Oil and Water or Oil and Vinegar? Evidence-Based
Medicine Meets Recovery
Larry Davidson Æ Æ Robert E. Drake Æ Æ
Timothy Schmutte Æ Æ Thomas Dinzeo Æ Æ
Received: 23 January 2008/Accepted: 13 July 2009
? Springer Science+Business Media, LLC 2009
of ‘‘recovery’’ and ‘‘recovery-oriented practice,’’ practi-
tioners, program managers, and system leaders are
increasingly asking about the relationship between ‘‘evi-
dence-based practices’’ and recovery. After reviewing the
concepts of recovery from mental illness, being in recovery
with a mental illness, recovery-oriented care, and evidence-
based medicine, the authors argue for a complementary
relationship between recovery and evidence-based prac-
tices. This relationship is neither simple nor straightfor-
ward, but results in a whole that is greater than the sum of
its parts through which each element benefits from the
influence of the other.
With the increasing prominence of the notions
practice ? Recovery ? Self-determination ? Choice
Evidence-based medicine ? Evidence-based
With the increasing prominence of the notions of ‘‘recov-
ery’’ and ‘‘recovery-oriented practice,’’ practitioners, pro-
gram managers, and system leaders are increasingly asking
about the relationship between ‘‘evidence-based practices’’
and recovery. It seems to some that just as they were
beginning to get used to the notion of evidence-based
practices in mental health, they began to be pressured to
make the services they provide recovery-oriented as well.
This has naturally led to such questions as: ‘‘Is recovery
evidence-based?’’ or ‘‘What is the evidence for recovery-
oriented practice?’’ At the other end of the spectrum, some
mental health consumer advocates view the emphasis on
evidence-based practices to place a further restriction on
their ability to exercise choice in their care. They also are
suspicious of scientific claims to offer a privileged access
to truth, arguing instead for relying on first-hand experi-
ences as at least an equally valid source of information
about the utility of psychiatric interventions.
This article attempts to address these issues by exam-
ining several possible relationships that may exist between
the emphasis on evidence-based practice and the intro-
duction of the notion of recovery into mental health care.
We use as a metaphor for this examination the relationship
suggested by a colleague, who asked one of us if the
relationship between recovery and evidence-based practice
was like that between oil and water; in other words, a
relationship between fundamentally different substances or
concepts that cannot be blended or integrated. As we
responded to him, and as we will argue in the following, we
view the relationship between evidence-based practices
and recovery to be more like the relationship between oil
and vinegar than that between oil and water. Whether
raspberry or balsamic-flavored, we believe the two taken
together can produce a tasty vinaigrette, dynamically
generating a whole that is greater than the sum of its parts.
In order to build this case, we first review briefly the
core aspects of recovery and recovery-oriented practice as
they have emerged in mental health and then describe some
of the key principles of evidence-based medicine. We
L. Davidson (&) ? R. E. Drake ? T. Schmutte ? T. Dinzeo ?
Department of Psychiatry, Program for Recovery and
Community Health, Yale University School of Medicine,
Erector Square 6 West, Suite #1C, 319 Peck Street,
New Haven, CT 06513, USA
L. Davidson ? R. E. Drake ? T. Schmutte ? T. Dinzeo ?
Psychiatric Research Center, Dartmouth Medical School,
Hanover, NH, USA
Community Ment Health J
acknowledge that both of these developments have been
prominent primarily in English-speaking countries over the
last decade, but hope that some of the lessons that we have
been learning might also be relevant to other countries into
which these concepts are just beginning to be introduced.
Following these reviews, we consider several possible
relationships between the two concepts, which we con-
ceptualize as follows: (1) the role of evidence within the
context of shared decision-making processes; (2) the
recovery paradigm as shifting the standard of what are
considered to be desirable and meaningful outcomes (i.e.,
what constitutes ‘‘evidence’’); and (3) the need to gather
evidence on the efficacy of certain newly emerging,
recovery-oriented practices. Exploring each of these pos-
sibilities will demonstrate how the relationship between
recovery and evidence-based practices is far from simple or
straightforward, suggesting a more thoughtful approach in
which each can inform the work of the other.
Recovery and Recovery-Oriented Practice
in Mental Health Care
We focus this brief review on the following three concepts:
recovery from serious mental illness, being ‘‘in recovery’’
with a serious mental illness, and recovery-oriented prac-
tice. It first is necessary to clarify the differences between
the first two concepts—recovering from and being in
recovery in relation to a serious mental illness—as it has
been our impression that a good deal of the confusion
surrounding recent uses of the term recovery in mental
health has been due to a confusion between these two
distinct concepts. For the purposes of this discussion, we
suggest taking the phrase ‘‘recovery from’’ serious mental
illness to refer to eradication of the symptoms and ame-
lioration of the deficits caused by the illness. This sense of
recovery has been the traditional focus of treatment
effectiveness and longitudinal outcome studies and refers
to a person getting to a point (i.e., recovering) at which he
or she no longer has a mental illness. The notion of being
‘‘in recovery,’’ however, comes from a very different
source and refers to a very different phenomenon (Jacobson
2001). Developed and promoted initially by the mental
health consumer movement, championed by the field of
psychiatric rehabilitation, and only recently moving into
the mainstream of mental health, this sense of recovery
refers to learning how to live a safe, dignified, full, and
self-determined life, at times in the face of the enduring
symptoms of a serious mental illness. At its most basic,
recovery from refers to getting rid of or overcoming a
mental illness, while being in recovery refers to reclaiming
and living one’s life in the face of a mental illness which
has not and may not go away (Davidson and Roe 2007).
This distinction has important implications for both
research and practice, as we explain below.
Recovery from Serious Mental Illness
Recovery from serious mental illness is not a new concept.
