The online version of this article can be found at:
2010 6: 243 originally published online 7 July 2010Chronic Illness
Robert Horne, Jane Clatworthy, Matthew Hankins and On behalf of the ASCOT Investigators
High adherence and concordance within a clinical trial of antihypertensives
can be found at:
Additional services and information for
What is This?
- Jul 7, 2010 OnlineFirst Version of Record
- Dec 6, 2010 Version of Record >>
by guest on October 22, 2013 by guest on October 22, 2013 by guest on October 22, 2013 by guest on October 22, 2013 by guest on October 22, 2013 by guest on October 22, 2013 by guest on October 22, 2013 by guest on October 22, 2013 by guest on October 22, 2013 by guest on October 22, 2013chi.sagepub.comchi.sagepub.comchi.sagepub.comchi.sagepub.com chi.sagepub.comchi.sagepub.comchi.sagepub.comchi.sagepub.com chi.sagepub.comchi.sagepub.com Downloaded from Downloaded from Downloaded from Downloaded from Downloaded from Downloaded from Downloaded from Downloaded from Downloaded from Downloaded from
Chronic Illness (2010) 6, 243–251
High adherence and concordance within a clinical
trial of antihypertensives
ROBERT HORNE*, JANE CLATWORTHY* and MATTHEW HANKINSy
On behalf of the ASCOT Investigators
*Centre for Behavioural Medicine, The School of Pharmacy, University of London,
Mezzanine Floor, BMA House, Tavistock Square, London WC1H 9JP, UK
yDivision of Primary Care & Public Health, Brighton & Sussex Medical School,
Mayfield House, University of Brighton, Falmer, BN1 9PH, UK
Received 22 November 2009, Accepted 20 March 2010
Objectives: To explore hypertensive patients’ beliefs about their condition and its treatment and their adherence
within the context of a clinical trial. To assess the degree of concordance between patients’ beliefs about
hypertension and the medical model of the condition.
Methods: This was a questionnaire-based study, involving 230 patients participating in the Anglo-Scandinavian
Cardiac Outcomes Trial (ASCOT), a randomized controlled trial comparing two pharmaceutical approaches to the
management of hypertension. A comparison group of 106 hypertensive patients who were screened for ASCOT but
did not meet the entry criteria was also recruited. Outcome measures were beliefs about hypertension and
antihypertensive medication, and adherence to medication (self-report and tablet count).
Results: Adherence to medication was higher than anticipated, with 45% participants reporting complete
adherence over the 18-month study and a further 40% only rare non-adherence. Patients’ beliefs about their
condition and treatment were generally concordant with the medical model of hypertension. High concordance was
associated with high medication adherence (p50.001).
Discussion: Clinical trial volunteers may have beliefs that are unusually concordant with the medical model of
hypertension and may demonstrate atypically high adherence. This has implications for the transferability of trial
findings to the general hypertensive population.
Keywords: Adherence, Clinical trial, Hypertension, Illness perceptions, Treatment perceptions
Anti-hypertensive medications are effective
in reducing high blood pressure and have
been shown to significantly reduce the risk of
cardiovascular illness.1It is thought, how-
ever, that many patients fail to accrue the full
research has indicated that about a third to
half of hypertensive patients do not adhere to
their anti-hypertensive medication.3–5
(such as forgetting) or can be intentional,
whereby patients make a decision not to take
treatment based on their personal beliefs
about their illness and treatment.6Studies in
other conditions have found that patients
develop ‘common-sense’ models of their
illness and treatment that can be at odds
with the scientific view.7The recognition of
the importance of patients’ personal beliefs
about illness and treatment as barriers to
adherence has led to the development of the
concept of concordance to describe the degree
of match or mismatch between practitioner
Reprint requests to: Robert Horne.
? The Author(s), 2010. Reprints and permissions:
and patient views about the illness and
models of illness may be particularly salient
in the case of hypertension. The clinical
model of hypertension is an asymptomatic
condition where intervention is prophylactic
and the patient is required to take regular
long-term medication to prevent future com-
plications. However, the notion of an asymp-
tomatic condition may be counter-intuitive
to many patients if their experiences of other
illnesses have been symptomatic. In the
absence of a clearly defined symptomatol-
ogy, patients may assign common, everyday
symptoms to their hypertension to ‘make
sense’ of their condition. For example, in a
recent study, approximately 30% partici-
pants attributed tiredness and hot flushes
to their hypertension.9Patients who have a
symptomatic model of their condition are
more likely to doubt their personal need for
regular medication.10,11Without a clear
rationale for the necessity of medication,
patients are less likely to adhere, especially if
they are concerned about potential adverse
Clinical trials, however, are a special case.
