-
PA57CH14-Campbell ARI 26 November 2016 13:14
Changing Provider Behaviorin the Context of ChronicDisease
Management: Focuson Clinical InertiaKim L. Lavoie,1,2 Joshua A.
Rash,3
and Tavis S. Campbell31Department of Psychology, University of
Quebec at Montreal (UQAM), Montreal, QuebecH3C 3P8, Canada2Montreal
Behavioural Medicine Centre (MBMC), Research Centre, Hôpital du
Sacré-Coeurde Montréal, Montreal, Quebec H2J 1C5,
Canada3Department of Psychology, University of Calgary, Calgary,
Alberta T2N 1N4, Canada;email: [email protected]
Annu. Rev. Pharmacol. Toxicol. 2017. 57:263–83
First published online as a Review in Advance onSeptember 7,
2016
The Annual Review of Pharmacology and Toxicologyis online at
pharmtox.annualreviews.org
This article’s doi:10.1146/annurev-pharmtox-010716-104952
Copyright c© 2017 by Annual Reviews.All rights reserved
Keywords
clinical inertia, therapeutic inertia, diagnostic inertia,
clinical practiceguidelines, evidence-based medicine
Abstract
Widespread acceptance of evidence-based medicine has led to the
prolifer-ation of clinical practice guidelines as the primary mode
of communicatingcurrent best practices across a range of chronic
diseases. Despite overwhelm-ing evidence supporting the benefits of
their use, there is a long history ofpoor uptake by providers.
Nonadherence to clinical practice guidelines isreferred to as
clinical inertia and represents provider failure to initiate
orintensify treatment despite a clear indication to do so. Here we
review ev-idence for the ubiquity of clinical inertia across a
variety of chronic healthconditions, as well as the organizational
and system, patient, and providerfactors that serve to maintain it.
Limitations are highlighted in the emergingliterature examining
interventions to reduce clinical inertia. An evidence-based
framework to address these limitations is proposed that uses
behav-ior change theory and advocates for shared decision making
and enhancedguideline development and dissemination.
263
Click here to view this article'sonline features:
• Download figures as PPT slides• Navigate linked references•
Download citations• Explore related articles• Search keywords
ANNUAL REVIEWS Further
Ann
u. R
ev. P
harm
acol
. Tox
icol
. 201
7.57
:263
-283
. Dow
nloa
ded
from
ww
w.a
nnua
lrev
iew
s.or
g A
cces
s pr
ovid
ed b
y C
onco
rdia
Uni
vers
ity -
Mon
trea
l on
07/1
0/17
. For
per
sona
l use
onl
y.
http://www.annualreviews.org/doi/full/10.1146/annurev-pharmtox-010716-104952
-
PA57CH14-Campbell ARI 26 November 2016 13:14
Evidence-basedmedicine (EBM): asystematic applicationof the
scientificmethod into health-care practice with thegoal of
providingoptimal care topatients
Clinical practiceguidelines
(CPGs):systematicallydevelopedrecommendationsinformed by a reviewof
evidence andassessment of thebenefits and harms ofalternative
careoptions
Clinical inertia:failure to initiate orintensify
treatmentdespite a clearindication andrecognition to do so
Evidence-based medicine (EBM), whose origins date back to the
mid-nineteenth century (1), is thesystematic application of the
scientific method into health-care practice with the goal of
providingoptimal clinical care to patients (2). Inherent in this
definition is the expectation of explicit andjudicious use of
current best evidence to guide clinical decision making. EBM is a
constantlyevolving process; as more evidence becomes available, old
tests and treatments are replaced withmore accurate, powerful,
effective, and safer ones. Clinical practice guidelines (CPGs) have
becomethe primary mode of communicating current best practices
across a range of clinical disorders (3).The ultimate goal of
practice guidelines is to facilitate the translation of the most
recent evidenceinto practice to improve patient care and outcomes
(4).
Despite their widespread availability and strong evidence
supporting the benefits of their use(5–7), there is a long history
of poor uptake of CPGs by providers, with many studies reporting
ratesof nonadherence at or exceeding 50% (8–10). Rates of provider
nonadherence to guidelines are saidto be responsible for up to 80%
of myocardial infarctions and strokes in the context of
suboptimallytreated hypertension, diabetes, and dyslipidemia (11).
Practice guidelines have been criticized forbeing overly
simplistic, impractical, biased, and not broadly applicable and for
representing a threatto professional autonomy and the
provider-patient relationship (4, 12). However, the intentionof
CPGs is not to provide a black and white, cookbook approach to
diagnosis and treatment, butrather to facilitate a bottom-up
approach that integrates the best external evidence with
clinicalexpertise that considers individual patients’ goals,
values, and preferences when making decisionsabout care (1).
Provider nonadherence to CPGs is increasingly referred to as
clinical inertia. This term wasinitially introduced by Phillips et
al. (13) in 2001 and defined as provider failure to initiate
orintensify treatment despite a clear indication and recognition of
the need to do so. Other termshave been used to describe the same
behavior, including therapeutic inertia, physician inertia,and
diagnostic inertia (14–16), but they are generally synonymous and
reflect conscious providerinaction in the face of available and
explicit evidence-based guidelines and recognition of the needto
act.
In an era of chronic disease that demands the practice of EBM to
achieve optimal clinicaloutcomes, overcoming the problem of
clinical inertia is imperative. Here we review current def-initions
of clinical inertia and summarize its prevalence and impact across
a range of chronicdiseases. We also review barriers to provider
adherence to CPGs and summarize the efficacy ofprovider-focused
interventions. Finally, we propose the adoption of a theoretical
framework forunderstanding and overcoming the problem of clinical
inertia that is grounded in health behaviorchange theory and can be
used to improve future implementation strategies.
DEFINING CLINICAL INERTIA
To consider provider behavior as reflecting clinical inertia, at
least two conditions must be met:(a) The patient fails to meet
clearly defined and measurable treatment targets, and (b) the
patientfails to receive appropriate intensification of therapy
within a defined and reasonable period oftime (11). Clinical
inertia may also apply to the failure to stop or reduce therapy
that may nolonger be needed; although this side of clinical inertia
also has important clinical consequences(17), it has received far
less attention.
One issue regarding the definition of clinical inertia is how to
distinguish true clinical inertiafrom what may in fact be
appropriate clinical inertia, which reflects reasonable decisions
not tointensify treatment despite the available evidence. This may
occur with more complex cases (e.g.,a frail elderly patient with
diabetes and hypertension) or when age, comorbidities,
polypharmacy,and potential adverse drug reactions may render
guideline-recommended therapies inappropriate
264 Lavoie · Rash · Campbell
Ann
u. R
ev. P
harm
acol
. Tox
icol
. 201
7.57
:263
-283
. Dow
nloa
ded
from
ww
w.a
nnua
lrev
iew
s.or
g A
cces
s pr
ovid
ed b
y C
onco
rdia
Uni
vers
ity -
Mon
trea
l on
07/1
0/17
. For
per
sona
l use
onl
y.
-
PA57CH14-Campbell ARI 26 November 2016 13:14
(13) or unsafe (18). More recent definitions of clinical inertia
take these concerns into account(19) and suggest that true clinical
inertia occurs only when the following criteria are met:
Theprovider (a) is aware of the existence of implicit or explicit
guidelines; (b) believes that this guidelineapplies to the patient;
(c) has resources available to apply the guideline; and (d ) does
not followthe guideline despite awareness, belief, and available
resources to do so. Although this may bethe most appropriate and
balanced definition of clinical inertia, most studies to date have
definedclinical inertia based on observations of providers’ failure
to intensify treatment in the context ofpatients not meeting
measurable therapeutic targets. However, without systematically
consultingclinical data such as medical charts, it is impossible to
verify the extent to which clinical inertia maybe appropriate
(e.g., in the context of severe comorbidities) or the result of
patient nonadherence.
THE PROBLEM OF CLINICAL INERTIA: PREVALENCE AND IMPACT
Evidence for clinical inertia comes from epidemiological
studies, direct observation, or an analysisof provider behavior
during clinical visits. The challenge in reporting prevalence rates
of clinicalinertia is that it has been defined inconsistently and
measured imprecisely across studies. In general,it is easier to
measure clinical inertia in relation to conditions where treatment
targets, intensifi-cations, and timelines are well defined, such as
diabetes, hypertension, and dyslipidemia. Rates ofclinical inertia
have been studied most extensively within the context of these
conditions and to alesser extent in the context of asthma and
chronic obstructive pulmonary disease (COPD) and aretypically
quantified as instances when providers fail to intensify treatment
among diagnosed pa-tients who are not meeting targets. These cases
do not consider the reasons for provider inaction,some of which may
be appropriate, representing a major limitation in this area of
research.
Diabetes
There is little question that the timely initiation and
intensification of insulin, glucose-loweringmedication, or both in
diabetes is associated with clinically relevant benefits, including
improvedglycemic control and a reduction in microvascular
complications (20–23). Despite this, a reviewby Phillips et al.
(13) revealed that in the United States, only 65% of patients were
diagnosedaccurately, and among those, only 73% were given
pharmacological therapy. It is not surprising,therefore, that
hemoglobin A1c values met American Diabetes Association targets in
only 7% ofpatients. A more recent study using a retrospective
cohort of more than 81,000 patients with type 2diabetes from the
United Kingdom reported that over 50% of patients with poor
glycemic controldid not receive intensification of oral
antidiabetic medication within 7 years of treatment (24).Finally, a
study from Canada using administrative data from more than 80,000
patients compared4-month drug intensification by specialists and
general practitioners among 2,652 matched caseswith uncontrolled
diabetes (25). Although specialists intensified treatment in a
greater proportionof cases (45.1%) than general practitioners did
(37.4%), overall rates were less than 50% (26).
