Content uploaded by Tom Fahey
Author content
All content in this area was uploaded by Tom Fahey on Mar 18, 2014
Content may be subject to copyright.
Cardiovascular medication: improving adherence
Search date April 2010
Liam Glynn and Tom Fahey
ABSTRACT
INTRODUCTION: Adherence to medication is generally defined as the extent to which people take medications as prescribed by their
healthcare providers. It can be assessed in many ways (e.g., by self-reporting, pill counting, direct observation, electronic monitoring, or by
pharmacy records). This review reports effects of intervention on adherence to cardiovascular medications however adherence has been
measured. METHODS AND OUTCOMES: We conducted a systematic review and aimed to answer the following clinical question: What
are the effects of interventions to improve adherence to long-term medication for cardiovascular disease in adults? We searched: Medline,
Embase, The Cochrane Library, and other important databases up to April 2010 (Clinical Evidence reviews are updated periodically; please
check our website for the most up-to-date version of this review). We included harms alerts from relevant organisations such as the US
Food and Drug Administration (FDA) and the UK Medicines and Healthcare products Regulatory Agency (MHRA). RESULTS: We found
39 systematic reviews, RCTs, or observational studies that met our inclusion criteria.We performed a GRADE evaluation of the quality of
evidence for interventions. CONCLUSIONS: In this systematic review we present information relating to the effectiveness and safety of the
following interventions: patient health education, prescriber education, prompting mechanisms, reminder packaging (calendar [blister] packs,
multi-dose pill boxes), and simplified dosing.
QUESTIONS
What are the effects of interventions to improve adherence to long-term medication for CVD in adults? ...... 4
INTERVENTIONS
INTERVENTIONS TO IMPROVE ADHERENCE TO
CARDIOVASCULAR MEDICATION
Likely to be beneficial
Prompting mechanisms ...................... 4
Simplified dosing .......................... 12
Unknown effectiveness
Patient health education ..................... 21
Prescriber education ....................... 17
Reminder packaging ........................ 18
Key points
•Adherence to medication is generally defined as the extent to which people take medications as prescribed by
their healthcare providers.
It can be assessed in many ways (e.g., by self-reporting, pill counting, direct observation, electronic monitoring,
or through pharmacy records). In this review, we have reported adherence to cardiovascular medications however
it has been measured.
•The RCTs we found used a variety of different interventions in different populations, measured adherence differ-
ently, and expressed and analysed results differently.
The diversity and complexity of interventions employed in RCTs makes it difficult to separate out any individual
components that might be of benefit.
•We found evidence that simplified dosing regimens may increase adherence compared with more complex regimens.
While simplifying the frequency of dosage may increase adherence, it is not known whether simplified regimens
may increase adherence when someone is taking multiple drugs, as may be the case with cardiovascular
medicines.
In altering a drug regimen simply to increase adherence, any changes could potentially affect the effectiveness
of the treatment, and could also potentially increase adverse effects.
•Prompting mechanisms may also increase adherence to medication.
Some prompting mechanisms may be simple and inexpensive (e.g., mailed reminders), while others (e.g., daily
telephone calls, installing videophones) seem impracticable for use in routine practice.
•Patient health education may also increase adherence to medication but more data are needed to draw conclusions.
Adherence behaviour is complex.Traditional education methods may fail to address this. However, more patient-
centred approaches, particularly those that are nurse- or pharmacist-led, using video or telephone strategies,
may be beneficial and require further investigation.
We found some evidence that a combination of strategies, such as education plus prompting, may be more
successful than a single educational strategy.
•We found no evidence from one RCT that reminder packaging (a calendar blister pack) was effective, and found
insufficient evidence on other types of reminder packaging such as multi-dose pill boxes.
•We found one RCT of prescriber education in a developing country, which showed that a 1-day intensive training
session of general practitioners on hypertension improved medication adherence compared with usual care but
Cardiovascular disorders
© BMJ Publishing Group Ltd 2011. All rights reserved. ..................... 1..................... Clinical Evidence 2011;04:220
..................................................
these data are not generalisable to the range of people taking cardiovascular medication so we cannot draw firm
conclusions about this intervention.
DEFINITION Definition of adherence: Adherence to a medication regimen is generally defined as the extent
to which people take medications as prescribed by their healthcare providers. [1] Adherence,
compliance, and concordance are often used interchangeably when studying health behaviour,
but their meanings are in fact different, particularly in the context of RCTs examining interventions
aimed at improving adherence. Adherence takes into account that people choose to take their
medicines, have control over their use, and develop an agreement with healthcare professionals
about their management. [2] The main difference between the terms "adherence" and "compliance"
is on a motivational level, with the latter suggesting that the patient is passively following the
physician's orders, and that the treatment plan is not based on a therapeutic alliance or contract
established between the patient and the physician. [1] Unfortunately, the term "concordance" has
occasionally, and not always appropriately, replaced the terms "compliance" or "adherence". [3]
"Concordance" aims to describe an agreement between patient and healthcare professional about
the whole process of medication-taking as part of a wider consultation, rather than describing the
specific extent to which medication is taken. [4] For the purposes of this review, "adherence" will
be defined as the extent to which people take medications as prescribed by their healthcare
providers.The reporting of adherence varies, with some studies reporting adherence as a dichoto-
mous outcome, and using an artificial cut-off point (e.g., 80% "adherent"), whereas other studies
compare study arms using continuous outcomes (e.g., a count of pills taken of 75% v 91%). Mea-
surement of adherence: The ideal measurement of adherence should: be usable over a prolonged
period; be unobtrusive; be non-invasive; be practicable and cheap; yield immediate results; and
not be open to manipulation. [5] Based on these stringent criteria, the objective measurement of
adherence is difficult, and poses a challenge for researchers and clinicians. Measurement of ad-
herence can be divided into "direct" (which demonstrate drug ingestion) and "indirect" (which do
not demonstrate drug ingestion) methods. Direct methods include observing people taking medica-
tion, or the measurement of medicine, metabolites, or biological markers in the blood. Although
objective and accurate, direct adherence measures are often impractical or too expensive for the
RCT setting. A variety of indirect adherence measures are commonly employed in RCTs, and each
one has strengths and weaknesses.These include self-reporting by patients, prescribing data, pill
counting, measurement of physiological markers, and electronic monitoring. Patient self-reporting
of adherence is simple, inexpensive, and probably the most practical and useful in the clinical setting.
It is, however, subject to considerable bias, as the person may wish to please the investigator, be
worried about admitting to not taking medication, or simply not accurately remember. Prescribing
data, such as the rate of prescription refills or cessation of refills (discontinuation rate), are easy
to obtain through pharmacies, but require a closed-pharmacy system to be accurate, and cannot
be regarded as equivalent to ingestion of medication. However, this information affords a useful
proxy, and may be easier to measure over long follow-up periods. [6] Pill counts provide a direct
measure of adherence. However, they may be manipulated by people if they are aware that the
pills are being counted (e.g., pill dumping), and it does not necessarily mean that medication has
been taken at the correct time. Measurement of physiological markers (e.g., measuring heart rate
in patients taking beta-blockers) is easy to perform, but is greatly limited by its assumption of a
cause-and-effect relationship, which is rarely applicable. Electronic monitoring methods have
greatly advanced recently and allow recordings of the timing and frequency of drug ingestion, which
make them the only method to provide data on drug-taking patterns. However, they are expensive,
and there is no guarantee that opening of the medication container is followed by ingestion of the
correct dose. It could also be argued that placing an electronic cap to measure compliance is an
intervention in itself as people are aware that they are being monitored (Hawthorne effect). [7] This
effect may or may not persist in the longer term when people become used to the electronic cap.
Although electronic monitoring is closest to a "gold standard" in measuring adherence, it has so
far been used mainly as a research tool owing to its relatively high cost.
INCIDENCE/
PREVALENCE Not applicable for this review.
AETIOLOGY/
RISK FACTORS The reasons for not adhering to prescribed cardiovascular medication are complex, and non-adher-
ence may lead to various sequelae. For example, the prescribing clinician may alter or discontinue
a regimen believing it not to be working when, in fact, it may have been taken only inconsistently
or not at all. Failure to adhere to a prescribed regimen may increase adverse effects from the reg-
imen, in that medication is taken incorrectly, and may fail to improve symptoms from the underlying
condition for which it was prescribed.Interventions to improve adherence: Interventions to improve
adherence can potentially be divided into a variety of different categories or groupings. In this review
we have grouped RCTs under the categories of: prescriber education; prompting mechanisms;
patient health education; simplified dosing; and reminder packaging (blister packs and pill boxes),
and have explained what we have included under each category where necessary. However, inter-
© BMJ Publishing Group Ltd 2011. All rights reserved. ........................................................... 2
Cardiovascular medication: improving adherence
Cardiovascular disorders
ventions to improve adherence are complex by nature and will often be combined in a multi-facto-
rial or "complex intervention" approach.This approach is necessary as there are many factors that
contribute to poor adherence, although this does make it difficult to tease out the individual compo-
nents of many adherence interventions. Educational interventions can be directed at prescribers,
patients, and their family members using written material, videotapes, or individual or group training.
Prompting mechanisms are intended to stimulate medication-taking through mailed or telephoned
reminders or through the use of electronic medication-reminder caps. Simplified dosing is intended
to improve adherence through the reduction of dosing frequency (e.g., once-daily regimens v twice-
daily regimens, or twice-daily regimens v 3-times-daily regimens). Reminder packaging falls into
two distinct categories: those that are packaged in pill boxes (multi-compartment compliance aid,
dose administration aid) or those that are pre-packaged into blister packs (calendar blister, unit
dose, monitored dosage system). Definitions of terms relating to reminder packaging are reported
in table 1, p 30 .
PROGNOSIS Patterns of medication-taking behaviour and adherence: Patterns of medication-taking behaviour
have been accurately described using electronic monitoring devices. Six general patterns of taking
medication emerge among people treated for chronic illnesses who continue to take their medica-
tions: approximately one sixth come close to perfect adherence to a regimen; one sixth take nearly
all doses, but with some timing irregularity; one sixth miss an occasional single day's dose and
have some timing inconsistency; one sixth take drug holidays three to four times a year, with occa-
sional omissions of doses; one sixth have a drug holiday monthly or more often, with frequent
omissions of doses; and one sixth take few or no doses while giving the impression of good adher-
ence. [8] [9] Most deviations in taking medication occur as omissions of doses (rather than additions)
or delays in the timing of doses. [10] [11] Levels of adherence are poorly described, with those
studies of higher quality limited by smaller numbers, and those studies of larger populations limited
by crude measures of adherence. However, in terms of adherence to cardiovascular medication,
most studies have examined adherence in relation to lipid-lowering drugs. It is evident that target
cholesterol concentrations are only achieved in less than 50% of people receiving lipid-lowering
drugs, and that only one in four people continue taking cholesterol-lowering drugs long term. [12]
[13] In adherence studies of people without CHD taking lipid-lowering drugs for the purposes of
primary prevention, discontinuation rates are higher compared with people taking lipid-lowering
drugs for the purpose of secondary prevention, indicating a possible relationship between adherence
and awareness of illness. [14] [15]
AIMS OF
INTERVENTION To increase adherence to cardiovascular medication in order to achieve treatment goals; to prevent
relapse of disease; to reduce morbidity; to reduce mortality; to improve quality of life, with minimal
adverse effects.
OUTCOMES Adherence to medication, however measured. Adherence is often measured by pill count, pre-
scription renewal requests, self-reporting, and electronic monitoring. Adverse effects.
METHODS Clinical Evidence search and appraisal April 2010.The following databases were used to identify
studies for this systematic review: Medline 1966 to April 2010, Embase 1980 to April 2010, and
The Cochrane Database of Systematic Reviews 2010, Issue 3 (1966 to date of issue). An additional
search within The Cochrane Library was carried out for the Database of Abstracts of Reviews of
Effects (DARE) and Health Technology Assessment (HTA). We also searched for retractions of
studies included in the review. Abstracts of the studies retrieved from the initial search were assessed
by an information specialist. Selected studies were then sent to the contributor for additional as-
sessment, using predetermined criteria to identify relevant studies. In addition, we use a regular
surveillance protocol to capture harms alerts from organisations such as the FDA and the MHRA,
which are added to the reviews as required. Study design criteria for inclusion in this review were:
published systematic reviews of RCTs and RCTs in any language, undertaken in adults. Further
studies were identified from a search of bibliographies of identified systematic reviews.We included
RCTs whatever the level of blinding (whether double-blind, single-blind, or open). RCTs had to
contain at least 20 people in total, or at least 10 per study arm, of whom more than 80% were fol-
lowed up.The minimum length of follow-up required to include studies was 6 weeks.We have in-
cluded RCTs in people with CVD and excluded RCTs in mixed populations (i.e., RCTs that also
included people with other diseases, in which people with CVD did not form the majority).We ex-
cluded RCTs in hospitalised people, and included RCTs in people in the community or seen as
outpatients, who were responsible for administering their own medication. Difficulties in evaluating
RCTs included analysing a multiplicity of different interventions and combinations of different ele-
ments that were not easily categorised. Therefore, in each treatment option, we have explicitly
stated what we have included under that option heading.We have excluded RCTs that employed
complex interventions (i.e., mixtures of different elements) in which the individual effects of our in-
tervention of interest could not be separately assessed.We have also excluded RCTs that did not
© BMJ Publishing Group Ltd 2011. All rights reserved. ........................................................... 3
Cardiovascular medication: improving adherence
Cardiovascular disorders
directly report adherence as an outcome, or reported an adherence outcome that was not clearly
defined.There was a wide variation between RCTs in how adherence was measured (e.g., whether
by pill count, self-reporting, electronic methods, or the number of repeat prescriptions obtained),
with no standard method employed.We have therefore included RCTs however adherence was
measured, but explicitly stated the adherence outcome measure employed in each RCT. We
identified a number of systematic reviews that employed different inclusion criteria, and that cate-
gorised interventions in different groupings.The systematic reviews did not pool data because of
differences between included RCTs (including study designs, interventions employed, and outcome
measures assessed). We have therefore reported each of the RCTs included in the systematic
reviews separately. Measures to increase adherence may or may not have adverse effects (e.g.,
regular contact and stressing the importance of medication and possible adverse effects of non-
compliance may increase anxiety in some people). For adverse events we have reported harms
data relating directly to the adherence intervention employed.We have not reported harms data
relating to the drug treatments used, as adherence is our outcome of interest. The exception to
this is in the simplified-dosing option, where we have reported drug adverse effects, as the inter-
vention directly alters the drug regimen (e.g., to once-daily dosage, rather than twice-daily). Hence,
in this case, differences in drug adverse effects between the regimens are due to the adherence
intervention itself.To aid readability of the numerical data in our reviews, we round many percentages
to the nearest whole number. Readers should be aware of this when relating percentages to
summary statistics such as relative risks (RRs) and odds ratios (ORs). We have performed a
GRADE evaluation of the quality of evidence for interventions included in this review (see table, p
31 ). The categorisation of the quality of the evidence (high, moderate, low, or very low) reflects
the quality of evidence available for our chosen outcomes in our defined populations of interest.
