Start Submission Become a Reviewer

Reading: Enhanced Integrated Behavioral Health Model Improves Depressive Symptoms in a Low-Income, Un...

Download

A- A+
Alt. Display

Research & theory

Enhanced Integrated Behavioral Health Model Improves Depressive Symptoms in a Low-Income, Uninsured, Primarily Hispanic Population Served by a Free and Charitable Clinic

Authors:

Amy Flynn ,

Health Resources in Action, Boston, MA, US
About Amy
MS
X close

Erika Gaitan,

Health Resources in Action, Boston, MA, US
X close

Rebecca Stocker,

Hope Family Health Center, McAllen, TX, US
X close

Elizabeth Showalter,

Health Resources in Action, Boston, MA, US
X close

Karen Sautter Errichetti

Health Resources in Action, Boston, MA, US
X close

Abstract

Background: Residents living in US-Mexico border communities have some of the worst health outcomes nationally. This randomized trial evaluated whether patients receiving enhanced integrated behavioral health (IBH) care at a southern Texas free and charitable were more likely to improve health outcomes after 12 months compared to patients receiving standard care.

Theory & Methods: The IBH intervention featured brief intervention by a behavioral health specialist and enhanced coordinated care. The primary outcome was systolic blood pressure. Secondary outcomes were diastolic blood pressure, HbA1c, BMI, and depressive symptoms. Linear regression models were utilized to assess the impact of IBH on participants; secondary analyses examined possible effect modification.

Results: After 12 months, intervention participants (n = 172) were more likely to have a lower PHQ-9 score than control participants (n = 198) (β = –1.67, p = 0.01). There was significant modification of the intervention effect by age; there was a different effect on older participants (β = –2.08, p = 0.01). There were no statistically significant findings for other outcomes.

Conclusions: Collaborative, integrated behavioral health and primary care can improve depressive symptoms for low-income or uninsured individuals living in southern Texas border communities. These findings provide evidence that may help develop IBH programs to improve health of vulnerable populations experiencing health inequities.

How to Cite: Flynn A, Gaitan E, Stocker R, Showalter E, Errichetti KS. Enhanced Integrated Behavioral Health Model Improves Depressive Symptoms in a Low-Income, Uninsured, Primarily Hispanic Population Served by a Free and Charitable Clinic. International Journal of Integrated Care. 2020;20(4):15. DOI: http://doi.org/10.5334/ijic.5421
93
Views
18
Downloads
5
Twitter
  Published on 24 Nov 2020
 Accepted on 01 Jul 2020            Submitted on 30 Aug 2019

Introduction

Residents of the Rio Grande Valley (RGV), Texas have among the worst health outcomes nationally. Rates of chronic disease such as obesity, diabetes, and related mortality in the RGV exceed those in most other regions of Texas and the nation [1, 2]. Poverty is pervasive, placing residents at high risk for poor health; they are more likely to be exposed to environmental hazards and have higher rates of chronic physical and mental health concerns [3, 4, 5].

At its foundation, integrated behavioral health care (IBH) aims to address multiple health concerns, and related social determinants, by bringing behavioral health and primary care services together. No single model of IBH can be applied universally to health or social service settings; however, each approach is team-based and involves collaboration between multidisciplinary health and social service providers to achieve shared patient and community outcomes [6]. The SAMSHA-HRSA Center for Integrated Health Solutions categorizes these approaches into three levels: coordinated, focused on communication between providers; co-located, models bringing different services into the same physical space; and integrated, which involves practice and workflow changes [7]. A 2016 update on research on integration of behavioral health and primary care emphasized the increasing strength of evidence highlighting the success of these approaches [8]. IBH has been effective in various populations [9, 10, 11, 12] with some studies focusing on specialty populations such as adolescents and those with substance use disorders or severe mental illness, which may need tailored and focused integrated care [13, 14, 15].

A growing body of evidence demonstrates IBH care approaches improve population health and access to health care in areas similar to South Texas [16, 17, 18]. In Austin, Texas for example, People’s Community Clinic used an IBH model to enable adult primary care patients diagnosed with depression and anxiety to receive psychiatric medication, counseling, and education. The program achieved results typically seen only in controlled clinical trials. The IBH model improved patients’ mental health outcomes with a minimal investment of resources [19]. Similarly, Bridges et al. revealed that Latinos participating in IBH care had significant improvements in symptoms and expressed high satisfaction with IBH treatment [20].

