A three year, prospective, longitudinal study of components of effective fertility regulation is proposed. A total of approximately 2,100 women attending family planning clinic will be enrolled as they accept or re-accept either oral contraceptives or the diaphragm. The study population will comprise all age groups (1,050 women aged 19 or less and 1,050 women aged 20 or older) and three ethnic groups (Black, Hispanic and White). Participants will be followed at four month intervals for two years regardless of whether or not they continue to use the method they accepted on enrollment. Periods of non-use of any method and method changes will be documented as will any pregnancies as they occur. The decision to carry pregnancies to term or to undergo abortion will be examined as will return to contraception if it occurs within the two year period of follow-up. Follow-up contacts will be de-coupled from scheduled clinic re-visits to minimize loss to follow-up resulting from discontinuation or change of clinic attendance. The over-all objective of the study will be to determine which factors, combinations of factors and changes in factors correlate with correct, continuous and effective contraceptive practice. Oral contraceptives and the diaphragm were chosen for primary comparison because of their relative popularity in our clinics, their innate differences and the importance of attempting to explain the gap between theoretical effectiveness and use-effectiveness of both methods. The focus will be on the user's perspective and follow-up will track the individual as she uses, fails to use or changes contraceptive methods. Factors to be examined will include classical socio-demographic variables, social situational variables, psychological and cognitive factors, method attributes and service provider characteristics as correlates behavior, and the occurrence of unplanned pregnancy. An attempt will be made to structure hypothesized relationships among variables according to the model proposed by Luker (1975) to explain contraceptive risk taking behavior. Alternatives to this model will also be explored and tested. A randomly chosen sub-sample of 300 participants, proportionately distributed among initial method, age and ethnic sub-groups, will be """"""""set aside."""""""" Half will be interviewed at yearly intervals and half only at the end of two years to estimate the magnitude of any """"""""treatment bias"""""""" introduced by the follow-up process. The long term goal of this research is to enhance the ability of family planning service providers to improve the quality of service, to assist more effectively in the choice of contraceptive method, and to aid individuals and couples to achieve maximal use-effectiveness with their chosen method of fertility regulation.
