Hypertension, Diabetes, Obesity, Tobacco Dependence
Conditions
Brief summary
This is a multi-center randomized controlled trial comparing the effectiveness of community health worker (CHW) vs. usual clinician support in helping chronically-ill patients with low socioeconomic status to improve their health outcomes.
Detailed description
Community Health Workers (CHWs) have the potential to improve chronic disease outcomes among patients with low socioeconomic status (SES). Yet, widespread use of CHWs has been hampered by lack of standardized, scalable and evidence-based models. Our community-academic-health system team used participatory action research with low-SES patients to design IMPaCT (Individualized Management for Patient-Centered Targets). A randomized controlled trial (RCT) of 446 hospitalized patients with varied diseases demonstrated that a 2-week dose of IMPaCT improved access to care, mental health, activation, and quality of communication, while reducing recurrent hospital readmissions. We have adapted IMPaCT for use among low-SES primary care patients with multiple chronic conditions. This RCT is designed to evaluate the newly adapted IMPaCT model in 3 primary care settings - academic, federally qualified health center, and VA. Upon enrollment in the trial, patients will collaboratively set a chronic disease management goal with their primary care provider. Patients will then be randomized to collaborative goal-setting versus goal-setting augmented by 6 months of support from an IMPaCT CHW. Follow-up will be conducted at 6- and 9- months post-enrollment. The study design is a single-blinded, 2- armed, multi-site randomized controlled trial involving three clinic sites and 444 patients.
Interventions
BEHAVIORALIMPaCT
The IMPaCT intervention has three stages: 1. Goal-setting: CHWs will help patients to deconstruct the chronic disease management goal they set with their PCP into patient-driven short-term goals and action plans. 2. Tailored Support: CHWs will conduct weekly follow-up for 6 months through either telephone or home visit in order to support the achievement of patients' short-term goals. 3. Connection with longitudinal support: IMPaCT CHWs will also facilitate a weekly patient support group.
OTHERUsual care
Sponsors
Patient-Centered Outcomes Research Institute
University of Pennsylvania
Study design
Allocation
RANDOMIZED
Intervention model
SINGLE_GROUP
Primary purpose
HEALTH_SERVICES_RESEARCH
Masking
SINGLE (Outcomes Assessor)
Eligibility
Sex/Gender
ALL
Age
18 Years to No maximum
Healthy volunteers
No
Inclusion criteria
* Patient of specific practice (Mutch/St.Leonards Court, Spectrum, PVAMC) for at least 1 yr defined as having one office visit within preceding 12 months * Resident of home ZIP code including ANY of the following 19104, 19131, 19139, 19142, 19143, 19145, 19146, & 19151 * Has 2 or more of the following conditions: 1. Obesity: BMI30 2. HTN: ICD9 relating to HTN from EMR problem list or EMR ICD9 encounter diagnosis x2) 3.DM: ICD9 relating to DM from EMR problem list or EMR ICD9 encounter diagnosis x2) 4. tobacco (from EPIC Social History/Vital Signs) * Has at least one poorly controlled condition based on the most recent value prior to enrollment: Obesity (BMI\>=35); HTN (SBP \>= 160); DM (HgBA1c \>=9); Tobacco Use (\>0 cigarettes) * Is uninsured, insured by Medicaid, or dually eligible for Medicare and Medicaid or VA Insurance * Has a scheduled appointment at a study clinic in the future. * Community Health Workers/CHWs/IMPaCT Partners involved in the care of patients of the three practices will be included in the qualitative portion of this study.
Exclusion criteria
* Will not provide informed consent for this study. * Does not have the capacity to provide informed consent for this study. * Previously enrolled in this study. * Currently enrolled in another study focusing on chronic disease management. * Currently has a CHW * No one will be excluded on the basis of sex or race.