As long as there has been a concept of mental illness, there
have been people who have recovered from it; historical
examples include Percival, Judge Schreber, and Clifford
Beers. Some people are fortunate to recover fully from a
first and only episode of mental illness, while for others
recovering from the disorder may take place over a longer
period of time. While the lineage of this insight can be
traced back to Bleuler, at least since the 1970s we have
documented, in fact, that many people, between 25 and 65%
of a given sample of people diagnosed with a serious mental
illness, will experience significant improvements in their
condition over time resulting in a good outcome ranging
from only mild impairment to full recovery—at times even
after many years of profound suffering and disability. (e.g.,
Davidson et al. 2005; Davidson and McGlashan 1997; Jobe
and Harrow 2005; McGlashan 1988; WHO 1979). Recov-
ery in this sense, as we noted above, refers to the absence of
symptoms, signs, and functional impairments; people who
recover no longer have a mental illness.
We know relatively little about this form of recovery and
about the extent to which it is promoted by existing treat-
ments. For example, despite the initial fanfare accompa-
nying the release of a second generation of anti-psychotic
medications, two recent large, multi-site, randomized,
double-blind studies offer persuasive evidence that these
medications are no more effective than their predecessors
(Jones et al. 2006; Lieberman et al. 2005). It remains the
case that only about 70% of people diagnosed with
schizophrenia will experience a decrease in symptoms in
response to medication and that this decrease will be limited
primarily to the domain of positive symptoms. The majority
of the disability associated with mental illness, however,
appears to be due to the hidden impacts of negative symp-
toms and neurocognitive deficits (Davidson and McGlashan
1997). While both first and second generation medications
tend to be associated with improvements in psychotic
symptoms, it also remains unclear the degree to which these
gains make significant differences in the actual day-to-day
lives of people with schizophrenia. The surprisingly high
rates of medication discontinuation and switching in the two
large-scale medication studies mentioned above (e.g., 74%
over an 18-month period in the CATIE study) suggest that
whatever the perceived benefits of these medications may
be they often are outweighed by their serious side effects.
Cognitive approaches have shown promise in aug-
menting medication and in addressing the neurocognitive
deficits that medications do not affect. Adding cognitive
Community Ment Health J
behavioral therapy to medication, for example, may
increase the rate of reduction of positive symptoms from 41
to 59% and appears to produce results which are sustained
over time (Kingdon and Turkington 2005; Zimmermann
et al. 2005). A meta-analysis of seventeen RCTs of cog-
nitive remediation, training, or rehabilitation—which
involve the person’s participation in exercises which aim to
learning, attention, executive functioning)—found positive
findings in fourteen studies, with small-to-moderate effects
for improvements in neuropsychological performance and
cognitive functioning (Twamley et al. 2003). While all of
these findings show promise, few studies have included an
assessment of real world functioning and further study is
required to assess the degree to which these gains gener-
alize to everyday life tasks.
Thus, while people may be experiencing improvements
in their condition over time, some to the point of recov-
ering fully from it, these improvements cannot be attrib-
uted directly to existing treatments. First of all, we know
that only one-third of individuals with serious mental ill-
nesses receive any specialty mental health care (DHHS
1999) and that among those who do receive care, most
people do not receive evidence-based practices. Secondly,
even among those individuals who are fortunate to receive
evidence-based treatments, these treatments demonstrate
only small to moderate effects. An example is provided by
the case of Family Psychoeducation (FP), which represents
one of the few evidence-based practices for which the real
world impact of the intervention has been estimated. A
recent Cochrane meta-analysis that included 43 random-
ized and quasi-randomized controlled trials found evidence
that FP reduces hospital admission, encourages medication
adherence, and improves general social functioning and
levels of expressed emotion within the family. The authors
of this review noted, however, that intervention effects
across these studies tended to be moderate, with the pri-
mary benefit of FP being that it reduces the risk of relapse.
In this case, they suggested that the data accounted for
prevention of one relapse per year for every eight families
participating in the intervention (Pharoah et al. 2006).
‘‘Recovery’’ in the sense of the remission of all symp-
toms and deficits therefore can not be attributed directly to
our existing treatments and remains poorly understood.
Although a host of factors have been suggested as con-
tributing to recovery, from a shorter duration of untreated
psychosis (e.g., Marshall and Rathbone 2006; Perkins
et al. 2005) to the more supportive social networks or
nutritional/dietary habits of developing countries (e.g.,
Christensen and Christensen 1988; Gupta 1992; Hopper
and Wanderling 2000; Peet 2004), we have made little
progress in accounting for the heterogeneity in outcome
that has been established for well over a 100 years.
Being in Recovery
During these same 30 years, however, there have been
dramatic developments in the consumer and family
movements in mental health, and along with these devel-
opments the emergence of the second form of recovery
described above. This notion of being ‘‘in recovery’’ refers
to a process rather than to an outcome (e.g., Anthony 1993;
Bellack 2006; Deegan 1988; Jacobson 2001; Spaniol and
Koehler 1994). This sense of recovery was borrowed by the
consumer movement from their peers in the addiction self-
help community, who considered themselves to be ‘‘in
recovery’’ as long as they were making active efforts to
manage their sobriety and rebuild a meaningful life in the
wake of their addiction. What appears to have been most
appealing about this notion to people with serious mental
illnesses was that their peers with addictions had been
reclaiming their lives and the responsibility for making
their own decisions even without first being cured of what
they considered to be a life-long condition. As there also is
not yet a cure for serious mental illnesses, people with
these conditions argued in a similar vein that they should
be able to reclaim their lives and autonomy without first
having to be cured from mental illness.
This notion of being in recovery therefore does not have
as much to do with a person’s level or degree of symptoms,
deficits, or pathology as much as it does with how the
person is managing his or her life in the presence of an
enduring illness—or perhaps how the person is managing
an enduring illness in the context of his or her ongoing life.
This form of recovery thus represents a personal, social,
and political reality as much as it does a medical one, as it
is significantly impacted by who the person is, the nature
and density of his or her social support network and milieu,
and the rights and responsibilities accorded or denied to the
person by virtue of the society in which he or she lives
(e.g., through stigma and discrimination; Davidson 2006).
