Strategies for Managing Hypertension: Is the Paradigm Shifting?

Print

Written By

Reviewed By

Citation

Mills KT, Obst KM, Shen W et al. Comparative Effectiveness of Implementation Strategies for Blood Pressure Control in Hypertensive Patients: A Systematic Review and Meta-analysis. Ann Intern Med. 2018;168(2):110-120.

Hypertension, considered a leading risk factor for cardiovascular morbidity and mortality, is poorly managed in our healthcare delivery system today.  Forty-five percent of all adults in the United States have high blood pressure — that’s over 100 million people!1,2  Of those treated with pharmacotherapy, more than half are not achieving their blood pressure (BP) goals.2 Thus, millions of Americans are receiving suboptimal care and this has significant consequences for individuals, their families, and our economy.  Multiple barriers to achieving optimal BP have been identified.3

 

A recently published systematic review and meta-analysis examines implementation strategies for improving BP control in patients with high blood pressure.4 The specific aim was to determine which implementation strategies work best.  See Table 1.

 

Table 1. Three Targets and Eight Implementation Strategies

 

Patient-Level

Provider-Level

Multilevel

Health Coaching

Directed at lifestyle changes and medication adherence

Provider training

Education for providers – guideline adherence and /or patient communications

Multilevel without team-based care

Address multiple barriers – i.e. provider training combined with health coaching with only one healthcare practitioner lead

Home BP monitoring

Self-monitoring; BP readings given to providers

Audit and feedback

Periodic patient outcome summaries given to providers for evaluation

Team-based care with a physician titrating medication

Collaborative care with >2 providers

 

Electronic decision-support

Computerized alerts, reminders, or order sets

Team-based care with non-physician providers titrating medications

Collaborative care with >2 providers

 

The investigators searched MEDLINE and Embase for randomized, controlled hypertension management studies published through September 11, 2017.  Studies were divided into 8 categories based on the implementation strategy used. Two independent reviewers examined the trials for inclusion based on the following criteria: 1) adults with an average SBP >140 and/or DBP > 90 mmHg or treated with antihypertensive medication; 2) the primary outcome was a net change in systolic BP (SBP) and/or diastolic BP (DBP); 3) the intervention targeted 1 or more barriers; 4) the control arm was limited to usual care; and 5) trial duration was at least 6 months. 

 

Study heterogeneity was evaluated using the Cochran Q test and quantified with the I2 index. Publication bias was also addressed. Weights for the models were calculated using a random effects meta-analysis. A sensitivity analysis was conducted for all participants with uncontrolled BP.  The study was funded by the National Institutes of Health.

 

A total of 100 articles with 121 comparisons and 55,920 participants were included in the analysis.  See Table 2.  The baseline median age derived from the mean values of the included studies was 60 years (range: 33-77); the median SBP was 148 mmHg (range: 124-181); and the median DBP was 86 mmHg (range: 70-105).  The median trial duration was 6 months (range: 6 months to 5 years).

 

Table 2:  Abbreviated Summary for the 8 Implementation Strategies

 

Implementation strategy

Number of studies

Number of participants

Range of mean baseline SBP and DBP mmHg

Study design:

% of the trials

Patient-level

Health coaching

39

10656

124-181; 70-105

Parallel RCT: 97 Cluster RT: 30

Factorial RCT:3

Home BP testing

29

7966

126-170; 72-104

Parallel RCT: 100
Cluster RT: 8

Provider-level

Provider training

6

17642

127-153; 74-96

Parallel RCT: 100 Cluster RT: 100

Audit and feedback

2

2121

133-146;73-89

Parallel RCT: 100 Cluster RT: 100

Electronic decision-support system

6

8229

136-158;75-89

Parallel RCT: 100 Cluster RT: 100

Multilevel

Multilevel without team-based care

8

3436

133-169; 73-95

Parallel RCT: 100 Cluster RT: 75

Team-based care with physician titrating meds

20

6680

127-162;76-93

Parallel RCT: 100 Cluster RT: 30

Team-based care with non-physician titrating meds

11

3417

136-174;76-99

Parallel RCT: 100 Cluster RT: 18

RCT, randomized controlled trial; RT, randomized trial

 

When adjusted for covariates, the three multilevel strategies were the most effective.  See Table 3.  Team-based care with medication titration by a non-physician resulted in the greatest reduction in mean SBP, followed by team-based care with physician medication titration, and multilevel strategies (i.e., combined patient and provider level) without team-based care.  Patient-level strategies (i.e., health coaching and home BP monitoring) also resulted in significant BP reductions. Of the provider-level support systems, only electronic decision support produced a modest but statistically significant improvement in BP.