Hypertension
Similar to diabetes, clinical inertia is prevalent in the
management of hypertension. The reviewby Phillips et al. (13)
indicated that in the United States, only 69% of patients were
diagnosedaccurately, and among those, just over half (53%) were
given pharmacological therapy. Thisexplains why only about 45% of
patients had adequate blood pressure (BP) control. In a morerecent
study of more than 21,000 respondents of a nationally
representative survey in five Europeancountries and the United
States, Wolf-Maier et al. (27) reported that among those with
poorly
www.annualreviews.org • Provider Behavior in Disease Management
265
Ann
u. R
ev. P
harm
acol
. Tox
icol
. 201
7.57
:263
-283
. Dow
nloa
ded
from
ww
w.a
nnua
lrev
iew
s.or
g A
cces
s pr
ovid
ed b
y C
onco
rdia
Uni
vers
ity -
Mon
trea
l on
07/1
0/17
. For
per
sona
l use
onl
y.
-
PA57CH14-Campbell ARI 26 November 2016 13:14
controlled BP, only 14–26% of patients in Europe and 32% of
patients in the United States receivedtreatment intensification.
Finally, in a study conducted within Veteran Affairs (VA) primary
careclinics, clinical inertia was identified in 66% of cases.
Moreover, among victims of clinical inertia,nearly one-quarter had
no follow-up appointment scheduled, and nearly 77% of those who
didsee their provider did so after a delay of 45 days (range: 29–78
days) (28).
Suboptimal management of hypertension has been shown to have a
major impact on BP control.For example, in a sample of more than
7,000 hypertensive patients from the United States whowere followed
for an average of 6.4 appointments over the course of a year,
Okonofua et al. (16)calculated that clinical inertia accounted for
19% of the variance in BP control. Furthermore, theauthors
estimated that BP could be controlled in 20% more patients (an
increase from 45.1% to65.9%) in one year if clinical inertia could
be reduced by 50%.
Dyslipidemia
Clinical inertia is also common in the context of dyslipidemia,
for which the review by Phillipset al. (13) revealed that only 47%
of patients in the United States were diagnosed accurately,
andamong those, the minority (17–23%) were given pharmacological
therapy. This helps to explainwhy low-density lipoprotein
cholesterol (LDL-C) levels reach National Cholesterol
EducationProgram targets in only 14–38% of patients. More recently,
clinical inertia in dyslipidemia wasevaluated in a sample of 22,888
patients with cardiovascular disease (CVD) (29). Only 32% ofthe
6,538 patients who failed to meet treatment targets (LDL-C ≥ 100
mg/dL) received intensi-fication of cholesterol-lowering therapy
within 45 days of their elevated laboratory test. Finally,an
18-month, retrospective cohort study (30) used medical records from
253,238 members of theKaiser Permanente Medical Care Program who
had poor control of dyslipidemia, diabetes, orBP to quantify
clinical inertia, which was defined as the failure of the provider
to intensify phar-macotherapy within 6 months. Clinical inertia was
observed in 41.4% of patients with elevatedLDL-C levels, as well as
30.3% of patients with elevated hemoglobin A1c levels, 28.8%
withelevated systolic BP, and 17.6% with elevated diastolic BP.
Asthma
In developed countries, asthma is often the most prevalent
chronic disease in children and one ofthe most common conditions
affecting adults (31, 32). Despite the availability of guidelines
forthe treatment of asthma and robust evidence that following
guideline recommendations improvesoutcomes (33, 34), provider
adherence to CPGs to manage asthma is poor. For example, a
retro-spective study of asthma care delivered to 345 patients at a
tertiary adult emergency department(ED) in Canada reported 69.6%
overall compliance with guidelines (35). Controller (i.e.,
inhaledcorticosteroid) use was prescribed in only one-third of
children and adults in the ED and on dis-charge. Studies have also
shown that in the nonacute care setting, few physicians prescribe
ongoingdaily controller medication or written self-management
plans, even in adults and children with arecent acute care visit
for asthma (36). Moreover, even when they are prescribed, the
cumulativeduration of available prescriptions covers less than 50%
of the follow-up period (37).
One possible explanation for these high rates of clinical
inertia may be diagnostic inertia.Lougheed et al. (38) assessed ED
management of asthma in 2,671 children and 2,078 adultstreated at
16 Ontario hospitals by means of questionnaires and chart reviews.
Objective measuresof airflow rate were documented in only 27.2% of
pediatric visits and 44.3% of adult visits.Given that CPGs make
specific recommendations about treatment intensification based on
lungfunction (39, 40), the failure to measure this at the point of
care will make it difficult to implementguideline-recommended
therapy.
266 Lavoie · Rash · Campbell
Ann
u. R
ev. P
harm
acol
. Tox
icol
. 201
7.57
:263
-283
. Dow
nloa
ded
from
ww
w.a
nnua
lrev
iew
s.or
g A
cces
s pr
ovid
ed b
y C
onco
rdia
Uni
vers
ity -
Mon
trea
l on
07/1
0/17
. For
per
sona
l use
onl
y.
-
PA57CH14-Campbell ARI 26 November 2016 13:14
Chronic Obstructive Pulmonary Disease
Chronic obstructive pulmonary disease (COPD) affects 65 million
people worldwide (41) andis the third-leading cause of death in the
United States after CVD and cancer (42). Althoughstudied less
extensively than in other diseases, diagnostic and clinical inertia
in COPD are alsoprevalent concerns. A comprehensive review by Cooke
et al. (43) in 2012 reports that in 2002, only10 million adults in
the United States were diagnosed with COPD, despite Third National
Healthand Nutrition Examination Survey estimates that 24 million
adults had impaired lung function.The most common reason for the
underdiagnosis of COPD was the lack of objective lung
functiontesting (spirometry). One survey of primary care practices
revealed that despite 66% of providersowning a spirometer, 38% said
they were unfamiliar with the test and 34% said they were
nottrained to administer or interpret it (44).
With regards to clinical inertia, Cooke et al. (43) reported
that 23–38% of COPD patients fail toreceive any
guideline-recommended drug therapy (45–47). However, these rates
vary dependingon disease stage, with one study reporting higher
rates of treatment (81%) among patients withsevere COPD compared to
those with moderate (72%) and mild (44%) disease (48). Amongthose
receiving treatment, one study conducted among patients seen at a
university-based familymedicine clinic reported that treatment
intensity is often suboptimal, with only 55% of patientsreceiving
stage-recommended therapy (49). Specifically, Chavez & Shokar
(49) reported thatonly 22%, 5%, 28%, and 13% of patients with mild,
moderate, severe, and very severe COPD,respectively, were being
treated at stage-recommended levels.
FACTORS ASSOCIATED WITH CLINICAL INERTIA
To address the problem of clinical inertia, factors that explain
its high prevalence must be eluci-dated. Reasons for clinical
inertia involve a complex interaction between three types of
factors:organizational and system, patient, and provider factors,
for which previous reports have estimatedtheir relative
contributions at 20%, 30%, and 50%, respectively (11) (see Figure
1). Reviews (4,11, 13, 15, 50) and textbooks (19) that detail these
factors have been published, so we considerthem only briefly here
with an emphasis on modifiable, provider-related factors.
Organizational and System Factors
Time constraints are one of the most cited organizational and
system-related factors associatedwith clinical inertia. Providers
often have several competing demands [e.g., high patient
volumes,teaching and research responsibilities, acquiring
continuing medical education (CME) credits,office management, staff
supervision] (13, 14, 51) that may interfere with the ability to
keep upwith constantly evolving clinical guidelines and prevent the
thorough assessment, diagnosis, andtimely provision of treatment
initiation or intensification (52). This may be complicated
furtherby the lack of available resources to implement
guideline-recommended therapy (e.g., limited sup-port staff,
diagnostic equipment, access to laboratory services, office space),
as well as inadequatereimbursement for implementing
guideline-recommended therapy (4). Factors associated with
thepractice setting (e.g., primary care, inner-city or rural
setting, lack of or limited access to multidis-ciplinary expertise
or specialists) may also contribute to clinical inertia (50). For
example, someresearchers have argued that the lack of availability
of multidisciplinary team–based care, whichis often the case in
primary care, may be an important factor associated with increased
clinicalinertia in primary (relative to tertiary) care settings
(50, 53–55). Finally, access to care, prescriptiondrugs, or
insurance (which may also be considered a patient factor) is an
important contributor to
www.annualreviews.org • Provider Behavior in Disease Management
267
Ann
u. R
ev. P
harm
acol
. Tox
icol
. 201
7.57
:263
-283
. Dow
nloa
ded
from
ww
w.a
nnua
lrev
iew
s.or
g A
cces
s pr
ovid
ed b
y C
onco
rdia
Uni
vers
ity -
Mon
trea
l on
07/1
0/17
. For
per
sona
l use
onl
y.
-
PA57CH14-Campbell ARI 26 November 2016 13:14
Organizational andsystem factors Provider factors Patient
factors
Time constraints• High patient volumes• Competing demands
Knowledge• Insufficient awareness or
familiarization with guidelines• Large volume of guidelines•
Inaccessibility of guidelines
Demographics• Older age, female sex, socioeconomic
and cultural characteristics, and lowhealth literacy and
education
Medical history• ComorbiditiesAgreement with and
applicability
of guidelines• Uncertainty about applicability• Uncertainty
about implementation• Uncertainty about accuracy or
consistency of risk factor values
Beliefs, attitudes, and preferences• Unwillingness to accept
diagnosis
and/or take medication• Preference for or against treatment
Resources• Lack of support staff and equipment• Inadequate or
poor reimbursement
Setting• Primary care, inner-city, or rural• No access to
multidisciplinary team
Access to care• Poor access to care or inadequate
(or absent) insurance coverage
Cognitive biases• Overestimating care provided• Underestimating
treatment needed
Self-efficacy• Lack of confidence in
ability to enact guidelinesTreatment adherence
• Patient beliefs about risks and benefitsof treatment
• Provider perceptions of nonadherencemay create negative
expectancy biasMotivation
• Overcoming old habits and routines• Readiness to make a
change
Lifestyle factors
• Smoking, poor diet, physical inactivity,and alcohol use
Figure 1Organizational and system, provider, and patient factors
associated with clinical inertia.
clinical inertia. In one study, 38% of patients with COPD
reported that insurance-related issueslimited access to
prescription drugs and 14% reported limited access to physician
office visits (48).Furthermore, 58–67% of providers reported that
insurance coverage for needed treatment wasinadequate or
unreasonable (48). However, it is noteworthy that clinical inertia
is common in VAhospitals and in countries such as Canada, where
medication costs may be less of an issue (37, 56).
