Use of strategies to improve antihypertensive medication adherence within United States outpatient health care practices, DocStyles 2015‐2016

Abstract

Patients’ adherence to antihypertensive medications is key to controlling high blood pressure. Evidence‐based strategies to improve adherence exist, but their use, individually and in combination, has not been described. 2015‐2016 DocStyles data were analyzed to describe health care professionals’ and their practices’ use of 10 strategies to improve antihypertensive medication adherence across 3 categories: prescribing, education, and tracking/encouragement. Among 1590 respondents, a mean of using 5 strategies was reported, with individual strategy use ranging from 17.2% (providing patients adherence‐related rewards) to 69.4% (prescribing once‐daily regimens). Those with higher odds of using ≥7 strategies and strategies across all 3 categories included: (1) nurse practitioners compared to family practitioners/internists and (2) health care professionals in practices with standardized hypertension treatment protocols who routinely recommend home blood pressure monitor use compared to respondents without those characteristics. Despite using an array of evidence‐based adherence‐promoting strategies, additional opportunities exist for health care professionals to provide adherence support among hypertensive patients.

1. INTRODUCTION

Approximately 1 in every 3 American adults have hypertension1 and only about half have their blood pressure controlled.2 In addition to therapeutic lifestyle modifications as a primary intervention for blood pressure control among those with hypertension, adherence to prescribed antihypertensive medication (antihypertensives) regimens is also needed. High adherence to antihypertensives has been associated with greater odds of blood pressure control.3, 4 Yet, cardiovascular‐related medication non‐adherence is common5, 6, 7 and improving adherence requires multifaceted approaches from various stalkeholders.8 Thus, efforts to improve antihypertensive adherence is a focus of national efforts to improve blood pressure control and cardiovascular disease outcomes, such as the US Department of Health and Human Services’ Million Hearts^© initiative.9

Patient‐reported barriers to antihypertensive medication adherence include having negative perceptions about the affordability and necessity of the medications to manage hypertension, forgetting or being physically unable to take the medications as prescribed, and having real or perceived side effects or adverse events related to taking the medication.10 Health care professionals (HCPs) are well positioned to help their patients overcome many of these barriers to adherence by working in collaboration with their patients and other providers to identify, remove, or minimize the barriers. While there is no single gold standard strategy for improving adherence, it is generally recognized that using multi‐pronged interventions is the best approach.6, 11 These strategies can include prescribing generic medications12, 13 and fixed‐dose combinations,14, 15, 16, 17 synchronizing medication fills (eg, coordinating refills),14, 18, 19 using longer days’ supply per fill,20 monitoring and assessing adherence,21 using devices that provide reminders about dosing,21 improving patient knowledge about the importance of medications,22 and using team‐based care for medication management and education.23

Although prior studies have examined the effectiveness of individual strategies for HCPs to improve adherence to antihypertensives, a study assessing the uptake of these strategies in health care practices across the US is lacking. Our null hypothesis was that there would be no difference in the use of strategies to improve medication adherence by provider and practice characteristics. Thus, this study used data from the 2015‐2016 DocStyles survey to describe HCPs and their practices’ use of strategies to improve patient adherence to antihypertensives in the outpatient setting, as well as methods HCPs use to measure adherence. Furthermore, the study determined if there were any differences in the use of these strategies by HCP and practice characteristics.

2. METHODS

2.1. Data source

We utilized data collected in 2015 and 2016 via the DocStyles survey, which is a web‐based survey of HCPs about their attitudes and behaviors towards health issues. The annual survey is administered by Porter Novelli Public Services from June‐July each year. Respondents were sampled from SERMO's Global Medical Panel, which includes >350 000 medical professionals in the US.24 Quotas were set for 1000 primary care physicians and 250 nurse practitioners (NPs) to participate in the survey annually. Individuals were eligible for the survey if they practiced in the United States, have been in practice for at least 3 years, actively saw patients, and worked in an individual, group, or hospital practice. HCPs who were defined by SERMO to be active members (those who have a history of higher survey completion rates) were invited to participate, followed by less active members, until all of the aforementioned quotas were filled for the year. Invitations were sent by email and respondents were able to terminate the survey at any time.