Despite the promise of integrated care strategies in improving physical and mental health, key gaps exist in evidence of the generalizability of these approaches in diverse populations. To the authors’ knowledge, no randomized control trials (RCT) of integrated care approaches have been conducted in free and charitable clinic settings or in clinics exclusively engaging volunteer primary care providers. Moreover, literature describing the impact of integrated care strategies on primarily Hispanic populations living in poverty at the US-Mexico border is sparse.

Hope Family Health Center (HOPE), a nonprofit free and charitable clinic in McAllen, Texas, provides medical and behavioral health services to uninsured patients living at or below the 200% federal poverty level (FPL). HOPE’s primary care providers are exclusively volunteers. In order to improve patient physical and mental health outcomes, HOPE adapted the collaborative care model to its clinic operations [19, 21]. HOPE aimed to remove barriers between behavioral and primary care through service co-location and integration supported by care management. Without effective intervention, it is likely individuals living in HOPE’s service area would not receive timely integrated care due to regional healthcare disparities, poverty, and lack of insurance.

In this RCT, we evaluated the effectiveness of HOPE’s integrated care model in a predominantly Hispanic, uninsured population receiving services from a free and charitable clinic located near the US-Mexico border. We hypothesized that integrated care would reduce blood pressure, HbA1c, body mass index, and depressive symptoms among adult patients with chronic illness living at or below the 200% FPL.

Theory & Methods

Study Setting & Design

HOPE provides free medical, counseling, and case management services to over 1,800 uninsured residents in the RGV annually, an area where almost one-third of the population are living in poverty [22]. The clinic is staffed with volunteer primary care providers and employed behavioral health counselors. Prior to the study, HOPE had co-located primary care and behavioral health services in the same physical space but had not fully integrated services through care coordination within that space.

We conducted an RCT to compare outcomes of intervention participants receiving HOPE’s IBH model with control participants receiving usual care. Participants enrolled in the study were followed for approximately 12 months and assessed for physical and behavioral health outcomes at baseline, 6- and 12-months. HOPE concurrently studied the implementation of their model as part of a separate research study [23].

Study Participants

All HOPE patients are uninsured and living at or below 200% of the FPL as eligibility criteria for either primary care or behavioral health services at the clinic. HOPE patients were eligible to participate in the study if they were receiving behavioral health services from HOPE at the time of enrollment; resided in Cameron, Hidalgo, Willacy, or Starr County; and had a diagnosis of one or more chronic conditions: hypertension (blood pressure ≥140/90 mmHg), obesity (body mass index ≥30.0), poorly controlled diabetes (HbA1c ≥ 6.8%), or moderate depression (Patient Health Questionnaire (PHQ-9) score ≥10).

HOPE identified potential participants by pre-screening their data system for eligibility criteria. Patients flagged as potentially eligible were then assessed for eligibility in person at their next visit. Patients meeting the eligibility criteria were invited to participate in the study. Those who agreed to participate underwent an informed consent process including the study overview, information about the randomization process, and an understanding that voluntary participation included taking a baseline and all follow-up assessments. The study research protocol was approved by New England Independent Review Board (Reference #120160447).

688 patients were identified as eligible to participate via pre-screening of records; 585 agreed to participate when approached by clinic staff and were allocated 1:1 to the intervention and control groups. Patients were excluded if they were pregnant or did not meet the eligibility criteria (Figure 1).

Figure 1 

Study Flow Diagram.

Randomization and Baseline Assessment

After the informed consent process, participants were allocated to the intervention or control group through simple random assignment using a computer-generated random number calculator and participant selection of number from individual, unmarked envelopes. After assignment, participants were seen by a nurse manager to complete baseline assessment, including the PHQ-9. A medical assistant assessed vitals including height, weight, and blood pressure. Baseline sociodemographic data were collected from HOPE’s data system. Two follow-up assessments were conducted in person when participants attended medical appointments to collect subsequent health data, at 6 months (midpoint) and 12 months (endpoint) after enrollment.

HOPE intended for intervention participants to receive care from a group of volunteer primary care providers separate from those treating the usual care participants. These individuals were to participate in the integrated program serving as “intervention only” providers. In practice, providers saw both intervention and control participants. This shift in design was primarily due to capacity of providers and HOPE’s priority to ensure patient care continuity. Additionally, volunteer providers’ variable schedules posed a challenge to implementing this aspect of the study design.

