Skip to content

The Effectiveness of a Decision-Support Tool for Adult Consumers With Mental Health Needs and Their Care Managers

Researching the Effectiveness of a Decision-Support Tool for Adult Consumers With Mental Health Needs and Their Care Managers

Status
Completed
Phases
NA
Study type
Interventional
Source
ClinicalTrials.gov
Registry ID
NCT02761733
Enrollment
240
Registered
2016-05-04
Start date
2013-09-30
Completion date
2016-10-31
Last updated
2017-04-26

For informational purposes only — not medical advice. Sourced from public registries and may not reflect the latest updates. Terms

Conditions

Mental Disorders

Keywords

Shared Decision Making, Case Management, Decision Aid

Brief summary

The primary goal of the proposed study is to investigate the implementation and effectiveness of the mPOWR (Moving Patient-centered Outcomes through Wellness and Recovery) in diverse urban and rural community mental health settings. The study compares patient participation and outcomes using the mPOWR system to a usual care control condition. Four community mental health agencies participate in the research: two in San Francisco (urban) and two in N.M. (rural). One site in each setting serves as the mPOWR implementation site and the other serves as the control site. Service sites were randomly assigned to intervention or usual care conditions. A quasi experimental design was used; only eligible participants were enrolled in the study (e.g., exclusion criteria of moderate to severe cognitive impairment, patient services structured for provision of mPOWR implementation, etc.). The study design will employ repeated quantitative measures to assess change in outcomes within and across conditions over time. Qualitative methods in the form of focus group interviews will also be used to round out the information obtained about patient and provider expectations and experiences. Primary outcomes of interest include: Short Form Health Survey-12 (SF-12; physical and mental health aspects of health and well-being); Outcome Rating Scale (ORS; general well-being, personal well-being, close relationships, and work/school/friend relationships); Shared Decision Making Questionnaire (congruence of patient's and provider's participation in therapeutic decision making and patient's understanding of treatment and treatment options); Working Alliance Inventory (perception of therapeutic alliance); and Satisfaction Questionnaire (communication patterns between physicians and their patients).

Detailed description

Research on the impact of decision-support tools for providers and patients has not included frontline care managers within community mental health settings. In particular, the care management field lacks effective models for integrating results-based and patient-centered care into routine practice with diverse adults with complex mental health and social service needs. For over five years, Family Service Agency of San Francisco (FSA) developed and piloted a mental health decision-support tool entitled the Assessment Diagnostic Evaluation and Planning Tool (ADEPT) for use by patients and front-line providers. The ADEPT was developed by a team of researchers, frontline providers, and FSA patients to collect data and track patient progress over time. Yet, its strong focus on diagnostic screening and excessive length limited its relevance to patients in guiding service decisions. Over the past two years, a team of FSA patients and care managers revised the tool, eliminating its diagnostic component and retaining two measures of quality of life and community living skills that were reviewed and/or adapted by patients to reflect meaningful outcomes in the delivery of services. Decision aids and a communications toolkit were also developed to support the use of shared-decision making (SDM) processes in the provision of services. Together, these instruments and tools make up the Moving Patient Outcomes toward Wellness and Recovery (mPOWR) system. mPOWR ensures that the patient-provider relationship remains focused on patient identified outcomes and use of SDM throughout care. The primary goal of the study is to investigate the implementation and effectiveness of the mPOWR in diverse urban and rural community mental health settings. FSA has partnered with senior faculty from the University of New Mexico to serve as independent researchers for this study. The study involves four community mental health agencies: two in San Francisco (urban) and two in N.M. (rural). One (randomly chosen) site in each setting serves as the mPOWR implementation site and the other serves as the control site. The study population initially enrolled included 240 (60/site X 4 sites) diverse adults with chronic mental health conditions. The specific aims for this study are: (1) To improve patient and provider participation in shared-decision making and engagement in mental health treatment, to improve \[patient\] personal Quality of Life, and to improve \[patient\] access to community/social services; (2) To increase patient understanding of their treatment and of treatment options; to increase their personal treatment progress; (3) To increase patient functionality and sense of perceived support for their therapeutic outcomes. Primary outcomes of interest include: Short Form Health Survey-12 (SF-12; physical and mental health aspects of health and well-being); Outcome Rating Scale (ORS; general well-being, personal well-being, close relationships, and work/school/friend relationships); Shared Decision Making Questionnaire (congruence of patient's and provider's participation in therapeutic decision making and patient's understanding of treatment and treatment options); Working Alliance Inventory (perception of therapeutic alliance); and Satisfaction Questionnaire (communication patterns between physicians and their patients). Qualitative methods in the form of focus group interviews will also be used to round out the information obtained about patient and provider expectations and experiences. Data will be collected in a time series design.