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Short Form Health Survey (SF-12) Physical Component Score (PCS) | Baseline, 6 months | The main dependent variable is mean change in standardized score for SF-12 PCS. The SF-12 is a survey designed for use with patients with multiple chronic conditions. This 12-item scale can be used to assess the physical and mental health of respondents. 10 of the 12 questions are answered on a 5 point likert scale and 2 are answered on a 3 point likert scale. The questions are then scored and weighted into 2 subscales, physical health and mental health. Respondents can have a score that ranges from 0-100 with 100 being the best score and indicating high physical or mental health. A 3 point change in SF-12 score reflects a meaningful difference. We measure the between-arm difference in mean change in SF-12 PCS between baseline and 6-month follow-up assessment. |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Change in Chronic Disease Control - Obesity | Baseline, 6 months, 9 months | We will asses change in chronic disease control using biometric testing (kg/m\^2). At six and nine months after enrollment, all patients underwent a clinic visit and the appropriate laboratory testing: height and weight. Patients' measurements on this parameter will be used to determine their change in standardized score for their outcome of interest. |
| Change in Chronic Disease Control - Tobacco Use | Baseline, 6 months, 9 months | We will asses change in chronic disease control using patient self-report (cigarettes per day). At six and nine months after enrollment, all patients underwent a clinic visit and the appropriate laboratory testing. Patients' measurements on this parameter will be used to determine their change in standardized score for their outcome of interest. |
| Change in Chronic Disease Control - Hypertension | Baseline, 6 months, 9 months | We will asses change in chronic disease control using biometric testing (systolic blood pressure in mm Hg). At six and nine months after enrollment, all patients underwent a clinic visit and the appropriate laboratory testing. Patients' measurements on this parameter will be used to determine their change in standardized score for their outcome of interest. |
| Short Form Health Survey (SF-12) - Mental Component Summary (MCS) | Baseline, 6 months, 9 months | The SF-12 is a survey designed for use with patients with multiple chronic conditions. This 12-item scale can be used to assess the physical and mental health of respondents. 10 of the 12 questions are answered on a 5 point likert scale and 2 are answered on a 3 point likert scale. The questions are then scored and weighted into 2 subscales, physical health and mental health. Respondents can have a score that ranges from 0-100 with 100 being the best score and indicating high physical or mental health. A 3 point change in SF-12 score reflects a meaningful difference. We will assess this outcome using the SF-12 Mental Component Summary (MCS) score. The MCS reliably detects differences in mental health over time. We will measure the between-arm difference in mean change in SF-12 MCS score between baseline, 6- and 9- month follow-up. |
| Number of Participants Reporting Highest Rating for Quality of Patient-centered Care - Comprehensiveness | 6 months, 9 months | We will assess this outcome using the Consumer Assessment of Healthcare Providers and Systems Patient-Centered Medical Home (CAHPS PCMH) survey. This survey assesses the quality of patient-centered primary care and can be used by any practice (not just PCMH practices). We will measure the CAHPS PCMH domains pertaining to Self-Management Support and Comprehensiveness of Care. We will measure the number of patients who gave the highest rating of care for the comprehensiveness question at 6 and 9 months post-enrollment. |
| Number of Participants With Any Hospital Admission | 6 months and 9 months | We will measure admission to hospital at 6- and 9 months after enrollment. This data will be obtained through 1) self-report, 2) the Pennsylvania Cost Containment Council (PHC4), a state-based initiative that tracks utilization data across the state of Pennsylvania, 3) the Veterans Affairs electronic medical record, 4) the Penn electronic medical record |
| Change in Chronic Disease Control - Diabetes | Baseline, 6 months, 9 months | We will asses change in chronic disease control using biometric testing (HgA1c). At six months after enrollment, all patients underwent a clinic visit and the appropriate laboratory testing: HgbA1c. Patients' measurements on this parameter will be used to determine their change in standardized score for their outcome of interest. |