Integrated Behavioral Health Intervention

HOPE adapted the collaborative care model in December 2015 (Figure 2) [19, 21, 24, 25]. HOPE’s approach included activities from the Sanchez and Watt model that have been linked to improved health outcomes, including care management and access to behavioral health specialists [19]. In addition, HOPE’s program activities were adapted for their unique border community, including bilingual programming and psychoeducation.

Figure 2 

HOPE IBH Program Logic Model.

Participants randomized to the intervention group received enhanced integrated primary care and behavioral health services (Table 1). As part of the intervention, participants received a care plan developed collaboratively by providers of different services. These plans included at least one visit with a behavioral health specialist (BHS) and chronic care management visits with a transitional nurse. The addition of the BHS and transitional nurse enhanced the behavioral health services previously available at HOPE by integrating these services into the primary care workflow. The BHS conducted assessments, provided initial counseling, and coordinated referrals to care management and/or community-based health services. The BHS also made warm-handoffs as appropriate (e.g. to primary care or additional behavioral health counseling). The chronic care management visits focused on enabling patients to understand their chronic conditions (e.g. diabetes, obesity) and strategies to managing these conditions. The transitional nurse served as an additional resource to services addressing nutrition, exercise, smoking cessation, and support groups. Intervention participants also received services from a care coordinator who made warm-handoffs to either the BHS, transitional nurse, internal pharmacy services, or coordinated referrals to other support services available in the community.

Table 1

Availability of Services for Usual and Integrated Care Participants.

HOPE Usual Care HOPE Integrated Care

Coordinated information sharing (e.g. development of care plans, record keeping)
Primary care services
Individual and group behavioral health counseling services
Behavioral Health Specialist (BHS) assessments
BHS initial individual counseling and follow-up sessions
Chronic care management visits (with transitional nurse)
Referrals to social and health education services
Coordination of referrals to external social and health education services
Warm hand-offs between services within HOPE

Usual Care

Control group participants received HOPE’s usual care including primary care services and behavioral health counseling co-located within the clinic, but unlike the intervention participants, they were not offered the enhanced services from the transitional nurse or BHS. Control participants could be referred to additional specialty and community-based services. However, external referrals were not coordinated by the care coordinator and these participants did not receive warm-handoffs to other HOPE providers from the care coordinator.

Outcomes, Covariates, and Modifiers

The primary outcome for this study was blood pressure as measured by a medical assistant using a manometer [26]. Secondary outcomes were HbA1c (a measure of glucose control), body mass index (BMI), and depressive symptoms (assessed through PHQ-9 score) [27]. The PHQ-9 assessment was administered through provider interview. HbA1c was measured by blood test [28]. BMI was measured using a clinical scale and height tool [29].

We collected participant sociodemographic measures at baseline including age, sex (male; female), ethnicity (Hispanic; non-Hispanic), employment (employed; unemployed), marital status (married; not married, which included those who were divorced, partnered, separated, single, or widowed), and primary language spoken (English or Spanish). Clinic administrative data regarding the number and type of visits received by each participant were collected.

Statistical Analysis

Our primary analyses used an intent-to-treat approach to assess the difference in health outcomes between individual participants in the intervention and control groups at 12 months. T-tests were conducted to test within-group differences across time (baseline to endpoint) and between groups at study baseline and endpoint for each of the continuous outcomes. Chi square tests were used to assess equivalence at baseline. We then estimated multivariate linear regression models adjusting for important covariates and potential confounders identified in the assessment of baseline equivalence and bivariate analyses using backward elimination model selection for parsimony [30]; covariates were removed from the model where p > 0.15. Pearson correlation coefficients were calculated to identify collinearity among variables in each model. Effect sizes for statistically significant results were calculated using the Cohen’s d statistic. We subsequently explored the potential for effect modification by baseline health condition (e.g., baseline depression as an effect modifier for impact on PHQ-9 score at 12 months), age (below versus equal to or above the mean study age of 51 years), and sex. Stratified analyses were performed if significant interaction terms were detected. All analyses were conducted using SAS version 9.4 (Cary, NC), and differences where p < 0.05 were considered statistically significant.

