Using Feedback From Patient-Reported Outcome Measures in Mental Health Services: A Scoping Study and Typology
Abstract
Objective:
Methods:
Results:
Conclusions:
Methods
Scoping Study
Search Strategy
Categorization of Articles
Results
Category and study | Design | Sample | PROM | Procedure for PROM use | Results |
---|---|---|---|---|---|
Category 1 | |||||
Christensen et al., 2004 (41) | Randomized controlled trial (RCT) | 134 married couples with serious and chronic distress undergoing a free therapy program in 2 U.S. cities | 32-item Dyadic Adjustment Scale, a self-report measure of marital satisfaction; 3 subscales from the Marital Satisfaction Inventory–Revised: 22 items on global distress, 19 on problem-solving communication, and 13 on affective communication; 14-item Marital Status Inventory, a measure of thoughts and tentative and actual steps undertaken toward divorce; 68-item Mental Health Index, a measure of current symptoms, life satisfaction, and well-being (subscale of the Compass Outpatient Treatment Assessment System) | Couples were randomly assigned to 1 of 2 treatment conditions (comparing 2 treatment types). All completed various screening measures before and at intake. At intake and 13 and 26 weeks, couples completed all PROMs. At the end of treatment, clients completed measures of relationship satisfaction and an evaluation of services. | The 2 treatment types were compared on change in PROM scores. |
Hannan et al., 2005 (42) | Single-group posttest | 618 clients at a U.S. university outpatient clinic | 45-item Outcome Questionnaire (OQ-45) (9), a measure of client progress on 3 dimensions: subjective discomfort (25 items), interpersonal relationships (9 items), and social role performance (11 items) | Clients completed the OQ-45 before each therapy session. Routine feedback to therapists was suspended for a period of 3 weeks to examine therapists’ ability to estimate client progress. | Therapists tended to overpredict client improvement and not to predict deterioration. |
Category 2 | |||||
Berking et al. 2006 (43) | RCT | 118 inpatients at a German psychosomatics, psychotherapy, and behavioral medicine clinic | 11-item German version of the Brief Symptom Inventory (BSI); 12-item German version of the Inventory of Interpersonal Problems (IIP), a self-rated measure of interpersonal difficulties; 10-item Incongruence Questionnaire (INK), assessing extent of congruence of current situation with one’s motivations and goals; 42-item Questionnaire to Assess Changes in Experiencing and Behavior (VEV), a measure of therapy-induced changes in experience and behavior | Patients receiving cognitive-behavioral therapy (CBT) were randomly allocated to either a feedback or a no-feedback condition. All patients completed the Emotionality Questionnaire, BSI, IIP, and INK on day 1, day 3, and then weekly. In the feedback condition, therapists received results the following day. At the end of therapy, patients completed the VEV. | Average improvement on all outcome measures was significantly greater in the feedback group. |
Bickman et al., 2011 (44) | RCT (substantial attrition) | 340 youths (ages 11–18) receiving home-based services from a private, for-profit, behavioral health organization at 28 U.S. sites | 32-item Symptoms and Functioning Severity Scale, a measure of the frequency of emotions or behaviors linked to typical mental disorders of youths | Clients were randomly allocated to an experimental or a control group. At the end of a treatment session, clients completed a paper questionnaire. Clinicians of clients in the experimental group received weekly feedback (mean scores and alerts) and cumulative feedback every 90 days. Clinicians of clients in the control group received only the 90-day feedback. | As indicated by PROMs, clients in the experimental group improved significantly faster than those in the control group. |
de Jong et al., 2012 (45) | RCT | 413 outpatients receiving psychiatric treatment at a Dutch medium-sized health care institution | OQ-45, Dutch version | Patients were randomly allocated to an experimental feedback group or a no-feedback control group. All patients completed the PROM after sessions 1, 3, and 5 and then after every 5th session. After each PROM completion, therapists in the feedback group received an e-mail with information on the patient’s PROM progress. No alarms were used, but therapists were able to identify “not-on-track” patients. The study also examined to what extent therapist characteristics moderate effects of feedback, and therapists completed a use-of-feedback questionnaire at the end of the study. | For clients identified as “not on track,” feedback resulted in a significant positive effect on PROMs when therapists reported using feedback with their clients. |
