Yes. Antipsychotic medicines can raise blood sugar and body weight, and a large new UK study pinpoints exactly when the risk climbs fastest: the first 90 days after starting treatment. In nearly 97,000 adults, average blood sugar (HbA1c) rose about 14 times faster in the first three months on a second-generation antipsychotic than in the months before treatment began, while weight rose about nine times faster. This does not prove the medicine caused diabetes in any individual person, but it does identify the first few months as the critical window to check both weight and blood sugar.
The study behind the headline
Published on 14 October 2026 in The Lancet Psychiatry and funded by the UK National Institute for Health and Care Research (NIHR), the research was led by Dr Juan Carlos Bazo-Alvarez at UCL Epidemiology and Health. The team reviewed anonymous NHS GP health records from nearly 97,000 adults in England, aged 18 to 59, who had severe mental illness and started taking a second-generation antipsychotic between 2000 and 2023. The four medicines covered were olanzapine, quetiapine, risperidone and aripiprazole. For each person, the records from two years before starting treatment were compared with the two years after, with the focus on weight and blood test results. It is one of the largest studies of its kind, and its size is what makes the early timing signal hard to ignore.
What changed in the first 90 days
The standout finding is the speed of change right after treatment began. In the first 90 days, HbA1c, which reflects average blood sugar over several weeks, rose about 14 times faster than it had before treatment started. Weight rose about nine times faster in the same period. The biggest metabolic changes clustered in this early window, which is why the researchers describe the first three months as critical for monitoring.
There is another important detail. Blood sugar and weight changes did not always happen together. Some people developed rising blood sugar without dramatic weight gain, which means that checking the bathroom scales alone can miss a metabolic problem developing underneath. The scale and the blood test are telling two different stories, and both need to be heard.
Olanzapine showed the largest early increases
Of the four medicines studied, olanzapine showed the largest early increases. For people taking olanzapine, blood sugar rose about 50 times faster and weight about 10 times faster in the first 90 days. That is a striking figure, but it needs careful interpretation. This is what happened in the records; it is not a league table to make switching decisions from. Each medicine has its own balance of benefits and side effects for each person, and any change of medication must be decided with the prescribing doctor, never made alone.
It is also worth remembering that weight is shaped by many influences beyond any one medicine. Diet, activity, sleep and hormones all play a part, and nutrient status can be one overlooked factor: our explainer on whether low vitamin D can cause weight gain walks through that evidence in plain language. The study’s message is about monitoring, not about blaming any single cause.
Prediabetes and women faced the greatest risk
One figure in this study concentrates the mind. Among people who had prediabetes when they started treatment, about 19% had developed type 2 diabetes within two years, compared with just 3% of those whose blood sugar was normal at the start. That is a gap large enough to change practice: it makes the case for knowing your blood sugar status before the first prescription is written, because prevention works best when the risk is spotted early.
Women were disproportionately affected by the metabolic changes in the study. The researchers do not fully explain why, but the pattern is consistent enough that early monitoring matters for everyone starting treatment, and especially for women.
None of this changes the underlying reality. Second-generation antipsychotics are important, life-changing treatments for schizophrenia and bipolar disorder. Schizophrenia affects around 23 million people worldwide, and about 37 million people live with bipolar disorder, according to UCL’s release on the study. The link between these medicines, weight gain and diabetes risk has been established for years. What is genuinely new here is the timing: the danger concentrates in the first 90 days, far earlier than most monitoring schedules assume.
What the researchers say should happen now
The study’s practical message is refreshingly clear. Healthcare professionals should check both weight and blood sugar before treatment starts, and check both again during the first few months afterwards. Both, because the two do not always move together, and because catching a rise early means that lifestyle steps, medication reviews or diabetes prevention measures can begin while the change is still small.
If you or someone you support is starting an antipsychotic, that gives you two concrete questions to bring to the appointment: what are my weight and blood sugar readings today, and when will we check them again? Writing the answers down matters, because a reading that drifts upward over three months is much easier to act on than one discovered by accident a year later.
Looking after general health helps too. Sleep quality, for example, interacts closely with metabolic health, and sleep problems are common alongside mental health conditions. Our piece on whether low vitamin D can affect sleep explores one common and often fixable thread. Small, steady habits support the bigger picture while the clinical monitoring does its job.
What this study cannot tell us
This was observational research, and that matters for how we read it. The team studied what happened in real GP records, which is a strength because it reflects everyday care across two decades. But records cannot prove that a particular drug caused a particular person’s diabetes. Prescribing reasons, doses, how regularly medicines were taken, diet, activity levels and other health conditions could all have played a part in the patterns seen.