All participants enrolled had complete baseline sociodemographic data collected. While efforts were made to collect full data for outcome measures at baseline, there were some challenges. Missing data for outcome measures blood pressure, BMI, and depressive symptoms ranged 1–21% at baseline and 11–19% at 12 months. Prior to analysis, we multiply imputed the missing outcome data with the mi procedure in SAS using all available baseline sociodemographic and health data, generating 10 datasets. Analyses of these four outcomes employed the MIANALYZE procedure in SAS. Because HbA1c level was not universally collected from participants, multiple imputation was not applied to the primary model of this outcome.

We conducted separate sensitivity analyses to assess for differential attrition by study group and potential contamination of intervention effect. We compared the proportion of participants who were lost to follow-up in the intervention versus control groups using a Chi-square test. Due to clinical staff and volunteer providers treating both intervention and control participants, there was concern about contamination of the control group. Thus, sensitivity analyses using a per protocol approach were conducted to verify the validity of results found in the intent-to-treat analyses following the approach described above. Clinic administrative data regarding visits with various providers were used to understand the extent participants received the level of care to which they were randomly assigned. For these analyses, intervention receipt was operationally defined as any evidence of care coordination in the medical record. Control participants with evidence of having received care coordination (n = 38) and intervention participants with no evidence of care coordination (n = 45) were removed from the sample for these sensitivity analyses.

Results

Baseline Participant Characteristics

Most participants were female (73.5%) and Hispanic (83.2%) with a mean age of 51 years. No statistically significant differences were identified between the intervention and control groups in terms of observed sociodemographic measures or health outcomes at baseline, except for PHQ-9 score (Table 2). The intervention group had a significantly higher mean PHQ-9 score (7.1) than the control group (5.0) at baseline (p = 0.001).

Table 2

Study Participant Characteristics at Baseline by Intervention Group.

Intervention (n = 270) Control (n = 312) p-value

n % n %

Demographics
Sex
      Female 199 73.7 229 73.4 0.93
      Male 71 26.3 83 26.6
Ethnicity
      Hispanic 217 80.4 267 85.6 0.09
      Non-Hispanic 53 19.6 45 14.4
Age, mean (SD) 51.2 (10.3) 50.6 (10.7) 0.51
Employment Status
      Employed 2 0.7 5 1.6 0.34
      Unemployed 268 99.3 307 98.4
Marital Status
      Married 135 50.0 161 51.6 0.70
      Not Married 135 50.0 151 48.4
Primary Language Spoken
      English 28 10.4 40 12.8 0.36
      Spanish 242 89.6 272 87.2
Health, mean (SD)
      Systolic Blood Pressure 133.4 (21.3) 134.5 (19.4) 0.48
      Diastolic Blood Pressure 81.0 (9.6) 82.0 (9.2) 0.17
      BMI 33.9 (7.5) 33.6 (6.6) 0.85
      PHQ-9 score 7.1 (7.1) 5.0 (6.3) 0.001
      HbA1c 8.0 (2.5) 7.6 (2.3) 0.12

370 participants completed a 12-month follow-up assessment; 172 intervention and 198 control. The intervention and control groups had the same rate of retention (63%; p = 0.95), indicating no differential attrition.

Intent-to-Treat Analysis

In adjusted analyses, the intervention and control groups differed significantly on depressive symptoms at 12-months; the groups were similar on HbA1c, blood pressure, and BMI (Table 3). On average, those in the intervention group had a PHQ-9 score that was 1.67 points lower than those in the control group (p = 0.01, d = 0.29).

Table 3

Effect of IBH Intervention on Health Outcomes at 12 Months, Intent-to-Treat Analyses.

n Intervention Mean (SD) Control Mean (SD) Intervention – Control Adjusted Mean Difference (SE) p-value

Systolic Blood Pressurea 370 128.4 (18.3) 130.0 (20.3) –2.47 (1.70) 0.15
Diastolic Blood Pressureb 370 78.1 (7.1) 79.0 (8.5) –0.93 (0.75) 0.22
HbA1cc 146 8.0 (1.9) 7.7 (1.9) –0.11 (0.24) 0.67
BMId 370 34.3 (7.7) 33.1 (6.9) 0.14 (0.22) 0.52
PHQ-9 scoree 370 4.6 (5.4) 5.3 (6.2) –1.67 (0.66) 0.01

a adjusted for age, sex, marital status, baseline systolic blood pressure, and baseline PHQ-9 score b adjusted for marital status, baseline diastolic blood pressure, and number of comorbidities at baseline c adjusted for baseline HbA1c level d adjusted for baseline BMI and baseline PHQ-9 score e adjusted for age, primary language, employment status, and baseline PHQ-9 score.