Lambert et al., 2001 (46) | RCT | 609 clients at a U.S. university counseling center | OQ-45 | Clients were randomly allocated to an experimental or a control group. All clients completed the OQ-45 at intake and before each treatment session. Data for the control group were not shared with clients or therapists. In the experimental group, therapists received results on a graph and were alerted to the client’s progress with a color-coding system. Clinicians’ reactions to the feedback were not managed, with no mechanism to use feedback in any systematic way. | For clients identified as “not on track,” feedback resulted in significantly better outcome scores and significantly longer treatment duration. For clients “on track,” no significant differences were noted in outcome measures, and number of treatment sessions was significantly fewer for the feedback condition. |
Lambert et al., 2002 (47) | Quasi-experimental; intervention conducted after data for the control group had been collected | 1,020 clients at a U.S. university counseling center | OQ-45 | Intended as a replication of Lambert et al. (46) with a larger sample. Clients in 1999 summer and fall semesters were assigned to the control group; clients in 2000 winter and spring semesters were assigned to the experimental (feedback) condition. All clients completed the OQ-45 at intake and before each treatment session. Data from the control group were not shared with clients or therapists. In the experimental group, therapists received results on a graph and were alerted to the client’s progress with a color-coding system. Clinicians’ reactions to the feedback were not managed, with no mechanism to use feedback in any systematic way. Therapists with clients in the feedback group received a tracking form, which suggested possible clinician actions in response to feedback. | For “not-on-track clients,” feedback resulted in significantly better outcome scores and significantly longer treatment duration. For “on-track” clients, no significant differences were noted in outcome measures or treatment duration. |
Lutz et al., 2012 (48) | RCT | 1,708 clients receiving outpatient psychotherapy in 1 of 3 regions in Germany | German version of the BSI; German version of the IIP; 12-item Short-Form Health Status Instrument; additional measures depending on the patient’s main diagnosis. | Clinicians were randomly allocated to an experimental or a control group (treatment as usual). In both groups, PROMs were used at intake, discharge, and 1 year later. In the experimental group, patients completed PROMs 5 times during treatment. In the experimental group, therapists received immediate PROM feedback (summary and graphs) about their patients. There were no prescriptive guidelines on use of PROM feedback; therapists could incorporate this information into therapy at their discretion. Detailed information can be found in the final report of the so-called TK model. Lutz and colleagues later noted that the study results need to be interpreted with caution because of some compromising externally imposed design modifications. | Feedback did not affect PROM scores. The groups also did not differ in treatment length. |
Probst et al., 2013 (39) | RCT | 252 inpatients recruited from a psychosomatics department of a hospital and a psychosomatics hospital, both in Germany. Probst et al. (39) reported results from 43 patients at risk of outcome deterioration. Probst et al. (40) reported on 209 patients considered to be “on track.” | OQ-45, German version | Patients were randomly allocated to an experimental or a control group. All patients completed the OQ-45 every weekend. On Mondays, therapists of patients in the experimental group received feedback reports. Therapists could choose to discuss feedback with patients. Also included was the Assessment of Signal Cases scale, which measures therapeutic alliance, motivation for change, social support, and critical life events. This is part of clinical support tools (CST), which provide empirically based problem-solving strategies. | For patients at risk of deterioration, feedback significantly improved outcome scores (39). For patients “on track,” feedback did not have a significant effect (40). |
Whipple et al., 2003 (49) | Quasi-experimental; random assignment to experimental and intervention groups; nonrandom assignment to 1 of the experimental groups | 358 adult clients in a U.S. university counseling center | OQ-45 | Clients were randomly allocated to an experimental (feedback) or a control group. All clients completed the OQ-45 at intake and before each treatment session. In the feedback group, therapists received results on a graph, along with suggested decision rules, and were alerted to the client’s progress with a color-coding system. Therapists of clients in the feedback group who were considered “not on track” received a tracking form, which suggested possible clinician actions in response to feedback. The experimental group was further divided into a feedback-only group and a feedback plus CST group. However, use of CST was nonrandom and depended on therapists’ decisions to use CST. | For clients “not on track,” feedback plus CST resulted in significantly higher outcome scores than feedback only, which in turn resulted in significantly higher scores than no feedback. For clients “on track,” no significant group differences were noted. Clients considered “not on track” in either of the 2 feedback groups remained in therapy significantly longer than “not-on-track” clients in the control group. For “on-track” clients, therapy duration was significantly longer for the control group than for the 2 feedback groups. |
