Welcome to From Insults To Respect. Today’s topic involves life and death issues.

When some people become concerned about their emotions, moods, behavior, or the ups and downs of life, they often seek professional mental health services. If they go to someone in the medical profession, they typically find that after a visit of somewhere between ten minutes to an hour they have been labeled as having a mental disorder and then are prescribed one of the psychiatric drugs, or even a whole cocktail of them. This pattern of dealing with such concerns leads, in some circles, to a conflict. Some involved in the conflict think that the pathologizing of these very human experiences and the encouraging of the use of drugs to manage them is the very opposite of a healthy approach. Meanwhile, other people feel that drug treatment for these types of concerns are enormously helpful, and anyone questioning this position is irresponsible. Insults have been known to fly back and forth from people on both sides of this conflict.

Now it just so happens, I have extensive training in the biological, psychological, social, and cultural factors that influence emotions, moods, and behavior. I have also taken specific coursework in psychopharmacology, statistics, and evaluating research methodology. To my dismay, after studying the relevant research, I have found that the current psychiatric medical approach for addressing these types of concerns appear to be causing far more harm than good. Please don’t misunderstand me. I do think the medical model can be enormously helpful when it comes to treating such problems as lesions, tumors, microbe infections, bone fractures, tissue tears, and body organ blockages. It is the psychiatric medical model that I have found seriously flawed. You can hear me, for free, being interviewed on this topic by the insightful Emily Whyte Rubin on a recently recorded Feeling Deeply podcast by clicking HERE.

Whenever I share my conclusions on this topic, I have found, at times, insults flying my way. This occurs even though I really don’t have a conflict with those who accept unquestionably the medical model for addressing psychological concerns.

You see, when I look to see if someone has a conflict, I think of the word DIG. This reminds me to dig to find what the person DESIRES in this situation, what is INTERFERING with the person’s desire, and whether or not the person believes someone is GUILTY of doing something wrong.

Being guilty, as I use the term, means the person would be deserving of some punishment, such as being insulted.

So, from my perspective, I do have a desire-incompatible problem with those who support the psychiatric medical model because I desire that they stop pathologizing these concerns and stop advocating that people take psychiatric drugs. However, I don’t view them as guilty for the position that they have come to accept. I believe that they mean well. I seek to treat them respectfully, while doing my best to advocate that they change their minds.

That’s not to say that I advocate that those people already taking these drugs should immediately stop taking them. There are some physiological withdrawal effects that occur if someone suddenly stops, which some experience as awful. If some of you who have been taking these drugs decide that you want to try to wean yourself off, a good resource to help can be accessed HERE.

Although I don’t have a conflict with those who believe in the psychiatric medical approach, I do find that as I go about advocating for my own view on this issue, many with opposing views do end up having a conflict with me.

There are, however, other people who do support the psychiatric medical model, and yet, like me, treat people with opposing views respectfully. Today’s post comes out of one such respectful discussion. This person, without insulting me, defended his support of the psychiatric drugs by claiming that all of the research clearly indicates that the “antipsychotic” drugs, when used by people diagnosed as having schizophrenia, decrease the risk of dying. I then, very respectfully, asked that he supply me with the references that he is relying on to make his assertion, and within a very brief period, he did just that.

Since this is a public forum, and since I will be critiquing his position, it strikes me as unkind to mention his name. However, despite this being a public forum, I do invite him, and anyone else, to critique anything I say in this, or any of my post.

With that said, let us move on to my critique.

What Evidence Exists For Informing Us As To Whether or Not These Drugs Reduce The Risk of Dying

Prior to looking at the articles that the respectful criticizer of my position provided to defend his belief that the “antipsychotics” reduce mortality for people being treated for schizophrenia, let’s first look at some relevant research studies that he left out. You will note that each time I use the term “antipsychotics” I put it within quotation marks. This is to help remind readers that these drugs do not work like antibiotics, which are designed to kill bacteria that have infected an organism. Nor am I at all convinced that these drugs work to counter any of the causes that lead someone to be given a psychotic label.

The Studies That Were Left Out

Patients diagnosed as having schizophrenia have a 15-20 year shorter life expectancy than the general population, according to a 2014 study published in the Annual Review of Clinical Psychology titled “Excess Early Mortality in Schizophrenia.” Just in case there is any confusion of what “mortality” means, it means risk of death.

Earlier studies suggested that part of the reason for these early deaths might be due to the “antipsychotic drugs” that these patients are prescribed. For example, in 2006, the British Journal of Psychiatry published an article titled “Schizophrenia, Neuroleptic Medication and Mortality.”  “Neuroleptic medication” is another name used to refer to “antipsychotics.” The authors found that over the 17-year follow-up period of their study there was a graded relationship between the number of “antipsychotic” drugs prescribed and mortality of those with schizophrenia.