While this form of recovery has been identified and
developed primarily by the consumer movement, it has
been defined in various ways by people in recovery, the
U.S. government, psychiatric rehabilitation practitioners,
and researchers (DHHS 2003, 2005, 2007; Onken et al.
2007). Where all authorities seem to agree is that this form
of recovery involves the person’s self-determined pursuit
of a meaningful life in the communities of his or her choice
in the face of an enduring impairment.
Several factors have been identified as appearing to be
essential to this form of being in recovery—as they may
also be to persons’ efforts to manage other prolonged
health conditions—which have yet to be investigated as
Community Ment Health J
components of the active treatments described above. As
listed, for example, in the SAMHSA (Substance Abuse
and Mental Health Services Administration of the U.S.
Government) consensus definition of recovery, these fac-
tors include hope, dignity, respect, choice, social support,
involvement in meaningful activities, and having a sense of
meaning and purpose in one’s life (DHHS 2007). Care
which is oriented to promoting this form of recovery has, as
a result, been defined as person-centered, strengths-based,
collaborative, and empowering. Consistent with the goals
of self-determination and enabling people to pursue
meaningful lives in the community despite the lack of a
cure for their condition, advances in psychiatric rehabili-
tation focus on enhancing functioning and supporting
people in taking part in routine adult roles in the commu-
nity such as employment, education, and socialization.
Supported housing that offers a high degree of consumer
choice has been highly effective in enabling people to
obtain and maintain safe and affordable independent
housing with community-based support (Tsemberis et al.
2004). Other rehabilitative strategies such as supported
employment and assertive community treatment also
appear to demonstrate greater impact than the active
treatments reviewed above.
Supported Employment (SE) programs help individuals
to find jobs of their choice in integrated work settings that
pay minimum wage or higher and offer ‘on the job’
training and support when requested by the worker (as
opposed to offering training and then ‘placing’ people in
jobs). SE tailors placements to individual capabilities and
interests, offers time-unlimited vocational support, and
may include collaboration with a person’s supervisors and
coworkers (Bond et al. 2001a). A recent meta-analysis of
11 RCTs examined the efficacy of SE for people with
schizophrenia. In the 5 investigations that compared this
approach to conventional rehabilitation services, over half
(51%) of participants receiving SE worked competitively
versus less than twenty percent (18%) of those in the
comparison groups (weighted mean effect size was
large, = 0.79). Overall, those receiving SE were about four
times more likely than control participants to obtain com-
petitive work (Bond et al. 2001a). When quasi-experi-
mental studies of converting day treatment to SE are
included, between 40 and 60% of participants enrolled in
SE obtained competitive employment compared to less
than 20% of those not enrolled in SE, resulting in a large
unweighted effect size of .85 (Becker et al. 2007).
Since its inception, Assertive Community Treatment
(ACT) has focused on in vivo skills training and commu-
nity support and has aimed to improve quality of life by
delivering services in the community around the clock,
7 days per week. A Cochrane review found that those
receiving high-fidelity ACT were significantly more likely
to remain in contact with services and less likely to be
admitted to the hospital (Marshall and Lockwood 2000). In
addition, a more recent literature review of 25 RCTs con-
cluded that ACT significantly increased housing stability
and moderately improved symptoms and quality of life
(Bond et al. 2001b). As a service delivery modality which
can also incorporate other interventions such as SE, cog-
nitive behavioral therapy, and the emerging and promising
practice of peer support (Davidson et al. 2006; Solomon
2004), ACT remains one of the most effective services
currently available for people with serious mental illnesses.
Most recently, additional modifications to ACT have been
proposed to enhance its potential to be a more recovery-
oriented practice, stressing the importance of such issues as
self-determination, choice, and empowerment (e.g., Salyers
and Tsemberis 2007).
In summary, then, recovery-oriented practices identify
or her condition while pursuing a meaningful life in the
community. Those interventions that aim to support the
person in managing the illness appear to have a more robust
evidence base than those that aim to reduce or eliminate the
or might not relate to an emphasis on evidence-based prac-
tice in mental health.
Evidence-Based Practice in Mental Health
Evidence-based practices in mental health derive from the
general movement toward evidence-based medicine in
general health care (Sackett et al. 1996). The evidence-
based practices movement assumes that mental health
programs and practitioners have an ethical obligation to
offer the most effective treatments available to people with
mental illnesses (IOM 2001; National Advisory Mental
Health Council Behavioral Science Workgroup 1999;
DHHS 1999, 2003) and asserts that effectiveness is to be
established and confirmed by scientific evaluation. What
this means and how to implement the concept have been
debated extensively (Drake et al. 2003, 2005).
Evidence-Based Medicine Builds on Basic Health Care
The history and philosophy of medicine recognize a variety
of basic values as the foundation for ethical and effective
practice (Culver and Gert 1982). These include treating
patients and their families with respect, being truthful in
providing patients and families with accurate information,
and first avoiding harm and then attempting to be helpful to
people in distress. Chief among the values of modern
Community Ment Health J
medicine also is the principle of autonomy: people have a
fundamental right to make decisions about their own
bodies. Evidence-based medicine fully endorses these and
other basic health care values.
Sharing the scientific evidence collaboratively is the sine
qua non. Without knowing the evidence, practitioners
cannot be certain that they are avoiding harm and doing
their best to help their patients. Without understanding the
evidence, people cannot be truly informed and thus cannot
make informed choices and decisions about their care.
Only in exceptional situations are coercion, involuntary
treatment, or other contradictions to these basic health care
values considered morally justified. For example, the per-
son who is in a delirium or coma due to a stroke and is not
competent to understand his or her current medical needs
may require intervention prior to being able to provide
informed consent. In this situation, relying on next-of-kin,
a court-appointed guardian, or medical urgency is consid-
ered morally justified.
