We appreciate the opportunity to comment on the letters that have been received in response to our recent consensus statement on antipsychotic drugs and obesity and diabetes.

Before addressing the specific issues raised in each of these letters, I think it is important to note why the American Diabetes Association (ADA) and other organizations produce consensus statements (1). As stated in our clinical practice recommendations, “the need for a consensus statement arises when clinicians or scientists desire guidance on a subject for which there is a relative deficiency of comprehensive evidence that might otherwise allow for a more definitive statement to be made.” Therefore, it should be noted that such statements represent the expert opinion of the panel based on the presentations they heard, the literature they reviewed, and the considerable discussion among panel members while writing the statement. If there was a reasonable number of randomized control trials on the subject, there would be no need for a consensus statement, but rather the associations would issue an official “Position Statement” or clinical guideline. Thus, a consensus statement can be viewed as an expert recommendation that often precedes more definitive recommendations issued when sufficient additional data become available. These may or may not be different from those in the initial consensus statement.

The letter from Holt (2) comments that he was a member of another group that reviewed the evidence surrounding this issue and reached conclusions “that differed in some respects” from those of the ADA consensus panel. Their recommendations are due to be published soon. We are, of course, delighted that another group has deliberated seriously on this important clinical question and look forward to seeing their conclusions and recommendations. We hope that much new data come to bear on this issue in the near term. We also support Holt and each of the other correspondents in pointing out that antipsychotic medications are essential for people with schizophrenia and that their differential effectiveness is very important. We purposefully used the word “consider” throughout our statement to suggest that the advice we gave with regard to avoiding undesirable side effects should be one of many factors in deciding which medication to use. Such risk-benefit considerations are essential.

Citrome and Volevka (3) comment that “the report probably overreaches available evidence when suggesting that clinicians should consider prescribing one antipsychotic over another with the aim of avoiding diabetes.” We disagree with this statement and believe that the risk of developing diabetes must also be considered given the devastating effects of this disease and its complications. The efficacy of antipsychotic therapy is an extremely important factor but not the only one. They also express concern that our report might lead the inexperienced practitioner to inappropriately make changes in medications, exposing the patient to increased risk from an inadequately treated psychiatric illness. We would hope that practitioners think carefully about all the consequences of changing medications. We believe it would also be most unfortunate if they ignored the possibility that a patient could develop one or more of the serious side effects associated with these drugs. For that reason we strongly advocate and carefully described a monitoring regimen.

Isaac and Isaac (4) point out that the evidence may not support the view that there is a difference in the incidence of obesity in diabetes between those given first-generation antipsychotics and patients treated with second-generation antipsychotics (SGAs). Our panel did not address this issue since SGAs are far more widely used and preferred (with specific exceptions noted). They also point out that weight gain is common and can have many causes. We acknowledged that point, but it does not obviate the concern that many patients on SGAs gained substantial weight within weeks of drug initiation. There appears to be a differential propensity for weight gain depending on the drug selected.

The letter from Boehm et al. (5) supports the preceding sentence, and we agree that proof is lacking, despite weight gain being widely recognized as a risk factor for diabetes and our conclusion that the weight gain from SGAs correlates with the new-onset cases of diabetes. We appreciate that more data may be required before the U.S. Food and Drug Administration’s (FDA) Division of Neuropharmacological Drug Products itself adopts a ranking of diabetes risk for the various SGAs. However, the panel’s review of the available information, which also included data presented by the FDA supporting a differential risk of diabetes among the SGAs, was sufficient to unanimously reach the judgment we made. Again, a consensus statement denotes expert judgement and opinion and provides clinical guidance. It is understandable that in order for drug labeling to change, a higher level of evidence is required. We point out that based on our analysis, the FDA correctly judged that there was sufficient evidence available on the first SGAs to become clinically available to merit a warning label. However, we believe that there is currently insufficient epidemiologic data to either implicate or exculpate the newer SGAs in this regard. We anticipate that the monitoring recommended in the consensus conference report will, if implemented, accelerate the accumulation of this valuable data and improve our ability to care for these complex patients.

In closing, our statement attempted to bring attention to important factors related to the care of patients with psychiatric disorders. In reaching consensus, the panel provided expert opinion on the available data, including which patients might be at increased risk of developing adverse metabolic sequelae and specifying the baseline and follow-up monitoring that would be appropriate. We do not believe our statement provides the final word on the subject. Indeed we look forward to all future publications on the issue, which will undoubtedly enhance our knowledge and provide even better clinical guidance.

1
American Diabetes Association: Consensus development conference on antipsychotic drugs and obesity and diabetes (Consensus Statement).
Diabetes Care
27
:
596
–601,
2004
2
Holt RIG: Consensus development conference on antipsychotic drugs and obesity and diabetes (Letter).
Diabetes Care
27
:
2086
–2087,
2004
3
Citrome L, Volavka J: Consensus development conference on antipsychotic drugs and obesity and diabetes (Letter).
Diabetes Care
27
:
2087
–2088,
2004
4
Isaac MT, Isaac MB: Consensus development conference on antipsychotic drugs and obesity and diabetes (Letter).
Diabetes Care
27
:
2088
,
2004
5
Boehm G, Racoosin JA, Laughren TP, Katz R: Consensus development conference on antipsychotic drugs and obesity and diabetes (Letter).
Diabetes Care
27
:
2088
–2089.
2004