Adjusted for age and gender, people with schizophrenia taking either no “antipsychotic,” one, two and three or more antipsychotics had relative mortality risks of 1.29, 2.97, 3.21, and 6.83 respectively. Said in a simpler manner, individuals who were labelled as having schizophrenia who did not take any “antipsychotic” drugs had the least chance of dying during the 17-year follow-up period, compared to the other patients who were labeled as having schizophrenia but were taking “antipsychotics.”  If the patients took just one of the “antipsychotic” drugs, they were more likely to die, and if they took more than one “antipsychotic” drug, they were even more likely to die. The association remained stable throughout the observation period.

The authors note that it has been claimed that the contemporary high natural mortality in schizophrenia results from a variety of lifestyle factors. Of these factors, the study that they did was able to consider several (smoking, exercise, body mass index, blood pressure, serum total and HDL cholesterol). After adjustment for these factors the excess mortality of people with schizophrenia persisted. The association with “antipsychotics” and mortality was, according to the authors, very clear.

A 2007 study was published in the Archives of General Psychiatry titled “A Systematic Review of Mortality in Schizophrenia: The Differential Mortality Gap Worsening Over Time.” The authors wrote:

Mental health services have advanced in many parts of the world during the past few decades. Apart from a different mix of community-based care, the introduction of the second-generation antipsychotic medications [also referred to as atypical antipsychotics] in the early 1990s was initially found to be associated with better quality of life and reduced risk of relapse.^77–79 More recent trials have questioned the clinical superiority of second-generation antipsychotic medication,^80,81 and concern is now widespread about the adverse effects associated with these medications.⁸² In particular, compared with typical antipsychotics, several of the second-generation antipsychotics are more likely to cause weight gain and metabolic syndrome.⁸³ Because the metabolic syndrome is associated with a 2- to 3-fold increase in cardiovascular mortality and a 2-fold increase in all-cause mortality,⁸⁴ these adverse effects would be expected to contribute to even higher SMRs [Standard Mortality Ratio] in the next few decades.^85,86

During the 1970s through the 1990s, the authors note, mortality rates were improving for the general population. In contrast to this general trend, mortality of those classified as having schizophrenia was trending in the opposite direction. The median SMR for those diagnosed in the 1970s, 1980s, and 1990s were 1.84, 2.98, and 3.20, respectively. The higher the SMR, the higher the rate of mortality.

This study also indicates that the use of “antipsychotics” that are prescribed for people diagnosed as having schizophrenia is least for the least developed countries, while countries with emerging economies use a little more, and the developed countries use the most. Here are the SMRs for these patients: the median all-cause SMRs for least developed, emerging economy, and developed countries were 2.02, 2.19, and 2.79, respectively.

The authors conclude,

Adverse health outcomes associated with weight gain and/or metabolic syndrome (eg, myocardial infarction, cerebrovascular accidents, or cancer) may take decades to fully emerge. Thus, it seems likely that studies undertaken in the 1990s (ie, the most recent studies included in this review) would capture only a small fraction of the eventual burden of mortality associated with the adverse effect profile of the second-generation antipsychotic medications. In light of the rising secular trends in SMRs already identified by this review, the prospect of further increases in mortality risks for schizophrenia is alarming.

The following year, 2008, a study published in the journal Epidemiologic Reviews, titled “Schizophrenia: A Concise Overview of Incidence, Prevalence, and Mortality,” found further evidence that supported the 2007 Archives of General Psychiatry concerns.

In 2009, another study looked at the relevant research, this one appearing in the journal Schizophrenia Research. Titled, “Influence of Antipsychotics On Mortality in Schizophrenia: Systematic Review,” it concluded, “There is some evidence that long-term exposure to antipsychotics increases mortality in schizophrenia. More rigorously designed, prospective studies are urgently needed.”

Also in 2009, there was a relevant study published in the prestigious New England Journal of Medicine. Titled, “Atypical Antipsychotic Drugs and the Risk of Sudden Cardiac Death.” It first reviews the relevant research regarding the “typical” antipsychotic drugs and concludes that there is extensive data linking them to an increased risk of sudden cardiac death. It then notes that less is known about whether the same risk is associated with the newer second generation “antipsychotic” drugs often referred to as atypical antipsychotics. After doing the best that these researchers could do to match the users of this type of drug with those who were not users, they found that the atypical drugs had an adjusted rate of sudden cardiac death that was twice that for nonusers and that this did not differ significantly from the rate for users of the typical “antipsychotic” drugs. Moreover, users of each of the six most frequently prescribed antipsychotic drugs had a significantly increased rate of sudden cardiac death, and the risk of sudden cardiac death increased with an increasing dose among current users of typical or atypical antipsychotic drugs.