| Number of Participants Reporting the Highest Rating for Quality of Patient-centered Care - Supportiveness of Self-management | 6 months, 9 months | We will assess this outcome using the Consumer Assessment of Healthcare Providers and Systems Patient-Centered Medical Home (CAHPS PCMH) survey. This survey assesses the quality of patient-centered primary care and can be used by any practice (not just PCMH practices). We will measure the CAHPS PCMH domains pertaining to Self-Management Support and Comprehensiveness of Care. We will measure the number of patients who gave the highest rating of care for the supportiveness of disease self-management question at 6 and 9 months post-enrollment. |
| Number of Participants With Multiple Hospital Admissions | 6 months and 9 months | We will measure admission to hospital at 6- and 9 months after enrollment. This data will be obtained through 1) self-report, 2) the Pennsylvania Cost Containment Council (PHC4), a state-based initiative that tracks utilization data across the state of Pennsylvania, 3) the Veterans Affairs electronic medical record, 4) the Penn electronic medical record |
| Number of Participants With 30 Day Hospital Readmissions | 30 days | We will measure admission to hospital at 30 days after enrollment. This data will be obtained through 1) self-report, 2) the Pennsylvania Cost Containment Council (PHC4), a state-based initiative that tracks utilization data across the state of Pennsylvania, 3) the Veterans Affairs electronic medical record, 4) the Penn electronic medical record |
| Hospital Admission - Total Hospital Days | 6 months and 9 months | We will measure admission to hospital at 6- and 9 months after enrollment. This data will be obtained through 1) self-report, 2) the Pennsylvania Cost Containment Council (PHC4), a state-based initiative that tracks utilization data across the state of Pennsylvania, 3) the Veterans Affairs electronic medical record, 4) the Penn electronic medical record |
| Hospital Admission - Mean Number of Hospitalizations | 6 months and 9 months | We will measure admission to hospital at 6- and 9 months after enrollment. This data will be obtained through 1) self-report, 2) the Pennsylvania Cost Containment Council (PHC4), a state-based initiative that tracks utilization data across the state of Pennsylvania, 3) the Veterans Affairs electronic medical record, 4) the Penn electronic medical record |
| Hospital Admission - Mean Length of Stay (Among Participants With Hospitalization) | 6 months and 9 months | We will measure admission to hospital at 6- and 9 months after enrollment. This data will be obtained through 1) self-report, 2) the Pennsylvania Cost Containment Council (PHC4), a state-based initiative that tracks utilization data across the state of Pennsylvania, 3) the Veterans Affairs electronic medical record, 4) the Penn electronic medical record |
| Qualitative Assessment of Intervention and Mechanisms Affecting Achievement of Primary Outcome | 6 months | At 6-months post-enrollment, a trained qualitative interviewer on our study team will conduct an in-depth qualitative semi-structured interview with 40 intervention arm patients and their CHWs. Qualitative interviews will be audio-taped and transcribed. Patients will be purposively sampled across each study site in order to be able to make comparisons between those who achieved a minimally important improvement in the primary outcome and those who did not. These interviews will be guided by the Integrative Behavior Model (IBM). UPDATE: After beginning interviews, our team decided that 26 interviews was sufficient to reach thematic saturation. |
Participant flow
Recruitment details
Study enrollment took place at: a Veterans Affairs medical center primary care practice, a federally qualified health center and an academic family medicine clinic. Patients were recruited between January 28th, 2015 and March 28, 2016 at which time the trial was stopped because we had reached our pre-specified sample size target.
Participants by arm
| Arm | Count |
|---|---|
| Usual Clinician Support Prior to randomization, patients in this arm will already have set a chronic disease management goal with their primary care provider (who will have received training in collaborative goal-setting). After randomization, patients in the usual clinician support arm will receive usual care in accordance with guidelines at each site.
Usual care | 288 |
| CHW Support Prior to randomization, patients in this arm will already have set a chronic disease management goal with their primary care provider (who will have received training in collaborative goal-setting). After randomization, patients in the CHW arm will receive the IMPaCT intervention.
IMPaCT: The IMPaCT intervention has three stages:
1. Goal-setting: CHWs will help patients to deconstruct the chronic disease management goal they set with their PCP into patient-driven short-term goals and action plans.
2. Tailored Support: CHWs will conduct weekly follow-up for 6 months through either telephone or home visit in order to support the achievement of patients' short-term goals.
3. Connection with longitudinal support: IMPaCT CHWs will also facilitate a weekly patient support group. | 304 |
| Total | 592 |
Withdrawals & dropouts
| Period | Reason | FG000 | FG001 |
|---|---|---|---|
| 6 Month Follow-Up | Lost to Follow-up | 63 | 47 |
