Under The Microscope
Systematic Reviews on the Effectiveness of Natural Family Planning Methods
Basing clinical decisions on best possible evidence and high-quality research findings is an important topic in medicine and health care. This is because much of health care practice lacks a strong evidence base and many health care providers fail to integrate the best and latest research into practice. There have been a number of movements in health care and medicine that have been developed to promote the use of the best research findings in clinical practice. One of the best and most useful has been the development of systematic (also called integrative) research reviews (Mulrow, 1994; Higgins and Green, 2011).
Systematic research reviews are summaries of past research in which the reviewer (or reviewers) extract, analyze and integrate findings from peer reviewed scientific studies and produce conclusions that can aid practice. Systematic reviews are based on a wide and exhaustive search of existing studies. These reviews have prescribed criteria, steps and methods in how to report the findings. The general steps include selecting a topic or issue that is important to practice, conducting an exhaustive search of the existing evidence, establishing criteria for selecting the best evidence, making a table of the studies in the review, analyzing the evidence, and providing a summary and conclusion.
The research studies selected for the reviews are also ranked according to level of evidence – i.e., from the most convincing to the least credible studies or evidence. The evidence is then provided a grade from A – F, e.g., strong scientific evidence = A, good scientific evidence = B, mixed or conflicting evidence = C, strong negative evidence = D, and lack of evidence = F.
An example of a simple evidence hierarchy is as follows:
Levels of evidence (from strongest to weakest):
• Systematic review of controlled comparison studies
• Randomized controlled studies
• Non-randomized comparison studies
• Quasi-experimental studies
• Non-experimental studies
• Program evaluations, quality improvement projects, case reports
• Opinions of respected experts and authoritative professional committees
Another movement in health care and medicine is the call for comparative effectiveness research (CER) that involves providing evidence for treatments, methods or approaches for managing major health concerns and conditions. The purpose of CER is to identify what works best for which patients under what circumstances and is patient focused. The Institute of Medicine (IOM) has delineated the 100 most significant areas that need better research and comparative evidence for treating or managing health problems (IOM, 2009). In the top 25 list of the IOM’s CER is the effort to compare the effectiveness of innovative strategies for preventing unintended pregnancies. Which method of NFP works best for which couples (for achieving or avoiding pregnancy) and under what circumstances is of interest to those health providers who provide or refer for a moral (and integrative) approach to family planning.
Natural Family Planning and Evidenced Based Reviews
Two review articles on NFP methods were recently published that used some form of assessing, grading and ranking research evidence for the effectiveness of major NFP methods provided in the United States (US). The first review study reviewed observational trials and provided a list of the data sources from which the studies were obtained. The authors ranked the Creighton Model (a mucus only based method) as the most effective and the Standard-Days Method (SDM), a fixed day calendar method, as the least effective. The authors graded the evidence for perfect use of six major NFP methods as a “B” and the evidence for the Creighton Model and the Sympto-Thermal Method as being the most effective NFP methods as “B.” Table 1 displays perfect and typical use pregnancy rates for major U.S. NFP methods as developed by the authors of this review.
The second review article was developed by a group of individuals that represented several modern fertility awareness based methods (FABM) of family planning (i.e., Creighton Model, Sympto-Thermal, Standard-Days Method, and the Two-Day simplified mucus method). This review used a Strength of Recommendation Taxonomy (SORT) that was developed by experts in NFP and based on the recommendations of Lamprecht and Trussell (1997). The SORT criteria were developed for prospective cohort effectiveness studies, since the authors felt, that for pragmatic reasons, FABM methods do not lend themselves to randomized controlled trials. The SORT criteria were further categorized or ranked as being a) critical criteria, b) important criteria, and c) useful criteria. Critical criteria could receive a potential score from 0-4, important criteria a score from 0-2, and useful criteria 0-1. The maximum score that a study or method could receive is 56.