Interventions

BEHAVIORALMoving Patient Outcomes toward Wellness and Recovery (mPOWR)

An mPOWR intervention and toolkit that consists of strategies, prompts, and decision aids designed to foster patient involvement in treatment. The mPOWR intervention includes a questionnaire that assesses quality of life and consumer functioning for use in the development of consumer-centered goals and outcomes within the mPOWR intervention.

Sponsors

Patient-Centered Outcomes Research Institute
CollaboratorOTHER
University of New Mexico
CollaboratorOTHER
Felton Institute
Lead SponsorOTHER

Study design

Allocation
NON_RANDOMIZED
Intervention model
PARALLEL
Primary purpose
SUPPORTIVE_CARE
Masking
NONE

Eligibility

Sex/Gender
ALL
Age
18 Years to No maximum
Healthy volunteers
No

Inclusion criteria

* Men and women over the age of 18 who are receiving mental health services.

Exclusion criteria

Temporary Exclusion: * Disruptive, aggressive, or severely disorganized behaviors; * Visibly intoxicated or under the influence of illicit drugs. General Exclusion: * Moderate to severe cognitive impairment; * Developmental disability that precludes comprehension; * Language issues: Intervention is only offered in English and Chinese; * Patient services structured for provision of mPOWR implementation (e.g., not in and out of inpatient or crisis stabilization services).

Design outcomes

Primary

MeasureTime frameDescription
Short Form Health Survey-12 (SF-12), Physical Symptoms SubscaleChange in scores on the SF-12 from Baseline to 24 month follow-upThe physical health subscale (PCS-12; Physical Component Summary) of the SF-12 (Health Survey Short Form-12) was utilized in the current study to assess physical aspects of health and well-being48. The measure includes twelve questions asking about overall health, limitations from health conditions, physical health, emotional well-being and daily activities, and feelings over the past four weeks, utilizing variable Likert scale response choice options. The aggregate PCS subscale score of the SF-12 is calculated utilizing norm-based scoring with a weighted sum (Ware, Kosinski, & Keller, 1995). PCS scores in the present study ranged from 13.2 to 65.6, with higher values indicating better physical health.
Short Form Health Survey-12 (SF-12) Mental SymptomsChange in scores on the SF-12 from Baseline to 24 month follow-up Description: The Health Survey Short Form-12 (SF-12) includes 12 items that assess for physical and mental aspects of health and well-being.The mental health subscale (MCS-12; Mental Component Summary) of the SF-12 (Health Survey Short Form-12) was utilized in the current study to assess mental aspects of health and well-being48. The measure includes twelve questions asking about overall health, limitations from health conditions, physical health, emotional well-being and daily activities, and feelings over the past four weeks, utilizing variable Likert scale response choice options. The physical health subscale of the SF-12 was utilized as a key client functioning outcome in the current study. The aggregate MCS subscale score of the SF-12 is calculated utilizing norm-based scoring with a weighted sum (Ware, Kosinski, & Keller, 1995). MCS scores in the present study ranged from 9.6 to 72.0, with higher values indicating better physical health.
Outcome Rating Scale (ORS)Change in scores on the Outcome Rating Scale from Baseline to 24 month follow-upThe Outcome Rating Scale (ORS) was utilized as a repeated measure of general therapy outcomes and quality of life changes during the course of therapy. The Outcome Rating Scale includes a visual analog scale (a horizontal line on which the participants marks how well they are doing within the last week from low to high) that records four questions about general well-being, personal well-being, close relationships, and work/school/friend relationships. Physical marks for each of four domains on the visual analog scale are measured by research team members with a ruler and converted to a score from 1 to 100. The four items are then averaged for an overall therapy outcome score. The total averaged ORS score ranges from 1 to 100, with higher scores indicating a better outcome. Analyses will examine treatment progress via change in ORS scores from pre- to post-intervention.
Shared Decision Making QuestionnaireChange in scores on the Shared Decision Making Questionnaire from Baseline to 24 month follow-upA 6-item modified version of the Shared Decision Making Questionnaire (SDM-Q-9) 46 was utilized to assess client reports about the degree to which their provider involved them in understanding and making a treatment decision. Examples items included My provider discussed the advantages and disadvantages of options and strategies or My provider helped me understand all the information measured on a 6-point scale (completely disagree, strongly disagree, somewhat disagree, somewhat agree, strongly agree, and completely agree). The total average score ranges from 1 to 6 with higher scores indicating a better outcome of greater shared decision making. Analyses will examine change in Shared Decision Making Questionnaire scores from pre- to post-intervention.
Working Alliance InventoryChange in scores on the Working Alliance Inventory from Baseline to 24 month follow-upThe Working Alliance Inventory measures the perception of therapeutic alliance in a clinical dyad during the process of developing a relationship required for effective psychotherapy. The current study utilized the client version of the Working Alliance Inventory included 7 items measured on a 7-point scale (never, rarely, occasionally, sometimes, often, very often, always). Example items included I am confident in my provider's ability to help me and My provider and I trust one another. The Working Alliance Inventory total average score ranges from 1 to 7, with high scores indicating a more positive outcome. Analyses will examine change in patient-reported Working Alliance Inventory scores from pre- to post-intervention.
Communication Satisfaction QuestionnaireChange in scores on the Communication Satisfaction Questionnaire from Baseline to 24 month follow-upCommunication satisfaction was measured utilizing a modified version of a 19-item measure of communication patterns between physicians and their clients (Campbell et al., 2007). Thirteen items focusing on the client's satisfaction with communication with their provider and their engagement in treatment were measured on a 7-point scale (strongly agree, agree, agree somewhat, undecided, disagree somewhat, disagree, strongly disagree). Example items included My provider checks to be sure that I understand everything or My provider involves me in decisions as much as I want. Total average scores range from 1 to 7 with higher scores indicating better communication satisfaction. Analyses will examine change in Satisfaction Questionnaire scores from pre- to post-intervention.

