|Home | About | Journals | Submit | Contact Us | Français|
Over 2,200 North American psychotherapists completed a Web-based survey concerning their clinical work, including theoretical orientation, client characteristics, and use of specific psychotherapy techniques. Psychotherapeutic integration was common, with the majority of respondents identifying with more than one theoretical orientation or as having an eclectic orientation. The modal patient was a White female adult suffering from a mood or anxiety disorder and interpersonal problems. Individual psychotherapy was the preferred treatment modality. The most frequently endorsed techniques were relationship-oriented such as conveying warmth, acceptance, understanding, and empathy. The least frequently endorsed techniques were biofeedback, neurofeedback, body and energy therapies, and hypnotherapy. Efforts to disseminate empirically based therapies require understanding and accommodating clinicians’ tendencies to integrate techniques.
There is limited information about what constitutes the routine conduct of psychotherapy in community settings. To aid in the dissemination of empirical evidence to frontline clinicians, researchers need to understand the patients being seen, their problems, and currently used treatments (Hohmann & Shear, 2002; Kazdin, Siegel, & Bass, 1990). Such information could also allow for monitoring changes and trends in the delivery of mental health services.
Some data concerning what practitioners actually do in therapy comes from a survey of psychologists over two decades ago (Wogan & Norcross, 1985). Over 300 psychologists working with adults were asked about use of 99 techniques. Relational interventions were widely employed, whereas some highly specialized techniques such as flooding were rarely used. A more recent investigation examined the practice patterns of members of the United Kingdom Council of Psychotherapy (Tantam, 2006). Although analytically informed conceptualizations guided most treatments, only 39% actually used psychoanalytic techniques.
This article describes theoretical orientations, caseloads of and techniques used by a large and diverse sample of U.S. and Canadian mental health professionals. The composition of the mental health workforce has changed radically over the past several decades and may be unique in North America. One trend is proliferation in the number of core disciplines (psychology, psychiatry, social work, psychiatric nursing, and marriage and family therapy) and additional categories of providers such as addiction and pastoral counselors (Robiner, 2006). There has also been an increase in clinicians who are female and those representing racial and ethnic diversity, a decline of psychoanalysis, and an increase in client self-change and cognitive– behavioral interventions (Norcross, Hedges, & Prochaska, 2002; Norcross, Karpiak, & Santoro, 2005). In addition, there is more access to psychotherapy by socioeconomically disadvantaged individuals, a rise in psychotherapy for mood disorders (Olfson, Marcus, Druss, & Pinkus, 2002), and more positive attitudes toward mental health treatment (Mojtabai, 2007).
The Columbia University-New York State Psychiatric Institute Institutional Review Board (IRB) approved this study. The Editor of a popular psychotherapy magazine, the Psychotherapy Networker, sent email invitations twice to subscribers with known email addresses (~40%; N = 22,000), directing them to a secure Website where they could read a study description, offer consent, and complete the survey. Between September 2006 and April 2007, 2,739 participants registered and at least partially completed the survey.
Response metrics such as view, participation, and completion rates are recommended for reporting of Web-based surveys (Eysenbach, 2004). To guarantee participant confidentiality, the IRB did not allow tracking devices. Thus, lack of information about the number of unique visitors to the Website prevented calculation of exact view and participation rates. An estimate of participation rate is the number of Website survey registrations divided by the number of subscribers who were sent emails (13%), which is conservative because it is unlikely that every subscriber who was sent an email visited the Website. The completion rate among individuals who consented to participate was 72%.