These categorisations are not necessarily a reflection of the overall methodological quality of any
individual study, because the Clinical Evidence population and outcome of choice may represent
only a small subset of the total outcomes reported, and population included, in any individual trial.
For further details of how we perform the GRADE evaluation and the scoring system we use, please
see our website (www.clinicalevidence.com).
QUESTION What are the effects of interventions to improve adherence to long-term medication for CVD
in adults?
OPTION PROMPTING MECHANISMS.....................................................
•For GRADE evaluation of interventions for Cardiovascular medication: improving adherence, see table, p 31 .
•Prompting mechanisms may increase adherence to medication.
•Some prompting mechanisms may be simple and inexpensive (e.g., mailed reminders), while others (e.g., daily
telephone calls, installing videophones) seem impracticable for use in routine practice.
Benefits and harms
Prompting mechanisms versus usual care:
We found 11 systematic reviews (search dates 1996; [16] 2000; [17] 2002; [18] [19] [20] 2003; [21] [22] 2004; [23] 2007;
[24] 2008; [25] 2009 [26] ), which identified 5 RCTs of sufficient quality. [7] [27] [28] [29] [30] The reviews did not pool
data. Some of the RCTs had weak methods, and completeness of reporting varied widely among trials. Adherence
was measured in a variety of ways in the 5 RCTs (pill counts, pharmacy refill records, electronic caps) and overall
compliance was calculated in different ways, with no standard method employed. For full details of prompting
mechanisms used in RCTs, see further information on studies.
-
Adherence to medication
Compared with usual care Prompting interventions (including daily and weekly telephone calls, video-telephone
calls, mailed reminders, and electronic medication-reminder caps) may be more effective than usual care at improving
adherence in people taking cardiovascular medication. However, the practicality of some of these interventions in
routine clinical practice is unclear (very low-quality evidence).
Favours
Effect
size
Results and statistical
analysisOutcome, InterventionsPopulation
Ref
(type)
Adherence to medication
P <0.05 among groupsCompliance (baseline to post-
intervention) monitored by
60 people age 65
years or older, diag-
[7]
RCT Direct statistical analysis of tele-
phone group or video-telephone
electronic caps on medication
bottles , 6-week intervention
nosis of chronic
heart failure, had
© BMJ Publishing Group Ltd 2011. All rights reserved. ........................................................... 4
Cardiovascular medication: improving adherence
Cardiovascular disorders
Favours
Effect
size
Results and statistical
analysisOutcome, InterventionsPopulation
Ref
(type)
3-armed
trial group versus usual care not re-
ported, but higher rates of adher-
ence in prompting mechanism
groups
phase followed by 2-week post-
intervention compliance moni-
toring period
76% to 74% with daily telephone
calls
to have telephone
socket, home not
in high-crime area,
Mini-Mental State
Examination
(MMSE) score of
20 or better. 82% to 84% with daily video-
telephone calls
81% to 57% with usual care
Absolute numbers not reported
prompting interven-
tion (electronic
cap)
P = 0.0002Mean % compliance (defined
as doses consumed/doses
prescribed x 100) , number of
remaining doses in each vial
counted at 12 weeks
70 people with hy-
pertension on long-
term treatment,
age 50 years or
older, on one or
more drugs
[27]
RCT
95% with electronic cap on medi-
cation vial
78% with standard cap
Absolute numbers not reported
Not significant
Reported as not significant
P value not reported
Compliance (measured by pill
and packet counts) , 6 weeks
92% for lovastatin and 93% for
colestipol with telephone call
30 people with
CABG or PTCA in
the last 7 to 30
days, baseline
fasting LDL
130 mg/dL or high-
[28]
RCT
89% for lovastatin and 90% for
colestipol with no telephone call
er, on lovastatin
and colestipol, with Absolute numbers not reported
a telephone in their
home
Not significant
Reported as not significant
P value not reported
Compliance (measured by pill
and packet counts) , 12 weeks
88% for lovastatin and 90% for
colestipol with telephone call
30 people with
CABG or PTCA in
the last 7 to 30
days, baseline
fasting LDL
130 mg/dL or high-
[28]
RCT
86% for lovastatin and 88% for
colestipol with no telephone call
er, on lovastatin
and colestipol, with Absolute numbers not reported
a telephone in their
home
prompting interven-
tion (telephone
call)
P <0.05Compliance (measured by
contacting pharmacies to ob-
tain document refill informa-
tion) , 1 year
30 people with
CABG or PTCA in
the last 7 to 30
days, baseline
fasting LDL
[28]
RCT
71% for lovastatin and 54% for
colestipol with telephone call
130 mg/dL or high-
er, on lovastatin
and colestipol, with 47% for lovastatin and 27% for
colestipol with no telephone call
a telephone in their
home Absolute numbers not reported
prompting interven-
tion (telephone
call)
P <0.05Compliance (measured by
contacting pharmacies to ob-
tain document refill informa-
tion) , 2 years
30 people with
CABG or PTCA in
the last 7 to 30
days, baseline
fasting LDL
[28]
RCT
63% for lovastatin and 48% for
colestipol with telephone call
130 mg/dL or high-
er, on lovastatin
and colestipol, with 39% for lovastatin and 23% for
colestipol with no telephone call
a telephone in their
home Absolute numbers not reported
© BMJ Publishing Group Ltd 2011. All rights reserved. ........................................................... 5
Cardiovascular medication: improving adherence
Cardiovascular disorders
Favours
Effect
size
Results and statistical
analysisOutcome, InterventionsPopulation
Ref
(type)
prompting interven-
tion (reminder
postcard)
P <0.05
Post hoc analysis
Mean compliant events, which
equalled the number of refills
divided by the possible number
of refills , outcome measured
for 3 months
311 people on car-
diovascular medica-
tions, attended pri-
mary care or spe-
ciality clinic at a
university health
[29]
RCT
3-armed
trial
0.58 with no reminder
centre, medication
refill due in 2 days, 0.65 with postcard reminder 2
working days before medication
refill due
people selected
from a computer
database
Third arm evaluat-
ed telephone re-
minder
prompting interven-
tion (telephone
call)
P <0.05
Post hoc analysis
Mean compliant events, which
equalled the number of refills
divided by the possible number
of refills , outcome measured
for 3 months
311 people on car-
diovascular medica-
tions, attended pri-
mary care or spe-
ciality clinic at a
university health
[29]
RCT
3-armed
trial
0.58 with no reminder
centre, medication
refill due in 2 days, 0.64 with telephone call 1 working
day before medication refill due
people selected
from a computer
database
Third arm evaluat-
ed postcard re-
minder
prompting interven-
tion
P = 0.0001 for either intervention
v usual care
Mean % compliance (compli-
ance assessed by counting
tablets; % compliance defined
636 people with
newly diagnosed or
uncontrolled mild
[30]
RCT No direct statistical comparison
of telephone intervention alone
as total number of consumed
tablets/total number of tablets
to moderate hyper-
tension, aged 18 to
3-armed
trial or mailed intervention alone ver-
sus usual care reported, but
that should have been con-
sumed x 100) , assessed at 5
80 years, on single
therapy higher rates of adherence in
prompting mechanism groups
clinic visits: inclusion visit, and
4 follow-up visits at 26, 52, 106,
and 155 days
90% with usual care
99% with telephone intervention
97% with mailed intervention
Absolute numbers not reported
prompting interven-
tion
P = 0.0001 for either intervention
v usual care
Proportion of compliers (partic-
ipants with 80–110% drug con-
sumption ) (compliance as-
636 people with
newly diagnosed or
uncontrolled mild
[30]
RCT Between-group analysis
sessed by counting tablets; %
compliance defined as total
to moderate hyper-
tension, aged 18 to
3-armed
trial No direct statistical comparison
of telephone intervention alone
number of consumed
tablets/total number of tablets
80 years, on single
therapy or mailed intervention alone ver-
sus usual care reported, but
that should have been con-
sumed x 100) , assessed at 5 higher rates of adherence in
prompting mechanism groups
clinic visits: inclusion visit, and
4 follow-up visits at 26, 52, 106,
and 155 days
69% with usual care
96% with telephone intervention
91% with mailed intervention
Absolute numbers not reported
-
Adverse effects
-
-
© BMJ Publishing Group Ltd 2011. All rights reserved. ........................................................... 6
Cardiovascular medication: improving adherence
Cardiovascular disorders
No data from the following reference on this outcome. [7] [27] [28] [29] [30]
-
-
Prompting mechanism plus usual care versus unit-of-use packaging plus usual care versus unit-of-use
packaging plus prompting mechanism plus usual care versus usual care alone:
We found 11 systematic reviews (search dates 1996; [16] 2000; [17] 2002; [18] [19] [20] 2003; [21] [22] 2004; [23] 2007;
[24] 2008; [25] 2009 [26] ), which identified two RCTs of sufficient quality. [6] The reviews did not pool data.The two
RCTs were undertaken by the same research group, and employed a similar methodology.
-
Adherence to medication
Prompting mechanism plus usual care compared with unit-of-use packaging plus usual care compared with unit-of-
use packaging plus prompting mechanism plus usual care compared with usual care alone A prompting intervention
(mailed reminder 10 days before refill date), unit-of-use packaging, and combined prompting intervention plus unit-
of-use packaging may all be more effective than usual care at improving adherence to medication in people with
mild to moderate hypertension; and the combined prompting intervention plus unit-of-use packaging may be more
effective than the prompting intervention alone or unit-of-use packaging alone. However, the unit-of-use packaging
intervention was not fully defined (a 30-day inventory tray in 1 RCT; not defined in another RCT), which makes
drawing conclusions on it difficult.We don't know whether a mailed prompting reminder is more effective than unit-
of-use packaging at improving adherence (very low-quality evidence).
