This commentary will suggest how the research community can be instrumental in improving DSM-V and helping it avoid unintended consequences. According to several converging, anonymous (but I think quite reliable) sources to which I have had access, the draft options for DSM-V will finally be posted between mid-January and mid-February of 2010.
This commentary will suggest how the research community can be instrumental in improving DSM-V and helping it avoid unintended consequences. According to several converging, anonymous (but I think quite reliable) sources to which I have had access, the draft options for DSM-V will finally be posted between mid-January and mid-February 2010. There will then be just one additional month until mid-March for collecting comments. The good news is that the products of a previously closed process will finally be available for wide review and correction. The bad news is that there will be only a very brief period allotted for this absolutely crucial input from the field.
The research community has a central role and a great responsibility in taking advantage of this precious opportunity to carefully review and identify the problems in the DSM-V drafts and to suggest solutions.
Problems with the DSM-V process
The dangers of the “everything is on the table,”1 ambitious, innovative bias of DSM-V have been amplified by its secrecy and weak methods. There has been a remarkable lack of the free flow of ideas that is necessary to prevent any DSM process from becoming idiosyncratic and arbitrary. Many of the work groups have functioned mostly on their own without sufficient monitoring from the DSM-V Task Force, a large group of diverse advisors, or the field as a whole.
The original DSM-V timeline had the fatal flaws of scheduling field trials before the proposed changes could be vetted by the field and an impossible publication deadline of May 2012. Fortunately, my sources suggest that this plan has been shelved, and that a new timeline has field trials following the posting of options and a new DSM-V publication date of May 2013.
Unfortunately, there are still numerous process problems. There is a continued bewildering secrecy concerning timelines and methods. My sources indicate that a grant request for external funding for the DSM-V field trials has been rejected, and there is no indication that there is sufficient money, time, or expertise to conduct meaningful field trials that would measure the impact of changes on the rates of disorder. The few papers published to date by the DSM-V leadership1-3 (and the wordings of the few work group criteria sets that have surfaced at meetings or informally) display a lack of the one skill that is absolutely essential in crafting an acceptable diagnostic manual-the ability to write clearly and consistently.
It will be no surprise if the draft criteria sets that appear early next year are written poorly and include many worrisome suggestions. This should not be at all blamed on the DSM-V work group members. It is my experience (repeated 3 times with DSM-III, DSM-IV-TR, and DSM-IV) that early work group drafts are always, and probably inherently, riddled with serious problems.
Work group members are selected because of their special contribution to research in their own narrow area of expertise. They tend to overvalue their own section and make decisions based on highly selected research and clinical experiences. Thus, work groups routinely have an overconcern about false negatives; an underconcern about false positives; and insufficient concern about how suggestions will eventually play out in the general psychiatric and primary care settings, where most people receive their diagnosis. Add to this that work group members lack experience in the difficult art of criteria writing, and it is guaranteed that their first products will usually need many months of extensive internal and external review and detailed editing before being ready for field testing.
The iterative polishing and disciplining of work group product must come from an integrated effort that includes contributions from the DSM-V leadership and editorial staff; the task force as a whole; a large and diverse group of advisors; the oversight committee; and, finally and most important, the field at large. The value of the first DSM-V drafts will be only that they serve as a starting point for public comment and the painstaking revision process.
What harm can DSM-V do?
Elsewhere, I have outlined the 3 harmful unintended consequences that emerged unexpectedly from DSM-I ; namely, a contribution to the false epidemics of autism and attention deficit disorder, and a forensic disaster that has led to the inappropriate psychiatric commitment of sexually violent offenders.4-6 These unpleasant surprises occurred despite the fact that DSM-IV was stubbornly unambitious, discouraged all changes, required extensive empirical documentation, and was widely reviewed by the field at large and by numerous advisors. The risks of unintended consequences from an ambitious, secretive, and poorly organized DSM-V are numerous and significant. My focus here will be only on the ways in which DSM-V may be costly and risky to the research enterprise.
The criteria sets for the most widely studied disorders have been quite stable since the publication of DSM-III in 1980, and indeed since the publication of the Research Diagnostic Criteria in 1978. These DSM criteria sets served as the foundation of the structured and semi-structured interview instruments widely used in all clinical and epidemiologic research. Whenever DSM-V makes a change in a criteria set, this will necessitate that changes be made in the instruments used to assess that diagnosis. Aside from the considerable cost and inconvenience occasioned by such changes, they have the potential to break the highly desirable continuity between the past and ongoing research and all future research findings. The new diagnostic criteria will have untested psychometric performance characteristics and may result in a very different definition of “caseness.” This would make it extremely difficult to interpret differences in findings across time, because the studies will have been done with the different criteria. For example, this apples and oranges problem will greatly complicate the already difficult interpretation of the often radically different rates of mental disorder determined by different epidemiological studies.7-10
A prime example of how far the ambitions of the DSM-V Task Force has exceeded its grasp is its goal to develop and market a set of new interviewing instruments to be used in conjunction with DSM-V.3 While the commercial motivation is understandable, the disruption of methods continuity would be unfortunate, and the costs of switching to a new system of instruments would be prohibitive and wasteful. Moreover, nothing in the work to date by the DSM-V Task Force inspires confidence in its ability to produce and test useful new interviewing instruments, and it would seem to have its hands more than full producing DSM-V itself without needing other distractions.