The researchers conclude, “Our data show that in a large retrospective cohort of adults, current users of the atypical antipsychotic drugs had a dose-dependent increase in the risk of sudden cardiac death that was essentially identical to that among users of the typical agents.”

The researchers mention some limitations of their study:

The primary limitation of our study is the potential for confounding by factors associated with the use of antipsychotic drugs. For persons with serious mental illness, these factors include cardiovascular and other somatic disease; concurrent use of other proarrhythmic medications; mood disorders; behavioral risk factors, including substance abuse, poor self-care, and smoking; and other effects of mental illness.¹² However, both the study design and analysis included several provisions to manage confounding.

In 2010, the British Journal of Psychiatry published a study titled “Twenty-five Year Mortality of a Community Cohort with Schizophrenia.” Here, data from 1981 to 2006 was examined. The authors concluded that:

This study suggests that the natural cause mortality of schizophrenia is increasing, a finding that must be of concern to everyone involved with this disease. Further large-scale long term follow-up studies are needed to establish the reasons behind this increase and to suggest useful interventions.

So, taken together, these studies express concern about patients classified as having schizophrenia dying at an earlier age than the general population. Moreover, because the newer generation of “antipsychotics” are more likely to cause weight gain and metabolic syndrome, clear risk factors for increasing the risk of early death, the theory that the “antipsychotics” may be a significant factor for some of these early deaths has been discussed in several peer reviewed journals. Several studies provided evidence for this theory.

The Studies That Were Cited By The Person Who Criticized My Position

Recall that what prompted the respectful critique of my position was my opinion, derived from reviewing the research, that the “antipsychotics” appear to cause more harm than good, particularly in the long run. As science writer Robert Whitaker summarizes this research:

“The drugs are supposed to provide the benefit of reducing psychotic symptoms. But the research reveals that, over the long term, this benefit turns into a negative, and so, over the long-term, there are only negatives to be chalked up: the increased chronicity of psychotic symptoms, the impaired functional outcomes, the worse cognitive functioning, and, of course, a broad range of “side effects,” such as tardive dyskinesia, metabolic problems, sexual dysfunction, and so forth. Such is the bottom-line arithmetic that makes the case against antipsychotics.”

For a free PDF that fully reviews the evidence for Whitaker’s above statement, see HERE.

At the time that I heard from my respectful criticizer, I had really not fully examined the existing evidence that bears on his theory that this class of drugs actually decrease the risk of dying.

My first approach to looking into this was to contact Robert Whitaker to see if he had heard about that theory and whether he knew of any evidence to support it. He replied:

I intend to write about this idea that antipsychotics reduce mortality as soon as I get time. Antipsychotics of course cause all sorts of adverse effects associated with increased mortality.
Bob 

Disappointed, but undeterred, I decided to do a Google Scholar search to see what relevant research studies I could find, and above I summarized the evidence that the drugs actually increase the likelihood of a person dying. Then I retrieved all of the studies that my respectful criticizer provided that he said supports his contention. Below, I shall critique each and every one of them. But in doing so, the length of this post becomes quite a bit longer than most of my readers are willing to read. So, for their sake, I will first summarize the whole group of them, and for readers who want to delve into my critique of each specific study mentioned by my respectful criticizer, they can do so.

My Summary

The first study on my critic’s list used a meta-analysis approach that looked at 596 studies, all of which looked at patients that were randomly assigned to either an “antipsychotic” or a placebo for 13 weeks or less. For the patients in this analysis that had been labeled as having schizophrenia, there was no statistically significant difference in mortality between the two comparison groups.

Upon reading this article, I was left puzzled as to why this article was included in my critic’s list. As I have said, his contention was that every research paper published demonstrated “antipsychotics” decrease mortality for schizophrenia labeled individuals. Rather than supporting his position, this article supports the theory that given the evidence that the “antipsychotic” drugs have a number of serious side effects, and once started, physiological withdrawal effects can be awful, it is prudent to safely hold off on the use of these drugs for at least 13 weeks to see if some patients will begin to recover without them. Despite this evidence, I am under the impression from dialogues with many psychiatric patients that the standard practice is to prescribe these types of drugs within somewhere between 10 minutes to an hour after the patient is first brought to the attention of a prescribing doctor.  

Before moving on, I would like readers to note that this was the only randomized controlled study on my critic’s list of references that he said supports his position. The rest of the studies compared different groups of patients that were not randomly chosen to be in each of the groups. These studies compared a group of patients described as not taking “antipsychotics,” with others said to be either taking just one of them, more than one, different dosages of them, or another drug such as a benzodiazepine. Without random placement to each of the groups, these studies, from a scientific point of view, cannot speak to whether it was the prescribed drug that was the cause of the different rates of mortality, or other reasons. This is an essential point to keep in mind.

