Dr. Ranz is affiliated with the Department of Psychiatry, New York State Psychiatric Institute, Columbia University, 1051 Riverside Dr., Box 111, New York, NY 10032 (e-mail: firstname.lastname@example.org). Dr. Mancini is with the Department of Counseling and Clinical Psychology, Teachers College, Columbia University, New York City.
Over the past two decades "recovery" has become the dominant ideology among consumers and others who advocate for people with severe mental illness. Recovery has also been a central focus of policy makers, as exemplified in two recent federal reports (1,2), and has become identified as a model of care that an increasing number of professionals use when working with people with severe mental illness, especially in community settings. These various groups sometimes use different definitions of recovery, but most people who use the term agree that it represents a process that encourages people diagnosed as having severe mental illness to feel "hope, healing, empowerment, and connection" (3), giving them a sense of control over their lives toward the goal of maximizing their ability to function in the world.
Ideas regarding recovery gradually emerged both from first-person accounts of consumer-survivors in the 1970s and 1980s (4,5) and from long-term studies of schizophrenia that began to be published in the early 1980s, demonstrating partial or full recovery in 25% to 65% of each sample (6). These ideas have increasingly been translated into mental health practices, including an array of programmatic, clinical, and personal aspects of care that have been identified as supporting recovery (7,8,9). Although these recovery-oriented practices are heterogeneous and have been variously defined, there is increasing agreement on a number of broad principles (3,10,11). These include ensuring that services are client driven (for example, programmatic efforts to incorporate client preferences in service design and service provision), focused on life roles and community integration (for example, employment), and designed to maximize autonomous functioning outside of the mental health system (for example, independent housing and peer advocacy).
In an effort to further refine, measure, and support these practices, researchers have developed self-report measures of recovery-oriented practices (12). However, most or all of these measures have been specifically designed for frontline clinicians and program administrators. To our knowledge, no self-report measure of recovery-oriented practices has been developed for psychiatrists and there have been few if any prior investigations of psychiatrists' recovery-oriented practices. We view the development of a greater knowledge base on psychiatrists' recovery-oriented practices as a pressing need. Therefore, we developed a new survey specifically for psychiatrists. We based our survey content on previous guidelines for recovery-oriented practices (13). We hewed to conventional ideas of recovery-oriented practices, focusing on interventions that seek to enhance autonomous functioning and community integration. Given that our scale was newly developed, we described practices for each item in straightforward descriptive language to maximize face validity of the scale.
We administered our survey to alumni of Columbia University's Public Psychiatry Fellowship. Since its inception in 1981, the fellowship has been training psychiatrists who have expressed their intention to devote their careers to working in the public sector. Previous surveys have demonstrated that over 95% of fellowship alumni devote their careers to the public sector, 75% in leadership roles (14,15).
Our primary goal was to ascertain what recovery-oriented practices are actually employed by fellowship alumni who receive specific training in rehabilitation and recovery. Secondarily, we explored whether exposure to recovery concepts and attitudes toward leverage practices (for example, outpatient commitment) influenced psychiatrists' recovery-oriented practices.
After receiving institutional review board approval from the New York State Psychiatric Institute, the first author disseminated information about this study via e-mail to all alumni of Columbia University's Public Psychiatry Fellowship for whom contact information was available. Included in each e-mail was a description of the aims of the study, an invitation to complete an online survey of recovery-oriented practices, and a link to the survey instrument. Responses were coded, and names were removed from the database. Contact information was available for 172 of 192 alumni (90%), and 144 of 172 (84%) completed the survey in the fall of 2005.
The mean±SD age of participants was 42.1±7.8 years. The sample had an approximately equal proportion of men (N=73, or 51%) and women (N=71, or 49%). Respondents were primarily Caucasian (N=89, or 62%), with African Americans (N= 15, or 10%), Asians (N=24, or 17%), Latinos (N=10, or 7%), and South Asians (N=6, or 4%) making up the rest of the sample. Most respondents worked in either community-based mental health centers (N=43, or 30%), hospital-based settings (outpatient, N=32, or 22%; inpatient, N=26, or 18%; and other hospital-based settings, such as emergency room or consultation liaison, N=10, or 7%). A minority also worked in community-based social service or rehabilitation agencies (N=11, 8%), in private practice (N=7, or 5%), in community-based health centers (N=5, or 3%), for governmental agencies (N=3, or 2%), and in other settings (N=4, or 3%) or unspecified settings (N=3, or 2%).
