Research ArticleOriginal ResearchA

Community Health Workers as Trust Builders and Healers: A Cohort Study in Primary Care

Robert L. Ferrer, Carolina Gonzalez Schlenker, Inez Cruz, Polly Hitchcock Noël, Raymond F. Palmer, Ramin Poursani and Carlos Roberto Jaén
The Annals of Family Medicine September 2022, 20 (5) 438-445; DOI: https://doi.org/10.1370/afm.2848

Abstract

PURPOSE Improving patients’ self-care for chronic disease is often elusive in the context of social deprivation. We evaluated whether a practice-integrated community health worker (CHW) intervention could encourage effective long-term self-management of type 2 diabetes mellitus (T2DM).

METHODS This cohort study, in a safety-net primary care practice, enrolled patients with uncontrolled T2DM and psychosocial risk factors. Patients were identified through a practice diabetes registry or by clinicians’ referrals. The CHWs engaged patients in trust building and sensemaking to understand their social context, identify goals, navigate health care, and connect to community resources. Primary outcome was progress through 3 prospectively defined stages of self-care: outreach (meeting face-to-face); stabilization (collaborating to address patients’ life circumstances); and self-care generativity (achieving self-care competencies). Secondary outcomes were change in hemoglobin A1c (HbA1c) and need for urgent care, emergency department, or hospital visits.

RESULTS Of 986 participating patients, 27% remained in outreach, 41% progressed to stabilization, and 33% achieved self-care generativity. Repeated measures ANOVA demonstrates an overall decline in HbA1c, without group differences, through the 4th HbA1c measurement (mean follow-up 703 days). Beginning at the 5th HbA1c measurement (mean 859 days), the self-care generativity group achieved greater declines in HbA1c, which widened through the 10th measurement (mean 1,365 days) to an average of 8.5% compared with an average of 8.8% in the outreach group and 9.0% in the stabilization group (P = .003). Rates of emergency department and hospital visits were lower in the self-care generativity group.

CONCLUSIONS Practice-linked CHWs can sustainably engage vulnerable patients, helping them advance self-management goals in the context of formidable social disadvantage.

Key words:

INTRODUCTION

Chronic diseases are the leading cause of disability and mortality, accounting for 73% of deaths globally and 88% nationally (United States).1 Burdens of chronic disease fall most heavily on socioeconomically disadvantaged populations, meaning that those with greatest need for care have the fewest resources.2

A prime example of the remarkable acceleration of chronic disease and associated disparities is type 2 diabetes mellitus (T2DM).3–5 US diabetes prevalence increased from less than 1% in 1958 to 11% in 2020,6 with increasingly steep social gradients in health.

Avoiding diabetes complications such as ischemic heart disease, renal insufficiency, and limb amputation is more likely if glycosylated hemoglobin, blood pressure, and serum lipids are controlled.7–10 Unfortunately, many patients living with social disadvantage fail to achieve targets for risk markers and suffer a disproportionate share of complications.11–13

Successful management of T2DM requires patients to actively participate in their care, but their efforts are undermined by poverty, lack of insurance, or difficult home and neighborhood environments.14 Implementing self-care strategies is an important function of primary care that provides a platform for reaching populations across the socioeconomic spectrum.15,16 Clinicians, however, often lack effective systems to help patients thrive in difficult circumstances.17

In response, primary care practices have deployed new care models. A common strategy is to engage interdisciplinary teams, which may include nurse care managers, health educators, behavioral health clinicians, or community health workers (CHWs).18 A large body of literature supports the role of CHWs in chronic disease management, but the relatively short duration of randomized trials (typically 6-12 months) limits our understanding of longer-term outcomes.19 Importantly, the optimal design and duration of the role of CHWs in different practice and community contexts has yet to be established.20,21

This cohort study evaluated 4-year outcomes of a primary care intervention for T2DM control in a predominantly Latine, inner-city cohort. The intervention emphasized skills for self-care in home and community, building trust with the health care team, and acknowledging the impact of living under conditions of imposed scarcity and injustice. This study assessed patients’ engagement with the intervention and followed hemoglobin A1c (HbA1c) levels and needed urgent care, emergency department, or hospital visits over 4 years.

METHODS

Context

We evaluated a quality improvement initiative in a safety-net primary care practice in San Antonio, Texas.22 Launched in 2011 with financial support from Bexar County, Texas, the initiative transitioned in 2012 to Medicaid 1115 Waiver funding from the Center for Medicare and Medicaid Services to the Texas Department of Health and Human Services (DHHS). Medicaid 1115 Waivers allow organizations to use state Medicaid funds for specified activities beyond traditional medical services.23,24

The Texas DHHS approved goal for this study was to decrease the proportion of patients with uncontrolled diabetes mellitus, defined as HbA1c greater than 9% (National Quality Forum measure 005925). The Institutional Review Board at UT Health San Antonio reviewed the protocol, classifying it as quality improvement and therefore not human subjects research.