Because of differences in formal training, licensure, and practice circumstances as well as primary focus on practicing psychotherapists, those living outside of the U.S. and Canada (92; 3%) and students (40; 2%) were excluded from further analyses, leaving 2,607. Because scope of practice and patient profiles are different for M.D.s versus others, 14 (<1%) psychiatrists and 4 (<1%) other physicians were excluded. Of the remaining 2,589 participants, 433 (17%) were excluded because they had <25% missing responses. In the remaining sample of 2,156 participants reported on here, the modal clinician was White (n = 1981, 92%), female (n = 1644, 77%), 59.30 (SD = 9.89) years old, with an average of 15.26 (SD = 9.86) years of clinical experience. The largest group were social workers (36%; n = 775), next were professional counselors (23%; n = 488), followed by psychologists (17%; n = 374), marriage and family therapists (17%; n = 360), and others (n = 158, 7%) including certified drug/alcohol and pastoral counselors. Half were in independent practice (n = 1129, 53%), and the rest worked in outpatient mental health clinics (n = 450, 21%) or other institutional settings. Most practiced in the Eastern region of the U.S. (n = 1056, 50%), followed by Pacific area (n = 451, 21%), Central zone (n = 390, 19%), Mountain region (n = 140, 7%), and Canada (n = 70, 3%).
In brief, the construction of the survey was a systematic, sequential, and iterative process, which started with interviews of seven psychotherapists, representatives from several disciplines, regarding their clinical practices and influences on practice. Additional items were added from a variety of professional sources (e.g., Websites and publications). Several national clinical researchers reviewed the initial document and six clinicians completed the survey to provide feedback on clarity, redundancies, and response burden. The survey addressed numerous areas, including influences on uptake and sustained use of psychotherapies (Cook, Schnurr, Biyanova, & Coyne, 2009), influential figures, books and authors on practice (Cook, Biyanova, & Coyne, 2009a), and perceived barriers to adoption of treatments (Cook, Biyanova, & Coyne, 2009b). None of the information presented here is duplicated in earlier papers.
The primary focus of this paper is to report information on theoretical orientation, caseloads, and psychotherapy techniques. Theoretical orientation was assessed by the following question, “How would you describe your current theoretical orientation? Please fill in percentages to total 100%.” Additionally, information was gathered regarding client characteristics such as sex, age, and types of presenting problems. Practice over the past 30 days was also assessed, including number of clients, number of sessions, mean session length, number of new clients, and whether therapy was individual, group, or family and couples. Participants were asked to indicate using 5-point ordinal scale (none/very few, some, half, most, and almost all/all) what proportion of their clients in the preceding month received each of 60 psychotherapy techniques.
Forty-eight percent of clinicians worked mostly with female clients, 13% saw primarily men, and 40% had about equal distribution of genders in their caseloads. Most clinicians (91%) reported treating predominantly White clients, with few working mostly with Black (4%) or Hispanic clients (5%). Sixty-nine percent worked mostly with adults, 14% with children and adolescents, and the rest had a mixed clientele. Mood and anxiety disorders were most prevalent in the caseloads of 40% of the clinicians and 21% reported that majority of their clients had social problems such as family discord or lack of social support. Only 8% of clinicians devoted most of their time to posttraumatic stress disorder clients; less than 8% reported that majority of their clients had issues such as homelessness, unemployment, inadequate housing; 6% worked mostly with substance use problems; 5% primarily with serious behavioral problems such as aggression; 4% addressed mostly health behavior problems; 4% specialized in personality disorders; 3% in serious comorbid medical conditions; 2% in serious psychopathology and less than 1% devoted most of their time to clients with eating disorders. Forty-one percent saw primarily clients with income below $19,000; 27% worked mostly with those whose income ranged from $20,000 to $49,999, 26% with clients in the $50,000-$99,000 income category; and only 7% saw those whose income exceeded $100,000. Individual psychotherapy was the preferred treatment modality in 84% of clinicians; 9% saw mostly families and couples; and the rest specialized in group therapy. In a typical week, most therapists saw 19.28 (SD = 10.95) clients, with 2.5 (SD = 4.35) new clients each week. Most clients (67%) were seen weekly; 16% monthly; 8% two or more times a week; and 7% less than once a month. Most clients (70%) had 50–60 min sessions; 14% had 45 min sessions; for 10% sessions lasted more than 1 hr; and 6% were seen for less than 30 min.