A survey instrument was developed de novo to assess practices and attitudes associated with recovery orientation. We conceptualized recovery-oriented practices as described in the introduction, using recovery guidelines, prior measures of recovery, and other source documents to generate item content. We assessed psychiatrists' perceptions of their own practices and those of the agency in which they work. We additionally were interested in psychiatrists' attitudes about the burden of mental illness, awareness of recovery concepts, and attitudes toward leverage practices. The resulting survey consisted of 52 items with Likert and categorical item response alternatives. [An appendix detailing the background of the survey and listing the questions in the survey is available as an online supplement at ps.psychiatryonline.org.]
To identify scales we subjected the 37 survey items on the Likert scale to an exploratory principal components factor analysis with varimax rotation. Ten factors with eigenvalues greater than 1 emerged, but inspection of the scree plot suggested a more parsimonious solution of six primary factors that accounted for 57% of the variance explained. We described these factors as agencies' recovery-oriented practices, psychiatrists' recovery-oriented practices, recovery awareness, leverage practices, environmental interventions, and perceived burden of mental illness. In this brief report, we report on psychiatrists' recovery-oriented practices, recovery awareness, and leverage practices because they impinge on the questions described above. Future articles will address the other factors.
Psychiatrists' recovery-oriented practices were measured by averaging nine items from the survey. These items assessed the extent to which respondents reported that they focus on recovery goals, involve significant others, discuss life goals, seek to change a client's environment, promote peer advocacy or consumer-run programs, and ask about support systems, housing, work, and non-mental health issues, such as spirituality.
Recovery awareness was measured by averaging three items assessing the extent to which the respondent was aware of the concept of recovery in mental health; reported an understanding of it; and endorsed a belief in its viability for persons with schizophrenia. These items were scored on a 6-point Likert scale ranging from 0, not at all, to 5, very great extent. The coefficient α for psychiatrist recovery-oriented practices was .89, and for recovery awareness, .90.
Leverage practices comprised four separate items, which we elected not to combine into a scale because of low internal reliability. These items were designed to assess practices and attitudes related to authoritarian medication management, authoritarian decision making, perceived usefulness of outpatient commitment, and perceived usefulness of involuntary hospitalization. Items were measured on a 5-point scale ranging from 0 to 4, with descriptive content defining the scale points. For example, authoritarian medication management ranged from 0, "support the person's choice, even if I feel it is not optimal," to 4, "apply for permission to medicate the patient involuntarily." For all four items, higher scores indicated greater endorsement of leverage practices.
We first examined the intercorrelations of the identified scales, which we expected to show moderate to strong correlations (see Table 1). Psychiatrists' recovery-oriented practices were significantly correlated with recovery awareness (r=.28) and negatively correlated with authoritarian medication management (r=-.18). Consistent with expectations, authoritarian medication management was correlated with authoritarian decision making (r=.34) and with perceived usefulness of both involuntary hospitalization (r=.36) and involuntary outpatient commitment (r=.28). In addition, perceived usefulness of involuntary hospitalization was correlated with perceived usefulness of involuntary outpatient commitment (r=.42) and authoritarian decision making (r=.23).
We were interested in which recovery-oriented practices were used routinely and which were used more infrequently. Thus we examined item-level mean scores for psychiatrist practices. The practices that appear to be most widely employed included asking about the consumers' social support system (4.0±1.0; 77% endorsing the item to a great or very great extent), asking about their housing situation (4.0±1.2; 80% endorsing the item to a great or very great extent), asking about their work life (3.9±1.1; 73% endorsing the item to a great or very great extent), and asking about life goals beyond medication (3.7±1.1; 65% endorsing the item to a great or very great extent). The practices used to the least extent included encouraging clients to participate in peer advocacy (2.8±1.4; 34% endorsing the item to a great or very great extent), discussing non-mental health issues (3.0±1.2; 35% endorsing the item to a great or very great extent), seeking to change a person's environment (3.1±1.1; 33% endorsing the item to a great or very great extent), and involving significant others in service planning (3.2±1.3; 39% endorsing the item to a great or very great extent).
Psychiatrists reported somewhat mixed feelings about both involuntary hospitalization and outpatient commitment. Mean scores (3.0±.8 and 2.9±.9, respectively) indicated that participants generally considered these interventions necessary but not always beneficial. Regarding authoritarian medication management, psychiatrists indicated a willingness to apply some pressure on the patient by bringing the issue up in every session (2.0±.9). In contrast, regarding authoritarian decision making, psychiatrists indicated they explain why they prefer a certain course of action but leave the person free to choose his or her own course of action (1.0±.7).
