Where the scientific method fails

How often do different methods of contraception fail?

The authors evaluated 139 studies from January 1990 onwards. 47 dealt with combined oral contraceptives (COCs), one with a special POP (progesterone only pill), three with hormone patches, and three with the vaginal ring. Implants were examined 15 times, injectable contraceptives 16 times. 31 times it was about copper-containing spirals, nine times about those with levonorgestrel. Condoms for men were the subject three times, four times other barrier methods, eleven times natural methods and four times sterilization of women.

Pearl index with weaknesses

The Pearl Index is a simple and well-accepted way of showing the failure of contraceptive methods. One serious limitation is that failure rates decrease with prolonged use. The reasons are likely in part because unsuccessful users are "filtered out" and that use of the methods generally improves with practice.

The problem is avoided by means of “life table analyzes”. They provide cumulative failure rates, labeled either “large” or “net”, depending on whether tables with single or multiple causes are used for the dropout. Life table analyzes instead of Pearl indices were v. a. used in longer term procedures such as the use of IUDs.

Hormone IUD before copper IUD

The following order of efficiency resulted:

1. Sterilization of the woman and long-term hormonal contraception (coil with levo-norgestrel and implants),

2. copper-containing spirals with surfaces of 300 mm2 and above,

3. copper-containing spirals less than 300 mm2 Surface and short-acting hormonal contraceptives (injections, oral contraceptives, plasters and vaginal rings),

4. Barrier and natural methods.

The authors discuss, among other things, that the Pearl indices for combined oral contraceptives with 20 and 30 µg ethinylestradiol overlap considerably. In general, it is possible that the data does not reflect real life. Among other things, Studies like to be carried out only with women aged 18 and over. Discrepancies between studies and practice are likely to affect the methods that require a certain degree of compliance. Compliance issues could be partly responsible for known differences between study results from the USA compared to other regions.

Although no method is 100% safe, the protection offered by implants and a levonorgestrel IUD is at least comparable to the results of sterilization in the first year.

“Of course” often fails

The highest failure rates are with “natural methods”. They should not be recommended if an unplanned pregnancy can have deleterious effects. Furthermore, both highly effective and less reliable methods can be combined with the use of condoms to prevent sexually transmitted infections. If natural methods are used, they also offer additional protection during the period of highest risk of pregnancy. SN

Source:Mansour, D: Efficacy of contraceptive methods: a review of the literature, journal: EUROPEAN JOURNAL OF CONTRACEPTION AND REPRODUCTIVE HEALTH CARE, issue 15 (2010), pages: 4-16