The situation is the same in psychiatry. A person who is
seen by a psychiatrist for treatment of depression has a
right to receive full information about the illness, to
understand the available treatments and their expected
outcomes and potential side effects, to make an informed
choice, and to give consent regarding treatment. The per-
son’s rights may be overlooked only in specific circum-
stances, such as when he or she presents an immediate life-
threatening risk to self or others. In such extreme situa-
tions, state laws typically provide for a process of substi-
tutive decision-making, for example, by a panel of doctors
and lawyers, to protect the person and/or the threatened
Evidence-Based Medicine Assumes that Scientific
Evidence is Often Complicated, Hierarchical,
Ambiguous, and Limited
researchers often disagree about the scientific evidence,
that the evidence is often limited to specific types of dis-
orders or settings, and that there is no strong evidence to
inform many common clinical decisions. These legitimate
concerns have been considered extensively in the evidence-
based medicine literature, resulting in consensus that the
available scientific evidence has many limitations and must
be used judiciously and in conjunction with other important
considerations (Guyatt and Rennie 2002).
The gold standard for scientific evidence regarding
treatment is the systematic review of several double-blind
randomized controlled trials. In practice, however, ran-
domized controlled trials are rare and systematic reviews
are even more rare. Even when available, randomized
controlled trials are usually conducted under constrained
research conditions, such as using university research
settings, highly selected patients, highly trained clinicians,
carefully monitored treatments, and other special condi-
tions. In other words, research trials often do not include
typical settings (such as rural clinics), minority groups
(such as Americans of Hispanic origin), clinical popula-
tions (such as persons with multiple previous episodes or
serious co-morbidities), practicing clinicians (such as
those with large case loads), or other features of routine
practice. Thus, the available scientific evidence always
has serious limitations. One way in which evidence-based
medicine responds to these concerns is by considering
various types and levels of evidence within a loosely
conceived hierarchy. For example, in the absence of
randomized controlled trials, quasi-experimental studies
are to be considered, though they have less inferential
Some clinical decisions can be guided by higher-level,
more reliable, and more extensive evidence than others, but
there are always limitations to the quality of the evidence
in general and to the applicability of the evidence to spe-
cific people and situations (Whitley 2007). At the same
time, some useful and relevant evidence is almost always
available. In the typical clinical situation, the available
evidence, even if strong, does not match the current patient
perfectly. For example, several randomized clinical trials
show the effectiveness of a particular intervention, but the
current patient differs from the patients in those trials. How
do the clinician and patient use the evidence in this situa-
tion? This is one area in which clinical experience and
understanding the effects of physiological variation, co-
occurring illnesses, treatments, and other factors are criti-
cal. For example, a competent physician has to know that
older patients, those of Asian heritage, or those with
impaired liver function may be vulnerable to certain side
effects and may require smaller dosages of some
Evidence-Based Medicine Recognizes that Factors
Other than Scientific Evidence are Also Critically
Important in Medical Decisions
A second way in which evidence-based medicine deals
with the limitations of evidence is to assume that scientific
evidence is only one important component of decision-
making (Guyatt and Rennie 2002). Additional components
include the person’s own values, goals, and preferences; his
or her interest in being an active participant in decision-
making; his or her cultural context and local support sys-
tem; and the local resources. All of these factors must be
weighed in decisions. In fact, these other factors often
override the scientific evidence, especially in situations
where the evidence is weak (O’Connor et al. 2007).
Community Ment Health J
Without access to accurate and understandable scientific
information, patients and practitioners are likely to be
influenced by misinformation from the press and internet,
the prejudices of their families or social contacts, the biases
of vested interest groups (e.g., marketing from industry), or
the idiosyncratic opinions of local providers. With access
to accurate information, people can and do make choices
that are highly differentiated. That is, their treatment
choices often differ from what their providers might
recommend, from what their peers choose, and from what
they would choose in the absence of such information
(Wennberg 1988, 1991). Because situations, information,
and preferences are complicated, people often report that
decision aids help them to clarify their understanding of the
information, their values, and their preferences (O’Connor
et al. 2007).
This emphasis on accurate information and shared deci-
sion-making requires a fundamental shift away from the
do. Shared decision-making suggests that the patient and
clinician consider information about reasonable and avail-
effects), share their expertise and perspectives, discuss
alternatives, and arrive at plans that both can endorse
(Deegan and Drake 2006). Most medical decisions today
involve complicated trade-offs that should be weighed and
considered carefully in this manner (O’Connor et al. 2007).
A critical implication of the emphasis on patient pref-
erences and shared decision-making is that it affords a
different view of outcomes as well as of treatments. For
example, many people value functional status and quality
of life as much or more than they value symptom control
and certainly more than they value physiologic measures.
For example, people with mental illnesses often value
working, avoiding medication side effects, and improving
quality of life more than they value complete symptom
control (Deegan 2007).
One common misconception regarding evidence-based
practice in the mental health field is that it prescribes a
standard, invariant approach for all people. As the fore-
going discussion makes clear, the philosophy of evidence-
based medicine is that the available evidence is just one
factor in the decision-making process. A treatment with the
strongest evidence may not be the optimal treatment for a
particular individual, who may have specific risk factors
that preclude the treatment, for whom the treatment may
not be available or affordable in a given situation, or who
may have a different preference. Consider the example of
clozapine. This drug is an evidence-based alternative for
patients with schizophrenia who do not respond well to
other antipsychotic medications (Lehman et al. 2004).
However, many factors might exclude clozapine as the
treatment of choice for a particular person. The person may
have a history of medical problems that would make clo-
zapine especially dangerous; the person may prefer not to
have weekly blood monitoring; the person may choose to
avoid the characteristic side effects of clozapine, such as
severe sedation and metabolic risks; or clozapine may not
be available or affordable for a variety of reasons. All of
these factors could override a decision to try clozapine. In
shared decision-making, they would be considered care-
fully and weighed against potential benefits by the patient
and the physician.