Intervention Design and Theory of Change

The intervention was designed as quality improvement, with pre and post analysis, and the ability to evolve through iterative learning. Its design drew on several theoretical frameworks relevant to primary care. Antonovsky’s concept of salutogenesis26 asserts healthy people live lives that are comprehensible, manageable, and meaningful. A meaningful clinical encounter connects with a person’s sense of purpose. From Human Development economics,27–29 we identified a core goal of expanding people’s feasible opportunities to be and do the things they value. And drawing on Freire,30 we placed trust at the center of transformative human interactions, while acknowledging the importance of naming past injustices. The Supplemental Appendix contains additional notes on the framework.

Setting and Participants

The study took place in San Antonio, Texas, where 15.5% of 1.56 million residents have been diagnosed with diabetes mellitus31 with prevalence varying threefold across income and education strata. Among the 10,000 patients in the Family and Community Medicine practice, point-in-time estimates identified approximately 2,500 with T2DM. Resident and faculty physicians practice alongside nurse care managers, behavioral health consultants, clinical pharmacists, and medical assistants who participate in care management activities. These professionals are available to both clinicians and CHWs.

We identified potential participants from a practice registry of patients with T2DM, listing those with HbA1c of 9% or higher, and accepted other clinicians’ referrals of patients having difficulty controlling their diabetes. From October 1, 2013 through September 30, 2017, of the 8,647 patients in our diabetes registry we identified 1,270 eligible patients who agreed to meet with a CHW. We excluded 237 patients who received only brief, task-oriented assistance (eg, scheduling an appointment), and 47 patients who did not have a period of care recorded at disenrollment, or had HbA1c less than 6.5% at baseline, or no HbA1c results post baseline. This yielded a cohort of 986 people. For patients referred more than once during the 4 years (n = 34), we only included data from their last enrollment.

Initial meetings with CHWs occurred at patients’ homes, the practice, or another venue chosen by the patient. CHWs then engaged patients in a series of face-to-face meetings we called Nosotros [“we”] events, designed to foster trust and patient progress by iteratively cycling through 4 questions: What do we want? Why do we want it? What do we do? What do we get?

The first 2 questions focus on patients’ desired outcomes and motivation. The second 2 questions establish a course of action and its link to desired outcomes. Patients’ goals could be disease specific (“I don’t want to end up on dialysis”) or more general (“I need a part-time job to help pay for groceries and meds.”) Questions are posed in the first-person plural to emphasize commitment to the Nosotros from both patient and CHW.

Patients’ intermediate goals were framed in a rubric, the 6-piece puzzle (Supplemental Appendix), which includes healthy food, sufficient activity, appropriate medication use, active participation in health care, understanding the numbers measuring self-management success, and developing trust in key partners. Assessing practical opportunities for healthy behaviors is a starting point for planning change.

Community health workers were not limited to assisting patients with their diabetes management and engaged help as needed for other patient concerns. For instance, CHWs were able to pass patient concerns to the care team regarding other unresolved health or social issues.

Patients and CHWs negotiated their encounter frequency based on patients’ needs. To describe patients’ trajectories, CHWs conceptualized 3 periods of care, each with observable behavioral end points that were prospectively assigned by the CHWs. The periods of care were: (1) outreach where the patient and CHW agreed to meet face-to-face; (2) stabilization in which the CHW and patient identified and addressed barriers to self-care; and (3) self-care generativity in which the patients displayed motivation for self-care and demonstrated the skills to manage diabetes in their home (Table 1). These activities often had an unexpected, original nature transcending external advice. Community health workers planned to work with patients for 12 weeks, though in practice the range was flexible. The CHWs classified patients’ progress through the 3 care periods in real time.

View this table:
Table 1.

Periods of Care Definitions

Community health workers were trained by a physician (C.G.S.) with extensive experience working with CHWs in different settings.32 The CHWs assembled weekly, for 1 half-day, for training, reflection, and program adaptations. They also met one-on-one with the physician (C.G.S.) weekly to formulate approaches to their patient panel.

Program Evaluation

We examined process, clinical, and utilization outcomes for patients enrolled from October 1, 2013 through September 30, 2017, retaining the full cohort in the analytical data set.

To describe the features of self-care generativity, 3 study investigators reviewed the primary qualitative data from CHW narratives that signaled patients’ progress to self-care generativity. Data on common themes was discussed to reach consensus and then an investigator checked back with CHWs to confirm working themes were congruent with CHWs’ understanding. Investigators were blinded to HbA1c values and health services outcomes while defining the periods of care variables. Once the 3 periods of care were operationally defined (Table 1), we assessed patients’ progress through the periods and the relationship to diabetes control and health care events. No further revisions were made to the criteria for assigning patients to specific periods of care.