Favours
Effect
size
Results and statistical
analysisOutcome, InterventionsPopulation
Ref
(type)
Adherence to medication
prompting interven-
tion (mailed re-
minder)
P <0.05Mean number of days' supply
of medication obtained over
360-day study period, ex-
pressed as "medication posses-
304 people, previ-
ously untreated
mild to moderate
hypertension, <65
[6]
RCT
4-armed
trial sion ratio" (defined as the
number of days' supply of
years old, on vera-
pamil once daily, medication obtained through-refill medication out the study period expresseddispensed in 30-
day supplies as a ratio against the number
of days that should have been
supplied)
The remaining
arms evaluated 0.64 with standard care plus
mailed reminder 10 days prior to
medication refill date
standard care plus
unit-of-use packag-
ing and standard
care plus mailed 0.56 with standard care
reminder plus unit-
of-use packaging
reminder packag-
ing (unit-of-use
packaging)
P <0.05
The unit-of-use packaging was
reported to be "a sequentially
Mean number of days' supply
of medication obtained over
360-day study period, ex-
pressed as "medication posses-
304 people, previ-
ously untreated
mild to moderate
hypertension, <65
[6]
RCT
4-armed
trial numbered 30-day supply invento-
ry tray with easy-access compart-
ments" but was not further de-
fined
sion ratio" (defined as the
number of days' supply of
medication obtained through-
out the study period expressed
as a ratio against the number
years old, on vera-
pamil once daily,
refill medication
dispensed in 30-
day supplies of days that should have been
supplied)
The remaining
arms evaluated 0.67 with standard care plus unit-
of-use packaging
standard care plus
mailed reminder 10
days prior to medi- 0.56 with standard care
cation refill date
and standard care
plus mailed re-
minder plus unit-of-
use packaging
prompting mecha-
nism (mailed re-
P <0.05
The unit-of-use packaging was
reported to be "a sequentially
Mean number of days' supply
of medication obtained over
360-day study period, ex-
pressed as "medication posses-
304 people, previ-
ously untreated
mild to moderate
hypertension, <65
[6]
RCT
4-armed
trial
minder) plus re-
minder packaging
(unit-of-use packag-
ing)
numbered 30-day supply invento-
ry tray with easy-access compart-
© BMJ Publishing Group Ltd 2011. All rights reserved. ........................................................... 7
Cardiovascular medication: improving adherence
Cardiovascular disorders
Favours
Effect
size
Results and statistical
analysisOutcome, InterventionsPopulation
Ref
(type)
ments" but was not further de-
fined
sion ratio" (defined as the
number of days' supply of
medication obtained through-
years old, on vera-
pamil once daily,
refill medication out the study period expresseddispensed in 30-
day supplies as a ratio against the number
of days that should have been
supplied)
The remaining
arms evaluated 0.56 with standard carestandard care plus
mailed reminder 10 0.79 with standard care plus
mailed reminder plus unit-of-use
packaging
days prior to medi-
cation refill date
and standard care
plus unit-of-use
packaging
Not significant
Reported as not significant
P value not reported
Mean number of days' supply
of medication obtained over
360-day study period, ex-
pressed as "medication posses-
304 people, previ-
ously untreated
mild to moderate
hypertension, <65
[6]
RCT
4-armed
trial The unit-of-use packaging was
reported to be "a sequentially
numbered 30-day supply invento-
sion ratio" (defined as the
number of days' supply of
medication obtained through-
years old, on vera-
pamil once daily,
refill medication ry tray with easy-access compart-
out the study period expresseddispensed in 30-
day supplies ments" but was not further de-
fined
as a ratio against the number
of days that should have been
supplied)
The remaining
arms evaluated 0.64 with standard care plus
mailed reminder 10 days prior to
medication refill date
standard care
alone and standard
care plus mailed
reminder plus unit-
of-use packaging 0.67 with standard care plus unit-
of-use packaging
prompting mecha-
nism (mailed re-
P <0.05
The unit-of-use packaging was
reported to be "a sequentially
Mean number of days' supply
of medication obtained over
360-day study period, ex-
pressed as "medication posses-
304 people, previ-
ously untreated
mild to moderate
hypertension, <65
[6]
RCT
4-armed
trial
minder) plus re-
minder packaging
(unit-of-use-packag-
ing)
numbered 30-day supply invento-
ry tray with easy-access compart-
ments" but was not further de-
fined
sion ratio" (defined as the
number of days' supply of
medication obtained through-
out the study period expressed
as a ratio against the number
years old, on vera-
pamil once daily,
refill medication
dispensed in 30-
day supplies of days that should have been
supplied)
The remaining
arms evaluated 0.64 with standard care plus
mailed reminder 10 days prior to
medication refill date
standard care
alone and standard
care plus unit-of-
use packaging 0.79 with standard care plus
mailed reminder plus unit-of-use
packaging
prompting mecha-
nism (mailed re-
P <0.05
The unit-of-use packaging was
reported to be "a sequentially
Mean number of days' supply
of medication obtained over
360-day study period, ex-
pressed as "medication posses-
304 people, previ-
ously untreated
mild to moderate
hypertension, <65
[6]
RCT
4-armed
trial
minder) plus re-
minder packaging
(unit-of-use-packag-
ing)
numbered 30-day supply invento-
ry tray with easy-access compart-
ments" but was not further de-
fined
sion ratio" (defined as the
number of days' supply of
medication obtained through-
out the study period expressed
as a ratio against the number
years old, on vera-
pamil once daily,
refill medication
dispensed in 30-
day supplies of days that should have been
supplied)
The remaining
arms evaluated 0.67 with standard care plus unit-
of-use packaging
standard care
alone and standard
care plus mailed 0.79 with standard care plus
mailed reminder plus unit-of-use
packaging
reminder 10 days
prior to medication
refill date
© BMJ Publishing Group Ltd 2011. All rights reserved. ........................................................... 8
Cardiovascular medication: improving adherence
Cardiovascular disorders
Favours
Effect
size
Results and statistical
analysisOutcome, InterventionsPopulation
Ref
(type)
prompting interven-
tion (mailed re-
minder)
P <0.05Mean number of days' supply
of medication obtained over
360-day study period, ex-
pressed as "medication posses-
128 people with
previously untreat-
ed mild to moder-
ate hypertension,
[31]
RCT
4-armed
trial sion ratio" (defined as the
number of days' supply of
on verapamil once
daily, mean age medication obtained through-approximately 54 out the study period expressedyears, refill medica- as a ratio against the numbertion dispensed in
30-day supplies of days that should have been
supplied)
The remaining
arms evaluated 0.71 with standard care plus
mailed reminder 10 days prior to
medication refill date
standard care plus
unit-of-use packag-
ing and standard 0.64 with standard care
care plus mailed
reminder plus unit-
of-use packaging
reminder packag-
ing (unit-of-use
packaging)
P <0.05
The unit-of-use packaging was
not further defined
Mean number of days' supply
of medication obtained over
360-day study period, ex-
pressed as "medication posses-
sion ratio" (defined as the
128 people with
previously untreat-
ed mild to moder-
ate hypertension,
on verapamil once
[31]
RCT
4-armed
trial number of days' supply ofdaily, mean age medication obtained through-approximately 54 out the study period expressedyears, refill medica- as a ratio against the numbertion dispensed in
30-day supplies of days that should have been
supplied)
The remaining
arms evaluated 0.64 with standard care
standard care plus 0.75 with standard care plus unit-
of-use packaging
mailed reminder 10
days prior to medi-
cation refill date
and standard care
plus mailed re-
minder plus unit-of-
use packaging
prompting mecha-
nism (mailed re-
P <0.05
The unit-of-use packaging was
not further defined
Mean number of days' supply
of medication obtained over
360-day study period, ex-
pressed as "medication posses-
sion ratio" (defined as the
128 people with
previously untreat-
ed mild to moder-
ate hypertension,
on verapamil once
[31]
RCT
4-armed
trial
minder) plus re-
minder packaging
(unit-of-use-packag-
ing)
number of days' supply of
medication obtained through-
daily, mean age
approximately 54 out the study period expressedyears, refill medica- as a ratio against the numbertion dispensed in
30-day supplies of days that should have been
supplied)
The remaining
arms evaluated 0.87 with standard care plus
mailed reminder plus unit-of-use
packaging
standard care plus
mailed reminder 10
days prior to medi- 0.64 with standard care
cation refill date
and standard care
plus unit-of-use
packaging
Not significant
Reported as not significant
P value not reported
Mean number of days' supply
of medication obtained over
360-day study period, ex-
pressed as "medication posses-
128 people with
previously untreat-
ed mild to moder-
ate hypertension,
[31]
RCT
4-armed
trial The unit-of-use packaging was
not further defined
sion ratio" (defined as the
number of days' supply of
on verapamil once
daily, mean age medication obtained through-approximately 54 out the study period expressedyears, refill medica- as a ratio against the numbertion dispensed in
30-day supplies of days that should have been
supplied)
The remaining
arms evaluated
© BMJ Publishing Group Ltd 2011. All rights reserved. ........................................................... 9
Cardiovascular medication: improving adherence
Cardiovascular disorders
Favours
Effect
size
Results and statistical
analysisOutcome, InterventionsPopulation
Ref
(type)
0.71 with standard care plus
mailed reminder 10 days prior to
medication refill date
standard care
alone and standard
care plus mailed
reminder plus unit-
of-use packaging 0.75 with standard care plus unit-
of-use packaging
prompting mecha-
nism (mailed re-
P <0.05
The unit-of-use packaging was
not further defined
Mean number of days' supply
of medication obtained over
360-day study period, ex-
pressed as "medication posses-
sion ratio" (defined as the
128 people with
previously untreat-
ed mild to moder-
ate hypertension,
on verapamil once
[31]
RCT
4-armed
trial
minder) plus re-
minder packaging
(unit-of-use-packag-
ing)
number of days' supply of
medication obtained through-
daily, mean age
approximately 54 out the study period expressedyears, refill medica- as a ratio against the numbertion dispensed in
30-day supplies of days that should have been
supplied)
The remaining
arms evaluated 0.71 with standard care plus
mailed reminder 10 days prior to
medication refill date
standard care
alone and standard
care plus unit-of-
use packaging 0.87 with standard care plus
mailed reminder plus unit-of-use
packaging
prompting mecha-
nism (mailed re-
P <0.05
The unit-of-use packaging was
not further defined
Mean number of days' supply
of medication obtained over
360-day study period, ex-
pressed as "medication posses-
sion ratio" (defined as the
128 people with
previously untreat-
ed mild to moder-
ate hypertension,
on verapamil once
[31]
RCT
4-armed
trial
minder) plus re-
minder packaging
(unit-of-use-packag-
ing)
number of days' supply of
medication obtained through-
daily, mean age
approximately 54 out the study period expressedyears, refill medica- as a ratio against the numbertion dispensed in
30-day supplies of days that should have been
supplied)
The remaining
arms evaluated 0.75 with standard care plus unit-
of-use packagingstandard care
alone and standard 0.87 with standard care plus
mailed reminder plus unit-of-use
packaging
care plus mailed
reminder 10 days
prior to medication
refill date
-
Adverse effects
-
-
No data from the following reference on this outcome. [31] [6]
-
-
Prompting mechanism plus patient health education versus usual care:
We found 11 systematic reviews (search dates 1996; [16] 2000; [17] 2002; [18] [19] [20] 2003; [21] [22] 2004; [23] 2007;
[24] 2008; [25] 2009 [26] ), which identified one RCT of sufficient quality. [32] See further information on studies for full
details on interventions.
-
Adherence to medication
Compared with usual care A prompting intervention (including a telephone call and mailed reminder) plus patient
health education (including an educational programme, newsletter, and general health advice) may be more effective
than usual care at improving adherence to medication in people with newly diagnosed hypertension and in people
with existing hypertension at 1 year (low-quality evidence).
© BMJ Publishing Group Ltd 2011. All rights reserved. .......................................................... 10
Cardiovascular medication: improving adherence
Cardiovascular disorders
Favours
Effect
size
Results and statistical
analysisOutcome, InterventionsPopulation
Ref
(type)
Adherence to medication
prompting interven-
tion plus patient
health education
P <0.05Results expressed as medica-
tion possession ratio (defined
as mean number of days' sup-
ply of medication obtained
453 outpatients,
mild to moderate
hypertension, on
once-daily atenolol,
[32]
RCT
over 360-day trial period) , fol-
low-up at 6 months
either new cases
or existing (previ-
ously treated); 0.82 with prompting intervention
plus patient health education
subgroup analysis
of 344 people with
existing hyperten-
sion 0.48 with usual care
Subgroup analysis
prompting interven-
tion plus patient
health education
P <0.05Results expressed as medica-
tion possession ratio (defined
as mean number of days' sup-
ply of medication obtained
453 outpatients,
mild to moderate
hypertension, on
once-daily atenolol,
[32]
RCT
over 360-day study period) ,
follow-up at 6 months
either new cases
or existing (previ-
ously treated); 0.93 with prompting intervention
plus patient health education
subgroup analysis
of 109 people with
new hypertension 0.52 with usual care
Subgroup analysis
-
Adverse effects
-
-
No data from the following reference on this outcome. [32]
-
-
Prompting mechanisms versus prescriber education, patient health education, or simplified dosing:
We found 11 systematic reviews (search dates 1996; [16] 2000; [17] 2002; [18] [19] [20] 2003; [21] [22] 2004; [23] 2007;
[24] 2008; [25] 2009 [26] ), which identified no RCTs of sufficient quality.
-
-
-
Further information on studies
[7] Prompting mechanism Daily calls, which lasted 3 to 5 minutes and were made by research assistant on
Monday to Friday. RCT methods Method of randomisation was described. Results based on 50/60 (83%) of
those randomised.Withdrawals in each individual group not reported. Participants offered $20 to take part in
study. Electronic caps placed on maximum of 4 medication bottles for each person. No significant difference in
adherence reported between telephone and video-telephone groups (reported as not significant; P value not
reported).
[27] Prompting mechanism Electronic cap on medication bottles: digital timepiece displayed when last opened,
alarm bleeped when dose due, flashed when dose missed. RCT methods Participants blinded, investigators
not blinded. Method of randomisation not described. Loss to follow-up not reported. Blood pressure results were
measured but presented as baseline analysis; no between-group analyses reported. Factoral design: only first
randomisation reported here.
[28] Prompting mechanism Telephone calls: same pharmacist telephoned people in their homes every week for
12 weeks. Standard set of questions, with emphasis on the importance of therapy, and asking reasons for non-
compliance where appropriate. RCT methodsThe method of randomisation was described, and follow-up was
100%. Different measures of adherence in short term (up to 12 weeks) and long term (up to 2 years). Small
RCT (15 people in each group). Changes in total cholesterol, LDL, HDL, and triglyceride level were not signifi-
cantly different between groups at 6 or 12 weeks. Compared with the no-telephone group, the telephone inter-
© BMJ Publishing Group Ltd 2011. All rights reserved. .......................................................... 11
Cardiovascular medication: improving adherence
Cardiovascular disorders
vention significantly reduced total cholesterol (P = 0.03), LDL (P = 0.02), and triglyceride levels (P = 0.04) at 1
and 2 years.
[29] RCT methods Method of randomisation not described. Level of blinding not reported. A telephone call was also
made to people in all three groups who were 3 days late obtaining the medication.This was to determine: if
postcard group had received postcard; if medication obtained at different pharmacy; and reasons for not refilling.
Calls made to all groups (including control) may have influenced the results. Of 40/311 (13%) total withdrawals,
35 were in group (1).These people were excluded from analysis as not contacted by telephone (unlisted or
telephone disconnected). Hence, withdrawals varied between groups.The RCT found no significant difference
between postcard and telephone groups in mean compliant events (P value not reported).
[30] Prompting mechanism Telephone intervention: three calls in total by nurses after scheduled visits to reinforce
compliance, standard call, with good compliance praised. Mailed intervention: three mailed communications
reinforcing compliance, health education, and reminding people of clinic visits.RCT methods Method of ran-
domisation was described. Results based on follow-up of 538/636 (86%) people. Significantly superior control
of blood pressure with telephone intervention compared with usual care (63% with telephone intervention v
47% with usual care; P <0.05).
