The healthy adherer-effect.
Reply to: Simpson SH, Eurich DT, Majumdar SR, Padwal RS, Tsuyuki RT,
Varney J, et al. A meta-analysis of the association between adherence to
drug therapy and mortality. Bmj 2006;333(7557):15.
The healthy adherer-effect.
BLG van Wijk1, B.M.A. van Wieren – de Wijer1, J Urquhart2, E.R.
Heerdink1, A de Boer1, O.H. Klungel1.
1Department of Pharmacoepidemiology & Pharmacotherapy, Utrecht
Institute for Pharmaceutical Sciences, Utrecht University
2Chief Scientist, AARDEX Ltd; Emeritus Professor of Pharmacoepidemiology,
Simpson and colleagues demonstrate in their thoughtful meta-analysis that
patients adherent to medication regimens have better health outcomes than
those who are non-adherent, regardless of whether the drug consumed was a
placebo or contained an active ingredient.  The authors suggest that
patients more adherent to medication intake also have a more overall
healthy behaviour. To verify this suggestion, we requested data from the
PHARMO-database, a Dutch prescription database, currently covering more
than 2,000,000 patients. We identified a sub sample of patients taking
blood pressure lowering medication between 1991 and 2003 for which
detailed information on lifestyle was available. Pharmacy records can be
used to assess adherence to medication in an easy and non-invasive way,
although generally overestimating actual drug intake. We calculated the
refill rate for antihypertensive drugs in the year preceding the date for
which lifestyle information was available. In a number of age and gender
adjusted logistic regression models, we analysed the association between
various indicators of healthy behaviour that are known to be associated
with cardiovascular outcomes  and suboptimal adherence.
A total of
4,857 patients showed an average refill rate of 93.8% (+/- 11.4), while
445 patients (9.2%) had a refill rate below 80%. Infrequent exercising (OR
1.67 [95% CI: 1.02-2.70]) was associated with non-adherence. A trend
towards worse adherence was seen for patients who did not follow a low
salt diet (OR 1.29 [95% CI: 0.70-2.33). A body mass index above 30 kg/m2
was not associated with non-adherence (OR 0.97 [95%CI: 0.75-1.26).
Compared to patients who consumed of 1 or 2 units of alcohol a day,
complete abstinence (OR 1.83 [95%CI: 0.98-3.39]) and more frequent
drinking (OR 1.73 95% CI: 1.01- 2.93) were associated with non-adherence.
Patients who stopped smoking were as adherent as patients who never smoked
(OR 0.99 [95% CI: 0.79-1.24]) in contrast to smokers, who were
significantly more often non-adherent (OR 1.36 [95% CI: 1.02-1.80]).
Although we probably underestimated the prevalence of non-adherence by
using pharmacy records, thereby inevitably misclassifying non-adherent
patients, introducing bias towards the null, the results at least suggest
that some indicators of healthy behaviour could indeed be associated with
adequate medication refilling. Next to non-adherence with other
medications, as suggested by Urquhart , part of the positive
association between adherence and outcomes observed by Simpson and
colleagues  may be explained by the absence of information on
indicators of healthy behaviour.
1. Simpson SH, Eurich DT, Majumdar SR, Padwal RS, Tsuyuki RT, Varney
J, Johnson JA. A meta-analysis of the association between adherence to
drug therapy and mortality. Bmj. 2006 Jul 1;333(7557):15.
2. Osterberg L, Blaschke T. Adherence to medication. N Engl J Med. 2005
3 1: Yusuf S, Hawken S, Ounpuu S, Dans T, Avezum A, Lanas F, McQueen M,
Budaj A, Pais P, Varigos J, Lisheng L; INTERHEART Study Investigators.
Effect of potentially modifiable risk factors associated with myocardial
infarction in 52 countries (the INTERHEART study): case-control study.
Lancet. 2004 Sep 11-17;364(9438):937-52.
4 Urquhart J. A different view of the 'effect' of adherence to dosing
regimens for trial placebos. Posted on:
Competing interests: No competing interests