Even when a study does have random assignment to each of the different comparison groups, it is quite common that when researchers try to replicate its findings by doing another similar study, very different results occur. This inability to always replicate such studies occurs for a variety of reasons. First, random assignment only increases the likelihood, but does not guarantee, that the different groups being compared will be so similar prior to any intervention that any difference found after the intervention is due solely to the different interventions each group does or does not receive. Thus, with random assignment the pre-intervention groups can end up quite different. When this happens, the researchers can end up attributing any difference between the groups after the intervention to the intervention when the difference might be due to the groups being poorly matched prior to the intervention. A second reason why random assignment studies are not perfect is, those who did the experiment may report the results incorrectly because of a mistake, subconscious biases, or monetary motivations that lead to bending the truth.

Now, I am not faulting researchers for trying to get some relevant evidence using the nonrandom assignment to groups methodology. Although random assignment to groups is considered the gold standard, carrying out long-term studies with psychiatric patients is almost impossible. Patients who are assigned to take the drugs often don’t take them, or don’t take them as prescribed, and the longer the study, the more likely this will happen. A significant number of patients wander off and can’t be found at various follow-up points. Many doctors refuse to participate in long-term studies that utilize a placebo group believing it is unethical to withhold a promising treatment for such a long period of time when they see their patients suffering; and over the decades newer, more promising interventions come along, so to continue to use the drugs that were popular at the start of the study would be considered unethical.

So, given the absence of long-term random assigned studies, some weaker methodologies make sense but it is crucial to be very tentative in drawing any conclusions.

The most important finding from the set of nonrandom to groups studies my critic provided, from my perspective, came out of a 2009 study. There we find that patients that were on “antipsychotics” for less than 6 months had a lower rate of dying than those who were on the drug for longer periods. The authors note this in the results section of their article with a single sentence, stating,”Patients who used antipsychotics for less than 6 months had especially low mortality rates.” This is accompanied by a graph showing this. Then, in the discussion section, the authors again mention this in a single sentence.

If you read just the abstract of the article, as many practitioners do because time reading research studies is not billable, you would find no mention of this statistically significant finding, a finding that could have life saving implications. And this finding can easily get lost in the numerous paragraphs of the study which focused instead on the evidence that those on the drugs for longer periods had a lower mortality than those who were said to be in the “no antipsychotic group.” However, the vast majority of the patients said to be in the “no antipsychotic group,” and probably all of them, actually were on these drugs at various points throughout the study. While they were hospitalized, they were almost certainly on the drugs, according to one team of researchers who had engaged in a similar study using the same data set, and many were hospitalized multiple times. What the researchers actually meant by labelling the group the “no antipsychotic group” were patients who didn’t take them each time they were released from the hospital. Thus, each time they left the hospital, they were suffering from withdrawal reactions.

There are a number of other serious problems with the design of these types of studies that was chronicled in a peer reviewed article that came out shortly after the 2009 study, including incomplete reporting of data, questionable selection of drug groups and comparisons, important unmeasured risk factors, inadequate control for potentially confounding variables, exclusion of deaths occurring during hospitalization leading to exclusion of 64% of deaths on current “antipsychotics” from the analysis, and survivorship bias due to strong and systematic differences in illness duration across the treatment groups. The writers of this critique concluded:

It is likely that many of these patients died from the effects of suicide or cardiovascular disease while admitted to a hospital. As this approach likely underestimates mortality in patients treated with antipsychotic medications, this critical methodological factor might help to explain why the previous literature arrived at different findings with regard to antipsychotic-related mortality risk.

Another huge problem with these types of studies is this: 90% of the patients at the end of them were still living. Thus, it is very possible that although a relatively higher percentage of patients not taking “antipsychotics” during periods when they were not in the hospital died when compared to the other patients during the five to eleven year periods that were looked at during the course of these studies, we still don’t know what would have happened if we could have looked at what age all of these patients ended up dying. The added risk factors that come with ingesting these drugs, such as weight gain and metabolic syndrome, may not exert their influence on mortality rates until a longer period of time.

And yet another major problem with concluding from these studies that the “antipsychotics” reduce the risk of dying is this: It appears from one of these studies that schizophrenia patients during the time frame looked at had no increased risk of having a diagnosis of treatable nonfatal ischemic heart disease or cancer but had a far greater risk of dying from these conditions, suggesting substantial underdiagnosis and/or undertreatment.

This presents the theory that when compared to those patients in the other comparison groups, the group of individuals who did not take “antipsychotics” when they were not hospitalized may have had a higher percentage of members who not only avoid “antipsychotic” treatment, they also avoid all medical treatment. With this theory, it is not taking the “antipsychotics” that protects the patients from dying, but rather it is the increased likelihood that more “antipsychotic” compliant patients are more likely to comply with all medical treatment. Said another way, with this theory, we have less of a reason to conclude “antipsychotics” provide some protection from dying from these conditions; rather, it is this non-compliance to all medical treatment characteristic of some in the “no antipsychotic” treatment group that is the real cause for the statistical increased rate of death for those in this group during the relatively short window of time that these researchers were able to peek into.