Category 3 | |||||
Cheyne and Kinn, 2001 (50) | Pilot RCT | 42 consecutive referrals for alcohol counseling at a range of U.K. community-based cognitive-behavioral counseling services | Schedule for the Evaluation of Individual Quality of Life (SEIQoL) (68), on which respondents rate the importance of life areas to their overall quality of life | Clients were randomly allocated to an experimental or a control group. Clients in the experimental group completed the SEIQoL together with the therapist at the first and final counseling sessions and at 4- and 8-week review appointments. Four weeks after completion of treatment, all participants were mailed a questionnaire about satisfaction with services and outcomes achieved (42% response rate). | The experimental condition had a larger proportion of clients with favorable outcomes (not statistically significant). A separate publication (36) reported qualitative data on therapists’ positive experiences of completing the SEIQoL with clients. |
de Jong et al., 2014 (32) | RCT | 475 outpatients at Dutch private psychotherapy practices and mental health institutes | OQ-45, Dutch version | Patients were randomly allocated to a no-feedback control group, a therapist-only feedback group, or a therapist-patient feedback group. All patients completed the OQ-45 online (typically on a laptop in the therapist’s waiting room) before each therapy session but not more than once a week. In the 2 feedback conditions, PROM scores and feedback messages were generated immediately, and subsequent discussion of feedback was at the therapists’ discretion. | Group differences in OQ-45 scores at treatment end were not significant, although the therapist-client feedback group had the smallest number of deteriorated cases. For “not-on-track” clients, feedback prevented negative outcomes. |
Hansson et al., 2013 (51) | RCT | 262 patients in 2 general psychiatry outpatient clinics in Sweden | OQ-45, Swedish version | Patients were randomly allocated to an experimental or a control group. Patients completed the OQ-45 at intake and at each clinic visit but not more than once a week. Therapists of patients in the experimental group received patients’ OQ-45 scores via a Web application before each subsequent visit; these scores were also handed to the patient. In the control group, neither therapist nor patient received feedback. | Patients in the experimental group had greater improvements in their outcome scores (not statistically significant). |
Slade et al., 2006 (52) | RCT | 160 patients of 8 U.K. community mental health teams | 12-item Manchester Short Assessment (MANSA); a measure of quality of life | Patients were randomly allocated to an experimental or a control group. Both groups received treatment as usual. Patients and therapists in the experimental group also completed a monthly mailed questionnaire and were sent identical feedback every 3 months in the form of graphics and text that also highlighted areas of disagreement between patient and therapist. | No significant group differences were noted in quality-of-life scores or in scores of patient-rated unmet needs and other secondary measures rated by therapists. Patients in the experimental group had significantly fewer psychiatric inpatient days. |
Category 4 | |||||
Newnham et al., 2010 (53) | Historical cohort design | 1,308 consecutive inpatients and day patients participating in a 10-day CBT group at a private psychiatric hospital in Australia | 5-item World Health Organization Well-Being Index (WHO-5), a measure of positive mental health; 4 subscales (4-item vitality, 2-item social functioning, 3-item role emotion, and 5-item mental health) of the Short Form–36 Health Survey (SF-36); 21-item Depression Anxiety Stress Scale (DASS-21), a measure of negative emotional symptoms | Patients in cohort 1 received treatment as usual. Patients in cohort 2 completed the WHO-5 every second day but did not receive feedback (scores and a graph with explanation) until the final therapy day, when they could discuss their scores during the group session. Patients in cohort 3 completed the WHO-5 every second day and received the same WHO-5 feedback from their therapists midway through treatment (day 5) and on the final day, also with opportunities to discuss scores. Therapists were not given specific instructions on use of feedback. Patients in all cohorts also completed the DASS-21 and SF-36 at admission and discharge. | No effect of feedback on WHO-5 scores was noted. For patients “not on track,” feedback was significantly associated with decreased depressive symptoms (DASS-21) and the vitality and role emotion subscales of the SF-36 but not with any other subscale score. It was later noted that after treatment, “on-track” patients in cohort 3 were significantly less likely than “on-track” patients in cohort 2 to be readmitted. |
