Oral Corticosteroids and First Do No Harm

July 12, 2017

Even short-term use is associated with an increased risk of adverse events, this study shows.

One of every 5 privately insured American adults younger than 64 years received a prescription for short-term use of oral corticosteroids over a 3-year period. The short-term prescribing of oral corticosteroids in this group was associated with higher rates of sepsis, venous thromboembolism, and bone fractures.

Long-term systemic corticosteroid use is known to lead to adverse effects, but Akbar Waljee and colleagues at the University of Michigan pointed out that studies examining complications associated with short-term steroid use are lacking.

Corticosteroids are a common reason for admission to a hospital for drug-associated adverse events, the researchers noted. Extended use of oral corticosteroids is associated with infections, deep vein thrombosis, avascular necrosis, and bone fractures and may lead to chronic disease states, such as diabetes, hypertension, osteoporosis, and drug-induced Cushing syndrome.

The authors presented the results of their investigation into the consequences of short-term oral corticosteroid treatment in a recent British Medical Journal article.

The study

In this large retrospective, population-based, cohort study, the investigators assigned 1,548,945 adults younger than 64 years to 2 groups, those who received at least 1 outpatient prescription for short-term oral corticosteroids and those who received no oral steroids. They were assessed for 3 acute adverse events: sepsis, venous thromboembolism, and fractures.

The results

• 21% of subjects received at least 1 outpatient prescription for short-term oral corticosteroids during the study period.

• The median number of days the oral steroids were used was 6; 47.4% of users took them for more than 7 days.

• The median prednisone equivalent daily dose was 20 mg/d; 23.4% of subjects took ≥ 40 mg/d.

• The methylprednisolone 6-day “dosepak” was the most frequently prescribed corticosteroid, accounting for 46.9% of prescriptions.

• In the steroid use group, 70.5% had a single course of treatment, 20.7% received 2 courses, and 8.8% were prescribed 3 or more courses of oral steroids.

• The subjects most often prescribed short-duration oral steroids were older white women with multiple comorbidities (P<0.001).

• The most common indications for short-term oral steroid prescriptions were upper respiratory infections, spinal conditions, and intervertebral disc disorders.

• Family practitioners and general internal medicine physicians wrote the bulk of steroid prescriptions, followed by emergency, otolaryngology, and orthopedics doctors.

• The risks of sepsis, venous thromboembolism, and bone fracture were statistically higher in the steroid user group when compared with nonusers.

• The risk of a complication developing in the 5- to 90-day period after visiting the clinic for treatment in the short-term steroid use group was as follows: hospital admission for sepsis, 0.05% vs 0.02% in nonusers; development of venous thromboembolism, 0.14% vs 0.09%; and the risk of bone fracture, 0.51% vs 0.39%.

• The short-term risk of bone fractures was higher in Caucasian patients than in other races (P=0.006).

Implications for physicians

• Oral corticosteroids are often prescribed for short-term use by several medical specialties.

• Short-term use of oral corticosteroids increases the risk of sepsis, venous thromboembolism, and bone fractures.

• The increase in acute adverse events in short-term users of oral corticosteroids was 2 to 5 times higher than in nonusers.

• The increase in risk was most prominent in the first 30 days after starting steroid treatment and was consistent across doses; the risk decreased after 30 days from initial exposure.


The United States Veterans Affairs, the Michigan Institute for Data Science, Research to Prevent Blindness and The WK Kellogg Foundation supplied funding for the authors.


Akbar K Waljee, Mary A M Rogers, Paul Lin, et al. “Short term use of oral corticosteroids and related harms among adults in the United States: population based cohort study.” BMJ. 2017;357:j1415. doi: 10.1136/bmj.j1415.