In considering how dependable the evidence is from this entire set of studies, it is worthwhile to look how frequent they came up with contradictory findings. A few of the studies, for example, found a statistically significant decreased risk of suicide for those taking “antipsychotics” when compared to the so called “no antipsychotic” group, while another study that looked at the same data set found no significant difference. One study found a significant reduction in risk of dying when the patients used the “antipsychotic” clozapine, whereas another study did not find this, and found instead that clozapine was one of the drugs that appeared to slightly increase the risk of dying. A couple of studies found that even patients who took the highest doses of “antipsychotics” had a similar risk of dying when compared to a moderate dose, while another found the highest dosed patients were at an increased risk of dying. Thus, anyone making firm conclusions based on these types of studies is trying to build a house on shifting sands.

One finding in the set of studies provided by my critic is that a common way to prescribe “antipsychotics” is to prescribe it along with a benzodiazepine, a pill that has some similarities to ingesting alcohol. In one study, nearly 40 percent of the patients were prescribed a cocktail of both drugs, and these patients had a statistically higher likelihood of dying than those who were just prescribed an “antipsychotic.” Making the case stronger that the benzodiazepines were causing the increased risk of death was that the higher the dose of this drug, the higher the rate of mortality.

So, my very tentative conclusion is that the weight of the evidence suggests that in the short run (13 weeks or less) the “antipsychotics” do not significantly affect mortality rates for patients labeled as having schizophrenia. Those patients who are on “antipsychotics” for more than 6 months are at a statistically significant increased risk of dying compared to those who were on this type of drug for shorter periods. There is some weak evidence that “antipsychotics” may provide some protection from dying for those patients taking them for periods of from 6 months to less than 11 years when compared with patients who come off the drugs each time they leave the hospital, but interpreting this statistically significant finding in that way is highly problematic for the numerous reasons that I mentioned above. For patients taking these drugs for longer periods of time, the available data suggest real concerns that these drugs significantly reduce the lifespan of patients.

Finally, all of the studies that provide some support for the theory that, “antipsychotics” reduce the rate of death had researchers with either pharmaceutical company conflicts of interest or were members of departments of psychiatry, the very profession that relies heavily on earning income by prescribing these drugs. It seems to me that given that the issues being studied have life and death consequences, there is an urgent need for high quality studies that are carried out by epidemiologists with no ties to psychiatry or the pharmaceutical companies.

I now provide a description and critique of all of the studies provided by my respectful criticizer.

First Study

The first study that appears on his list was published in The Lancet in 2018, and titled “Second-generation Antipsychotic Drugs and Short-term Mortality: A Systematic Review and Meta-analysis of Placebo-controlled Randomized Controlled Trials.” It utilized a meta-analysis approach that looked at 596 studies, all of which had patients that were randomly assigned to either an “antipsychotic” or a placebo for 13 weeks or less. Here’s how its authors summarized their findings:

Overall, and for the main indication of schizophrenia, there is no evidence from randomised trials that antipsychotic drugs increase mortality. However, vulnerable populations (particularly patients with dementia) might be at increased risk. This meta-analysis could only address acute treatment effects leading to death in the short-term, and not long-term effects of antipsychotic drugs on mortality.

Second and Third Study

These two studies were carried out by the same team of authors and applied a similar methodology using a data base from Finland. The first of these is titled, “Effectiveness of Antipsychotic Treatments in a Nationwide Cohort of Patients in Community Care After First Hospitalisation Due to Schizophrenia and Schizoaffective Disorder: Observational Follow-up Study,” and was published in 2006 in the British Medical Journal. It looked at  a nationwide cohort of 2230 consecutive adults hospitalized for the first time because of schizophrenia or schizoaffective disorder, from January 1995 to December 2001. All of the patients had been hospitalized at various points in time, and during their hospitalization, no record was kept to see if while they were there they had been placed on antipsychotics and/or other drugs. Those patients who chose not to fill their prescriptions whenever they left the hospital were said to be in the no antipsychotic group. However, it is very possible that many of the patients in this group had repeatedly gone into a hospital for various periods of time and while there were placed on an “antipsychotic,” or a cocktail of “antipsychotics” and other drugs, and each time they were discharged they went through drug induced withdrawal reactions.

As we will see, there is a number of other serious problems with this study, but let’s get to the authors of this study’s conclusion:

Patients who currently took any antipsychotic drug had decreased mortality compared with the no treatment group. However, not using antipsychotic drugs may be a marker of other conditions that affect the risk of mortality.