Per-Protocol Analysis

The results of sensitivity analyses were like the intent-to-treat analyses (Table 4). No statistically significant effects were detected for physical health measures; additionally, intervention participation was associated with a mean PHQ-9 score 1.64 points lower (p = 0.04) in the per protocol subsample.

Table 4

Effect of IBH Intervention on Health Outcomes at 12 Months, Per Protocol Analysis.

n Intervention – Control Adjusted Mean Difference (SE) p-value

Systolic Blood Pressurea 285 –1.11 (1.87) 0.55
Diastolic Blood Pressureb 285 –0.77 (0.76) 0.33
HbA1cc 102 –0.23 (0.28) 0.42
BMId 281 –0.01 (0.23) 0.97
PHQ-9 scoree 188 –1.64 (0.74) 0.04

a adjusted for age, marital status, baseline systolic blood pressure, and number of comorbidities at baseline b adjusted for age, baseline diastolic blood pressure, and number of comorbidities at baseline c adjusted for baseline HbA1c level d adjusted for baseline BMI e adjusted for baseline PHQ-9 score.

Effect Modification

Because primary and sensitivity analyses produced similar results, effect modification analyses were conducted using the intent-to-treat sample. Significant interactions were detected between sex and study group in analyses of 12-month blood pressure measures and HbA1c level. In sex-stratified models, the intervention was not statistically associated with blood pressure or HbA1c level for either sex (not shown). A significant interaction between study group and age was detected for 12-month PHQ-9 scores. In age-stratified models, intervention participants aged 51 years or older had a PHQ-9 score 2.08 points lower than those in the control group (p = 0.01, d = 0.34, Table 5) on average; however, the intervention effect was not significant among participants under 51 years of age. There were no significant interactions of study group and potential effect modifiers for 12-month BMI.

Table 5

Effect of IBH on PHQ-9 Score at 12 Months, Stratified by Mean Age Group, Intent-to-Treat Analysis.

n Intervention – Control Adjusted Mean Difference (SE) p-value

Participants 51+ Yearsa 206 –2.08 (0.81) 0.01
Participants Under 51 Yearsa 164 –1.34 (0.88) 0.13

a adjusted for baseline PHQ-9 score.

Discussion

This trial demonstrated that a behavioral health and care coordination intervention implemented in a free and charitable clinic with voluntary care providers can reduce depressive symptoms in uninsured patients living at or below the 200% FPL compared to standard care. When controlling for baseline measures and other covariates, intervention participants had significantly reduced PHQ-9 score at 12-months compared to control participants. These results align with findings from previous studies of IBH and its effect on mental health [31, 32]. The study also found evidence of effect modification by participation age, with greater impact identified among older intervention participants. This result has the potential to respond to the literature that has identified difficulties in treating depression in older adults and the need for approaches targeting symptoms in an older population [33, 34].

The finding that depressive symptoms changed but physical outcomes such as blood pressure did not is interesting and notable. HOPE’s integrated care approach emphasized brief intervention by the BHS and referral to counseling, which may be one mechanism by which depressive symptoms were mitigated. However, it is not surprising that physical outcomes did not change. First, there may have been insufficient time to observe a difference given the single year of follow up, and it is possible that physical changes may have occurred over a longer intervention period. Second, literature suggests that patients on anti-depressant drugs have increased risk of cardiovascular disease particularly if there is pre-existing cardiovascular disease [35]. However, we were unable to assess the impact of pharmacological intervention due to challenges in obtaining information about medications.

These findings fill an important gap in the integrated care literature by demonstrating the impact of behavioral health service integration with primary care in a previously understudied population. To our knowledge, this is the first RCT to examine an adapted integrated behavioral health model within the unique service and population characteristics of our context: a free and charitable clinic setting reliant on volunteer clinical providers serving uninsured predominately Hispanics living in poverty at the US-Mexican border. Our results demonstrate the positive impact integrated care can have on health outcomes.

The effects demonstrated in the trial occurred despite a lack of feasibility to implement two planned program components: psychiatry consultation and volunteer primary care provider engagement. Results suggest that psychiatric consultation was not necessary to improve depressive symptoms in a charitable clinic setting. Professional shortages are common in areas served by free and charitable clinics, and engagement of mental health professionals in an IBH intervention may improve feasibility and sustainability of the approach; however, additional research is needed to validate this finding and guide other clinics working to provide similar programming. Similarly, HOPE planned to engage volunteer primary care providers more directly in the intervention, but the episodic character of provider work at the clinic made this impossible. While the trial significantly improved mental health despite this, it is possible that physical health outcomes may have improved with more direct engagement of these providers.

