Patterns of contraceptive use among clients of the Barbados Family Planning Association.

This is a study of the behavior patterns of lower class Barbadian women during the fecund period of their lives with respect to the introduction of new types of contraceptives within a multiple method family planning program. The behavior patterns of these women are examined through their use of the facilities of The Barbados Family Planning Association (BFPA). In 1955, the BFPA was set up to disseminate information on family planning and to provide the material means for family limitation. From its inception, a variety of contraceptive methods have been made available to women utilizing the services. The study is developed within a diffusion theoretic framework that focuses on the trial and adoption stages of the diffusion process. Data are drawn for this thesis from 2 samples: 1) a systematic sample of records kept by the BFPA on its clients, and 2) a systematic stratified sample chosen from patients of the Population Council Research Project (PCRP). (The PCRP was an affiliate of the BFPA between 1965 and 1967.) Patients were interviewed at both the family planning clinics and in their homes. The BFPA has followed the policy of making available at nominal cost all existing contraceptive methods. This has led to very high dropout rates after admission and a considerable amount of method changing. However, changes take place generally in the direction of increasing the reliability of the method used. The lower the reliability of the method, the greater the rate of dropout as well as the rate of change from it. The longer the time spent in the program, the greater the number of methods used; but the longer the time a method is used, the less likely that it will be given up for another method. The transition probabilities for changing from user to nonuser in a year do not remain constant over the 12 years for which the program has been in operation. This is also true for changing from nonuser to user. In the case of the users these probabilities remain constant for time spans of 4 consecutive years; decreasing from one time span to another. The transition probabilities for going from nonuser to user remain constant for time spans of 4 consecutive years over the first 8 years of operation of the program and for the last 3 years. The year 1964 was found to be deviant. However, the probability of going from nonuser to user increases from one time span to the other. Therefore, 3 time subsystems are delimited, each characterized by constant probabilities. Each subsystem represents a unique period in the development of the program.