Characterization of diagnosis and treatment patterns in the Clinical Practice Research Database (CPRD): a population study of men with hypogonadism with and without type 2 diabetes

Presentation Number: OR32-3
Date of Presentation: June 17th, 2013

Emily F Shortridge*1, Stephan Lanes2, Paula K Polzer1, Charles Wentworth2, Elisa A Razzoli1, Melissa Jeffers2, Dara Putthoff Schuster1 and Glenn Matfin3
1Eli Lilly and Company, Indianapolis, IN, 2United Biosource, Chevy Chase, MD, 3International Diabetes Center, Minneapolis, MN


Background: Hypogonadism (HG) is interrelated with diabetes (T2DM), though there is controversy about the benefits of testosterone replacement therapy (TRT) for patients with HG+T2DM. There is little information about diagnosis and treatment patterns of men with HG+T2DM, so we examined these conditions using the UK Clinical Practice Research Database (CPRD). 

Methods: A retrospective cohort study using the CPRD followed male patients with at least one year of enrollment from January 2000-December 2010. Baseline and follow-up descriptive statistics for the HG-only and HG+T2DM cohorts and for patients who initiated TRT in the study period were estimated.

Results: 20,509 men had newly diagnosed HG at baseline; of these, 7,198 HG-only men were <18. 656 (3.2%) men had HG+T2DM, and these men tended to be older than men with HG-only (64 versus 34 years). Only 11.6% of men with a diagnosis of HG had a recorded baseline total testosterone level.

4,488 men initiated TRT during the study. For treated men, the HG+T2DM cohort was older than the HG-only group (62 years versus 54 years). More newly treated men had testosterone levels recorded at baseline compared to the overall sample (46.3%, HG-only; 49.4%, HG+T2DM). 82.4%had a testosterone level <320ng/dL. Most did not have levels assessed after TRT initiation; 16.4% had an additional testosterone test within 3 months of TRT initiation, 23.0% within 6 months, and 29.4% within 1 year. There were no differences between the cohorts in the likelihood of follow-up tests. The most frequent initial TRT formulations were gels (35.4%, HG-only; 44.7%, HG+T2DM) and injections (35.6%, HG-only; 31.8%, HG+T2DM). Regardless of formulation, long-term adherence to TRT was poor; persistence at 6 months was about 41% for both groups, and at one year it was about 25%.  

Conclusions: This large sample of UK men in primary care provides insights into diagnosis and treatment patterns for men with HG with and without T2DM. Overall, we found limited biochemical validation of a HG diagnosis. Despite a reported higher incidence of HG in the setting of T2DM, and despite regular screening blood work for T2DM, men with HG+T2DM did not have a higher occurrence of baseline or follow-up testosterone testing. With the recent adoption of guidelines for treatment of HG in the UK, it will be vital to revisit screening and follow-up patterns.


Disclosure: EFS: Employee, Eli Lilly & Company. SL: Independent Contractor (including contracted research), Eli Lilly & Company. PKP: Employee, Eli Lilly & Company. CW: Independent Contractor (including contracted research), Eli Lilly & Company. EAR: Employee, Eli Lilly & Company. MJ: Independent Contractor (including contracted research), Eli Lilly & Company. DPS: Employee, Eli Lilly & Company. GM: Consultant, Eli Lilly & Company.