The study has several limitations. Contamination occurred within the trial due to implementation challenges and the single clinic setting. However, sensitivity analyses suggested the impact of contamination was minimal and did not bias study results. Additionally, receipt of services outside of HOPE was not tracked and therefore not considered in these analyses. A limited number of patient characteristics were available for equivalence testing, and unmeasured differences between groups is possible. However, covariates included in our final models encompass biological and social factors known to be influential for health outcomes; additionally, randomization was successful based on the measured covariates, supporting the credibility of our results. Retention rates for both groups were lower than anticipated which may have limited our ability to see effects in some outcomes; however, no differential attrition was detected. The unique characteristics of our population and service setting may limit the generalizability of these results.

In conclusion, implementing an integrated care model with a focus on collaboration and communication between primary care and behavioral health providers in a free and charitable clinic setting can improve depressive symptoms after 12 months for individuals who are low-income or uninsured living in the border region of southern Texas. The results from this study can have funding implications when decisions on settings for IBH, or other innovative care models, are being made. Cost-effectiveness studies of HOPE’s approach would be very useful in understanding the feasibility of such models in similar clinics.

Reviewers

Two anonymous reviewers.

Competing Interests

AF, EG, and ES are employed by Health Resources in Action and this manuscript was completed during paid work hours. Health Resources in Action used grant funds from Methodist Healthcare Ministries who received federal funding for this project from the Corporation for National and Community Service, Social Innovation Fund for this project to pay for time used to conduct the study and prepare this manuscript. The grant also provided funds for travel for in-person meetings.

KSE was employed by Health Resources in Action and this manuscript was completed during paid work hours. Health Resources in Action used grant funds from Methodist Healthcare Ministries who received federal funding for this project from the Corporation for National and Community Service, Social Innovation Fund for this project to pay for time used to conduct the study and prepare this manuscript. The grant also provided funds for travel for in-person meetings.

RS was employed by Hope Family Health Center and this manuscript was completed during paid work hours. Hope Family Health Center used grant funds from Methodist Healthcare Ministries who received federal funding for this project from the Corporation for National and Community Service, Social Innovation Fund for this project to pay for time used to conduct the study and prepare this manuscript. The grant also provided funds for travel for in-person meetings.

References

  1. Centers for Disease Control and Prevention. Diagnosed Diabetes. 2016. Retrieved from: https://gis.cdc.gov/grasp/diabetes/DiabetesAtlas.html Accessed on: August 30, 2019. 

  2. Fisher-Hoch SP, Vatcheva KP, Laing ST, Hossain MM, Rahbar MH, Hanis CL, Brown HS, et al. Missed opportunities for diagnosis and treatment of diabetes, hypertension, and hypercholesterolemia in a Mexican American population, Cameron County Hispanic Cohort, 2003–2008. Prev Chronic Dis. 2012; 9(8). DOI: https://doi.org/10.5888/pcd9.110298 

  3. Cohen DA, Mason K, Bedimo A, Scribner R, Basolo V, Farley TA. Neighborhood physical conditions and health. Am J Public Health. 2003; 93(3): 467–71. DOI: https://doi.org/10.2105/AJPH.93.3.467 

  4. Diez Roux AV, Merkin SS, Arnett D, Chambless L, Massing M, Nieto FJ, et al. Neighborhood of residence and incidence of coronary heart disease. N Engl J Med. 2001; 345(2): 99–106. DOI: https://doi.org/10.1056/NEJM200107123450205 

  5. Quercia RG, Bates LK. The neglect of America’s housing: Consequences and policy responses. 18: Working Vibrant Neighborhoods, Successful Schools. 2002. 

  6. Lardieri MR, Lasky GB, Raney L. Essential Elements of Effective Integrated Primary Care and Behavioral Health Teams. Washington, DC: SAMHSA-HRSA Center for Integrated Health Solutions; 2014 Mar. Available here: https://www.integration.samhsa.gov/workforce/team-members/Essential_Elements_of_an_Integrated_Team.pdf. 