| 6 Month Follow-Up | Withdrawal by Subject | 3 | 9 |
| 9 Month Follow-Up | Lost to Follow-up | 52 | 59 |
| 9 Month Follow-Up | Withdrawal by Subject | 9 | 12 |
Baseline characteristics
| Characteristic | Usual Clinician Support | CHW Support | Total |
|---|---|---|---|
| Admissions in prior 12 months | 1.0 Avg # of admissions STANDARD_DEVIATION 2.5 | 0.7 Avg # of admissions STANDARD_DEVIATION 1.1 | 0.9 Avg # of admissions STANDARD_DEVIATION 1.9 |
| Adult Attachment Questionnaire (AAQ) Anxious | 2.9 units on a scale STANDARD_DEVIATION 1.8 | 2.8 units on a scale STANDARD_DEVIATION 1.8 | 2.9 units on a scale STANDARD_DEVIATION 1.8 |
| Adult Attachment Questionnaire (AAQ) Avoidant | 3.6 units on a scale STANDARD_DEVIATION 1.3 | 3.7 units on a scale STANDARD_DEVIATION 1.3 | 3.6 units on a scale STANDARD_DEVIATION 1.3 |
| Adult Attachment Questionnaire (AAQ) Secure | 3.4 units on a scale STANDARD_DEVIATION 0.9 | 3.4 units on a scale STANDARD_DEVIATION 0.9 | 3.4 units on a scale STANDARD_DEVIATION 0.9 |
| Age, Continuous | 52.1 years STANDARD_DEVIATION 11.5 | 53.1 years STANDARD_DEVIATION 10.7 | 52.6 years STANDARD_DEVIATION 11.5 |
| Alcohol Overuse | 73 Participants | 81 Participants | 154 Participants |
| Chronic Disease Prevalence - Diabetes | 134 Participants | 139 Participants | 273 Participants |
| Chronic Disease Prevalence - Hypertension | 257 Participants | 263 Participants | 520 Participants |
| Chronic Disease Prevalence - Obesity | 228 Participants | 240 Participants | 468 Participants |
| Chronic Disease Prevalence - Tobacco Dependence | 104 Participants | 124 Participants | 228 Participants |
| Delayed Health Need | 129 Participants | 143 Participants | 272 Participants |
| Diabetes Baseline Control | 10.5 HbA1c% STANDARD_DEVIATION 2 | 10.5 HbA1c% STANDARD_DEVIATION 2 | 10.5 HbA1c% STANDARD_DEVIATION 2 |
| Diabetes Diabetes Goal | -2.3 HbA1c% STANDARD_DEVIATION 1.8 | -2.3 HbA1c% STANDARD_DEVIATION 2.2 | -2.3 HbA1c% STANDARD_DEVIATION 2 |
| Drug Use | 69 Participants | 63 Participants | 132 Participants |
| Employed | 48 Participants | 47 Participants | 95 Participants |
| ER visits in prior 12 months | 2.2 Avg # of visits STANDARD_DEVIATION 3.5 | 1.8 Avg # of visits STANDARD_DEVIATION 2.1 | 2.0 Avg # of visits STANDARD_DEVIATION 2.9 |
| Ethnicity (NIH/OMB) Hispanic or Latino | 0 Participants | 11 Participants | 11 Participants |
| Ethnicity (NIH/OMB) Not Hispanic or Latino | 285 Participants | 285 Participants | 570 Participants |
| Ethnicity (NIH/OMB) Unknown or Not Reported | 3 Participants | 8 Participants | 11 Participants |
| Household Income < 15K | 186 Participants | 197 Participants | 383 Participants |
| Household Income > 15K | 69 Participants | 69 Participants | 138 Participants |
| Household Income Unknown | 33 Participants | 38 Participants | 71 Participants |
| Hypertension Baseline Control | 159.8 mmHg STANDARD_DEVIATION 26.5 | 163.4 mmHg STANDARD_DEVIATION 19.3 | 161.6 mmHg STANDARD_DEVIATION 23.3 |
| Hypertension Hypertension Goal | -26.4 mmHg STANDARD_DEVIATION 34.6 | -24.1 mmHg STANDARD_DEVIATION 17.9 | -25.3 mmHg STANDARD_DEVIATION 27.7 |
| Lack of Basic needs | 118 Participants | 120 Participants | 238 Participants |
| Low Social Support | 57 Participants | 76 Participants | 133 Participants |
| Obesity - Baseline Clinical Parameters | 42.2 kg/m^2 (BMI) STANDARD_DEVIATION 6.9 | 42.8 kg/m^2 (BMI) STANDARD_DEVIATION 8.1 | 42.5 kg/m^2 (BMI) STANDARD_DEVIATION 7.5 |
| Obesity - Health Goal | -15.1 pounds STANDARD_DEVIATION 5.6 | -14.7 pounds STANDARD_DEVIATION 6.3 | -14.9 pounds STANDARD_DEVIATION 6 |
| Patient Activation Measure (PAM) | 60.8 units on a scale STANDARD_DEVIATION 14.3 | 59.4 units on a scale STANDARD_DEVIATION 14.2 | 60.1 units on a scale STANDARD_DEVIATION 14.3 |
| Perceived Stress Scale (PSS) | 6.7 units on a scale STANDARD_DEVIATION 3.3 | 6.7 units on a scale STANDARD_DEVIATION 3.5 | 6.7 units on a scale STANDARD_DEVIATION 3.4 |
| Race/Ethnicity, Customized African American | 274 Participants | 284 Participants | 558 Participants |
| Selected Condition Diabetes | 35 Participants | 34 Participants | 69 Participants |
| Selected Condition Hypertension | 33 Participants | 30 Participants | 63 Participants |
| Selected Condition Obesity | 149 Participants | 157 Participants | 306 Participants |
| Selected Condition Tobacco Dependence | 71 Participants | 83 Participants | 154 Participants |
| Sex: Female, Male Female | 176 Participants | 194 Participants | 370 Participants |
| Sex: Female, Male Male | 112 Participants | 110 Participants | 222 Participants |
| Short Form Health Survey (SF-12) Mental Health Component | 43.1 score on a scale STANDARD_DEVIATION 12.9 | 42.2 score on a scale STANDARD_DEVIATION 14.1 | 42.7 score on a scale STANDARD_DEVIATION 13.5 |
| Short Form Health Survey (SF-12) Physical Health Component | 33.6 score on a scale STANDARD_DEVIATION 9.9 | 34.0 score on a scale STANDARD_DEVIATION 10.7 | 33.8 score on a scale STANDARD_DEVIATION 10.3 |
| Short Grit Scale (Grit-S) | 2.7 units on a scale STANDARD_DEVIATION 0.5 | 2.7 units on a scale STANDARD_DEVIATION 0.6 | 2.7 units on a scale STANDARD_DEVIATION 0.5 |
