Community Health Workers to Improve Asthma Outcomes


Written By

Kathleen J. Pincus, PharmD, BCPS

Reviewed By

Jessica Binz, Pharm.D.
Erica Crannage, Pharm.D., BCPS, BCACP
Peter Koval, Pharm.D., BCPS


Krieger J, Song L, Philby M. Community health worker home visits for adults with uncontrolled asthma: the HomeBASE trial randomized clinical trial. JAMA Intern Med 2015; 175:109-117.

Can deploying trained laypeople working directly with low income adults with asthma in the community improve outcomes? This is what the Home-Based Asthma Support and Education (HomeBASE) trial set out to answer.1 While comprehensive evidence-based guidelines and effective pharmacotherapies exist, there are still 3,345 asthma-associated deaths and 1.8 million emergency department (ED) visits each year.2,3 There is clearly a disconnect between optimal asthma therapy and actual medication use behaviors.

Community health workers (CHWs) are “trained laypeople who share life circumstances with the patients they serve," according to the IMPACT Toolkit developed by the Penn Center for CHWs.4 Annual wages for CHWs are significantly lower than health professionals ($16.64 median hourly wage).Specific duties of CHWs typically include conducting home visits, providing self-management education, and assisting with healthcare system navigation. But intangible effects on patient activation and community engagement are often thought to provide the greatest benefits from interactions with CHWs. This is particularly apparent in minority or low-income communities where patients can feel disconnected or distrustful of the medical system and healthcare providers.4 Since CHWs are recruited from target communities, barriers such as language and misconceptions about disease are reduced. When developing programs involving CHWs, it is important to remember that CHWs are not credentialed health care providers (HCP). Every program must include intensive and ongoing training to prepare and support CHWs.

The HomeBASE Trial randomized participants (Table 1) to an intervention group (n=177) who received five home visits from a CHW at enrollment and 0.5, 1.5, 3.5 and 7+ months or a control group (n=188) who received usual care. Symptom-free days, asthma-related quality of life scored on the Mini Asthma Quality of Life Questionnaire (MiniAQLQ),6 and asthma-related unscheduled health care use were assessed at baseline and after 12 months. Participants were recruited from area clinics, community agencies, a Medicaid health plan, and through self-referrals.  All participants received modest monetary compensation ($35-50). Baseline clinical evaluation performed at a research center prior to randomization included a questionnaire, spirometry, and skin prick test.

CHWs received 80 hours of classroom training prior to participation in the study, and then additional training every two weeks. Each CHW had a caseload of 40-45 participants. During each initial visit the CHW would use standardized protocols to assess the participant’s knowledge about asthma, current asthma control, self-management practices, exposure to triggers and challenges with asthma management. 2 The CHW developed an individualized asthma action plan, provided social support, and assisted the patient in obtaining community-based resources. When appropriate, they also provided the participant with bedding encasements, low-emission vacuum cleaners, medication boxes and spacers, food storage containers to limit the effect of roaches in the home, and air filters. A HomeBASE study nurse performed audits and observed CHW home visits to provide feedback and ensure consistency. Patients in the usual care group received information about community resources for self-management (e.g. pamphlets, classes and support groups), but additional interventions were not described by the study authors.

Table 1: Patient and Community Health Worker Requirements

Patient Enrollment

Community Health Worker Requirements

Inclusion Criteria

Exclusion Criteria

  • English or Spanish speaking
  • Age 18-65 years
  • Residents of one county in Washington State
  • “Not well controlled” or “very poorly controlled” asthma
  • Household income < 250% federal poverty line
  • Plans to move
  • Housing unsafe for CHWs
  • Physical disability restricting participation
  • Low functional status due to other medical conditions
  • Anti-inflammatory medication use other than NSAIDs
  • COPD or >15 pack-year smoking history
  • Completion of other asthma education program within 3 years
  • High-school education or GED
  • Native Spanish speaker
  • Strong ties to the community
  • Personal experience with asthma


