Objectives: To provide a comprehensive and objective summary of contraceptive failure rates for a variety of methods based on a systematic review of the literature.
Methods: Medline and Embase were searched using the Ovid interface from January 1990 to February 2008, as well as the reference lists of published articles, to identify studies reporting contraceptive efficacy as a Pearl Index or life-table estimate. Reports that recruited less than 400 subjects per study group and those covering less than six cycles/six months were excluded. In addition, unlicensed products or those not internationally available, emergency contraception, and vasectomy studies were excluded.
Results: Information was identified and extracted from 139 studies. One-year Pearl Indices reported for short-acting user-dependent hormonal methods were generally less than 2.5. Gross life-table rates for long-acting hormonal methods (implants and the levonorgestrel releasing-intrauterine system [LNG-IUS]) generally ranged between 0-0.6 per 100 at one year, but wider ranges (0.1-1.5 per 100) were observed for the copper intrauterine devices (0.1-1.4 per 100 for Cu-IUDs with surface area > or =300 mm( 2 ) and 0.6-1.5 per 100 for those with surface area <300 mm( 2 )). Barrier and natural methods were the least effective.
Conclusions: Our review broadly confirms the hierarchy of contraceptive effectiveness in descending order as: (1) female sterilisation, long-acting hormonal contraceptives (LNG-IUS and implants); (2) Cu-IUDs with > or =300 mm( 2 ) surface area; (3) Cu-IUDs with <300 mm( 2 ) surface area and short-acting hormonal contraceptives (injectables, oral contraceptives, the patch and vaginal ring), and (4) barrier methods and natural methods.