Objective
To assess how much patients with long-term conditions value self-efficacy (i.e., confidence in their ability to manage their condition) compared with other health outcomes, including measures of quality of life, and process outcomes including access to General Practitioners.
Methods
Discrete Choice Experiment (DCE) set in UK community settings. Participants: 367 patients (mean age 57.5) living in the community with a wide range of self-defined long-term conditions. Main outcome measures: the relative value that individuals place on four specific outcomes, namely, self-efficacy, Health Related Quality of Life (HRQoL), access to General Practitioners, and level of isolation.
Results
Most responders completed their questionnaire in a consistent manner. Most valuations of outcomes were in the expected direction and were statistically significant. A substantial minority of responders exhibited counter-intuitive preferences. The existence of a significant constant in all models raised concerns about model misspecification. Nevertheless, all models showed that participants were willing to trade substantial reductions in their HRQoL for improvements in their self-efficacy.
Conclusions
The majority of patients with chronic conditions were able to complete the DCE questionnaires. However, the existence of counter-intuitive preferences and evidence of model misspecification require further investigation. These issues are largely overlooked in the health economics literature. Self-efficacy is an important outcome for this group and is not included explicitly in conventional HRQoL measures. This is potentially important where decisions are made on the basis of cost-effectiveness using Quality Adjusted Life Years as the metric. Exclusion of these outcomes may lead to the cost-effectiveness of these interventions being understated.
Recent policy has targeted self-care as a means to improve
patient outcomes and reduce costs [1,2]). Self-care has been
defined as care taken by individuals toward their own health and
well-being [2]. Interventions have been designed that support
individuals’ ability to self-care, for example, the Expert Patient
Programme (EPP) based on the chronic disease self-management
program developed in the United States by Lorig [3,4]. The
“EPP” aims to provide self-care support to any individual with a
chronic condition in England and is a generic, lay led, group
program involving six-weekly sessions. The EPP is designed to
enable participants to develop appropriate self-care skills including
patient’s “self-efficacy” [5]. The concept of self-efficacy refers
to a psychological state which relates to an individual’s confidence
that they can achieve some task (such as managing the
symptoms of their disorder or engage in regular exercise) [6–8],
and is one of the most important concepts in modern psychological
approaches to understanding health behavior [9]. Selfefficacy
is increasingly accepted as a mediating variable and an
outcome of self-management programs but this has become normative
without evidence of what patients most value in managing
long-term conditions.
Arecent trial [5] demonstrated that aUKversion of the chronic
disease self-management program was effective at improving selfefficacy,
and a cost-effectiveness analysis based on the same clinical
trial [10] generated Quality Adjusted Life-Year (QALY) gains
for these interventions using the EQ5D instrument. The EQ5D
instrument measures Health Related Quality of Life (HRQoL),
across five dimensions, namely mobility, ability to self-care, ability
to perform usual activities, level of pain/discomfort, and level of
anxiety/depression. The cost-effectiveness analysis concluded that
the EPP intervention was likely to provide a cost-effective alternative
to usual care in people with long-term conditions at commonly
used threshold values of a QALY.
Thus QALYs, often generated from EQ5D, have a commonly
expressed value [11] but they may not incorporate all the outcomes
that are of interest. In contrast, while self-efficacy is
undoubtedly an outcome of interest (at least for practitioners and
researchers), we have no knowledge of whether self-efficacy is
“of value” per se, or indeed what that value might be. This leads
to problems of interpretation as decision-makers cannot assess
the relative merits of self-efficacy compared with HRQoL. In the
example above, the EPP was deemed to be likely to be costeffective
based solely on the QALY, although there was a large
degree of uncertainty around the decision. Valuing self-efficacy
(or other relevant outcomes) in terms of QALYs gives decisionmakers
additional information and can be incorporated into the
cost-effectiveness analysis and could reduce the uncertainty
around a decision.
Discrete Choice experiments (DCEs) are one method of either
expanding the measure of outcome or incorporating factors
other than health outcomes [12]. DCEs are based on the premise
that the benefits associated with health-care interventions can be
expressed in terms of the “attributes or characteristics” of that
intervention [13] and the “attributes or characteristics” of the
person valuing them [14,15].
Process outcomes may be important in the evaluation of
health-care technologies, and DCE enables the relative values of
these outcomes to be assessed [12]. For example, speed of access
to health care, who provides that health care and where it is
provided, are often considered as important aspects of health
care that may not be captured by a measure of HRQoL [12].DCEs have been used frequently in the health economics literature
to estimate preferences in miscarriage management [16],
management of prostate cancer [17], as well as in a variety of
other conditions and in numerous settings [18–29]. However,
DCEs have not been previously employed to investigate the
trade-off between HRQoL, psychological outcome measures
such as self-efficacy, and social outcomes such as isolation and/or
process outcomes such as General Practitioner (GP) access. One
advantage of the DCE methodology is that it enables the individual
to simultaneously compare and value a number of different
attributes. This may have the added benefit of limiting the
“salience” problem identified with standard gamble techniques,
where there is discordance between stated preferences and actual
choices because of the experiment focusing the individuals’ concentration
on one specific attribute [30], although it is acknowledged
that responders may still not focus on the true opportunity
cost of their choice [31].
This article describes a DCE conducted to examine the relationship
between HRQoL and other outcomes which may be of
relevance to patients with long-term conditions. In addition, the
estimation of rates of substitution between the QALY and selfefficacy
enables decision-makers to include these outcomes in
their assessment of cost-effectiveness.
Although self-efficacy has been identified as important to patients
with chronic conditions who have been exposed to interventions
designed to impact on levels of self-efficacy, it is likely that in
different patient groups other outcomes would be valued and
traded for HRQoL. These “important” outcomes should be identified
before commencing the study and appropriate techniques
should be used to ascertain the rate of substitution between these
outcomes and HRQoL.
The lack of concern in the health economics literature surrounding
counterintuitive preferences and the possibility of
model misspecification is disturbing. While these methodological
caveats should not preclude the use of DCEs in health economic
evaluation, these issues require further examination and suggest
that results of DCEs where these issues could be a factor should
be interpreted cautiously.
