Chapter 6. CHW Workforce Research and Evaluations

This study marks the first research effort that utilized a survey of verified employers i-n all 50 States to draw a profile of the community health workers (CHWs) workforce. Also, for this study, it was possible to use recent refinements in occupational and industry data,^[1] new reviews of the relevant literature, and collaborations with four concurrent, independently funded initiatives in CHW support,^[2] education,^[3] and research promotion.^[4]

Extent and Nature of Current Research

An indicator of the degree of involvement of the research community in any one topic is the number of published journal articles addressing that topic. Figure 6.1 compares journal publications since 1965 in five-year intervals.^[5] The increase in volume is significant: from 62 articles in the 1970s to nearly 400 in the 1990s.

[D]

Source: CHW National Workforce Study (CHW/NWS) (2006).

The quality and the scope of research within this pool of sources varied from few rigorous evaluations of specific medical interventions utilizing CHWs to many descriptive reports of CHW programs. Many studies suffered from small sample sizes, poor research designs, and lack of control groups. Rigorous longitudinal studies were needed to clearly isolate the CHW interventions and measure outcomes and cost effectiveness.

Findings From Literature Reviews

Nine literature reviews were published between 2002 and 2006 to evaluate the use of community health workers in specific primary care and medical specialty interventions. These reviews represent the best available assessments of findings from research on health interventions that included the use of CHWs. No peer-reviewed journal exists with a specific focus on CHW practice. All of the articles reviewed represent contributions to other fields such as pediatrics and health education. Most reported findings were statistically significant, but not all of them had clinical significance. Due to the variety of topics, methodologies, and results, the collective research did not provide a systematic evaluation of CHW effectiveness and best practices. It did present, however, valid-if fragmented-evidence of CHW contributions to the delivery of health care, prevention, and health education for underserved communities. Also, these literature reviews could provide a useful framework on which to base future research.

No well-documented differences were found between outcomes from programs involving paid CHWs and volunteers. And, there were no reports on the utilization of CHWs in the private sector, as competitive considerations kept the evaluation of proprietary projects from being made public.

Table 6.1 displays the number and dates of the studies examined, topics addressed, and populations served by the interventions reviewed.^[6] Then, each review is briefly described and followed by a summary of findings on cost effectiveness (Table 6.2).

Three of the nine reviews were limited to the involvement of CHWs in interventions addressing diabetes, heart disease/stroke, and pregnancy in minority women. They covered a total of 98 studies, of which 23 were included in more than one review. Two reviews included only randomized controlled trials (RCTs), and one excluded studies measuring only changes in knowledge or attitudes.

Table 6.1 Literature Reviews of CHW Research Studies, 2002-2006

Source: CHW/NWS (2006).

^† A study was not counted if the characteristic shown was not specifically mentioned in the review.
* HRSA (2002): 19 of 20 studies reviewed were in the U.S.; Lewin (2005): Of 24 U.S. studies, 21 were included and three were excluded because they primarily referred to the provision of paraprofessional clinical care; Norris (2006): 15 of 18 articles reviewed were in the U.S.; Persily (2003): 12 of 14 studies reviewed were in the U.S.
** National Fund for Medical Education.
*** Nine of the 18 studies included were program profiles in one report.^[7]

Andrews 2004^[8]

The evaluation, limited to research studies involving ethnic minority women, found that "[CHWs] are effective in increasing access to health services, increasing knowledge and promoting behavior change...."^[9] Only two of the cited studies described clinical outcomes (reduction in low birth weight (LBW) deliveries and weight loss). The remaining 12 lacked a clear reference to the theoretical framework supporting the methods employed; 10 were descriptive; 6 were quasi-experimental; seven were experimental; and one was a cross-sectional pre-post design.

Andrews found that most of the studies reported significant results for increasing access to services, but that the investigators differed in the definition of CHW roles and responsibilities and in the retention of participants, whose attrition ranged from 16 to 60 percent. Andrews concluded that CHW involvement in case management was more successful for retention than the more limited outreach role. Five of the seven studies on increasing knowledge on health behavior showed significant results; the validity of findings in the remaining two was limited by "high attrition rates, small sample size and lack of standardized instruments."