 

Table 3. Net change in BP by Implementation Strategy

 

Net change in BP

(95% CI), mm Hg

Implementation Strategy

Systolic BP

Diastolic BP

Team-based care with titration by non-physician

-7.1 (-8.9 to -5.2)*

-3.1 (-4.1 to -2.2)*

Team-based care with titration by physician

-6.2 (-8.1 to -4.2)*

-2.7 (-3.8 to -1.5)*

Multilevel strategy without team-based care

-5.0 (-8.0 to -2.0)*

-2.9 (-5.4 to -0.4)*

Health Coaching

-3.9 (-5.4 to -2.3)*

-2.1 (-2.9 to -1.3)*

Electronic decision-support systems

-3.7 (-5.2 to -2.2)*

-1.5 (-1.9 to -1.1)*

Home BP monitoring

-2.7 (-3.6 to -1.7)*

-1.5 (-2.3 to -0.8)*

Provider training

-1.4 (-3.6 to 0.7)

-1.0 (-2.2 to 0.1)

Audit and feedback

-0.8 (-2.1 to 0.5)

-0.6 (-1.3 to 0.1)

*statistically significant (p<0.05)

Mean reductions estimated using generalized estimating equations and adjusted for sex, age, baseline SBP, trial duration, type of control group, and all other intervention strategies.

 

The authors concluded team-based care with non-physicians or physicians titrating medications as well as multilevel strategies were the most effective for achieving improved blood pressure control.  Strategies addressing patient-level barriers (i.e., health coaching and home BP monitoring) were effective but produced relatively modest reductions in BP.  Provider education and audits with feedback were ineffective when implemented in isolation (e.g. not part of a multilevel strategy). From an implementation perspective, the computer-mediated provider-directed strategies may be the cheapest to develop, execute, and maintain when compared to the long-term cost of hiring trained non-physician healthcare team members; however, when implemented as a stand-alone strategy, they are unlikely to deliver optimal BP lowering results. 

 

Previous well-designed trials also support team-based care models. A meta-analysis of pharmacist-led team-based care showed significant reductions in SBP (-7.6 mmHg, 95% CI: -6.3 to -9.0) and DBP (-3.9 mmHg, 95% CI: -2.8 to -5.1) compared to usual care.5 A meta-analysis of nurse-led models compared to usual care also demonstrated a robust reduction in SBP (-8.9 mmHg, 95% CI: -5.3 to -12.5) and DBP (-4.0 mmHg, 95% CI: -2.7 to -5.3).6

 

There is a growing body of literature to support non-physician-led teams to manage hypertension.  In January 2018, the Black Barbershop study, a randomized control, cluster design, multi-level intervention trial for uncontrolled hypertension, was published in the New England Journal of Medicine.7  The intervention group had monthly visits with a pharmacist and the control group received lifestyle education combined with primary care provider follow-up.  The pharmacists prescribed antihypertensive therapy using a step-wise approach.  The mean baseline SBP was 153 mmHg in the intervention group (n= 139) versus 155 mmHg in the control (n= 180). The average SBP at six months was -27.0 mmHg lower in the intervention group compared to -9.3 mmHg in the control, with a 21.6 mmHg greater mean reduction in the intervention arm (p < 0.001).  Most notable was the poorly controlled mean baseline SBP in the Black Barbershop study population and the rather robust reductions in SBP obtained.

 

A majority of the studies included in this systematic review and meta-analysis were completed in the U.S. or other high-income countries, thus the implementation strategies and results are relevant to most practice environments in North America.  Additionally, the BP reductions from these trials were demonstrated in populations with relatively mild to moderate hypertension. A sensitivity analysis, which only included trials with uncontrolled BP at baseline, showed an even greater SBP lowering effect for team-based care with titration by a non-physician or physician and a greater DBP lowering effect for all multi-level strategies.

 

There were a few limitations to the analysis. Despite searching ClinicalTrials.gov for unpublished clinical trials with neutral or negative findings, only a small number of studies have been conducted for some strategies (See Table 2). Additionally, the I2, a measure study heterogeneity not due to chance, was high.  This suggests that some of the effects in the studies may be due to unaccounted environmental or implementation factors and may not be reproducible. For example, system-level barriers, such as lack of performance standards or leadership commitment, were seldom addressed in trials.  Such factors can have a potentially large impact on team or provider performance and, ultimately, BP control. Prediction intervals were not reported in the article. Another factor that needs to be addressed in future research is the use of financial incentives.

 

As with other implementation programs (i.e., National Diabetes Prevention Program) that impact large populations with chronic illness, the feasibility of large scale implementation, long-term sustainability, and total cost must be explored to fully evaluate the utility of the strategy.  Commentaries have supported team-based care models but many acknowledge the relative lack of economic data.8 In this systematic review, an economic analysis of the eight strategies could not be completed due to the lack of data.4 However, some RCTs have included economic analyses and these studies indicate team-based care is likely cost-effective.9

 

This meta-analysis used a well-designed and accepted scientifically valid approach to identify implementation strategies to improve BP control among patients with hypertension.  Team-based care with non-physician providers was the most effective strategy. This analysis supports the 1A recommendation in the new 2017 ACC/AHA guidelines which advocates for structured, team-based interventions to achieve optimal BP control.1 As value-based care payment models are implemented and more responsibility is placed on health-systems and providers to achieve benchmarks, financial incentives will promote the use of implementation strategies that have a proven benefit.  Furthermore, healthcare delivery models should address the social inequities that exist and the limitations of single provider care.10  Is public health policy keeping up with the evidence?  How can we proactively shift the paradigm toward team-based care?