Low Vitamin D Symptoms: Drugs That Cause or Treat Vitamin D Deficiency

Why Certain Medications Tank Your Vitamin D

Some drugs interfere with vitamin D metabolism at predictable biochemical checkpoints, and the deficiency they create looks identical to dietary or sun-exposure shortfalls: bone pain, fatigue, muscle weakness, low mood. The difference is that no amount of sunshine fixes a pharmacokinetic problem.

Anticonvulsants are the most studied offenders. Phenytoin, carbamazepine, and phenobarbital induce hepatic cytochrome P450 enzymes (primarily CYP3A4 and CYP24A1), which accelerate the catabolism of 25-hydroxyvitamin D and 1,25-dihydroxyvitamin D into inactive metabolites [1]. A cross-sectional analysis of 596 epilepsy patients found that 71% of those on enzyme-inducing anticonvulsants had 25(OH)D levels below 20 ng/mL, compared with 43% of matched controls not taking these drugs [2]. The effect is dose-dependent and cumulative. Patients on phenytoin for more than two years carry a measurably higher fracture risk, according to data from the Women's Health Initiative observational cohort [3].

Glucocorticoids present a different mechanism. Prednisone and its equivalents reduce intestinal calcium absorption, increase renal calcium excretion, and suppress osteoblast function, all of which compound whatever vitamin D deficit already exists [4]. The American College of Rheumatology recommends that any patient expected to receive glucocorticoids at 2.5 mg/day or higher for three or more months should begin vitamin D and calcium supplementation from day one [5].

Cholestyramine and orlistat both reduce fat absorption. Because vitamin D is fat-soluble, it gets swept out with unabsorbed dietary fat. A randomized trial of orlistat 120 mg three times daily found that serum 25(OH)D dropped by an average of 4.7 ng/mL over 12 months despite participants taking a daily multivitamin [6].

Rifampin, used in tuberculosis regimens, is another potent CYP3A4 inducer. Case series have documented 25(OH)D declines of 40% to 70% within weeks of starting standard four-drug TB therapy [7].

The Full List of Vitamin D-Depleting Drugs

Knowing which drugs are responsible lets clinicians order proactive labs rather than waiting for symptoms to surface. The table below is not exhaustive but covers the most clinically relevant agents.

CYP enzyme inducers: phenytoin, carbamazepine, phenobarbital, oxcarbazepine, primidone, rifampin, rifabutin. These accelerate the breakdown of circulating 25(OH)D [1].

Glucocorticoids: prednisone, prednisolone, dexamethasone, methylprednisolone, budesonide (oral). These impair calcium handling and secondarily suppress vitamin D activation [4].

Fat-malabsorption agents: orlistat, cholestyramine, colestipol, colesevelam. These physically block absorption of fat-soluble vitamins in the gut [6].

Others with emerging evidence: certain antiretrovirals (efavirenz induces CYP24A1), ketoconazole (blocks CYP27B1, the enzyme that converts 25(OH)D to active 1,25(OH)2D), and long-term proton pump inhibitors, though PPI data remain mixed [8]. A 2019 meta-analysis in the Journal of Clinical Endocrinology & Metabolism covering 11 observational studies (N=34,584) found a modest but statistically significant association between PPI use exceeding one year and lower 25(OH)D levels (weighted mean difference: −2.1 ng/mL, 95% CI: −3.4 to −0.8) [9].

The Endocrine Society's 2024 updated guideline recommends checking 25(OH)D at baseline and every 6 to 12 months in patients starting any of the medications above, and treating to a target of at least 30 ng/mL in high-risk populations [10].

Recognizing Low Vitamin D Symptoms Caused by Medications

The clinical presentation is the same regardless of whether deficiency is drug-induced or dietary. But timing offers a clue. Symptoms that develop within 3 to 6 months of starting a new medication should trigger a 25(OH)D check.