Two of the reviewed studies had positive results in breastfeeding behavior, and favorable reviews were given to single studies on weight loss, drug use, high-risk sexual behavior, and physical activity. A study on diabetes self-care did not show a measurable impact from the CHW intervention. Two studies showed both improved outcomes and reduced costs.

Brownstein 2005^[10]

Brownstein's review of six studies related to heart disease and stroke concluded that CHW interventions were associated with "significant improvements in participants' blood pressure care and control."^[11]

Home visits by outreach workers "to mobilize the patient's support system" were more effective in hypertension control than group education sessions.^[12]

CHWs providing blood pressure (BP) monitoring, education and follow-up (working with nurse practitioners) produced significant increases in appointment keeping and continuity of care.^[13]

CHWs teaming with a nurse and a physician increased entry to care and reduced blood pressure;^[14] a follow-up RCT combining hypertension (HTN) care and medications with CHW visits for education and for mobilizing family support led, over a 3-year period, to better care and better BP control than a "usual-care" control group.^[15]

A related Community-Based Participatory Research (CBPR) project provided further levels of training to CHWs, and compared more- and less-intensive CHW interventions. Both groups experienced significant increase in BP control with no significant differences in degree of improvement between the two intervention groups.^[16]

In a Medicaid population with diabetes and hypertension, CHW care management produced significant reductions in ER visits, hospital admissions, and total patient costs to the Medicaid program.^[17]

HRSA 2002^[18]

This review was developed for the Maternal and Child Health Bureau (MCHB) as an exploratory exercise in preparation for a national cost-effectiveness study on the use of CHWs in MCH programs. The evaluation studies reviewed were selected for their relevance to the design of the study, and the coverage was not meant to be comprehensive.^[19] The principal relevance of this review rests in identifying key considerations for research on CHWs.

Lewin 2005^[20]

This review of 43 RCTs excluded studies measuring only changes in knowledge, attitudes, or intentions, which "were not considered useful indicators of the effectiveness of [CHW] interventions."

The investigators concluded that CHWs "show promising benefits" in a limited range of health issues, including childhood immunizations.

National Fund for Medical Education 2006^[21]

This review was conducted to accompany a study on financing and sustainability of CHW services. It summarized findings of earlier literature reviews and examined seven RCTs published from 2002 to 2005. Of these, one^[22] reported no positive effect in measuring the role of CHWs in reduction of exposure of children to tobacco smoke. Two suggested positive effects but were included with reservations over "shortcomings in the design of the CHW role."^[23] The four remaining RCTs showed impact on blood glucose in African-American men with diabetes,^[24] on the participation by Hispanic women in an annual comprehensive clinical exam,^[25] on smoking cessation by adult Latinos,^[26] and on blood pressure control in urban African-Americans.^[27]

Nemcek 2003^[28]

Nemcek, writing from a nursing standpoint, concluded that "the rationale is strong for using CHWs to improve delivery of community-based preventive care" and that findings suggest roles for CHWs in three domains: (1) developing a "therapeutic alliance" between patient, provider, and family/community support systems; (2) risk reduction; and (3) improving patterns of health care utilization.

Of 18 programs reported in 10 articles, Nemcek found nine acceptable process and outcome evaluations, two with only outcome descriptions, and the remaining seven with process evaluations only. Improved utilization of services, including medical appointment-keeping and less frequent ER visits, were the most commonly reported types of outcomes. Clinical outcomes included reduction of low birth weight deliveries and changes in blood pressure and sugar levels. Changes in health-related knowledge, treatment compliance, and lifestyles were also included.

Nemcek found no useful information for evaluating the structure of CHW programs "because programs have lacked a standard structure" and noted there was "a dearth of CHW process and outcome evaluation evidence in the literature... most reports are not research studies and the use of rigorous controls was not documented."

Norris 2006^[29]

Norris et al. reviewed 18 articles evaluating CHW interventions focusing on adults with diabetes and showing client outcomes, including eight RCTs. Multiple CHW roles and activities were identified, and the investigators concluded that there were "some preliminary data demonstrating improvements in participant knowledge and behavior." Other research designs included six before/after designs, three non-randomized comparison studies, and one with post-intervention measures only.