Puschner et al., 2009 (54) | RCT | 264 adults receiving inpatient treatment at a German psychiatric hospital | OQ-45, German version | Clinicians were randomly allocated to an experimental or a control group. All patients completed German version of the OQ-45 at intake, every week thereafter, and at discharge. In the experimental group, patients and clinicians received summary information 1 or 2 days after PROM completion. This information consisted of graphs, text with treatment recommendations and possible alert messages, and encouragement for patients and clinicians to discuss the results. However, no guidelines for such discussion were provided. Patients and clinicians in the control group received no feedback. | No significant effect of feedback on treatment outcome was noted as measured by the OQ-45. Most patients found the feedback useful for motivation, but their views about its effectiveness were mixed. Most patients reported that they rarely discussed feedback with professionals or caregivers. |
Category 5 | |||||
Anker et al., 2009 (55) | RCT | 205 couples seeking outpatient couples therapy at a family counseling agency in Norway | 4-item Outcome Rating Scale (ORS) (63), derived from the OQ-45; 15-item Locke-Wallace Marital Adjustment Test (LW), covering aspects of marital functioning and satisfaction | Participants were randomly allocated to an experimental group (feedback) or a control group (treatment as usual). Participants completed the ORS and LW before the first session, the ORS before each subsequent session, and the ORS and LW 6 months after the final session. In the control group, the ORS was completed in the presence of a secretary, and results were not fed back to either participant or therapist. In the experimental group, the ORS was rated in the presence of the therapist before each session and scored immediately. Therapists were trained to incorporate into treatment the ORS feedback and associated computer-generated treatment and progress feedback. They were also advised to show the results to clients and initiate discussions, although this was not monitored. Clients also completed the Session Rating Scale (SRS) (66), a measure of the therapeutic alliance. | Improvements in ORS scores were significantly greater in the experimental group than in the control group, which was maintained at 6-month follow-up. |
Harmon et al., 2007 (30) | Quasi-experimental; nonrandom group allocation and a comparison group from archival data | 1,374 adult clients seeking treatment at a large U.S. university counseling center | OQ-45 | Because of attrition, not all clients could be allocated randomly to the 2 intervention groups (feedback to both therapists and clients and feedback to therapists only). Archival data (N=1,445) from the same clinic and therapists served as a no-feedback control group. Clients completed the OQ-45 at intake and weekly thereafter. Before each session, the previous week’s scores were made available as feedback in the form of graphs and a color-coding system to categorize client progress. In both groups, clients considered “not on track” were further randomly allocated to either CST feedback (results of additional measures of therapeutic alliance, stages of change, and social support) or no CST feedback. Clients who received feedback and were not responding well to treatment were encouraged to discuss their concerns about lack of progress and ideas for therapy modifications. Clinicians’ reactions to the PROM feedback were not managed. Therapists who received feedback plus CST were able to consult a CST manual for treatment suggestions based on feedback data. | Mean OQ-45 scores improved significantly more for the feedback groups than for the archival no-feedback control group. No significant difference was noted between the 2 intervention groups. However, CST feedback (in addition to PROM feedback to the therapist only or to the therapist and client) resulted in significantly improved outcomes, compared with feedback without CST. Clients considered “not on track” received significantly more sessions in the feedback conditions than clients in the control group. |