It has been suggested that non-adherence
may be less prevalent in clinical trials than in
practice,13which may account for poorer
reported efficacy of drugs in clinical practice
than in clinical trials.14It is unclear whether
this is a result of the specialist care and
attention received by patients in a clinical
trial, or reflects the type of people who are
prepared to volunteer for a trial.
The present study set out to address out-
standing questions relating to beliefs and
adherence among patients volunteering for
inclusion in a clinical trial. The aims were:
1. to identify the prevalence of non-adher-
ence to antihypertensive medication
within a clinical trial,
2. to assess hypertensive patients’ percep-
tions of their condition and their treat-
ment and the extent to which they are
concordant with the medical model of
3. to explore the relationship between con-
cordance with the medical model of
hypertension and adherence to medica-
4. to determine whether beliefs about
hypertension and its treatment and
those who are participating in a clinical
trial and those who volunteered but did
not meet the full entry criteria and
remained in routine care.
Clinical Trial Sample
Brighton arm of the Anglo-Scandinavian
Cardiac Outcomes Trial (ASCOT) were
invited to participate in this study by the
trial nurses. ASCOT was a randomized
receive either a beta blocker- or calcium
channel blocker-based treatment regimen
and those with total cholesterol of 6.5
mmol/L or less were additionally random-
ized to receive a statin or placebo. In order to
be eligible for the ASCOT, trial patients
needed to be hypertensive and aged between
40 and 80 years of age. They were also
required to have at least three other risk
factors for a cardiovascular event (e.g. type-2
diabetes, over 55 years old, regular smoker,
family history of coronary artery disease).
ASCOT but who had not met the full entry
criteria (e.g. insufficient additional risk fac-
tors for a cardiovascular event identified)
and therefore remained in routine care were
This was a prospective, longitudinal study in
questionnaires to assess their perceptions of
hypertension, beliefs about anti-hyperten-
sive medication and adherence to anti-
hypertensive medication on four occasions
over an 18-month period (baseline, 6
months, 12 months, 18 months).
The trial participants were invited to take
part in a study of ‘patients’ views about their
illness’ by the nurse at an ASCOT appoint-
ment. Patients were given an information
sheet and a consent form. The non-trial
participants were sent these by the post.
Those wishing to take part returned their
researcher in the stamped addressed enve-
lope provided. The researcher then sent four
questionnaires over the next 18 months at
6-month intervals. All participants were
telephoned shortly after each questionnaire
was sent to check that it had arrived and
address any questions they might have. One
postal reminder was sent if the questionnaire
had not been returned within a month.
Patients’ perceptions of hypertension were
assessed using an adapted version of the
Revised Illness Perception Questionnaire
(IPQ-R),16which has been validated and
used in a range of illness groups including
hypertension.17,18Eight of the sub-scales
comprising the IPQ-R were used to assess
beliefs about hypertension (Identity (symp-
toms), Timeline—acute/chronic, Timeline—
cyclical, Consequences, Personal control,
Treatment control, Coherence, Emotional
Perceived symptoms (Identity subscale)
were measured using a 19-item list compris-
ing 12 common symptoms (e.g. headache,
breathlessness and nausea) and 5 symptoms
often attributed by patients to their hyper-
tension (fast heart rate, ‘pins and needles’,
feeling flushed, loss of libido and impo-
tence). Participantswere required to indicate
which of the symptoms were experienced as
part of their hypertension. The other scales
were assessed through patients’ responses to
statements (e.g. ‘my high blood pressure will
last a long time’ (Timeline—acute/chronic),
‘my treatment can control my high blood
pressure’ (Treatment control)) on a 5-point
Likert scale ranging from ‘strongly agree’ to
‘strongly disagree’. For each of these sub-
scales, item responses were summed and a
mean score was computed.
Beliefs about anti-hypertensive medica-
tion were assessed using the Beliefs about
Medicines Questionnaire (BMQ), which has
been validated in a range of illness groups
including cardiac samples.19
comprises a five-item Medication Necessity
sub-scale that assesses participants’ beliefs
about their personal need for taking their
medication (e.g. ‘Without these medicines
I would be very ill) and a six-item Medication
Concerns sub-scale that assesses their con-
cerns about the potential adverse effects of
taking medication (e.g. ‘These medicines
give me unpleasant side effects’). Items
were rated on a 5-point Likert scale ranging
from ‘strongly disagree’ to ‘strongly agree’.