Participant flow

Recruitment details

The research team distributed recruitment letters (in English and Chinese) via mail and in person in all four research sites. With assistance from case managers, research assistants introduced themselves to patients and invited participation. Recruitment occurred from October 2013 to June 2014.

Pre-assignment details

Enrollment was not directly related to the assignment of intervention, which was randomized at the site level. The screening predetermined criteria (exclusion of cognitive impairment). The number of individuals found eligible to participate was 266, and the research team subsequently randomly selected 240 participants.

Participants by arm

ArmCount
Intervention Site - Urban Setting
The clinical case mangers at this site were trained to implement the intervention. This site is based in inner city San Francisco, CA. (Geriatric Services West)
60
Control Site - Urban Setting
Treatment as usual. This site is based in inner city San Francisco, CA. (Gough Street Clinic)
60
Intervention Site - Rural Setting
The clinical case mangers at this site were trained to implement the intervention. This site is based in rural New Mexico. (Tri-County Community Services in Taos, NM)
60
Control Site - Rural Setting
Treatment as usual. This site is based in rural New Mexico. (Mental Health Resources in Clovis, NM)
60
Total240

Withdrawals & dropouts

PeriodReasonFG000FG001FG002FG003
Overall StudyCognitive Impairment5010
Overall StudyDeath3511
Overall StudyDischarged from Study Site1092511
Overall StudyLost to Follow-up6703
Overall StudyPhysician Decision4000
Overall StudyWithdrawal by Subject2000