  7. Heath B, Wise RP, Reynolds K. A Review and Proposed Standard Framework for Levels of Integrated Healthcare. Washington, DC: SAMHSA-HRSA Center for Integrated Health Solutions. 2013 Mar. Available here: https://www.integration.samhsa.gov/integrated-care-models/A_Standard_Framework_for_Levels_of_Integrated_Healthcare.pdf. 

  8. Gerrity M. Evolving Models of Behavioral Health Integration: Evidence Update 2010–2015. New York, NY: Millbank Memorial Fund; 2016 May. Available here: https://www.milbank.org/wp-content/uploads/2016/05/Evolving-Models-of-BHI.pdf. 

  9. Bogner HR, de Vries HF. Integration of Depression and Hypertension Treatment: A Pilot, Randomized Controlled Trial. Ann Fam Med. 2008; 6(4): 295–301. DOI: https://doi.org/10.1370/afm.843 

  10. Foy R, Hempel S, Rubenstein L, Suttorp M, Seelig M, Shanman R, et al. Meta-analysis: Effect of Interactive Communication Between Collaborating Primary Care Physicians and Specialists. Ann Intern Med. 2010; 152(4): 247–258. DOI: https://doi.org/10.7326/0003-4819-152-4-201002160-00010 

  11. McGregor M, Lin EHB, Katon WJ. TEAMcare: An Integrated Multicondition Collaborative Care Program for Chronic Illnesses and Depression. J Ambul Care Manage. 2011; 34(2): 152–162. DOI: https://doi.org/10.1097/JAC.0b013e31820ef6a4 

  12. Balasubramanian BA, Cohen DJ, Jetelina KK, Dickinson LM, Davis M, Gunn R, et al. Outcomes of Integrated Behavioral Health with Primary Care. JABFM. 2017; 30(2): 130–139. DOI: https://doi.org/10.3122/jabfm.2017.02.160234 

  13. Richardson LP, McCarthy CA, Radovic A, Suleiman BS. Research in the Integration of Behavioral Health for Adolescents and Young Adults in Primary Care Settings: A Systematic Review. Journal of Adolescent Health. 2017; 60(3): 261–269. DOI: https://doi.org/10.1016/j.jadohealth.2016.11.013 

  14. Druss BG, von Esenwein SA, Glick GE, Deubler E, Lally C, Ward MC, et al. Randomized Trial of an Integrated Behavioral Health Home: The Health Outcomes Management and Evaluation (HOME) Study. Am J Psychiatry. 2017; 174: 246–255. DOI: https://doi.org/10.1176/appi.ajp.2016.16050507 

  15. Pollard RQ, Betts WR, Carroll JK, Waxmonksky JA, Barnett S, DeGruy FV, et al. Integrating Primary Care and Behavioral Health With Four Special Populations: Children with Special Needs, People With Serious Mental Illness, Refugees, and Deaf People. American Psychologist. 2014; 69(4): 377–387. DOI: https://doi.org/10.1037/a0036220 

  16. Bedoya CA, Traeger L, Trinh NHT, Chang TE, Brill CD, Hails K, et al. Impact of a culturally focused psychiatric consultation on depressive symptoms among Latinos in primary care. Psychiatric Services. 2014; 65(10): 1256–1262. DOI: https://doi.org/10.1176/appi.ps.201300088 

  17. Camacho Á, González P, Castañeda SF, Simmons A, Buelna C, Lemus H, et al. Improvement in Depressive Symptoms Among Hispanic/Latinos Receiving a Culturally Tailored IMPACT and Problem-Solving Intervention in a Community Health Center. Community Mental Health Journal. 2015; 51(4): 385–392. DOI: https://doi.org/10.1007/s10597-014-9750-7 

  18. Ell K, Katon W, Xie B, Lee PJ, Kapetanovic S, Guterman J, et al. Collaborative care management of major depression among low-income, predominantly Hispanic subjects with diabetes: a randomized controlled trial. Diabetes Care. 2010; 33(4): 706–713. DOI: https://doi.org/10.2337/dc09-1711 

  19. Sanchez K, Watt TT. Collaborative care for the treatment of depression in primary care with a low-income, Spanish-speaking population: Outcomes from a community-based program evaluation. Prim Care Companion CNS Disord. 2012; 14(6). DOI: https://doi.org/10.4088/PCC.12m01385 

  20. Bridges AJ, Andrews III AR, Villalobos BT, Pastrana FA, Cavell TA, Gomez D. Does Integrated Behavioral Health Care Reduce Mental Health Disparities for Latinos? Initial Findings. Journal of Latina/o Psychology. 2014; 2(1): 37–53. DOI: https://doi.org/10.1037/lat0000009 

  21. Watt T. A Process and Outcome Evaluation of Two Integrated Behavioral Health Care Models: People’s Community Clinic and Lone Star Circle of Care. 2009. Retrieved June 12, 2015, from https://www.austinpcc.org/documents/specialprograms/IBH_year_3_report.pdf. 