This cursory overview of the guiding principles of
recovery-oriented practice and evidence-based medicine—
both of which assume that people have a right to make
informed decisions about their own care—would seem to
suggest that the two are quite compatible. Why, then, do
questions persist about the relationship between the two? If
they mix well enough to produce a tasty vinaigrette, as we
have tried to show, what is all the fuss about? We next
consider three ways in which the two approaches may
appear to be in conflict, but we believe actually produce a
creative and useful tension.
Three Possible Relationships Between Recovery
and Evidence-Based Practice
Is Recovery Evidence-Based?
We consider this question separately in relation to each of
the two forms of recovery described above. In terms of
recovery from serious mental illnesses, the answer is yes, as
we have known at least since the time of Bleuler that people
at times experience complete remissions from psychosis
and that others often experience improvements in symp-
toms and functioning over time. These findings have now
been confirmed consistently and repeatedly over the last
30 years through a series of rigorous longitudinal studies in
which a significant portion of any given sample of people
with serious mental illnesses were found to experience at
least partial, and many up to full, recovery over time
(Davidson et al. 2005). We do not yet know why or how this
comes about, as this form of recovery typically occurs over
a longer period of time and does not appear to be a direct
result of the treatments we currently offer for the disorder.
But, yes, there is definitely an evidence base that has
established that many people will recover over time, despite
the fact that this form of recovery does not seem to be
attributable to the evidence-based practices we deploy.
In relation to being in recovery, there is a robust evi-
dence base that many, if not most, people can figure out
how to live with and manage the disorder over time
(Davidson et al. 2005). At the same time, this question may
Community Ment Health J
be considered irrelevant to the issue at hand, as what is
primarily at stake in the notion of being in recovery is not a
medical or scientific condition but the person’s need for,
and right to, finding a way to live a meaningful life in the
face of the illness. Unless, until, and only for as long as a
person poses a serious and imminent risk to self or others,
is gravely disabled, or is deemed to lack the capacity to be
in charge of his or her life by a judge, people with serious
mental illnesses retain the human, civil, and legal right to
self-determination. This is not a question of evidence, but a
question of statute. Some mental health practitioners may
at first be uncomfortable with this idea, due to the common
perception that people with mental illnesses demonstrate
poor judgment. On this score, however, there is no com-
pelling evidence. The evidence that exists, in fact, suggests
that the vast majority of people with serious mental ill-
nesses—despite whatever cognitive deficits they may
experience—nonetheless remain legally competent to
make their own informed choices and typically do so on a
par with people who do not have mental illnesses (Stroup
et al. 2005; Roe et al. 2001).
What the introduction of recovery adds to evidence-
based practice is thus twofold. One, it instills a sense of
humility in the face of the evidence that exists, high-
lighting for us the fact that many people do in fact recover
from mental illnesses, and many more find ways to live
meaningful lives in the face of enduring illnesses, with or
without help from mental health services. This fact—
which has been pointed out by the consumer movement
consistently over the last 30 years—may initially appear
discouraging, but should encourage us instead to be
hopeful that in future generations we will discover new
and more effective ways to help people manage and/or
overcome these illnesses. In this respect, the work is just
beginning. Two, the introduction of recovery helps to
move psychiatry closer to the mainstream of contemporary
medical practice, in which evidence-based medicine has
become the expectation rather than the exception. This is
because recovery reinforces the point that patients are
partners in the treatment and rehabilitation enterprise, as
they retain the right to make their own informed choices
based on, or despite, the evidence—just as is true in other
branches of medicine.
Are Evidence-Based Practices Recovery-Oriented?
While the field has yet to come to a consensus on what
precisely comprises recovery-oriented care, for the sake of
this discussion we will suggest, as we did above, that
recovery-oriented practices are distinguished by being
empowering. Given this definition, there are two possible
answers to the question above as well.
A first possibility would be to examine each evidence-
based practice individually and assess the degree to which
it was consistent with the principles of recovery-oriented
care. Tools are beginning to appear that could be used for
this purpose (e.g., O’Connell et al. 2005), asking such
questions as to what degree does this practice invite and
honor people’s own choices and preferences, to what
degree does this practice elicit, identify, and build on
people’s strengths, to what degree does this practice focus
on enabling people to participate in naturally occurring
community activities as opposed to remaining within the
confines of the mental health system, etc. In this way,
different practices would fare better or worse in terms of
recovery orientation. We might even suggest that certain of
the evidence-based practices described above, such as
ACT, SE, and supported housing, are effective in part
precisely because they are recovery-oriented. These, and
other similar, approaches may work because, when deliv-
ered correctly, they emphasize people choosing their own
goals, taking responsibility for managing their own con-
dition, pursuing their own dreams, etc.
Using this approach, supported employment, for exam-
ple, would score high by virtue of its built-in focus on
people’s own interests and choices, and by virtue of its aim
to help people to attain their own goals related to com-
petitive jobs in the community. Similarly, supported
housing that offers people a high degree of choice can be
considered to enhance recovery not only because it enables
people to secure a home of their own but also because it
has been found to increase their sense of mastery as well as
reduce psychiatric symptoms (Greenwood et al. 2005).
ACT, on the other hand, might be vulnerable to criticism
for using coercion and for historically emphasizing symp-
tom reduction and avoidance of hospitalization over a
person’s own goals (although this may now be changing).
Finally, in other cases, such as that of psychopharmacol-
ogy, the presence or absence of a recovery orientation
would be found in how the treatment was delivered rather
than in the treatment per se (e.g., Deegan 2007; Noordsy
et al. 2000).
