Reaching Urban Poor Hypertensive Patients: A Novel Model of Chronic Disease Care Versus a Traditional Fee-for-Service Approach
Abstract
Background: There is a significant disparity in hypertensive treatment rates between those with and without health insurance. If left untreated, hypertension leads to significant morbidity and mortality. The uninsured face numerous barriers to access chronic disease care. We developed the Community-based Chronic Disease Management (CCDM) clinics specifically for the uninsured with hypertension utilizing nurse-led teams, community-based locations, and evidence-based clinical protocols. All services, including laboratory and medications, are provided on-site and free of charge. Methods: In order to ascertain if the CCDM model of care was as effective as traditional models of care in achieving blood pressure goals, we compared CCDM clinics’ hypertensive care outcomes with 2 traditional fee-for-service physician-led clinics. All the clinics are located near one another in poor urban neighborhoods of Milwaukee, Wisconsin. Results: Patients seen at the CCDM clinics and at 1 of the 2 traditional clinics showed a statistically significant improvement in reaching blood pressure goal at 6 months (P < .001 and P < .05, respectively). Logistic regression analysis found no difference in attaining blood pressure goal at 6 months for either of the 2 fee-for-service clinics when compared with the CCDM clinics. Conclusion: The CCDM model of care is at least as effective in controlling hypertension as more traditional fee-for-service models caring for the same population. The CCDM model of care to treat hypertension may offer another approach for engaging the urban poor in chronic disease care.
Introduction
Millions of Americans lack health insurance and struggle to find affordable access to our country’s health system. For those uninsured Americans with chronic disease(s), the recurring need for affordable care is particularly challenging.1
A number of barriers limit access to care for the uninsured, among them appointment scheduling, lack of health literacy, minority group status, lack of access to good nutritional counseling, and travel distance.2-6 Not surprisingly, multiple studies have shown that people without health insurance are less able to adhere to treatment plans than those with health insurance.7-9
Despite years of effort, traditional models of health care, such as those involving fee-for-service or physician-led decisions, have not shown sustainable adaptability to care for the uninsured chronically ill. If we are to narrow the disparity in health outcomes for the uninsured, we need new ways of delivering care to this population.
The Community-based Chronic Disease Management (CCDM) clinic began in 2007 in Milwaukee to address this disparity by removing some of the barriers to treatment adherence for the uninsured. The CCDM clinic was designed specifically to eliminate the barriers that patients without insurance commonly encounter when trying to access health care such as high-priced drugs, appointment systems, paper charts, etc. CCDM intentionally sought out community-based partners to help with leveraging patient-centered resources such as convenient locations, health care team leadership, culturally attuned health education, and more.
The CCDM clinic was not conceived to be for all patients and for all complaints. Instead, CCDM focused on only 3 chronic diseases: high blood pressure (essential type), uncomplicated diabetes mellitus type 2, and hypercholesterolemia. People who had other diseases or other health complaints were to be referred elsewhere. CCDM located its clinics in community settings in order for easy pedestrian access or with public transportation.
Additionally, nurses led the health team; which helped keep the human resource costs down. In particular, CCDM used parish nurses because of their familiarity with the patient population and the community’s familiarity with them. Advanced practice nurses and physicians served only as consultants. Clinical protocols were developed specifically for CCDM and these protocols were evidence-based and had been shown to assist with quickly achieving blood pressure target goals in other settings.
CCDM also advocated for its uninsured patients to sign up for state-funded health insurance. Once insurance was successfully garnered, the patients would be referred on to local federally qualified health centers where more broadly-based cares might be found.
CCDM built an electronic record which was housed on a secure web server. This allowed the team to access patient data at any time and from any location.
CCDM was supported by generous funding from 2 local foundations: Columbia St Mary’s Foundation and the Healthier Wisconsin Partnership Program. The secured funding assisted with on-site phlebotomy and the distribution of free high-quality pharmaceuticals. But much CCDM’s operating budget (rent, utilities, furnishings, etc) was supported by in-kind donations from its community partners.10 All of CCDM services were available at no cost to the patient.
