Unintended Weight Changes: Drugs That Cause or Treat Them

Why Medications Cause Unintended Weight Changes

Drugs alter body weight through several distinct pathways. Some increase appetite via histamine H1 receptor blockade or serotonin 5-HT2C antagonism. Others shift glucose and lipid metabolism, promote fluid retention, or reduce resting energy expenditure. A 2022 meta-analysis in Obesity Reviews found that medication-induced weight gain accounts for an estimated 10 to 15% of obesity cases in adults seeking treatment at specialty clinics [1].

The direction and magnitude of weight change depend on the drug class, the dose, the duration of treatment, and individual genetic factors. For example, the melanocortin-4 receptor (MC4R) gene variant rs17782313 has been associated with a 1.5- to 2-fold increased risk of antipsychotic-induced weight gain [2]. This means two patients on the same medication at the same dose can have wildly different outcomes. Recognizing that a medication is the root cause of an unexplained shift on the scale is often the single most productive step a clinician can take, because the fix may be as simple as switching to a weight-neutral alternative within the same drug class.

Drugs That Commonly Cause Weight Gain

The list of weight-promoting medications is long. Three classes stand out for the size and speed of their effect: atypical antipsychotics, corticosteroids, and insulin/insulin secretagogues.

Atypical antipsychotics. Olanzapine and clozapine sit at the top. A Cochrane review of 212 trials found olanzapine produced a mean weight gain of 4.2 kg over 10 weeks compared to placebo [3]. Clozapine is comparable. Quetiapine and risperidone cause moderate gains (2 to 3 kg), while aripiprazole and ziprasidone are considered relatively weight-neutral [4]. The American Psychiatric Association guidelines recommend baseline metabolic screening and quarterly weight checks for all patients started on second-generation antipsychotics [5].

Corticosteroids. Prednisone, dexamethasone, and other glucocorticoids increase appetite, promote visceral fat deposition, and cause sodium and water retention. A retrospective cohort study of 93,000 patients in the UK found that oral corticosteroid users had a 1.36 relative risk of obesity compared to non-users (95% CI 1.30 to 1.42) [6]. Weight gain becomes clinically significant at doses above 5 mg/day of prednisone equivalent taken for more than three months.

Insulin and sulfonylureas. Insulin therapy produces average weight gains of 2 to 4 kg in the first year of treatment. The UKPDS trial reported that patients randomized to intensive insulin gained a mean of 4.0 kg more than those on conventional therapy over 10 years [7]. Sulfonylureas like glipizide and glyburide cause similar though smaller gains (1.5 to 2.5 kg), driven by increased insulin secretion and the caloric "rescue" of glucose that would otherwise be excreted.

Antidepressants. Not all antidepressants are equal here. Paroxetine, mirtazapine, and amitriptyline are the most likely to cause weight gain. A large network meta-analysis published in The Lancet in 2024, covering 385 trials and 116,477 participants, confirmed that mirtazapine and amitriptyline were associated with the greatest weight increases among commonly prescribed antidepressants [8]. Bupropion, by contrast, is the only antidepressant consistently associated with modest weight loss.

Other common offenders. Gabapentin (mean gain 2.2 kg over 12 weeks), pregabalin, valproic acid (8 to 13 kg in one year of use at anticonvulsant doses), pioglitazone (3 to 4 kg from fluid retention and adipogenesis), and certain beta-blockers like atenolol round out the list. Metoprolol succinate appears less prone to weight gain than older agents [9].

Drugs That Commonly Cause Weight Loss

Weight loss as a medication side effect ranges from the mildly inconvenient to the medically dangerous, depending on the patient's starting weight and nutritional status.

GLP-1 receptor agonists. Semaglutide and tirzepatide are the most potent pharmacologic weight-loss agents available. The STEP-1 trial (N=1,961) showed that semaglutide 2.4 mg weekly produced a 14.9% mean body weight reduction at 68 weeks versus 2.4% with placebo [10]. The SURMOUNT-1 trial (N=2,539) demonstrated that tirzepatide at its highest dose (15 mg weekly) achieved a 22.5% mean weight reduction at 72 weeks [11]. These drugs are now FDA-approved specifically for chronic weight management, but they can also cause unintended excessive weight loss in patients prescribed them for type 2 diabetes at lower doses.