A second approach is to raise the question of what
counts as evidence in the studies that have led to a par-
ticular practice being included in the list of practices that
are ‘‘evidence-based.’’ In this respect, very few studies
have used outcome measures that are recovery-oriented,
settling typically (as we saw above) for such measures as
reduction in symptoms, increases in neuropsychological
performance, or reductions in hospitalizations, with little
reference to the degree to which these changes made a
difference in a person’s day-to-day life. Extrapolations of
course can be made, such as it is better on the whole to
remain out of the hospital or to have fewer symptoms, but
these indicators speak only to minimizing illness and
Community Ment Health J
deficit and do not address maximizing the person’s
opportunities for a meaningful life. But the two processes
of minimizing illness and maximizing life may not be the
same (Davidson et al. 2009). In fact, for some people
taking risks to move ahead in their lives, such as returning
to school or taking a job, may temporarily increase their
level of symptoms and may even precipitate a hospital-
ization. In the longer term, however, these were indicators
of progress rather than dysfunction. When viewed from this
perspective, an entirely different domain of outcomes
becomes important, centering on the issues of self-agency
and self-efficacy. In this respect, research has only begun to
address the processes of developing hope, taking control of
one’s treatment and recovery, and defining and pursuing
one’s personal goals.
With this shift from symptom reduction to personal
growth, the introduction of recovery raises the bar related
to what can be considered a ‘good enough’ outcome in
mental health. With recovery, we can no longer focus
solely on reducing symptoms and suffering, which remain
nonetheless worthy goals as well. But we need to com-
plement this focus with one on the process of developing
achieving meaningful life goals, and, in the words of the
New Freedom Commission on Mental Health, ‘‘living,
working, learning, and participating fully in the commu-
nity’’ (DHHS 2003). This poses a significant challenge
both to mental health practice and to research, in that
assessing such outcomes may turn out to be nearly as
difficult as promoting them. This, however, is one of the
two main challenges posed to mental health by the intro-
duction of concepts of recovery, the other one, mentioned
above, being that of respecting individuals with serious
mental illnesses as partners in the treatment and rehabili-
Are Recovery-Oriented Practices Evidence-Based?
Finally, there are two answers to this last question as well.
First, to the degree that recovery-oriented practice refers to
a collaborative approach involving shared decision-making
between the person and the practitioner, the answer is both
yes and, once again, that the question is not relevant. Data
have shown that shared decision-making is a more effec-
tive approach to care than a traditional authoritarian one
(Priebe et al. 2007), and in this sense the answer to the
question is yes. To the degree that the question is posed to
challenge the person’s active role in determining his or her
own care, however, we have made the point above that this
is a question of right not of evidence. Even if we did not
already have evidence of its effectiveness, it remains a
cornerstone of a democratic society that each adult is free
to make his or her own choices unless there are clear and
compelling reasons for that freedom to be curtailed.
Emergency commitments are obviously an example of the
latter, as is when a judge determines that a person needs to
be conserved. But in all other circumstances, people with
mental illnesses remain in control of their ‘‘own person,
free from all restraint or interference of others, unless by
clear and unquestioned authority of law’’ (as determined by
the United States Supreme Court in Union Pacific Railway
Co. v. Botsford). The introduction of psychiatric advance
directives as a new tool for recovery-oriented practice aims
to enable people to retain these rights even when they have
become acutely or temporarily disabled, delineating
beforehand the person’s preferences should such situations
The second answer to the question of whether recovery-
oriented practices are evidence-based may be the least
satisfactory, as it is ‘‘not yet.’’ Because of the relative
newness of the concept of recovery and its application to
practice, the field has not yet accumulated an adequate
evidence base to speak to the effectiveness of certain
recovery-oriented practices. We are taking recovery-
oriented in this sense to refer not to the overall principles of
self-determination, strengths-based, etc., as we did above,
but to specific concrete practices that are currently emerg-
ing in the field. We do have a solid evidence base for sup-
ported housing and supported employment, as we noted
above, but other developments such as peer support and
peer-run programs and businesses are only now beginning
to be studied in rigorous ways. Thus far, the evidence for
peer support, for example, shows that people in recovery
can provide mental health services at least as effectively as
people who are not in recovery (or who have not disclosed
that history) (e.g., Davidson et al. 2007). What has yet to be
determined are the ways in which peer support may have a
value added component deriving specifically from the
provider’s personal history of recovery and his or her use of
that history within the relationship (Dixon et al. 1994;
Solomon 2004). On this matter, as is in relation to other
uniquely recovery-oriented practices (such as recovery
mentors or coaches, peer-run respite programs, social
cooperatives, recovery centers) much remains to be learned.
As recovery-oriented practices emerge and proliferate,
an emphasis on evidence-based practices will be essential
in preventing the mental health field from going down yet
another blind alley. Recovery advocates should in this way
welcome, rather than reject, a focus on evidence, as they
have nothing to lose and everything to gain in showing that
this new approach to care is at least as effective, if not more
so, than the approach which preceded it. As long as evi-
dence-based practice incorporates and honors the active
role of the person as a partner in a collaborative health care
enterprise, recovery advocates will surely want people with
serious mental illnesses to be able to make informed
Community Ment Health J
choices on their own behalf. How can such choices be
made in the absence of evidence? Integrating the emphasis
on evidence with recovery might require a broader defini-
tion of what constitutes evidence, to include, for example,
first-hand experiences. Although evidence-based medicine
recognizes a hierarchy of evidence, it has not thus far
included first-hand accounts. But it is possible, of course, to
do so (e.g., Flanagan et al. 2007).
As we hope is by now evident from the discussion above,
the relationship between recovery and evidence-based
medicine is neither contradictory nor mutually exclusive. It
is neither straightforward nor one-directional. There are
many possible relationships between the two value bases
and sets of principles, and through several of these rela-
tionships we think both recovery and evidence-based
practices can benefit. Stated simply, we need both to listen
to people in recovery about their needs, preferences, val-
ues, and choices and also to find out which interventions
are most effective in helping them to achieve the kinds of
lives they wish to lead. While involving people in recovery
in research through adoption of a participatory paradigm
may be one particularly promising strategy for accom-
plishing these goals (Wallcraft et al. 2009), the develop-
ment of respectful and collaborative relationships between
health care practitioners and individuals with mental ill-
nesses is the best way to ensure the delivery of evidence-
based medicine over the long term. This is the vinaigrette
which can be produced when each brings out and builds on
the best in the other.
Anthony, W. A. (1993). Recovery from mental illness: The guiding
vision of the mental health service system in the 1990s.