CCDM served neighborhoods in Milwaukee’s core, one of the poorest urban areas in the United States.11 In these neighborhoods, the average life expectancy is 3 years less than people living only a few miles away in more affluent neighborhoods. Households have 50% less income, on average, than other Milwaukee households, and the infant mortality rate of babies born in CCDM neighborhoods is 3.5 times higher than those born to families just a mile or two away.12,13
Our team published a pilot study in 2013 about the effectiveness of CCDM’s model.14 At that time, CCDM compared favorably to national benchmarks for controlling hypertension. In recent years, however, we have questioned the comparison. Hypertension treatment and control appear to be more difficult in populations with demographic characteristics similar to CCDM’s. Perhaps there are other benchmarks to which CCDM should be compared. This idea is substantiated by other published outcomes from clinics caring for poor urban populations, which demonstrated hypertension outcomes far below HEDIS (Healthcare Effectiveness Data and Information Set) benchmarks.15,16
Accordingly, we compared CCDM’s outcomes with those of 2 clinics, both near CCDM’s sites, that used traditional fee-for-service and physician-led models of care. We wanted to see if CCDM’s model achieved HTN outcomes comparable to traditional fee-for-service models which were largely caring for the same population. CCDM’s novel model of care should have achieved outcomes equal to or better than those achieved by traditional fee-for-service models if CCDM was to be considered a valid model of care.
Study Population
The study sample consisted of hypertensive patients aged 18 to 65 years, who returned for a visit at 6 months plus or minus 10 days after their initial visit. Patients were not excluded if they had diabetes mellitus type 2 (DM2). Goal blood pressure was defined as systolic blood pressure measurement <140 mm Hg and diastolic blood pressure <90 mm Hg (systolic <130 mm Hg and diastolic < 80 mm Hg for hypertension with DM2).
Clinical Settings
CCDM provided care for only patients who had hypertension, non–insulin-dependent DM2, and hypercholesterolemia. Protocols were developed for the screening and treatment of hypertension, DM2, and hypercholesterolemia. Patients with advanced complications, such as organ failure, or other medical concerns were referred on to appropriate care sites. (Discussion about CCDM’s work with diabetes and hypercholesterolemia is outside the scope of this article.) CCDM eventually grew to include 5 clinical sites.10
We compared the demographic characteristics of CCDM patients with those of patients served at 2 other clinics in the same part of Milwaukee. Clinic A and clinic B, both fee-for-service, full-spectrum family medicine clinics, are sites for Family Medicine residency training programs, and National Committee for Quality Assurance Level 3 patient-centered medical homes. We compared the demographics served by each of the 3 clinics based on ZIP code of patient residence, age, gender, and insurance status. During the study period, separate, competitive health systems administered clinics A and B.
Microsoft Access software was used to build the custom electronic medical record at CCDM. Clinics A and B both used the same electronic medical record from which data was retrieved.
Methods
To look at blood pressure data from the 3 clinics, we modified the definition for hypertension control that is used by the Office of Human Resources and Services Administration for federally qualified health centers. Our definition differed mainly in selecting patients with ages up to 65 years instead of up to 85 years.
For CCDM, the query period was October 24, 2007, through August 27, 2014. Between February and August of 2014, CCDM actively enrolled its patients into the expanded Wisconsin Medicaid program and CCDM’s overall patient census dropped considerably. For the 2 comparison clinics, the query period was January 1, 2010, through December 31, 2012.
Similar to the method we employed for CCDM’s pilot study, we compared the average blood pressure taken at the first visit against the average blood pressure taken within 10 days of 6 months after the initial visit.
We limited our search criteria by excluding diagnostic codes having to do with complicated hypertension or DM2 (renal failure, heart failure, etc).
The Medical College of Wisconsin’s Institutional Review Board approved this research under protocol number 00006704. Informed consent was waived by the Medical College of Wisconsin’s Institutional Review Board.
Statistical Methods
We used a chi-square test to compare gender and the Kruskal-Wallis test to compare ages. We used McNemar’s test for paired data to assess changes in goal blood pressure attainment at 6 months compared with initial visit. Finally, we used logistic regression analyses to examine changes in goal blood pressure attainment from the initial visit to the 6-month visit, and to investigate the effects of gender, age, and the clinic.
Statistical software was Stata (2015 Stata Statistical Software Release 14; StataCorp LP, College Station, TX).