Metformin. Weight-neutral to slightly weight-reducing. The Diabetes Prevention Program (DPP) trial found metformin produced a mean weight loss of 2.1 kg at 2.8 years compared to placebo [12]. It works partly by suppressing hepatic glucose output and partly by reducing appetite through GLP-1-related mechanisms.

Topiramate. Originally an anticonvulsant, topiramate causes dose-dependent appetite suppression. At 200 mg/day, mean weight loss ranges from 4 to 8% of body weight over 6 to 12 months [13]. The mechanism involves carbonic anhydrase inhibition and modulation of GABA and glutamate signaling. It is a component of the combination weight-loss drug Qsymia (phentermine/topiramate).

Stimulants and ADHD medications. Amphetamine-based drugs (Adderall, Vyvanse) and methylphenidate suppress appetite centrally. Lisdexamfetamine (Vyvanse) is FDA-approved for binge eating disorder in part because of its anorectic properties. Children and adolescents on stimulant therapy should have growth velocity monitored every 3 to 6 months [14].

Chemotherapy and targeted cancer therapies. Many oncologic agents cause weight loss through nausea, mucositis, altered taste, or direct catabolic effects. Cachexia, defined as involuntary weight loss exceeding 5% of body weight in 6 months, affects 50 to 80% of patients with advanced cancer and is an independent predictor of mortality [15].

Thyroid hormones. Excess levothyroxine (iatrogenic hyperthyroidism) increases basal metabolic rate and can produce unintended weight loss. TSH suppression below 0.1 mIU/L is associated with measurable decreases in fat mass and, in older adults, accelerated bone loss [16].

How Clinicians Evaluate Drug-Related Weight Changes

The assessment follows a structured approach. First, the clinician maps the temporal relationship between medication initiation (or dose change) and the onset of weight change. A gain of 5% or more within 3 months of starting a new drug is considered clinically significant by most endocrinology guidelines. Second, other causes must be excluded: thyroid dysfunction, Cushing syndrome, adrenal insufficiency, undiagnosed diabetes, malignancy, malabsorption, and psychiatric conditions including depression and eating disorders all require screening.

Lab work typically includes a complete metabolic panel, TSH, fasting glucose, hemoglobin A1c, and a lipid panel. If the clinical picture suggests Cushing syndrome, a 24-hour urinary free cortisol or overnight dexamethasone suppression test is warranted. A body composition assessment using dual-energy X-ray absorptiometry (DXA) can distinguish fat mass changes from lean mass or fluid shifts, which is especially relevant when evaluating corticosteroid-related gains.

The Endocrine Society's 2023 clinical practice guideline on obesity pharmacotherapy recommends that clinicians perform a comprehensive medication review for every patient presenting with unexplained weight gain, because switching to a weight-neutral alternative within the same drug class often resolves the problem without sacrificing therapeutic efficacy [17].

Switching Strategies: Weight-Neutral Alternatives

When a medication is identified as the likely cause, the evidence often supports a lateral switch.

For antipsychotics, moving from olanzapine to aripiprazole has been shown to produce a mean weight loss of 2 to 3 kg over 16 weeks without worsening psychotic symptoms in stable patients, according to a randomized trial published in the American Journal of Psychiatry [18]. For antidepressants, switching from paroxetine to sertraline or bupropion reduces weight gain liability while maintaining comparable efficacy for major depressive disorder. The choice depends on the patient's symptom profile. Bupropion may be preferred when appetite suppression is desirable. Sertraline suits patients who need serotonergic activity without the weight penalty of paroxetine.