[6] RCT methods Follow-up of 100%. Method of randomisation not described. Level of blinding not reported. "Unit-
of-use" packaging was reported to be "a sequentially numbered 30-day supply inventory tray with easy-access
compartments". It was not further described.
[32] Prompting mechanism Active intervention consisted of health education (educational programme, newsletter
discussing importance of compliance, nutrition, and lifestyle advice) plus prompting intervention (telephone
conversation 1 week prior to next medication refill initially, and then mailed reminder 10 days prior to refill each
month). RCT methods Method of randomisation not described. Level of blinding not described.
-
-
Comment: Clinical guide:
As outlined above, there is a variety of potential prompting mechanisms, from the simple and rela-
tively low-cost mailed reminder to the more expensive and labour-intensive use of telephone calls,
video-telephone calls, or electronic medication-reminder caps (and we have included RCTs that
assessed any form).There is a small amount of evidence for effect with all of the above mechanisms,
but several (e.g., daily telephone calls, installing videophones) remain impracticable for use in
routine clinical practice.
OPTION SIMPLIFIED DOSING...........................................................
•For GRADE evaluation of interventions for Cardiovascular medication: improving adherence, see table, p 31 .
•We found evidence that simplified dosing regimens may increase adherence compared with more complex reg-
imens.
•While simplifying the frequency of dosage may increase adherence, it is not known whether simplified regimens
may increase adherence when someone is taking multiple drugs, as may be the case with cardiovascular
medicines.
•In altering a drug regimen simply to increase adherence, any changes could potentially affect the effectiveness
of the treatment, and could also potentially increase adverse effects.
Benefits and harms
Simplified dosing regimens versus more complex regimens:
We found 10 systematic reviews (search dates 1996; [16] 2000; [17] 2002; [18] [19] 2003; [21] [33] [34] 2004; [23] 2007;
[24] 2008 [25] ), which identified 6 RCTs of sufficient quality. [35] [36] [37] [38] [39] [40] The reviews did not pool data.
We found no subsequent RCTs.
-
Adherence to medication
Compared with more complex dosing regimens Simplified dosing regimens may be more effective than more complex
dosing regimens (e.g., once-daily regimens compared with twice-daily regimens, or twice-daily regimens compared
with 3-times-daily regimens) at increasing adherence to medication in people with hypertension, hyperlipidaemia,
and angina (very low-quality evidence).
© BMJ Publishing Group Ltd 2011. All rights reserved. .......................................................... 12
Cardiovascular medication: improving adherence
Cardiovascular disorders
Favours
Effect
size
Results and statistical
analysisOutcome, InterventionsPopulation
Ref
(type)
Adherence to medication
simplified dosing
P = 0.009Compliance (defined as those
people taking at least 90% of
their medications) , assess-
389 people with
mild or moderate
hypertension,
[35]
RCT
ment by pill count at 6 and 10
weeks during clinic visit
mean age 53 to 54
years, adequately
controlled, on 93% of people with once-daily
regimen of metoprolol (slow re-
lease)
metoprolol or pro-
pranolol either as
monotherapy or in
conjunction with di- 81.5% of people with twice-daily
regimen of metoprolol
uretic. Participants
continued their Absolute numbers not reported
other medication
as normal 2 weeks of baseline monitoring
on original medication, then 8
weeks of intervention
simplified dosing
P = 0.0089Tablet count compliance over
test period, expressed as mean
rank , assessment by pill count
389 people with
mild or moderate
hypertension,
[35]
RCT
at 6 and 10 weeks during clinic
visit
mean age 53 to 54
years, adequately
controlled, on 123.38 with once-daily regimen
of metoprolol (slow release)
metoprolol or pro-
pranolol either as
monotherapy or in 100.92 with twice-daily regimen
of metoprolol
conjunction with di-
uretic. Participants Absolute numbers not reported
continued their
other medication
as normal 2 weeks of baseline monitoring
on original medication, then 8
weeks of intervention
simplified dosing
RCT reported that "all the differ-
ences were statistically signifi-
cant, P <0.001"
Self-reported — participants
asked to rate their compliance
(results presented as self-re-
ported compliance of 100%,
7274 people with
hypertension, suit-
able for treatment
with nicardipine,
[36]
RCT
80%, or 0% to 60%) , adherenceage 18 years and assessed at 3 months by stan-
dardised interview
older, mean age
50 years, 60% on
current treatment. 82%, 15%, and 3% with
nicardipine twice daily (slow re-
lease)
Other concomitant
antihypertensive
therapies allowed 71%, 24%, and 4% with
nicardipine 3 times daily
Absolute numbers not reported
simplified dosing
P = 0.01Compliance calculated using
computer that accounted for
drug supplies given, the recom-
29 men, partici-
pants in earlier
study, mean age
[37]
RCT
mended dosage, and a count49 years, on niacin
Crossover
design of returned medication; ex-
pressed as % of dose recom-
mended
4 times daily plus
lovastatin twice
daily plus colestipol
twice daily for 1
Intervention
continued
for 8 96% with niacin twice daily
year, adjusted to
maintain target
months,
then groups 85% with niacin 4 times daily
cholesterol 150 to
were Absolute numbers not reported
175 mg/dL, high
risk of cardiac
crossed
over for fur- events (elevated
ther 8
months apoprotein B or
stenosis or strong
family history).
Other medication
(lovastatin and
colestipol) contin-
ued as before
© BMJ Publishing Group Ltd 2011. All rights reserved. .......................................................... 13
Cardiovascular medication: improving adherence
Cardiovascular disorders
Favours
Effect
size
Results and statistical
analysisOutcome, InterventionsPopulation
Ref
(type)
Significance not assessedManual pill count (tablets con-
sumed, % of correct number) ,
follow-up at 12 weeks
31 people with sta-
ble angina, mean
age 63 to 64 years
(range 47–74
years)
[38]
RCT P value not reported
98% with isosorbide mononitrate
once daily
98% with isosorbide mononitrate
twice daily
Absolute numbers not reported
simplified dosing
P <0.05Compliance assessed by elec-
tronic bottle cap that measured
date and time bottle opened
31 people with sta-
ble angina, mean
age 63 to 64 years
[38]
RCT
(outcome expressed as % of(range 47–74
years) days with correct number of
openings) , follow-up at 12
weeks
97% with isosorbide mononitrate
once daily
88% with isosorbide mononitrate
twice daily
Absolute numbers not reported
simplified dosing
P <0.05Compliance assessed by elec-
tronic bottle cap that measured
date and time bottle opened
31 people with sta-
ble angina, mean
age 63 to 64 years
[38]
RCT
(outcome expressed as the %(range 47–74
years) of intervals between openings
within the correct time range)
, follow-up at 12 weeks
88% with isosorbide mononitrate
once daily
69% with isosorbide mononitrate
twice daily
Absolute numbers not reported
simplified dosing
P <0.01Compliance assessed by pill
count (outcome expressed as
% of doses taken) , follow-up
at 8 weeks (after crossover)
27 people, mild hy-
pertension, well
controlled on
monotherapy,
mean age 62 years
[39]
RCT
Crossover
design with enalapril once daily
Given initial
treatment with enalapril twice daily
for 4 weeks, Absolute results not reported
then groups
crossed
over
simplified dosing
P <0.001Compliance assessed by elec-
tronic bottle cap (outcome ex-
pressed as % of doses taken
27 people, mild hy-
pertension, well
controlled on
[39]
RCT
by electronic count) , follow-up
at 8 weeks (after crossover)
monotherapy,
mean age 62 years
Crossover
design
with enalapril once daily
Given initial
treatment with enalapril twice daily
for 4 weeks,
then groups
crossed
over
simplified dosing
P <0.001Outcome expressed as % of
days with correct number of
doses taken , follow-up at 8
weeks (after crossover)
27 people, mild hy-
pertension, well
controlled on
monotherapy,
mean age 62 years
[39]
RCT
Crossover
design with enalapril once daily
Given initial
treatment with enalapril twice daily
for 4 weeks,
© BMJ Publishing Group Ltd 2011. All rights reserved. .......................................................... 14
Cardiovascular medication: improving adherence
Cardiovascular disorders
Favours
Effect
size
Results and statistical
analysisOutcome, InterventionsPopulation
Ref
(type)
then groups
crossed
over
Not significant
Differential mean compliance was
+1.1%
Mean compliance measured by
an electronic monitoring sys-
tem (MEMS) , 5 months
405 people with
chronic heart fail-
ure and left ventric-
ular dysfunction
[40]
RCT
3-armed
trial
95% CI –4.4% to +6.6%
P = 0.62
89% with carvedilol twice daily
(immediate release formulation)The remaining arm
evaluated 88% with carvedilol once daily
(controlled release formulation)
carvedilol (con-
trolled release for-
mulation) taken in Absolute numbers not reported
the morning and
placebo taken in
the afternoon
-
Adverse effects
-
Favours
Effect
size
Results and statistical
analysisOutcome, InterventionsPopulation
Ref
(type)
Adverse effects
simplified dosing
P = 0.043Pruritus
with nicardipine twice daily (slow
release)
7274 people with
hypertension, suit-
able for treatment
with nicardipine,
age 18 years and
[36]
RCT
with nicardipine 3 times daily
older, mean age Absolute results not reported
50 years, 60% on
current treatment.
Other concomitant
antihypertensive
therapies allowed
simplified dosing
P = 0.033Palpitations
with nicardipine twice daily (slow
release)
7274 people with
hypertension, suit-
able for treatment
with nicardipine,
age 18 years and
[36]
RCT
with nicardipine 3 times daily
older, mean age Absolute results not reported
50 years, 60% on
current treatment.
Other concomitant
antihypertensive
therapies allowed
simplified dosing
P = 0.004People with at least 1 adverse
event
7274 people with
hypertension, suit-
able for treatment
[36]
RCT with nicardipine twice daily (slow
release)
with nicardipine,
age 18 years and
older, mean age with nicardipine 3 times daily
50 years, 60% on Absolute results not reported
current treatment.
Other concomitant
antihypertensive
therapies allowed
simplified dosing
P = 0.024Hot flushes
with nicardipine twice daily (slow
release)
7274 people with
hypertension, suit-
able for treatment
with nicardipine,
age 18 years and
[36]
RCT
with nicardipine 3 times daily
older, mean age Absolute results not reported
50 years, 60% on
current treatment.
Other concomitant
© BMJ Publishing Group Ltd 2011. All rights reserved. .......................................................... 15
Cardiovascular medication: improving adherence
Cardiovascular disorders
Favours
Effect
size
Results and statistical
analysisOutcome, InterventionsPopulation
Ref
(type)
antihypertensive
therapies allowed
complex dosing
P <0.005Flushing
14 people with niacin twice daily
29 men, partici-
pants in earlier
study, mean age
49 years, on niacin
[37]
RCT
Crossover
design 6 people with niacin 4 times daily
Absolute numbers not reported
4 times daily plus
lovastatin twice
daily plus colestipol
Intervention
continued twice daily for 1
for 8 year, adjusted to
months, maintain target
then groups cholesterol
were 150 mg/dL to
crossed 175 mg/dL, high
over for fur- risk of cardiac
ther 8
months events (elevated
apoprotein B or
stenosis or strong
family history).
Other medication
(lovastatin and
colestipol) contin-
ued as before
-
No data from the following reference on this outcome. [35] [38] [39] [40]
-
-
Simplified dosing versus patient health education, prompting mechanisms, prescriber education, or reminder
packaging:
We found 10 systematic reviews (search dates 1996; [16] 2000; [17] 2002; [18] [19] 2003; [21] [33] [34] 2004; [23] 2007;
[24] 2008 [25] ), which identified no RCTs of sufficient quality.The reviews did not pool data. We found no subsequent
RCTs.
-
-
-
Further information on studies
[35] Method of randomisation not described. Level of blinding not reported. Loss to follow-up 50/389 (13%). No
significant difference between groups in blood pressure control measured in clinic (P value not reported). Sub-
group also participated in home blood pressure monitoring; these results not presented here.The RCT found
no significant difference in adverse effects between once-daily and twice-daily metoprolol.
[36] Method of randomisation was described. Open RCT. Results based on 6813/7274 (94%) of those randomised.
The treating physicians' estimates of participants' compliance were consistent with the participants' estimates.
Adherence measured by self-reporting. Acceptability of twice-daily treatment was rated significantly higher by
participants compared with three-times-daily (P <0.001). No significant difference between groups in blood
pressure control (P = 0.185).
[37] Method of randomisation not described. Level of blinding not reported. Crossover RCT — results should be in-
terpreted with caution.
[38] Method of randomisation not described. Participants were aware that cap was recording opening of bottle.
Follow-up 29/31 (94%). Small RCT. No significant differences between groups in number of angina attacks or
mean number of rescue GTN tablets taken (P value not reported).The RCT reported that none of the differences
in tolerability were significantly different between once- and twice-daily isosorbide mononitrate (absolute numbers
and P value not reported).
[39] Randomisation method not described. Evaluation blinded. Follow-up 25/27 (93%). All groups received home
visits every 2 weeks for duration of study.Third 4-week treatment period incorporated into study design to detect
carryover effects, as no wash-out period between treatments. Crossover RCT — results should be interpreted
© BMJ Publishing Group Ltd 2011. All rights reserved. .......................................................... 16
Cardiovascular medication: improving adherence
Cardiovascular disorders
with caution. No significant difference between groups in blood pressure measurements, although differences
approached significance in favour of the twice-daily regimen.
