Abstract
This study aimed to determine the predictors of contraceptive use among married female youths and their husbands using the behavioral theory of the Health Belief Model (HBM). A community-based survey was conducted in a rural area of Myanmar in 2008. A total of 444 respondents (222 couples) were interviewed separately using a pretested, structured questionnaire. Significant predictors of contraceptive use were determined by univariate and multivariate analysis. Wife’s HBM perception was a highly significant predictor of contraceptive use in married youths (adjusted odds ratio = 10; 95% confidence interval = 2.7, 37.6). Wives aged 20 to 24 years and having their own income, experience of spousal communication, and shorter distance from home to health center were also significant predictors of contraceptive use. A poor agreement on HBM perception between wife and husband was noted. This study highlights the importance of HBM perceptions, wife’s income, spousal communication, and geographic barriers in contraceptive use among married youths in rural Myanmar.
Introduction
Half of the world’s population is aged below 25 years, of whom the majority are youths, especially in developing countries.1,2 A youth, according to World Health Organization (WHO), is defined as a person between the age of 15 and 24 years, further classified as late adolescent (15-19 years) and young adult (20-24 years). 3 Many countries in the Asia-Pacific region have shown a trend toward increasing age at marriage among both sexes; however, marriage during adolescence is not uncommon. 4 A review on teenage pregnancy reported that 25% of all pregnancies in Sub-Saharan Africa and Asia resulted in an induced abortion. 5 In developing countries, married adolescents have a higher exposure to risky sexual intercourse and unintended pregnancies because of unprotected intercourse and having sex more frequently, and they are less likely to protect themselves than unmarried adolescents. 6 Improving contraceptive use among married youths is an important reproductive health (RH) issue; as a result, predictors of contraceptive use among married youths needs to be addressed.
Most studies investigating associated factors of contraceptive use have been conducted among women of reproductive age and not specifically among youths. Factors include individual, husband, and service aspects. Individual factors identified in previous studies comprised sociodemographic factors, 7 knowledge8,9 and attitude, 10 as well as spousal communication on contraception.11-13 Accessibility and availability of contraception were found to be associated service factors in a number of studies.10,14,15
The Health Belief Model (HBM) is an important behavioral theory related to seeking behaviors. 16 It consists of 6 constructs—namely, perceived susceptibility, perceived severity, perceived barriers, perceived benefits, perceived efficacy, and cues to action. There have been several studies in health using HBM to predict seeking behavior; however, only some studies used HBM in relation to RH behaviors.17-21 These studies showed the influence of HBM constructs on different RH behaviors in various perspectives, and therefore, it will be useful to apply this theory to identify the relationship with contraceptive use.
There have been few studies investigating perceptions of both wife and husband on contraception in relation to HBM. Moreover, there is a lack of information on the factors influencing contraceptive use among married youths from the perspectives of both wife and husband, in addition to service factors. Therefore, this study aimed to determine the predictors of contraceptive use of married female youths and their husbands and to assess the agreement of perception on contraceptive use with respect to HBM between female youths and their husbands.
Methods
Study Setting
A community-based, cross-sectional survey was conducted during May to October 2008 in Ayeyarwaddy division of Myanmar where the maternal mortality rate is 300 per 100 000 live births, 22 the second highest figure among all states and divisions of Myanmar. Of 14 townships in Ayeyarwaddy division under RH projects, one township was randomly selected as the study setting using a computer-generated random number, which resulted in Hinthada township. The total population was 464 456 in 2005, of which 17.4% were aged between 15 and 24 years. A total of 70% of the population resides in rural areas, and RH services are the responsibility of 9 rural health centers (RHCs), where contraceptive pills, injections, condoms, and intrauterine devices are available. Nationally, the contraceptive prevalence rate was 32% in this age group, lower than that in neighboring countries.
Study Sample
Currently married females aged between 15 and 24 years and their husbands, who had resided in rural areas of Hinthada township for at least 6 months before the survey was conducted, were included in the study. Those with a desire to get pregnant or were currently pregnant, not willing to participate, and mentally retarded were excluded. According to a 20% difference of outcome interest on contraception, a ratio of contraceptive user to nonuser of 1:1, and a power of 80% to detect this difference, at least 98 couples who used and 98 couples who did not use contraception were required for the study.
Data Collection
Preparatory phase
A draft of a structured English questionnaire was first developed that included the topics on attitude and HBM perceptions related to contraceptive use. It was then validated on the issue of relevance, conciseness, and clarity by 2 experts in RH fields and research using the content validity index; the indices were 0.91, 0.91, and 0.85, respectively. After validation, it was translated into Myanmar language and pretested in a similar study setting on participants who were not recruited in the main study.
