Use of the Chronic Care Model, Including Self‐Reported Home Blood Pressure Measurement and Pharmacist Management, Improves Blood Pressure Control Rates
Although treating hypertension significantly reduces cardiovascular morbidity and mortality, hypertension control rates in the United States still remain less than ideal. The Effectiveness of Home Blood Pressure Monitoring, Web Communication, and Pharmacist Care on Hypertension Control (e‐BP) study was a randomized controlled trial set in an integrated group practice designed to determine whether an integrated model of care would improve blood pressure (BP) control rates compared with the usual approach.
In the study, patients with uncontrolled hypertension who were already taking antihypertensive medication were randomly assigned to 1 of 3 groups: usual care, home BP monitoring with secure Web‐based services training, or home BP monitoring with Web‐based training plus pharmacist care management delivered through Web‐based communications. The primary study outcomes were change in systolic and diastolic BP and the percentage of patients with controlled BP (office BP <140/90 mm Hg). Differences in outcome were measured between each of the 2 intervention groups as compared with the usual care group (control). This was a single‐site study conduced at 10 medical centers within Group Health, a large nonprofit integrated health plan and group practice that provides clinical care to more than 540,000 individuals in Washington State and Idaho. All Group Heath centers use a single commercially available electronic medical record (EMR) that is integrated with Web‐based services and available to all Group Health patients.
Clinical and administrative data were used to identify Group Health patients aged 25 to 75 years with a diagnosis of hypertension who were taking antihypertensive medications. They could not have a history of known diabetes mellitus, established cardiovascular disease, renal disease, or other serious medical conditions. Research assistants subsequently telephoned these patients to confirm the above eligibility requirements and to ensure that they were able to use a computer, had regular access to the Internet, had an e‐mail address, and were willing to attend screening visits and obtain their antihypertensive medications at Group Health–owned pharmacies. Patients who met all inclusion criteria were invited to attend screening visits at which office BP was measured according to a strict protocol. Patients whose mean BP level was 140–199/90–109 mm Hg at both screening visits were eligible to participate.
All enrolled patients were registered to use Group Health’s secure patient Web services and receive educational pamphlets describing the Web services as well as general information about hypertension. In addition, those in the “home BP and patient Web” group received a home BP monitor (Omron HEM‐705CP, Omron Healthcare, Bannockburn, IL) and training on its use. They were instructed to check BP at home at least 2 days a week with 2 measurements at each sitting. The goal BP for home measurement was <135/85 mm Hg. They also received more comprehensive training on how to use the Web‐based services and were encouraged to contact their provider through the Web if their BP was not controlled. Those in the “home BP and patient Web + pharmacist care” group received the home BP monitor and Web‐based training as described above plus the assistance of a clinical pharmacist. The clinical pharmacist initially contacted the patient by phone and introduced a care plan that included patient‐specific lifestyle recommendations, stepped‐care changes in antihypertensive medications, and follow‐up instructions. Every 2 weeks, the clinical pharmacist contacted patients via e‐mail until their home BP was controlled. Once controlled, calls occurred periodically. Pharmacists responded to patient information and queries with specific recommendations, when necessary, including medication additions and changes. The pharmacists’ initial action plan and subsequent recommendations were made available to the patients’ physician through the EMR. Clinical concerns or deviations from the stepped‐care medication protocol were referred back to the usual primary care physician. Regardless of study group allocation, all patients were invited to return to the clinic for a final study visit at 12 months, at which point office BP was measured using the same technique performed at baseline. The research assistants performing both the baseline and the final office BP measurements were blinded to study group allocation. The primary analyses included only patients who attended the final study visit; only 6% of patients were lost to follow‐up. Additional analyses performed using last observation carried forward did not lead to a substantial difference in the results.
Of the 7279 potential participants who initially answered the telephone survey questions, 20.7% were excluded because they did not have access to a computer or the Web or have an e‐mail address. Of the remaining potential patients, 2573 attended a screening visit and had their BP measured. Of these, 1537 were excluded because their BP was <140/90 at 1 of the 2 screening visits. After all inclusion and exclusion criteria were considered, 1212 eligible patients remained, 778 of whom provided written informed consent and were randomized. These 778 patients represent approximately 10.8% of the originally identified cohort. Mean BP on entry was 152/89 mm Hg and was similar among the treatment groups. More than 50% of the randomized participants already owned a home BP monitor. Mean age at baseline was 59 years, and >80% of the study population were white.