Process variables for each patient included the number of CHW encounters, the duration spent with a CHW, and CHWs’ narrative observations of each encounter. The primary clinical outcome required for our Delivery System Reform Incentive Payment project was HbA1c level less than 9%. Hemoglobin A1c predicts microvascular complications with consequences for length and quality of life.33 To quantify potentially serious health events during follow-up, we also examined pre- and post-intervention trends in visits to urgent care, emergency departments, and hospitals.

Statistical Methods

In the quality improvement context, the timing of HbA1c measurements was determined by clinicians rather than a study timetable. We examined HbA1c trends averaged over sequential observations, 1st HbA1c, 2nd HbA1c, etc (measurement episodes) following each patient’s enrollment with their CHW. Supplemental Appendix displays mean days elapsed from study entry for successive measurement episodes. In the analysis, patients appear only once and were assigned to the most advanced period of care cohort attained. To determine whether hierarchical models were needed to account for patient clustering within CHWs, we examined intraclass correlation coefficients (ICC), noting negligible level-2 variance (estimated ICC for HbA1c was 0.00; for urgent care, 0.021; for emergency department, 0.005; and for hospital visit, 0.008).

We assessed between-group HbA1c differences using repeated measures ANOVA with period of care as the between-group factor and HbA1c assessments as the within-group repeated measure.

Count data for urgent care, emergency department, and hospital visits were markedly skewed with a predominance of zero values. A test of overdispersion for each health services variable in a Poisson distribution found greater variance than expected (P <.001). A negative binomial model (allowing the variance to be estimated separately from the mean) was a better fit than a zero-inflated model. We therefore applied negative binomial regressions to predict the number of health care visits for each period of care group in the 12 months following CHW intervention, adjusting for the count of health care visits in the previous 12 months. Models were adjusted for covariates that we expected a priori to be important predictors: age, sex, health insurance, immigrant status, and language preference.

RESULTS

Table 2 displays patient demographics. The 3 cohorts were well balanced with respect to insurance status and baseline HbA1c. Patients in the stabilization group were slightly older. There were proportionately more females and undocumented persons in the self-care generativity group than the stabilization group. The self-care group had proportionately fewer participants with English as their primary language.

View this table:
Table 2.

Cohort Demographics and Baseline Hemoglobin A1c Values

Differences in HbA1c Over Time

Mean HbA1c at entry was between 10.1% and 10.5% in the 3 groups (P values for differences not significant). All 3 groups then demonstrated a significant drop in HbA1c (Figure 1), without group differences, through the 4th HbA1c measurement (mean follow-up 859 days). Subsequently, the self-care group declined relative to the other groups (P = .03), reaching mean HbA1c less than 9% by the 7th measurement (mean follow up 1,133 days). This trend continued through the 10th measurement (P = .003), where the HbA1c in the self-care group continued to decline, reaching an average of 8.49% (mean follow up 1,365 days). Covariates in the model included age, sex, health insurance status, language preference, and immigration status. Figure 1 depicts the HbA1c trends, with Table 3 listing sample sizes for sequential measurement episodes.

Figure 1.
Figure 1.

Glycoslyated hemoglobin trajectories by care periods.

HbA = hemoglobin A

Note: Marginal means for HbA are estimated for sequential measurement episodes during follow-up. Covariates appearing in the model were evaluated at age 55.7 and insurance 0.59. Difference between outreach and stabilization groups significant at P <.001 and between outreach and self-care generatively groups significant at P <.002.

View this table:
Table 3.

Sequential HbA1c Measurement Episodes and Elapsed Days

Health Care Utilization Outcomes (Table 4)

View this table:
Table 4.

Health Care Utilization Outcomes for Negative Binomial Models

In our statistical model, we estimated separate negative binomial models for each health care outcome, by period of care group, adjusting for age, sex, insurance status, language preference, immigration status, and health care utilization in the 12 months before intervention. Exponentiating the estimated raw coefficients generated incidence-rate ratios (IRR; or incidence rates, in the case of the intercept). Figure 2 displays predicted counts of visits to urgent care, emergency department, and hospitals for each cohort.

Figure 2.
Figure 2.
Figure 2.

Predicted average count of health care visits, unadjusted and adjusted.

a Values were adjusted for age, sex, health insurance status, and prior-year utilization.

In analyses adjusted for age, sex, health insurance status, immigrant status, language preference, and prior year utilization (Supplemental Appendix), IRR for hospital visits were twofold higher in the stabilization group than in self-care. Hospital visits in the outreach group were 6% higher than self-care, but this was not significant. Hospital visits were 90% higher in stabilization than in outreach (P = .014) (not shown in the Supplemental Appendix).