| Single Item Literacy Screener (SILS) | 1.9 units on a scale STANDARD_DEVIATION 1.1 | 2.1 units on a scale STANDARD_DEVIATION 1.3 | 2.0 units on a scale STANDARD_DEVIATION 1.2 |
| Tobacco Use | 9.7 number of cigarettes STANDARD_DEVIATION 7.2 | 8.9 number of cigarettes STANDARD_DEVIATION 6.4 | 9.3 number of cigarettes STANDARD_DEVIATION 6.8 |
| Trauma History | 280 Participants | 297 Participants | 577 Participants |
| Unmet health Need | 81 Participants | 76 Participants | 157 Participants |
Adverse events
| Event type | EG000 affected / at risk | EG001 affected / at risk |
|---|---|---|
| deaths Total, all-cause mortality | 3 / 288 | 1 / 304 |
| other Total, other adverse events | 0 / 288 | 0 / 304 |
| serious Total, serious adverse events | 32 / 288 | 35 / 304 |
Outcome results
Primary
Short Form Health Survey (SF-12) Physical Component Score (PCS)
The main dependent variable is mean change in standardized score for SF-12 PCS. The SF-12 is a survey designed for use with patients with multiple chronic conditions. This 12-item scale can be used to assess the physical and mental health of respondents. 10 of the 12 questions are answered on a 5 point likert scale and 2 are answered on a 3 point likert scale. The questions are then scored and weighted into 2 subscales, physical health and mental health. Respondents can have a score that ranges from 0-100 with 100 being the best score and indicating high physical or mental health. A 3 point change in SF-12 score reflects a meaningful difference. We measure the between-arm difference in mean change in SF-12 PCS between baseline and 6-month follow-up assessment.
Time frame: Baseline, 6 months
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Usual Clinician Support | Short Form Health Survey (SF-12) Physical Component Score (PCS) | 2.3 score on a scale | Standard Deviation 11.3 |
| CHW Support | Short Form Health Survey (SF-12) Physical Component Score (PCS) | 0.6 score on a scale | Standard Deviation 12.7 |
Comparison: Determination of targeted sample size was based on detecting a between-arm difference in mean change in SF-12 Physical Component Summary of 3 points, which is the minimal clinically significant difference for this instrument. To achieve at least 80% power with a type I error rate of 5%, we required 444 total participants. To account for 25% attrition, we aimed to accrue 592 participants.p-value: 0.3Regression, Linear
Secondary
Change in Chronic Disease Control - Diabetes
We will asses change in chronic disease control using biometric testing (HgA1c). At six months after enrollment, all patients underwent a clinic visit and the appropriate laboratory testing: HgbA1c. Patients' measurements on this parameter will be used to determine their change in standardized score for their outcome of interest.
Time frame: Baseline, 6 months, 9 months
Population: Each participant was only analyzed for their chosen chronic condition.
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Usual Clinician Support | Change in Chronic Disease Control - Diabetes | -6.5 HbA1c% | Standard Deviation 26 |
| CHW Support | Change in Chronic Disease Control - Diabetes | -10.9 HbA1c% | Standard Deviation 16.3 |
| Usual Clinician Support - 9 Months | Change in Chronic Disease Control - Diabetes | -10.4 HbA1c% | Standard Deviation 22.5 |
| CHW Support - 9 Months | Change in Chronic Disease Control - Diabetes | -17.9 HbA1c% | Standard Deviation 19.5 |
p-value: 0.2195% CI: [-1.3, 0.9]Regression, Linear
Secondary
Change in Chronic Disease Control - Hypertension
We will asses change in chronic disease control using biometric testing (systolic blood pressure in mm Hg). At six and nine months after enrollment, all patients underwent a clinic visit and the appropriate laboratory testing. Patients' measurements on this parameter will be used to determine their change in standardized score for their outcome of interest.
Time frame: Baseline, 6 months, 9 months
Population: Each participant was only analyzed for their chosen chronic condition.
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Usual Clinician Support | Change in Chronic Disease Control - Hypertension | -6.2 mm Hg | Standard Deviation 26 |
| CHW Support | Change in Chronic Disease Control - Hypertension | -10.9 mm Hg | Standard Deviation 16.3 |
| Usual Clinician Support - 9 Months | Change in Chronic Disease Control - Hypertension | -10.4 mm Hg | Standard Deviation 22.5 |
| CHW Support - 9 Months | Change in Chronic Disease Control - Hypertension | -17.9 mm Hg | Standard Deviation 19.5 |
p-value: 0.2195% CI: [-14.3, 1.8]Regression, Linear
Secondary
Change in Chronic Disease Control - Obesity
We will asses change in chronic disease control using biometric testing (kg/m\^2). At six and nine months after enrollment, all patients underwent a clinic visit and the appropriate laboratory testing: height and weight. Patients' measurements on this parameter will be used to determine their change in standardized score for their outcome of interest.