Patient Factors
Patient-related factors may also underlie clinical inertia. One
such factor involves patient demo-graphics, including older age,
female sex, and socioeconomic or cultural background (which
mayundermine health literacy) (50). For example, in a
cross-sectional retrospective study of 1,729medical and insurance
claims, Nau & Mallya (57) reported that among patients with
diabetes,men were more likely than women to receive lipid tests
(82.4% versus 79.4%) and lipid-loweringmedication (45.5% versus
33.2%). Furthermore, patients with more education and better
healthliteracy appear less likely to experience clinical inertia
(50).
Another factor relates to patients’ medical history. Many
studies have indicated that treatmentis less likely to be initiated
or intensified if patients have complex comorbidities (e.g., a
psychiatricor neurological disorder, substance abuse, terminal
illness) because this may raise questions aboutthe applicability or
appropriateness of existing guidelines (50).
Patient-related beliefs, attitudes, and preferences have also
been linked to clinical inertia. Forexample, clinical inertia may
reflect a patient’s unwillingness to accept their diagnosis or take
a
268 Lavoie · Rash · Campbell
Ann
u. R
ev. P
harm
acol
. Tox
icol
. 201
7.57
:263
-283
. Dow
nloa
ded
from
ww
w.a
nnua
lrev
iew
s.or
g A
cces
s pr
ovid
ed b
y C
onco
rdia
Uni
vers
ity -
Mon
trea
l on
07/1
0/17
. For
per
sona
l use
onl
y.
-
PA57CH14-Campbell ARI 26 November 2016 13:14
Cognitive biases:systematic deviationsfrom standards
injudgment
Self-efficacy: beliefor confidence in theability to
implementguidelines and enactmeaningful change
medication to manage what they may experience as an asymptomatic
disease (e.g., hypertensionand dyslipidemia) (50). EBM emphasizes
patient preferences, and in some cases, patients mayopt for
lifestyle modifications (e.g., dietary changes or increases in
physical activity) prior to theinitiation or intensification of
medications (58).
Patient nonadherence, the reasons for which are multifactorial
(e.g., medication cost, diseasenonacceptance, fear of side effects,
poor outcome expectancies, forgetfulness) and exacerbated
byprovider factors [e.g., poor communication skills (14)], may also
contribute to clinical inertia. Forexample, medication adherence
was found to predict 3-year treatment intensification in a cohort
of2,065 insured patients with type 2 diabetes newly started on
hypoglycemic therapy (59). Patients inthe lowest quartile of
adherence were less likely to have their medication appropriately
intensifiedthan patients in the highest quartile (27% versus 37%),
which equated to 53-fold lower oddsof treatment intensification
after having an elevated hemoglobin A1c. Established or
suspectedpatient nonadherence by providers may also contribute to
clinical inertia by creating a negativeexpectancy bias (50). In
these cases, providers may fail to provide guideline-recommended
therapyowing to low expectations of adherence and perceived
helplessness to change patient behavior.Some researchers have
argued that clinical inertia and lack of patient adherence go hand
in handand that this represents a shared failure to give preference
to the long-term benefit of treatmentintensification (60).
Finally, lifestyle factors (e.g., smoking, poor diet, physical
inactivity), by virtue of raising thebar for achieving clinical
targets, may also be linked to clinical inertia (14). For example,
theEUROASPIRE study reported that, despite treatment
intensification, patients with coronaryheart disease failed to
achieve BP targets, and nearly 50% of patients remained above
target lipidlevels 6 months after percutaneous intervention,
coronary artery bypass graft, or hospitalizationfor acute ischemia
or myocardial infarction (61). This study further revealed
concomitant increasesin obesity (from 25% to 38%) and diabetes
(from 17% to 28%) between EUROASPIRE I andIII, suggesting an
important role for lifestyle factors in explaining the treatment
failures.
Provider Factors
The strongest contributors to clinical inertia are factors
related to the provider and have beenthe most intensely studied (4,
11, 13, 28, 50). In general, five factors have been identified: (a)
lackof knowledge or awareness of clinical guidelines, (b) lack of
agreement with guidelines or theirapplicability, (c) cognitive
biases, (d ) motivational factors, and (e) low self-efficacy to
implementguidelines.
Lack of knowledge or awareness. Lack of awareness or familiarity
with evidence-based guide-lines for chronic disease management has
been reported as a major contributor of clinical inertia(4, 50, 52,
62). In a review of 46 surveys, Cabana et al. (4) observed that
lack of awareness wasreported as a barrier to guideline
implementation in a median of 54.5% of respondents. Similarly,41%
of respondents had not heard of nationally endorsed BP guidelines
in a US survey of pri-mary care physicians (63). In the context of
COPD, only about half of primary care physicians arereportedly
aware of CPGs (48), and only 25% said they used them for clinical
decision making(64). Recent reviews continue to note this as an
important barrier (65), which may be exacer-bated by the large
number of guidelines and the time required to keep them constantly
updated(62).
Lack of agreement and applicability. Another important
provider-related factor is lack ofagreement with guidelines or
their applicability to certain patients. Multiple reasons for
this
www.annualreviews.org • Provider Behavior in Disease Management
269
Ann
u. R
ev. P
harm
acol
. Tox
icol
. 201
7.57
:263
-283
. Dow
nloa
ded
from
ww
w.a
nnua
lrev
iew
s.or
g A
cces
s pr
ovid
ed b
y C
onco
rdia
Uni
vers
ity -
Mon
trea
l on
07/1
0/17
. For
per
sona
l use
onl
y.
-
PA57CH14-Campbell ARI 26 November 2016 13:14
problem have been identified, including doubting the credibility
of the evidence, doubting thatthe benefits of therapy outweigh the
risks, the perception that guidelines would reduce
providerautonomy, and the belief that guidelines undermine the
provider-patient relationship by reduc-ing patient choice (4).
Guidelines have also been criticized for being overly simplified,
leadingto disagreement about their applicability to individual
patients or populations (4, 50). This lat-ter claim is not entirely
unfounded, given that guidelines are typically informed by
randomizedcontrolled trials with strict inclusion or exclusion
criteria that may limit their applicability tocertain patients.
Provider beliefs about applicability may also be influenced by
patient factorssuch as demographics (e.g., age, sex), medical
history, comorbidities, patient preferences, andperceptions of
patient adherence. Uncertainty regarding the accuracy, consistency,
or both ofrisk factor values may also contribute to perceptions of
guideline applicability. This is partic-ularly prevalent in the
treatment of hypertension, in which discrepancies between office
andhome BP are common (53). Other reasons have to do with the
nature of the guideline it-self. Guidelines have been criticized
for being written in a way that does not always facilitatetheir
use. Cabana et al. (4) noted that across 23 surveys, 17%, 11%, 10%,
and 4.5% of physi-cians reported that guidelines were not easy to
use, inconvenient, cumbersome, and confusing,respectively.
Cognitive biases. Cognitive biases represent systematic
deviations from standards in judgment.In the context of clinical
inertia, providers may have systematic cognitive biases that
underminethe timely delivery and intensification of treatment. For
example, providers routinely underesti-mate the need to intensify
therapy (11). A study by el-Kebbi et al. (66) reported that
physiciansdid not intensify diabetes therapy over a 2–3-month
period among diabetic patients owing to mis-perceptions of control
in 41% of cases—despite most patients being obese (body mass index
=32 kg/m2). Similarly, health-care providers have been shown to
overestimate the care they pro-vide (67). In a survey about the use
of CPGs for the treatment of hypertension, US primary
careclinicians overestimated the proportion of patients who were
prescribed guideline-recommendedmedication (75% perceived versus
65% actual) as well as the proportion of patients whose BPlevels
were below target levels set at their previous visit (68% perceived
versus 43% actual)(68).
Motivational factors. Motivation reflects a general desire or
willingness to engage in a particularbehavior and may be influenced
by both extrinsic (e.g., money, praise, status, power) and
intrin-sic motivators (e.g., pleasure, behavior is consistent with
personal goals or values) (69). Clinicalinertia may reflect a lack
of provider motivation to change practice behavior. Habits are
hardto change, even those related to clinical practice, where as
many as 20% of providers report alack of motivation (i.e., desire
or perceived importance) as a barrier to the provision of
guideline-recommended therapy (4). A related factor that may serve
to undermine motivation is outcomeexpectancies, which is the
expectation that behavior will lead to a particular outcome (70).
In thereview by Cabana et al. (4), 26% of providers reported
negative outcome expectancies, whichcould be influenced by
perceived lack of treatment efficacy, guideline nonapplicability,
or patientnonadherence, to be a barrier to following CPGs.
Self-efficacy. Providers’ belief or confidence in their ability
to implement guidelines and enactmeaningful practice change
represents another barrier to the timely delivery or
intensificationof treatment. Cabana et al. (4) observed that a
median of 13% of respondents across 19 surveysreported limited
self-efficacy as a barrier to the implementation of CPGs. Similar
results were
270 Lavoie · Rash · Campbell
Ann
u. R
ev. P
harm
acol
. Tox
icol
. 201
7.57
:263
-283
. Dow
nloa
ded
from
ww
w.a
nnua
lrev
iew
s.or
g A
cces
s pr
ovid
ed b
y C
onco
rdia
Uni
vers
ity -
Mon
trea
l on
07/1
0/17
. For
per
sona
l use
onl
y.