Compared with the self-care group, the average count of emergency department visits was 74% higher in the stabilization group (P = <.001) and 31% higher in the outreach group (P = .03). There were no significant differences in IRRs across groups for urgent care visits.

Quantifying Regression to the Mean

A major concern in observational studies is regression to the mean, the tendency for sampled point-in-time values to become less extreme on follow-up assessment.34 To assess this, we analyzed a 5-year data set (that included the study years) of over 27,000 HbA1c measurements from our practice. Among patients whose initial HbA1c was greater than 9%—the eligibility cutoff for our study—and followed for intervals similar to our CHW intervention (6 to 36 months), absolute HbA1c reductions averaged −0.055% by the 10th HbA1c measurement. In comparison, the intervention group’s decrease was −2% (absolute). Over shorter time frames, HbA1c tended to decrease, but much less than we observed in our study population. For patients in the practice data set for 6-24 months prior to the start of this study, mean change in HbA1c was −0.37% (absolute). Regression to the mean is therefore unlikely to account for a substantial fraction of the HbA1c trends observed.

DISCUSSION

We evaluated program effects within a large, multi-year cohort of patients with uncontrolled T2DM. Community health workers were able to help 32% of patients attain a behaviorally defined level of self-management we named “self-care generativity” that was associated with sustained, clinically meaningful reductions in HbA1c for up to 4 years, and a decreased risk of emergency department visits and hospitalization. Another 41% achieved an intermediate level of engagement and HbA1c reduction, while 27% did not engage with the intervention.

The intervention was integrated within a safety-net primary care practice prepared to respond to patients’ social adversity by integrating CHWs, behavioral health clinicians, nurse care managers, and medical assistants into population management. Community health workers’ detailed understanding of patients’ circumstances help to tailor their care rather than apply fixed interventions. That flexibility required a CHW workforce prepared to improvise in response to the challenging environments they navigated.

Our evaluation applied a naturalistic design, identifying patients with HbA1c of 9% or more from a practice registry. Clinicians could also refer patients not making progress in glycemic control. The intervention evolved through continuous quality improvement as the team reviewed processes and outcomes.35 Outcomes assessment relied on objective end points of HbA1c values and health care events. Selection bias across the 3 cohorts at entry was mitigated by all included patients initially agreeing to work with a CHW.

Still, the findings must be interpreted in the context of a quality improvement intervention rather than a randomized trial.36 Although the study population was not subject to the selection filter of agreeing to participate in a randomized controlled trial (RCT), our results could be influenced by the biases of nonrandomized intervention studies. Accepting help from a CHW may indicate general willingness to engage in health-promoting behaviors. Attrition over the 4-year study could have selected a population with greater commitment to engage with health care, or higher attrition may have occurred among patients experiencing poor outcomes. It should be noted, however, that we did not select patients based on a discrete event—such as a hospital visit—likely to mark a temporary decline in health from which recovery was likely.37 Also, reviews comparing cohort studies with RCTs find relatively small differences in effect estimates, with the ratio of the odds ratios calculated to be 1.04 (95% CI, 0.89-1.21).38

We attempted to quantify expected regression to the mean in longitudinal data from our patients with T2DM and elevated HbA1c, and found it was 25% the magnitude of the observed main effect.

The time necessary to achieve clinically meaningful results is not surprising among patients facing multiple challenges, including unmet social needs. Time was needed for the practice to understand how to help specific patients and for patients to sustain new habits in old settings. CHWs met weekly to discuss continuous quality improvement and potential program adaptations.

This study adds to a growing body of literature on the value of CHWs in chronic disease management.39 Systematic reviews of CHW interventions for T2DM have generally concluded that interventions are effective, often with modest effects and heterogeneity across studies. For example, a Task Force on Community Preventive Services review of 44 CHW interventions for diabetes estimated a median HbA1c decrease of 0.49%, (interquartile range, −0.76 to −0.27) but median study duration was just 6-12 months, leaving questions about those interventions’ durability.18,19,40,41 As most studies of CHW interventions report results from less than a year of observation, there may be insufficient time allotted for adaptive learning by CHWs. We observed positive effects up to 4 years.