Results
Hypertension Control
Of those patients who met the criteria, 292 were seen at CCDM, 77 were seen at clinic A, and 59 were seen at clinic B during the study periods. Of these patients, 41%, 43%, and 37%, achieved their blood-pressure goals at 6 months at CCDM and clinics A and B, respectively. The change in goal blood pressure attainment reached statistical significance at clinic A and CCDM (Table 1).
| Location | Baseline, n (%) | 6-Month Follow-up, n (%) |
|---|---|---|
| CCDM*** | 44 (15) | 119 (41) |
| Clinic A* | 21 (27) | 33 (43) |
| Clinic B | 15 (25) | 22 (37) |
| Clinics A and B** | 36 (26) | 55 (40) |
Abbreviation: CCDM, Community-based Chronic Disease Management.
a
Difference at each location tested with McNemar’s test for paired data.
*
P < .05; **P < .01; ***P < .0001.
Logistic regression analysis found no difference in attaining blood pressure goal at 6 months for either of the 2 fee-for-service clinics (clinic A, clinic B or clinics A and B) when compared with CCDM (Table 2).
| Characteristic | Odds Ratio | 95% Confidence Interval |
|---|---|---|
| Not at goal initially | 1.0 | Referent |
| At goal blood pressure at initial visit | 2.5 | 1.5-4.2 |
| Clinic | ||
| CCDM | 1.0 | Referent |
| Clinic A | 1.0 | 0.6-1.6 |
| Clinic B | 0.8 | 0.4-1.4 |
a
Age and sex were entered as covariates; neither was found to be significant. We also examined an interaction between clinic and being at goal on initial visit. This term was also not significant.
Clinical Locations
Clinics A and B are in the same area of Milwaukee and serve many of the same neighborhoods (Figure 1). A ZIP code analysis of median household income revealed that Milwaukee’s center is largely defined by a severity of poverty not shared in more peripheral neighborhoods.12 Each of the 3 health clinics is in this poorer central section of Milwaukee.
Demographics
All 3 clinics cared predominantly for patients from the same 7 ZIP codes, all of them among Milwaukee’s lowest socioeconomic group. CCDM’s patients were 55% male, whereas those from clinics A and B were 39% and 36%, respectively. CCDM’s patients were slightly older, 50.2 years old, than clinic A’s, 45.4 years, or Clinic B’s, 46.8 years. As expected, insurance designation was very different among the patients: 99% of CCDM patients lacked health insurance, compared with 7% in both clinics A and B (Table 3).
| Characteristic | CCDM (N = 292) | Clinic A (N = 77) | Clinic B (N = 59) | Clinics A and B Combined (N = 136) |
|---|---|---|---|---|
| Gender n (%)** | ||||
| Male | 161 (55) | 30 (39) | 21 (36) | 51 (38)** |
| Female | 131 (45) | 47 (61) | 38 (64) | 85 (62) |
| Age, years, mean (SD)** | 50.2 (8.5) | 45.4 (11.0) | 46.8 (11.9) | 46.0 (11.3)*** |
| With health insurance, % | 1* | 93 | 93 | |
| ZIP Code’s Local Socioeconomic Rank | Frequency of ZIP Code of Residence for Patients Seen at Each Clinic | |||
| 5th | 1st | 6th | 1st | N/A |
| 2nd | 2nd | 4th | 3rd | N/A |
| 9th | 3rd | 3rd | 6th | N/A |
| 11th | 4th | 1st | 2nd | N/A |
| 7th | 5th | 2nd | 4th | N/A |
| 8th | 6th | 7th | 7th | N/A |
| 10th | 7th | 5th | 5th | N/A |
Abbreviation: CCDM, Community-based Chronic Disease Management; N/A, not applicable.
a
Gender compared with a chi-square test; age compared with the Kruskal-Wallis test; clinics A and B combined tested vversus CCDM; 30% of CCDM charts were missing insurance designation.
*
P < .05; **P ≤ .01; ***P < .0001.
Discussion
Outcomes show that CCDM’s model of care appears as effective as—and perhaps better than—traditional fee-for-service, physician-led models in controlling hypertension in a poor, inner-city population.
Results also showed congruence among the 3 clinics’ patients’ demographics with regard to ZIP Codes of origin: all 3 clinics shared in having most of their patients living within Milwaukee’s poorest neighborhoods.12 The social determinants of illness are frequent companions to people who live in these neighborhoods. Low levels of education, poverty, environmental pollution, overcrowding, poor-quality food stuffs, to name just a few, are all highly prevalent. These environmental and social challenges are not unique to Milwaukee. Many of these social determinants are present within the lives of poor patients elsewhere. These pernicious burdens can act like a bulwark against a person’s efforts to attain health, especially accessing the kind of recurring health care services necessary for the management of chronic diseases.
