by Geetha Jayaram, MD, MBA
Dr. Jayaram is Associate Professor, Departments of Psychiatry, Health Policy and Management
Johns Hopkins University, Baltimore, Maryland.
Kessler first introduced MEDWATCH in 1993[1] and indicated a need for a national reporting system, explaining that even well-designed clinical trials may fail to uncover problems that surface after initial widespread use of a medication. Combinations of drugs never researched may have new interactions never reported. Various factors influence the frequency of these events, their reporting, and the ultimate recall of medications.
Lazarou’s meta-analysis of 39 prospective studies in JAMA (1998)[2] revealed that the overall incidence of serious adverse drug events (ADEs) was 6.7 percent in US hospitals, ranking it as the 4th to 6th leading cause of death in hospitalized patients. Many factors influence ADEs, including type of hospital and medication, demographics, sampling strategies, honesty in reporting, protocols for reporting, and vigor of audits conducted.
Moore’s research in September, 2007, in the Archives of Internal Medicine[3] underscores the 2.6-fold increase in the reporting of serious adverse events to the US Food and Drug Administration in the last seven years. The paper provoked extensive media interest and raised the question, “What accounts for this increase?” Authors attempt a detailed explanation and urge better systems of post-marketing surveillance.
What about their pertinence to psychiatry? Rothschild[4] reported more medication errors among psychiatric inpatients and noted that atypical antipsychotics are responsible for 37 percent of errors/adverse events. In our unpublished work, we found that opiates and benzodiazepines were the most error-prone psychotropics, followed by antidepressants, antipsychotics, and mood stabilizers.
Our error rate by retrospective chart review is 2.8 per 1000 orders, far lower than published rates. Yet this rate may be meaningless to others and will have broad applicability only if we have a ‘best practices’ method that compares favorably with peers in the region.
Also, what about medication errors in psychiatry causing permanent disability or death? Rates may vary based on design, taxonomy of safety, and absence of reporting of “near misses.” In conducting a root cause analysis of the error, both individual and systems problems must be considered, with an evaluation of harm level, need for protocol changes, and education that is needed.[5]
Are we comparing apples and oranges when we compare state and privately run systems, community, and teaching hospitals? There are no standard reporting forms for psychiatry. Differences exist in admitted populations, organization of systems, reporting requirements, medication delivery processes, and use of information technology to reduce errors. Data collection and analysis varies. Finally, what about the patient’s role in error? This problem has never been studied nor published in psychiatry.
Clearly the increase in recent reporting is due to increased vigilance, the pharmaceutical industry, direct marketing and education of patients, internet access, and lawyers soliciting complaints through television advertisements. Errors increase with complexity of care processes and multiplicity of medications. Errors also increase with absence of reconciliation processes and poor hand-off mechanisms, two commonly blamed systems problems.
As psychiatrists, where do we go from here? Our solutions lie in examination of institutions with similar patient and systems characteristics. Definitions can be agreed upon between teaching hospitals, for example, and protocols outlined. There must be transparency in reporting within a culture of learning from errors. We must also learn from best practices resulting in low error rates and identify techniques and protocols that reduce error. Besides medications themselves, it is their unsafe use, under unsafe conditions, for certain populations that pose a risk. Psychiatry is new to this enterprise. It is time that we made a concerted effort to address this large public health problem through carefully conducted research, training, and dissemination of successful practices. Incentives through payment systems that correctly identify better quality are forthcoming and may promote the effort.
References
1. Kessler DA. Introducing MED Watch- A new approach to reporting medication and device adverse effects and product problems. JAMA 1993;269:2765–8.
2. Lazarou J, Pomeranz BH, Corey PN. Incidence of adverse drug reactions in hospitalized patients. JAMA 1998;279:1200–5.
3. Moore TJ, Cohen MR, Furberg CD. Serious adverse drug events reported to the food and drug administration, 1998–2005. Arch Intern Med 2007;167:175–9.
4. Rothschild JM, Mann K, Keohane C, et al. Medication safety in a psychiatric hospital. Gen Hos Psychiatry 2007;29:156–62.
5. Chang A, Schyve PM, Croteau RJ, et al. The JCAHO patient safety event taxonomy: A standardized terminology and classification schema for near misses and adverse events. Int J Qual Health Care 2005;17:95–105.
