Ever since the introduction of an array of newer or “second-generation” antipsychotics over a decade ago, hospital networks, managed care organizations, and public mental health systems across the United States have watched psychiatrists and primary care physicians use ever-higher doses and ever-more complex combinations of these medications. The physicians’ rationale is that as long as psychotic symptoms remain, more medication is the necessary response — despite a wealth of studies of the older medications revealing that time-on-medication, not dosage, is usually the deciding factor. Hence the growth of antipsychotic poly-pharmacy. A rising tide of metabolic side effects, notably significant weight gain and Type II diabetes, resulted from this practice. Another result was breaking the bank, not only of public payers but of private ones, as well.
Daring the odds, the New York State Office of Mental Health (NYS-OMH) undertook, in late 2003, its project in "Rational Pharmacotherapy of Schizophrenia." The problem was how to move OMH’s very large health care system (800 psychiatrists, 200 internists, 26 inpatient hospitals serving 4,500 patients, and a network of outpatient clinics serving another 20,000 patients) in what its medical and administrative leadership believed to be a positive direction, in the face, generally, of front-line resistance. The hypothesis was that by asking LESS, not MORE, of practitioners, along with providing the necessary measurement tools and reinforcing the message, change could be accomplished because practitioners might be willing to take a chance on using fewer medications and prescribing less of each per patient.
The project began in November 2003, with a non-threatening but highly informative “kick-off” presentation by one of the nation's recognized experts in the pharmacotherapy of schizophrenia, George Simpson, MD, Chair of Psychiatry at The University of Southern California. Breakout groups spent the day discussing the component practices that tended to support poly-pharmacy. So that practitioners would be able to tell whether they could reverse this trend in measurable ways, experts in the change process itself, recruited from the Institute for Healthcare Improvement (IHI), were on hand to explain how to use effective measurement tools and strategies. Over the ensuing months, also following the format recommended by IHI, Learning Sessions were conducted regionally and across the state, to elicit suggestions from front-line practitioners for changing the system. The Chronic Care Model, described in several IHI publications in relation to chronic systemic illnesses such as diabetes, was easily transferable to the care of patients with schizophrenia. That model describes clearly how other resources, such as nurse-run health-and-wellness groups, and nutrition groups, can augment the role of primary care physicians and increase the likelihood of better patient outcomes.
It was at these Learning Sessions that things began to get interesting. At first, only the usual opinion leaders were willing to speak out, and several of these had their own reservations about changing their practices. Notably, some wanted to formulate their own lists of “acceptable” poly-pharmacy practices, rather than work toward a system-wide consensus. But others, initially less comfortable with speaking out on psychopharmacology issues, reached down and tapped their leadership potential, acknowledging potential large gains down the line. Those in this group reinforced each other between meetings by communicating through email, something that had not previously occurred around such purely clinical issues.
The next thing they told us — something agency management might have suspected but would never have dared to say out loud — was that the usual oversight structures, such as medical director approvals and local pharmacy and therapeutics committees, contributed more to the problems than to the solutions. That is, they often were committed to justifying the status quo — specifically, describing and calibrating ongoing practices — rather than uncovering problems and promoting changes. One hospital, for example, figured this out and set up a new structure consisting of the chief pharmacist, head nurse, and treating physician to determine best actions. Finally, the physicians at each local site demanded access to the technology that management had been using to browbeat them in the first place. They wanted access to the software programs, developed by inspired IT staff, that summarized, sliced, and diced all the prescribing practices, and they wanted it on their desktops, for daily use, not hauled out only at meetings. The take-home lesson for every participant, not very original but very persuasive, was that management may set the direction, but only the front-line practitioners know how to get there.
At the most successful of the 26 hospitals, there was a decrease in antipsychotic poly-pharmacy of nearly 15 percent within 6 months, as shown in the following figure:
These results continued for the remainder of 2004, leveled off, and then began to drop further.
At the next most successful site, there was a drop of 10 percent. (Given the large number of patients treated, these outcomes were statistically significant.) Best of all, no one was complaining. Here is how some of the local physicians described the change:
- “I'm developing a sense for who is on too much medication, and people are getting better as we are lowering their medications.”
- “Non-physicians are especially reluctant to consider lowering medications, fearing that the patient may become violent. It takes lots of reassurance, but it's worth it.”
- “Remember [patient] XY? He was in the hospital over 8 years and the staff had given up. Nothing worked. We reduced the number of medications he was on and worked on dosages, and he was discharged. No one had thought it was possible. ”
- “One of our doctors actually said to me, when I inquired of a patient who had been admitted a few months back, ‘Of course he's better; he's on less medication!’ ”
One source of the problem was that once a patient was started on a particular medication, the physician was reluctant to discontinue it, even if it was ineffective. Rather, a second or third medication of the same class was added to the regimen. Thus, a further measure of success was the length of the medication trial, with the assumption that shorter was better in getting to the effective agent. The graph below shows this process at the same hospital, with a generally similar trend:
We anticipate that the rate of antipsychotic poly-pharmacy will continue to fluctuate in a general downward direction as psychiatrists and other mental health staff grow increasingly accustomed to this new prescribing practice, and they experience the improved clinical functioning of patients on simplified, reduced regimens.
For further information about this project, contact NYS-OMH’s current Acting Chief Medical Officer, Lewis Opler, MD, at: lao1@columbia.edu. More information is also available at: http://www.omh.state.ny.us/omhweb/resources