For each sub-scale, responses were summed
and mean scoreswere computed, with higher
scores indicating stronger beliefs in the con-
struct represented by the scale.
Adherence to anti-hypertensive medica-
tion was assessed by patient self-report and
Patient self-report: The six-item Medication
Adherence Report Scale (MARS) elicits
patients’ reports of non-adherence.11In
order to diminish the social pressure on
patients to report high adherence, items are
phrased in a non-threatening manner and
patients are assured that their responses will
be anonymous and confidential. Participants
were asked to rate the frequency with which
they engaged in each of the six aspects of
non-adherent behaviour listed (e.g. ‘I forget
to take these medicines’, ‘I stop taking these
medicines for a while’) using a 5-point scale
ranging from ‘never’ to ‘always’. Scores for
each item are summed to give a total
score ranging from 6 to 30, with higher
scores indicating higher levels of adherence.
CLINICAL TRIAL OF ANTIHYPERTENSIVES245
MARS scores were calculated for each time
point and a mean MARS score across the
four time points was computed.
Tablet count: As part of the trial protocol,
all patients were required to return their
medication containers at each ASCOT
appointment when new medication was
issued. The number of tablets returned by
each patient at each appointment was
counted and the percentage of prescribed
medication removed from the containers
Data analyses were conducted using the
Statistical Package for the Social Sciences
A repeated measures ANOVA found no
significant differences in patients’ beliefs
about hypertension and antihypertensive
medication over the 18 months of the
study, with the exception of Treatment
Concerns that decreased (F[3, 654]¼4.93,
p50.01). This decrease was small, despite
being statistically significant (in a possible
range of 1–5, mean scores fell from 2.43 to
2.26). Therefore, all subsequent analyses
used baseline beliefs data.
frequencies) were used to explore levels of
non-adherence using both the self-report
and tablet count data.
Descriptive statistics were also used to
compare participants’ perceptions of their
condition and its treatment with the medical
model. The medical model of hypertension
was defined as an asymptomatic, chronic
condition, controllable through medicines
that are both safe and necessary. Patients’
mean scores on the IPQ-R Identity, IPQ-R
Timeline (acute–chronic), IPQ-RTreatment
Control, BMQ Treatment Necessity and
BMQ Treatment Concerns sub-scales were
dichotomized into ‘in agreement with the
medical model’ (scored 1) and ‘not in agree-
ment with medical model’ (scored 0).
Agreement with the medical model was char-
acterized by a mean score greater than the
midpoint on the Timeline (acute–chronic),
Treatment Control and Treatment Necessity
sub-scales (43), a mean score lower than the
midpoint on the Treatment Concerns sub-
scale (53) and a report of no symptoms
experienced as part of hypertension on the
Identity subscale. A total concordance score
(ranging from 0 to 5) was computed for each
participant by summing their concordance
scores on the individual subscales. An inde-
pendent samples t-test was conducted to
compare those reporting lower and higher
levels of adherence on their concordance with
the medical model of hypertension.
Independent samples t-tests were used to
compare ASCOT patients and non-trial
patients on their beliefs about hypertension
and its treatment and self-reported adher-
ence to medication. They were also com-
pared in terms of age (independent samples
t-test) and gender (chi-square).
In total, 336 patients took part in the study
(i.e. returned one or more questionnaire):
230 were ASCOT patients (88% male, mean
age 67 years) and 106 were non-trial partic-
ipants (57% male, mean age 68 years).
Levels of attrition were lower than expected
for this questionnaire-based study, with a
mean response rate of 93% for ASCOT
patients and 90% for non-trial participants.
Prevalence of Non-adherence
to Anti-hypertensive Medication
within a Clinical Trial
Mean self-reported adherence to medication
was very high, with a mean baseline MARS
score of 29.3 (SD¼1.5) out of a maximum
possible score of 30. High rates of adherence
were maintained over the course of the
study, with 103 (45%) participants reporting
complete adherence over the 18-month
study (mean MARS score of 30) and
246 HORNE ETAL.
a further 92 (40%) participants reporting
that they were rarely non-adherent (mean
MARS score 29–29.75).
Tablet count data also revealed extremely
high levels of adherence, with a median
percentage of tablets taken of 99.5%.
However, of the 230 ASCOT participants
in this study, only 161 had tablet count data.