Attitude and HBM constructs were pretested to identify the most reliable items. Attitudes were first developed on 10 items using a 5-point Likert scale ranging from strongly disagree to strongly agree. Finally, 6 items of attitudes were kept resulting in a Cronbach’s α coefficient of .7. Of 6 constructs of HBM perceptions, each construct consisted of 4 items giving rise to a total of 24 items. After pretesting, all items of HBM perceptions were kept resulting in a Cronbach’s α of .84.
A 2-day workshop for training 3 interviewers was carried out in the Department of Medical Research, Lower Myanmar. The workshop aimed to make them understand the concepts of the study and the process of data collection, and to standardize the interview technique.
Data collection phase
Of all 9 targeted RHCs, 7 RHCs were visited, covering about 80% of the rural area of the township. Two areas were excluded because of transportation difficulty during the rainy season. Midwives in the RHCs assisted in obtaining a list of married youth living in the villages. At least 28 couples living within the area of each RHC from each village were selected and interviewed. The first household containing a married female youth living with her husband was randomly selected, and an appointment was made with the couple. At the appointment, the couples were invited to participate in the study. An information sheet was provided, and written informed consent was obtained from those agreeing to participate. The wife and husband were interviewed separately, face-to-face, using a validated and reliable structured questionnaire. The average time of each interview was approximately 20 minutes.
Measures
The main outcome measure was contraceptive use, defined as continued use of any modern contraceptive method for at least 6 months prior to the interview. Independent variables included individual characteristics, marital and obstetric characteristics, service factors on contraception, spouse communication on contraception, knowledge, attitude, and HBM perceptions. All variables were constructed in the questionnaires for both wives and their husbands.
Individual characteristics were age, education, and own income. Marital and obstetric characteristics were age at marriage, duration of marriage, number of children, and desired family size. Service factors were considered on the geographic accessibility to service in terms of distance from home to health center (km) and the availability of contraceptive methods in the living areas. Spousal communication was determined by asking whether they had ever discussed the use of contraception with their spouse (yes or no).
Knowledge of contraception emphasized appropriate prescription and common side effects, comprising 8 items. The responses were scored as “0” for an incorrect answer and “1” for a correct answer. The total score was then obtained by summing all items, which could range from 0 to 8 points. This variable was analyzed using mean or median as appropriate. High scores showed high knowledge on contraception.
In all, 6 items of attitude focused on contraceptive use, and these were scored according to the direction of the attitude statement. The responses of attitudes were scored from “1” to “5” regarding favoring contraceptive use. A total attitude score was calculated by summing each individual score, which ranged from 6 to 30. This variable was analyzed using mean or median as appropriate. High scores represented higher favorable attitudes toward contraceptive use.
The 6 constructs of HBM perceptions were perceived susceptibility to pregnancy, perceived severity of pregnancy, perceived benefits of contraception, perceived barriers to contraception, self-efficacy, and cues to action on contraceptive use. Susceptibility emphasized their perception of the likelihood of being pregnant or a tendency to be pregnant. The dangers and consequences of being pregnant were mentioned in the perceived severity of pregnancy. The benefit was described in terms of the benefits of contraception to prevent their pregnancy. The barriers to contraception were assessed based on the accessibility to and costs of contraceptives and method of contraception. Self-confidence in making a decision to use contraceptives was explored in self-efficacy. Finally, the motivation from health providers and peers and information from the media were considered in cues to action. All items of HBM perception were scored from “1” to “5,” and high scores represented high perception regarding the use of contraception. The subtotal scores of each construct (range 4 to 20) were summed to give a total score for all HBM items (range 24 to 120). The subtotal scores of each construct and total score were then categorized into a dichotomous scale as either “negative” or “positive” using the median score as the cutoff point.
Statistical Analysis
Data entry and validation were done using Epidata 3.1 software and analyzed by R software version 2.7.1 (the R Foundation for Statistical Computing 2008, Austria). The effect of each HBM construct to overall HBM perception on the use of contraception was assessed by fitting a logistic regression model. The associations between overall HBM perception and other independent variables with contraceptive use were analyzed using the unpaired t test or Wilcoxon Rank sum test for continuous data and χ2 test for categorical data, as appropriate. Consequently, HBM perceptions of couples and other variables that showed a P value less than .2 were selected and included in a multiple logistic regression model to identify independent predictors for contraceptive use. A P value less than .05 was considered as significant. Finally, the agreement of perceptions in HBM on contraceptive use between female youths and their husbands was determined by the κ coefficient.