Participants were asked to assign a percentage value to each theoretical model, comprising their total theoretical orientation (equal to 100%). Then, the percentages for each model were summated across participants and divided by the number of participants in the sample. Cognitive-Behavior Therapy (CBT) was the most popular approach, that is, it constituted a larger percentage of the practitioners’ models when all the endorsed percentages were combined. CBT was followed by the family systems, psychodynamic/analytic, and acceptance/mindfulness based (in that order). When the participants were summed to include everyone who mentioned using a given approach in their practice, CBT was still the leading orientation (n = 1,940; 79%); family systems was the second (1,212; 49%), mindfulness was the third (1,013; 41%), psychodynamic/analytic was the fourth (885; 36%), and Rogerian/client-centered/humanistic the fifth (758; 31%). Only 59 participants (2%) identified themselves completely with one orientation. The rest either endorsed “eclectic” approach or specified the exact percentages each orientation informs their practice.
Table 1 enumerates use of specific therapeutic practices. These practices are arranged in descending order, from techniques used by most clinicians with all or almost all clients to the least used techniques. The top 10 practices were endorsed by more than half of participants: trying to convey warmth and respect with all or almost all of their clients, followed by communicating that client is accepted and prized, communicating understanding of client’s experience, empathizing with the client, promoting clear, direct expression of client’s feelings, making reflective or clarifying comments, focusing on cultivating therapeutic relationship, encouraging client to develop healthy recreational activities, encouraging emotional processing of distressing experiences, and raising awareness of how client relates to others. The following 10 practices were endorsed by less than five percent as used most or all of the time and included biofeedback, neurofeedback, body therapies, hypnotherapy, paradoxical techniques, Eye Movement Desensitization and Reprocessing (EMDR), psychological testing, energy therapies, dream analysis, and use of empty chair or roleplaying techniques.
This study documents use of a wide range of mental health practices by a broad scope of psychotherapists from various disciplines and settings across North America. The trend of synthesis or merging of theoretical influences continues. Similar to the findings from a survey of psychologists conducted over 25 years ago (Smith, 1982), the majority of participants in this sample identified themselves with more than one theoretical orientation or as having an eclectic orientation. Clearly, any attempts to disseminate evidenced-based practices to community psychotherapists should understand and accommodate tendencies to integrate techniques. In addition there is a strong trend in the continued strength and impact of CBT as it is the most frequently endorsed theoretical orientation. This is consistent with a national survey of marriage and family therapists in which the most frequently endorsed primary treatment modality was CBT (Northey, 2002), and a review of clinical psychologists reported orientations from 1960 to 2003 (Norcross, Karpiak, & Santoro, 2005).
Consistent with results from a representative sample of psychologists conducted over 25 years ago, individual therapy was the most popular modality (Prochaska & Norcross, 1983). Relationship-oriented common-factor techniques were the most frequently utilized. These techniques are at the core of Rogers’ (1957) client-centered approach and focus on building and sustaining a good therapeutic alliance. A meta-analysis of 79 psychotherapy studies indicates empirical support for a number of these techniques (Martin, Garske, & Davis, 2000). Their reported common usage may suggest that this is what clinicians believe are the most important mechanisms that facilitate patient improvement in psychotherapy.
Techniques that likely fall under the rubric of CBT (e.g., assign homework or behavioral tasks outside of session; challenge clients’ thoughts or use other cognitive restructuring techniques; follow a treatment manual; and assign readings or self-help books) also appear quite popular in use. Since development over 40 years ago, CBT has proven effective for a variety of psychiatric conditions as well psychosocial issues (Beck, 2005; Butler, Chapman, & Forman, 2006). In addition, CBT appears to be gaining influence and popularity with clinicians in terms of self-rated influential figures, authors, and books (Cook et al., 2009a). Its use may also be due, in part, to insurance panels requiring providers to offer CBT.
Psychodynamic techniques (e.g., relate current problems to childhood and family experiences; help clients explore unconscious processes) were also used often. Indeed, almost one in three therapists used them with most or all of their clients, and a similar number used them with some of their clients. This is consistent with results of a prior study that found over one third of clinicians used psychodynamic treatment with anxiety disordered patients in their practice (Goisman, Warshaw, & Keller, 1999). Using the Delphi methodology, a panel of psychotherapy experts predicted that the use of classical analytic techniques, such as free association and dream interpretation, would decrease over time (Norcross et al., 2002). Whether the popularity and use of dynamic therapy reduces over the coming years, however, remains to be seen.