A further 26 participants had returned med-
ication inconsistently over the course of the
trial, such that it was unclear which returned
period of time. It was therefore not possible
to calculate a percentage medication taken
for this group. The remaining 43 partici-
pants did not return any medication over the
18 months of the study. It was hypothesized
that those whowere non-compliant in return-
ing their medication may also have been
non-compliant in taking their medication.
Indeed, a chi-square analysis revealed that
these two groups without pill-count data
were more likely to report non-adherence
on the MARS than those who did have
p50.05). This indicates that the tablet
count data was biased in favour of high
MARS adherence data was used in sub-
Level of Concordance Between
Patients’ Beliefs about their
Condition and Treatment and
the Medical Model
The percentage of participants matching and
differing from the medical model on each of
the IPQ-R and BMQ sub-scales are shownin
Fig. 1. The majority of participants held
views that were concordant with the medical
Symptom beliefs were those least concor-
dant with the medical model of hyperten-
sion; a third of patients attributed one or
more symptoms to their hypertension. As
shown in Fig. 2, the symptoms most often
associated with hypertension were fatigue
(9.1%), breathlessness (8.3%), dizziness
(6.1%), feeling flushed (5.7%) and head-
0 10 20
Percentage of patients concordant with the medical model
30405060 70 8090 100
Strong perceived need
Low concerns about
effective in controlling
Belief that hypertension
is a chronic condition
Belief that hypertension
FIG. 1. Level of concordance with the ‘medical model’ of hypertension.
CLINICAL TRIAL OF ANTIHYPERTENSIVES247
Relationship Between Concordance
with the Medical Model of Hypertension
and Reported Adherence to Medication
As shown in Fig. 3, those reporting lower
rates of adherence (mean MARS scores529)
reported significantly lower concordance
with the medical model of hypertension
than those reporting higher rates of adher-
ence (t(221)¼4.05, p50.001).
Differences Between ASCOT
and Non-trial Participants
As shown in the Table, independent sample
t-tests revealed no significant differences in
perceptions of hypertension and beliefs
about antihypertensives between those in
the ASCOT trial (n¼230) and those not
accepted for the trial (n¼106), with the
exception of their perceived impact of their
Fast heart rate
Pins and needles
Loss of strength
Loss of libido
Percentage of patients
FIG. 2. Percentage of patients attributing symptoms to their hypertension.
Mean concordance with the medical model
of hypertension (95% CI)
(mean MARS <29,
(mean MARS >=29,
FIG. 3. Relationship between concordance with the medical model of hypertension and adherence
248 HORNE ETAL.
p50.05), where those in the trial reported
a greater impact (M¼2.3, SD¼0.62) than
those in routine care (M¼2.1, SD¼0.60).
Self-reported adherence to medication
was very high both among those taking part
in ASCOT (29.3, SD¼1.5) and those not
accepted for ASCOT (29.2, SD¼1.8).
There was no significant difference in self-
reported adherence between these groups
This study explored levels of adherence in
hypertensive patients in relation to their
beliefs about their condition and their
Adherence rates (as assessed by self-report
and tablet count) were much higher than
those found in previous research.3–5There
are a number of possible explanations of this
finding. Initially, it was thought that the high
levels of adherence to medication might be
the result of patients receiving specialist care
and attention in a clinical trial. Enstrom
et al.,20for example, reported that patients’
blood pressure control deteriorated when
they left a clinical trial and returned to
routine care, attributing this to poorer adher-
ence when the specialist trial care ended. In
the present study, however, the patients who
had been screened for ASCOT but did not
meet the full entry criteria reported equally
high levels of adherence as those in the trial.
A second explanation is that patients who
volunteer for a clinical trial are atypically
adherent. This has implications for the
transferability of clinical trial findings to
‘real’ practice, where patients may not be
so adherent and the treatment effects found
in the trial may not be maintained. A
community-based study would be needed
to establish whether adherence levels are
indeed lower amongst patients in routine
Finally, it may be that previous estimates
of non-adherence to antihypertensives have
exaggerated the problem by using misleading
criteria for classifying patients as ‘adherent’
and ‘non-adherent’. Some studies have
failed to document the criteria on which
patients were classified ‘non-adherent’,21
whilst others have used somewhat extreme
criteria. For example, in one study, patients
were classified as non-adherent if they
reported having ‘ever’ forgotten to take
their medication.5If ASCOT and non-trial
patients in this study had been classified as
TABLE. A comparison of the ASCOT sample and the non-trial sample
Age in years (mean, SD)
Gender (number/percentage male)
IPQ Timeline (acute–chronic) (mean, SD)
IPQ Timeline (cyclical) (mean, SD)
IPQ Consequences (mean, SD)
IPQ Personal Control (mean, SD)
IPQ Treatment Control (mean, SD)
IPQ Coherence (mean, SD)
IPQ Emotional Representation (mean, SD)
IPQ Identity (mean no. symptoms/SD)
BMQ Necessity (mean, SD)
BMQ Concerns (mean, SD)
Baseline MARS (mean, SD)
Mean MARS over four time points (mean, SD)
Note: Results in bold indicate significant differences between the two samples.