At baseline, study participants reported an average of 3.9 symptom free days during the previous 2 weeks (4.2 intervention, 3.7 control group, p=0.36), a MiniAQLQ score of 3.7 (3.7 intervention, 3.8 control, p=0.80), and 4.4 annual episodes requiring urgent health care visits (4.4 intervention, 4.3 control, p=0.87). Participants in the intervention group had greater increases in symptom-free days (2.02/2 weeks, 95% CI 0.94-3.09, p<0.001) and MiniAQLQ score (0.5, 95% CI 0.28-0.71, p<0.001). Unplanned healthcare utilization decreased similarly between groups (-1.47 intervention group vs. -1.34 control group; p=0.83).

Strengths of this study included rigorous auditing to ensure consistency of the intervention and the relatively high enrollment and retention of participants. The study was powered to demonstrate an improvement in the primary endpoints, but was slightly underpowered when participants that completed the entire protocol were included. The study authors ran the analysis with both the total enrolled population using a multiple imputation method for unavailable exit data and with the per-protocol population and found negligible differences in results which strengthens the inferences that can be drawn. The measures included traditional measures of asthma-control recommended in treatment guidelines and a validated measure of asthma related quality of life.  However, some measures of disease severity, including nighttime awakenings and rescue medication use, were not reported.2 On the MiniAQLQ a 0.5 change is considered clinically significant – albeit minimally so.6 As only the patient-reported outcomes were found to be different between treatment groups and patients were not blinded to their group assignment, there is the potential for bias. Use of symptom diaries throughout the study may have reduced the subjectivity of the reports.

The results indicated a small number needed to treat to gain benefits from the intervention: 7.4 patients treated to increase symptom-free days by 2 days/2 weeks and 2.6 patients treated to improve MiniAQLQ score by 0.5 points. This translates to only 1 additional symptom-free day per week. It is important to note that neither group achieved “well-controlled” asthma.2

The cost of the intervention was not discussed in sufficient detail in this manuscript. The authors simply state that the cost per participating was $1300. A more robust cost-benefit analysis would be needed in order for a health system or plan to contemplate such an intervention.  The results indicate there were no changes in health care service utilization measures including ED visits or hospitalizations. The authors mention that the cost of this intervention is less than a year’s supply of inhaled corticosteroids, though one could argue that the improvement seen with appropriate use of inhaled corticosteroids would likely be greater.

Given the bundled nature of the intervention, we are unable to analyze which aspects most contributed to the improvements in quality of life and number of symptom-free days. Analysis of the effect and costs of the environmental measures and supplies separated from the educational and patient activation components of the intervention would be of interest. The authors did not provide any details regarding the care received by patients in the control arm.  Readers are left wondering what factors led to reduced health-care utilization from baseline in the control group. It is possible there were other initiatives generally available to all participants that led to reduced health-care utilization – but these were not identified by the study authors.

A large portion of the CHW training manual focused on the appropriate use of asthma medications, specifically: emphasizing the difference between controller and reliever medications, adherence coaching, and how to manage exacerbations. While CHW was responsible for assessing inhaler technique, no step-by-step instructions for proper technique were included in the training materials. While medication use behaviors are a cornerstone of appropriate asthma care, there was no mention of involving pharmacists in either the intervention or the CHW training in this program. Pharmacists are certainly qualified to provide patient education regarding the disease, self-management, and medication use. But employing pharmacists to perform home visits to provide this education would be far more costly. Would home visits from a health professional lead to better outcomes? Where do CHWs fit within the current healthcare model?

Community and home-based interventions by CHWs showed measurable improvement in symptom control for low income adults with asthma. However, reductions in ED visits, hospital admissions, and costs were not demonstrated. While symptom control is important, demonstration of measurable cost savings is a necessary driving factor.  Given the increasing pressure put on health systems to reduce utilization and costs, relatively low-cost interventions involving CHWs are attractive options.  But without a demonstrative positive impact on cost, implementation of such programs seems unlikely.  What do you think?  Are CHW programs worth it?  And what role should community pharmacists play?