Hawkins et al., 2004 (8) | RCT | 201 adults seeking outpatient psychotherapy services at a U.S. hospital-based clinic | OQ-45 | Clients were nonrandomly assigned to therapists on the basis of therapist availability, but clients were subsequently assigned randomly to 1 of 2 treatment conditions (feedback to both therapist and client or feedback to therapist only) or the control condition (treatment as usual with no PROM feedback). All clients completed the OQ-45 at intake and after each treatment session. In the feedback conditions, the previous week’s scores were made available before each session in the form of graphs and a color-coding system to categorize client progress and make treatment recommendations (similar to 46,47). However, clinicians’ reactions to the PROM feedback were not managed or monitored. In the client-therapist feedback condition, clients also received written feedback messages, and those identified as not progressing were encouraged to discuss personal concerns about their progress and potential treatment modifications. A format was available to discuss treatment progress, although interactions with patients were not monitored. | The greatest improvement in OQ-45 scores was for clients in the client-therapist feedback condition, followed by therapist-only feedback and the control condition (statistically significant). For clients considered “not on track,” no significant group differences were noted, although this may have been attributable to small sample size. No significant group effects on treatment duration were noted. |
Murphy et al., 2012 (56) | RCT | 110 adult clients at a university counseling service in Ireland | ORS | The ORS is typically administered in conjunction with the SRS, a measure of therapeutic alliance. The purpose was to test the effects of ORS on its own. Clients were randomly allocated to an experimental group (feedback to both therapist and client) or a no-feedback control group. All clients completed the ORS at intake and before each subsequent session. In the control group, clients completed the ORS in the presence of a researcher (except for the first administration), and neither client nor therapist received feedback on ORS scores. In the experimental group, clients completed the ORS in front of the therapist by using a software program, which instantly generated score feedback, such as in the form of progress graphs. Therapists could decide whether to react to this feedback, such as whether to discuss it with clients. Therapists received an ORS and SRS manual that offered strategies and recommendations for appropriate courses of action in response to ORS scores. | Feedback resulted in significant differences for clients with anxiety issues but not for clients with depression, relationship issues, or other concerns. No effect of feedback on treatment duration was noted. |
Priebe et al., 2007 (57) | RCT | 507 patients with severe and enduring mental illness who used community psychiatric services in 1 of 6 European countries (Germany, the Netherlands, Spain, Sweden, Switzerland, and the United Kingdom) | MANSA | Clinicians were randomly allocated to an experimental or control group (treatment as usual). Clinicians in the experimental group implemented a manualized computer-mediated intervention. In this feedback intervention, patients rated their quality of life approximately every 2 months during routine care; ratings were followed up by questions about whether patients wanted additional support for particular domains. Patients in the control group completed the quality-of-life questionnaire before treatment and 12 months later. Other measures included satisfaction with treatment and unmet care needs. | Quality-of-life scores were significantly higher for the experimental group 12 months posttreatment, despite the presence of ceiling effects in the measure. The effect size for this group difference was higher when only results of participants with a low initial score were analyzed. |
Reese et al., 2009 (58) | RCT | Study 1: 74 clients at a U.S. university counseling center; study 2: 74 clients receiving individual therapy at a U.S. graduate training clinic for a marriage and family therapy master’s program | ORS | Study 1: clients were randomly assigned to an experimental (feedback) or control group. Clients in the control group were given the ORS at intake and the end of treatment. Responses were not analyzed by the therapist, nor were any scores made available to the therapist. In the feedback condition, clients completed the ORS at the beginning of each session and the SRS toward the end of each session. ORS graphs were generated as feedback, and general guidelines were available on how the therapist could proceed, although this was not monitored or managed. Study 2: unlike study 1, therapists rather than clients were randomly allocated to either feedback or no-feedback groups. Another difference was that clients in the control group completed the ORS at the beginning of each session. However, results were not seen by the therapists in the control condition. | In both studies, clients in the experimental (feedback) group had significantly larger gains in ORS scores than clients in the control group. No significant differences in number of sessions attended were noted. |