Ethical Considerations
The proposal of this study was approved by the Institute Ethics Committee of the Faculty of Medicine, Prince of Songkla University, Hat Yai, Songkhla, Thailand, document reference number SUB.EC 51/ 354-004 on April 29, 2008, and the letter of permission from Ministry of Health, Myanmar on March 18, 2008, before the study was conducted.
Results
A total of 444 individuals (222 eligible couples) were included in the study. The age of the wives ranged from 16 to 24 years (mean ± standard deviation [SD] = 21 ± 2.1) and that of their husbands from 18 to 40 years (mean ± SD = 25 ± 4.3). Almost all were Buddhists (97%), with the remaining being Christian (2%) and Muslim (1%). The distribution of education among wives and husbands was similar, and one third of couples had completed primary school education. One third of the women were housewives without having their own income. Among employed wives, the median monthly income was 30 000 Kyats (US$31) and that of their husbands was 42 500 Kyats (US$44). Almost all husbands were employed, and 49% of them were farmers or running their own business. The mean age at marriage of the women was 19 years (range = 14 to 24 years). Duration of marriage ranged from 1 to 96 months (mean ± SD = 29.4 ± 21.5). Among all couples, 60% had at least 1 child at the time of the interview. The wife’s and husband’s desired family sizes were similar and ranged from 2 to 7, with a median size of 4.
When asked about knowledge of contraceptives, well-known methods, about in wives responded spontaneously, were the contraceptive pills (94%) and 3-month injection (91%). Similarly, these 2 methods were identified spontaneously by the husbands (78% and 80%, respectively). The wife and husband stated that they had heard of female sterilization (76% and 81%), condoms (67% and 49%), and male sterilization (66% and 60%) when the interviewers mentioned these methods. In contrast, many couples did not recognize emergency oral contraceptive pills (87% of wives and 86% of husbands). Only 46% of wives and 32% of husbands replied spontaneously that they could obtain contraceptives from RHCs. The majority of wives and husbands (75% and 62%) mentioned that they would obtain contraceptives from drug shops, whereas one fourth to one third of couples mentioned private clinics. The wives said that they gained their knowledge of contraception from health providers (37%), whereas the husbands stated that they obtained this knowledge from friends (41%).
The effect of each HBM construct to overall HBM of wife’s perception on contraceptive use by logistic regression is shown in Table 1. Of 6 constructs, perceived susceptibility to pregnancy and perceived barriers to contraception were independently significant predictors of contraceptive use. Positive perception of pregnancy susceptibility (OR = 6.3; 95% CI = 2.3-17.4) and negative perception of barriers (OR = 2.9; 95% CI = 1.5-5.7) were highly associated with positive overall HBM perception. Table 2 shows the univariate analysis of all independent factors for wives and husbands between contraceptive users and nonusers. Higher use was observed in wives having a positive HBM perception about contraception, wives having discussions with their spouses on contraception from both wife’s and husband’s perspectives, and wives aged 20 to 24 years, and for duration of marriage greater than 24 months and shorter distance from house to health center. Wife’s HBM perception, age group, and own income as well as spousal communication and distance to the health center remained significant in the final model (Table 3). Wives who had a positive HBM perception, communicated with their spouse on the use of contraception, and aged 20 to 24 years had higher odds of using contraceptives. Longer distance from home to health center reduced the odds of use.
Effect of Each HBM Construct on Overall HBM of Wife’s Perception of Contraceptive Use, by Logistic Regression a
Abbreviations: HBM, Health Belief Model; OR, odds ratio; CI, confidence interval.
*P < .05; ** P < .01; *** P < .001.
Univariate Analysis of Individual, Husband, and Service Factors on Contraceptive Use
Abbreviations: HBM, Health Belief Model; IQR, interquartile range; SD, standard deviation.
Logistic Regression Model for the Predictors of Contraceptive Use a
Abbreviations: HBM, Health Belief Model; OR, odds ratio; CI, confidence interval.
*P < .05; ** P < .01; *** P < .001.
The agreement on each HBM construct and overall HBM perception between wife and husband is shown in Table 4. Of 6 constructs of HBM, fair agreement was seen in the positive perceived severity and negative perceived barriers between wife and husband, with kappa values of 0.2 and 0.3, respectively. Perceived susceptibility to pregnancy and cues to action to the use of contraception, however, showed poor agreement (κ coefficients of 0.12 and 0.18, respectively). Poor agreement was also found in overall perception.
Agreement on HBM Perceptions Between Young Wife and Husband
Abbreviations: HBM, Health Belief Model.
Discussion
The perceptions measured by HBM strongly predicted contraceptive use among married female youths in this rural area of Myanmar. Specifically, perceived susceptibility to pregnancy and perceived barriers to contraception were the 2 most influential constructs. In addition, wife’s age and having her own income as well as spousal communication and distance from home to health center were also significant predictors. However, poor agreement on constructs of HBM between wife and husband was shown.