-
PA57CH14-Campbell ARI 26 November 2016 13:14
Science-practice gap:the divide betweenresearch
literatureconcerning clinicalinterventions and itsapplication to
patients
reported in a survey of 154 clinicians treating COPD patients
(71). Self-efficacy beliefs may beinfluenced heavily by
organizational barriers such as time and available resources.
INTERVENTIONS TO REDUCE CLINICAL INERTIA
Various strategies have been employed to encourage providers to
bridge the science-practice gapby changing clinical practice
behavior. Interventions may be grouped into five broad
categories:educational approaches, practice audit and feedback,
decision support approaches, incentives, andmultifaceted
interventions (see Table 1 for a summary of definitions and
examples).
Educational approaches range from passive interventions, such as
the dissemination of printedmaterials, the use of opinion leaders
to impart knowledge, and traditional didactic lectures
andconference presentations, to more active approaches, including
academic detailing and engagingforms of CME (e.g., practical
workshops). Educational approaches target provider knowledge
orawareness of guideline-recommended therapy. A recent review of
105 CME studies reported that58% of 105 studies improved physician
practice behavior (e.g., prescribing), with more activeforms of CME
and those using multiple media formats (e.g., slides, videos),
multiple instructiontechniques (e.g., didactic lectures,
interactive group discussions, practical exercises), and
multipleexposures showing greater efficacy (72). In contrast, more
passive forms of CME show smaller butreliable effects on changing
practice behavior.
Audit and feedback involves reviewing clinical performance over
a specific period of time viachart audits, patient surveys, or
direct observation and giving providers specific feedback on
thequality of their performance. This approach helps providers
recognize cognitive biases (e.g., over-estimation of care).
According to reviews by Mostofian et al. (73) and Yen (74), audit
and feedbackhas been associated with a range of effects on provider
behavior, with studies showing small (16%decrease in physician
compliance) to large effects (70% increase in physician
compliance). Fur-thermore, the larger positive effects (74) were
associated with lower baseline compliance amongproviders, and
feedback was most effective when delivered prior to making
decisions about clinicalcare (73).
Decision support approaches are information systems designed to
improve clinical decisionmaking by analyzing patient-specific
variables (e.g., clinical data) and using the data provided
togenerate treatment recommendations. These approaches are passive
and often involve remindersor simplified decision algorithms for
preventive interventions, prescribing, and dosing. They
targetprovider knowledge or awareness of guideline-recommended
therapy, cognitive biases, and self-efficacy. Decision support
systems reportedly improved provider performance in 64% of
studies(75) and were particularly effective when triggered
automatically during clinic practice (68% ofstudies) (76).
Reminders can also have large effects on practice behavior. One
review reported thatcomputerized prompts for prevention activities
improved physician performance in 76% of trials(75). However,
decision support systems need sufficient data to code and trigger a
response andthe support system and reminders in place to trigger
appropriate provider behavior.
Incentives come in the form of financial or other rewards (e.g.,
institutional accreditation).Incentives target provider motivation.
Although effective in 70% of studies (77), this approachtargets
extrinsic rather than intrinsic motivations, meaning that the
behavior may extinguish inthe absence of ongoing reinforcement.
Multifaceted interventions do not encompass a single approach
but rather seek to combinemultiple approaches to optimize efficacy.
In general, approaches that combine more than onestrategy (e.g.,
education, audit and feedback, reminders, simplification of
treatment regimens)have been found to be highly effective in
changing physician practice behavior, with overall successrates
exceeding 70% (74, 78, 79).
www.annualreviews.org • Provider Behavior in Disease Management
271
Ann
u. R
ev. P
harm
acol
. Tox
icol
. 201
7.57
:263
-283
. Dow
nloa
ded
from
ww
w.a
nnua
lrev
iew
s.or
g A
cces
s pr
ovid
ed b
y C
onco
rdia
Uni
vers
ity -
Mon
trea
l on
07/1
0/17
. For
per
sona
l use
onl
y.
-
PA57CH14-Campbell ARI 26 November 2016 13:14
Tab
le1
Cha
ract
eris
tics
ofin
terv
enti
ons
tore
duce
clin
ical
iner
tia
Beh
avio
rch
ange
met
hod
Defi
niti
onB
ehav
ior
chan
geth
eory
App
roac
hP
rovi
der
barr
iers
addr
esse
dE
ffica
cyE
xam
ple
Oth
erco
nsid
erat
ions
Edu
cati
onal
appr
oach
es
Pri
nted
educ
atio
nal
mat
eria
lsD
istr
ibut
ion
ofpr
inte
dre
com
men
datio
nsfo
rcl
inic
alca
re(e
.g.,
prac
tice
guid
elin
es)
Non
eP
assi
veL
ack
ofkn
owle
dge
and
awar
enes
sL
owM
aile
dgu
idel
ines
and
mai
led
guid
elin
espl
used
ucat
ion
outr
each
did
notc
hang
epr
escr
iptio
nof
NSA
IDs
rela
tive
tono
inte
rven
tion
(90)
.
Hig
hly
vari
able
and
diffi
cult
tode
term
ine
qual
ities
ofm
ore
succ
essf
ulin
terv
entio
ns(9
1)
Loc
alop
inio
nle
ader
sT
rans
mis
sion
ofop
inio
nsof
heal
th-c
are
prov
ider
swho
are
deem
edin
fluen
tial
Non
eP
assi
veL
ack
ofkn
owle
dge
and
awar
enes
sL
owA
one-
page
,dis
ease
-spe
cific
sum
mar
yen
dors
edby
anop
inio
nle
ader
show
edsm
all
impr
ovem
ents
inph
ysic
ian
pres
crip
tion
ofca
rdio
vasc
ular
med
icat
ion
(92)
.
Effe
ctiv
ely
used
with
othe
rst
rate
gies
;hig
hly
vari
able
and
diffi
cult
tode
term
ine
best
way
toop
timiz
eus
e(9
3)
Aca
dem
icde
taili
ngA
ctiv
ein
form
atio
ntr
ansf
erth
roug
hpr
esen
tatio
nsby
atr
aine
dpe
rson
Non
eP
assi
veL
ack
ofkn
owle
dge
and
awar
enes
sL
owR
elat
ive
toco
ntro
l,el
ectr
onic
and
dire
ct(f
ace-
to-f
ace)
acad
emic
deta
iling
incr
ease
dlik
elih
ood
oflip
idte
stin
gfo
rdi
abet
ics(
94).
Pro
duce
ssm
allb
utre
liabl
eef
fect
sw
hen
used
alon
eor
com
bine
dw
ithot
her
met
hods
(95)
CM
EP
assi
vefo
rms
incl
ude
info
rmat
ion
prov
ided
via
educ
atio
nalc
onfe
renc
es,
lect
ures
,or
mee
tings
;act
ive
form
sin
clud
eta
ilore
dle
arni
ngan
dpr
actic
alw
orks
hops
Var
iabl
eA
ctiv
eor
pass
ive
Pas
sive
:lac
kof
know
ledg
ean
daw
aren
ess
Act
ive:
mul
tiple
Mod
erat
eT
wo
2–3-
hin
tera
ctiv
ese
min
ars
toim
prov
eco
mm
unic
atio
nw
ithch
ildre
nw
ithas
thm
ain
crea
sed
use
ofco
ntro
ller
med
icat
ion
and
redu
ced
asth
ma-
rela
ted
hosp
ital
visi
ts(9
6).
Mos
teffe
ctiv
ew
hen
(a)b
oth
inte
ract
ive
and
dida
ctic
,(b
)hig
hly
atte
nded
,and
(c)c
hang
ing
sim
ple
beha
vior
s(97
)
Aud
itan
dfe
edba
ckap
proa
ches
Aud
itan
dfe
edba
ckR
esul
tsof
revi
ewso
fclin
ical
perf
orm
ance
(e.g
.,ch
arts
,su
rvey
s,ob
serv
atio
n)ar
efe
dba
ckto
the
prov
ider
Non
eA
ctiv
eC
ogni
tive
bias
esM
oder
ate
Aud
itsof
acut
eas
thm
apr
oced
ures
wer
ean
onym
ized
and
fed
back
topr
ovid
ers,
whi
chin
crea
sed
use
ofpe
akflo
wby
45%
(98)
.
Mos
teffe
ctiv
ew
hen
(a)d
eliv
ered
bya
supe
rvis
oror
colle
ague
,(b
)per
form
ance
islo
wto
begi
nw
ith,(
c)pr
ovid
edm
ultip
letim
es,
and
(d)c
lear
targ
ets
are
used
(99)
Dec
isio
nsu
ppor
tap
proa
ches
Ana
lysi
sof
patie
ntda
taN
ewcl
inic
alin
form
atio
nco
llect
eddi
rect
lyfr
ompa
tient
san
dgi
ven
toth
epr
ovid
er
Non
eP
assi
veL
ack
ofkn
owle
dge
and
awar
enes
sM
oder
ate
Pat
ient
self-
repo
rts
ofde
pres
sed
moo
din
unre
cogn
ized
depr
esse
dpa
tient
sw
ere
fed
back
toph
ysic
ians
,whi
chim
prov
ed12
-mon
thre
cogn
ition
and
trea
tmen
tofd
epre
ssio
n(1
00).