During the study years, 3284 primary care physicians and NPs were invited to participate, and 2507 completed the survey (76.3%). Reasons for non‐completion included terminating the survey prior to completion, not meeting screening questions, survey quotas being fulfilled, and non‐response to the survey invitation. HCPs who completed the survey were paid an honorarium between $69 and $76, depending on the number of questions they were asked to answer.

2.2. Study design

We limited our analytic cohort to: unique respondents from 2015 and 2016, using the most recent data available, including those who practiced in outpatient settings and were family practitioners (n = 687), internists (n = 598), or NPs (n = 305). HCPs’ perceptions of strategies used in their practice to improve antihypertensive adherence were assessed using the question, “Which of the following supports does your practice consistently use to improve patients’ adherence to their antihypertensive medication regimens?” Response options were randomized and included 10 strategies in 2015, including an additional option for “none of these/not sure.” In 2016, 1 additional strategy was added (medication synchronization), and was included in our descriptive unadjusted analyses, but not in any models. HCPs were instructed to select all answer choices that applied. We secondarily categorized each strategy into 1 of 3 overarching categories for analyses: prescribing, education, and tracking/encouragement strategies.

In 2016, among HCP respondents who indicated that their practice uses strategies to regularly assess adherence, a follow‐up question about methods the HCP uses was asked with, “How do you regularly assess your patients’ adherence to their antihypertensive medication regimens?” Response options were randomized and included 5 methods, in addition to “none of these/not sure.” HCPs were instructed to select all answer choices that applied.

2.3. Statistical analyses

Covariates included in the analyses comprised of HCP respondent characteristics, including: age, sex, race/ethnicity, years practicing medicine, HCP type (family practitioner, internist, NP), average number of patients treated per week, and recommendations to use home blood pressure monitoring (HBPM) as part of the hypertension treatment regimen; as well as information about the HCP's main practice setting, including individual vs group practice, urbanicity of the practice location, and the practice's status of using a standardized hypertension treatment protocol (referred to here as hypertension protocols). Variables were categorized based on the distribution of data for continuous variables as well as methods from prior studies. Hypertension protocol use was categorized as either “yes” or “no” (includes “no, but implementing,” “no and not implementing,” and “don't know”).

The prevalence of individual strategy use was calculated. Additionally, the prevalence of methods HCPs use to measure adherence was calculated. Three multivariate logistic regression models were developed to calculate odds ratios (ORs) and 95% confidence intervals (CIs). First, to examine the odds of HCPs reporting that their practice uses each individual strategy, models were constructed for each of the 10 strategies. Second, to examine HCPs’ perceptions of their practice using a wide breadth of strategies to improve adherence, the odds of reporting at least 1 strategy from each of the 3 pre‐defined categories (prescribing, education, and tracking/encouragement) was calculated. Third, to assess HCPs’ perceptions of their practice using a high intensity of strategies, the odds of reporting using ≥7 strategies to improve adherence was calculated. To create the most parsimonious models, backwards selection was used to individually remove variables that were not statistically significant at P < .05; however, HCP type, patient volume, use of hypertension protocols, and HBPM were determined to be important variables of interest a priori and were kept in all final models. All analyses were conducted using SAS version 9.3.

3. RESULTS

A total of 1590 HCPs met the inclusion criteria for this study (535 in 2015; 1055 in 2016). The mean age of included HCPs was 47.3 years and 60.6% were male (Table 1); 43.2% were family practitioners, 37.6% were internists, and 19.2% were NPs.

Table 1.