Our reflections on what worked are consistent with a review of CHW interventions,19 emphasizing community embeddedness, supportive supervision, continuous education, cultural congruence,42 and adequate logistical support. We add our perspective that the change agent is the sense-making collaboration between the CHW and patient. Also important was CHWs coordination with practitioners, who benefitted from CHWs’ observations of patients’ homes, families, and neighborhoods, that provided the contextual details with the power to stall or support progress.43

Despite promise, CHW programs can be difficult to maintain with time-limited funding. Programs’ ability to demonstrate cost savings44 may promote sustainability. After the Delivery System Reform Incentive Payment funding ended, our CHW program has been partially supported under pay-for-performance contracts with Medicaid and Medicare managed care payers. We would argue that investment in CHWs should not be strictly contingent on demonstrating savings. Community Health Workers help advance patients’ health and promote health equity,45 making them essential, fully integrated members of the primary care team. Community health workers leverage personal and professional experience to navigate social complexity, raising both practitioners’ and patients’ normative expectations for primary care practice.47,48

CONCLUSIONS

We evaluated a multi-year intervention for uncontrolled T2DM in a safety-net primary care practice and found sustained HbA1c reductions and fewer emergency department and hospital visits. These outcomes suggest a path to financial sustainability. Also, it is important to acknowledge the limits of applying individual-level interventions to health conditions whose origins lie in social injustices.50,51 Community-level change is necessary to address root causes.52,53

Footnotes

  • Conflicts of interest: The authors have no conflicts of interest related to the manuscript. The funder had no role in study design, data collection, analysis or interpretation of data, manuscript preparation, or the decision to submit the manuscript. All authors had access to the study data and approved the final manuscript. Dr Ferrer is an associate editor at Annals of Family Medicine.

  • Read or post commentaries in response to this article.

  • Funding support: Texas Department of Health and Human Services, Texas Delivery System Reform Incentive Payment.

  • Previous presentation: “Nosotros” Approach: Community Health Workers as Trust Builders and Community Healers. Ferrer RL, Schlenker CG, Poursani RS, Cruz I, Noel P, Palmer R, Jaén CR. North American Primary Care Research Group Annual Meeting; November 16-20, 2019; Toronto, Canada.

  • Supplemental materials

  • Received for publication August 24, 2021.
  • Revision received February 8, 2022.
  • Accepted for publication May 10, 2022.