CLINICAL TRIAL OF ANTIHYPERTENSIVES249
‘non-adherent’ using a similar criterion, the
baseline non-adherence rate would have
been 38%, in line with previous estimates.
However, not only would it have lacked face
validity to label those who reported ‘rarely’
not taking their medication ‘non-adherent’ it
would also have been inappropriate in this
study in light of the tablet count data indi-
cating very high levels of adherence.
In contrast to other studies,9,22,23partic-
ipants in this study were found to have illness
perceptions and treatment beliefs that were
broadly consistent with the medical model of
hypertension.The majority of
viewed their hypertension as a chronic,
through medication, and regarded their
medication as both necessary and safe. It is
perhaps unsurprising that patients willing to
participate in a medical trial would have a
medical model of their condition. A third of
patients, however, did deviate from the
medical model by attributing symptoms to
their condition. This is consistent with
research suggesting that patients assign
common everyday symptoms to hyperten-
sion in order to ‘make sense’ of their condi-
tion in light of their experience of other
Although there was little variation in
adherence to medication, the variation that
did exist was associated with concordance
with the medical model of hypertension.
Patients reporting lower adherence to anti-
hypertensives reported lower concordance
with the medical model.
Ideally, electronic monitoring of adherence
would have been used in this study. This is
considered to give the most accurate avail-
able assessment of levels of non-adherence.
However, it was not possible to obtain
funding for this equipment and self-report/
tablet count methods were used instead.
Self-report measures are often criticized for
overestimating levels of adherence as a result
of self-presentational bias.24However, care
was taken to limit self-presentational bias in
this study. The adherence questionnaire was
worded in a non-threatening manner, and
participants were given a range of response
options (‘always’ to ‘never’), thought to
increase the accuracy of adherence measure-
ment.25In addition, the tablet count con-
ducted on medication returned by ASCOT
participants confirmed high levels of adher-
ence to medication, indicating that 95% of
patients took over 80% of their medication,
which is considered to be the minimum
amount needed for treatment to be effec-
tive.26This figure should, however, be inter-
preted cautiously given that those who did
report non-adherence were significantly less
likely to return their medication, hence the
tablet count data may be biased.
Clinical Implications and
This study has demonstrated the importance
of eliciting patients’ illness and treatment
beliefs, with a relationship between the
degree of concordance with the medical
model of hypertension and reported adher-
ence to medication. Although this study
found limited variation in levels of adher-
ence, the findings support the view that
further studies of adherence in hypertension
should take patients’ perceptions of their
illness and treatment into account.
The unexpected findings of patients’
models of hypertension being highly concor-
dant with the medical model of hypertension
and the extremely high levels of adherence
may be the result of our sample being
atypical of hypertensive patients in general,
given that both groups were volunteers to a
clinical trial. There is now a need for a
community-based investigation to clarify the
generalizability of the adherence levels found
in this study.
Acknowledgements. This research was
funded by the NHS Executive South East
Project Grant Scheme (SEO 018). We would
also like to thank Natalie Lambert and
Richard Jenkins for data collection and
entry, the ASCOT nurses for introducing
250 HORNE ETAL.
the study to participants and collecting Download full-text
returned medication, Neil Poulter and the
Cardiovascular Studies Unit at Imperial
College for access to the medical data, and
not least to all participants who gave up their
time to take part in the study.
1. Chobanian AV, Bakris GL, Black HR, et al. The
seventh report of the Joint National Committee on
Prevention, Detection, Evaluation, and Treatment
of High Blood Pressure: the JNC 7 report. JAMA
2003; 289: 2560–72.
2. DiMatteo MR, Giordani PJ, Lepper HS, et al.
a meta-analysis. Med Care 2002; 40: 794–811.
3. Hekler EB, Lambert J, Leventhal E, et al.
Commonsense illness beliefs, adherence behaviors,
Americans. J Behav Med 2008; 31: 391–400.