Health care professional and practice characteristics of respondents, DocStyles 2015‐2016 (N = 1590)

Characteristic	Respondents (n)	%
Health care professional (HCP) characteristics
Mean age, y (SD)	1590	47.3 (10.3)
Sex, %
Male	963	60.6
Female	627	39.4
Race/ethnicity, %
White, non‐Hispanic	1002	63.0
Black, non‐Hispanic	51	3.2
Asian, non‐Hispanic	367	23.1
Other/multiracial, non‐Hispanic	110	6.9
Hispanic	60	3.8
Mean years in practice (SD)	1590	16.3 (8.8)
HCP type, %
Family practitioner	687	43.2
Internist	598	37.6
Nurse practitioner	305	19.2
HCP's average number of patients per week, %
≤75	440	27.7
76‐125	800	50.3
≥126	350	22.0
Recommend HBPM as part of treatment regimens, %
Yes	1168	73.5
No	422	26.5
Practice characteristics
Practice setting, %
Individual practice	378	23.8
Group practice	1212	76.2
Metropolitan statistical area of practice, %
Metropolitan	1212	76.2
Micropolitan	110	6.9
Rural/small town	63	4.0
Missing	205	12.9
Hypertension treatment protocol use within practice, %
Yes, and EHRs used	443	27.9
Yes, and EHRs not used	120	3.8
No/don't know	1027	64.6
Strategies
Mean number of adherence promoting strategies used (SD)	1590	5.2 (2.9)

EHR, electronic health record; HBPM, home blood pressure monitoring; HCP, health care professional; SD, standard deviation.

3.1. Use of individual strategies

Out of 10 strategies, HCPs reported their practice used a mean of 5.2 (standard deviation: 2.9) strategies to improve patient adherence to antihypertensives (Table 1). The prevalence of practices’ use of each of the 10 strategies ranged from 17.2% (providing patients adherence‐related rewards) to 69.4% (prescribing once‐daily regimens; Table 2). A higher proportion of HCPs reported that their practice uses at least one of the strategies within the prescribing category (86.3%) compared to strategies in the education and tracking/encouragement categories (78.5% and 70.4%, respectively). When adjusted logistic regression models were constructed for each of the 10 strategies, both family practitioners and internists compared with NPs had statistically significantly lower odds of reporting strategy use for 3 out of the 6 total strategies in the education and tracking/encouragement categories (Figure 1). In contrast, differences by HCP type for strategy use within the prescribing category were non‐significant. Additionally, HCPs in practices that use HBPM as part of hypertension treatment regimens consistently showed significantly higher odds of using individual strategies across all 3 categories compared to practices that do not use HBPM (Table S1). Furthermore, HCPs who reported seeing a higher patient volume had a trend towards having lower odds of using each of the individual strategies, particularly strategies within the prescribing and education categories.

Table 2.

Prevalence of health care practices’ utilization of strategies to improve adherence to antihypertensive medication regimens, DocStyles 2015‐2016 (N = 1590)

Individual strategies and strategy categories	%	95% LCL	95% UCL
1. Prescribing strategies	86.3a
Prescribe once‐daily regimens, when possible	69.4	67.2	71.7
Prescribe medications included in the patient's insurance coverage formulary, when possible	61.8	59.4	64.2
Prescribe medications for longer intervals (90 d fills vs 30 d fills), when appropriate	59.9	57.5	62.3
Prescribe fixed‐dose combination pills, when possible, for patients needing more than one medication to control their blood pressure	48.2	45.8	50.7
Attempt to synchronize their medication refills to limit pharmacy tripsb	38.6	35.6	41.5
2. Education strategies	78.5a
Discuss with patients potential side effects of any medications when initially prescribed and at every office visit thereafter	56.2	53.7	58.6
Provide all prescription instructions clearly in writing and verbally	55.0	52.6	57.5
Ensure patients understand their risks if they do not take medications as directed. Ask patients about risks, and have them restate the positive benefits of taking their medications	53.7	51.3	56.1
3. Tracking & encouragement strategies	70.4a
Encourage patients to use medication reminders (eg, pill boxes, alarms)	53.6	51.2	56.1
Regularly assess their adherence (eg, via self‐reported adherence, pill counts, surveys)	45.2	42.7	47.6
Provide rewards for medication adherence (eg, praise adherence, arrange incentives)	17.2	15.3	19.03
Other response option
None of these/not sure	4.5	3.5	5.5

LCL, lower confidence limit; UCL, upper confidence limit.