Reese et al., 2010 (59) | RCT | 46 heterosexual couples receiving couples therapy at a U.S. graduate training clinic (master’s program) for marriage and family therapy | ORS | The study was intended as a replication of Anker et al. (55) with a U.S. sample. Couples were randomly assigned to an experimental (feedback) or control (treatment as usual) condition. All clients completed the ORS at the start of each session and the SRS at the end of each session. The feedback group received ORS graphs as feedback, and general guidelines were available on how the therapist could proceed, although this was not monitored or managed. | Couples in the experimental (feedback) group made significantly greater and faster gains in ORS scores than clients in the control group. |
Schmidt et al., 2006 (60) | RCT | 61 patients with bulimia nervosa or eating disorder not otherwise specified at a U.K. specialist eating disorder unit received guided self-help CBT | 6-item Short Evaluation of Eating Disorders (SEED), a self-rated measure of severity of anorexia and bulimia symptoms; 14-item Hospital Anxiety and Depression Scale, a self-rated assessment of anxiety and depression symptoms | Patients were randomly assigned to an experimental (feedback) or control (no feedback) group. Patients in the feedback group received a personalized letter after initial assessment, including feedback from physical examination and blood tests. A symptom feedback form was completed collaboratively by the patient and therapist halfway through treatment, and patients also received an end-of-treatment feedback letter from their therapist. All patients completed all PROMs before allocation to groups and at the end of treatment and the SEED only at 6-month follow-up. Throughout treatment, patients in the feedback group received computerized PROM feedback every 2 weeks. Patients in the control group completed the same number of computerized assessments during treatment but did not receive any of the feedback. Feedback in the experimental group was also guided by an outcome-monitoring and feedback system, providing automated feedback about progress. | Feedback did not have an effect on treatment uptake or dropout. Feedback resulted in significantly greater improvements on scores for dietary restriction but not on scores for bingeing, vomiting, or exercise. |
Simon et al., 2012 (61) | RCT | 370 adults seeking psychotherapy services at a U.S. hospital-based outpatient clinic | OQ-45 | Clients were randomly assigned to an experimental (feedback) or control (no feedback) condition. All clients completed the OQ-45 before each session. CST was used for clients who were “not on track” in the feedback condition; the tool provided therapists with, for example, decision trees for problem solving, treatment suggestions, and progress alerts and tools to deal with patients who were “not on track.” Therapists were instructed to present the PROM feedback to their clients, although this was not monitored. | OQ-45 scores in the feedback group improved significantly more than those in the no-feedback control group (small effect size). The mean number of sessions was not significantly different between groups. |
Simon et al., 2013 (62) | RCT | 133 adults seeking inpatient treatment at a U.S. eating disorder hospital | OQ-45 | The procedure was identical to that of Simon et al. (61). The purpose was to extend investigations of the effect of PROM feedback to a new client population. | PROM scores in the feedback group improved significantly more than those in the no-feedback control group (small effect size). Body mass index increased in both conditions, with no significant group differences. |
Slade et al., 2008 (31) | Quasi-experimental; random assignment to 1 of 2 feedback types, with a comparison group from archival data | 1,101 adult clients in a U.S. university counseling center, compared with archival data from 2,818 clients under no-feedback and feedback conditions in the same clinic (30,46,47,49). Data reported only for patients considered “not on track” | OQ-45 | Clients were randomly assigned to 1 of 2 conditions (feedback to both therapist and client or feedback to therapist only). Archival data from the same clinic and therapists allowed comparisons with no-feedback conditions and delayed feedback conditions. Unlike previous studies in the same clinic in which feedback was delayed by 1 week (30,46,47,49), this study used an electronic feedback system that provided instant PROM feedback. In the therapist-only feedback condition, therapists were encouraged to use feedback in treatment, but their reactions to PROM feedback were not managed or monitored. In the client-therapist feedback condition, clients also received written feedback messages, and those identified as not progressing were encouraged to discuss personal concerns about their progress and potential treatment modifications. CST feedback and decision trees were also provided to clients and therapists for clients considered “not on track.” The focus was only on patients “not on track.” | No significant differences were noted between the 2 treatment conditions, but these groups showed significant improvements compared with the no-feedback group. Immediate electronic feedback did not lead to significantly larger gains in outcome scores. Clients in the no-feedback condition received significantly more treatment sessions. |