Baseline characteristics

CharacteristicIntervention Site - Urban SettingControl Site - Urban SettingIntervention Site - Rural SettingControl Site - Rural SettingTotal
Age, Categorical
<=18 years
0 Participants0 Participants0 Participants0 Participants0 Participants
Age, Categorical
>=65 years
47 Participants41 Participants6 Participants8 Participants102 Participants
Age, Categorical
Between 18 and 65 years
13 Participants19 Participants54 Participants50 Participants136 Participants
Race/Ethnicity, Customized
Race/Ethnicity
American Indian or Alaskan Native
0 Participants1 Participants1 Participants0 Participants2 Participants
Race/Ethnicity, Customized
Race/Ethnicity
Asian or Pacific Islander
16 Participants5 Participants0 Participants0 Participants21 Participants
Race/Ethnicity, Customized
Race/Ethnicity
Black, not Hispanic
6 Participants10 Participants0 Participants0 Participants16 Participants
Race/Ethnicity, Customized
Race/Ethnicity
Hispanic
5 Participants4 Participants32 Participants20 Participants61 Participants
Race/Ethnicity, Customized
Race/Ethnicity
Mixed
0 Participants1 Participants0 Participants0 Participants1 Participants
Race/Ethnicity, Customized
Race/Ethnicity
Other
3 Participants1 Participants1 Participants0 Participants5 Participants
Race/Ethnicity, Customized
Race/Ethnicity
Unknown
0 Participants0 Participants1 Participants3 Participants4 Participants
Race/Ethnicity, Customized
Race/Ethnicity
White, not Hispanic
30 Participants38 Participants25 Participants37 Participants130 Participants
Region of Enrollment
United States
60 Participants60 Participants60 Participants60 Participants240 Participants
Sex/Gender, Customized
Female
34 Participants30 Participants28 Participants35 Participants127 Participants
Sex/Gender, Customized
Male
26 Participants29 Participants32 Participants25 Participants112 Participants
Sex/Gender, Customized
Other
0 Participants1 Participants0 Participants0 Participants1 Participants

Adverse events

Event typeEG000
affected / at risk
EG001
affected / at risk
EG002
affected / at risk
EG003
affected / at risk
deaths
Total, all-cause mortality
— / —— / —— / —— / —
other
Total, other adverse events
0 / 600 / 600 / 600 / 60
serious
Total, serious adverse events
3 / 605 / 601 / 601 / 60

Outcome results

Primary

Communication Satisfaction Questionnaire

Communication satisfaction was measured utilizing a modified version of a 19-item measure of communication patterns between physicians and their clients (Campbell et al., 2007). Thirteen items focusing on the client's satisfaction with communication with their provider and their engagement in treatment were measured on a 7-point scale (strongly agree, agree, agree somewhat, undecided, disagree somewhat, disagree, strongly disagree). Example items included My provider checks to be sure that I understand everything or My provider involves me in decisions as much as I want. Total average scores range from 1 to 7 with higher scores indicating better communication satisfaction. Analyses will examine change in Satisfaction Questionnaire scores from pre- to post-intervention.

Time frame: Change in scores on the Communication Satisfaction Questionnaire from Baseline to 24 month follow-up

Population: Several factors influenced variability in the number of participants analyzed across time points. First, patients may have responded to some items but opted out of others because of questionnaire length or comprehension considerations. Second, some patients were available at baseline but were unavailable at subsequent time points, or vice versa.

ArmMeasureGroupValue (MEAN)Dispersion
Intervention Site - Urban SettingCommunication Satisfaction QuestionnaireBaseline4.283 units on a scaleStandard Deviation 0.53
Intervention Site - Urban SettingCommunication Satisfaction Questionnaire24 Month Follow-up3.947 units on a scaleStandard Deviation 0.478
Control Site - Urban SettingCommunication Satisfaction Questionnaire24 Month Follow-up4.093 units on a scaleStandard Deviation 0.529
Control Site - Urban SettingCommunication Satisfaction QuestionnaireBaseline4.024 units on a scaleStandard Deviation 0.706
Intervention Site - Rural SettingCommunication Satisfaction QuestionnaireBaseline4.363 units on a scaleStandard Deviation 0.564
Intervention Site - Rural SettingCommunication Satisfaction Questionnaire24 Month Follow-up4.135 units on a scaleStandard Deviation 0.769
Control Site - Rural SettingCommunication Satisfaction QuestionnaireBaseline4.473 units on a scaleStandard Deviation 0.578
Control Site - Rural SettingCommunication Satisfaction Questionnaire24 Month Follow-up4.065 units on a scaleStandard Deviation 0.69
p-value: 0.028395% CI: [-0.5599, -0.0335]Mixed Models Analysis
p-value: 0.740795% CI: [-0.1969, 0.2771]Mixed Models Analysis
p-value: 0.061995% CI: [-0.4778, 0.0102]Mixed Models Analysis
p-value: 0.000995% CI: [-0.599, -0.1596]Mixed Models Analysis
Primary

Outcome Rating Scale (ORS)

The Outcome Rating Scale (ORS) was utilized as a repeated measure of general therapy outcomes and quality of life changes during the course of therapy. The Outcome Rating Scale includes a visual analog scale (a horizontal line on which the participants marks how well they are doing within the last week from low to high) that records four questions about general well-being, personal well-being, close relationships, and work/school/friend relationships. Physical marks for each of four domains on the visual analog scale are measured by research team members with a ruler and converted to a score from 1 to 100. The four items are then averaged for an overall therapy outcome score. The total averaged ORS score ranges from 1 to 100, with higher scores indicating a better outcome. Analyses will examine treatment progress via change in ORS scores from pre- to post-intervention.