^aPrevalence of using at least 1 strategy within this category.

^bResponse option for the 2016 DocStyles survey only (n = 1055).

Figure 1.

Odds ratios and 95% confidence intervals of family practitioners and internists reporting their practice uses individual strategies to improve adherence to antihypertensives compared to nurse practitioners (referent). Models included adjustment for age, health care professional type, patient volume, use of recommendations for home blood pressure monitoring, and the practices’ use of standardized hypertension treatment protocols

3.2. Use of a wide breadth and high intensity of strategies

Our results showed that 58.2% of HCPs reported that their practice used a wide breadth of strategies and 37.6% used a high intensity of strategies (Table 3). Saturated adjusted models with all covariates are provided in Table S2. In the final adjusted model that examined the intensity of strategies used compared to NPs, the odds of reporting that their practice uses a higher intensity of strategies was lower among internists (OR, 0.59; 95% CI, 0.43‐0.80) and family practitioners (OR, 0.68; 95% CI, 0.50‐0.92; Table 3). Additionally, HCPs who worked in practices with hypertension protocols and who recommended using HBPM as part of hypertension treatment regimens had significantly higher odds of reporting that their practice uses a high intensity of strategies (OR, 1.32 and 4.89, respectively). Finally, HCPs who had a higher volume of patients (≥126 per week) had lower odds of their practice using a wide breadth of strategies compared to those with a lower volume (≤75 per week; OR, 0.71; 95% CI, 0.51‐0.98). Similar patterns were observed in the analysis that examined the breadth of strategies used (Table 3).

Table 3.

Final models for multivariate logistic regression on HCPs reporting their health care practice uses a wide breadth and high intensity of strategies to improve adherence to antihypertensives, DocStyles 2015‐2016 (N = 1590)

Characteristic	Wide breadth of strategies (≥1 strategy from each category)			High intensity of strategies (≥7 strategies)
	Prevalence, %	OR (95% CI)	P	Prevalence, %	OR (95% CI)	P
Total	58.2			37.6
Health care professional (HCP) characteristics
Agea
<47	53.1b	0.64 (0.52‐0.79)	<.0001	34.6b	0.77 (0.62‐0.95)	.0169
≥47	63.5	1.00		40.6	1.00
HCP type
Family practitioner	57.8b	0.82 (0.60‐1.12)	.2171	37.1b	0.68 (0.50‐0.92)	.0130
Internist	54.0	0.75 (0.55‐1.03)	.0733	31.8	0.59 (0.43‐0.80)	.0007
Nurse practitioner	67.2	1.00		49.8	1.00
HCP's average number of patients per week
≤75	63.4b	1.00		41.4b	1.00
76‐125	58.9	0.93 (0.71‐1.20)	.5693	40.4	1.18 (0.91‐1.54)	.2060
≥126	50.0	0.73 (0.53‐0.99)	.0459	26.3	0.71 (0.51‐0.98)	.0402
Recommend HBPM as part of treatment regimen
Yes	65.8b	3.14 (2.48‐4.00)	<.0001	46.0b	4.89 (3.62‐6.61)	<.0001
No	37.2	1.00		14.2	1.00
Practice characteristics
Hypertension treatment protocol use within health care practice
Yes	63.2b	1.45 (1.16‐1.81)	.0012	41.6b	1.32 (1.05‐1.66)	.0156
No/don't know	55.4	1.00		35.4	1.00

CI, confidence interval; HBPM, home blood pressure monitoring; HCP, health care professional; OR, odds ratio.

^aCategorization developed using the mean age of respondents.

^bStatistically significant group differences in prevalence rates using the Chi‐square test.

