Millions of Patient Records Link Tobacco Use to Higher Odds of Golfer’s Elbow Surgery

The paper notes that more than 90% of cases are unrelated to sport, with repetitive occupational tasks accounting for much of the problem. Golf, tennis, baseball and throwing activities are familiar examples of tendon strain, though the name can obscure how often the condition affects people doing physically demanding jobs.

Pain on the inside of your elbow can make gripping a tool, lifting a bag or doing your usual job uncomfortable, even if you have never picked up a golf club. A study published in Scientific Reports, titled "Exploring the link between comorbidities, tobacco usage, and surgical intervention in medial epicondylitis," examined how tobacco use and other health conditions relate to golfer's elbow and surgery.

Ernesto Joubran and colleagues reviewed insurance records covering approximately eight million patients, finding a small association between recorded tobacco use and a golfer's elbow diagnosis, plus a stronger association with surgery. The findings describe patterns in medical records and cannot establish that tobacco caused the condition or made an operation necessary.

Golfer's Elbow Beyond the Golf Course

Golfer's elbow, medically called medial epicondylitis, affects the tendon that connects several forearm muscles to the inside of the elbow. These muscles help move the wrist and grip objects, making repeated bending, twisting and forceful hand movements potential sources of strain. Repeated tiny injuries can change the tendon's collagen, the structural material that gives it strength, leading to thickening, scar tissue and pain that interferes with work and everyday activities.

The paper notes that more than 90% of cases are unrelated to sport, with repetitive occupational tasks accounting for much of the problem. Golf, tennis, baseball and throwing activities are familiar examples of tendon strain, though the name can obscure how often the condition affects people doing physically demanding jobs.

What millions of patient records revealed

The researchers examined Humana insurance claims stored in the PearlDiver database from 2007 to 2015, identifying 23,573 patients with golfer's elbow, approximately 0.29% of the database population. For the tobacco comparison, they matched 1,657,810 patients with a recorded tobacco use disorder to the same number without that diagnosis. Matching covered age, sex, alcohol use, diabetes, high cholesterol, high blood pressure and obesity, helping make the groups comparable on those recorded characteristics.

Among all patients with golfer's elbow, the largest age group was 60–69 years, accounting for 25.64% of diagnoses. People aged 50–59 represented 23.08%, and those aged 40–49 represented 17.26%; together, ages 40–69 accounted for roughly two-thirds of cases. Men made up 52.46% of the diagnosed population.

Other health conditions were common: high blood pressure appeared in 58.66% of patients, high cholesterol in 56.98%, diabetes in 26.22%, tobacco use disorder in 23.69%, obesity in 16.57% and alcohol abuse in 2.93%. These figures describe conditions recorded alongside golfer's elbow, without showing that each contributed to its development.

Nearly 89.21% received their initial diagnosis in a physician's office, followed by outpatient hospitals at 5.40% and urgent care facilities at 1.26%. Inpatient hospitals and rural clinics accounted for very small shares, and the setting was unknown for 3.88%.

Tobacco's stronger association was with surgery

In the matched groups, golfer's elbow was recorded in 5,584 tobacco users, or 0.34%, compared with 5,300 people without recorded tobacco use disorder, or 0.32%. The odds ratio was 1.05, representing approximately 5% higher odds of diagnosis.

The surgery comparison produced a larger difference: 158 patients in the tobacco group underwent an operation associated with golfer's elbow, compared with 98 in the comparison group. The reported odds ratio of 1.62 means 62% higher odds across the full matched populations, with a 95% confidence interval of 1.25–2.08.

That number does not mean every tobacco user already diagnosed with golfer's elbow had a 62% greater chance of needing surgery. Among patients with the diagnosis in each matched group, the observed surgery proportions were 2.83% for tobacco users and 1.85% for the comparison group.

Across the entire golfer's elbow population, 525 patients underwent surgery, representing 2.23%. People aged 50–59 accounted for the largest share of operations, at 36.57%, followed by ages 40–49 at 28.57% and ages 60–69 at 17.90%.

Recorded tobacco use was more common among surgical patients than among all diagnosed patients, at 30.10% versus 23.69%. High blood pressure appeared less often, at 50.48% versus 58.66%; differences involving diabetes, high cholesterol, obesity and alcohol abuse did not reach statistical significance. These comparisons were descriptive because the surgical patients were included in the larger group used for comparison, and the researchers did not adjust for multiple differences.

What the findings can tell readers

Annual recorded golfer's elbow diagnoses increased 103% between 2007 and 2015. Coded elbow tendon-release and tissue-removal procedures increased 348%, with both counts peaking in 2014. Those procedure codes also cover other forms of epicondylitis, so the increase cannot be treated as a golfer's elbow surgery trend alone. The small decline in 2015 may partly reflect changes in diagnostic coding and missed records.

Tobacco codes did not reveal the product used, amount, duration or whether someone had stopped, and undocumented use could place exposed patients in the comparison group. Occupation, sporting activity and socioeconomic circumstances were not accounted for.

The records could not reliably distinguish which elbow was affected, confirm earlier treatments had failed, or measure recovery after surgery. Coding errors, possible overlap with tennis elbow procedures, relatively few operations in the matched groups and reliance on one insurer's older records limit how widely the findings apply.