That does not make the findings dismissible. Large, careful studies of real records are exactly how medicine learns about long-term effects that clinical trials, which are shorter and smaller, cannot easily capture. It does mean the headline figures are signals to monitor for, not verdicts on any individual. The same careful reading applies to any study that links a medicine to an outcome, a habit we practised when decoding the research on whether statins prevent Alzheimer’s. Observational data tells us where to look and what to watch for; it cannot tell us what would have happened to one specific person on a different medicine.
The safety rule that matters most
The most important paragraph in this article is this one. Do not stop or change a prescribed antipsychotic because of diabetes worries without talking to the prescribing doctor. Stopping suddenly can trigger relapse or severe withdrawal effects, and untreated schizophrenia or bipolar disorder carries far greater risks than the metabolic effects described in this study. The goal of this research is better monitoring, not less treatment.
If the first-90-day numbers concern you, the safe move is to book a review and ask about weight and blood sugar checks, not to adjust the dose yourself. Concern is a good reason to ask more questions, never a good reason to skip a prescription.
Frequently asked questions
Do antipsychotics cause diabetes?
They can raise blood sugar and body weight, which increases the risk of developing type 2 diabetes. In the Lancet Psychiatry study of nearly 97,000 adults, HbA1c rose about 14 times faster in the first 90 days of treatment than before treatment began. Because this is observational data, it cannot prove the medicine caused diabetes in any individual person, but the pattern across so many records is strong enough to justify close monitoring.
Which antipsychotic carries the highest diabetes risk?
Of the four medicines studied, olanzapine showed the largest early increases: blood sugar rising about 50 times faster and weight about 10 times faster in the first 90 days. Quetiapine, risperidone and aripiprazole were also included in the study but showed smaller early changes. Treatment choices are individual and depend on far more than metabolic risk, so discuss any concerns with your prescribing doctor rather than comparing medicines on this one figure.
Why are the first 90 days after starting antipsychotics so important?
The largest metabolic changes in the study clustered in the first three months after treatment began. Blood sugar and weight both accelerated fastest in this window, which makes it the period when monitoring catches problems earliest and when prevention measures have the most effect. Checks later in treatment still matter, but the first 90 days are the window that most monitoring schedules have historically underemphasised.
What should be monitored when starting an antipsychotic?
The researchers recommend checking both weight and blood sugar before treatment starts, and checking both again during the first few months afterwards. Checking only weight is not enough, because in the study blood sugar and weight changes did not always occur together. Ask your doctor for the actual readings at each check so you can see the trend for yourself.
I have prediabetes and take an antipsychotic. What should I do?
Talk to your doctor about closer monitoring and diabetes prevention steps. In the study, about 19% of people with prediabetes developed type 2 diabetes within two years of starting antipsychotics, compared with 3% of those with normal blood sugar. Knowing your status early gives you the best chance of prevention. Do not stop or change your medication on your own; the safe first step is a review appointment.
Can I stop my antipsychotic if I am worried about diabetes?
No, not without medical advice. Stopping suddenly can cause serious harm, including relapse and withdrawal effects, and untreated schizophrenia or bipolar disorder carries far greater risks than the metabolic effects described in this research. If you are worried, book a review with your prescribing doctor and ask about weight and blood sugar monitoring. Concern is a reason to ask more questions, not to skip a prescription.
Related reading
- Will Low Vitamin D Cause Weight Gain?: one overlooked factor in weight changes, explained with the evidence.
- Can Low Vitamin D Affect Sleep?: sleep, metabolism and one common, fixable thread.
- Do Statins Prevent Alzheimer’s?: another medicines study decoded, with the same careful reading of observational evidence.
Sources
- UCL News (14 Oct 2026): First 90 days after starting antipsychotics are critical for monitoring diabetes risk
- Medical Xpress: First 90 days of antipsychotics critical for diabetes risk
- Drugs.com: Antipsychotics diabetes risk runs in the first few months, study warns
- News-Medical: Study: First 90 days of antipsychotic treatment critical for metabolic changes
- Original study: The Lancet Psychiatry, published 14 October 2026, led by Dr Juan Carlos Bazo-Alvarez, UCL Epidemiology and Health; funded by the UK National Institute for Health and Care Research (NIHR).
Medical disclaimer: This article is for information only and is not medical advice. It explains published research in plain language; it cannot diagnose, treat or advise on any condition. Always talk to your doctor, pharmacist or another qualified health professional about medication questions. If you take an antipsychotic, do not stop or change it without medical guidance.