The
circ
umst
ance
sun
der
whi
chth
isst
rate
gyis
mos
teffi
caci
ous
are
noty
etkn
own
272 Lavoie · Rash · Campbell
Ann
u. R
ev. P
harm
acol
. Tox
icol
. 201
7.57
:263
-283
. Dow
nloa
ded
from
ww
w.a
nnua
lrev
iew
s.or
g A
cces
s pr
ovid
ed b
y C
onco
rdia
Uni
vers
ity -
Mon
trea
l on
07/1
0/17
. For
per
sona
l use
onl
y.
-
PA57CH14-Campbell ARI 26 November 2016 13:14
Dec
isio
nsu
ppor
tsys
tem
sIn
form
atio
nsy
stem
sth
atan
alyz
epa
tient
-spe
cific
clin
ical
vari
able
s(e
.g.,
prev
entiv
eca
re,d
isea
sem
anag
emen
t,pr
escr
ibin
g)
Non
eP
assi
veL
ack
ofkn
owle
dge
and
awar
enes
s,la
ckof
agre
emen
t
Mod
erat
eR
elat
ive
tous
ualc
are,
aco
mpu
ter-
assi
sted
deci
sion
supp
ortp
rogr
amta
rget
edan
dre
duce
dth
epr
escr
ibin
gof
9in
appr
opri
ate
med
icat
ions
inan
ED
sett
ing
(101
).
Mos
teffe
ctiv
ew
hen
(a)a
dvic
egi
ven
for
patie
nts
and
prac
titio
ners
,(b)
requ
irin
gpr
ovid
erre
ason
for
over
ride
,an
d(c
)eva
luat
edby
deve
lope
rs(7
6,10
2)L
imita
tions
:ale
rts
may
bedi
srup
tive,
prov
ider
may
not
agre
ew
ithad
vice
,and
cont
inge
ntup
onqu
ality
ofda
ta(1
5)
Rem
inde
rsM
anua
lor
com
pute
rize
dpr
ompt
sdi
rect
ing
phys
icia
nsto
perf
orm
asp
ecifi
ccl
inic
alac
tion
Non
eP
assi
veL
ack
ofkn
owle
dge
and
awar
enes
sM
oder
ate
Com
pute
rize
dre
min
ders
ofpa
tient
-spe
cific
reco
mm
enda
tions
for
elev
ated
gluc
ose
impr
oved
the
timel
yin
tens
ifica
tion
oftr
eatm
ent
(103
).
Use
def
fect
ivel
yw
ithot
her
stra
tegi
es(1
04)
Sim
plifi
catio
nof
regi
men
The
use
ofsi
mpl
ified
deci
sion
algo
rith
ms
for
trea
tmen
tre
gim
ens
Non
eP
assi
veL
ack
ofkn
owle
dge
and
awar
enes
s(gu
idel
ine
com
plex
ity)
Mod
erat
eA
sim
plifi
edfo
ur-s
tep
algo
rith
mto
man
age
hype
rten
sion
impr
oved
timel
yin
tens
ifica
tion
ofan
tihyp
erte
nsiv
em
edic
atio
ns(1
05).
The
circ
umst
ance
sun
der
whi
chth
isst
rate
gyis
mos
teffi
caci
ous
are
noty
etkn
own
Ince
ntiv
eap
proa
ches
Eco
nom
icin
cent
ives
Fina
ncia
lrew
ards
orpe
nalti
esfo
ren
gagi
ngin
spec
ific
clin
ical
prac
tices
Lea
rnin
gth
eory
(pos
itive
rein
forc
emen
tor
puni
shm
ent)
Pas
sive
Lac
kof
mot
ivat
ion
Mod
erat
eP
rovi
der
finan
cial
bonu
s(e
.g.,
$1,0
00fo
ra
20%
impr
ovem
ent)
impr
oved
phys
icia
nim
mun
izat
ion
beha
vior
by25
.3%
(106
).
Effe
ctsd
ono
ttra
nsfe
rto
patie
ntou
tcom
es(7
7)
Mul
tifa
cete
dap
proa
ches
Mul
tifac
eted
Use
oftw
oor
mor
ein
terv
entio
nsV
aria
ble
depe
ndin
gon
inte
rven
tions
used
Act
ive
and
pass
ive
Mul
tiple
Hig
hE
duca
tiona
lout
reac
h,au
dita
ndfe
edba
ck,r
emin
ders
,opi
nion
lead
ered
ucat
ion,
and
regi
men
sim
plifi
catio
nim
prov
edth
ead
optio
nof
targ
eted
prac
tices
intr
eatm
entI
CU
s(1
07).
Has
been
rate
das
the
mos
tsu
cces
sful
inte
rven
tion
stra
tegy
toch
ange
phys
icia
nbe
havi
orfo
ra
desi
red
outc
ome
(73)
Abb
revi
atio
ns:C
ME
,con
tinui
ngm
edic
aled
ucat
ion;
ED
,em
erge
ncy
depa
rtm
ent;
ICU
,int
ensi
veca
reun
it;N
SAID
,non
ster
oida
lant
i-in
flam
mat
ory
drug
.
www.annualreviews.org • Provider Behavior in Disease Management
273
Ann
u. R
ev. P
harm
acol
. Tox
icol
. 201
7.57
:263
-283
. Dow
nloa
ded
from
ww
w.a
nnua
lrev
iew
s.or
g A
cces
s pr
ovid
ed b
y C
onco
rdia
Uni
vers
ity -
Mon
trea
l on
07/1
0/17
. For
per
sona
l use
onl
y.
-
PA57CH14-Campbell ARI 26 November 2016 13:14
Behavior changetheory (BCT):the application oftheoretically
informedmethods to targetidentified barriers andbring about
desiredbehavior change
Shared decisionmaking (SDM):empowering patientsto assume a
central orshared role in makingdecisions about theirmedical
care
LIMITATIONS OF CURRENT APPROACHES TO OVERCOMINGCLINICAL
INERTIA
Despite the existence of several provider-focused interventions
and evidence for their efficacy,current rates of clinical inertia
suggest that they remain inadequate for changing practice
behavior.An examination of the objectives, designs, and
intervention strategies employed across approachesreveals several
limitations. As summarized in Table 1, with the exception of some
multifacetedinterventions, most approaches have been designed to
address a single provider barrier. Giventhe range of provider
factors associated with clinical inertia, those focusing on a
single barriermay be less likely to succeed than those that aim to
address a range of provider factors becauseof one major flaw: They
make the assumption that the targeted barrier is the problem, when,
infact, barriers may be multiple and vary across providers.
Furthermore, the majority of approachesreviewed involve the passive
dissemination of knowledge, which is less effective than more
activeapproaches (e.g., practice audits with feedback) that involve
provider participation (73).
Most approaches to date have targeted one barrier in particular:
provider knowledge or aware-ness of CPGs. Interventions that target
behavior change should logically be inspired by behaviorchange
theories (BCTs); however, with the exception of incentives [which
are inspired by learningtheory and operant conditioning (80)] and
possibly some forms of active CME, few approachesappear to be based
on BCTs. In fact, a study of 110 accredited CME programs offered to
physiciansin Canada reported that 96%, 47%, and 26% used strategies
that targeted knowledge, compre-hension, and practice skills,
respectively. Finally, few approaches address barriers inherent
inimplementing guidelines (e.g., impractical, inconvenient, or
biased guidelines) (4, 12).
TOWARD THE USE OF AN EVIDENCE-BASED FRAMEWORKFOR OVERCOMING
CLINICAL INERTIA
We propose the use of an evidence-based framework to adequately
address clinical inertia that em-phasizes the use of BCT. We also
make recommendations for integrating strategies for
overcomingpatient-level barriers that promote shared decision
making (SDM) and suggest a framework forimproving guideline
development and dissemination.
Overcoming Provider Barriers Using Behavior Change Theory
BCT provides a framework for designing interventions that
address specific behavior gaps in thecontext of clinical inertia.
First, BCT can help us understand what barriers should be
targeted.For example, behavioral assessment can be used to identify
whether CPGs are not being adoptedfor reasons related to lack of
knowledge or awareness, lack of agreement, cognitive biases, lack
ofmotivation, lack of self-efficacy, or a combination of these
reasons. This information can be usedto group providers in terms of
their barriers and offer interventions that target those
issues.
Second, BCT offers a framework for overcoming particular
barriers. For example, BCT wouldrecommend adopting motivational
approaches (e.g., motivational interviewing, inspired by
self-determination theory) (81) that are designed to enhance
intrinsic motivation and confidence (82)to overcome barriers in
motivation or self-efficacy. For a complete list of BCT-inspired
approachesthat could be used to overcome specific provider
barriers, see Table 2.
Finally, BCT provides a mechanism for understanding why some
interventions fail. For ex-ample, although some audit and feedback
approaches have been successful, others have actu-ally decreased
physician compliance with CPGs. This may be explained by learning
theory,which posits that receiving negative feedback about poor
performance may be experienced
274 Lavoie · Rash · Campbell
Ann
u. R
ev. P
harm
acol
. Tox
icol
. 201
7.57
:263
-283
. Dow
nloa
ded
from
ww
w.a
nnua
lrev
iew
s.or
g A
cces
s pr
ovid
ed b
y C
onco
rdia
Uni
vers
ity -
Mon
trea
l on
07/1
0/17
. For
per
sona
l use
onl
y.
-
PA57CH14-Campbell ARI 26 November 2016 13:14
Tab
le2
Pro
vide
rin
terv
enti
ons
toov
erco
me
clin
ical
iner
tia
Pro
vide
rba
rrie
rIn
terv
enti
onty
peB
ehav
ior
chan
geth
eory
Defi
niti
onSt
rate
gies
Lac
kof
know
ledg
ean
daw
aren
ess
Edu
catio
nN
AT
hepr
oces
sof
lear
ning
via
the
acqu
isiti
onof
know
ledg
eP
rint
edm
ater
ials
,lec
ture
s,co
nfer
ence
s,ac
adem
icde
taili
ng,t
heIn
tern
et,a
ndw
ebin
ars
Lac
kof
agre
emen
tA
ttitu
deor
beha
vior
chan
geC
ogni
tive
beha
vior
alth
eory
(108
)B
ehav
ior
isth
ere
sult
ofou
rin
terp
reta
tions
(tho
ught
s)of
our
envi
ronm
ent(
exte
rnal
stim
uli).
