processing....
Psychiatric medication errors have not been sufficiently studied despite the presence of several risk factors for their existence and for harmful clinical consequences. We found 2 medication error incident reports:
We found no studies in psychiatric settings that used an independent audit to identify medication errors and then examine a potential causal link between errors and clinical harm.
Psychiatry professional organizations have only recently identified medication errors as a patient safety and quality concern. The American Psychiatric Association (APA) first convened its Task Force on Patient Safety in 2002.[53] Adverse medication events were identified as number 1 out of 4 areas of priority. The Task Force recommendations were approved by the Board of Trustees on November 24, 2002, and by the Assembly Executive Committee on January 24, 2003, leading to the inception of the APA Committee on Patient Safety.
Independent audits. Medication errors in inpatient psychiatry deserve more study. If the studies of inpatient psychiatric settings and general care settings are considered, self-reporting of error has missed between 244 and 1000 errors per each error detected by other means.[7] Ideally, all hospitals would conduct an independent audit to determine the validity of their self-reported medication error rate.
Success in JCAHO, CMS, and state licensing surveys may give false assurance that a hospital's medication usage system is not error prone and may not be a basis for confirming that a successful hospital's self-reporting of medication errors is valid.[52]
Identification of bias in self-reporting is integral to the empiricism and statistical models of validation used in contemporary science. Independent validation of experimental observation is a requirement for scientific publication. It should also be a requirement during internal surveillance of complex medication usage systems in hospitals and other psychiatric healthcare settings.
We recommend use of a standardized, empirically validated methodology for medication error detection and reporting in all hospitals. Benchmarks for acceptable medication error rates in inpatient psychiatric settings should be established. Hospital-based psychiatrists need to know a hospital's methodology of error detection, the reported error rates, and current or future performance improvement initiatives that are under way to better detect, report, and rectify errors.
Adopt lexicon of medical terms and definitions. We also believe that psychiatry should adopt the widely accepted lexicon of medical error terms and definitions. Patients would benefit if psychiatrists and their medical and surgical colleagues spoke a common language when detecting, reporting, and managing ADEs. This might also increase awareness among psychiatrists that ADEs include not only ADRs but also medication errors. Broader incorporation of medical error terminology and performance improvement initiatives might also enable a more objective comparison of quality among psychiatric hospitals.
Medication errors of omission. Medication errors of omission occur in circumstances where medication treatment was not given, either because of a missed diagnosis or because of lack of medication access. The latter occurs when formularies are restricted in order to cut costs and/or increase profits. Medicaid, private insurers, and those agencies who will oversee the new 2006 Medicare drug benefit all use formulary restrictions. We believe this area needs further study. One possible strategy for increased reporting of medication errors of omission is the inclusion of medication errors of omission in FDA postmarketing surveillance.
The true costs of insurance. Increasing pressure to cut Medicaid budgets and to maintain or increase shareholders' earnings for private insurers will lead to ongoing attempts at restricting formularies. However, if the reports of Gorman, Lichtenberg, Soumerai, and others are correct, formulary restrictions for Medicaid, Medicare, and privately insured beneficiaries are actually increasing overall healthcare costs.[26-29] Given the cost of medication coverage and pressure on legislators to "do something," it is important to have better cost, safety, and healthcare quality data upon which to base legislative decisions. Matters of cost, safety, and healthcare quality are further complicated by ethics concerns when considering the financial incentive that formulary restrictions pose in the for-profit medical insurance industry -- an industry that posts annual earnings of billions of dollars.
Open reporting should be rewarded, not punished. Fear of punishment may deter the reporting of errors. The establishment of a nonpunitive culture must accompany the incorporation of the mainstream lexicon of medication error terminology and a broader awareness of medical errors in psychiatry. Administrative and clinical leaders are responsible for establishing systems and an ethos that rewards frequent and open reporting of errors and the use of educational review to improve error-prone processes. Inherent in a nonpunitive culture is the recognition that most errors are systems issues and not issues of badly performing individuals. Fear of punishment should not interfere with developments that diminish harm to patients.