Persily 2003^[30]

This review encompassed 14 studies, of which one was not from the United States and one was purely descriptive, limited to programs intended to improve pregnancy outcomes. Persily found that, although "home visiting by lay workers may be more accepted by pregnant women," published studies showed "mixed results." Among the 14 studies on "lay home visiting programs," eight showed positive impact on use of prenatal care; three of five, examining low birth weight delivery, showed impact; and one study showed impact on pre-term delivery. Three studies reported impact on "social support." Only one study (on child abuse) showed no significant impact. However, the review described weaknesses in the studies such as the use of descriptive or quasi-experimental designs, poorly specified interventions, and lack of cost analyses.

Swider 2002^[31]

This review covered 19 CHW effectiveness studies of various design from 1981 through 1999. Swider concluded that there was some evidence for supporting CHWs in increasing access to care, particularly for underserved populations, but "inconclusive results" regarding knowledge acquisition, clinical outcomes, and behavioral changes. In most of the studies reviewed, the CHWs' "primary role expectations were not reported, nor were details of the intervention they provided." Therefore, only one of four studies with a primary CHW role of "outreach and case finding" had positive outcomes.

Cost Effectiveness

Ten published studies^[32] were found that dealt with cost effectiveness of, or return on investment (ROI) from, CHW activities. In only two of these studies did cost considerations constitute the main topic of the published article.^[33] The limited number of studies and the variety of measures used did not allow meaningful conclusions overall.

In Table 6.2, the articles' health-related objectives, outcome measures, and cost-effectiveness results are displayed by author in alphabetical order.

Table 6.2 Studies

Source: CHW/NWS (2006).

[1] Appendix B.

[2] The CHW Programs Inventory initiated by the Center for Sustainable Health Outreach (CSHO) of The University of Southern Mississippi under a grant from the W. K. Kellogg Foundation (WKKF) became the starting point for the CHW National Employer Inventory (CHW/NEI) through a partnership agreement with CSHO. Also, the Albuquerque, Miami, Northern Manhattan, Oakland, and FirstHealth (North Carolina) Community Voices sites provided feedback to develop contacts for the CHW/NEI in their respective States.

[3] The Community Health Worker National Education Collaborative (CHW-NEC) initiative explored, under a grant from the U.S. Department of Education's Fund for the Improvement of Postsecondary Education, best practices for CHW education and training and provided a taxonomy of key areas for developing employable CHWs (discussed in Chapter 4).

[4] The preparatory work for a forthcoming invitational conference to set a National research agenda on CHWs, supported by the California Endowment, The Northwest Area Foundation, The California Health Care Foundation, The Health Care Education-Industry Partnership of Minnesota, and The California Wellness Foundation, enhanced the material used in this chapter.

[5] The list of journal articles was obtained from the bibliographic database of 1,068 entries compiled for this study. The 2005+ year group in Figure 6.1 includes nine articles from 2006.

[6] Appendix F contains a table that shows selected articles by author, date of publication, and health issue addressed, which were included in the nine reviews.

[7] Health Resources and Services Administration. Impact of community health workers on access, use of services, and patient knowledge and behavior. Bureau of Primary Health Care, Health Resources and Services Administration, U.S. Department of Health and Human Services 1998.

[8] Andrews JO, Felton G, Wewers ME et al. Use of community health workers in research with ethnic minority women. J Nurs Scholarsh 2004; 36 (4):358-65.

[9] Ibid. (p.358)

[10] Brownstein JN, Bone LR, Dennison CR et al. Community health workers as interventionists in the prevention and control of heart disease and stroke. Am J of Prev Med 2005; 29 (5S1):128-33.

[11] Ibid. (p.132).

[12] Morisky DE, Levine DM, Green LW et al. Five-year blood pressure control and mortality following health education for hypertensive patients. Am J Public Health 1983; 73 (2):153-62.

