Clinical practice changes in mysterious ways. You might think that after 2003, physicians who prescribe atypical antipsychotics would be ordering plenty of lab tests for glucose values and lipid profiles. That was the year that the Food and Drug Administration mandated drug label warnings and issued a letter to health care professionals about increased risk for diabetes and hyperglycemia when taking second-generation antipsychotics. Also at that time, a consensus statement by the American Diabetes Association and American Psychiatric Association issued monitoring protocols for all patients on these drugs that include getting baseline fasting lipid profiles and fasting plasma glucose levels.
Since then, clinicians have changed some of their prescribing practices–they’re prescribing olanzapine less often–but they’ve not increased laboratory screenings as recommended, according to a new study of data on 109,451 Medicaid recipients who started second-generation antipsychotics and 203,527 control patients who started albuterol but not second-generation antipsychotics.
The authors could only speculate on why that is. Theories about behavior change suggest that people are more likely to adopt behaviors if they see a clear relative advantage that’s consistent with existing experience (like switching to a lower-risk drug, or deciding not to treat). People are less likely to adopt a behavior simply because it will lower the risk for some future event, especially if the new behavior is complex (like metabolic monitoring may be for some psychiatrists). These phenomena seemingly were at play, for example, when warnings about increased suicide risk with pediatric use of antidepressants led to a decline in prescriptions but no change in the frequency of follow-up visits.
Alternatively, these trends may be related to physician specialty, with different levels of awareness about the risks of second-generation antipsychotics and behavior change in psychiatrists compared with primary care physicians, the authors suggest. No one really knows.
If you’re a physician who prescribes these drugs, do you order metabolic screening and monitoring? If not, why not? Your answers might prove illuminating.