JCAHO, CMS, and state licensing surveys could be more successful in detecting error-prone medication usage systems if regulatory standards included the following:
Education of primary care providers. Better education of primary care providers about the symptoms and management of depression and anxiety is essential. Programs designed to increase primary care clinicians' recognition of depression and other psychiatric disorders have not yet proved to improve treatment outcome.[54,55] However, studies that have added training in depression management in addition to education on better recognition of its presence have yielded better outcomes.[56]
The feasibility and sustainability of programs focusing on better education of primary care clinicians has not been established. Because of the time constraints of standard primary care practice, fast and efficient methods of screening, such as the Primary Care Evaluation of Mental Disorders (PRIME-MD) for anxiety, depression, and other psychiatric disorders, are recommended.[36] The Patient Health Questionnaire, a 9-item subset of the PRIME-MD, is a more concise screen for depression, if time constraints preclude use of the full PRIME-MD questionnaire.[57] The routine use of such simple methods for screening and ongoing monitoring might provide a sustainable way of reducing diagnostic errors in ambulatory medical settings.
Incentive programs to improve mental health diagnosis. A final recommendation is for managed care and insurance companies, as well as health systems and provider networks, to consider the implementation of incentives for better detection and treatment of mood and anxiety disorders. Pilot "pay for performance" programs, such as Bridges to Excellence[58] or the Integrated Healthcare Association initiative in California,[59] have focused on physical conditions such as diabetes or asthma. Quality measures do exist for the medication treatment and follow-up of diagnosed depression, such as performance standard and report card measures.[60] Adding incentives for better detection and treatment to the use of quality measures is a recommended first step.
In recent years, considerable progress has been made in developing and researching strategies for decreasing ADEs.[61,62] Specific prevention strategies have been divided into[63]:
Two elements are essential as the foundation for successful prevention strategies:
Examples of short-term strategies relevant to inpatient psychiatry include the following:
In addition, clinical and pharmacy staff would benefit from systematically monitoring the literature for published reports of preventable adverse events and from reviewing these reports in scheduled multidisciplinary team (such as patient safety team) meetings, which would stimulate performance improvement initiatives directed at further prevention of adverse events. Safety alerts from the JCAHO, the Institute for Safe Medication Practices, the FDA, and others can also be valuable sources of anecdotal reports on patient safety that are not reported in published journals and are not accessible through MEDLINE searches.
Informatics is being used to automate portions of the inpatient medication process by improving checks and controls in pharmacy software systems, incorporating machine-readable coding (bar coding) to ensure that the right drug is administered to the right patient, and preventing prescribing errors by using CPOE.[4,63,67,68] CPOE can dramatically reduce the rates of many types of prescribing errors, such as:
When utilizing integrated software, the prescriber has current drug information prior to entering an order, including:
CPOE also provides automated checks for:
CPOE systems can be integrated into other hospital data systems that include patient-specific clinical, demographic, and laboratory data.
There is empirical support for the success of medication error reduction after implementation of CPOE. One study conducted at Brigham and Women's Hospital in Boston demonstrated a 55% decrease in error rates,[11] and serious errors dropped by 88% in a subsequent study by the same group.[69] In another study conducted at LDS Hospital in Salt Lake City, CPOE implementation resulted in a 70% reduction in ADEs.[70]
Limitations of CPOE include:
A number of additional computerized interventions appear to have the potential to further decrease medication error rates in the inpatient setting. Among these are[62]:
Within psychiatry, no data are available regarding the efficacy of these technologies, but bar-coding linked to medication administration records would be expected to be especially important given the frequency of use of solid dose forms. Outside the hospital, tools that help patients track what medications they have actually taken may be especially beneficial in this population.
We conclude that several approaches are available that should reduce medication error rates in the short term in psychiatry that do not involve automation. Long-term, computerized physician order entry is likely to result in the largest benefit, although this remains to be demonstrated. Bar-coding linked to computerized medication administration records may also be beneficial.
The authors gratefully acknowledge the substantial contributions made by Susan Gile, MBA RN.