Abstract
For millions of uninsured Americans who have hypertension, quality medical care is too expensive to access with any regularity. The Community-based Chronic Disease Management (CCDM) Clinic was created to deliver clinical care for medically uninsured patients in a setting of low resources and high need. CCDM’s model melds nurse-led teams with the chronic disease model and uses evidence-based clinical decision protocols. This new model of care differs from traditional models. CCDM conducted a nonrandomized prospective trial of the effectiveness of this new model of care. The intervention included free education, medications, and laboratory investigations. For hypertensives treated for 6 months and 1 year, national benchmark goals were reached for 45% (50/110, P < .00005) and 56% (43/77, P < .00005) of patients, respectively, compared with 18% and 22% being at goal at initial presentation. The CCDM model may have implications for health service delivery in insured populations as well. Further study is warranted.
The health consequences of essential hypertension are well established: It is costly, totaling more than $76 billion per year in the United States; it is progressive, contributing to the majority of strokes, heart attacks, and congestive heart failure cases; and, it is deadly, taking an average of 5 years off the life expectancy of those older than 50 years.1,2 Despite the evidence of hypertension’s morbidities and the relative ease by which it can be treated, current systems of medical care fall short of consensus treatment goals, getting only about half of all US hypertensives to their blood pressure goal.2-5
There is good evidence that best practice approaches yield improved treatment adherence and outcomes. 6 For millions of hypertensive Americans who lack health insurance, research has clearly shown that episodic, disjointed care is costly, inconvenient, and not particularly successful in reaching health goals.7-10
To address this gap between the best-practice approach and barriers to care faced by the uninsured, we designed the Community-based Chronic Disease Management (CCDM) Clinic. We borrowed from models that were well established and well evidenced, including the Chronic Disease Model and the Community Nursing Center.6,11-15 By so doing, we broadened the human resource potential of competent, basic primary health care (eg, professional nursing) from conventional practice models and wed it with specific interventions focused on a distinct disease. We hypothesized that uncomplicated high blood pressure could be treated to goal at levels consistent with national benchmarks using an integrated nursing model of care, clinical decision protocols, information technology, and low-cost medications.
Methods
Theoretical Framework
We approached the development of CCDM to match limited resources with patients’ needs. CCDM only provided care for patients who had hypertension, non-insulin-dependent type 2 diabetes mellitus, and hypercholesterolemia. Those patients with advanced complications, such as organ failure, were referred to specialized care. Discussion about CCDM’s work with diabetes and hypercholesterolemia is outside the scope of this article. Patients who had other medical concerns were referred to appropriate care sites.
The CCDM hypertension protocols were developed by reviewing a number of sources: recommendations from respected authorities,16-18 protocols used elsewhere with success,19,20 and the randomized controlled trials literature. 21 Treatment protocols were developed for common comorbidities such as diabetes mellitus type 2 and renal insufficiency. An example can be seen in Figure 1.
Community-based Chronic Disease Management (CCDM)—Protocol for the management of hypertension
CCDM’s first site opened in a local food pantry for 4 hours once a week. After about 1 year, a second site opened at a second food pantry for another 4 hours once a week. Budgetary constraints forced the closing of the second location after about 1 year. Consequently, CCDM retreated to one location but expanded its hours to a full 8-hour day once per week.
Patient Experience
First-time presenters were screened for hypertension by CCDM nurses. The definition used was either (a) a prior diagnosis of hypertension or (b) a systolic blood pressure >140 mm Hg or a diastolic blood pressure >90 mm Hg for patients without diabetes mellitus, or a systolic blood pressure >130 mm Hg or a diastolic blood pressure >80 mm Hg for patients with diabetes mellitus, taken on 2 separate occasions. CCDM nurses were trained in methods of blood pressure measurement by a national consultant. Final blood pressure was the average of the 2 blood pressures taken from whichever of the patient’s arms had the higher initial blood pressure measured. Each patient has a minimum of 3 blood pressures taken during each visit.
Those screening positive were offered enrollment into the CCDM program. Enrolled patients had their diagnosis confirmed by a nurse practitioner or doctor (either on-site or via real-time telephonic communication) and started on the appropriate treatment protocol for their particular clinical situation.
Labs were drawn on-site and sent on to a local laboratory. A Web-based database was used to track patient care, create disease registries, and allow for multicenter point-of-care data retrieval. Generic medications were dispensed at no cost on-site.
Statistical Methods
Comparisons were made between average blood pressure at the first visit and average blood pressure within 10 days of 6 months and within 10 days of 1 year after the initial visit. The 6-month and 1-year cohorts had 40 patients in common but were otherwise distinct patients. Goal blood pressure was defined as less than 140/90 mm Hg or less than 130/80 mm Hg if the patient had diabetes. McNemar’s test for paired data assessed changes in attaining goal blood pressure. Body mass index (BMI, kg/m2) was computed from height and weight measurements and categorized into underweight (<18.5), normal weight (18.5-24.9), overweight (25-29.9), obese (≥30), and unknown. Random-effects linear regression analyses were used to examine changes in systolic and diastolic blood pressures from initial visit to 6-month or 1-year visits while allowing testing for significant effects of covariates, gender, age, and BMI category, and to account for repeated measures on patients. None of the covariates were found to be significant, and they were removed from the regression models.
Microsoft Access software was used to build the custom electronic medical record. Statistical software was Stata (Stata: Release 12, Statistical Software, College Station, TX: StataCorp LP).
Results
A total of 707 patients with hypertension had 5033 visits from October 24, 2007 through October 31, 2011; half (50.6%) had 3 visits or less. At initial presentation, the average blood pressure was 155 mm Hg systolic and 95 mm Hg diastolic (standard deviations [SD] 23.7 and 14.9 mm Hg, respectively), and 16% were at goal blood pressure. Seventy-three percent were classified as being overweight or obese based on BMI stratification, 56% were male, and the average age was 48 years (SD = 8.8 years).