Time frame: Change in scores on the Outcome Rating Scale from Baseline to 24 month follow-up

Population: Several factors influenced variability in the number of participants analyzed across time points. First, patients may have responded to some items but opted out of others because of questionnaire length or comprehension considerations. Second, some patients were available at baseline but were unavailable at subsequent time points, or vice versa.

ArmMeasureGroupValue (MEAN)Dispersion
Intervention Site - Urban SettingOutcome Rating Scale (ORS)Baseline58.085 units on a scaleStandard Deviation 24.192
Intervention Site - Urban SettingOutcome Rating Scale (ORS)24 Month Follow-up54.083 units on a scaleStandard Deviation 29.634
Control Site - Urban SettingOutcome Rating Scale (ORS)24 Month Follow-up56.987 units on a scaleStandard Deviation 28.737
Control Site - Urban SettingOutcome Rating Scale (ORS)Baseline49.003 units on a scaleStandard Deviation 29.995
Intervention Site - Rural SettingOutcome Rating Scale (ORS)Baseline57.522 units on a scaleStandard Deviation 23.978
Intervention Site - Rural SettingOutcome Rating Scale (ORS)24 Month Follow-up47.083 units on a scaleStandard Deviation 29.46
Control Site - Rural SettingOutcome Rating Scale (ORS)Baseline59.583 units on a scaleStandard Deviation 24.171
Control Site - Rural SettingOutcome Rating Scale (ORS)24 Month Follow-up61.433 units on a scaleStandard Deviation 21.688
p-value: 0.658895% CI: [-10.4728, 6.616]Mixed Models Analysis
Comparison: a priori testp-value: 0.263395% CI: [-2.2237, 11.0155]Mixed Models Analysis
Comparison: a priori testp-value: 0.034395% CI: [-17.2107, -0.7299]Mixed Models Analysis
Comparison: a priori testp-value: 0.312695% CI: [-3.4864, 10.9458]Mixed Models Analysis
Primary

Shared Decision Making Questionnaire

A 6-item modified version of the Shared Decision Making Questionnaire (SDM-Q-9) 46 was utilized to assess client reports about the degree to which their provider involved them in understanding and making a treatment decision. Examples items included My provider discussed the advantages and disadvantages of options and strategies or My provider helped me understand all the information measured on a 6-point scale (completely disagree, strongly disagree, somewhat disagree, somewhat agree, strongly agree, and completely agree). The total average score ranges from 1 to 6 with higher scores indicating a better outcome of greater shared decision making. Analyses will examine change in Shared Decision Making Questionnaire scores from pre- to post-intervention.

Time frame: Change in scores on the Shared Decision Making Questionnaire from Baseline to 24 month follow-up

Population: Several factors influenced variability in the number of participants analyzed across time points. First, patients may have responded to some items but opted out of others because of questionnaire length or comprehension considerations. Second, some patients were available at baseline but were unavailable at subsequent time points, or vice versa.

ArmMeasureGroupValue (MEAN)Dispersion
Intervention Site - Urban SettingShared Decision Making QuestionnaireBaseline4.698 units on a scaleStandard Deviation 1.289
Intervention Site - Urban SettingShared Decision Making Questionnaire24 Month Follow-up4.512 units on a scaleStandard Deviation 0.872
Control Site - Urban SettingShared Decision Making Questionnaire24 Month Follow-up4.363 units on a scaleStandard Deviation 0.999
Control Site - Urban SettingShared Decision Making QuestionnaireBaseline4.005 units on a scaleStandard Deviation 1.293
Intervention Site - Rural SettingShared Decision Making QuestionnaireBaseline4.995 units on a scaleStandard Deviation 0.907
Intervention Site - Rural SettingShared Decision Making Questionnaire24 Month Follow-up5.270 units on a scaleStandard Deviation 0.827
Control Site - Rural SettingShared Decision Making QuestionnaireBaseline5.220 units on a scaleStandard Deviation 0.673
Control Site - Rural SettingShared Decision Making Questionnaire24 Month Follow-up5.239 units on a scaleStandard Deviation 0.665
p-value: 0.58495% CI: [-0.5675, 0.3193]Mixed Models Analysis
p-value: 0.240495% CI: [-0.1754, 0.7038]Mixed Models Analysis
p-value: 0.107495% CI: [-0.0667, 0.7051]Mixed Models Analysis
p-value: 0.767795% CI: [-0.3238, 0.439]Mixed Models Analysis
Primary