We next examined the association of recovery awareness, leverage, and psychiatrists' recovery-oriented practices. To examine these associations, we conducted a hierarchical regression equation, using recovery awareness and items regarding attitudes toward leverage practices as independent predictors and psychiatrists' recovery-oriented practices as the dependent variable. This analysis controlled for age and gender and is summarized in Table 2. Demographic variables entered in the first step did not contribute significant variance to the model. Entry of recovery awareness and items regarding attitudes toward leverage practices on a second step resulted in a significant increase in the proportion of variance explained, and psychiatrists' recovery-oriented practices (p≤.001), with recovery awareness (p≤.001), and less authoritarian medication management (p≤.01) emerged as unique predictors of recovery-oriented practices. These findings suggest that both exposure to recovery concepts and less authoritarian medication management are strongly associated with psychiatrists' use of recovery-oriented practices.
The aim of this study was twofold: to elucidate psychiatrists' practices designed to support their patients' recovery and to explore correlates of their use of these practices. Psychiatrists' recovery-oriented practices that appear to be most widely employed included asking about the consumer's social support system, housing situation, work life, and life goals beyond medication. The practices used to the least extent included encouragement of peer advocacy, discussing non-mental health issues, seeking to change a person's environment, and involving significant others in service planning.
These findings suggest that although psychiatrists feel comfortable asking about a range of recovery-oriented issues involving their patients, they are less likely to address aspects of functioning that are external to the treatment setting, such as encouraging peer advocacy, involving significant others in service planning, and discussing non-mental health issues. This further suggests a basic tension between the spirit of recovery-oriented practices, which emphasizes environmental interventions and community integration, and the nature of public psychiatrists' recovery-oriented practices.
In our regression analyses, we found that psychiatrists' recovery-oriented practices were associated with recovery awareness and less authoritarian medication management. This finding suggests that more education and advocacy to improve awareness of recovery-oriented practices may increase their use. Moreover, the association between less authoritarian medication management and psychiatrists' recovery-oriented practices underscores that psychosocial interventions and prescribing practices are intertwined and suggests that the more psychiatrists embrace the broader suite of practices labeled "recovery oriented," the more they also will embrace less authoritarian medication management. The pattern of correlations among the four leverage items suggests that psychiatrists who endorse one leverage practice tend to endorse the others. However, although psychiatrists in our sample tended not to endorse authoritarian decision making, they also indicated, on balance, relatively positive views of both outpatient commitment and involuntary hospitalization and generally endorsed milder forms of authoritarian medication management. Clearly this group of public psychiatrists who have received specialized training in recovery nonetheless view both involuntary inpatient and outpatient commitment as necessary interventions under certain circumstances. Furthermore, these respondents indicated a willingness to apply some pressure (by bringing the issue up in every session) on patients who relapsed repeatedly as a result of nonadherence.
In light of the above, we believe psychiatrists need to educate consumers in advance as to the circumstances under which they may initiate specific leverage practices that might be experienced as coercive and explore ways to give patients more options even in these difficult situations, such as through the creation of advance directives.
A number of limitations in the study presented here should be borne in mind. As mentioned above, alumni of Columbia University's Public Psychiatry Fellowship are self-selected and are not necessarily representative of all public psychiatrists. Another important limitation is that all of the data are self-report and are thus vulnerable to problems with shared method variance. We further note that we developed a measure de novo for this study. This was necessitated by the absence of extant instruments assessing psychiatrists' recovery-oriented practices. We would note, however, that the absence of a validated instrument is a less serious flaw unless the purpose is to assess internal constructs or states, such as psychopathology or personality variables. In our case, we were assessing straightforward clinical practices by using items with high face validity and standard Likert response alternatives. Further, our factor solution was consistent with expectations, and the resulting scales showed predictable internal relationships and adequate internal reliability coefficients. Thus we consider our measure a reasonable, if not ideal, approach for measuring psychiatrists' recovery-oriented practices.
From the standpoint of recovery advocacy, it is gratifying to learn that Columbia University's Public Psychiatry Fellowship alumni are addressing areas that most recovery proponents would endorse as important. As the concept of recovery continues to mature it is to be expected that recovery-oriented practices will continue to expand among all psychiatrists. However, our findings with regard to involuntary inpatient and outpatient commitment and authoritarian medication management suggest that even psychiatrists who receive specialized training in rehabilitation and recovery are going to be reluctant to condone medication refusal among patients who have repeatedly relapsed as a result of nonadherence and are going to continue to be willing to initiate involuntary commitment when patients' behavior makes them a danger to themselves or others. This is self-evident to most psychiatrists, but many recovery advocates continue to regard these practices as coercive. These differing perspectives represent a potential ongoing conflict.
The authors thank Thomas M. White, M.D., who developed the Dialogix Web-based survey tool and assisted the authors in the process of converting the survey instrument into a Dialogix-based survey form. They also thank Hunter McQuistion, M.D., for his assistance indeveloping the survey questions.
The authors report no competing interests.