Study | Effect of feedback on PROM scoresa | Treatment length |
---|---|---|
Category 2 | ||
Berking et al., 2006 (43) | Significant (d=.47–.50) | Not reported |
Bickman et al., 2011 (44) | Significant (d=.18) | Not reported |
de Jong et al., 2012 (45) | Significant positive effect only for “not-on-track” patients and when therapists reported use of feedback | Not reported |
Lambert et al., 2001 (46) | Significant for “not-on-track” clients (d=.44); not significant for “on-track” clients | Feedback associated with significantly longer treatment for “not-on-track” clients and significantly fewer days for “on-track” clients |
Lambert et al., 2002 (47) | Significant for “not-on-track” clients (d=.40); not significant for “on-track” clients | Feedback associated with significantly longer treatment for “not-on-track” clients |
Lutz et al., 2012 (48) | Not significant | Feedback associated with significantly shorter treatment; “not-on-track” patients received longer treatment and “on-track” patients less treatment |
Probst et al., 2013 (39) | Significant for “at-risk” patients (d=.54); not significant for “on-track” patients (40) | Not significant |
Whipple et al., 2003 (49) | Significant for “not-on-track” clients (d=.70 and d=.28); not significant for “on-track” clients | Feedback associated with significantly longer treatment for “not-on-track” clients and significantly fewer days for “on-track” clients |
Category 3 | ||
Cheyne and Kinn, 2001 (50) | Not significant | No difference in number of appointments |
de Jong et al., 2014 (32) | Not significant | Not significant |
Hansson et al., 2013 (51) | Not significant | No difference in number of clinic visits |
Slade et al., 2006 (52) | Not significant | Feedback associated with significantly reduced inpatient days |
Category 4 | ||
Newnham et al., 2010 (53) | Significant only for clients “not on track” and only for some of the measures | Not applicable (10-day program) |
Puschner et al., 2009 (54) | Not significant | Not reported |
Category 5 | ||
Anker et al., 2009 (55) | Significant (d=.50) | Not reported |
Harmon et al., 2007 (30) | Both categories 2 and 5 significantly more improved than category 1 (d=.23 and d=.33, respectively); not significant for category 2 versus category 5 | Feedback associated with significantly longer treatment for “not-on-track” clients |
Hawkins et al., 2004 (8) | Category 5 significantly more improved than both categories 2 and 1 (η2=.02 and η2=.04, respectively); categories 2 and 5 combined significantly more improved than category 1 (η2=.02) | Not significant |
Murphy et al., 2012 (56) | Significant for only a subgroup of the sample | Not significant |
Priebe et al., 2007 (57) | Significant (d=.20 or d=.43) only for participants with low initial PROM scores | Not reported |
Reese et al., 2009 (58) | Significant (η2=.07 and η2=.10) | Not significant |
Reese et al., 2010 (59) | Significant (d=.81) | Not reported |
Schmidt et al., 2006 (60) | Significant for only some measures | Not significant |
Simon et al., 2012 (61) | Significant (η2=.02) | Not significant |
Simon et al., 2013 (62) | Significant (d=.30) | Not significant |
Slade et al., 2008 (31) | Categories 2 and 5 both significantly more improved than category 1 (d=.35 and d=.48, respectively); not significant for category 2 versus category 5 | Significantly more treatment sessions for category 1 (control group) |
Discussion
Conclusions
Acknowledgments
References
Information & Authors
Information
Published In
History
Authors
Metrics & Citations
Metrics
Citations
Export Citations
If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download.
For more information or tips please see 'Downloading to a citation manager' in the Help menu.
View Options
View options
PDF/EPUB
View PDF/EPUBLogin options
Already a subscriber? Access your subscription through your login credentials or your institution for full access to this article.
Personal login Institutional Login Open Athens loginNot a subscriber?
PsychiatryOnline subscription options offer access to the DSM-5-TR® library, books, journals, CME, and patient resources. This all-in-one virtual library provides psychiatrists and mental health professionals with key resources for diagnosis, treatment, research, and professional development.
Need more help? PsychiatryOnline Customer Service may be reached by emailing [email protected] or by calling 800-368-5777 (in the U.S.) or 703-907-7322 (outside the U.S.).