Short Form Health Survey-12 (SF-12) Mental Symptoms

The mental health subscale (MCS-12; Mental Component Summary) of the SF-12 (Health Survey Short Form-12) was utilized in the current study to assess mental aspects of health and well-being48. The measure includes twelve questions asking about overall health, limitations from health conditions, physical health, emotional well-being and daily activities, and feelings over the past four weeks, utilizing variable Likert scale response choice options. The physical health subscale of the SF-12 was utilized as a key client functioning outcome in the current study. The aggregate MCS subscale score of the SF-12 is calculated utilizing norm-based scoring with a weighted sum (Ware, Kosinski, & Keller, 1995). MCS scores in the present study ranged from 9.6 to 72.0, with higher values indicating better physical health.

Time frame: Change in scores on the SF-12 from Baseline to 24 month follow-up Description: The Health Survey Short Form-12 (SF-12) includes 12 items that assess for physical and mental aspects of health and well-being.

Population: Several factors influenced variability in the number of participants analyzed across time points. First, patients may have responded to some items but opted out of others because of questionnaire length or comprehension considerations. Second, some patients were available at baseline but were unavailable at subsequent time points, or vice versa.

ArmMeasureGroupValue (MEAN)Dispersion
Intervention Site - Urban SettingShort Form Health Survey-12 (SF-12) Mental SymptomsBaseline43.828 units on a scaleStandard Deviation 14.15
Intervention Site - Urban SettingShort Form Health Survey-12 (SF-12) Mental Symptoms24 Month Follow-up39.931 units on a scaleStandard Deviation 12.795
Control Site - Urban SettingShort Form Health Survey-12 (SF-12) Mental Symptoms24 Month Follow-up39.570 units on a scaleStandard Deviation 13.714
Control Site - Urban SettingShort Form Health Survey-12 (SF-12) Mental SymptomsBaseline38.647 units on a scaleStandard Deviation 13.623
Intervention Site - Rural SettingShort Form Health Survey-12 (SF-12) Mental SymptomsBaseline34.153 units on a scaleStandard Deviation 12.288
Intervention Site - Rural SettingShort Form Health Survey-12 (SF-12) Mental Symptoms24 Month Follow-up35.850 units on a scaleStandard Deviation 11.915
Control Site - Rural SettingShort Form Health Survey-12 (SF-12) Mental SymptomsBaseline39.970 units on a scaleStandard Deviation 13.033
Control Site - Rural SettingShort Form Health Survey-12 (SF-12) Mental Symptoms24 Month Follow-up37.768 units on a scaleStandard Deviation 11.092
p-value: 0.098895% CI: [-7.5302, 0.6254]Mixed Models Analysis
p-value: 0.87695% CI: [-4.1799, 3.5625]Mixed Models Analysis
p-value: 0.187895% CI: [-1.2234, 6.3054]Mixed Models Analysis
p-value: 0.33395% CI: [-4.9, 1.6526]Mixed Models Analysis
Primary

Short Form Health Survey-12 (SF-12), Physical Symptoms Subscale

The physical health subscale (PCS-12; Physical Component Summary) of the SF-12 (Health Survey Short Form-12) was utilized in the current study to assess physical aspects of health and well-being48. The measure includes twelve questions asking about overall health, limitations from health conditions, physical health, emotional well-being and daily activities, and feelings over the past four weeks, utilizing variable Likert scale response choice options. The aggregate PCS subscale score of the SF-12 is calculated utilizing norm-based scoring with a weighted sum (Ware, Kosinski, & Keller, 1995). PCS scores in the present study ranged from 13.2 to 65.6, with higher values indicating better physical health.

Time frame: Change in scores on the SF-12 from Baseline to 24 month follow-up

Population: Several factors influenced variability in the number of participants analyzed across time points. First, patients may have responded to some items but opted out of others because of questionnaire length or comprehension considerations. Second, some patients were available at baseline but were unavailable at subsequent time points, or vice versa.