Time frame: Baseline, 6 months, 9 months
Population: Each participant was only analyzed for their chosen chronic condition.
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Usual Clinician Support | Change in Chronic Disease Control - Obesity | -0.3 kg/m^2 (BMI) | Standard Deviation 2.5 |
| CHW Support | Change in Chronic Disease Control - Obesity | -0.5 kg/m^2 (BMI) | Standard Deviation 2.3 |
| Usual Clinician Support - 9 Months | Change in Chronic Disease Control - Obesity | -0.4 kg/m^2 (BMI) | Standard Deviation 2.7 |
| CHW Support - 9 Months | Change in Chronic Disease Control - Obesity | -0.5 kg/m^2 (BMI) | Standard Deviation 3 |
p-value: 0.2195% CI: [-0.7, 0.4]Regression, Linear
Secondary
Change in Chronic Disease Control - Tobacco Use
We will asses change in chronic disease control using patient self-report (cigarettes per day). At six and nine months after enrollment, all patients underwent a clinic visit and the appropriate laboratory testing. Patients' measurements on this parameter will be used to determine their change in standardized score for their outcome of interest.
Time frame: Baseline, 6 months, 9 months
Population: Each participant was only analyzed for their chosen chronic condition.
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Usual Clinician Support | Change in Chronic Disease Control - Tobacco Use | -3.1 cigarettes per day | Standard Deviation 7.1 |
| CHW Support | Change in Chronic Disease Control - Tobacco Use | -3.5 cigarettes per day | Standard Deviation 6.4 |
| Usual Clinician Support - 9 Months | Change in Chronic Disease Control - Tobacco Use | -3.5 cigarettes per day | Standard Deviation 6.5 |
| CHW Support - 9 Months | Change in Chronic Disease Control - Tobacco Use | -4.1 cigarettes per day | Standard Deviation 5.4 |
p-value: 0.2195% CI: [-2.2, 1.2]Regression, Linear
Secondary
Hospital Admission - Mean Length of Stay (Among Participants With Hospitalization)
We will measure admission to hospital at 6- and 9 months after enrollment. This data will be obtained through 1) self-report, 2) the Pennsylvania Cost Containment Council (PHC4), a state-based initiative that tracks utilization data across the state of Pennsylvania, 3) the Veterans Affairs electronic medical record, 4) the Penn electronic medical record
Time frame: 6 months and 9 months
Population: This outcome was only assessed for participants who had a hospital admission. Therefore the number of participants analyzed is smaller than that for the overall study.
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Usual Clinician Support | Hospital Admission - Mean Length of Stay (Among Participants With Hospitalization) | 9.9 days | Standard Deviation 17.5 |
| CHW Support | Hospital Admission - Mean Length of Stay (Among Participants With Hospitalization) | 4.1 days | Standard Deviation 3.1 |
| Usual Clinician Support - 9 Months | Hospital Admission - Mean Length of Stay (Among Participants With Hospitalization) | 9.2 days | Standard Deviation 15.3 |
| CHW Support - 9 Months | Hospital Admission - Mean Length of Stay (Among Participants With Hospitalization) | 5.8 days | Standard Deviation 6.5 |
p-value: 0.06Regression, Linear
Secondary
Hospital Admission - Mean Number of Hospitalizations
We will measure admission to hospital at 6- and 9 months after enrollment. This data will be obtained through 1) self-report, 2) the Pennsylvania Cost Containment Council (PHC4), a state-based initiative that tracks utilization data across the state of Pennsylvania, 3) the Veterans Affairs electronic medical record, 4) the Penn electronic medical record
Time frame: 6 months and 9 months
Population: This outcome was only assessed for participants who had a hospital admission. Therefore the number of participants analyzed is smaller than that for the overall study.