[40] The three-arm randomisation format of this trial was designed to evaluate the effect of a twice-daily versus
once-daily formulation of carvedilol in a double-blinded manner as well as to evaluate the two dosing regimens
in a real-world effectiveness format (twice-daily carvedilol IR compared with the open-label arm of once-daily
controlled release carvedilol CR). Randomisation method not described. Follow-up 401/405 (99%).There were
also no significant differences in quality of life, treatment satisfaction, or physiological measures among the
study arms.This review only analyses the treatment arms of immediate release versus controlled release, as
these are the comparisons relevant for assessing adherence.
-
-
Comment: In this option we have included any RCTs that compared any form of simplified dosing (i.e., a re-
duction in the number of tablets taken daily). All included RCTs compared different regimens of
the same drug.We excluded RCTs that compared different drugs in each arm (e.g., one drug once
daily v a different drug twice daily), as the different drugs in each arm may have different accept-
abilities, which may affect adherence in each arm, making interpretation of adherence between
groups difficult.
Fixed-dose combinations: We found one systematic review (search date 2008, 15 studies [5
RCTs, 4 CCTs, 6 retrospective cohort studies], 32,331 people), [41] which was a meta-analysis of
the compliance, safety, and effectiveness of fixed-dose combinations (FDCs) of antihypertensive
agents.The review did not present results separately from RCTs for our outcome of interest (ad-
herence). It found that the use of FDCs significantly improved compliance compared with individual
drugs given as separate tablets. It found no significant difference in persistence with therapy, systolic
or diastolic blood pressure, or adverse effects between groups.
Clinical guide:
The relatively limited evidence (1 meta-analysis, small number of RCTs and short follow-up period)
supports the strategy of simplifying the dosage of medication when prescribing cardiovascular
medicines.Whether simplification of dosage influences adherence when a patient is taking multiple
drugs (the usual situation in secondary prevention of CHD) is not known. Similarly, the trade-off
between simplification of dosage to enhance adherence balanced against the risk of altered phar-
macodynamics and pharmacokinetics, particularly in older patients at risk of adverse drug reactions,
is also unknown. Nevertheless, provision of simple, clear instructions alongside simplification of
the dosage regimen seems a sensible and pragmatic strategy. [1]
OPTION PRESCRIBER EDUCATION .....................................................
•For GRADE evaluation of interventions for Cardiovascular medication: improving adherence, see table, p 31 .
•We found one RCT of prescriber education in a developing country which showed that a 1-day intensive training
session of general practitioners on hypertension improved medication adherence compared with usual care but
these data are not generalisable to the range of people taking cardiovascular medication so we cannot draw firm
conclusions about this intervention.
Benefits and harms
Prescriber education versus usual care:
We found 8 systematic reviews (search dates 1996; [16] 2000; [17] 2002; [18] [19] 2003; [21] 2004; [23] 2007; [24] 2008
[25] ), which identified one RCT of sufficient quality. [43] We found no subsequent RCTs.
-
Adherence to medication
Compared with usual care Prescriber education may be more effective than usual care at improving adherence in
people in developing countries who are taking antihypertensive medication (low-quality evidence).
Favours
Effect
size
Results and statistical
analysisOutcome, InterventionsPopulation
Ref
(type)
Prescriber education
prescriber educa-
tion
P = 0.048Percentage of days on which
correct dose of medication was
taken, measured by an electron-
200 people with
hypertension and
taking antihyperten-
[43]
RCT
ic measuring system (MEMS) ,
6 weeks
sive medications
© BMJ Publishing Group Ltd 2011. All rights reserved. .......................................................... 17
Cardiovascular medication: improving adherence
Cardiovascular disorders
Favours
Effect
size
Results and statistical
analysisOutcome, InterventionsPopulation
Ref
(type)
(see further infor-
mation on studies) 48% with prescriber education
32% with usual care
Absolute numbers not reported
-
Adverse effects
-
-
No data from the following reference on this outcome. [43]
-
-
Prescriber education versus prompting mechanisms, patient health education, simplified dosing, or reminder
packaging:
We found 8 systematic reviews (search dates 1996; [16] 2000; [17] 2002; [18] [19] 2003; [21] 2004; [23] 2007; [24] 2008
[25] ), which found no RCTs of sufficient quality.We found no subsequent RCTs.
-
-
-
Further information on studies
[43] This cluster randomised controlled trial sought to determine the impact of a simple educational package for
general practitioners on adherence to antihypertensive drugs in a developing world setting. Six randomly selected
communities in Karachi, Pakistan, from which 200 people with hypertension taking antihypertensive drugs and
being treated by 78 general practitioners, were randomised. Method of randomisation was described as a
multi-stage cluster random sampling technique using computer-generated codes.InterventionThe intervention
was a 1-day intensive training session on hypertension, which focused on standard treatment algorithms for
the management of hypertension. Of the 200 people who were enrolled, 178 (89%) successfully completed 6
weeks of follow-up.
-
-
Comment: In this option, "prescriber education" refers to a prescribing clinician who has received a directed
intervention (educational) aimed at improving medication adherence in people to whom he or she
has prescribed; this is compared with adherence achieved by another clinician of a similar overall
training level, but who has not received the educational intervention.
Clinical guide:
Prescribers play a key initial role in the process of adherence to medication within the setting of a
therapeutic alliance between clinician and patient. However, they still remain removed from the
process of medication-taking, which is the ultimate determinant of adherence. Despite this, educa-
tional interventions can be directed at prescribers with the aim of improving adherence, but studies
of sufficient methodological rigour have yet to be carried out.
OPTION REMINDER PACKAGING (CALENDAR [BLISTER] PACKS; MULTI-DOSE PILL BOXES).....
•For GRADE evaluation of interventions for Cardiovascular medication: improving adherence, see table, p 31 .
•We found no evidence from one RCT that reminder packaging (a calendar blister pack) was effective, and found
insufficient evidence on other types of reminder packaging such as multi-dose pill boxes. Reminder packaging
using a calendar blister pack seems effective, but we don't know whether other types of reminder packaging,
such as multi-dose pill boxes, improve adherence.
© BMJ Publishing Group Ltd 2011. All rights reserved. .......................................................... 18
Cardiovascular medication: improving adherence
Cardiovascular disorders
Benefits and harms
Calendar (blister) pack versus usual care:
We found 11 systematic reviews (search dates 1996; [16] 2000; [17] 2002; [18] [19] 2003; [21] [34] 2004; [23] [44] 2007;
[24] 2008; [25] 2009 [26] ), which identified two RCTs of sufficient quality. [45] [46]
-
Adherence to medication
Compared with usual care We don't know whether packaging medication in calendar (blister) packs is more effective
than packaging medication in traditional (usual) vials at improving adherence to medication in people with hypertension
(very low-quality evidence).
Favours
Effect
size
Results and statistical
analysisOutcome, InterventionsPopulation
Ref
(type)
Adherence to medication
Not significant
Reported as not significant
P value not reported
People taking >80% of pills
(self-reported) , at 3 months
56% with special packaging
180 people aged
20 to 80 years with
elevated diastolic
blood pressure
>90 mmHg on at
[45]
RCT
54% with traditional (usual) pill
vials
least 1 visit in the
2 years prior to the Absolute numbers not reported
study; recruited
from people receiv- The special packaging was a
commercially available system
ing care at a com-
munity hospital- comprising 28 doses of medica-
based family
medicine practice tion. Each pill was enclosed in a
single plastic blister sealed with
foil, on which was printed the day
of week and time of day the
medication was due
Not significant
Reported as not significant
P value not reported
People taking >80% of pills
(measured by pill count) , at 3
months
180 people aged
20 to 80 years with
elevated diastolic
blood pressure
[45]
RCT
84% with special packaging
>90 mmHg on at
least 1 visit in the 75% with traditional (usual) pill
vials
2 years prior to the
study; recruited Absolute numbers not reported
from people receiv-
ing care at a com- 158/180 (88%) of people anal-
ysed
munity hospital-
based family
medicine practice The special packaging was a
commercially available system
comprising 28 doses of medica-
tion. Each pill was enclosed in a
single plastic blister sealed with
foil, on which was printed the day
of week and time of day the
medication was due
daily dose blister
packaging
P = 0.012Prescription refill regularity ,
12 months
85 people 65 years
of age or older with
hypertension in the
US
[46]
RCT 80% with daily dose blister pack-
aging
66% with traditional pill bottles
Absolute numbers not reported
daily dose blister
packaging
P = 0.039Medication possession ratio ,
12 months
85 people 65 years
of age or older with
hypertension in the
US
[46]
RCT 0.93 with daily dose blister pack-
aging
0.87 with traditional pill bottles
Absolute numbers not reported
-
© BMJ Publishing Group Ltd 2011. All rights reserved. .......................................................... 19
Cardiovascular medication: improving adherence
Cardiovascular disorders
Adverse effects
-
-
No data from the following reference on this outcome. [45] [46]
-
-
Calendar (blister) packs versus simplified dosing, patient health education, prompting mechanisms, prescriber
education, or multi-dose pill boxes:
We found 12 systematic reviews (search dates 1996; [16] 2000; [17] 2002; [18] [19] 2003; [21] [34] 2004; [23] [44] 2007;
[24] 2008; [25] [46] 2009 [26] ), which identified no RCTs of sufficient quality.
-
-
Multi-dose pill box (unit-of-use packaging) plus usual care versus usual care alone versus prompting
mechanism plus usual care versus multi-dose pill box (unit-of-use packaging) plus prompting mechanism
plus usual care:
See option on prompting mechanisms, p 4 .
-
-
Multi-dose pill boxes versus calendar (blister) packs, simplified dosing, patient health education, or prescriber
education:
We found 12 systematic reviews (search dates 1996; [16] 2000; [17] 2002; [18] [19] 2003; [21] [34] 2004; [23] [44] 2007;
[24] 2008; [25] [46] 2009 [26] ), which identified no RCTs of sufficient quality.
-
-
-
Further information on studies
[45] The RCT also found no significant difference between groups in average diastolic blood pressures at 3 months
(165 people assessed). Method of randomisation not described. Physicians treating people were blinded to the
study group; it was not reported whether assessment was blinded. Loss to follow-up at first follow-up visit was
15/180 (8%). In contrast to previously reported work, this RCT did not demonstrate any significant improvement
in compliance with special packaging of antihypertensive medications.
[46] The RCT found that patients using daily-dose blister packaging had lower diastolic blood pressure (P = 0.01)
than patients who had their medications packaged in traditional bottles of loose tablets.Open study Method
of randomisation was described as randomisation logs provided by the university department of biostatistics.
No losses to follow-up reported.
-
-
Comment: In addition to the studies above, we found one open-label crossover RCT (784 people in Poland
with uncontrolled hypertension), which assessed the impact of an electronic reminder and monitoring
device compared with usual care over 12 months. [47] It found a significant difference in adherence
in favour of the device between groups at 6 months, but this difference diminished after crossover.
Adherence was assessed by use of a self-reporting questionnaire. Blood pressure was not affected.
Method of randomisation was not described. Losses to follow-up were substantial, at 50% (386/784),
and for this reason this study did not meet Clinical Evidence reporting inclusion criteria.
Clinical guide:
Reminder packaging now appears commonly in clinical practice as more and more drug companies
produce their medications in calendar packs, and the use of multi-dose pill boxes, especially for
older people, continues to rise. Although the benefit seems self-evident, there are few data published
on the subject.The only RCT of sufficient quality included here on calendar (blister) packs, which
compared medications dispensed in special packaging versus medications dispensed in traditional
pill vials, [45] found no significant difference between groups in adherence.
© BMJ Publishing Group Ltd 2011. All rights reserved. .......................................................... 20
Cardiovascular medication: improving adherence
Cardiovascular disorders
OPTION PATIENT HEALTH EDUCATION...................................................
•For GRADE evaluation of interventions for Cardiovascular medication: improving adherence, see table, p 31 .
•Patient health education may also increase adherence to medication.
•Adherence behaviour is complex.Traditional education methods may fail to address this. However, more patient-
centred approaches, particularly those that are nurse- or pharmacist-led, using video or telephone strategies,
may be beneficial and require further investigation.
•We found some RCT evidence that a combination of strategies, such as education plus prompting, may be more
successful than a single educational strategy.
Benefits and harms
Patient health education versus usual care:
We found 10 systematic reviews (search dates 1996; [16] 2000; [17] 2002; [18] [19] 2003; [21] [33] 2004; [23] 2007; [24]
2008; [25] 2009 [26] ), which between them identified 12 RCTs (34–2618 people) [48] [49] [50] [51] [52] [53] [54] [55]
[56] [57] [58] [59] including one extended follow-up report of one RCT, [60] and we found one additional RCT. [61] The
RCTs employed a number of educational interventions and assessed different measures of adherence; for full details
on interventions, see further information on studies. Some of the RCTs were of poor methodological quality, and
completeness of reporting varied widely between trials.
-
Adherence to medication
Compared with usual care Patient health education may be no more effective than usual care at improving adherence
in people taking cardiovascular medication. However, the education interventions used were diverse, and results
varied by the specific educational intervention employed (very low-quality evidence).
Favours
Effect
size
Results and statistical
analysisOutcome, InterventionsPopulation
Ref
(type)
Adherence to medication
Similar rates of adherence be-
tween groups, but differences not
tested statistically
Assessment by pill count
(compliance defined as the %
of medication prescribed that
was removed from the bottle;
230 male steelwork-
ers, with hyperten-
sion, not on current
treatment. Of 115
[48]
RCT
defined as "compliant" if com-in each group, 80 pliance pill count of 80% or
more) , 6 months follow-up
(70%) in education
group and 64
(56%) in no educa- 40/80 (50%) with educational in-
tervention
tion group received
drug treatment.