3.3. Adherence assessment methods

In 2016, 432 (40.9%) HCPs indicated that their practice regularly assesses patients’ adherence to antihypertensives. Among those HCPs, the assessment method most commonly used was self‐reported adherence (90.7%) and the least commonly used was structured surveys (eg, Morisky Medication Adherence Scale;25 10.4%; Table 4).

Table 4.

Among practices that regularly assess antihypertensive medication adherence, prevalence of HCPs’ strategies to measure adherence, DocStyles 2016 (N = 432)

Medication adherence assessment method	Prevalence, %	95% LCL	95% UCL
Self‐reported adherence during visits	90.7	88.0	93.3
Feedback from pharmacies on patients’ refill status	47.0	42.5	51.6
Feedback from health plans on patients’ refill status	27.3	23.3	31.4
Pill counts	19.9	16.3	23.5
Structured surveys (eg, Morisky Medication Adherence Scale)	10.4	7.6	13.1
None of these/not sure	2.3	1.0	3.7

HCP, health care professional; LCL, lower confidence limit; UCL, upper confidence limit.

4. DISCUSSION

Among the 1590 respondents to the 2015‐2016 DocStyles surveys, we found a wide range of evidence‐based strategies used by HCPs and their practices to improve antihypertensive medication adherence. 86.3% of HCPs reported that their practice used at least one individual prescribing strategy and 78.5% and 70.4% reported using at least one education and tracking/encouragement strategy, respectively. NPs were more likely to report that their practice uses a wider breadth and higher intensity of strategies, especially those categorized as education and tracking/encouragement strategies. Less than half of practices regularly assessed patients’ adherence, with self‐reported adherence being the most common method used by HCPs.

Our study found that around half of HCPs reported that their practice used education and tracking/encouragement strategies, such as providing prescription instructions clearly in writing and verbally, discussing potential side effects, and regularly assessing adherence. These potential missed opportunities to improve care need to be identified and improved on within health care practices. Even though prescribing strategies were overall more frequently used, only 48% of HCPs reported that their practice prescribed fixed‐dose combination pills when possible and only 39% stated their practice used medication synchronization as a strategy to reduce trips to pharmacies. These are likely important gaps, where increased uptake is needed to support improved adherence to antihypertensives. Medication synchronization involves the coordination of a patient's prescriptions to be filled at a common time (eg, monthly or quarterly fill date) and the use of fixed‐dose combination pills is an opportunity to reduce the complexity and cost of medications. Prior studies have found that medication complexity and less refill consolidation are associated with lower adherence to cardiovascular medications.14, 18, 19 Additionally, a national‐level study of Medicare beneficiaries found that those who were not prescribed fixed‐dose combinations, and who were prescribed more than 1 class of antihypertensives had lower adherence.5 Strategies characterized as tracking and encouragement were reported to be less frequently used in our study. There may be future growth in this area, especially with programs that harness mobile health technology to track, remind, and reinforce patients’ medication taking behavior and transmit adherence summaries to their HCPs to improve blood pressure control.26

In 2014, the Community Preventive Services Task Force found strong evidence that patients who received care from nurses and pharmacists through team‐based care had improved blood pressure control.23 Community pharmacists play a vital role in improving care among those with uncontrolled hypertension, particularly when they participate in community‐clinical linkages with HCPs in clinical settings.27 Pharmacists can improve blood pressure control through a variety of mechanisms, including communicating with providers to coordinate medication synchronization, participating in collaborative practice agreements to reduce care fragmentation, and using e‐prescribing that has bi‐directional messaging with prescribers.28, 29 Although our study found that almost half of HCPs reported receiving feedback from pharmacies regarding their patients’ adherence to antihypertensives, the regularity in which this occurs and the effectiveness of the communication to drive additional action to address the non‐adherence needs to be assessed further.