REFERENCES

  1. 1.↵
    1. Bauer UE,
    2. Briss PA,
    3. Goodman RA,
    4. Bowman BA.
    Prevention of chronic disease in the 21st century: elimination of the leading preventable causes of premature death and disability in the USA. Lancet. 2014; 384(9937): 45-52. doi:10.1016/S0140-6736(14)60648-6
    OpenUrlCrossRefPubMed
  2. 2.↵
    1. Shaw KM,
    2. Theis KA.
    3. Self-Brown S,
    4. Roblin DW,
    5. Barker L.
    Chronic disease disparities by county economic status and metropolitan classification, behavioral risk factor surveillance system, 2013. Prev Chronic Dis. 2016; 13:E119. doi:10.5888/pcd13.160088
    OpenUrlCrossRef
  3. 3.↵
    1. Saydah S,
    2. Lochner K.
    Socioeconomic status and risk of diabetes-related mortality in the U.S. Public Health Rep. 2010; 125(3): 377-388. doi:10.1177/003335491012500306
    OpenUrlCrossRefPubMed
  4. 4.
    1. Zimmet PZ,
    2. Magliano DJ,
    3. Herman WH,
    4. Shaw JE.
    Diabetes: a 21st century challenge. Lancet Diabetes Endocrinol. 2014; 2(1): 56-64. doi:10.1016/S2213-8587(13)70112-8
    OpenUrlCrossRefPubMed
  5. 5.↵
    1. NCD Risk Factor Collaboration (NCD-RisC)
    . Worldwide trends in diabetes since 1980: a pooled analysis of 751 population-based studies with 4.4 million participants. Lancet. 2016; 387(10027): 1513-1530. doi:10.1016/S0140-6736(16)00618-8
    OpenUrlCrossRefPubMed
  6. 6.↵
    1. Centers for Disease Control and Prevention
    . National diabetes statistics report 2020. Published 2020. Accessed Aug 22, 2021. https://www.cdc.gov/diabetes/library/features/diabetes-stat-report.html
  7. 7.↵
    1. Turnbull FM,
    2. Abraira C,
    3. Anderson RJ, et al; Control Group
    . Intensive glucose control and macrovascular outcomes in type 2 diabetes. Diabetologia. 2009; 52(11): 2288-2298. doi:10.1007/s00125-009-1470-0
    OpenUrlCrossRefPubMed
  8. 8.
    1. Hasan R,
    2. Firwana B,
    3. Elraiyah T, et al.
    A systematic review and meta-analysis of glycemic control for the prevention of diabetic foot syndrome. J Vasc Surg. 2016; 63(2): 22S-28S.e2. doi:10.1016/j.jvs.2015.10.005
    OpenUrlCrossRef
  9. 9.
    1. Qaseem A,
    2. Wilt TJ,
    3. Kansagara D, et al; Clinical Guidelines Committee of the American College of Physicians
    . Hemoglobin A1c targets for glycemic control with pharmacologic therapy for nonpregnant adults with type 2 diabetes mellitus: a guidance statement update from the American College of Physicians. Ann Intern Med. 2018; 168(8): 569-576. doi:10.7326/M17-0939
    OpenUrlCrossRefPubMed
  10. 10.↵
    1. UK Prospective Diabetes Study (UKPDS) Group
    . Effect of intensive blood-glucose control with metformin on complications in overweight patients with type 2 diabetes (UKPDS 34). Lancet. 1998; 352(9131): 854-865.
    OpenUrlCrossRefPubMed
  11. 11.↵
    1. Heidemann DL,
    2. Joseph NA,
    3. Kuchipudi A,
    4. Perkins DW,
    5. Drake S.
    Racial and economic disparities in diabetes in a large primary care patient population. Ethn Dis. 2016; 26(1): 85-90. doi:10.18865/ed.26.1.85
    OpenUrlCrossRef
  12. 12.
    1. Rawshani A,
    2. Rawshani A,
    3. Franzén S, et al.
    Risk factors, mortality, and cardiovascular outcomes in patients with type 2 diabetes. N Engl J Med. 2018; 379(7): 633-644. doi:10.1056/NEJMoa1800256
    OpenUrlCrossRefPubMed
  13. 13.↵
    1. Skrepnek GH,
    2. Mills JL Sr.,
    3. Armstrong DG.
    A diabetic emergency one million feet long: disparities and burdens of illness among diabetic foot ulcer cases within emergency departments in the United States, 2006-2010. PLoS One. 2015; 10(8): e0134914. doi:10.1371/journal.pone.0134914
    OpenUrlCrossRefPubMed
  14. 14.↵
    1. Spanakis EK,
    2. Golden SH.
    Race/ethnic difference in diabetes and diabetic complications. Curr Diab Rep. 2013; 13(6): 814-823. doi:10.1007/s11892-013-0421-9
    OpenUrlCrossRefPubMed
  15. 15.↵
    1. Starfield B,
    2. Shi L,
    3. Macinko J.
    Contribution of primary care to health systems and health. Milbank Q. 2005; 83(3): 457-502. doi:10.1111/j.1468-0009.2005.00409.x
    OpenUrlCrossRefPubMed
  16. 16.↵
    1. Ferrer RL.
    Pursuing equity: contact with primary care and specialist clinicians by demographics, insurance, and health status. Ann Fam Med. 2007; 5(6): 492-502. doi:10.1370/afm.746
    OpenUrlAbstract/FREE Full Text
  17. 17.↵
    1. DeVoe JE,
    2. Bazemore AW,
    3. Cottrell EK, et al.
    Perspectives in primary care: a conceptual framework and path for integrating social determinants of health into primary care practice. Ann Fam Med. 2016; 14(2): 104-108. doi:10.1370/afm.1903
    OpenUrlFREE Full Text
  18. 18.↵
    1. Hartzler AL,
    2. Tuzzio L,
    3. Hsu C,
    4. Wagner EH.