As I have argued elsewhere, there are serious risks of including a number of prodromal and subthreshold conditions as official diagnostic categories in DSM-V.11-13 The most appealing subthreshold conditions (minor depression, mixed anxiety depression, minor cognitive disorder, and prepsychotic disorder) are all characterized by nonspecific symptoms that are present at extremely high frequencies in the general population. These proposed “disorders” might well become among the most common diagnoses in the general population-particularly once they are helped along by drug company marketing-resulting in excessive use of medications that often have serious long-term complications associated with weight gain. Early case finding is a wonderful goal, but it requires a happy combination of a specific diagnostic test and a safe intervention. Instead, we would now have the peculiarly unhappy combination of a wildly false positive set of criteria with potentially dangerous interventions.
Prospective epidemiological research suggests that DSM-IV is already quite overinclusive.14 Further widening the net would go even further in both medicalizing normality and trivializing psychiatric diagnosis. Altogether, in my view, the costs and risks of the subthreshold diagnoses far outweigh any possible current gains.
What can be done to save DSM-V from itself?
The DSM-V process would not be in its current state if it had been self-correcting and/or open to external suggestions. Influencing its direction now will not be easy, but is certainly not impossible. It will require a sustained external pressure that the research community is well positioned to apply.
Optimism that DSM-V can be saved from itself springs from the fact that external pressure has already resulted in the following improvements, however reluctantly made:
1. Appointment by the American Psychiatric Association (APA) Board of Trustees of an oversight committee to monitor the work on DSM-V
2. Postponement of field trials until after options have been posted and reviewed
3. Reduction of hype about a “paradigm shift”
4. Increased recognition of the value of caution
5. Likely postponement in the DSM-V publication date to May 2013.
There are 3 levers of pressure that the research community can exert to affect a more open, empirically based, and accurate DSM-V.
1. Most immediately, starting in January, individual researchers can each have a valuable correcting role by pointing out the specific problems in their areas that will be caused by the various DSM-V suggestions for change.
2. Within the APA itself, the most relevant components are the Council on Research and Quality and the newly appointed oversight committee, which includes prominent spokespeople for the research community.
3. The APA will be exquisitely sensitive to pressure from the research community-most especially if it comes from NIMH, NIDA, and/or NIAAA, but also from other relevant research-oriented organizations within psychiatry, psychology, and the neurosciences. The APA realizes that it holds the franchise to publish the DSMs only by historical accident, and that this is easily revocable if enough interested organizations lose confidence in its competence and its ability to control its inherent conflict of interest.
Another possible contribution to DSM-V that has excited many psychiatric researchers-but which is certainly premature-is the proposal to go beyond the descriptive method used in the DSM system and instead to attempt to base the classification on the exciting new findings from the revolution in neuroscience.15 This goal would certainly be highly desirable, but, in my view, should not play any current role in creating the DSM-V diagnostic criteria. As an official nomenclature, DSM-V must follow behind research and include only well-established and widely agreed on findings. The next 6 months are certain to be the most important in the development of DSM-V -especially because the field trials will probably not measure impact on rates and are thus likely not to be very informative. Researchers should carefully review DSM-V drafts as they emerge and make their concerns known.
1. Kupfer D, Regier D, Kuhl E. On the road to DSM-V and ICD-11. Eur Arch Psychiatry Clin Neuroscience. 2008;258(suppl 5):2-6.
2. Kupfer D, Kuhn E, Regier D. Research for improving diagnostic systems: consideration of factors related to later life development. Am J Geriatric Psychiatry. 2009;17:355-358.
3. Regier D, Narrow W, Kuhl E, Kupfer D. The conceptual development of DSM-V. Am J Psychiatry. 2009;166:645-650.
4. First M, Halon R. Use of DSM paraphilia diagnoses in sexually violent predator commitment cases. J Am Acad Psychiatry Law. 2008;36:443-454.
5. First M, Frances A. Issues for DSM-V; Unintended consequences of small changes: the case of paraphilias Am J Psychiatry. 2008;165:1240-1241.
6. Frances A, Sreenivasan S, Weinberger LE. Defining mental disorder when it really counts- DSM-IV-TR and SVP/SDP Statutes. J Am Acad Psychiatry Law. 2008;36:375-384.
7. Robins L, Helzer J, Weissman M, et al. Lifetime prevalence of specific psychiatric disorders in three sites. Arch Gen Psychiatry. 1984;41:949-958.
8. Kessler R, McGonagle K, Zhoa S, et al. Lifetime and 12-month prevalence of DSM-III-R psychiatric disorders in the United States. Results from the National Comorbidity Survey. Arch Gen Psychiatry. 1994;51:8-19.
9. Conway K, Compton W, Stinson F, Grant B. Lifetime comorbidity of DSM-IV mood and anxiety disorders and specific drug use disorders: results from the National Epidemiologic Survey on Alcohol and Related Conditions. J Clin Psychiatry. 2006;67:247-257.
10. Kessler R, Chiu W, Demler O, et al. Prevalence, severity, and comorbidity of 12-month DSM-IV disorders in the National Comorbidity Survey Replication. Arch Gen Psychiatry. 2005;62:617-627.
11. Frances A. A warning sign on the road to DSM-V: beware of its unintended consequences. Psychiatric Times. August 2009;26(8):1-9.
12. Woods S, Addington J, Cadenhead K, et al. Validity of the prodromal risk syndrome for first psychosis: findings from the North American Prodrome Longitudinal Study. Schizophr Bull. 2009;35:894-908.
13. Carpenter W. Anticipating DSM-V: should psychosis risk become a diagnostic class? Schizophr Bull. 2009;35:841-843.
14. Moffitt T, Caspi A, Taylor A, et al. How common are common mental disorders? Evidence that lifetime prevalence rates are doubled by prospective versus retrospective ascertainment. Psychol Med. 2009(September).
15. Andrews G, Goldberg D, Krueger R, et al. Exploring the feasibility of a meta-structure for DSM-V and ICD-11: could it improve utility and validity? Psychol Med. In press.