67% female, 66% white Control N=113 Intervention N=119	University affiliated primary care clinics in Seattle, San Diego, and Los Angeles	Remission, treatment response, anxiety sensitivity, depression symptoms, QoL and functional disability, utilization

CALM ⁵¹ RCT Randomized across sites	Assess collaborative care vs. usual care for adult patients with anxiety disorders, including GAD, PTSD, PD, and SAD	Study period 18 months. Recruitment not complete	English speaking adult current patients with GAD, PTSD, PD and SAD, N to be1040, 260 at each site	Seattle, WA, Los Angeles and San Diego, CA, and Little Rock, AK	Anxiety disorder symptoms, functioning and QoL, satisfaction, utilization. Design only. No results yet

Rollman, 2005 ^52, ⁵³ RCT Randomized by patients	Assess telephone-based collaborative care vs. usual care for adult anxiety and panic disorder patients.	Recruitment July 2000 to April 2002. Study period 12 months	English speaking, adult current patients with anxiety disorders. Average age 44 years, 81% female, 95% white. Control N=75 Intervention N=116	13 PCPs in Pittsburgh area, urban academic, suburban, and rural.	Anxiety disorder symptoms, depression symptoms, QoL, utilization, employment status

Price, 2000 ⁵⁴ Matched Cohort	Assess integrated care vs. usual care for adult patients with generalized anxiety disorder (GAD) and GAD secondary to depression.	Study period 6 months. Recruitment not reported	English speaking, adult current patients with GAD. Mean age 49 years, 80% female, 86% white. Control N=111 Intervention N=113	Kaiser Permanente clinics in Westminster, CO. Intervention patients family practice, control patients internal medicine	Anxiety symptoms, satisfaction

Other
Katon, 1992 ⁵⁵ RCT Randomized by patients, stratified by physician and blocked	Assess effect of psychiatric consultation vs. usual care for distressed high utilizers of medical care.	Study period 12 months. Recruitment not reported.	Top 10% adult ambulatory care utilizers of appropriate age group with psychiatric distress. Average age 47, 61% female. Control N=127 Intervention N=124	2 primary care clinics of Group Health Cooperative of Puget Sound.	Psychiatric distress, functional disability, utilization, use of and adherence to antidepressants

Epstein, 2007 ⁵⁶ RCT Randomized by pediatricians	Assess collaborative care consultative service for titration and monitoring vs. usual care to improve ADHD care.	Study period 1 year. Recruitment not reported.	1^st through 5^th grade children with ADHD Control N=215 Intervention N=162	12 community-based pediatric practices without onsite psychiatry or psychologist.	ADHD symptoms. Titration trials, medication management, dosage, adherence

PRISM-E (for at-risk alcohol use) ^4, ^5, ⁵⁷ RCT Randomized by patients	Assess integrated vs. enhanced referral care for managing at-risk alcohol use in elderly patients	Recruitment March 2000 to March 2002. Study period 6 months (on-going)	Elderly primary care patients. 92% male, 70% white, mean age 72 Intervention N=280 Referral N=280	9 practices with 34 urban, suburban, and rural clinics. 5 VA, 2 community health, 2 hospital networks in the Northeast, Miami, and Chicago	Drinking severity, MH QoL, Program use.

Backward Integration
Weisner, 2001 ^58, ⁵⁹ RCT Randomized by patients	Assess integrated vs. usual care for medical and substance abuse care	Recruitment April 1997 to December 1998. Study period 6 months	Adult patients admitted to a chemical dependency program. Mean age 37, 55% male, 74% white. Control N=307 Intervention N=285	Kaiser Permanente's Chemical Dependency Recovery Program, southern CA	Abstinence, treatment utilization. No primary care outcomes
Druss, 2001 ⁶⁰ RCT Randomized by patients Willenbring, 1999 ⁶¹ RCT Randomized by patients	Assess integrated medical health care vs. usual care for patients with serious mental illness	Study period 12 months. Recruitment not reported.	VA mental health patients without a current primary care provider. Mean age 45, 99% male, 70% white Control N=61 Intervention N=59	West Haven, CT, VAMC	Utilization, quality of preventive care, satisfaction, physical and mental health status, costs
Assess integrated outpatient treatment vs. usual care for alcohol-related medically ill alcohol abuse patients	Study period 2 years. Recruitment period not reported.	VA patients with current alcohol abuse behavior and alcohol-related medical illness. Mean age=55.1 Control N=53 Intervention N=48	Minneapolis, MN VA medical center	Drinking severity, quality of life, utilization

From: Appendix E: Evidence Table

Integration of Mental Health/Substance Abuse and Primary Care.

Evidence Reports/Technology Assessments, No. 173.

Butler M, Kane RL, McAlpine D, et al.

Rockville (MD): Agency for Healthcare Research and Quality (US); 2008 Oct.

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.