Our study revealed perceived susceptibility and barriers as strong independent predictors for contraceptive use after adjusting for other factors. This may be because the perception of susceptibility to pregnancy is a basic inside feeling, which is less influenced by other external factors or the environment. This finding was supported by a previous study that perceived susceptibility, barriers, and self-efficacy with the use of birth control in teens with diabetes. 17 Perceived barrier was an additional significant factor in our study supported by previous studies for intention to use birth control 19 and frequency of condom use. 18 In contrast, perceived susceptibility was not associated with the intention of abstinence. 20 Another study reported that contraceptive use was associated with self-efficacy but not with susceptibility or barriers. 21 As a result, the effect of each perception on the seeking behaviors depends on the nature or the characteristics of study participants, the context of interested behaviors, attitudes, and social norms.
In addition to these perceptions, young wives who have their own income and have engaged in spousal communication regarding contraception were more likely to use contraceptives according to our study. Lower frequency of contraceptive use in married youths aged 15 to 19 years compared with those aged 20 to 24 years was also detected, as in 2 previous studies.7,23 This may be because of the immaturity of youths aged 15 to 19 years, especially regarding lack of knowledge of contraception and a lack of income to support the costs of contraception. Young wives who have their own income may obtain services more easily than those who have no income. This explanation was supported by a study in China, which found that the knowledge score on reproduction and contraception of female college students was associated with age and family socioeconomic status of respondents. 24 Our study showed that spousal communication on contraception increased the likelihood of contraceptive use, a finding also seen in previous studies.11,12,25 Likewise, the data from the Sri Lanka Demographic and Health Survey 2000 showed that 80% of current users of contraception pointed out that their contraceptive use depended on the decision of both the women and their spouses. 26 Although one study 27 and 2 unpublished reports in Myanmar revealed the importance of the male factor in contraceptive use, according to our study, female factors were much more important.
In our study, geographic accessibility was also a significant factor, although the areas with RH service coverage were selected as our study setting. Longer distance to travel to obtain contraceptives was associated with lower frequency of contraceptive use. This finding is similar to that in previous studies.10,15 In addition, we found that knowledge about contraception in this age group was also low (data not shown), but this factor was not significant in our study. There have been several studies regarding knowledge related to contraceptive use in adolescents; however, the findings are inconsistent.14,28,29 Some studies have shown a positive relationship, 14 whereas others have found no association. 28 Many studies found that attitude on side effects of contraception was associated with contraceptive use.10,15,30
There were 2 limitations in this study. First, identification of contraceptive users and nonusers was made on the basis of self-report. However, we ensured that a good rapport was established between interviewer and respondent and the researchers explained the objectives of the study before the interview; thus, any deviation from self-reported use of contraception is probably small. Second, this study was restricted to a rural area of one township and may not be generalizable to married youths in urban areas. A further comparative study to identify the predictors of contraceptive use, including culture, lifestyles, and modernization of married female youths and their husbands living in rural and urban areas would be useful for national program planning and evaluation. Although the evidence showed that the perception measured by HBM reflects RH-seeking behaviors among adolescents, there was a variation of significant constructs.17-21 The strength of this community-based survey was to identify predictors of contraceptive use using HBM perceptions and individual, couple, and service factors among married youths and their husbands in a developing country with low rate of contraceptive use and high maternal mortality rates.
Conclusions
Multidimensional factors, including individual, spouse, and service levels, influenced contraceptive use among married youths and their husbands. Perceptions on contraceptive use, as measured by the HBM, spousal communication, age, and distance from home to health center were independent predictors of contraceptive use. The raising of awareness on correct perceptions of contraceptive use among married youths in rural areas can be implemented in the RH policy to increase use of contraception. A study looking at husband involvement in spousal communication would be interesting to further explore the effect it has on contraceptive use among married couples.
Footnotes
Acknowledgements
We wish to acknowledge the authorities from the Ministry of Health, Myanmar, for allowing us to conduct the study. Our sincere thanks are directed to the Medical Officer of Hinthada Township and research assistants from the Department of Medical Research (Lower Myanmar) for their kind help and cooperation in data collection. We are also grateful to all couples who participated in the study. We would like to thank Mr Edward McNeil for manuscript editing.
The author(s) declared no potential conflicts of interest with respect to the authorship and/or publication of this article.
The author(s) disclosed receipt of the following financial support for the research and/or authorship of this article: funding support for this study was provided by the World Health Organization through the Special Programme of Research, Development and Research Training in Human Reproduction (WHO/HRP).