effective than single techniques, especially if they simultan eously targeted education,
behaviour change and emotional responses. Techniques targeting adherence behaviour
included changing drug packaging, simplifying dose instructions, mailed reminders and
skills development approaches. Third, adherence interventions were more effective for
some conditions, especially diabetes, asthma, cancer and hypertension, suggesting that
it may be easier to increase adherence for some patients than others. Finally, the
researchers noted that a broader approach to identifying outcome measures could be
beneficial. For example, as well as boosting adherence, interventions may affect patient
satisfaction, patient understanding and quality of life. These may be important targets
in themselves. This suggestion links to a more general question about the evaluation
of health care interventions, that is, who decides what are the appropriate outcome
measures? While physical health is very important, is it always the most important
outcome? Consider the definition of health we began with in Chapter 1.
In a second meta-analysis, Haynes et al.(2005) examined the outcomes of randomized
controlled trials, which measured adherence to medication and included a clinical or
health outcome (that is, whether people in the intervention condition also showed greater
health benefit). For short-term prescriptions they found that 4 of 9 interventions (44 per
cent) had an effect on both adherence and at least one clinical outcome while, for longer-
term treatment, 26 of 58 (45 per cent) led to improvements in adherence but only 18
interventions (31 per cent) led to improvement in at least one clinical outcome. The
researchers concluded that for short-term drug treatments counselling, written inform -
ation and a personal phone call could boost adherence, but for long-term treatments,
no particular technique and only some complex interventions led to improvements in
health outcomes. Those that were successful in improving health included combina-
tions of more convenient care, providing information, counselling, reminders, self-
monitoring, reinforcement, family therapy, psychological therapy, crisis intervention,
telephone follow-up and additional supervision. Thus it is challenging to improve
adherence to long-term medication to the extent that such improvements impact on
clinical outcomes. However, while health improvement is a critical outcome it may not
be the only one and it is a challenging target for interventions designed to change behaviour
because even successful interventions (e.g. those generating increases of drug adherence
using objective measures) may not make a difference to health.
Although it is sometimes challenging to promote improved adherence, health care
professionals can maximize adherence by improving patients’ understanding, recall and
satisfaction. Patients must understand advice before they can follow it and they must
remember it beyond the consultation if it is to shape behaviour. Both understanding
and recall are associated with patient satisfaction. Consequently, combining these
factors, Ley proposed the model of adherence shown in Figure 10.1. Understanding
is correlated with adherence (r= 0.36) (Ley, 1988) and in Chapter 8 we discussed
how health care professionals can enhance patient understanding. Even when
information is understood it may be forgotten. For example, in an early study, patients
were found to have forgotten around half of the verbal instructions given to them,
after only 5 minutes (Ley, 1973). Again, we noted in Chapter 8 how recall could be
improved (e.g. see our consideration of logical order, explicit categorization, specific
advice and emphasizing and repeating important points).
In a review of reviews, van Dulmen et al.(2007) concluded that adherence can also
be promoted by simplifying treatments, for example, reducing the numbers of pills or
236 RELATING TO PATIENTS