Cog
nitiv
ere
stru
ctur
ing
Hea
lthbe
liefm
odel
(109
)B
ehav
ior
chan
geis
influ
ence
dby
perc
eive
dsu
scep
tibili
tyto
apr
oble
m,s
erio
usne
ss,b
enefi
tsof
actio
n,an
dba
rrie
rs.
Cog
nitiv
ebi
ases
Att
itude
orbe
havi
orch
ange
Cog
nitiv
ebe
havi
oral
theo
ry(1
08)
Cog
nitiv
ere
stru
ctur
ing
Hea
lthbe
liefm
odel
(109
)
Lac
kof
mot
ivat
ion
Att
itude
orbe
havi
orch
ange
Self-
dete
rmin
atio
nth
eory
(81)
Beh
avio
ris
dete
rmin
edby
the
degr
eeto
whi
chit
isdr
iven
auto
nom
ousl
yan
dco
nsis
tent
with
anin
divi
dual
’sgo
als
and
valu
es.
Mot
ivat
iona
lint
ervi
ewin
g(6
9)
Tra
nsth
eore
tical
mod
el(1
10)
Lev
elof
read
ines
sto
chan
gede
term
ines
beha
vior
chan
ge.
We
mov
eth
roug
hfiv
est
ages
ofch
ange
:pre
cont
empl
atio
n,co
ntem
plat
ion,
prep
arat
ion,
actio
n,an
dm
aint
enan
ce,t
hela
tter
thre
eof
whi
chin
dica
tere
adin
ess.
Mot
ivat
iona
lint
ervi
ewin
g(6
9)
(Con
tinue
d)
www.annualreviews.org • Provider Behavior in Disease Management
275
Ann
u. R
ev. P
harm
acol
. Tox
icol
. 201
7.57
:263
-283
. Dow
nloa
ded
from
ww
w.a
nnua
lrev
iew
s.or
g A
cces
s pr
ovid
ed b
y C
onco
rdia
Uni
vers
ity -
Mon
trea
l on
07/1
0/17
. For
per
sona
l use
onl
y.
-
PA57CH14-Campbell ARI 26 November 2016 13:14
Tab
le2
(Con
tinu
ed)
Pro
vide
rba
rrie
rIn
terv
enti
onty
peB
ehav
ior
chan
geth
eory
Defi
niti
onSt
rate
gies
The
ory
ofpl
anne
dbe
havi
or(1
11)
Inte
ntio
nis
influ
ence
dby
attit
ude
tow
ard
the
beha
vior
,su
bjec
tive
norm
s,an
dpe
rcei
ved
beha
vior
alco
ntro
l.
Mot
ivat
iona
lint
ervi
ewin
g(6
9)
Reg
ulat
ory
focu
sth
eory
(112
)B
ehav
ior
isde
term
ined
fund
amen
tally
byth
ede
sire
topu
rsue
plea
sure
(pos
itive
outc
omes
)and
avoi
dpa
in(n
egat
ive
outc
omes
)
Per
sona
llyre
leva
ntin
cent
ives
Soci
alco
gniti
veth
eory
(70)
Beh
avio
ris
influ
ence
dby
mod
elin
got
hers
we
iden
tify
with
,pos
itive
outc
ome
expe
ctan
cies
,and
confi
denc
e(s
elf-
effic
acy)
inou
rab
ility
tosu
cces
sful
lyen
gage
ina
beha
vior
.
Mot
ivat
iona
lint
ervi
ewin
g(6
9),
cogn
itive
rest
ruct
urin
g,ex
posu
re(b
ehav
iora
lex
peri
men
ts),
and
goal
sett
ing
and
prob
lem
solv
ing
Lac
kof
self-
effic
acy
Att
itude
orbe
havi
orch
ange
Cog
nitiv
ebe
havi
oral
ther
apy
(108
)C
ogni
tive
rest
ruct
urin
g
The
ory
ofpl
anne
dbe
havi
or(1
11)
Mot
ivat
iona
lint
ervi
ewin
g(6
9)
Abb
revi
atio
n:N
A,n
otap
plic
able
.
276 Lavoie · Rash · Campbell
Ann
u. R
ev. P
harm
acol
. Tox
icol
. 201
7.57
:263
-283
. Dow
nloa
ded
from
ww
w.a
nnua
lrev
iew
s.or
g A
cces
s pr
ovid
ed b
y C
onco
rdia
Uni
vers
ity -
Mon
trea
l on
07/1
0/17
. For
per
sona
l use
onl
y.
-
PA57CH14-Campbell ARI 26 November 2016 13:14
negatively (as a punishment) and result in reduced motivation
and frequency of enacting thetarget behavior. Anticipation of
negative feedback may also create anxiety and lead to avoid-ance of
participating in such interventions. Similarly, according to
self-determination theory,incentive-based interventions would only
be expected to work if (a) providers were highly moti-vated by
financial rewards, (b) accepting financial rewards did not conflict
with other values, and(c) the financial rewards are offered.
Furthermore, behavior change strategies that rely on extrin-sic
rewards may undermine behavior change for intrinsic reasons (i.e.,
I want to practice EBMbecause I value excellence, altruism, and
accountability) and may not be feasible in the longterm.
Overcoming Patient Barriers by Promoting Shared Decision
Making
Two important, modifiable, patient-related factors associated
with clinical inertia are treatmentnonadherence and unhealthy
lifestyle behaviors. One of the most promising methods for
im-proving patient adherence (both to therapy and lifestyle
recommendations) is adopting an SDMapproach (83). SDM aims to
empower patients to assume a central role in decision making
abouttheir care and includes the following elements: (a) reciprocal
exchange of information betweenthe patient and provider, (b)
negotiation of treatment options and outcomes, and (c) reaching
con-sensus about the course of action. This approach has succeeded
at increasing patient adherenceacross a variety of conditions (84,
85), and although it may appear to undermine the practice ofEBM, we
argue that this is not the case. In fact, integrating SDM into EBM
may actually en-hance both provider compliance and patient
adherence to guideline-recommended therapies byhelping to
simultaneously overcome barriers related to provider knowledge and
agreement withthe applicability of guidelines, as well as cognitive
biases related to overestimates of the quality ofcare.
Overcoming Guideline-Related Barriers via Improved
Developmentand Dissemination
Often overlooked are limitations inherent to guidelines
themselves. Improving the way in whichwe develop and disseminate
guidelines may have a major, positive, and rapid impact on
guide-line uptake. Rogers (86) describes guideline characteristics
that affect provider adoption andcould be used to guide
intervention development and dissemination practices. They include
rel-ative advantage (is the new recommendation significantly
superior to the previous one?), com-patibility (is the guideline
consistent with the provider’s beliefs and values?), complexity
(howdifficult is it to understand and implement the guideline?),
trialability (can the provider testsome or all of the
recommendations with relative ease?), and observability (are there
oppor-tunities for the provider to observe the results of guideline
implementation among respectedpeers?). One study validated these
criteria in 23 trials and reported that trialability,
observabil-ity, and low complexity were the three guideline
characteristics associated with greater guidelineadoption (87).
This knowledge could be used to improve guideline development and
dissemi-nation strategies in conjunction with existing frameworks.
For example, the Guidelines Inter-national Network, which
represents 103 organizations from 47 countries, maintains a
databaseof more than 6,100 guidelines and offers a Guideline for
Guidelines that includes training ma-terials (88). This group has
also made available the Appraisal of Guidelines for Research
andEvaluation (AGREE I and II) instruments for guideline evaluation
(89). The GuideLine Imple-mentability Appraisal (GLIA) instrument
is also available for assessing the quality of
guidelineimplementation.
www.annualreviews.org • Provider Behavior in Disease Management
277
Ann
u. R
ev. P
harm
acol
. Tox
icol
. 201
7.57
:263
-283
. Dow
nloa
ded
from
ww
w.a
nnua
lrev
iew
s.or
g A
cces
s pr
ovid
ed b
y C
onco
rdia
Uni
vers
ity -
Mon
trea
l on
07/1
0/17
. For
per
sona
l use
onl
y.
-
PA57CH14-Campbell ARI 26 November 2016 13:14
CONCLUSIONS AND FUTURE DIRECTIONS
Clinical inertia is a major barrier to achieving optimal
clinical outcomes among patients with a widerange of chronic
diseases. CPGs are not without their limitations, and development
and dissemi-nation strategies could be improved by simplifying them
and making them more accessible for trialpurposes. In an effort to
improve the effectiveness of provider-focused intervention
strategies, wepropose using an evidence-based framework that
incorporates BCT that identifies what barriersto target and how to
target them. The fact that most provider-focused interventions to
date havenot been inspired by any evidence-based BCT is a major
limitation of current approaches. Adher-ence to CPGs by providers
does not guarantee good outcomes. Strategies that engage patients
inthe treatment process are also important for overcoming clinical
inertia, and we propose adopt-ing an SDM model to optimize both
patient and provider adherence to
guideline-recommendedtherapies.
SUMMARY POINTS
1. Despite the widespread availability of CPGs and strong
evidence supporting the benefitsof their use, provider nonadherence
to CPGs is prevalent in chronic diseases such ashypertension,
diabetes, and dyslipidemia, with rates that exceed 50%.
2. True clinical inertia occurs when a patient fails to meet
clearly defined and measurabletreatment targets and their provider
does not intensify treatment in accordance withguideline
recommendations despite awareness of the guidelines, belief in
their applica-bility, and available resources to do so.