Moving on to the other Finland study by this team of authors, it was published in 2009 in Lancet and titled “11-year follow-up of mortality in patients with schizophrenia: a population-based cohort study (FIN11 study).” As I mentioned, it is very similar to the previous Finland study, but extends the follow-up period an extra few years. Both studies have very similar results. Those who are labeled as not using any “antipsychotics” but probably did use them when they were in the hospital, had relatively higher rates of death than most of the other comparison groups. However, the patients that had the lowest rate of deaths were those in the group labeled as using “antipsychotics” for less than 6 months. In the main body of the journal article the authors note this in a single sentence, stating,”Patients who used antipsychotics for less than 6 months had especially low mortality rates.” There is a graph that illustrates this. It does repeat this again in another single sentence in the “Discussion” section of the report. Nothing else in the write-up of the report bothers to discuss this, and it can easily get lost in the numerous paragraphs defending the evidence that those on the drugs for longer periods had a lower mortality than those who were in the “no antipsychotic group” even though all of those patients in that group had probably been on such drugs each time they were hospitalized, and were only not taking them each time they were released from the hospital.

Note that the lead author of both of these articles is Dr. Jari Tiihonen. In the conflict of interest section of these studies it indicates he has served as a consultant, adviser, or speaker for AstraZeneca, Bristol-Myers Squibb, Eli Lilly, GlaxoSmithKline, Hoffman-La Roche, Janssen-Cilag, Lundbeck, Novartis, Organon, Otsuka, and Pfizer, which are all major pharmaceutical companies.

Dr. Tiihonen, along with the other authors of the article, did mention that although this study lends some support for the theory that, “antipsychotics” may lead to a reduced rate of mortality over an 11-year period, “a longer time might be needed for some adverse events to become apparent.” I applaud them for recognizing this.

Now, I can certainly understand why my respectful critic included these two articles on his list because it appears to offer some support for his position. That said, I am puzzled why he did not include the powerful critique of them that appeared after the 2009 study. It is titled, “Do antipsychotic medications reduce or increase mortality in schizophrenia? A critical appraisal of the FIN-11 study,” and appears in the peer reviewed journal Schizophrenia Research. Although it says in the title that it is critiquing the 2009 FIN-11 study, in the body of the article it notes that the same problems exist in the 2006 article. What are those problems? Here’s a summary that appears in the critique:

A number of methodological and conceptual issues make the interpretation of these findings problematic, including incomplete reporting of data, questionable selection of drug groups and comparisons, important unmeasured risk factors, inadequate control for potentially confounding variables, exclusion of deaths occurring during hospitalization leading to exclusion of 64% of deaths on current antipsychotics from the analysis, and survivorship bias due to strong and systematic differences in illness duration across the treatment groups….

It is likely that many of these patients died from the effects of suicide or cardiovascular disease while admitted to a hospital. As this approach likely underestimates mortality in patients treated with antipsychotic medications, this critical methodological factor might help to explain why the previous literature arrived at different findings with regard to antipsychotic-related mortality risk (Osborn et al., 2007; Weinmann et al., 2009).

Arguably, the biggest problem with the two studies that used the Finland data is this: 90% of the patient’s at the end of this study were still living. Thus, we still don’t know what would have happened if we could have looked at what age all of these patients ended up dying.

Finally, given the life and death importance of this study, I can’t help wondering why the country of Finland did not hire epidemiologists with no ties to psychiatry or the pharmaceutical companies to carry it out.

Fourth Study

The next study on my critic’s list is titled “Mortality and Cumulative Exposure to Antipsychotics, Antidepressants, and Benzodiazepines in Patients With Schizophrenia: An Observational Follow-Up Study” and was published in a 2016 edition of the American Journal of Psychiatry. The lead author is, once again, Dr. Jari Tiihonen, the same researcher of the last two studies that we looked at and who has numerous pharmaceutical industry conflicts of interest.

Using a very similar methodology as the Finland studies, it found that over a period of 5 years the prescription of moderate or high-dose “antipsychotics” for people labeled in Sweden as having schizophrenia was associated with lower overall mortality, compared to those said to be in a “no antipsychotic” group.” For those prescribed a benzodiazepine along with an “antipsychotic,” the results were quite different in that this group had the highest likelihood of dying when compared to the other groups. Thirty-eight percent of the patients were prescribed a benzodiazepine. Common benzodiazepines are Valium (diazepam)  and Xanax (alprazolam).

The problems with the study in concluding that these drugs protect people from dying are similar to the others carried out with Dr. Tiihonen. There were no randomization to groups. “Antipsychotic” drugs that may be used in hospitals were not recorded in the Prescribed Drug Register used to estimate drug use. Many of the patients in the group said to be not using “antipsychotics” very likely used them each time they went into the hospital, and each time they left the hospital they stopped using them and thus went through an awful withdrawal reaction.

