Dose–Effect Relations and Responsive Regulation of Treatment Duration: The Good Enough Level Michael Barkham and Janice Connell University of Leeds William B. Stiles Miami University Jeremy N. V. Miles University of York Frank Margison Manchester Mental Health and Social Care National Health Service Trust and University of Manchester Chris Evans Nottinghamshire Healthcare National Health Service Trust John Mellor-Clark CORE Information Management Systems, Ltd. and University of Leeds This study examined rates of improvement in psychotherapy as a function of the number of sessions attended. The clients (N = 1,868; 73.1% female; 92.4% White; average age = 40), who were seen for a variety of problems in routine primary care mental health practices, attended 1 to 12 sessions, had planned endings, and completed the Clinical Outcomes in Routine Evaluation—Outcome Measure (CORE–OM) at the beginning and end of their treatment. The percentage of clients achieving reliable and clinically significant improvement (RCSI) on the CORE–OM did not increase with number of sessions attended. Among clients who began treatment above the CORE–OM clinical cutoff (n = 1,472), the RCSI rate ranged from 88% for clients who attended 1 session down to 62% for clients who attended 12 sessions (r =-.91). Previously reported negatively accelerating aggregate curves may reflect progres- sive ending of treatment by clients who had achieved a good enough level of improvement. Keywords: dose– effect, good enough level, psychotherapy effectiveness, responsiveness In their classic meta-analysis of dose– effect relations in psy- chotherapy, Howard, Kopta, Krause, and Orlinsky (1986) charac- terized the path of client improvement as a negatively accelerating function of treatment length, in which 30% of clients made mea- surable improvement after 2 sessions, 41% after 4 sessions, 53% after 8 sessions, and 62% after 13 sessions. Subsequent research by Kopta, Howard, Lowry, and Beutler (1994) elaborated this finding by showing that the effective dose varied across different types of symptoms. Fewer sessions were required for improvement in dis- tress symptoms, and more sessions for characterological symp- toms, but Kopta et al. described the negatively accelerated pattern as common across problem types. From these classic reports has sprung a lively and substantial research literature on dose– effect relations in psychotherapy (e.g., Barkham et al., 1996, 2001; Barkham, Rees, Stiles, Hardy, & Shapiro, 2002; Dekker et al., 2005; Feaster, Newman, & Rice, 2003; Given, 2002; Gray, 2003; Grissom, Lyons, & Lutz, 2002; Hansen & Lambert, 2003; Hansen, Lambert, & Forman, 2002, 2003; Hoagwood, 2000; Howard, Lueger, Maling, & Martinovich, 1993; Kadera, Lambert, & An- drews, 1996; Kopta, 2003; Lueger et al., 2001; Lutz, Lowry, Kopta, Einstein, & Howard, 2001; Lutz, Martinovich, Howard, & Leon, 2002; Salzer, Bickman, & Lambert, 1999; Steenbarger, 1994; Warner et al., 2001). The negatively accelerating pattern has usually been interpreted as reflecting diminishing strength of each successive session. As Kopta (2003) put it, “the effect of therapy is greater in earlier sessions and increases more slowly at higher dosage levels” (p. 728). Reflecting concerns about administrative efficiency and fair allocation of scarce resources, discussion has focused on the op- timum number of sessions that clients should be offered. That is, “How much is enough?” (Kopta, 2003, p. 728). For example, Howard et al. (1986) suggested, “The present meta-analysis indi- cates that by 26 sessions, about 75% of patients have shown some improvement.... [I]n clinics that serve a large population with Michael Barkham and Janice Connell, Psychological Therapies Re- search Centre, University of Leeds, Leeds, United Kingdom; William B. Stiles, Department of Psychology, Miami University; Jeremy N. V. Miles, Department of Health Sciences, University of York, York, United King- dom; Frank Margison, Manchester Mental Health and Social Care National Health Service (NHS) Trust and Department of Psychiatry, University of Manchester, Manchester, United Kingdom; Chris Evans, Rampton Hospi- tal, Nottinghamshire Healthcare NHS Trust, Nottingham, United King- dom; John Mellor-Clark, CORE Information Management Systems, Ltd., Rugby, United Kingdom, and Psychological Treatments Research Centre, University of Leeds. Michael Barkham and Janice Connell were supported by the NHS Priorities and Needs Research & Development Levy. Correspondence concerning this article should be addressed to Michael Barkham, Psychological Therapies Research Centre, University of Leeds, 17 Blenheim Terrace, Leeds LS2 9JT, United Kingdom. E-mail: m.barkham@leeds.ac.uk Journal of Consulting and Clinical Psychology Copyright 2006 by the American Psychological Association 2006, Vol. 74, No. 1, 160 –167 0022-006X/06/$12.00 DOI: 10.1037/0022-006X.74.1.160 160