The 2 areas where there was noticeable demographic discrepancy were in health insurance and gender. This is not surprising given the fee-for-service business model of clinics A and B and the structure of CCDM. Whereas not having health insurance had been a selection criterion for enrollment into CCDM’s program, the patients at clinics A and B almost all carried health insurance (93%); the vast majority of which was either Medicaid or Medicare (>80%). Medicaid, in turn, is extended to impoverished families with children younger than 18 years. Typically, the primary caregivers for impoverished children in CCDM’s neighborhoods are women. Thus, by virtue of their gender and their poverty, many more women than men qualify for Medicaid insurance.
All 3 clinics described in this article have blood pressure goal outcome rates well below national norms. For example, the 2014 HEDIS benchmark for Medicaid patients achieving blood pressure goal is 57.1%, but the average hypertension goal achievement for this study’s 3 clinics is 40%.17 This average is not dissimilar from other hypertension outcome averages in cities with similar populations suggesting that more study is needed to understand the barriers that the urban poor face in trying to gain access to health care and achieve health goals—especially in the context of chronic disease.15,16
Table 3 suggests that CCDM’s patients skewed more toward the poorest of the ZIP codes than did patients of clinic A or B. Perhaps some aspects of CCDM’s care model make it particularly well-suited for the most challenging subpopulation groups. Replicating CCDM’s work in other cities or other uninsured urban populations would be a good next step.
Limitations
Both of the clinics with fee-for-service, physician-led decision-making models are sites for postgraduate residency education. The physician turnover associated with residency clinics might result in poorer patient outcomes.
Additionally, CCDM treated only hypertension cases considered “uncomplicated.” The 2 comparison clinics treat patients with more complicated health profiles. Despite screening criteria used in the data search, the fee-for-service clinics’ data sets might represent patients with more complicated comorbidities than CCDM’s. CCDM did not track data about its patients’ comorbidities. Consequently, comparing the study’s 3 clinics by a measurement of comorbidity is not possible.
Conclusion
We compared the effectiveness of a unique model for delivering chronic disease care to the medically uninsured with 2 traditional fee-for-service clinics that serve largely the same demographic. CCDM has demonstrated that its model works even where the pressures of poverty are significant and barriers to HTN control are greatest, such as areas where patients lack health insurance.
All 3 clinics in this study had outcomes below national standards, suggesting that more study is needed to understand the barriers that the urban poor face in trying to gain access to health care and achieve health goals—especially in the context of chronic disease.
Acknowledgments
CCDM clinical staff includes Brenda Buchanan, RN, Julia Means, RN, Nancy Leahy, APNP, Christy Tolbert, Bill Solberg, MSW, David Goines, Johnny Ayers, Robert Ramerez, and Carla Harris, RN. Data analyst: Haydee Zimmerman, BA; Graphic Map: Sara Kohlbeck, MPH; Manuscript review: Jeffrey Whittle, MD, MPH, Michael Gauger, and Dennis Butler, PhD.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was funded in part by the Advancing a Healthier Wisconsin endowment at the Medical College of Wisconsin (Grant Nos. 2007I-06, 2010I-07) and the Columbia St Mary’s Foundation.
References
1. Nardin R, Sayah A, Woolhandler S, McCormick D. Reasons why patients remain uninsured after Massachusetts’ health care reform: a survey of patients at a safety-net hospital. J Gen Intern Med. 2012;27:250-256.
2. Shelley D, Tseng TY, Andrews H, et al. Predictors of blood pressure control among hypertensives in community health centers. Am J Hypertens. 2011;24:1318-1323.
3. Blanchard J, Ogle K, Thomas O, Lung D, Asplin B, Lurie N. Access to appointments based on insurance status in Washington, D.C. J Health Care Poor Underserved. 2008;19:687-696.
4. Hadley J, Cunningham P. Availability of safety net providers and access to care of uninsured persons. Health Serv Res. 2004;39:1527-1546.
5. Persell SD, Osborn CY, Skripkauskas RR, Wolf MS. Limited health literacy is a barrier to medication reconciliation in ambulatory care. J Gen Intern Med. 2007;22:1523-1526.
6. Russell BE, Gurrola E, Ndumele CD, et al. Perspectives of non-Hispanic Black and Latino patients in Boston’s urban community health centers on their experiences with diabetes and hypertension. J Gen Intern Med. 2010;25:504-509.