3. Clinical inertia involves a complex interaction between
organizational and system factors,patient factors, and provider
factors, with their relative contributions at 20%, 30%, and50%,
respectively.
4. Provider-focused interventions to change clinical practice
behavior have had modestsuccess, but most often target one
particular barrier: provider knowledge or awarenessof CPGs.
5. Interventions designed to improve clinical inertia should be
grounded in BCT and in-corporate SDM and improved guideline
development and dissemination.
DISCLOSURE STATEMENT
The authors are not aware of any affiliations, memberships,
funding, or financial holdings thatmight be perceived as affecting
the objectivity of this review.
ACKNOWLEDGMENTS
Kim L. Lavoie receives salary support in the form of
investigator awards from the CanadianInstitutes of Health Research
(CIHR) and the Fonds de la recherche du Québec – Santé
(FRQS).Joshua A. Rash is supported by trainee awards from Alberta
Innovates Health Research (AIHS)and the Canadian Institutes of
Health Research (CIHR). Tavis S. Campbell receives support inthe
form of investigator awards from CIHR.
278 Lavoie · Rash · Campbell
Ann
u. R
ev. P
harm
acol
. Tox
icol
. 201
7.57
:263
-283
. Dow
nloa
ded
from
ww
w.a
nnua
lrev
iew
s.or
g A
cces
s pr
ovid
ed b
y C
onco
rdia
Uni
vers
ity -
Mon
trea
l on
07/1
0/17
. For
per
sona
l use
onl
y.
-
PA57CH14-Campbell ARI 26 November 2016 13:14
LITERATURE CITED
1. Sackett DL, Rosenberg WM, Gray JA, Haynes RB, Richardson WS.
1996. Evidence based medicine:what it is and what it isn’t. BMJ
312:71–72
2. Rosenberg W, Donald A. 1995. Evidence based medicine: an
approach to clinical problem-solving. BMJ310:1122–26
3. Graham R, Mancher M, Wolman DM, Greenfield S, Steinberg E,
eds. 2011. Clinical Practice GuidelinesWe Can Trust. Washington,
DC: Natl. Acad. Press
4. Cabana MD, Rand CS, Powe NR, Wu AW, Wilson MH, et al. 1999.
Why don’t physicians followclinical practice guidelines? A
framework for improvement. JAMA 282:1458–65
5. Carlhed R, Bojestig M, Wallentin L, Lindstrom G, Peterson A,
et al. 2006. Improved adherence toSwedish national guidelines for
acute myocardial infarction: the Quality Improvement in
CoronaryCare (QUICC) study. Am. Heart J. 152:1175–81
6. Grimshaw JM, Russell IT. 1993. Effect of clinical guidelines
on medical practice: a systematic review ofrigorous evaluations.
Lancet 342:1317–22
7. Lugtenberg M, Burgers JS, Westert GP. 2009. Effects of
evidence-based clinical practice guidelines onquality of care: a
systematic review. Qual. Saf. Health Care 18:385–92
8. Sager HB, Linsel-Nitschke P, Mayer B, Lieb W, Franzel B, et
al. 2010. Physicians’ perception ofguideline-recommended
low-density lipoprotein target values: characteristics of
misclassified patients.Eur. Heart J. 31:1266–73
9. Balder JW, Scholtens S, de Vries JK, van Schie LM, Boekholdt
SM, et al. 2015. Adherence to guidelinesto prevent cardiovascular
diseases: the LifeLines cohort study. Neth. J. Med. 73:316–23
10. McGlynn EA, Asch SM, Adams J, Keesey J, Hicks J, et al.
2003. The quality of health care delivered toadults in the United
States. N. Engl. J. Med. 348:2635–45
11. O’Connor PJ, Sperl-Hillen JM, Johnson PE, Rush WA, Biltz G.
2005. Clinical inertia and outpatientmedical errors. In Advances in
Patient Safety: From Research to Implementation, Vol. 2: Concepts
and Method-ology, ed. K Henriksen, JB Battles, ES Marks, DI Lewin,
pp. 293–308. Rockville, MD: Agency Healthc.Res. Qual.
12. Grahame-Smith D. 1995. Evidence based medicine: Socratic
dissent. BMJ 310:1126–2713. Phillips LS, Branch WT, Cook CB, Doyle
JP, El-Kebbi IM, et al. 2001. Clinical inertia. Ann. Intern.
Med. 135:825–3414. Allen JD, Curtiss FR, Fairman KA. 2009.
Nonadherence, clinical inertia, or therapeutic inertia? J.
Manag.
Care Pharm. 15:690–9515. Faria C, Wenzel M, Lee KW, Coderre K,
Nichols J, Belletti DA. 2009. A narrative review of clinical
inertia: focus on hypertension. J. Am. Soc. Hypertens.
3:267–7616. Okonofua EC, Simpson KN, Jesri A, Rehman SU, Durkalski
VL, Egan BM. 2006. Therapeutic inertia is
an impediment to achieving the Healthy People 2010 blood
pressure control goals. Hypertension 47:345–51
17. Garfinkel D, Mangin D. 2010. Feasibility study of a
systematic approach for discontinuation of multiplemedications in
older adults: addressing polypharmacy. Arch. Intern. Med.
170:1648–54
18. Hicks PC, Westfall JM, Van Vorst RF, Bublitz Emsermann C,
Dickinson LM, et al. 2006. Action orinaction? Decision making in
patients with diabetes and elevated blood pressure in primary care.
DiabetesCare 29:2580–85
19. Reach G. 2014. Clinical Inertia: A Critique of Medical
Reason. New York: Springer20. Asche C, Bode B, Busk A, Nair S.
2012. The economic and clinical benefits of adequate insulin
initiation
and intensification in people with type 2 diabetes mellitus.
Diabetes Obes. Metab. 14:47–5721. UK Prospect. Diabetes Study
Group. 1998. Intensive blood-glucose control with sulphonylureas
or
insulin compared with conventional treatment and risk of
complications in patients with type 2 diabetes(UKPDS 33). Lancet
352:837–53
22. Ohkubo Y, Kishikawa H, Araki E, Miyata T, Isami S, et al.
1995. Intensive insulin therapy prevents theprogression of diabetic
microvascular complications in Japanese patients with
non-insulin-dependentdiabetes mellitus: a randomized prospective
6-year study. Diabetes Res. Clin. Pract. 28:103–17
www.annualreviews.org • Provider Behavior in Disease Management
279
Ann
u. R
ev. P
harm
acol
. Tox
icol
. 201
7.57
:263
-283
. Dow
nloa
ded
from
ww
w.a
nnua
lrev
iew
s.or
g A
cces
s pr
ovid
ed b
y C
onco
rdia
Uni
vers
ity -
Mon
trea
l on
07/1
0/17
. For
per
sona
l use
onl
y.
-
PA57CH14-Campbell ARI 26 November 2016 13:14
23. Holman RR, Paul SK, Bethel MA, Matthews DR, Neil HAW. 2008.
10-year follow-up of intensiveglucose control in type 2 diabetes.
N. Engl. J. Med. 359:1577–89
24. Khunti K, Wolden ML, Thorsted BL, Andersen M, Davies MJ.
2013. Clinical inertia in people withtype 2 diabetes: a
retrospective cohort study of more than 80,000 people. Diabetes
Care 36:3411–17
25. Shah BR, Hux JE, Laupacis A, Zinman B, van Walraven C. 2005.
Clinical inertia in response to inadequateglycemic control: Do
specialists differ from primary care physicians? Diabetes Care
28:600–6
26. McEwen LN, Bilik D, Johnson SL, Halter JB, Karter AJ, et al.
2009. Predictors and impact of inten-sification of
antihyperglycemic therapy in type 2 diabetes: translating research
into action for diabetes(TRIAD). Diabetes Care 32:971–76
27. Wolf-Maier K, Cooper RS, Kramer H, Banegas JR, Giampaoli S,
et al. 2004. Hypertension treatmentand control in five European
countries, Canada, and the United States. Hypertension 43:10–17
28. Roumie CL, Elasy TA, Wallston KA, Pratt S, Greevy RA, et al.
2007. Clinical inertia: a common barrierto changing provider
prescribing behavior. Joint Comm. J. Qual. Patient Saf.
33:277–85
29. Pittman DG, Fenton C, Chen W, Haffner S, Pendergrass M.
2012. Relation of statin nonadherence andtreatment intensification.
Am. J. Cardiol. 110:1459–63
30. Rodondi N, Peng T, Karter AJ, Bauer DC, Vittinghoff E, et
al. 2006. Therapy modifications in responseto poorly controlled
hypertension, dyslipidemia, and diabetes mellitus. Ann. Intern.
Med. 144:475–84
31. Erzen D, Carriere KC, Dik N, Mustard C, Roos LL, et al.
1997. Income level and asthma prevalenceand care patterns. Am. J.
Respir. Crit. Care Med. 155:1060–65
32. Habbick BF, Pizzichini MMM, Taylor B, Rennie D,
Senthilselvan A, Sears MR. 1999. Prevalence ofasthma, rhinitis and
eczema among children in 2 Canadian cities: the International Study
of Asthma andAllergies in Childhood. CMAJ 160:1824–28
33. Kattan M, Crain EF, Steinbach S, Visness CM, Walter M, et
al. 2006. A randomized clinical trial ofclinician feedback to
improve quality of care for inner-city children with asthma.
Pediatrics 117:e1095–103
34. Szefler SJ, Mitchell H, Sorkness CA, Gergen PJ, O’Connor GT,
et al. 2008. Management of asthmabased on exhaled nitric oxide in
addition to guideline-based treatment for inner-city adolescents
andyoung adults: a randomised controlled trial. Lancet
372:1065–72
35. Krym VF, Crawford B, MacDonald RD. 2004. Compliance with
guidelines for emergency managementof asthma in adults: experience
at a tertiary care teaching hospital. CJEM 6:321–26
36. Cydulka RK, Tamayo-Sarver JH, Wolf C, Herrick E, Gress S.
2005. Inadequate follow-up controllermedications among patients
with asthma who visit the emergency department. Ann. Emerg. Med.