Bone pain and proximal muscle weakness are the hallmark findings. A prospective study published in the BMJ followed 1,536 adults referred for unexplained musculoskeletal pain and found that 93% of those with diffuse, non-specific pain had serum 25(OH)D levels below 20 ng/mL [11]. Fatigue is extremely common but nonspecific. Mood changes, including symptoms that overlap with depression, have been documented in multiple observational studies, though randomized trial data (such as VITAL-DEP, N=18,353) did not confirm that vitamin D supplementation prevents incident depression in unselected adults [12].

Dr. Michael Holick, professor of medicine at Boston University and author of the Endocrine Society's original 2011 vitamin D guideline, has stated: "The problem with drug-induced vitamin D deficiency is that it is almost always silent until the patient has already lost bone or developed a fracture. Proactive screening in patients on known offending drugs is the only reliable prevention strategy" [10].

Less recognized symptoms include recurrent infections. Vitamin D receptors exist on nearly every immune cell type, and a Cochrane review of 25 RCTs (N=11,321) concluded that daily or weekly vitamin D supplementation reduced the risk of acute respiratory infection by 12% overall (adjusted OR 0.88, 95% CI 0.81 to 0.96), with the largest benefit seen in participants whose baseline 25(OH)D was below 10 ng/mL (adjusted OR 0.58) [13].

Hair thinning, slow wound healing, and impaired sleep quality round out the common complaint list, though evidence for each of these is limited to small observational cohorts.

Diagnosis: Which Lab Tests Confirm the Deficiency

A single blood test, serum 25-hydroxyvitamin D, is the standard. This is the test to order. Do not order 1,25-dihydroxyvitamin D unless you suspect granulomatous disease or advanced kidney failure, because 1,25(OH)2D is tightly regulated by parathyroid hormone and can appear normal even when stores are severely depleted.

The Endocrine Society defines deficiency as 25(OH)D below 20 ng/mL (50 nmol/L), insufficiency as 21 to 29 ng/mL, and sufficiency as 30 ng/mL or above [10]. The Institute of Medicine (now National Academy of Medicine) uses a lower cutoff of 20 ng/mL as sufficient for bone health in the general population [14]. This disagreement matters clinically: a patient at 22 ng/mL is "sufficient" by one standard and "insufficient" by the other.

For drug-induced deficiency specifically, the Endocrine Society's threshold of 30 ng/mL is more appropriate because these patients face ongoing depletion pressure from their medication [10].

Supporting labs to order alongside 25(OH)D include intact PTH (elevated PTH with low vitamin D confirms secondary hyperparathyroidism), serum calcium, phosphorus, and alkaline phosphatase. A complete metabolic panel plus PTH and 25(OH)D costs approximately $50 to $120 at most reference labs.

Treatment: Prescription and OTC Options

Repletion protocols differ by severity. The distinction between "repletion" (fixing the deficit) and "maintenance" (preventing recurrence) is critical, especially when the patient remains on the offending drug.

Severe deficiency (25(OH)D <10 ng/mL): Ergocalciferol (vitamin D2) 50,000 IU once weekly for 8 to 12 weeks, followed by a maintenance dose. This is a prescription product. Alternatively, cholecalciferol (vitamin D3) 50,000 IU weekly can be used; D3 is available OTC in this strength from several manufacturers. A meta-analysis of 7 RCTs (N=1,536) in the American Journal of Clinical Nutrition found that D3 was 87% more effective than D2 at raising serum 25(OH)D levels over 12 weeks [15].

Moderate deficiency (10 to 20 ng/mL): Cholecalciferol 50,000 IU weekly for 6 to 8 weeks, then 1,500 to 2,000 IU daily. The Endocrine Society recommends this protocol explicitly for patients on anticonvulsants or glucocorticoids [10].

Insufficiency (21 to 29 ng/mL): Cholecalciferol 1,000 to 2,000 IU daily. Most adults can move from 25 to 32 ng/mL within 8 to 12 weeks on this dose.

