Sandra Steingard is a psychiatrist who has been treating people who are psychotic for 30 years. She recently published a piece in the Washington Post that spelled out some changes in her thinking about how we can best help people who struggle, over a lifetime, with things like schizophrenia and bipolar disorder. She has important things to say about what it means to live with mental illness and how to treat it.

The piece was prompted by a tragic shooting last fall, in which a man who told people he had been “hearing voices” killed 12 people at a Washington, DC, area Navy Yard. As she puts it, “Almost every day I meet with individuals who hear voices that no one else hears, are sure the TV or radio is talking to them or have such confused thinking that it is hard to understand what they are trying to tell me.”

Dr. Steingard explains that accepted practice, for many years, has been that patients with schizophrenia should be on antipsychotic medications, not just to control symptoms when they occur, but indefinitely. Often-cited studies from the 1970s and 1980s had found that patients who stopped taking medication after symptoms subsided were more likely to relapse than those who kept taking it. And while Dr. Steingard notes that very, very few people who hear voices or suffer delusions are a danger to anyone else, those voices and delusions can be very problematic for the person who’s having them, not to speak of those who love him. Still, preventing relapses isn’t the only priority. Staying on meds didn’t necessarily mean that the patient’s quality of life was better. And since antipsychotics have undesirable side effects, including dangerous weight gain, she began to have second thoughts.

Then she saw the results of a Dutch study that showed that over 7 years, patients who were not continuously on antipsychotic medication—”they stopped taking drugs when they became well but restarted them if symptoms emerged”—were much likelier to be able to work and have meaningful relationships than those who took medication continuously. And they had no higher rate of relapse. As NIMH director Tom Insel wrote in a blog post about the study, “Clearly, some individuals need to be on medication continually to avoid relapse. At the same time, we need to ask whether in the long-term, some individuals with a history of psychosis may do better off medication.”

So Dr. Steingard took a big step, and began discussing with a group of her patients the option of reducing their medication. Some two-thirds of the patients in the group decided to try cutting back, with close monitoring. “The Dutch study shifted the focus away from the belief that we need to eradicate all symptoms of schizophrenia to a focus on improving the quality of patients’ lives and health, the relationships they have, the work they do.”

Dr. Steingard’s account of the evolution in her thinking (and practice) is very compelling and it proposes, as Dr. Insel’s piece does, less reliance on medication and more on combining medication with things like family support and behavior therapy. And it involves seeing a person with mental illness as an active participant—indeed, the most important participant—in her treatment. “The point is that this is not a choice I should be making for my patients,” she writes. “It is a choice I need to make with them.”