1114 — Ascertainment of Testosterone Prescribing Practices in the VA

Jasuja GK, CHOIR, Bedford VAMC; Bhasin S, Brigham and Women's Hospital; Reisman JI, CHOIR, Bedford VAMC; Berlowitz DR, CHOIR, Bedford VAMC; Rose AJ, CHOIR, Bedford VAMC;

Objectives:
Prescribing of exogenous testosterone is increasing in the Veterans Health Administration (VHA). Since testosterone therapy is associated with known risks, it is important to follow certain evidence-based procedures when initiating therapy. We evaluated whether prescription of testosterone therapy was preceded by appropriate diagnostic evaluation of androgen deficiency and ascertainment of contraindications for testosterone therapy according to established guidelines.

Methods:
We examined a national cohort of 112,263 men who received their first testosterone prescription within VHA during FY 09-12. As signs and symptoms are difficult to ascertain from an electronic medical record (EMR), we considered a patient to have undergone potentially appropriate testing if he had two or more low (total: < 300 ng/dL or free: < 70 pg/mL) testosterone levels. We examined testosterone levels for one year prior to the index date (date of first testosterone prescription) for each patient. We used VA-CMS data to examine care received outside VHA.

Results:
New testosterone prescriptions in VHA increased by 78% over 4 years, from 20,581 in FY09 to 36,544 in FY12. Only 3% of men who received a testosterone prescription had two or more low AM testosterone levels, LH and/or FSH measured, and no contraindications to testosterone therapy. 16.8% of men did not have their testosterone level checked at all. Among those prescribed therapy, 1.4% had prostate cancer, 7.5% had obstructive sleep apnea, and 3.5% had elevated hematocrit ( > 50%) at baseline, all absolute or relative contraindications to therapy. 40% of patients with fee-for-service Medicare who had a documented ICD-9 code for male hypogonadism, but did not have any testosterone testing within VHA, had undergone Medicare-paid testing.

Implications:
Only a small proportion of men receiving testosterone therapy in VHA underwent appropriate diagnostic workups, and some received testosterone in spite of important contraindications.

Impacts:
These findings potentially suggest a quality of care issue that needs to be addressed. Promoting a more uniform application of clinical guidelines for prescribing testosterone may facilitate appropriate use of this therapy. Investigating the factors that contribute to suboptimal prescribing would be an important preparatory step before designing interventions to optimize testosterone use in VHA.