7. Wilson J, Axelsen K, Tang S. Medicaid prescription drug access restrictions: exploring the effect on patient persistence with hypertension medications. Am J Manag Care. 2005;11:SP27-SP34.
8. Gai Y, Gu N. Association between insurance gaps and continued antihypertension medication usage in a US national representative population. Am J Hypertens. 2009;22:1276-1280.
9. Huttin C. The role of different types of health insurance on access and utilization of antihypertensive drugs. An empirical study in a U.S. hypertensive population. Dis Manage Outcomes. 2002;10:419-430.
10. Sanders J, Solberg B, Gauger M. Breaking barriers to care: a community of solution for chronic disease management. J Am Board Fam Med. 2013;26:311-315.
11. Tolan T, Herzog K. Poverty numbers spike in Milwaukee. Milwaukee Journal Sentinel, September 21, 2011. http://www.jsonline.com/news/milwaukee/Poverty-numbers-spike-in-Milwaukee.html. Accessed March 24, 2016.
12. Greer DM, Baumgardner DJ, Bridgewater FD, et al. Milwaukee Health Report 2013: Health Disparities in Milwaukee by Socioeconomic Status. Milwaukee, WI: Center for Urban Population Health. http://www.cuph.org/2013-milwaukee-health-report.html. Accessed February 27, 2016.
13. Herzog K. Milwaukee infant mortality rates headed in wrong direction. Milwaukee Journal Sentinel. June 3, 2014. http://www.jsonline.com/news/milwaukee/milwaukee-infant-mortality-rates-heading-in-wrong-direction-b99282687z1-261616401.html. Accessed March 25, 2016.
14. Sanders J, Guse C, Onuoha BC. Pilot study of a new model for managing hypertension in an uninsured population. J Prim Care Community Health. 2013;4:44-49.
15. Olomu AB, Gourineni V, Huang JL, et al. Rate and predictors of blood pressure control in a federal qualified health center in Michigan: a huge concern? J Clin Hypertens (Greenwich). 2013;15:254-263.
16. Baumann LC, Chang M, Hoebeke R. Clinical outcomes for low-income adults with hypertension and diabetes. Nurs Res. 2002;51:191-198.
17. NCQA. Controlling high blood pressure. http://www.ncqa.org/ReportCards/HealthPlans/StateofHealthCareQuality/2015TableofContents/ControllingHighBloodPressure.aspx. Accessed January 14, 2016.
Biographies
Jim Sanders, MD, MPH is an associate professor in the Department of Family and Community Medicine at the Medical College of Wisconsin. His expertise lies in health systems design for the poor and underserved.
Clare E. Guse, MS is senior Biostatistician in the Department of Family and Community Medicine at the Medical College of Wisconsin. She has over 25 years of experience in the design and analysis of medical research studies.
Cite article
Cite article
Cite article
OR
Download to reference manager
If you have citation software installed, you can download article citation data to the citation manager of your choice
Information, rights and permissions
Information
Published In
Article first published online: August 9, 2016
Issue published: January 2017
Keywords
Authors
Metrics and citations
Metrics
Journals metrics
This article was published in Journal of Primary Care & Community Health.
View All Journal MetricsArticle usage*
Total views and downloads: 1064
*Article usage tracking started in December 2016
Articles citing this one
Receive email alerts when this article is cited
Web of Science: 5 view articles Opens in new tab
Crossref: 4
- Improving hypertension control and cardiovascular health: An urgent call to action for nursing
- Health Care Access Among Essential Critical Infrastructure Workers, 31 States, 2017-2018
- A systematic review of community nursing interventions focusing on improving outcomes for individuals exhibiting risk factors of cardiovascular disease
- Investigating the Enduring Impact of a Community-Based Health Education Program to Promote African American Elders’ Use of Technology Designed to Support Chronic Disease Self-Management
Figures and tables
Figures & Media
Tables
View Options
View options
PDF/EPUB
View PDF/EPUBAccess options
If you have access to journal content via a personal subscription, university, library, employer or society, select from the options below:
I am signed in as:
View my profileSign out
I can access personal subscriptions, purchases, paired institutional access and free tools such as favourite journals, email alerts and saved searches.
loading institutional access options
Alternatively, view purchase options below:
Purchase 24 hour online access to view and download content.
Access journal content via a DeepDyve subscription or find out more about this option.