These 144 men
were analysed 36/64 (56%) with no health edu-
cation
Not significant
P >0.05Outcome (% of pills taken as-
sessed by pill count) , follow-
up at 4 months
110 people, mean
age 56 years,
started on lipid-
lowering medica-
[49]
RCT
Some re-
ported data 88% with educational intervention
tion (fluvastatin)
mainly for primary
prevention
from sys-
tematic re-
view as well
84% with usual care
Absolute numbers not reported
as original
report of
RCT
educational inter-
vention
P <0.002Adherence assessed by pill
count , 6 months follow-up
110 people, with
either newly diag-
nosed or estab-
[50]
RCT 93% with educational intervention
lished treated hy-
pertension, mean
age 59 years
Some re-
ported data
from sys-
tematic re-
69% with usual care
Absolute numbers not reported
view as well
as original
report of
RCT
© BMJ Publishing Group Ltd 2011. All rights reserved. .......................................................... 21
Cardiovascular medication: improving adherence
Cardiovascular disorders
Favours
Effect
size
Results and statistical
analysisOutcome, InterventionsPopulation
Ref
(type)
educational inter-
vention
P <0.001Adherence assessed by pill
count , 2 years' follow-up
110 people, with
either newly diag-
nosed or estab-
[60]
RCT 96% with educational intervention
lished treated hy-
pertension, mean
age 59 years
RCT [60] is
a 2-year fol-
low-up re-
port of previ-
56% with usual care
Absolute numbers not reported
ous RCT
[50]
Some re-
ported data
from sys-
tematic re-
view [18] as
well as origi-
nal report of
RCT
Not significant
RCT reported "no significant dif-
ference between groups on com-
pliance"
Self-reported compliance (sur-
vey conducted by nurse,
household medicated survey,
which included questions on
115 people attend-
ing a primary care
clinic, <70 years
old, with hyperten-
[51]
RCT
4-armed
trial drugs and a count of all hyper-
tensive medications; outcome
sion, living near to
clinic reported as absolute numbers
of "good", "fair", and "poor"
compliance)
8 people reported as "good", 13
"fair", 8 "poor" with educational
intervention
7, 13, 5 people with daily self-
monitoring of blood pressure
9, 15, 6 people with education
and self-monitoring of blood
pressure
7, 12, 10 people with control
Not significant
Reported as not significant
P value not reported
Self-reported compliance at
interview
91% with educational intervention
417 people with
hypertension, on
medication
RCT had 2 differ-
ent educational
[52]
RCT
3-armed
trial 90% with no education
groups versus con- Absolute numbers not reported
trol. Results for
both educational
groups combined
in analysis
Not significant
Reported as not significant
P value not reported
Analysis of number of tablets
prescribed by pharmacy
records
417 people with
hypertension, on
medication
[52]
RCT
3-armed
trial 69% with educational intervention
68% with no education
RCT had 2 differ-
ent educational
groups versus con-
trol. Results for Absolute numbers not reported
both educational
groups combined
in analysis
Not significant
P = 0.05Self-reported compliance as
assessed by interview with
questionnaire (score measured
34 people with hy-
pertension on
treatment, age 16
[53]
RCT
on 6-point scale, where 0 = noyears or older (av- adherence and 5 = all tabletserage age 51–56 taken) (possible total score of
30) , 6 months
years), at tertiary
care medical cen-
tre 27.53 with educational interven-
tion
24.46 with usual care
© BMJ Publishing Group Ltd 2011. All rights reserved. .......................................................... 22
Cardiovascular medication: improving adherence
Cardiovascular disorders
Favours
Effect
size
Results and statistical
analysisOutcome, InterventionsPopulation
Ref
(type)
educational inter-
vention
P = 0.003
Not clear on what basis the
physician’s assessment of adher-
ence was made
Physician's assessment of ad-
herence (score measured on
6-point scale, where 0 = no ad-
herence and 5 = all tablets tak-
en) (possible total score of 30)
, 6 months
34 people with hy-
pertension on
treatment, age 16
years or older (av-
erage age 51–56
years), at tertiary
care medical cen-
tre
[53]
RCT
29.18 with educational interven-
tion
23.9 with usual care
Not significant
Reported as not significant
P value not reported
Index of compliance was medi-
cation possession ratio (MPR),
calculated from pharmacy
records (MPR defined as the
Participants cho-
sen from database
by prescription
(rather than diagno-
[54]
RCT
number of days' supply ofsis), 410 people medication obtained through-taking benazepril, out the study period expressed1728 taking meto- as a ratio against the numberprolol, and 568 of days that should have been
supplied) , over 9 months
taking simvastatin,
mean age 55 years
(range 20–97 0.71 with educational intervention
years), with refill of
medication every
30 days 0.72 with usual care
Subgroup analysis
This analysis is of
410 people taking
benazepril
Not significant
Reported as not significant
P value not reported
Index of compliance was medi-
cation possession ratio (MPR),
calculated from pharmacy
records (MPR defined as the
Participants cho-
sen from database
by prescription
(rather than diagno-
[54]
RCT
number of days' supply ofsis), 410 people medication obtained through-taking benazepril, out the study period expressed1728 taking meto- as a ratio against the numberprolol, and 568 of days that should have been
supplied) , over 9 months
taking simvastatin,
mean age 55 years
(range 20–97 0.74 with educational intervention
years), with refill of
medication every
30 days 0.73 with usual care
Subgroup analysis
This analysis is of
1728 people taking
metoprolol
Not significant
Reported as not significant
P value not reported
Index of compliance was medi-
cation possession ratio (MPR),
calculated from pharmacy
records (MPR defined as the
Participants cho-
sen from database
by prescription
(rather than diagno-
[54]
RCT
number of days' supply ofsis), 410 people medication obtained through-taking benazepril, out the study period expressed1728 taking meto- as a ratio against the numberprolol, and 568 of days that should have been
supplied) , over 9 months
taking simvastatin,
mean age 55 years
(range 20–97 0.73 with educational intervention
years), with refill of
medication every
30 days 0.70 with usual care
Subgroup analysis
This analysis is of
568 people taking
simvastatin
educational inter-
vention
P <0.001Mean compliance, expressed
as % of maximum number of
tablets that should have been
100 people >70
years of age with
chronic stable
[55]
RCT
© BMJ Publishing Group Ltd 2011. All rights reserved. .......................................................... 23
Cardiovascular medication: improving adherence
Cardiovascular disorders
Favours
Effect
size
Results and statistical
analysisOutcome, InterventionsPopulation
Ref
(type)
consumed (assessed by pill
count)
heart failure, aver-
age age 85 years,
excluded if mobility 93% with educational intervention
disorder or if Fol-
stein's Mental 51% with control
Health test score
was <21 Absolute numbers not reported
Not significant
Adjusted difference between
means –1
Adherence measured by elec-
tronic medication monitor
(electronic lid that registered
245 people with
hypertension, on 1
or more medica-
[58]
RCT 95% CI –5.1 to +3.1
time and date of opening);
mean "timing compliance" de-
tions, attending pri-
mary care P = 0.63
fined as number of doses taken
at 24- or 12-hour intervals for
a once- or twice-daily regimen
respectively, divided by the to-
tal number of days x 100% , 6
months
87% with educational intervention
90% with usual care
Absolute numbers not reported
educational inter-
vention
Difference +11%
95% CI +5.0% to +16.7%
Adherence measured by elec-
tronic medication monitor
(electronic lid that registered
time and date of opening);
314 people with
heart failure, 50
years or older, at-
tending primary
[61]
RCT
"taking adherence" was de-care, on at least 1
medication fined as the % of prescribed
medication taken , over 9
months
Intervention lasted
for 9 months 79% with educational intervention
68% with usual care
Absolute numbers not reported
educational inter-
vention
Difference +6%
95% CI +0.4% to +11.5%
Adherence measured by elec-
tronic medication monitor
(electronic lid that registered
time and date of opening);
314 people with
heart failure, 50
years or older, at-
tending primary
[61]
RCT
"scheduling adherence" wascare, on at least 1
medication defined as day-to-day deviation
in timing of dose (e.g., once-
Intervention lasted
for 9 months daily within 2.4 hours of dose
on previous day) , over 9
months
53% with educational intervention
47% with usual care
Absolute numbers not reported
Not significant
Difference +4%
95% CI –2.8% to +10.7%
Adherence measured by elec-
tronic medication monitor
(electronic lid that registered
time and date of opening);
314 people with
heart failure, 50
years or older, at-
tending primary
[61]
RCT
"taking adherence" was de-care, on at least 1
medication fined as the % of prescribed
medication taken , 3 months
after intervention had finished
Intervention lasted
for 9 months 71% with educational intervention
67% with usual care
Absolute numbers not reported
Not significant
Difference +0.3%
95% CI –5.9% to +6.5%
Adherence measured by elec-
tronic medication monitor
(electronic lid that registered
time and date of opening);
314 people with
heart failure, 50
years or older, at-
tending primary
[61]
RCT
"scheduling adherence" wascare, on at least 1
medication defined as day-to-day deviation
© BMJ Publishing Group Ltd 2011. All rights reserved. .......................................................... 24
Cardiovascular medication: improving adherence
Cardiovascular disorders
Favours
Effect
size
Results and statistical
analysisOutcome, InterventionsPopulation
Ref
(type)
in timing of dose (e.g., once-
daily within 2.4 hours of dose
Intervention lasted
for 9 months on previous day) , 3 months
after intervention had finished
48.9% with educational interven-
tion
48.6% with usual care
Absolute numbers not reported
Increase in self-rated adher-
ence (assessed using the 4-
636 adults with hy-
pertension, attend-
[56]
RCT item Morisky Self-reported
Medication-Taking Scale), re-
ported as a %
ing primary care,
using hypertensive
medication at the
time of baseline
visit 9% with nurse-delivered educa-
tional/behavioural intervention by
protocol, bi-monthly via telephone
for 2 years
1% with usual care
Absolute numbers not reported
MINT
Difference –14%
95% CI –0.2% to –27%
Adherence measured by elec-
tronic pill monitors , 12 months
57% with research assistant
(RA)-delivered motivational inter-
190 African-Ameri-
cans with hyperten-
sion (88% women;
mean age 54
years), attending
[59]
RCT
P = 0.027
viewing (MINT) sessions at 3, 6,
9, and 12 months
community-based
primary care prac-
tices 43% with usual care
Absolute numbers not reported
educational maga-
zine
P = 0.0001Overall compliance rate (indi-
viduals with a treatment com-
pliance of 80–110%) , 24 weeks
450 people with
hypertension at-
tending primary
care
[57]
RCT
83% with educational magazine
sent to the patient's home twice
monthly
49% with usual care
Absolute numbers not reported
educational maga-
zine
P = 0.0001Correct time compliers , 24
weeks
450 people with
hypertension at-
tending primary
care
[57]
RCT 74% with educational magazine
sent to the patient's home twice
monthly
42% with usual care
Absolute numbers not reported
-
Adverse effects
-
-
No data from the following reference on this outcome. [48] [49] [50] [51] [52] [53] [54] [55] [56] [57] [58] [59] [60] [61]
-
-
Patient health education plus prompting mechanism versus usual care:
See option on prompting mechanisms, p 4 .
-
-
© BMJ Publishing Group Ltd 2011. All rights reserved. .......................................................... 25
Cardiovascular medication: improving adherence
Cardiovascular disorders
Patient health education versus prompting mechanisms, prescriber education, simplified dosing, or reminder
packaging:
We found 10 systematic reviews (search dates 1996; [16] 2000; [17] 2002; [18] [19] 2003; [21] [33] 2004; [23] 2007; [24]
2008; [25] 2009 [26] ), which identified no RCTs of sufficient quality.
-
-
-
Further information on studies
[48] Educational intervention Educational programme on facts about hypertension, benefits of treatment, need
for compliance, slide–audiotape format and booklet, and "patient educator" (non-health professional) to reinforce
messages.RCT methods Method of randomisation not described. Factorial design. Steelworkers also randomised
to family doctor or industrial physician care at the same time to see if difference in outcome; these results not
reported here. Interpretation complicated by factorial design, and only 144 (62%) of those randomised received
drug treatment — results based on these 144 men.
[49] Educational intervention Small group training followed by postal information packages. RCT methods Final
end points compared, not adjusted for baseline differences.Clinical outcomes RCT found no significant differ-
ence between groups for mean total cholesterol (P = 0.26), mean LDL (P = 0.48), or mean HDL (P = 0.48). [25]
Educational intervention significantly reduced triglyceride compared with usual care (P <0.05). [25]
[50] [60]
Educational intervention Group education (units of 15 people over 90 minutes, information about blood
pressure management and importance of adherence) and additional postal education at 1, 3, and 5 months.
RCT methods Method of randomisation not described; not reported if outcome assessment blinded; withdrawal
15/110 (14%) at 6 months and 18/110 (16%) at 2 years. [18]
[51] Educational intervention Four 90-minute meetings, emphasising importance of hypertension, including
videotape and other standard material, and general health education. RCT methods Randomisation procedure
was described. Compliance categorised as "good", "fair", or "poor", the basis of which not explained. Hence,
difficult to interpret results.
[52] Educational intervention Either "Threatening message" group: print informational tabloid containing material
on hypertension, its effects, control measures, and instructions on following regimen (this version emphasised
severity of hypertension and consequences) or "Positive message" group: print tabloid as above, but emphasised
positive health aspects of treatment.RCT methods Factorial design — people sequentially allocated to different
interventions. Only the first randomisation reported here. Method of randomisation not described. Level of
blinding not reported. At outset, 87% of participants were on medication. Follow-up for self-report scores and
pharmacy scores unclear.
