8.1 HBPM and Medication Adherence

Medication adherence can be defined as the process by which patients take their medication as prescribed. Adherence is a dynamic process which includes treatment initiation, execution (correct implementation of the dose regimen) and persistence [3]. Low adherence is the most common cause of treatment resistance [4] and it is associated with an increased risk of cardiovascular events [5]. Strategies focused on improving adherence are complex and related to patient’s behaviour and education, physician attitude, complexity of drug regimen and other health care supportive measures [3].

HBPM has been proposed as one of the strategies which may improve adherence. To date, several studies, one systematic review and one meta-analysis have reported data on the effect of HBPM in adherence improvement. In 2006, Ogedegbe and Schoenthaler [6] reviewed eleven randomized controlled trials, from which six reported improvements in medication adherence. However, in most of these trials HBPM was used in combination with other interventions, such as patient counselling, additional education, or medication reminders, thus resulting in difficulties in establishing the specific role of each component.

In one of the studies of HBPM as an isolated intervention, the authors reported a significant increase in the number of pills taken per week, from 5.8 to 6.6, assessed by electronic monitoring after 6 weeks of intervention in 628 patients [7]. In a second study, with a considerable smaller sample (60 patients), HBPM produced a non-significant (p > 0.05) increase in adherence (94% versus 88%) [8].

A total of 28 trials with more than 7000 participants were included in a meta-analysis of the effect of HBPM on medication adherence [9]. Again, most studies [10] combined HBPM with other co-interventions, with only 11 examining the role of HBPM alone. Adherence was assessed by different methods, including electronic monitoring, pill count, pharmacy fill data, and self-report adherence. A pooled analysis of 13 studies with quantitative data revealed a significant modest effect of HBPM (isolated or associated to other co-interventions) in medication adherence with a standardized mean difference of 0.21 (95% CI: 0.08–0.34).

These results are not easy to interpret, as two important confounders may either enhance or reduce the importance of HBPM in the adherence issue. First, as aforementioned, most of the studies included a mixture of interventions, with HBPM being only a part of them. The positive results derived from these studies could be either an effect of HBPM, of other interventions or a mixture of them. Another important aspect to be taken into account is how representative was the sample included in studies of adherence of the general population with hypertension. It is well known that most participants in such studies are highly motivated often exhibiting high levels of adherence at baseline, not necessarily representing the actual level of medication adherence in the general population with hypertension where 25% will never even fill a prescription for a new medication. High levels of adherence at baseline obviously reduce the possibility of any intervention impacting on such adherence.

8.2 HBPM and BP Control

The use of HBPM is associated with improved BP control compared to conventional management based on office BP measurements [11–14]. Several meta-analyses have reported small although significant reduction in both systolic and diastolic BP in patients using HBPM. In 2001, Cappuccio et al. [11] reviewed 18 randomized controlled trials and found a reduction of 4.2/2.4 mmHg with HBPM, these differences remaining significant (2.2/1.9 mmHg) after adjusting for publication bias. Several years later Agarwal et al. [12] reviewed 37 randomized controlled trials and also found significant reductions (2.6/1.7 mmHg) in BP. Additional findings were that BP reduction were more pronounced when HBPM was complemented with telemonitoring. Furthermore, HBPM led to more frequent down-titrations in antihypertensive therapy and was associated with less therapeutic inertia.

Two more recent meta-analyses have also found small, although significant, effects of HBPM on BP values. Uhlig et al. [13], in a meta-analysis of 26 comparative studies, found a significant effect of HBPM at 6 months (3.9/2.4 mmHg). The results were considered of moderate-strength evidence. However, changes after 1 year (1.5/0.8 mmHg) were no longer significant. In studies combining HBPM with other co-interventions, the beneficial effect on BP remained significant after 1 year. Finally, Tucker et al. [14] reported another meta-analysis which was based on individual patient data from 25 trials which included more than 7000 patients. Results revealed a lack of significant effect of HBPM alone, although its combination with other supportive co-interventions was associated with lower BP at 12 months.

As with medication adherence, the impact of HBPM on BP values appears heterogeneous, depending on the presence of other lifestyle changes, the method of treatment modification and also the use of different or the same targets for clinic or home BP.

8.3 HBPM in the Context of Self-Management

A constant finding in studies assessing the effect of HBPM on either adherence or BP control is that such intervention seems to be more efficacious when is accompanied by other co-interventions. Among them, self-management and telemonitoring have been studied more in depth.

The TASMINH2 study [15] examined the effect of self-management, consisting of self-titration of antihypertensive medication, in comparison with usual care in patients with uncomplicated hypertension whose systolic BP was not controlled while on two or less antihypertensive medications. Telemonitoring was also included in the intervention group, for safety purposes. The primary outcome (reduction in systolic BP) was significant in favour of the intervention group (mean difference 3.7 mmHg and 5.4 mmHg at 6 and 12 months, respectively).

These results were also reproduced in high-risk individuals with hypertension, defined by systolic BP above 130/80 mmHg and a history of stroke, coronary heart disease, diabetes or chronic kidney disease [16]. Corresponding differences in systolic and diastolic BP at 12 months were 9.2 and 3.4 mmHg, respectively.

The main limitation of studies reporting benefits of self-management is the generalizability to the whole population with hypertension, as not all patients are suitable for engagement in such self-management.

8.4 HBPM in the Context of Telemonitoring

Telemonitoring can be an alternative to self-management. The use of automated devices that are able to transmit the data to health care providers (usually physicians, nurses or pharmacists) has allowed the study of the effect of remote management on BP control. A clinical trial in US individuals with hypertension reported 12-month reductions of 11/6 mmHg for systolic and diastolic BP, in patients using transmission of data and receiving treatment adjustments from their pharmacist, when compared to usual care [17]. An extended analysis of the same cohort suggested that the effect remained clinically significant 1 year after the intervention was stopped, but was negligible 4 years later [10].

Meta-analyses focused on the antihypertensive effect of intervention based on HBP telemonitoring have revealed a significant effect on BP control (16% increase in the possibility of achieving such control) with an effect on BP reduction of 4–5 mmHg for systolic and 2–3 mmHg for diastolic BP [18, 19]. These results were accompanied by an increase in antihypertensive medication use without differences in adherence, number of office consultations or changes in the quality of life.

A recent trial has examined the effect of telemonitoring added to self-measurement alone on BP reduction. The TASMINH4 study [20] enrolled 1182 patients with hypertension from the UK, randomized to usual care (therapeutic decisions based on clinic BP measurements), self-monitoring alone (therapeutic decisions based on HBPM, with readings sent to the physicians one per month) and telemonitoring (data transmitted through SMS). In both intervention groups, physicians were asked to review the data once per month and to adjust antihypertensive treatment based on HBPM readings. After 12 months, both intervention groups had lower systolic BP in comparison to usual care (3.5 mmHg for self-monitoring alone and 4.7 mmHg for telemonitoring) without significant differences between them. Moreover, as observed in previous TASMIN studies [15, 16], patients engaged in self-monitoring had an increased medication use [20].

These results confirm that HBPM is a useful tool for achieving a better BP control, if used on a scheduled basis and when adjustments in antihypertensive treatment are based on such readings. Telemonitoring did not seem to add advantages in such experimental context, although it could be obviously useful in other circumstances, such as difficulties in access to physicians or in cases of doubts regarding the reliability of the data.

In conclusion, most of the trials suggest that the impact of HBPM on BP control and treatment adherences is more powerful if used in a context of other interventions, such as lifestyles changes, as well as other interventions that are able to improve adherence and to increase the implication of the patient in his/her management.