There is also a concern about whether or not those patients included in groups said to be taking “antipsychotics” actually ingested them. Taking “antipsychotics” was assumed because someone, either the patient, or a family member, picked up the prescription at a pharmacy. In Sweden there is no financial cost for patients who pick up these drugs. There may have been some incentives to pick up the drugs, but once they were picked up, the patients may have tossed them in the garbage, or simply forgot to take them.

Also, as in the other two studies with Dr. Tiihonen as lead author, over 90% of the patient’s at the end of this study were still living. Recall that in the 2014 study that was not on my respectful critic’s list of references, over the course of 17 years schizophrenic labeled patients taking “antipsychotics” had died at a higher rate than those not taking “antipsychotics.” This provides some tentative support that it might take longer than 5 years, or even 11 years, before it becomes evident that patients that are on these drugs are more likely to die at an earlier age than those who don’t take them.

Given that this study was carried out after Dr. Tiihonen had done an earlier study that found patients on an “antipsychotic” for less than 6 months had the lowest likelihood of dying, it is of deep concern to me that he did not insist that his team do a similar analysis with this Sweden set of data. Could the reason be due to his conflicts of interest with the pharmaceutical industry? The loss of revenue for pharmaceutical companies that would occur if it became standard practice to not prescribe “antipsychotics” for longer than 6 months, rather than the current policy of prescribing it for an entire lifetime, would be enormous.

Fifth Study

This one was published in 2011, appears in the American Journal of Psychiatry, and is titled, “A Nationwide Cohort Study of Oral and Depot Antipsychotics After First Hospitalization for Schizophrenia.” The lead author, once again, is Dr. Tiihonen, and it looks at the same Finland data base that we discussed under the heading, “Second Study” and “Third Study.”

The major focus of this article was on comparing two groups of patients that took an antipsychotic either orally, or in the form known as “depot,” which is an injection given in a carrier liquid that releases it slowly so it lasts a lot longer. The authors looked at a period of approximately two years after initial discharge in the hospital. Many of the patients went in and out of the hospital at various times during this period, so those who are said in the study to be in the group that did not take “antipsychotics” were probably on it before they were initially discharged, and probably on them again during each of their rehospitalizations. As the authors duly note, “No information was available on the medications used in the hospital, but it can be assumed that virtually all patients had received some kind of antipsychotic treatment.” Nevertheless the authors report that the “no use antipsychotic group” had a higher risk of dying.

The problems with this finding are largely the same ones that we discussed regarding the other articles that Dr. Tiihonen served as lead author.

Sixth Study

This 2010 study, titled “Antipsychotic Polypharmacy and Risk of Death From Natural Causes In Patients with Schizophrenia: A Population-based Nested Case-Control Study” was published in the Journal of Clinical Psychiatry. It was conducted using data from Denmark. I’m not sure why my critic included this study on his list because it does not address the issue at hand. All of the subjects were on “antipsychotics” and the only relevant finding was that those who were on “antipsychotics” and benzodiazepines were at an increased risk of dying.

Seventh Study

This 2013 study, titled “Comorbidities and Mortality in Persons With Schizophrenia: A Swedish National Cohort Study,” was published in the American Journal of Psychiatry. It is pretty much the same type of study as the one’s carried out with Dr. Tiihonen and uses the same Swedish group of patients over a very similar time period as one of his studies, but here a different team of researchers looked at that data set. Again, lack of antipsychotic treatment was associated with elevated mortality. However, as hard as I looked in the article to see how the researchers defined “lack of antipsychotic treatment” I was not able to find it. As in other similar studies, most of those in the “lack of antipsychotic treatment group” were still alive, so we don’t know if they would, as a group, have begun to die at an earlier age if all were followed up until all in their group died.

Again my frustration is aroused because this study did not look to see if these patients from Sweden who were on “antipsychotics” for only 6 months or less were among those that were least likely to die, as was found by those who did look at this in the Finland data that we discussed above.

This study does add several additional relevant information than those we have so far discussed. First, in this large national cohort study, the leading causes of death were ischemic heart disease and cancer. However, schizophrenia patients had no increased risk of having a diagnosis of treatable nonfatal ischemic heart disease or cancer but had a far greater risk of dying from these conditions, suggesting substantial underdiagnosis and/or undertreatment.

This presents the theory that the group of individuals who did not take “antipsychotics” may have had a higher percentage of members who tend to avoid all medical treatment than the other comparison groups. With this theory, we have less of a reason to conclude that “antipsychotics” provide some protection from dying from these conditions; rather it is this special characteristic of some in the “lack of antipsychotic” treatment group that is the real cause for the statistical increased rate of death for those in this group during the relatively short window of time that these researchers were able to peek into.