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Usual Clinician Support | Hospital Admission - Mean Number of Hospitalizations | 1.5 Hospitalizations | Standard Deviation 0.7 |
| CHW Support | Hospital Admission - Mean Number of Hospitalizations | 1.1 Hospitalizations | Standard Deviation 0.4 |
| Usual Clinician Support - 9 Months | Hospital Admission - Mean Number of Hospitalizations | 1.6 Hospitalizations | Standard Deviation 1 |
| CHW Support - 9 Months | Hospital Admission - Mean Number of Hospitalizations | 1.4 Hospitalizations | Standard Deviation 0.8 |
p-value: 0.09Regression, Linear
Secondary
Hospital Admission - Total Hospital Days
We will measure admission to hospital at 6- and 9 months after enrollment. This data will be obtained through 1) self-report, 2) the Pennsylvania Cost Containment Council (PHC4), a state-based initiative that tracks utilization data across the state of Pennsylvania, 3) the Veterans Affairs electronic medical record, 4) the Penn electronic medical record
Time frame: 6 months and 9 months
| Arm | Measure | Value (NUMBER) |
|---|---|---|
| Usual Clinician Support | Hospital Admission - Total Hospital Days | 345 Days |
| CHW Support | Hospital Admission - Total Hospital Days | 155 Days |
| Usual Clinician Support - 9 Months | Hospital Admission - Total Hospital Days | 471 Days |
| CHW Support - 9 Months | Hospital Admission - Total Hospital Days | 300 Days |
Secondary
Number of Participants Reporting Highest Rating for Quality of Patient-centered Care - Comprehensiveness
We will assess this outcome using the Consumer Assessment of Healthcare Providers and Systems Patient-Centered Medical Home (CAHPS PCMH) survey. This survey assesses the quality of patient-centered primary care and can be used by any practice (not just PCMH practices). We will measure the CAHPS PCMH domains pertaining to Self-Management Support and Comprehensiveness of Care. We will measure the number of patients who gave the highest rating of care for the comprehensiveness question at 6 and 9 months post-enrollment.
Time frame: 6 months, 9 months
| Arm | Measure | Value (COUNT_OF_PARTICIPANTS) |
|---|---|---|
| Usual Clinician Support | Number of Participants Reporting Highest Rating for Quality of Patient-centered Care - Comprehensiveness | 83 Participants |
| CHW Support | Number of Participants Reporting Highest Rating for Quality of Patient-centered Care - Comprehensiveness | 137 Participants |
| Usual Clinician Support - 9 Months | Number of Participants Reporting Highest Rating for Quality of Patient-centered Care - Comprehensiveness | 97 Participants |
| CHW Support - 9 Months | Number of Participants Reporting Highest Rating for Quality of Patient-centered Care - Comprehensiveness | 139 Participants |
p-value: <0.001Regression, Logistic
Secondary
Number of Participants Reporting the Highest Rating for Quality of Patient-centered Care - Supportiveness of Self-management
We will assess this outcome using the Consumer Assessment of Healthcare Providers and Systems Patient-Centered Medical Home (CAHPS PCMH) survey. This survey assesses the quality of patient-centered primary care and can be used by any practice (not just PCMH practices). We will measure the CAHPS PCMH domains pertaining to Self-Management Support and Comprehensiveness of Care. We will measure the number of patients who gave the highest rating of care for the supportiveness of disease self-management question at 6 and 9 months post-enrollment.
Time frame: 6 months, 9 months
| Arm | Measure | Value (COUNT_OF_PARTICIPANTS) |
|---|---|---|
| Usual Clinician Support | Number of Participants Reporting the Highest Rating for Quality of Patient-centered Care - Supportiveness of Self-management | 95 Participants |
| CHW Support | Number of Participants Reporting the Highest Rating for Quality of Patient-centered Care - Supportiveness of Self-management | 157 Participants |
| Usual Clinician Support - 9 Months | Number of Participants Reporting the Highest Rating for Quality of Patient-centered Care - Supportiveness of Self-management | 112 Participants |
| CHW Support - 9 Months | Number of Participants Reporting the Highest Rating for Quality of Patient-centered Care - Supportiveness of Self-management | 145 Participants |
p-value: <0.001Regression, Logistic
Secondary
Number of Participants With 30 Day Hospital Readmissions
We will measure admission to hospital at 30 days after enrollment. This data will be obtained through 1) self-report, 2) the Pennsylvania Cost Containment Council (PHC4), a state-based initiative that tracks utilization data across the state of Pennsylvania, 3) the Veterans Affairs electronic medical record, 4) the Penn electronic medical record
Time frame: 30 days
Population: This outcome was only assessed for participants who had hospital admission. Therefore the number of participants analyzed is smaller than that for the overall study.