    Roles and functions of community health workers in primary care. Ann Fam Med. 2018; 16(3): 240-245. doi:10.1370/afm.2208
    OpenUrlAbstract/FREE Full Text
  19. 19.↵
    1. Community Preventive Services Task Force
    . Diabetes Management: Interventions Engaging Community Health Workers. Published Apr 2017. https://www.thecommunityguide.org/sites/default/files/assets/Diabetes-Management-Community-Health-Workers.pdf
  20. 20.↵
    1. Palmas W,
    2. March D,
    3. Darakjy S, et al.
    Community health worker interventions to improve glycemic control in people with diabetes: a systematic review and meta-analysis. J Gen Intern Med. 2015; 30(7): 1004-1012. doi:10.1007/s11606-015-3247-0
    OpenUrlCrossRefPubMed
  21. 21.↵
    1. Jacob V,
    2. Chattopadhyay SK,
    3. Hopkins DP, et al.
    Economics of community health workers for chronic disease: findings from community guide systematic reviews. Am J Prev Med. 2019; 56(3): e95-e106. doi:10.1016/j.amepre.2018.10.009
    OpenUrlCrossRefPubMed
  22. 22.↵
    1. Ferrer RL,
    2. Gonzalez Schlenker C,
    3. Lozano Romero R, et al.
    Advanced primary care in San Antonio: linking practice and community strategies to improve health. J Am Board Fam Med. 2013; 26(3): 288-298. doi:10.3122/jabfm.2013.03.120238
    OpenUrlAbstract/FREE Full Text
  23. 23.↵
    1. Rudowitz R,
    2. Artiga S,
    3. Musumeci M.
    The ACA and recent section 1115 Medicaid demonstration waivers. The Kaiser Commission on Medicaid and the Uninsured; 2014: 1–14. https://www.statecoverage.org/files/KFF_1115_Medicaid_Demonstration_Waivers.pdf
  24. 24.↵
    1. Revere L,
    2. Kavarthapu N,
    3. Hall J,
    4. Begley C.
    Achieving triple aim outcomes: an evaluation of the Texas Medicaid waiver. Inquiry. 2020; 57: 46958020923547. doi:10.1177/0046958020923547
    OpenUrlCrossRef
  25. 25.↵
    National Quality Forum (NQF0059): Diabetes: Hemoglobin A1c Poor Contorl (>9%). https://qpp.cms.gov/docs/QPP_quality_measure_specifications/CQM-Measures/2019_Measure_001_MIPSCQM.pdf
  26. 26.↵
    1. Antonovsky A.
    The salutogenic model as a theory to guide health promotion. Health Promot Int. 1996; 11(1): 1-8. doi:10.1093/heapro/11.1.11
    OpenUrlCrossRef
  27. 27.↵
    1. Sen A.
    Inequality Reexamined. Oxford University Press; 1992.
  28. 28.
    1. Ferrer RL,
    2. Carrasco AV.
    Capability and clinical success. Ann Fam Med. 2010; 8(5): 454-460. doi:10.1370/afm.1163
    OpenUrlAbstract/FREE Full Text
  29. 29.↵
    1. Ferrer RL,
    2. Cruz I,
    3. Burge S,
    4. Bayles B,
    5. Castilla MI.
    Measuring capability for healthy diet and physical activity. Ann Fam Med. 2014; 12(1): 46-56. doi:10.1370/afm.1580
    OpenUrlAbstract/FREE Full Text
  30. 30.↵
    1. Freire P.
    Pedagogy of the Oppressed. Continuum International Publishing Group; 2000.
  31. 31.↵
    1. San Antonio Metropolitan Health District
    . Diabetes in Bexar County - 2017/2018 Report. Accessed Aug 23, 2021. https://www.sanantonio.gov/Portals/0/Files/health/News/Reports/Diabetes/DiabetesReport2017-2018.pdf
  32. 32.↵
    1. Texas Department of State Health Services
    . Texas Community Health Worker Annual Report 2018. Accessed Aug 23, 2021. https://www.dshs.state.tx.us/legislative/2019-Reports/CommunityHealthWorkerAnnualReport2018.pdf
  33. 33.↵
    1. Cavero-Redondo I,
    2. Peleteiro B,
    3. Álvarez-Bueno C,
    4. Rodriguez-Artalejo F,
    5. Martínez-Vizcaíno V.
    Glycated haemoglobin A1c as a risk factor of cardiovascular outcomes and all-cause mortality in diabetic and non-diabetic populations: a systematic review and meta-analysis. BMJ Open. 2017; 7(7): e015949. doi:10.1136/bmjopen-2017-015949
    OpenUrlAbstract/FREE Full Text
  34. 34.↵
    1. Davis CE.
    The effect of regression to the mean in epidemiologic and clinical studies. Am J Epidemiol. 1976; 104(5): 493-498. doi:10.1093/oxfordjournals.aje.a112321
    OpenUrlCrossRefPubMed
  35. 35.↵
    A state policy framework for integrating health and social services. Commonwealth Fund. Accessed Jul 29, 2021. https://www.commonwealthfund.org/publications/issue-briefs/2014/jul/state-policy-framework-integrating-health-and-social-services
  36. 36.↵
    1. Mansournia MA,
    2. Higgins JPT,
    3. Sterne JAC,
    4. Hernán MA.
    Biases in randomized trials: a conversation between trialists and epidemiologists. Epidemiology. 2017; 28(1): 54-59. doi:10.1097/EDE.0000000000000564
    OpenUrlCrossRefPubMed
  37. 37.↵
    1. Lantz PM.
    “Super-Utilizer” interventions: what they reveal about evaluation research, wishful thinking, and health equity. Milbank Q. Published Feb 2020. Accessed Aug 23, 2021. https://www.milbank.org/quarterly/articles/super%E2%80%90utilizer-interventions-what-they-reveal-about-evaluation-research-wishful-thinking-and-health-equity/