The least frequently endorsed techniques (e.g., hypnotherapy) seem to require specialization or certification beyond what is provided in standard graduate education. Thus, their lack of use may not reflect their lack of popularity per se, but rather their additional requirements for training.
While there are numbers of well-known professional organizations such as the American Psychological Association, the National Association of Social Workers, and the American Association of Marriage and Family Therapy, there appears to be no joint organization currently serving the interests of a broad array of psychotherapists in their entirety. Similar to the United Kingdom’s Council for Psychotherapy, it might be a good idea to have a registry of all psychotherapists in North America. This would allow periodic assessments of the mental health practitioner field as well as act as conduits for dissemination. In addition, a national registry or entity could assist in the coordination, deployment, tracking, and control of the training for this work-force (Robiner, 2006).
The main strength of this study is also its primary limitation. It was conducted with a large group of therapists in diverse clinical settings across North America via a Web-based survey and exact generalizability is unknown for a number of reasons including lack of data on nationally representative samples of psychotherapists, and detailed information on Psychotherapy Networker subscribers with which to estimate representativeness of this sample. However, similar to a study with an international sample of over 4,000 psychotherapists (Orlinsky, Botermans, & Ronnestad, 2001; Orlinsky et al., 2005), this sample’s composition is likely to be more representative of the mental health workforce as a whole than those in the majority of clinician surveys, in which samples are drawn from single professional discipline or restricted geographical locations (e.g., Becker, Zayfert, & Anderson, 2004; Freiheit, Vye, Swan, & Cady, 2004; Rosen et al., 2004). Compared to Orlinsky et al.’s (2001) psychotherapists, this sample had a higher proportion of female therapists (77 vs. 53%), and therapists were about 17 years older (59 vs. 42). The majority of our participants were White (92%), but ethnic composition could not be compared to Orlinsky et al.’s (2001) sample, as this information was not provided in published reports. Psychologists were the largest professional group in the Orlinsky sample (57 vs. 17%, respectively); while social workers were the predominant profession in this study (36 vs. 6% in Orlinsky’s et al. study).
Another limitation of this study is that self-reported practices may not necessarily indicate what clinicians actually do (Hoyt, 2002). This may particularly be true of relationship-oriented common-factor techniques. Objective measures of psychotherapy utilization might include video samples of random therapy sessions or other verification such as patients’ perception of use or automated record-keeping. In addition, information on use of techniques does not indicate effectiveness (Prochaska & Norcross, 1983). In addition, there is some overlap in the therapeutic technique categories (e.g., follow a treatment manual and EMDR) making it difficult to definitively determine use of particular techniques. Because of small numbers as well as likely practice differences, another limitation is the noninclusion of psychiatrists. Previous research, however, has found that there are differences between psychologists and psychiatrists in terms of patient caseload and practice profile (Pingitore, Scheffler, Sentell, & West, 2002), psychotherapy techniques (Kazdin, Siegel, & Bass, 1990) as well as goals of treatment (Strupp, 1973).
Because most practitioners were experienced professionals, our sample underrepresents younger, less educated, or less credentialed peer counselors and paraprofessionals who provide a considerable proportion of counseling services, particularly to disadvantaged populations. Such persons are difficult to recruit for a representative sample or even count, but may be the most accessible sources of therapy available to members of low income and minority populations. In addition, most clinicians in this study were masters-level clinicians. Although there are currently no data on the proportion of masters- and doctoral-level clinicians in the mental health field as a whole to which study findings could compare, the data may reflect a trend in the field of the continuing increase in the number of master’s-level therapists (Norcross et al., 2002).
This project described was supported by Award Number K01 MH070859 from the National Institute of Mental Health. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institute of Mental Health or the National Institutes of Health.
JOAN M. COOK, Department of Psychiatry, Yale University, and National Center for PTSD, West Haven, CT.
TATYANA BIYANOVA, Department of Psychiatry, Yale University.
JON ELHAI, Department of Psychiatry, University of Toledo.
PAULA P. SCHNURR, National Center for PTSD, White River Junction, VT, and Dartmouth Medical School.
JAMES C. COYNE, Department of Psychiatry, University of Pennsylvania.