After 12 months of follow‐up, 730 patients returned for a follow‐up office BP measurement. At this final visit, BP control rates in each group were as follows: 31% (95% confidence interval [CI], 25%–37%) in the usual care group; 36% (95% CI, 30%–42%) in the home BP and patient Web group; and 56% (95% CI, 49%–62%) in the home BP and patient Web + pharmacist group. After adjustment for potential cofounders and as compared with the usual care group, this represented a nonsignificant 1.2 times (P=.20) improvement in BP control in the home BP and patient Web group and a significant 1.84 times (P<.001) improvement in BP control in the home BP and patient Web + pharmacist group. In the subgroup of patients with a baseline office systolic BP value ≥160 mm Hg, the differences were more substantial, with a more than tripling of BP control percentages in the home BP and patient Web + pharmacist group when compared with usual care. Overall reductions in systolic and diastolic BP for each group compared to baseline were as follows: −5.1/−3.3 mm Hg for usual care, −7.8/−4.1 mm Hg for home BP and patient Web, and −12.9/−6.2 mm Hg for home BP and patient Web + pharmacist. The home BP and patient Web + pharmacist group had significantly more hypertension medication classes added, 2.16, than either of the other 2 groups. Body mass index, physical activity, health‐related quality of life, and satisfaction with the health plan did not differ significantly among the 3 groups.
Among carefully selected patients enrolled in a single integrated health plan with poorly controlled office BP, use of home BP monitoring, a patient Web site integrated into the health plan’s EMR, and the assistance of a clinical pharmacist who could adjust medications through the EMR led to an improvement in BP control rates at 1 year when compared with usual care. A similar intervention that did not integrate a clinical pharmacist failed to result in substantial improvements in BP control.—Green BB, Cook AJ, Ralston JD, et al. Effectiveness of home blood pressure monitoring, Web communication and pharmacist care on hypertension control: a randomized controlled trial. JAMA. 2008;299:2857–2867.
Comment
Despite unprecedented attention over the past few decades, BP control rates in the United States still remain well below the Healthy People 2010 goal of 50% and remains low even among patients for whom medication has been prescribed and those who have access to health care. While the barriers to adequate BP control are complex and include patient, practitioner, and societal factors, it is important to realize that the biggest factors contributing to poor control rates may largely be the result of “systems issues” in health care delivery. Many would argue that our current health care delivery system, largely designed to deliver acute office‐based care, is not able to appropriately manage chronic conditions like hypertension that require out‐of‐office monitoring and consistent follow‐up. The present e‐BP study highlights the failings of our current health care system. Each potential participant contacted for this study had a previous diagnosis of hypertension, had previously been prescribed antihypertensive medications, and had access to health care including formulary coverage for their antihypertensive medications. Yet, despite this, 40% of potential participants who attended the screening visits did not have BP that was under adequate control. Among the 778 patients with poorly controlled hypertension who enrolled in the study, 85% were on <3 medications at baseline; since the Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure (JNC 7) defines resistant hypertension as BP not controlled despite the use of adequate doses of 3 antihypertensive medications (one of which is a diuretic), these data suggest that the majority of these patients had undertreated rather than resistant hypertension.
The Chronic Care Model may provide a better approach to the management of chronic medical conditions, such as hypertension. This model specifies the integration of 6 important domains of care: self‐management support, clinical information systems, delivery system redesign, decision support, health care organization, and community resources. Advocates of this model propose that integrating these domains into the health care delivery system will lead to the development of informed and activated patients interacting with prepared and motivated clinical practice teams. In the present clinical trial, investigators found that by employing a number of strategies consistent with the Chronic Care Model, BP control rates improved. The primary interventions studied included the use of home BP monitoring, patient‐centered Web‐based communication and education, and use of a clinical pharmacist.
Since only the intervention that used a clinical pharmacist achieved an improvement in BP control rates, this factor was likely the major contributor to improved BP control. Yet this study may not have been an adequate test of either home BP monitoring (since 50% of patients reported already having a home BP monitor at baseline) or Web‐based communication (since even the usual care group had access to a certain degree of Web‐based communication offered as part of the usual clinical practice in this group health plan). There were no changes from baseline in body mass index, physical activity, health‐related quality of life, and satisfaction with the health plan; thus, it appears that the main impact of the clinical pharmacist was (1) increasing the amount of communication between the patient and the clinical care team and (2) increasing the number of antihypertensive medications prescribed. In the intervention group that included use of a clinical pharmacist, the number of e‐mail interactions between patients and the care team, as measured by message threads, increased nearly 10‐fold compared with the usual care group.
Of importance, in this study, the clinical pharmacist was able to increase antihypertensive therapy, according to a prespecified care plan, without consulting a physician. This allowed the intervention group that included a clinical pharmacist to be on a greater number of antihypertensive agents than the other 2 groups. Previous studies have also found that nonphysician care teams that had the ability to increase medications according to a prespecified stepped‐care protocol led to improved BP control. In fact, this is the strategy employed by randomized clinical trials of antihypertensive therapy in which medications are usually administered and increased according to protocol by nonphysician study personnel, with physician oversight and input only when necessary.
Whether the present results are durable past 12 months or can be replicated in populations that are less highly selected, including those more ethnically diverse, with less access to home BP monitors, with less education, of a lower socioeconomic class, or less motivated should be the subject of future research. At present, in order to further improve BP control rates, health plans and provider groups should integrate elements of the Chronic Care Model into their health care delivery system if possible. This should include the increased use of clinical pharmacists, patient self‐monitoring of BP, and Web‐based communication between patients and their providers.