Psychosocial Rehabilitation Journal, 16(4), 11–23.
Becker, D., Whitley, R., Bailey, E. L., & Drake, R. E. (2007). Long-
term employment trajectories among participants with severe
mental illness in supported employment. Psychiatric Services,
Bellack, A. S. (2006). Scientific and consumer models of recovery in
schizophrenia: Concordance, contrasts, and implications. Schizo-
phrenia Bulletin, 32, 432–442.
Bond, G. R., Becker, D. R., Drake, R. E., Rapp, C. A., Meisler, N.,
Lehman, A. F., et al. (2001a). Implementing supported employ-
ment as an evidence-based practice. Psychiatric Services,
Bond, G. R., Drake, R. E., Mueser, K. T., & Latimer, E. (2001b).
Assertive community treatment for people with severe mental
illness. Disease Management and Health Outcomes, 9, 141–159.
Christensen, O., & Christensen, E. (1988). Fat consumption and
schizophrenia. Acta Psychiatrica Scandinavica, 78, 587–591.
Culver, C. M., & Gert, B. (1982). Philosophy in medicine: Concep-
tual and ethical issues in medicine and psychiatry. London:
Oxford University Press.
Davidson, L. (2006). What happened to civil rights? Psychiatric
Rehabilitation Journal, 30, 11–14.
Davidson, L., Chinman, M., Sells, D., & Rowe, M. (2006). Peer
support among adults with serious mental illness: A report from
the field. Schizophrenia Bulletin, 32, 443–450.
Davidson, L., Harding, C. M., & Spaniol, L. (2005). Recovery from
severe mental illnesses: Research evidence and implications for
practice (Vol. 1). Boston: Center for Psychiatric Rehabilitation
of Boston University.
Davidson, L., & McGlashan, T. (1997). The varied outcomes of
schizophrenia. Canadian Journal of Psychiatry, 42, 34–43.
Davidson, L., & Roe, D. (2007). Recovery from versus recovery in
serious mental illness: One strategy for lessening confusion
plaguing recovery. Journal of Mental Health, 16(4), 1–12.
Davidson, L., Tondora, J., O’Connell, M. J., Lawless, M. S., & Rowe,
M. (2009). A practical guide to recovery-oriented practice:
Tools for transforming mental health care. New York: Oxford
Deegan, P. E. (1988). Recovery: The lived experience of rehabilita-
tion. Psychosocial Rehabilitation Journal, 11(4), 11–19.
Deegan, P. E. (2007). The lived experience of using psychiatric
medication in the recovery process and a shared decision-making
program to support it. Psychiatric Rehabilitation Journal, 31,
Deegan, P. E., & Drake, R. E. (2006). Shared decision making and
medication management in the recovery process. Psychiatric
Services, 57, 1636–1639.
Department of Health, Human Services. (1999). Mental health: A
report of the surgeon general. Rockville, MD: U.S. Department
of Health and Human Services, U.S. Public Health Service.
Department of Health, Human Services. (2003). Achieving the
promise: Transforming mental health care in America. Presi-
dent’s new freedom commission on mental health. Final report
(DHHS Pub. No. SMA-03-3832.). Rockville, MD: U.S. Depart-
ment of Health and Human Services.
Department of Health, Human Services. (2005). Transforming mental
health care in America. Federal action agenda: First steps.
Rockville, MD: Substance Abuse and Mental Health Services
Department of Health and Human Services. (2007). National
consensus statement on mental health recovery. Retrieved
December 14, 2007, from http://mentalhealth.samhsa.gov/publi
Dixon, L., Krauss, N., & Lehman, A. (1994). Consumers as service
providers: The promise and challenge. Community Mental
Health Journal, 30, 615–625.
Drake, R. E., Merrens, M. R., & Lynde, D. W. (2005). Evidence-
based mental health practice: A textbook. New York: W. W.
Drake, R. E., Rosenberg, S. D., Teague, G. B., Bartels, S. J., &
Torrey, W. C. (2003). Fundamental principles of evidence-based
medicine applied to mental health care. Psychiatric Clinics of
North America, 26, 811–820.
Flanagan, E., Davidson, L., & Strauss, J. S. (2007). Incorporating
patients’ subjective experiences into the DSM-V. American
Journal of Psychiatry, 164(3), 391–392.
Greenwood, R. M., Schaefer-McDaniel, N., Winkel, G., & Tsemberis,
S. (2005). Decreasing psychiatric symptoms by increasing
choice in services for adults with histories of homelessness.
American Journal of Community Psychology, 36(4), 223–238.
Gupta, S. (1992). Cross-national differences in the frequency and
outcome of schizophrenia: A comparison of five hypotheses.
Social Psychiatry and Psychiatric Epidemiology, 27, 249–252.
Community Ment Health J
Guyatt, G., & Rennie, D. (Eds.). (2002). Users’ guide to the medical Download full-text
literature: A manual for evidence-based clinical practice.
Chicago: American Medical Association Press.
Hopper, K., & Wanderling, J. (2000). Revisiting the developed versus
developing country distinction in course and outcome in
schizophrenia: Results from ISoS, the WHO collaborative
follow-up project. International study of schizophrenia. Schizo-
phrenia Bulletin, 26, 835–846.
Institute of Medicine (IOM). (2001). Crossing the quality chasm: A
new health system for the 21st century. Washington, DC:
National Academy Press.
Jacobson, N. (2001). In recovery: The making of mental health policy.
Nashville, TN: Vanderbilt University Press.
Jobe, T. H., & Harrow, M. (2005). Long-term outcome of patients
with schizophrenia: A review. Canadian Journal of Psychiatry,
Jones, P. B., Barnes, T. R. E., Davies, L., Dunn, G., Lloyd, H.,
Hayhurst, K. P., et al. (2006). Randomized controlled trial of the
effect on quality of life of second- vs. first-generation antipsy-
chotic drugs in schizophrenia. Archives of General Psychiatry,
Kingdon, D. G., & Turkington, D. (2005). Cognitive therapy of
schizophrenia. New York: Guilford Press.