[53] Educational intervention A 30- to 40-minute intervention with nurse practitioner and follow-up telephone call
4 weeks later; included reinforcement of regimen, brochure, 12-minute audiovisual presentation, discussion of
risk factors, and postcard reminder of next appointment.RCT methods Method of randomisation not described.
Level of blinding not described. Small study. Differential withdrawals — results based on 17/17 (100%) in inter-
vention group and 13/17 (76%) in control group.
[54] Educational intervention Single mailing of one relevant educational videotape on drug prescribed and inferred
disease state, 30 minutes long, including advice on compliance.RCT methods People selected by medication
from computerised database. RCT inferred that people taking benazepril and metoprolol had hypertension and
those taking simvastatin had hyperlipidaemia, which may or may not be the case. Also investigated people with
transdermal oestrogen; these results not reported here. Method of randomisation, level of blinding, and loss to
follow-up not reported. Not known if all participants in videotape group had access to a video player.
[55] Educational intervention Counselling programme including a standard written protocol employing verbal
counselling, medication calendars, and information leaflets.RCT methods Method of randomisation not de-
scribed. Follow-up 82/100 (82%).
[58] Educational intervention Adherence support session with practice nurse (20 minutes) followed by reinforcement
session (10 minutes), which explored patient concerns regarding medication, whether person understood diag-
nosis, and strategies to resolve medication problems. RCT methods Method of randomisation was described.
Open label. Follow-up 204/245 (83%).Trial noted that it found higher levels of adherence to medication than
in previous studies in similar populations.Clinical outcomes RCT found no significant difference between
groups at 6 months with regard to systolic (P = 0.24) or diastolic (P = 0.85) blood pressure.
[61] Educational intervention Pharmacist-delivered intervention by protocol, including verbal and written material,
exploring participant's understanding of disease or medication, and addressing low medication adherence. RCT
methods Method of randomisation was described. Assessment blinded. Analysis based on 270/314 (86%) of
those randomised. Found benefit when the intervention was being applied over the study period, which dissi-
pated over 3 months after the intervention had finished.
© BMJ Publishing Group Ltd 2011. All rights reserved. .......................................................... 26
Cardiovascular medication: improving adherence
Cardiovascular disorders
[56] Educational intervention Nurse-led intervention delivered by telephone every 8 weeks over 6 months. Patient
factors targeted in the tailored behavioural intervention include perceived risk of hypertension and knowledge,
memory, medical and social support, patients’ relationship with their health care provider, adverse effects of
medication therapy, weight management, exercise, diet, stress, smoking, and alcohol use; Self-rated adherence
was assessed using the 4-item Morisky Self-reported Medication-Taking Scale [Morisky 1986].The scale for
each item was revised to include the response categories "strongly agree", "agree", "disagree", and "strongly
disagree".Those individuals who reported "strongly agree", "agree", "do not know", or "refused" to any of the
4 items were classified as non-adherent. Reported as a percentage and measured at baseline and post-inter-
vention (6 months).RCT methods Method of randomisation not described. Follow-up reported as 96% retention
rate at 6 months; otherwise not specified.
[59] Educational intervention Motivational interviewing. RCT methods Method of randomisation was described.
After baseline assessment, patients were randomly assigned to either UC or MINT group by the study statistician,
using sealed envelopes. Separate randomisation schedules were developed from a computerised random-
number generator, balanced at set intervals, using permutated blocks, to assure equal numbers in each group.
Owing to the nature of the behavioural intervention, neither the patients nor the RAs were blinded to the inter-
vention. However, the clinic staff who recorded the blood pressure data were blinded to patient assignment. It
is important to note that medication adherence data were downloaded automatically into the computer from the
MEMS caps.Thus, neither the researchers nor the patients could affect MEMS adherence outcome. Results
based on 83% (79/95) of intervention group and 85% (81/95) of UC group.Clinical outcomes Motivational in-
terviewing over 12 months led to a steady maintenance of medication adherence, compared with significant
decline in adherence for the usual care group.This effect was associated with a modest, non-significant trend
towards a net reduction in systolic blood pressure in favour of intervention.
[57] Educational intervention Educational magazine sent to the patient's home twice monthly. RCT methods
Method of randomisation was not described. Follow-up 393/450(87%). Clinical outcomes Educational magazine
led to increased adherence and improved blood pressure control.
-
-
Comment: Clinical guide:
Most factors known to affect adherence to medication — such as knowledge, health beliefs, per-
ception of risk, convenience, and memory — relate to the patient. It is therefore not surprising that
interventions to address adherence should focus on patient education, but what is perhaps surprising
is that such interventions do not seem to have greater influence on adherence. However, the
phenomenon of adherence is complex, and traditional educational methods may fail to recognise
this. People's beliefs and preferences need to be acknowledged and incorporated into adherence-
enhancing interventions. [62] A combination of strategies including prompting mechanisms and
simplified dosing, alongside patient education with emphasis on the patient's perspective, may
have a more successful impact on adherence.The rationale that the complexity of adherence be-
haviour may respond better to a multi-factorial approach that is patient-centred is supported by an
RCT of an educational intervention plus a prompting intervention that significantly improved adher-
ence compared with usual care, [32] by an RCT that used a combination of face-to-face education
and follow-up postal education, [50] [60] by an RCT of a nurse-delivered telephone intervention that
increased self-reported medication adherence by 9% in the intervention group versus 1% in the
usual care group, [56] and by an RCT of motivational interviewing delivered every 3 months over
12 months, which led to a steady maintenance of medication adherence compared with significant
decline in adherence for the usual care group. [59]
GLOSSARY
Fixed-dose combination A formulation of two or more active ingredients combined in a single dosage form available
in certain fixed doses.
Low-quality evidence Further research is very likely to have an important impact on our confidence in the estimate
of effect and is likely to change the estimate.
Mini-Mental score A score derived from the Folstein Mini Mental State Examination.This examination is used to
evaluate dementia, and consists of a series of questions and tasks to assess a patient's orientation, attention, calcu-
lation, language, visuospatial, executive, and short-term memory abilities.The cut off for dementia is a score of less
than 24 out of a possible 30.
Very low-quality evidence Any estimate of effect is very uncertain.
SUBSTANTIVE CHANGES
Prescriber education New evidence added. [24] [25] [43] Categorisation unchanged (Unknown effectiveness) because
current evidence is in a single restricted population only.
© BMJ Publishing Group Ltd 2011. All rights reserved. .......................................................... 27
Cardiovascular medication: improving adherence
Cardiovascular disorders
Prompting mechanisms New evidence added. [24] [25] [26] Categorisation unchanged (Likely to be beneficial).
Reminder packaging (calendar [blister] packs; multi-dose pill boxes) New evidence added. [24] [25] [26] [46]
Categorisation unchanged (Unknown effectiveness) because of conflicting results among trials.
Simplified dosing New evidence added. [24] [26] [40] [42] Categorisation unchanged (Likely to be beneficial).
Patient health education New evidence added. [24] [25] [26] [56] [57] [59] Categorisation changed from Unlikely to
be beneficial to Unknown effectiveness because of conflicting results among trials; new evidence suggests that some
more patient-centred approaches to education, using newer media, may be effective whereas traditional education
fails to improve adherence.
REFERENCES
1. Osterberg L, Blaschke T. Adherence to medication.N Engl J Med
2005;353:487–497.[PubMed]
2. Brawley LR, Culos-Reed SN. Studying adherence to therapeutic regimens:
overview, theories, recommendations.Control Clin Trials 2000;21(5 Sup-
pl):156S–163S.[PubMed]
3. Dieppe P, Horne R. Soundbites and patient centred care.BMJ 2002;325:605.
4. Marinker M, Shaw J. Not to be taken as directed. BMJ
2003;326:348–349.[PubMed]
5. Pullar T, Feely, MP. Reporting compliance in clinical trials. Lancet
1990;336:1252–1253.[PubMed]
6. Skaer TL, Sclar DA, Markowski DJ, et al. Effect of value-added utilities in promot-
ing prescription refill compliance among patients with hypertension. Curr Ther
Res Clin Exp 1993;53:251–255.
7. Fulmer TT, Feldman PH, Kim TS, et al. An intervention study to enhance medi-
cation compliance in community-dwelling elderly individuals. J Gerontol Nurs
1999;25:6–14.[PubMed]
8. Urquhar t J. The electronic medication event monitor: lessons for pharmacother-
apy. Clin Pharmacokinet 1997;32:345–356.[PubMed]
9. Urquhar t J. The odds of the three nons when an aptly prescribed medicine isn't
working: non-compliance, non-absorption, non-response. Br J Clin Pharmacol
2002;54:212–220. [PubMed]
10. Burnier M. Long-term compliance with antihypertensive therapy: another facet
of chronotherapeutics in hypertension. Blood Press Monit 2000;5(Suppl
1):S31–S34.[PubMed]
11. Paes AH, Bakker A, Soe-Agnie CJ. Impact of dosage frequency on patient
compliance. Diabetes Care 1997;20:1512–1517.[PubMed]
12. Primatesta P, Poulter NR. Lipid concentrations and the use of lipid lowering
drugs: evidence from a national cross sectional survey. BMJ
2000;321:1322–1325.[PubMed]
13. Benner JS, Glynn RJ, Mogun H, et al. Long-term persistence in use of statin
therapy in elderly patients. JAMA 2002;288:455–461.[PubMed]
14. Tsuyuki RT, Bungard TJ. Poor adherence with hypolipidemic drugs: a lost oppor-
tunity.Pharmacotherapy 2001;21:576–582.[PubMed]
15. Jackevicius CA, Mamdani M, Tu JV. Adherence with statin therapy in elderly
patients with and without acute coronary syndromes. JAMA
2002;288:462–467.[PubMed]
16. Newell SA, Bowman JA, Cockburn JD, et al. A critical review of interventions to
increase compliance with medication-taking, obtaining medication refills, and
appointment-keeping in the treatment of cardiovascular disease.Prev Med
1999;29:535–548.[PubMed]
17. Peterson AM, Takiya L, Finley R, et al. Meta-analysis of interventions to improve
drug adherence in patients with hyperlipidemia. Pharmacotherapy
2003;23:80–87.[PubMed]
18. Schroeder K, Fahey T, Ebrahim S. Interventions for improving adherence to
treatment in patients with high blood pressure in ambulatory settings. In:The
Cochrane Library, Issue 3, 2010. Chichester, UK: John Wiley & Sons, Ltd. Search
date 2002.
19. Petrilla AA, Benner JS, Battleman DS, et al. Evidence-based interventions to
improve patient compliance with antihypertensive and lipid-lowering medications.
Int J Clin Pract 2005;59:1441–1451.[PubMed]
20. Bennett JW, Glasziou PP. Computerised reminders and feedback in medication
management: a systematic review of randomised controlled trials.Med J Australia
2003;178:217–222.[PubMed]
21. Van der Wal MHL, Jaarsma T, Van Veldhuisen DJ. Non-compliance in patients
with heart failure; how can we manage it? Eur J Heart Fail 2005;7:5–17.[PubMed]
22. Mistiaen P, Poot E.Telephone follow-up, initiated by a hospital-based health
professional, for postdischarge problems in patients discharged from hospital to
home. In:The Cochrane Library, Issue 3, 2010. Chichester, UK: John Wiley &
Sons, Ltd. Search date 2003.
23. Russell CL, Conn VS, Jantarakupt P. Older adult medication compliance: inte-
grated review of randomized controlled trials.Am J Health Behav
2006;30:636–650.[PubMed]
24. Haynes RB, Ackloo E, Sahota N, et al. Interventions for enhancing medication
adherence. In:The Cochrane Library, Issue 3, 2010. Chichester, UK: John Wiley
& Sons, Ltd. Search date 2007.
25. Schedlbauer A, Davies P, Fahey T. Interventions to improve adherence to lipid
lowering medication. In:The Cochrane Library, Issue 3, 2010. Chichester, UK:
John Wiley & Sons, Ltd. Search date 2008.
26. Christensen A, Osterberg LG, Hansen EH, et al. Electronic monitoring of patient
adherence to oral antihypertensive medical treatment: a systematic review. J
Hypertens 2009;27:1540–1551.[PubMed]
27. McKenney JM, Munroe WP, Wright JT Jr. Impact of an electronic medication
compliance aid on long-term blood pressure control. J Clin Pharmacol
1992;32:277–283.[PubMed]
28. Faulkner MA, Wadibia EC, Lucas BD, et al. Impact of pharmacy counseling on
compliance and effectiveness of combination lipid-lowering therapy in patients
undergoing coronary artery revascularization: a randomized, controlled trial.
