Background: Idiopathic frozen shoulder is a self-limiting regional skeletal problem of unknown etiology. Clinically, patients first experience a phase of pain, progressing to a freezing stage when glenohumeral motion is lost, followed by a thawing phase when pain gradually subsides and most of the lost motion returns.

Objectives: To identify possible specific and non-specific risk factors for idiopathic frozen shoulder.

Methods: We compared the medical histories, drug treatment, previous hospital as well as health management organization blood tests of 126 new consecutive frozen shoulder patients from a shoulder clinic to those of an age-matched control group of 98 consecutive patients from an orthopedic foot and ankle clinic and to the regional population disease prevalence registry. Frozen shoulder was classified as idiopathic only if there was no history of trauma and no evidence of a rotator cuff tear.

Results: Among the frozen shoulder patients 29.4% had diabetes and 13.5% had thyroid disorders. The risk ratio for diabetes in the frozen shoulder group was 5.9 for males (95% confidence interval 4.1–8.4, P < 0.001) and 5.0 for females (95% CI[1] 3.3–7.5, P < 0.001). The risk ratio for thyroid disorders among females with frozen shoulder was 7.3 (95% CI 4.8–11.1, P = 0.001). No significant difference was found in the prevalence of thyroid disorders between frozen shoulder and the control group, but there was a significantly higher prevalence of diabetes in males and a trend for higher prevalence in females in the frozen shoulder group.

Conclusions: Physicians should be aware that diabetes is a specific risk factor for idiopathic frozen shoulder in both males and females and thyroid disorders are a non-specific risk factor in females only.