  38. 38.↵
    1. Anglemyer A,
    2. Horvath HT,
    3. Bero L.
    Healthcare outcomes assessed with observational study designs compared with those assessed in randomized trials. Cochrane Database Syst Rev. 2014; 2014(4): MR000034. doi:10.1002/14651858.MR000034.pub2
    OpenUrlCrossRef
  39. 39.↵
    1. Vasan A,
    2. Morgan JW,
    3. Mitra N, et al.
    Effects of a standardized community health worker intervention on hospitalization among disadvantaged patients with multiple chronic conditions: a pooled analysis of three clinical trials. Health Serv Res. 2020; 55(Suppl 2): 894-901. doi:10.1111/1475-6773.13321
    OpenUrlCrossRef
  40. 40.↵
    1. Babamoto KS,
    2. Sey KA,
    3. Camilleri AJ,
    4. Karlan VJ,
    5. Catalasan J,
    6. Morisky DE.
    Improving diabetes care and health measures among hispanics using community health workers: results from a randomized controlled trial. Health Educ Behav. 2009; 36(1): 113-126. doi:10.1177/1090198108325911
    OpenUrlCrossRefPubMed
  41. 41.↵
    1. Pérez-Escamilla R,
    2. Damio G,
    3. Chhabra J, et al.
    Impact of a community health workers-led structured program on blood glucose control among latinos with type 2 diabetes: the DIALBEST trial. Diabetes Care. 2015; 38(2): 197-205. doi:10.2337/dc14-0327
    OpenUrlAbstract/FREE Full Text
  42. 42.↵
    1. Katigbak C,
    2. Van Devanter N,
    3. Islam N,
    4. Trinh-Shevrin C.
    Partners in health: a conceptual framework for the role of community health workers in facilitating patients’ adoption of healthy behaviors. Am J Public Health. 2015; 105(5): 872-880. doi:10.2105/AJPH.2014.302411
    OpenUrlCrossRefPubMed
  43. 43.↵
    1. Committee on Implementing High-Quality Primary Care, Board on Health Care Services, Health and Medicine Division, National Academies of Sciences, Engineering, and Medicine
    . Implementing High-Quality Primary Care: Rebuilding the Foundation of Health Care. National Academies Press; 2021.
  44. 44.↵
    1. Moffett ML,
    2. Kaufman A,
    3. Bazemore A.
    Community health workers bring cost savings to patient-centered medical homes. J Community Health. 2018; 43(1): 1-3. doi:10.1007/s10900-017-0403-y
    OpenUrlCrossRef
  45. 45.↵
    1. Albritton E.
    How states can fund community health workers through Medicaid to improve people’s health, decrease costs, and reduce disparities. Published 2019. Accessed Aug 23, 2021. https://familiesusa.org/resources/how-states-can-fund-community-health-workers-through-medicaid-to-improve-peoples-health-decrease-costs-and-reduce-disparities/
  46. 46.
    1. Mata-Cases M,
    2. Rodríguez-Sánchez B,
    3. Mauricio D, et al.
    The association between poor glycemic control and health care costs in people with diabetes: a population-based study. Diabetes Care. 2020; 43(4): 751-758. doi:10.2337/dc19-0573
    OpenUrlAbstract/FREE Full Text
  47. 47.↵
    1. Cottrell EK,
    2. Gold R,
    3. Likumahuwa S, et al.
    Using health information technology to bring social determinants of health into primary care: a conceptual framework to guide research. J Health Care Poor Underserved. 2018; 29(3): 949-963. doi:10.1353/hpu.2018.0071
    OpenUrlCrossRefPubMed
  48. 48.↵
    1. Rollow W,
    2. Cucchiara P.
    Achieving value in primary care: the primary care value model. Ann Fam Med. 2016; 14(2): 159-165. doi:10.1370/afm.1893
    OpenUrlAbstract/FREE Full Text
  49. 49.
    1. Nutting PA,
    2. Crabtree BF,
    3. Miller WL,
    4. Stange KC,
    5. Stewart E,
    6. Jaén C.
    Transforming physician practices to patient-centered medical homes: lessons from the national demonstration project. Health Aff (Millwood). 2011; 30(3): 439-445. doi:10.1377/hlthaff.2010.0159
    OpenUrlAbstract/FREE Full Text
  50. 50.↵
    1. Frank AW.
    From sick role to practices of health and illness. Med Educ. 2013; 47(1): 18-25. doi:10.1111/j.1365-2923.2012.04298.x
    OpenUrlCrossRefPubMed
  51. 51.↵
    1. Marmot M; Commission on Social Determinants of Health
    . Achieving health equity: from root causes to fair outcomes. Lancet. 2007; 370(9593): 1153-1163. doi:10.1016/S0140-6736(07)61385-3
    OpenUrlCrossRefPubMed
  52. 52.↵
    1. DeSalvo K,
    2. Harris A.
    Bending the trends. Ann Fam Med. 2017; 15(4): 304-306. doi:10.1370/afm.2101
    OpenUrlFREE Full Text
  53. 53.↵
    1. Miller PM,
    2. Brown T,
    3. Hopson R.
    Centering love, hope, and trust in the community. Urban Educ. 2011; 46(5): 1078-1099. doi:10.1177/0042085910395951
    OpenUrlCrossRef
Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools
Get Permissions

Jump to section

Keywords