Pharmacotherapy 2000;20:410–416.[PubMed]
29. Simkins CV, Wenzloff NJ. Evaluation of a computerized reminder system in the
enhancement of patient medication refill compliance. Drug Intell Clin Pharm
1986;20:799–802.[PubMed]
30. Marquez Contreras E, Vegazo Garcia O, Claros NM, et al. Efficacy of telephone
and mail intervention in patient compliance with antihypertensive drugs in hyper-
tension. ETECUM-HTA study.Blood Press 2005;14:151–158.[PubMed]
31. Skaer TL, Sclar DA, Markowski DJ, et al. Effect of value-added utilities on pre-
scription refill compliance and health care expenditures for hypertension. J Hum
Hypertens 1993;7:515–518.[PubMed]
32. Sclar DA, Chin A, Skaer TL, et al. Effect of health education in promoting pre-
scription refill compliance among patients with hypertension. Clin Ther
1991;13:489–495.[PubMed]
33. Van Wijk BL, Klungel OH, Heerdink ER, et al. Effectiveness of interventions by
community pharmacists to improve patient adherence to chronic medication: a
systematic review.Ann Pharmacother 2005;39:319–328.[PubMed]
34. Connor J, Rafter N, Rodgers A, et al. Do fixed-dose combination pills or unit-of-
use packaging improve adherence? A systematic review. Bull World Health Organ
2004;82:935–939.[PubMed]
35. Baird MG, Bentley-Taylor MM, Carruthers SG, et al. A study of efficacy, tolerance
and compliance of once-daily versus twice-daily metoprolol (Betaloc) in hyperten-
sion. Betaloc Compliance Canadian Cooperative Study Group. Clin Invest Med
1984;7:95–102.[PubMed]
36. Boissel JP, Meillard O, Perrin-Fayolle E, et al. Comparison between a bid and a
tid regimen: improved compliance with no improved antihypertensive effect. The
EOL Research Group. Eur J Clin Pharmacol 1996;50:63–67.[PubMed]
37. Brown BG, Bardsley J, Poulin D. Moderate dose, three-drug therapy with niacin,
lovastatin, and colestipol to reduce low-density lipoprotein cholesterol less than
100 mg/dl in patients with hyperlipidemia and coronary artery disease. Am J
Cardiol 1997;80:111–115.[PubMed]
38. Brun J. Patient compliance with once-daily and twice-daily oral formulations of
5-isosorbide mononitrate: a comparative study. J Int Med Res 1994;22:266–272.
Erratum in: J Int Med Res 1994;22:350.[PubMed]
39. Girvin B, McDermott BJ, Johnston GD. A comparison of enalapril 20 mg once
daily versus 10 mg twice daily in terms of blood pressure lowering and patient
compliance. J Hypertens 1999;17:1627–1631.[PubMed]
40. Udelson JE, Pressler SJ, Sackner-Bernstein J, et al. Adherence with once daily
versus twice daily carvedilol in patients with heart failure: the Compliance And
Quality of Life Study Comparing Once-Daily Controlled-Release Carvedilol CR
and Twice-Daily Immediate-Release Carvedilol IR in Patients with Heart Failure
(CASPER) Trial. J Card Fail 2009;15:385–393.[PubMed]
41. Gupta AK, Arshad S, Poulter NR, et al. Compliance, safety, and effectiveness
of fixed-dose combinations of antihypertensive agents: a meta-analysis.Hyper-
tension 2010;55:399–407.[PubMed]
42. Kardas P. Compliance, clinical outcome, and quality of life of patients with stable
angina pectoris receiving once-daily betaxolol versus twice daily metoprolol: a
randomized controlled trial. Vasc Health Risk Manag 2007;3:235–242.[PubMed]
43. Qureshi NN, Hatcher J, Chaturvedi N, et al. Effect of general practitioner education
on adherence to antihypertensive drugs: cluster randomised controlled trial.BMJ
2007;335:1030.[PubMed]
44. Heneghan CJ, Glasziou P, Perera R. Reminder packaging for improving adher-
ence to self-administered long-term medications. In:The Cochrane Library, Issue
3, 2010. Chichester, UK: John Wiley & Sons, Ltd. Search date 2004.
45. Becker LA, Glanz K, Sobel E, et al. A randomized trial of special packaging of
antihypertensive medications. J Fam Pract 1986;22:357–361.[PubMed]
46. Schneider PJ, Murphy JE, Pedersen CA, et al. Impact of medication packaging
on adherence and treatment outcomes in older ambulatory patients. J Am Pharm
Assoc (2003) 2008;48:58–63.[PubMed]
47. Christensen A, Christrup LL, Fabricius PE, et al. The impact of an electronic
monitoring and reminder device on patient compliance with antihypertensive
therapy: a randomized controlled trial.J Hyper tens 2010;28:194–200.[PubMed]
48. Sackett DL, Haynes RB, Gibson ES, et al. Randomised clinical trial of strategies
for improving medication compliance in primary hypertension. Lancet
1975;1:1205–1207.[PubMed]
49. Marquez Contreras E, Casado-Martinez JJ, Lopez DA, et al.Therapeutic compli-
ance in dyslipidaemias: a trial of the efficacy of health education. Aten Primaria
1998;22:79–84. [In Spanish][PubMed]
50. Marquez Contreras E, Casado Martinez JJ, Ramos Perez J, et al.Trial of the
efficacy of health education programs on therapeutic compliance in arterial hy-
pertension. Aten Primaria 1998;21:199–204. [In Spanish][PubMed]
51. Pierce JP, Watson DS, Knights S, et al. A controlled trial of health education in
the physician's office. Prev Med 1984;13:185–194.[PubMed]
© BMJ Publishing Group Ltd 2011. All rights reserved. .......................................................... 28
Cardiovascular medication: improving adherence
Cardiovascular disorders
52. Kirscht JP, Kirscht JL, Rosenstock IM. A test of interventions to increase adher-
ence to hypertensive medical regimens.Health Educ Q 1981;8:261–272.[PubMed]
53. Hamilton GA, Roberts SJ, Johnson JM, et al. Increasing adherence in patients
with primary hypertension: an intervention. Health Val 1993;17:3–11.
54. Powell KM, Edgren B. Failure of educational videotapes to improve medication
compliance in a health maintenance organization. Am J Health Syst Pharm
1995;52:2196–2199.[PubMed]
55. Goodyer L, Miskelly F, Milligan P. Does encouraging good compliance improve
patients' clinical condition in heart failure? Br J Clin Pract
1995;49:173–176.[PubMed]
56. Bosworth HB, Olsen MK, Neary A, et al. Take Control of Your Blood Pressure
(TCYB) study: a multifactorial tailored behavioral and educational intervention
for achieving blood pressure control.Patient Educ Couns
2008;70:338–347.[PubMed]
57. Marquez Contreras E, Martel Claros N, Gil Guillen V, et al. Non-pharmacological
intervention as a strategy to improve antihypertensive treatment compliance.
Aten Primaria 2009;41:501–510. [In Spanish][PubMed]
58. Schroeder K, Fahey T, Hollinghurst S, et al. Nurse-led adherence support in hy-
pertension: a randomized controlled trial. Fam Pract 2005;22:144–151.[PubMed]
59. Ogedegbe G, Chaplin W, Schoenthaler A, et al. A practice-based trial of motiva-
tional interviewing and adherence in hypertensive African Americans. Am J Hy-
pertens 2008;21:1137–1143.[PubMed]
60. Marquez Contreras E, Casado Martinez JJ, Celotti Gomez B, et al. Treatment
compliance in arterial hypertension. A 2-year intervention trial through health
education. Aten Primaria 2000;26:5–10. [In Spanish][PubMed]
61. Murray MD, Young J, Hoke S, et al. Pharmacist intervention to improve medication
adherence in heart failure: a randomized trial. Ann Intern Med
2007;146:714–725.[PubMed]
62. Wei L, Wang J, Thompson P, et al. Adherence to statin treatment and readmission
of patients after myocardial infarction: a six year follow up study.Heart
2002;88:229–233.[PubMed]
Liam Glynn
Senior Lecturer in General Practice
National University of Ireland
Galway
Ireland
Tom Fahey
Professor of General Practice
RCSI Medical School
Dublin
Ireland
Competing interests: LG and TF declare that they have no competing interests.
Disclaimer
The information contained in this publication is intended for medical professionals. Categories presented in Clinical Evidence indicate a
judgement about the strength of the evidence available to our contributors prior to publication and the relevant importance of benefit and
harms. We rely on our contributors to confirm the accuracy of the information presented and to adhere to describe accepted practices.
Readers should be aware that professionals in the field may have different opinions. Because of this and regular advances in medical research
we strongly recommend that readers' independently verify specified treatments and drugs including manufacturers' guidance. Also, the
categories do not indicate whether a particular treatment is generally appropriate or whether it is suitable for a particular individual. Ultimately
it is the readers' responsibility to make their own professional judgements, so to appropriately advise and treat their patients.To the fullest
extent permitted by law, BMJ Publishing Group Limited and its editors are not responsible for any losses, injury or damage caused to any
person or property (including under contract, by negligence, products liability or otherwise) whether they be direct or indirect, special, inci-
dental or consequential, resulting from the application of the information in this publication.
© BMJ Publishing Group Ltd 2011. All rights reserved. .......................................................... 29
Cardiovascular medication: improving adherence
Cardiovascular disorders
TABLE 1 Definitions of different types of reminder packaging*
Definitions of different types of reminder packaging
Monitored dosage system (MDS): medications are manually packed into blister/bubble trays under the supervision of a pharmacist
and then cold- or heat-sealed with foil. Examples of these systems are the Nomad® and Manrex®. Patients using an MDS are
provided with weekly or monthly blister packs
1Pill boxes
Multi-compartment compliance aid (MCA) or dose administration aid: these are plastic trays or boxes that hold 7 days of a patient's
medicine and are divided into days of the week. Each day of the week has a sliding lid, which covers compartments for different
dosing times (usually 4 compartments for each day).They are commonly but not exclusively used for multiple medications.
Examples of these are Dosett®, Medidos®, and the Mediset
2
Calendar blister: a blister package designed to aid a patient's memory by incorporating the day/time when each dose is to be
taken into the package design
1Pre-packaged blister packs
Unit dose: the prescribed amount of each dosage in a package.This type of packaging can incorporate a reminder system2
Unit of use: the exact amount of a drug treatment prepackaged by the manufacturer or pharmacist in standardised amounts.
This type of packaging can incorporate a reminder system
3
* Source: Heneghan CJ, Glasziou P, Perera R. Reminder packaging for improving adherence to self-administered long term medication. In:The Cochrane Library, Issue 3, 2010. Chichester, UK: John Wiley &
Sons, Ltd. Search date 2004. Copyright Cochrane Collaboration, reproduced with permission.
© BMJ Publishing Group Ltd 2011. All rights reserved. ............................................................................................................ 30
Cardiovascular medication: improving adherence
Cardiovascular disorders
GRADE Evaluation of interventions for Cardiovascular medication: improving adherence.
-
Adherence to medication
Important out-
comes
CommentGRADE
Effect
size
Direct-
ness
Consisten-
cyQuality
Type of
evidenceComparisonOutcome
Studies (Partici-
pants)
What are the effects of interventions to improve adherence to long-term medication for CVD in adults? Quality points deducted for incomplete reporting of re-
sults and weak methods (method of randomisation not
described, level of blinding not reported). Directness
points deducted for diverse interventions affecting gen-
eralisability, and diverse range of outcome assessment
and analysis
Very low0
–20–24Prompting mechanisms ver-
sus usual care
Adherence to
medication
5 (1107) [7] [27]
[28] [29] [30]
Quality points deducted for incomplete reporting of re-
sults and weak methods (method of randomisation not
described, level of blinding not reported). Directness
point deducted for unclear intervention (unit-of-use inter-
vention not fully defined)
Very low0
–10–24Prompting mechanism plus
usual care versus unit-of-use
packaging plus usual care
versus unit-of-use packaging
plus prompting mechanism
plus usual care versus usual
care alone
Adherence to
medication
2 (432) [31] [6]
Quality point deducted for weak methods (method of
randomisation not described, level of blinding not report-
ed). Directness point deducted for unclear validity of
outcome assessment/single measure of adherence used
Low0
–10–14Prompting mechanism plus
patient health education ver-
sus usual care
Adherence to
medication
1 (453) [32]
Quality points deducted for weak methods (method of
randomisation not described, level of blinding not report-
ed) and inclusion of 2 crossover RCTs. Directness point
deducted for diverse range of outcome assessment and
analysis
Very low0
–10–24Simplified dosing regimens
versus more complex regi-
mens
Adherence to
medication
6 (8155) [35] [36]
[37] [38] [39] [40]
Quality point deducted for uncertainty about randomisa-
tion method. Directness point deducted for restricted
population (limited to a developing country)
Low0
–10–14Prescriber education versus
usual care
Adherence to
medication
1 (200) [43]
Quality points deducted for incomplete reporting of re-
sults and for weak methods (method of randomisation
not described, level of blinding for outcome assessment
not reported). Consistency point deducted for conflicting
results
Very low0
0–1–24Calendar (blister) pack ver-
sus usual care
Adherence to
medication
2 (265) [45] [46]
Quality points deducted for incomplete reporting of re-
sults and weak methods (method of randomisation not
described, level of blinding not reported). Consistency
point deducted for conflicting results. Directness points
deducted for diverse interventions affecting generalis-
ability and diverse range of outcome assessment and
analysis
Very low0
–2–1–24Patient health education
versus usual care
Adherence to
medication
13 (at least
5437) [48] [49] [50]
[60] [51] [52] [53]
[54] [55] [58] [61]
[56] [57] [59]
We initially allocate 4 points to evidence from RCTs, and 2 points to evidence from observational studies. To attain the final GRADE score for a given comparison, points are deducted or added from this initial
score based on preset criteria relating to the categories of quality, directness, consistency, and effect size. Quality: based on issues affecting methodological rigour (e.g., incomplete reporting of results, quasi-
randomisation, sparse data [<200 people in the analysis]). Consistency: based on similarity of results across studies. Directness: based on generalisability of population or outcomes. Effect size: based on magnitude
of effect as measured by statistics such as relative risk, odds ratio, or hazard ratio.
© BMJ Publishing Group Ltd 2011. All rights reserved. ............................................................................................................ 31
Cardiovascular medication: improving adherence
Cardiovascular disorders
-
© BMJ Publishing Group Ltd 2011. All rights reserved. ............................................................................................................ 32
Cardiovascular medication: improving adherence
Cardiovascular disorders