Our study found that compared to HCPs who reported seeing fewer patients per week (≤75), those who saw more patients (≥126) had lower odds of perceiving that their practice uses a wide breadth and high intensity of strategies to improve medication adherence. While this relationship needs further exploration, this finding could be an indicator of missed opportunities due to time and/or resource constraints in larger volume practices. Adoption of standardized hypertension treatment protocols, especially protocols integrated within EHR systems, can help close gaps in missed opportunities by translating guideline‐based algorithms into clinical decision support tools that support the reduction of therapeutic inertia and promote medication adherence.30, 31 Use of protocols can promote adherence by having, as the default, recommendations for drug therapy that reduce costs to patients (eg, prescription of generic vs brand medications) and that simplify the medication regimen (eg, prescription of fixed‐dose combinations [when appropriate] and regularly scheduled medication synchronization reminders). Moreover, protocols can prompt HCPs across the entire health care team, unweighting the responsibility from any one provider, to regularly assess for medication adherence and then address barriers to adherence, such as cost, access, side effects, and need for further education. National efforts, including the Million Hearts^© initiative, provide resources such as sample protocols to support the need for protocol‐driven care to increase blood pressure control in clinical settings. Million Hearts and the Community Preventive Services Task Force also support the use of HBPM for hypertension management and control.32 HBPM provides the opportunity for patients to be more informed of their blood pressure readings and share regular measurements with their primary care provider to titrate medication regimens33 and provide further education. At least one study has found that the independent effect of HBPM is improved adherence to medication regimens,34 although further investigation of this association is needed.35

4.1. Limitations

This study has several limitations. First, the survey specifically asks respondents about strategies that their practice uses to improve adherence to antihypertensive medications. Therefore, our findings were limited to perceptions HCPs have about their practices’ use of strategies and it is unknown if their responses accurately describe use of the strategies by others in the practice or individually by the respondent. Second, the DocStyles survey invites respondents based on their engagement level, so those who have a history of a higher survey completion rate are sent survey invitations first. In addition, the survey is voluntary, Internet‐based, and has predetermined quotas. Social desirability bias may also be present, which would potentially result in over reporting of strategies. Third, our study combined data from unique respondents during 2 separately conducted survey cycles (2015 and 2016), with the assumption that factors such as practice patterns and efforts to improve medication adherence did not significantly change during this period. Despite this potential limitation, our combination of study years provided a more robust sample and statistical power to compare strategy use by HCP characteristics. Fourth, the survey does not capture data on estimated patient adherence levels to antihypertensives, or the uptake and adoption of the strategies by the patients. Thus, we are unable to state whether reports of using a higher intensity and breadth of strategies resulted in improved outcomes, such as higher adherence to antihypertensive medications. Finally, we designated each strategy into 1 of 3 overarching categories, but the strategies may overlap with other categories and may not function independently.

5. CONCLUSIONS

Adherence to antihypertensive medications is an integral component of achieving blood pressure control and requires the engagement of patients, health care professionals, health care delivery systems, and community‐based partners (eg, community pharmacists, community health workers, community organizations). HCPs in primary care settings have the opportunity to set the stage for high medication adherence through the multi‐pronged use of prescribing, education, and tracking strategies. This study found significant differences in health care practices’ use of evidence‐based strategies to improve medication adherence to antihypertensive medications. NPs and HCPs who worked in practices with standardized hypertension treatment protocols reported that their practice uses a wider breadth and higher intensity of strategies. There are potential opportunities for increased uptake of particular strategies to improve adherence and achieve blood pressure control. Implementation of strategies can be strengthened by using team‐based care, standardized protocols, health information technology, and existing resources. Further assessment of methods used by pharmacists to improve adherence, adherence assessment methods, and the effects of reported strategies on patient outcomes is needed.

DISCLOSURES

The authors report no conflicts of interest to disclose.

DISCLAIMER

The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and Prevention.

Supporting information

 

Chang TE, Ritchey MD, Ayala C, Durthaler JM, Loustalot F. Use of strategies to improve antihypertensive medication adherence within United States outpatient health care practices, DocStyles 2015‐2016. J Clin Hypertens. 2018;20:225–232. 10.1111/jch.13188