Essential Hypertension: The 6-Month Experience
In all, 110 patients met the 6-month visit criteria (±10 days). Eighteen percent of patients in this group were at goal on initial measurement and 45% were at goal at the 6-month interval. Compared with blood pressure at presentation, McNemar’s test showed a significant increased proportion of patients at goal (McNemar’s χ2 = 19.57; P < .00005).
Random-effects linear regression analysis of blood pressure in these 110 patients showed an average drop of 18.4 points (95% confidence interval [CI] = 13.2-23.6 mm Hg) in average systolic blood pressure and an average drop of 10.5 points in average diastolic blood pressure (95% CI = 7.7-13.2 mm Hg) from the first to the 6-month visit.
Essential Hypertension: The 12-Month Experience
Seventy-seven patients had a follow-up visit at the 12-month interval. Twenty-two percent patients of this group were at goal on initial measurement and 56% were at goal at the 12-month interval. Compared with initial blood pressure, McNemar’s test showed a significant increased proportion of patients at goal (McNemar’s χ2 = 18.78; P < .00005).
Random-effects linear regression analysis in these 77 patients showed an average drop of 22.8 points in average systolic blood pressure (95% CI = 17.3-28.3 mmHg) and an average drop of 13.4 points in average diastolic blood pressure (95% CI = 8.9-15.0 mm Hg) from the first to the 1-year visit.
Comparison With Those Not in Control Within the 10-Day Window
For both the 6-month follow-up mark and the 1-year follow-up mark we compared those at blood pressure goal with patients who were not at goal. Table 1 shows that those who met goal at the 6-month mark were significantly less likely to have diabetes, whereas those who met goal at the 1-year mark had a similar trend toward less diabetes, which did not reach significance. Those reaching goal at 6 months were also significantly younger.
Comparison of Hypertensive Patients at Goal With Those Not at Goal at 6-Month and 1-Year Visits a
Categorical variables were compared using the χ2 test; age was compared using the t test; total number of visits was compared using the Wilcoxon rank-sum test; months between visits was compared with cluster-adjusted linear regression of the transformed time variable.
Discussion
For patients following up at the 6-month and 12-month intervals, the data show that the CCDM model succeeds at bringing patients to American Heart Association goals for blood pressure. And it does so at levels on par with national rates associated with more resourced clinical settings.2,4
The CCDM model is an example of a creative enterprise that addresses the unmet needs of a vulnerable population. The 3 main components of CCDM, the Chronic Disease Model, treatment by protocol, and Community Nursing Model all work together to assure competent care that matches resources with needs. Although these models of care have been shown to be effective on their own individually, to our knowledge this is the first example of all 3 being used in a setting of uninsured patients.
Most studies that have looked at improving rates of control in hypertensives did so in more resourced settings. Using the OVID database and key words “Hypertension” and “Medically Uninsured” we found only one article that studied the effects of using a specific model of care (the Chronic Disease Model) in an uninsured population. 7 Our study expands on previous research by formally incorporating the Community Nursing Center model and relying heavily on clinical protocols.14,23,24
Part of CCDM’s success was our choice to treat only uncomplicated hypertension. Of course, that was exactly the point of CCDM’s model; by leveraging information technology, evidence-based models of care, and expert opinion, CCDM helps free up more intensive and expensive human and capital resources for those patients who truly need them.
Our data collection focused on the clinical and not the social and economic aspects of the patients’ disease. Milwaukee is one of the most impoverished cities within the United States and more patients came from the city’s 2 most impoverished ZIP codes than any other.25,26 Fewer personal resources also means that child care, work schedules, transportation, and so on, all compete for priority in a day’s logistical planning and could interfere with a patient’s ability to follow-up with the CCDM clinic regularly.
Regression to the mean did not appear to have much impact on the reduction in blood pressure seen at the 6-month and 1-year time points. The method used by CCDM to identify patients as hypertensive would be expected to mitigate regression to the mean: including patients with a history of hypertension rather than basing inclusion only on a cutoff value and requiring an elevated average blood pressure on 2 separate occasions.
One of the limitations of this study was that the presence or absence of a face-to-face visit with a physician or midlevel clinician was not tracked. Thus, it is impossible to say how much this private consultation might have affected the patients’ adherence to treatment recommendations. Another limitation was an inability to determine which of the original 707 hypertensive patients had complicated conditions necessitating referral on to specialized care. Additionally, the attrition rate from the CCDM program was high with half of the patients having only 3 visits or less. It was because of this high attrition rate that CCDM’s data were analyzed using the 6-month and 1-year cohort populations as the denominators. This is different from the national data, which is derived from clinical practice data using denominators composed of all hypertensive patients who are active with a practice at the time of a cross-sectional enquiry.
This pilot study is encouraging and shows that the CCDM model aids in bringing uninsured hypertensive patients blood pressures to goal. However, the sample numbers in this study are small and more study will be needed to further elucidate cause–effect relationships and implications that this model of care has for other populations.
Footnotes
Acknowledgements
CCDM clinical staff includes Brenda Buchanan, RN, Julia Means, RN, Heather Puente, MPH, Ebonie Gray, APNP, Christy Tolbert, and Bill Solberg, MSW, as project coordinator. The Family Medicine Residents from the Columbia St. Mary’s Family Medicine Residency Program are thanked for consultative services and Dr Jeff Whittle for his skillful presubmission editing.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by grants from the Healthiest Wisconsin Partnership Program (Grant Nos. 2007I-06, 2010I-07) and the Columbia St. Mary’s Foundation.