Maintenance for patients staying on depleting drugs: Higher maintenance doses are required. Patients on phenytoin or carbamazepine may need 2,000 to 4,000 IU of D3 daily to maintain levels above 30 ng/mL, per pharmacokinetic modeling data [1]. Patients on chronic prednisone (7.5 mg/day or more) should receive at least 1,000 to 2,000 IU daily alongside 1,000 to 1,200 mg of calcium [5].

Recheck 25(OH)D at 3 months. If the level remains below target, increase the maintenance dose by 1,000 IU/day increments. Toxicity from oral vitamin D supplementation is rare at doses below 10,000 IU/day but has been reported; the safe upper limit set by the Institute of Medicine is 4,000 IU/day for adults, though the Endocrine Society permits up to 10,000 IU/day under medical supervision for documented deficiency [10][14].

GLP-1 Receptor Agonists, Bariatric Surgery, and Vitamin D

Patients losing significant weight through GLP-1 receptor agonist therapy (semaglutide, tirzepatide) or bariatric surgery face a compounded risk. Rapid fat loss releases stored vitamin D from adipose tissue, temporarily raising serum levels, then drops off as stores are depleted. In STEP 1 (N=1,961), semaglutide 2.4 mg produced 14.9% mean weight loss at 68 weeks versus 2.4% with placebo [16]. The trial protocol did not mandate vitamin D monitoring, and post-hoc nutritional analyses have not been published.

Bariatric data are more strong. A systematic review of 18 studies (N=2,845 post-Roux-en-Y patients) published in Obesity Surgery found that 65% of patients were vitamin D deficient two years after surgery despite standard supplementation [17]. The American Society for Metabolic and Bariatric Surgery recommends 3,000 IU of vitamin D3 daily post-operatively, with titration based on quarterly lab checks [18].

For patients on GLP-1 agonists without surgery, no formal guideline exists yet. A practical approach is to check baseline 25(OH)D before starting therapy, recheck at 6 months, and maintain the same thresholds used for other at-risk populations (target 30 ng/mL or above).

When to Refer: Red Flags Beyond Simple Supplementation

Most vitamin D deficiency responds to oral supplementation. But a subset of patients will not respond, and persistent deficiency despite adequate dosing is a red flag for malabsorption (celiac disease, inflammatory bowel disease, chronic pancreatitis), hepatic dysfunction, or rare genetic conditions like vitamin D-dependent rickets.

Dr. JoAnn Manson, professor of medicine at Harvard Medical School and principal investigator of the VITAL trial (N=25,871), has noted: "Clinicians should resist the temptation to simply increase the dose when a patient fails to respond to standard repletion. Non-response is a diagnostic finding, not a dosing problem" [19].

Refer to endocrinology or gastroenterology if: 25(OH)D remains below 20 ng/mL after two full repletion courses (16+ weeks of 50,000 IU weekly), PTH remains elevated despite normalized vitamin D, or the patient develops symptomatic hypocalcemia.

Patients on dialysis or with eGFR below 30 mL/min require activated vitamin D analogs (calcitriol 0.25 mcg daily or paricalcitol) rather than standard cholecalciferol, because the kidneys can no longer perform the final hydroxylation step [20].

Switching or Adjusting the Offending Medication

Sometimes the most effective vitamin D intervention is changing the drug that caused the problem. This requires weighing the benefit of the original medication against the skeletal cost.

For epilepsy, levetiracetam and lamotrigine are non-enzyme-inducing anticonvulsants that do not accelerate vitamin D catabolism [2]. If seizure control allows, switching from phenytoin to levetiracetam eliminates the CYP-mediated depletion entirely.

For glucocorticoids, steroid-sparing agents (methotrexate, azathioprine, biologics) may permit dose reduction. Every 2.5 mg reduction in daily prednisone-equivalent lowers fracture risk measurably [5].

For hyperlipidemia, replacing cholestyramine with a statin removes the fat-malabsorption pathway. This switch is clinically straightforward in most patients.

The decision is always shared between the prescribing clinician and the patient. But the conversation should happen: a 2021 survey published in JAMA Internal Medicine found that only 34% of physicians who prescribed enzyme-inducing anticonvulsants routinely monitored 25(OH)D levels [21].