Lehman, A. F., Kreyenbuhl, J., Buchanan, R. W., Dickerson, F. B.,
Dixon, L. B., Goldberg, R., et al. (2004). The schizophrenia
patient outcomes research team (PORT): Updated treatment
recommendations 2003. Schizophrenia Bulletin, 30, 193–217.
Lieberman, J. A., Stroup, T. S., McEvoy, J. P., Swartz, M. S.,
Rosenheck, R. A., Perkins, D. O., et al. (2005). Effectiveness of
antipsychotic drugs in patients with chronic schizophrenia. New
England Journal of Medicine, 353, 1209–1223.
Marshall, M., & Lockwood, A. (2000). Assertive community
treatment for people with severe mental disorders. Cochrane
Database of Systematic Reviews, (2). Art. No.: CD001089.
Marshall, M., & Rathbone, J. (2006). Early intervention for psychosis.
Cochrane Database of Systematic Reviews, (4). Art. No.:
McGlashan, T. (1988). A selective review of recent North American
long-term follow-up studies of schizophrenia. Schizophrenia
Bulletin, 14, 515–540.
National Advisory Mental Health Council Behavioral Science
Workgroup. (1999). Translating behavioral science into action.
Rockville, MD: National Institute of Mental Health.
Noordsy, D. L., Torrey, W. C., Mead, S., Brunette, M., Potenza, D., &
Copeland, M. E. (2000). Recovery-oriented psychopharmacol-
ogy: Redefining the goals of antipsychotic treatment. Journal of
Clinical Psychiatry, 61(Suppl. 3), 22–29.
O’Connell, M. J., Tondora, J., Evans, A. C., Croog, G., & Davidson,
L. (2005). From rhetoric to routine: Assessing recovery-oriented
practices in a state mental health and addiction system.
Psychiatric Rehabilitation Journal, 28(4), 378–386.
O’Connor, A. M., Wennberg, J. E., Legare, F., Llewellyn-Thomas, H.
A., Moulton, B. W., Sepucha, K. R., et al. (2007). Toward the
‘tipping point’: Decision aids and informed patient choice.
Health Affairs, 26(3), 716–725.
Onken, S. J., Craig, C. M., Ridgway, P., Ralph, R. O., & Cook, J. A.
(2007). An analysis of the definitions and elements of recovery:
A review of the literature. Psychiatric Rehabilitation Journal,
Peet, M. (2004). International variations in the outcome of schizo-
phrenia and the prevalence of depression in relation to national
dietary practices: An ecological analysis. British Journal of
Psychiatry, 184, 404–408.
Perkins, D. O., Hongbin, G., Boteva, K., & Lieberman, J. A. (2005).
Relationship between duration of untreated psychosis and
outcome in first-episode schizophrenia: A critical review and
meta-analysis. American Journal of Psychiatry, 162, 1785–1804.
Pharoah, F., Mari, J., Rathbone, J., & Wong, W. (2006). Family
intervention for schizophrenia. Cochrane Database of Systematic
Reviews, (4). Art. No.: CD000088.
Priebe, S., McCabe, R., Bullenkamp, J., Hansson, L., Lauber, C.,
Martinez-Leal, R., et al. (2007). Structured patient–clinician
communication and 1-year outcome in community mental
healthcare. British Journal of Psychiatry, 191, 420–426.
Roe, D., Joseph, L., & Shmuel, F. (2001). Comparing patients’ and
staff members’ attitudes: Does patients’ competence to disagree
mean they are not competent? Journal of Nervous and Mental
Disease, 189, 307–310.
Sackett, D. L., Rosenberg, W. M. C., Gray, J. A. M., Haynes, R. B., &
Richardson, W. S. (1996). Evidence based medicine: What it is
and what it isn’t: It’s about integrating individual clinical
expertise and the best external evidence. British Medical
Journal, 312, 71–72.
Salyers, M. P., & Tsemberis, S. (2007). ACT and recovery:
Integrating evidence-based practice and recovery orientation on
assertive community treatment teams. Community Mental Health
Journal, 43, 619–641.
Solomon, P. (2004). Peer support/peer provided services underlying
processes, benefits, and critical ingredients. Psychiatric Reha-
bilitation Journal, 27, 392–401.
Spaniol, L., & Koehler, M. (1994). The experience of recovery.
Boston: Center for Psychiatric Rehabilitation of Boston
Stroup, S., Appelbaum, P., Swartz, M., Patel, M., Davis, S., Jeste, D.,
et al. (2005). Decision-making capacity for research participa-
tion among individuals in the CATIE schizophrenia trial.
Schizophrenia Research, 80, 1–8.
Tsemberis, S., Gulcur, L., & Nakae, M. (2004). Housing first,
consumer choice, and harm reduction for homeless individuals
with a dual diagnosis. American Journal of Public Health, 94,
Twamley, E. W., Jeste, D. V., & Bellack, A. S. (2003). A review of
cognitive training in schizophrenia. Schizophrenia Bulletin, 29,
Wallcraft, J., Amering, M., & Schrank, B. (2009). Handbook of
service user involvement in mental health research. London:
Wennberg, J. E. (1988). Improving the medical decision-making
process. Health Affairs, 7(1), 99–106.
Wennberg, J. E. (1991). Outcomes research, patient preference, and
the primary care physician. The Journal of the American Board
of Family Practice, 4, 365–367.
Whitley, R. (2007). Cultural competence, evidence-based medicine,
and evidence-based practices. Psychiatric Services, 58, 1588–
World Health Organization. (1979). Schizophrenia: An international
follow-up study. Chichester: Wiley.
Zimmermann, G., Favrod, J., Trieu, V. H., & Pomini, V. (2005). The
effect of cognitive behavioral treatment on the positive symp-
toms of schizophrenia spectrum disorders: A meta-analysis.
Schizophrenia Research, 77, 1–9.
Community Ment Health J