For diabetes medications, the shift from sulfonylureas or insulin to SGLT2 inhibitors (empagliflozin, dapagliflozin) or GLP-1 agonists offers both glycemic control and weight reduction. The SUSTAIN-7 trial demonstrated that semaglutide 1.0 mg produced 6.5 kg mean weight loss over 40 weeks in patients with type 2 diabetes, compared to a 3.0 kg loss with dulaglutide 1.5 mg [19].

For epilepsy, replacing valproic acid with lamotrigine or zonisamide eliminates one of the most potent weight-gain-promoting anticonvulsants. Zonisamide itself produces modest weight loss (3 to 4 kg at therapeutic doses) [20].

Gabapentin and pregabalin for neuropathic pain can sometimes be replaced with duloxetine (an SNRI with weight-neutral to mildly weight-reducing properties) or topical therapies like lidocaine patches.

When Unintended Weight Loss Requires Urgent Evaluation

Not all unintended weight loss is benign. The BMJ Best Practice guidelines define clinically significant involuntary weight loss as a reduction of 5% or more over 6 to 12 months [21]. In adults over 65, this threshold drops: even 3 to 4% unintentional loss predicts increased mortality.

Red-flag features that demand rapid workup include: weight loss exceeding 10% of baseline in under 6 months, new dysphagia, persistent abdominal pain, night sweats, lymphadenopathy, and unexplained anemia. A prospective Dutch cohort of 2,677 patients presenting with involuntary weight loss found that malignancy was the cause in 24%, gastrointestinal disorders in 14%, psychiatric illness in 10%, and no identifiable cause in 26% after full evaluation [22].

Drug-induced weight loss warrants the same evaluation. Patients on GLP-1 agonists who lose more than 15% of body weight should have lean mass monitored, given emerging data that 25 to 40% of weight lost on semaglutide may be lean mass rather than fat [23]. Resistance training and adequate protein intake (1.0 to 1.2 g/kg/day) are standard recommendations to preserve muscle during pharmacologic weight loss.

Pharmacologic Treatment of Unintended Weight Gain

When switching medications is not feasible (as in clozapine for treatment-resistant schizophrenia), adjunctive pharmacotherapy can offset drug-induced weight gain. The strongest evidence supports metformin as an adjunct to antipsychotic therapy: a meta-analysis of 12 randomized controlled trials found metformin reduced body weight by a mean of 3.2 kg compared to placebo in patients on atypical antipsychotics [24].

Topiramate as an adjunct has also shown benefit in antipsychotic-treated patients, with mean weight reductions of 2 to 5 kg, though cognitive side effects limit its use. GLP-1 agonists are being studied in this population. A 2024 pilot trial of semaglutide in patients with schizophrenia on clozapine or olanzapine reported a mean weight loss of 7.2% at 24 weeks, with no worsening of psychiatric symptoms [25].

For corticosteroid-induced weight gain, no specific pharmacotherapy is approved. Caloric counseling, sodium restriction, and exercise remain the standard interventions. Minimizing the steroid dose and transitioning to steroid-sparing immunosuppressants (azathioprine, mycophenolate) as quickly as the underlying disease allows is the primary strategy.

Monitoring and Long-Term Management

Any patient on a medication known to affect weight should have body weight recorded at every clinic visit. The American Diabetes Association's 2024 Standards of Care recommend measuring weight, waist circumference, and BMI at baseline and at least every 3 months for patients on insulin, sulfonylureas, or thiazolidinediones [26].

For patients on GLP-1 agonists for chronic weight management, monitoring should include periodic DXA scans or bioimpedance analysis to track lean mass, along with standard metabolic panels. The Endocrine Society recommends reassessing the risk-benefit ratio of any weight-altering medication at least annually [17].

Patients who experience more than 5% unintended weight change in either direction while on stable medication doses should receive a full metabolic and oncologic screening before the change is attributed solely to the drug. A medication may unmask a concurrent illness, and assuming causation without investigation is a common clinical error.

Scheduled quarterly weight checks, a documented medication timeline, and a low threshold for metabolic labs form the minimum standard of care for any patient on drugs with known weight effects.