46:316–22
37. Ducharme FM, Noya FJ, Allen-Ramey FC, Maiese EM, Gingras J,
Blais L. 2012. Clinical effectivenessof inhaled corticosteroids
versus montelukast in children with asthma: prescription patterns
and patientadherence as key factors. Curr. Med. Res. Opin.
28:111–19
38. Lougheed MD, Garvey N, Chapman KR, Cicutto L, Dales R, et
al. 2009. Variations and gaps inmanagement of acute asthma in
Ontario emergency departments. Chest 135:724–36
39. Br. Thorac. Soc., Scott. Intercoll. Guidel. Netw. 2014.
British guideline on the management of asthma.Thorax 69:i1–92
40. Natl. Heart Lung Blood Inst., Natl. Asthma Educ. Prev.
Program. 2007. Expert Panel Report 3: guidelinesfor the diagnosis
and management of asthma. Rep., Natl. Inst. Health, Bethesda,
MD
41. WHO (World Health Organ.). 2013. Burden of COPD. Geneva:
WHO. http://www.who.int/respiratory/copd/burden/en/index.html
42. Murphy SL, Kochanek KD, Xu J, Arias E. 2015. Mortality in
the United States, 2014. NCHS Data Brief,Natl. Cent. Health Stat.,
Hyattsville, MD
43. Cooke CE, Sidel M, Belletti DA, Fuhlbrigge AL. 2012. Review:
clinical inertia in the management ofchronic obstructive pulmonary
disease. COPD 9:73–80
44. Kaminsky DA, Marcy TW, Bachand M, Irvin CG. 2005. Knowledge
and use of office spirometry for thedetection of chronic
obstructive pulmonary disease by primary care physicians. Respir.
Care 50:1639–48
45. Glob. Initiat. Chronic Obstr. Lung Dis. 2016. Global
strategy for the diagnosis, management, and preven-tion of COPD.
Rep., Glob. Initiat. Chronic Obstr. Lung Dis. (GOLD).
http://goldcopd.org/global-strategy-diagnosis-management-prevention-copd-2016/
280 Lavoie · Rash · Campbell
Ann
u. R
ev. P
harm
acol
. Tox
icol
. 201
7.57
:263
-283
. Dow
nloa
ded
from
ww
w.a
nnua
lrev
iew
s.or
g A
cces
s pr
ovid
ed b
y C
onco
rdia
Uni
vers
ity -
Mon
trea
l on
07/1
0/17
. For
per
sona
l use
onl
y.
http://www.who.int/respiratory/copd/burden/en/index.htmlhttp://www.who.int/respiratory/copd/burden/en/index.htmlhttp://goldcopd.org/global-strategy-diagnosis-management-prevention-copd-2016/http://goldcopd.org/global-strategy-diagnosis-management-prevention-copd-2016/
-
PA57CH14-Campbell ARI 26 November 2016 13:14
46. Diette GB, Orr P, McCormack MC, Gandy W, Hamar B. 2010. Is
pharmacologic care of chronicobstructive pulmonary disease
consistent with the guidelines? Popul. Health Manag. 13:21–26
47. Heins-Nesvold J, Carlson A, King-Schultz L, Joslyn KE. 2008.
Patient identified needs for chronicobstructive pulmonary disease
versus billed services for care received. Int. J. Chronic Obstr.
Pulm. Dis.3:415–21
48. Barr RG, Celli BR, Martinez FJ, Ries AL, Rennard SI, et al.
2005. Physician and patient perceptions inCOPD: the COPD Resource
Network Needs Assessment Survey. Am. J. Med. 118:1415.e9–17
49. Chavez PC, Shokar NK. 2009. Diagnosis and management of
chronic obstructive pulmonary disease(COPD) in a primary care
clinic. COPD 6:446–51
50. Aujoulat I, Jacquemin P, Rietzschel E, Scheen A, Trefois P,
et al. 2014. Factors associated with clinicalinertia: an
integrative review. Adv. Med. Educ. Pract. 5:141–47
51. Byrnes PD. 2011. Why haven’t I changed that? Therapeutic
inertia in general practice. Aust. Fam. Phys.40:24–28
52. Byrne D, O’Connor L, Jennings S, Bennett K, Murphy AW. 2015.
A survey of GPs awareness and useof risk assessment tools and
cardiovascular disease prevention guidelines. Ir. Med. J.
108:204–7
53. Kiberd J, Panek R, Kiberd B. 2007. Strategies to reduce
clinical inertia in hypertensive kidney transplantrecipients. BMC
Nephrol. 8:10
54. Nicolucci A, Rossi MC. 2008. Incretin-based therapies: a new
potential treatment approach to overcomeclinical inertia in type 2
diabetes. Acta Biomed. 79:184–91
55. Redon J, Coca A, Lazaro P, Aguilar MD, Cabanas M, et al.
2010. Factors associated with therapeuticinertia in hypertension:
validation of a predictive model. J. Hypertens. 28:1770–77
56. Berlowitz DR, Ash AS, Hickey EC, Friedman RH, Glickman M, et
al. 1998. Inadequate managementof blood pressure in a hypertensive
population. N. Engl. J. Med. 339:1957–63
57. Nau DP, Mallya U. 2005. Sex disparity in the management of
dyslipidemia among patients with type 2diabetes mellitus in a
managed care organization. Am. J. Manag. Care 11:69–73
58. Howes F, Hansen E, Williams D, Nelson M. 2010. Barriers to
diagnosing and managing hypertension:a qualitative study in
Australian general practice. Aust. Fam. Phys. 39:511–16
59. Grant R, Adams AS, Trinacty CM, Zhang F, Kleinman K, et al.
2007. Relationship between patientmedication adherence and
subsequent clinical inertia in type 2 diabetes glycemic management.
DiabetesCare 30:807–12
60. Reach G. 2008. Patient non-adherence and healthcare-provider
inertia are clinical myopia. DiabetesMetab. 34:382–85
61. Kotseva K, Wood D, De Backer G, De Bacquer D, Pyorala K,
Keil U. 2009. Cardiovascular preventionguidelines in daily
practice: a comparison of EUROASPIRE I, II, and III surveys in
eight Europeancountries. Lancet 373:929–40
62. Handler J, Lackland DT. 2011. Translation of hypertension
treatment guidelines into practice: a reviewof implementation. J.
Am. Soc. Hypertens. 5:197–207
63. Hyman DJ, Pavlik VN. 2000. Self-reported hypertension
treatment practices among primary care physi-cians: blood pressure
thresholds, drug choices, and the role of guidelines and
evidence-based medicine.Arch. Intern. Med. 160:2281–86
64. Foster JA, Yawn BP, Maziar A, Jenkins T, Rennard SI,
Casebeer L. 2007. Enhancing COPD managementin primary care
settings. MedGenMed 9:24
65. Barth JH, Misra S, Aakre KM, Langlois MR, Watine J, et al.
2015. Why are clinical practice guidelinesnot followed? Clin. Chem.
Lab. Med. 54:1133–39
66. el-Kebbi IM, Ziemer DC, Gallina DL, Dunbar V, Phillips LS.
1999. Diabetes in urban African-Americans. XV. Identification of
barriers to provider adherence to management protocols.
DiabetesCare 22:1617–20
67. Bramlage P, Thoenes M, Kirch W, Lenfant C. 2007. Clinical
practice and recent recommendations inhypertension
management—reporting a gap in a global survey of 1259 primary care
physicians in 17countries. Curr. Med. Res. Opin. 23:783–91
68. Steinman MA, Fischer MA, Shlipak MG, Bosworth HB, Oddone EZ,
et al. 2004. Clinician awarenessof adherence to hypertension
guidelines. Am. J. Med. 117:747–54
www.annualreviews.org • Provider Behavior in Disease Management
281
Ann
u. R
ev. P
harm
acol
. Tox
icol
. 201
7.57
:263
-283
. Dow
nloa
ded
from
ww
w.a
nnua
lrev
iew
s.or
g A
cces
s pr
ovid
ed b
y C
onco
rdia
Uni
vers
ity -
Mon
trea
l on
07/1
0/17
. For
per
sona
l use
onl
y.
-
PA57CH14-Campbell ARI 26 November 2016 13:14
69. Miller WR, Rollnick S. 2012. Motivational Interviewing:
Helping People Change. New York: GuilfordPress
70. Bandura A. 1986. Social Foundations of Thought and Action: A
Social Cognitive Theory. Englewood Cliffs,NJ: Prentice-Hall,
Inc.
71. Perez X, Wisnivesky JP, Lurslurchachai L, Kleinman LC,
Kronish IM. 2012. Barriers to adherence toCOPD guidelines among
primary care providers. Respir. Med. 106:374–81
72. Davis D, Galbraith R. 2009. Continuing medical education
effect on practice performance: effectivenessof continuing medical
education: American College of Chest Physicians Evidence-Based
EducationalGuidelines. Chest 135(Suppl.):42S–48S
73. Mostofian F, Ruban C, Simunovic N, Bhandari M. 2015.
Changing physician behavior: What works?Am. J. Manag. Care
21:75–84
74. Yen BM. 2006. Engaging physicians to change practice. J.
Clin. Outcomes Manag. 13:103–1075. Garg AX, Adhikari NKJ, McDonald
H, Rosas-Arellano MP, Devereaux PJ, et al. 2005. Effects of
comput-
erized clinical decision support systems on practitioner
performance and patient outcomes: a systematicreview. JAMA
293:1223–38
76. Kawamoto K, Houlihan CA, Balas EA, Lobach DF. 2005.
Improving clinical practice using clinicaldecision support systems:
a systematic review of trial