Does long term use of psychiatric drugs cause more harm than good?

Yes that is what I meant. A confusion on my part.

The first few comments posted here didn't even describe the article, so I thought I should at least do that. You can, of course, read the whole article yourself to form your own opinion. The article is organized as a pro-con debate (with the "con" side first) on whether the broad classes of medicines prescribed for psychiatric conditions (mostly mood disorders and schizophrenia) are helpful or not.

The writer who says that the drugs' "benefits would need to be colossal to justify [patient deaths associated with their use], but they are minimal" is Peter C Gøtzsche, a professor at the Nordic Cochrane Centre in Copenhagen. I would ordinarily expect someone with an affiliation with the Cochrane centers to have an evidence-based perspective on evaluating treatments for human disease, and I think he makes some good criticisms of study designs about drug effectiveness for treating psychiatric disorders.

The writers who say that "Psychiatric drugs are as beneficial as other treatments used for common, complex medical conditions. Leucht and colleagues reviewed the efficacy of psychiatric and general medicine drugs by analysing meta-analyses: they found that psychiatric drugs were generally as efficacious as other drugs" are Allan H Young, professor of mood disorders at King’s College London and John Crace, psychiatric patient and a writer for The Guardian. They in turn make several thoughtful criticisms of the studies Gøtzsche relies on to infer harm. My personal impression is that they have the better of the argument, because the conditions treated with the drugs mentioned in this article are themselves fatal, and if left untreated greatly increase patient mortality.

I think everyone who follows this research closely (as I do, as part of my journal club participation with researchers on human behavior genetics) has settled on the conclusion that patients are genetically diverse even if they have the same diagnosis, and therefore a drug that works for one patient may not work for another. But a drug that works for an immediate family member of the patient probably will work for the patient, and "talk therapy" of a kind proven to be safe and effective is an important both-and to add to treatment of major psychiatric disorders. Patients with major psychiatric disorders ARE living longer and enjoying better day-by-day functioning than ever before in my lifetime, so the statistics trump the anecdotes in showing that something about current treatment is working to help patients.

For any drug trial, if a patient switches from drug X to placebo, compared to patients who switch from drug X to drug Y, that could cause the placebo to seem worse than it actually is, due to withdrawal.

All the drug research I read regarding psychiatric drugs had a 3-6 month timeframe max.

Another flaw is that "Is the patient doing well?" was based on the subjective opinion of the doctor. (Even if the doctor does not know which patients are taking placebo, a competent psychiatrist should be able to tell. "Compliance with medication" is one thing doctors are trained to look for.) I would have liked to see other objective metrics, such as "How high can the patient score at Tetris?"

A multiyear psychiatric study is very hard, because many patients will struggle to follow any treatment plan. Also, placebos might cause the patient to initially get worse before they get better, whereas a drug might cover up symptoms while hurting long-term recovery. It's hard to find patients who would follow through for years.