[13] Bone LR, Mamon J, Levine DM et al. Emergency department detection and follow-up of high blood pressure: use and effectiveness of community health workers. Am J Emerg Med 1989; 7 (1):16-20.

[14] Hill MN, Bone LR, Kim MT et al. A clinical trial to improve high blood pressure care in young urban black men: recruitment, follow-up, and outcomes. Am J Hypertens 1999; 12:548-54.

[15] Dennison CR, Hill MN, Bone LR et al. Comprehensive hypertension care in underserved urban black men: high follow-up rates and blood pressure improvement over 60 months. Circulation 2003; 108:381.

[16] Levine DM, Bone LR, Hill MN et al. The effectiveness of a community/academic health center partnership in decreasing the level of blood pressure in an urban African-American population. Ethn Dis 2003; 13 (3):354-61.

[17] Fedder DO, Chang RJ, Curry S et al. The effectiveness of a community health worker outreach program on healthcare utilization of West Baltimore City Medicaid patients with diabetes, with or without hypertension. Ethn Dis 2003; 13 (1):22-7.

[18] Health Resources and Services Administration. A literature review and discussion of research studies and evaluations of the roles and responsibilities of community health workers (CHWs). Maternal and Child Health Bureau, Health Resources and Services Administration, U.S. Department of Health and Human Services July 5, 2002.

[19] Ibid. (p.19).

[20] Lewin SA, Dick J, Pond P et al. Lay health workers in primary and community health care. Cochrane Database of Systematic Reviews, 2005.

[21] National Fund for Medical Education. Advancing Community Health Worker Practice and Utilization: The Focus on Financing. San Francisco (CA): Center for the Health Professions, University of California at San Francisco, 2006.

[22] Conway TL, Woodruff SI, Edwards CC et al. Intervention to reduce environmental tobacco smoke exposure in Latino children: null effects on hair biomarkers and parent reports. Tob Control 2004; 13 (1):90-2.

[23] Krieger JW, Takaro TK, Song L et al. The Seattle-King County Healthy Homes Project: a randomized, controlled trial of a community health worker intervention to decrease exposure to indoor asthma triggers. Am J Public Health 2005; 95 (4):652-9; Hill MN, Han H-R, Dennison CR et al. Hypertension care and control in underserved urban African American men: behavioral and physiologic outcomes at 36 months. Am J Hypertens 2003; 16 (11):906-13.

[24] Gary TL, Bone LR, Hill MN et al. Randomized controlled trial of the effects of nurse case manager and community health worker interventions on risk factors for diabetes-related complications in urban African Americans. Prev Med 2003; 37 (1):23-32.

[25] Hunter JB, de Zapien JG, Papenfuss M et al. The impact of a promotora on increasing routine chronic disease prevention among women aged 40 and older at the U.S.-Mexico border. Health Educ Behav 2004; 31 (4 Suppl):18S-28S.

[26] Woodruff SI, Talavera GA, Elder JP. Evaluation of a culturally appropriate smoking cessation intervention for Latinos. Tob Control 2002; 11 (4):361-7.

[27] Levine DM et al. (2003).

[28] Nemcek MA, Sabatier R. State of evaluation: community health workers. Public Health Nurs 2003; 20 (4):260-70.

[29] Norris SL, Chowdhury FM, Van Le K et al. Effectiveness of community health workers in the care of persons with diabetes. Diabet Med 2006; 23 (5):544-56.

[30] Persily CA. Lay home visiting may improve pregnancy outcomes. Holist Nurs Pract 2003; 17 (5):231-8.

[31] Swider S. Outcome effectiveness of community health workers: an integrative literature review. Public Health Nurs 2002; 19 (1):11-20.

[32] Published studies in Table 6.2 are referenced in Appendix I.

[33] Whitley EM, Everhart RM, Wright RA. Measuring return on investment of outreach by community health workers. J Health Care Poor Underserved 2006; 17 (1 Suppl):6-15; Wolff N, Helminiak TW, Morse GA et al. Cost-effectiveness evaluation of three approaches to case management for homeless mentally ill clients. Am J Psychiatry 1997; 154 (3):341-8.