  22. US Census. QuickFacts: Willacy County, Texas; Starr County, Texas; Hidalgo County, Texas; Cameron County, Texas. 2018. Retrieved from: https://www.census.gov/quickfacts/fact/table/cameroncountytexas,hidalgocountytexas,willacycountytexas,starrcountytexas/PST045218 Accessed on: August 30, 2019. 

  23. Gaitan E, Auerbach A, Stocker R, Marquez N, Sautter Errichetti K. Integrating Behavioral Health and Primary Care at the US-Mexico Border: Results from an Implementation Study from a Safety Net Clinic with Volunteer Providers. Under Review. 

  24. Guide to Community Preventive Services. Improving mental health and addressing mental illness: collaborative care for the management of depressive disorders. 2010. Retrieved June 12, 2015, from http://www.thecommunityguide.org/mentalhealth/collab-care.html. 

  25. Gilbody S, Bower P, Fletcher J, Richards D, Sutton AJ. Collaborative care for depression: A cumulative meta-analysis and review of longer-term outcomes. Arch Intern Med. 2006; 166(21): 2314–2321. DOI: https://doi.org/10.1001/archinte.166.21.2314 

  26. UK NCGC. Hypertension. UK: Royal College of Physicians; 2011. http://www.ncbi.nlm.nih.gov/pubmed/22855971. Accessed July 8, 2019. 

  27. Kroenke K, Spitzer RL, Williams JB. The PHQ-9: validity of a brief depression severity measure. J Gen Intern Med. 2001; 16(9): 606–613. DOI: https://doi.org/10.1046/j.1525-1497.2001.016009606.x 

  28. Gonen B, Rubenstein A, Rochman H, Tanega SP, Horwitz DL. Haemoglobin A1: An indicator of the metabolic control of diabetic patients. Lancet (London, England). 1977; 2(8041): 734–737. DOI: https://doi.org/10.1016/S0140-6736(77)90237-9 

  29. UK NCGC. Obesity. UK: National Institute for Health and Care Excellence; 2014. http://www.ncbi.nlm.nih.gov/pubmed/25535639. Accessed July 8, 2019 

  30. Liebschutz JM, Xuan Z, Shanahan CW, LaRochelle Keosaian J, Beers D, Guara G, et al. Improving Adherence to Long-term Opioid Therapy Guidelines to Reduce Opioid Misuse in Primary Care. JAMA Intern Med. 2017; 177(9): 1265–1272. DOI: https://doi.org/10.1001/jamainternmed.2017.2468 

  31. Balasubramanian BA, Cohen DJ, Jetelina KK, Dickinson LM, Davis M, Gunn R, et al. Outcomes of Integrated Behavioral Health with Primary Care. JABFM. 2017; 30(2): 130–139. DOI: https://doi.org/10.3122/jabfm.2017.02.160234 

  32. Eghaneyan BH, Sanchez K, Killian M. Integrated health care for decreasing depressive symptoms in Latina women: Initial findings. Journal of Latina/o Psychology. 2017; 5(2): 118–125. DOI: https://doi.org/10.1037/lat0000067 

  33. Fiske A, Wetherell JL, Margaret G. Depression and Older Adults. Ann Rev Clin Psy. 2009; 5: 363–389. DOI: https://doi.org/10.1146/annurev.clinpsy.032408.153621 

  34. Licht-Strunk E, Van Marwijk HWJ, Hoekstra T, Twisk JWR, De Haan M, Beekman ATF. Outcome of depression in later life in primary care: Longitudinal cohort study with three years’ follow-up. BMJ (Online). 2009; 338(7692): 463–466. DOI: https://doi.org/10.1136/bmj.a3079 

  35. Waring WS. Clinical use of antidepressant therapy and associated cardiovascular risk. Drug Healthc Patient Saf. 2012; 4: 93–101. DOI: https://doi.org/10.2147/DHPS.S28804