ArmMeasureGroupValue (MEAN)Dispersion
Intervention Site - Urban SettingShort Form Health Survey-12 (SF-12), Physical Symptoms SubscaleBaseline38.167 units on a scaleStandard Deviation 11.067
Intervention Site - Urban SettingShort Form Health Survey-12 (SF-12), Physical Symptoms Subscale24 Month Follow-up34.386 units on a scaleStandard Deviation 8.801
Control Site - Urban SettingShort Form Health Survey-12 (SF-12), Physical Symptoms Subscale24 Month Follow-up35.884 units on a scaleStandard Deviation 12.343
Control Site - Urban SettingShort Form Health Survey-12 (SF-12), Physical Symptoms SubscaleBaseline38.601 units on a scaleStandard Deviation 10.595
Intervention Site - Rural SettingShort Form Health Survey-12 (SF-12), Physical Symptoms SubscaleBaseline37.187 units on a scaleStandard Deviation 12.585
Intervention Site - Rural SettingShort Form Health Survey-12 (SF-12), Physical Symptoms Subscale24 Month Follow-up36.354 units on a scaleStandard Deviation 9.734
Control Site - Rural SettingShort Form Health Survey-12 (SF-12), Physical Symptoms SubscaleBaseline37.108 units on a scaleStandard Deviation 10.148
Control Site - Rural SettingShort Form Health Survey-12 (SF-12), Physical Symptoms Subscale24 Month Follow-up37.163 units on a scaleStandard Deviation 12.555
p-value: 0.067995% CI: [-6.9559, 0.2255]Mixed Models Analysis
p-value: 0.099995% CI: [-6.2957, 0.5317]Mixed Models Analysis
p-value: 0.558895% CI: [-4.4957, 2.3779]Mixed Models Analysis
p-value: 0.853295% CI: [-2.6771, 3.2367]Mixed Models Analysis
Primary

Working Alliance Inventory

The Working Alliance Inventory measures the perception of therapeutic alliance in a clinical dyad during the process of developing a relationship required for effective psychotherapy. The current study utilized the client version of the Working Alliance Inventory included 7 items measured on a 7-point scale (never, rarely, occasionally, sometimes, often, very often, always). Example items included I am confident in my provider's ability to help me and My provider and I trust one another. The Working Alliance Inventory total average score ranges from 1 to 7, with high scores indicating a more positive outcome. Analyses will examine change in patient-reported Working Alliance Inventory scores from pre- to post-intervention.

Time frame: Change in scores on the Working Alliance Inventory from Baseline to 24 month follow-up

Population: Several factors influenced variability in the number of participants analyzed across time points. First, patients may have responded to some items but opted out of others because of questionnaire length or comprehension considerations. Second, some patients were available at baseline but were unavailable at subsequent time points, or vice versa.

ArmMeasureGroupValue (MEAN)Dispersion
Intervention Site - Urban SettingWorking Alliance InventoryBaseline5.896 units on a scaleStandard Deviation 1.065
Intervention Site - Urban SettingWorking Alliance Inventory24 Month Follow-up5.842 units on a scaleStandard Deviation 0.941
Control Site - Urban SettingWorking Alliance Inventory24 Month Follow-up5.810 units on a scaleStandard Deviation 1.344
Control Site - Urban SettingWorking Alliance InventoryBaseline5.494 units on a scaleStandard Deviation 1.086
Intervention Site - Rural SettingWorking Alliance Inventory24 Month Follow-up5.723 units on a scaleStandard Deviation 1.182
Intervention Site - Rural SettingWorking Alliance InventoryBaseline6.203 units on a scaleStandard Deviation 0.751
Control Site - Rural SettingWorking Alliance Inventory24 Month Follow-up5.751 units on a scaleStandard Deviation 1.042
Control Site - Rural SettingWorking Alliance InventoryBaseline6.058 units on a scaleStandard Deviation 1.045
p-value: 0.904695% CI: [-0.4507, 0.3987]Mixed Models Analysis
p-value: 0.251895% CI: [-0.1591, 0.6107]Mixed Models Analysis
p-value: 0.033895% CI: [-0.837, -0.037]Mixed Models Analysis
p-value: 0.155595% CI: [-0.6086, 0.0958]Mixed Models Analysis

Source: ClinicalTrials.gov · Data processed: Feb 4, 2026