Behavioral methods of family planning: A comparative study of efficacy and safety

Background and PurposePublished reports indicate that over 100 million women worldwide use the birth control pill (Oral Contraceptives, or OCPs).1,2 However, in a national study conducted by the Battelle Centers for Public Health Research and Evaluation, rates of discontinuation of some birth control methods were as high as 90%, and had averages between 40-61% (depending on marital status) for women who were followed over a two-year period.3 This suggests that women are searching for new methods, perhaps with fewer side effects or better effectiveness rates. Studies show up to 60% of women would be interested in using Fertility Awareness Based Methods (FABMs) if given information.4 Both FABMs and OCPs entail behavior modification on a daily basis, so the efficacy and side effects of FABMs and OCPs were chosen for comparison during this study.Most commonly reported unintended pregnancy rates for FABMs and OCPs are based on low quality retrospective surveys. A popularly cited review from J. R. Trussell and colleagues reports typical use failure rates of 24% for FABMs and 9% for OCPs.5 These numbers are based on estimates of the probability of pregnancy drawn from the 1995 and 2002 National Surveys of Family Growth.5'6 There are at least three major problems with these surveys. One is that these are lower quality retrospective surveys based primarily on patient recall, with the data collected via a series of phone surveys. Another major problem is that 86% of the purported FABM users surveyed identified the calendar method-a much older and less effective method-as their prima- ry form of contraception. Then, the rates for all types of FABMs were pooled together, including some methods which are not even classified as FABMs by trained providers. This lumping together of all FABMs including rhythm masks important differences in effectiveness among the variety of modern methods; a fact acknowledged by Trussell, the author of this estimate.7'8'9 The third major issue with this review is that the rates of effectiveness were adjusted to account for the researchers' assumption of underreporting of abortion. As a result of the issues with these popularly cited statistics, we sought to design objective criteria with which to evaluate the primary literature available for FABMs and OCPs, in order to understand both the quality of the studies available as well as the true effectiveness rates.MethodsIn total, three major search engines (Medline, PubMed, and EMBASE) were used with keyword searches to identify relevant articles. For FABMs, 30 articles from 1980 on were included in the study.6 For OCPs, 47 articles from 2000 on were included in the study. (The year 2000 was chosen in part because studies from earlier decades had used higher dose hormone formulations which were proven to be unsafe for women, and the trend in later years has been towards seeking lower dose hormonal formulations to provide the same efficacy with fewer side effects1). Strength of Recommendation Taxonomy (SORT) criteria10 were established first for the FABM review6 and then adapted for the OCP review (Adapted SORT). The adaptation was based on fundamental differences between FABMs and OCPs such as lack of a learning phase for OCPs. The purpose of the SORT and Adapted SORT criteria was to provide a framework to determine the level of evidence for individual studies.Two independent reviewers scored each of the identified articles for the FABMs and the OCPs reviewed. Major discrepancies in scores were resolved by sending the articles in question to a third reviewer for a tie-break score.ResultsFor the FABMs reviewed, there were 56 maximum possible points with the SORT criteria. The scores ranged from 31-55 points, and 10 of the 30 articles (30%) were found to earn a SORT Level 1 status (qualified by earning a positive score in all of the critical criteria).6For the OCPs reviewed, there were 42 maximum possible points with the Adapted SORT criteria. …