4. World Health Organization. Adherence to long-term
therapies: evidence for action. Geneva: World Health
5. Patel RP, Taylor SD. Factors affecting medication
Pharmacother 2002; 36: 40–5.
6. Horne R. Adherence to medication: a review of
existing research. In: Myers L, Midence K, eds.
Adherence to treatment in medical conditions. London:
Harwood Academic, 1998: 285–310.
7. Horne R, Weinman J. Predicting treatment adher-
In: Myers L, Midence K, eds. Adherence to treatment
in medical conditions. London: Harwood Academic,
8. Royal Pharmaceutical Society of Great Britain.
From compliance to concordance: achieving shared
goals in medicine-taking. London: RPSGB, 1997.
9. Okken VS, Niemeijer MG, Dijkstra A, et al. The
effect of physical, social and psychological factors
on drug compliance in patients with mild hyper-
tension. Neth Heart J 2008; 16: 197–200.
10. Baumann LJ, Cameron LD, Zimmerman RS, et al.
Illness representations and matching labels with
symptoms. Health Psychol 1989; 8: 449–69.
11. Horne R, Weinman J. Self regulation and self
management in asthma: exploring the role of illness
perceptions and treatment beliefs in explaining
non-adherence to preventer medication. Psychol
Health 2002; 17: 17–32.
12. Horne R, Weinman J. Patients’ beliefs about pre-
scribed medicines and their role in adherence to
treatment in chronic physical illness. J Psychosom
Res 1999; 47: 555–67.
13. Miller LG, Hays RD. Measuring adherence to
antiretroviral medications in clinical trials. HIV
Clin Trials 2000; 1: 36–46.
14. Lucas GM, Chaisson RE, Moore RD. Highly active
antiretroviral therapy in a large urban clinic: risk
factors for virologic failure and adverse drug reac-
tions. Ann Intern Med 1999; 131: 81–7.
15. Dahlof B, Sever PS, Poulter NR, et al. ASCOT
investigators. Prevention of cardiovascular events
with an antihypertensive regimen of amlodipine
adding perindopril as required versus atenolol
adding bendroflumethiazide as required, in the
Blood Pressure Lowering Arm (ASCOT-BPLA): a
multicentre randomised controlled trial. Lancet
2005; 366: 895–906.
16. Moss-Morris R, Weinman J, Petrie K, et al. The
revised illness perception questionnaire (IPQ-R).
Psychol Health 2002; 17: 1–16.
17. Chen SL, Tsai JC, Lee WL. The impact of illness
perception on adherence to therapeutic regimens of
patients with hypertension in Taiwan. J Clin Nurs
2009; 18: 2234–44.
18. Ross S, Walker A, MacLeod MJ. Patient compli-
ance in hypertension: role of illness perceptions and
treatment beliefs. J Hum Hypertens 2004; 18:
19. Horne R, Weinman J, Hankins M. The Beliefs
about Medicines Questionnaire: the development
and evaluation of a new method for assessing the
cognitive representation of medication. Psychol
Health 1999; 14: 1–24.
20. Enstrom I, Pennert K, Lindholm LH. Durability of
improvement achieved in a clinical trial. Is compli-
ance an issue? J Fam Pract 2000; 49: 634–7.
21. Balazovjech I, Hnilica Jr P. Compliance with anti-
hypertensive treatment in consultation rooms for
hypertensive patients. J Hum Hypertens 1993; 7:
22. Weir MR, Maibach EW, Bakris GL, et al.
Implications of a health lifestyle and medication
analysis for improving hypertension control. Arch
Intern Med 2000; 160: 481–90.
23. Meyer D, Leventhal H, Gutmann M. Common-
sense models of illness: the example of hyperten-
sion. Health Psychol 1989; 4: 115–35.
24. Zeller A, Ramseier E, Teagtmeyer A, Battegay E.
Patients’ self-reported adherence to cardiovascular
medication using electronic monitors as compara-
tors. Hypertens Res 2008; 31: 2037–43.
25. Haynes RB, Taylor DW, Sackett DL, et al. Can
simple clinical measures detect patient noncompli-
ance? Hypertension 1980; 2: 757–64.
26. Guerrero D, Rudd P, Bryant-Kosling C, et al.
Antihypertensive medication-taking. Investigation
of a simple regimen. Am J Hypertens 1993; 6:
CLINICAL TRIAL OF ANTIHYPERTENSIVES251