Table 1: Effectiveness of Modern Natural Family Planning Methods
Natural Family Planning Method Typical Use Perfect Use
Standard-Days Method 8-25 5
Billings Ovulation Method 3-22 3
TwoDay Method 14 4
Marquette Model 11-14 1-2
Sympto-Thermal Method 2-8 < 1 Creighton Model 2-5 < 1 The Critical Criteria are as follows: – Prospective study design – Involved only sexually active women participants – Large enough sample size to address the effectiveness question – Efficacy calculations began with the start of method use – Involved at least a one year follow-up – Statistical analysis involved either life table or survival analysis – All pregnancies were included – Intention to avoid (or achieve) pregnancy was determined prospectively (and ideally at the beginning of each menstrual cycle) – Typical use analysis included all pregnancies and all cycles/months of use – Correct use analysis included only those cycles in which the method was used consistently and correctly – The study was reviewed for human rights compliance by an institutional review board The Important Criteria are: – Users/participants obtained from multiple centers and multiple countries – Included a diverse population – Recorded all sexual behavior (e.g., intercourse, abstinence, withdrawal, and use of barriers) Useful Criteria are: – Participant demographic profiles were presented and motivation was measured – Coital frequency was monitored The authors were able to obtain 29 peer-reviewed studies that were published since 1980. It is not clear what criteria was used to reject or include the studies, but the authors indicate that one robust “level one” cohort study was included for all of the modern FABM. The two highest scored studies (with a score of 56) were the efficacy studies of the Standard-Days Method and the TwoDay simplified mucus method developed through the Institute of Reproductive Health at Georgetown University. Four European sympto-thermal studies received a score of 55, one Marquette Model study received a score of 54, two Billings Ovulation Method studies received a score of 52, and two Creighton Model studies received a score of 43. The authors mentioned that it was difficult to determine the effectiveness of the Creighton Model studies due to the unique method of interpreting unintended pregnancies. For example, with the Creighton Model of NFP, if a couple knows they are in the estimated fertile phase and have intercourse it is classified as an achieving related pregnancy, even if the intention was to avoid pregnancy. The standard method of classifying an unintended pregnancy is based on the couple’s intention. Therefore, most efficacy studies of NFP methods classify a couple who take a chance and have intercourse during the estimated fertile phase as an unintended pregnancy. The authors point out that the reviewed studies involved 8200 women participants with over 107,000 cycles of use. Based on the 29 studies and NFP methods reviewed, the overall correct-use effectiveness was from 0.4-5.0 per 100 women years. Except for the European Sympto-Thermal studies (which had a typical use rate of 1-2 unintended pregnancies per 100 women years) the overall typical use ranged from 10-14 unintended pregnancies per 100 women years. Comments The authors of both of these review articles on NFP effectiveness studies should be commended for their efforts as there are few systematic reviews on NFP other than the Cochrane review of older (1980s) randomized controlled trials (RCT) of NFP methods (Grimes, Gallo, Grigorieva, Nanda, and Schulz, 2010). It is also commendable that they provided some ranking and grading of the research studies included in their reviews. However, even with the use of their stated criteria for reviews there is some bias in the interpretation, as evidenced by the high ranking of the Creighton Model NFP in one study and the low ranking in the other. Future systematic reviews of NFP effectiveness studies might be best conducted by researchers and health professionals that are not tied to one of the NFP methods. As a NFP researcher and developer, I obviously have a bias toward the Marquette Method, and given that fact I do have some concerns with the two reviews. First of all, I disagree with the reviewers of the Manhart et al. (2013) study that categorizes NFP as a sub-set of FABM. Natural Family Planning methods were developed before FABM. FABM accept the use of barriers during the estimated fertile time and hence are not considered a moral or true natural method of family planning. I understand the reason for using the term FABM as being more acceptable for those who have no problem with the use of contraception. But FABM came out of the long development and history of NFP methods not the other way around. I also disagree that RCTs or CER studies of NFP methods are not practical or cannot be conducted. In fact they are needed so that health professionals can best decide which NFP methods work best for which couples and in what circumstances. I found it remarkable that the cohort comparison of the Marquette Method with the Creighton Model study was not even mentioned (Fehring, Schneider, Barron, and Raviele, 2009). Furthermore, RCTs of NFP studies, although very difficult to conduct, are possible and desired – as testified by Dr. Thomas Bouchard in this current issue of Current Medical Research. Other issues of concern are that many of the studies reviewed were not conducted in the United States (U.S.) and the findings might not relate to women and couples in the U.S. Many of the U.S. studies primarily include only middle to upper class white educated Catholic participants, i.e., they lack diversity. Furthermore, most of the studies reviewed included women with only regular length menstrual cycles. At a given time, many women in a population will have irregular length menstrual cycles due to being postpartum, breastfeeding, peri-menopausal, or have irregular menstrual cycles as a result of health problems. Finally, some of the best prospective efficacy studies were not included in the reviews like the World Health Organization (1981) five country study of the ovulation method of NFP. Sources 1. Fehring, R., Schneider, M., Barron, M.L., and Raviele, K. 2009. Cohort comparison of two fertility awareness methods of family planning. Journal of Reproductive Medicine 54: 165-170. 2. Grimes, D. A., M. F. Gallo, V. Grigorieva, K. Nanda, and K. F. Schulz. 2010. Fertility awareness-based methods for contraception. Cochrane database of systematic reviews (Online), (4), CD004860. Digital object identifier number: 10.1002/14651858.CD004860.pub2. 3. Higgins. J. P. T., and S. Green (Editors). Cochrane Handbook for Systematic Reviews of Interventions. Version 5.1.0 [updated March 2011]. Available at: http://188.8.131.52/openlearning/html/mod0.htm. 4. Institute of Medicine (IOM). Initial National Priorities for Comparative Effectiveness Research. Released, June 30, 2009. Available at: http://www.iom.edu/Reports/2009/ComparativeEffectivenessResearchPriorities.aspx. 5. Lamprecht, V., and J. Trussell. 1997. Effectiveness studies on natural methods of natural family planning. Advances in Contraception 13:155-165. 6. Manhart, M. D., M. Duane, A. L. Lind, I. Sinai, and J. Golden-Tevald. 2013. Fertility awareness-based methods of family planning: A review of effectiveness for avoiding pregnancy using SORT. Osteopathic Family Physician 5: 2-8. 7. Mulrow, C. D. 1994. Rationale for systematic reviews. British Medical Journal 309: 597-599. 8. Smole, B. A., and C. M. Robinson. 2012. Natural family planning. American Family Physician 86: 924-928. 9. World Health Organization. 1981. A prospective multicentre trial of the ovulation method of natural family planning. II. The effectiveness phase. Fertility and Sterility 36: 591-598