This Swedish data is also useful for highlighting that these non-randomly assigned to groups studies can find some remarkably different results when looking at the same set of questions. Thus, in the Finland study, the lowest risk of death was associated with the “antipsychotic” drug clozapine; whereas in the Sweden study the authors reported, “we did not confirm that study’s finding of decreased mortality with clozapine, which was associated with a nonsignificant, modestly greater mortality in our study.” Moreover, unlike the Finland study that had found a statistically significant increased risk of suicide for those in the “no antipsychotic drug group,” in this Sweden study there was no significant increased risk in suicide for patients in their “lack of antipsychotic treatment group.”

Eighth Study

This 2017 study, titled “Reduced All-Caused Mortality With Antipsychotics and Antidepressants Compared to Increased All-Cause Mortality with Benzodiazepines in Patients With Schizophrenia observed in Naturalistic Treatment Settings,” was published in Evidence-Based Mental Health. This study examined the same Sweden data set as our “Fourth Study,” but utilized a different set of researchers. The outcomes were largely the same. Five years were looked at. Compared to the “no antipsychotic” group, the prescription of “antipsychotics” for people labeled in Sweden as having schizophrenia was associated with lower overall mortality, unless they were also prescribed a benzodiazepine, in which case they had the highest likelihood of dying. As in the earlier Sweden study that I discussed, inferences from this study’s findings are limited due to lack of randomization to the various comparison groups which prevent differentiation of causal from non-causal associations. This study actually does not really compare patients who took no “antipsychotics” with those who did, because all of the patients took “antipsychotics” at various points in their lives. Despite statistical adjustment for some relevant baseline variables, highly relevant variables were unaccounted for, including direct measures of schizophrenia severity, unhealthy lifestyle behaviors, degree of substance use, medical comorbitities and treatment, health service use pattern, and treatment adherence. And, of course, the study just looked at what could be observed in a five year period at which point the vast majority of the patients studied were still alive. Such studies don’t reveal what the average age of death is for those who are labeled as having schizophrenia and are taking “antipsychotics” versus those who were also labeled as having schizophrenia but did not take “antipsychotics.” Like other studies that were done after the 2009 study, this one did not bother to find out if patients that were on “antipsychotics” for less than six months was the group least likely to die during the followup period.

Ninth Study

This 2017 study, titled “Mortality and Antipsychotic Drug Use in Elderly Patients With Parkinson’s Disease in Nursing Homes,” was published in the Journal of the American Medical Directors Association. It looked at some data over a period of 18 months and does not directly address the issue under discussion. That is, these patients were elderly and had Parkinson’s Disease, rather than being labeled as having schizophrenia. For this group, prescribed antipsychotics for an 18 month period was not associated with the risk of death.

Tenth Study 

This 2015 study, titled “Antipsychotic Treatment and Mortality in Schizophrenia,” was published in Schizophrenia Bulletin. It looked at the same Sweden data set as our “Fourth Study” and “Eighth Study.” Each had looked at the follow-up that began on January 1, 2006 and ended on December 31, 2010.

The study did do some statistical analyses that I was not able to identify in the earlier similar studies. For example, it found that death due to respiratory disease was highest for those in the high-dose antipsychotic group, whereas those in the no exposure group had the lowest mortality due to respiratory disease.

Also, one of its statistical analyses indicated that no medication use was associated with elevated mortality only for patients who had been hospitalized within 1 year prior to start of follow-up (ie, during year 2005), but not for patients who had never received inpatient treatment.  Patients who were hospitalized were probably placed on “antipsychotics” during their stay. Those who were not hospitalized were far less likely to have ever been placed on these types of drugs at any point. Thus, this statistical finding suggests that the group of patients who really had not been exposed to these types of drugs may have been no more likely to die during this 5 year period than those who had prescriptions for “antipsychotics” and picked them up, or had a family member that pick them up at a pharmacy.

High antipsychotic use was associated with higher mortality than medium use. Mortality risk increased more in women than in men with high antipsychotic exposure. Finally, Jari Tiihonen, who has a great many connections with the pharmaceutical industry and was involved in most of the other studies that suggest a possible reduction in mortality with the use of “antipsychotics,” was, once again, one of the authors of this study.

Conclusion

The available research is not so perfect that we can make any definite conclusions regarding whether or not “antipsychotics” either increase or decrease mortality. That said, when we look at all of the research studies on the list provided by my respectful critic, along with the research left off of his list, my impression, certainly fallible, is as follows: For people experiencing the challenging concerns that result in people getting labeled as having schizophrenia, it would be healthier, and lead to longer lives, if we took all of the cost of promoting, manufacturing, and prescribing these drugs, and use it instead to provide safe, clean, kind, supportive places for them to get help.

antipsychotic drugsantipsychotics and mortality