| Arm | Measure | Value (COUNT_OF_PARTICIPANTS) |
|---|---|---|
| Usual Clinician Support | Number of Participants With 30 Day Hospital Readmissions | 9 Participants |
| CHW Support | Number of Participants With 30 Day Hospital Readmissions | 3 Participants |
| Usual Clinician Support - 9 Months | Number of Participants With 30 Day Hospital Readmissions | 14 Participants |
| CHW Support - 9 Months | Number of Participants With 30 Day Hospital Readmissions | 6 Participants |
p-value: 0.04Regression, Logistic
Secondary
Number of Participants With Any Hospital Admission
We will measure admission to hospital at 6- and 9 months after enrollment. This data will be obtained through 1) self-report, 2) the Pennsylvania Cost Containment Council (PHC4), a state-based initiative that tracks utilization data across the state of Pennsylvania, 3) the Veterans Affairs electronic medical record, 4) the Penn electronic medical record
Time frame: 6 months and 9 months
| Arm | Measure | Value (COUNT_OF_PARTICIPANTS) |
|---|---|---|
| Usual Clinician Support | Number of Participants With Any Hospital Admission | 35 Participants |
| CHW Support | Number of Participants With Any Hospital Admission | 38 Participants |
| Usual Clinician Support - 9 Months | Number of Participants With Any Hospital Admission | 51 Participants |
| CHW Support - 9 Months | Number of Participants With Any Hospital Admission | 52 Participants |
p-value: 0.9895% CI: [0.6, 1.5]Regression, Logistic
Secondary
Number of Participants With Multiple Hospital Admissions
We will measure admission to hospital at 6- and 9 months after enrollment. This data will be obtained through 1) self-report, 2) the Pennsylvania Cost Containment Council (PHC4), a state-based initiative that tracks utilization data across the state of Pennsylvania, 3) the Veterans Affairs electronic medical record, 4) the Penn electronic medical record
Time frame: 6 months and 9 months
Population: This outcome was only assessed for participants who had hospital admission. Therefore the number of participants analyzed is smaller than that for the overall study.
| Arm | Measure | Value (COUNT_OF_PARTICIPANTS) |
|---|---|---|
| Usual Clinician Support | Number of Participants With Multiple Hospital Admissions | 14 Participants |
| CHW Support | Number of Participants With Multiple Hospital Admissions | 4 Participants |
| Usual Clinician Support - 9 Months | Number of Participants With Multiple Hospital Admissions | 21 Participants |
| CHW Support - 9 Months | Number of Participants With Multiple Hospital Admissions | 15 Participants |
p-value: 0.02Regression, Logistic
Secondary
Qualitative Assessment of Intervention and Mechanisms Affecting Achievement of Primary Outcome
At 6-months post-enrollment, a trained qualitative interviewer on our study team will conduct an in-depth qualitative semi-structured interview with 40 intervention arm patients and their CHWs. Qualitative interviews will be audio-taped and transcribed. Patients will be purposively sampled across each study site in order to be able to make comparisons between those who achieved a minimally important improvement in the primary outcome and those who did not. These interviews will be guided by the Integrative Behavior Model (IBM). UPDATE: After beginning interviews, our team decided that 26 interviews was sufficient to reach thematic saturation.
Time frame: 6 months
Population: This data point contains qualitative, not quantitative data. The study team is still analyzing this data set and has not yet completed analysis for this outcome. Below, we report the number of interviews completed for each clinical site (VA, Federally Qualified Health Center, Academic Site). We will not complete any more interviews.
| Arm | Measure | Category | Value (COUNT_OF_PARTICIPANTS) |
|---|---|---|---|
| Usual Clinician Support | Qualitative Assessment of Intervention and Mechanisms Affecting Achievement of Primary Outcome | FQHC | 10 Participants |
| Usual Clinician Support | Qualitative Assessment of Intervention and Mechanisms Affecting Achievement of Primary Outcome | Academic Site | 6 Participants |
| Usual Clinician Support | Qualitative Assessment of Intervention and Mechanisms Affecting Achievement of Primary Outcome | VA | 10 Participants |
Secondary
Short Form Health Survey (SF-12) - Mental Component Summary (MCS)
The SF-12 is a survey designed for use with patients with multiple chronic conditions. This 12-item scale can be used to assess the physical and mental health of respondents. 10 of the 12 questions are answered on a 5 point likert scale and 2 are answered on a 3 point likert scale. The questions are then scored and weighted into 2 subscales, physical health and mental health. Respondents can have a score that ranges from 0-100 with 100 being the best score and indicating high physical or mental health. A 3 point change in SF-12 score reflects a meaningful difference. We will assess this outcome using the SF-12 Mental Component Summary (MCS) score. The MCS reliably detects differences in mental health over time. We will measure the between-arm difference in mean change in SF-12 MCS score between baseline, 6- and 9- month follow-up.
Time frame: Baseline, 6 months, 9 months
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Usual Clinician Support | Short Form Health Survey (SF-12) - Mental Component Summary (MCS) | 1.7 score on a scale | Standard Deviation 14.1 |
| CHW Support | Short Form Health Survey (SF-12) - Mental Component Summary (MCS) | 2.2 score on a scale | Standard Deviation 13.3 |
| Usual Clinician Support - 9 Months | Short Form Health Survey (SF-12) - Mental Component Summary (MCS) | 1.2 score on a scale | Standard Deviation 13.7 |
| CHW Support - 9 Months | Short Form Health Survey (SF-12) - Mental Component Summary (MCS) | 2.2 score on a scale | Standard Deviation 14.1 |
p-value: 0.41Regression, Linear