bs_bs_banner
doi:10.1111/jpc.12426
ORIGINAL ARTICLE
Prevalence of unintended pregnancy and its associated factors among sexually active never-married youth in Shanghai Huan He and Robert W Blum Department of Population, Family and Reproductive Health, Bloomberg School of Public Health, Johns Hopkins University, Baltimore, Maryland, United States
Aims: This study was conducted to determine the prevalence and the associated factors for unintended pregnancy among never-married sexually active youth (15–24 years old) in Shanghai. Methods: Using a probability sample (n = 765, including 502 males and 263 females) from 2006, we estimated prevalence of unintended pregnancy involvement for females and males by demographic factors and its associations with contextual and behavioural factors, guided by a social-ecological framework. Results: Weighted proportion analyses estimated that 13.8% of females and 17.7% of males had been involved in an unintended pregnancy, and 99.0% and 90.5% of such pregnancies were reported to have ended in abortion by females and males, respectively. Controlling for sexual behaviour factors and residence history, multivariate hazard analyses with Weibull functions found increased hazard among older female youth (19–21 vs. 15–18 years old) but decreased hazard among older male youth (15–18 vs. 19–21/22–24 years old) as compared with the younger groups. Parallel analyses of unintended pregnancy’s associations with contextual factors by sex also suggested that female unintended pregnancy involvement was more likely to be associated with family factors (family living arrangements and maternal education), while unintended pregnancy involvement among males was only associated with school factors (school climate and sex education on pregnancy) and perceived neighbourhood contraceptive services access. Conclusions: Sexually active and never-married youth are at considerable risk of unintended pregnancy and abortion in Shanghai. Age- and sex-specific strategies will be needed if China is going to be successful in addressing unintended pregnancy. Key words:
abortion; contraception; sex; Shanghai; unintended pregnancy; youth.
What is already known on this topic
What this paper adds
1 As a result of the social changes in sexual values in the past 40 years, never-married Chinese youth are likely to report a higher sexual activity level than previously. 2 Never married youth are disadvantaged in obtaining contraceptive service under China’s ‘One Child’ family planning policy. 3 Studies in other regions of the world have found that unintended pregnancy has caused huge societal, familial and individual burdens.
1 The relatively high prevalence of pregnancy involvement and subsequent abortion among never-married youth in Shanghai is confirmed in a probability sample, which shows higher pregnancy involvement reported by males as compared with that reported by females, though not statistically significantly so. 2 Hazard of involvement in unintended pregnancy is increased among older female youth (19–21 vs. 15–18 years old), but in younger male youth (15–18 vs. 19–21/22–24 years old), as compared with their counterparts. 3 Compared with other areas of the world, there is a distinctly different pattern of involvement in unintended pregnancy, suggesting that in Shanghai, female youth are more sensitive to family influence, while males are more sensitive to school and neighbourhood factors.
Correspondence: Dr Huan He, Department of Population, Family and Reproductive Health, Johns Hopkins Bloomberg School of Public Health, 615 N. Wolfe Street E4527, Baltimore, MD 21205, USA. Fax: +1 410 955 2303; email: [email protected] Conflict of interest: The authors have no conflict of interest related to the study or the present manuscript. Accepted for publication 30 August 2013.
912
Since the 1980s, China has undergone tremendous changes related to social norms of sexual behaviour. In this process of ‘opening up’ to Western culture,1 the high value attached to women’s premarital virginity has decreased,2,3 and the societal disapproval of premarital sex has weakened.2–5 Therefore, it has become more commonplace6 for youth to initiate intercourse prior to marriage2,7 and for never-married women to obtain an abortion.8
Journal of Paediatrics and Child Health 49 (2013) 912–918 © 2013 The Authors Journal of Paediatrics and Child Health © 2013 Paediatrics and Child Health Division (Royal Australasian College of Physicians)
H He et al.
However, never-married youth are disadvantaged in the current Chinese family policy schema because publicly funded contraceptive services are mainly targeted at married couples.6 Given inadequate sex education9 and the ‘One Child Policy’ restrictions, unmarried pregnant youth are expected to get married soon or obtain an abortion.10,11 Thus, Chinese female youth might be at increased risk of unintended pregnancy (UP). As a consequence of its rising prevalence, increased medical risk for maternal and child health, and social costs,12–15 UP is an emerging public health concern for youth in China, especially in coastal metropolitan areas, such as Shanghai.16 However, there is very little information on UP prevalence and its associated factors among Chinese youth. There is still no population-level estimation of youth UP; neither has there been adequate analysis of its associated risk and protective factors. Most Chinese studies related to UP have been limited to selected subpopulations, such as college students11,17 and women attending premarital health checks16 or seeking clinic-based abortions.18 Few studies have explored individual-level factors, such as early sexual debut and rates of contraceptive use,11 rather than contextual-level factors, such as school and neighbourhood factors. Learning more about the associated factors of UP in a social-ecological framework19 will contribute to developing school- or community-based interventions with multifaceted strategies, which tend to be more effective than the usual ‘oneshot’ and merely individual-targeted programmes.20 Using a representative sample in Shanghai interviewed in 2006, this study aims to estimate the UP prevalence for never-married Chinese youth by sex and demographics and then assess its associated factors guided by a modified Bronfenbrenner’s model.21
Methods Data This study is based on data from the Three Asian City Youth Study.7 A cohort of 6299 Shanghainese youth aged 15–24 were recruited using a multistage probability sampling method from both household residential areas and group living facilities (GLFs) in Shanghai in 2006. In the household survey, dwellings with residents aged 15–24 were organised into three levels: from largest to smallest, district, ward and residential group. Districts were randomly selected by economic stratum; then wards were randomly sampled, and then residential groups and dwellings. The GLF samples were from five universities, with a random sample from each, and from all identified informal facilities. Sampling weights were generated to represent the 2.6 million youth population (both household residents and migrants) in Shanghai. One thousand thirty-two youth in the total sample were sexually active. Our study sample was restricted to those who were both sexually active and never married (n = 765, with 502 males and 263 females). The general individual and family background information was collected through face-to-face interviews, while the sensitive questions, such as those about sexual behaviours and pregnancy history, were collected by computer-assisted selfinterview. The study protocol was approved by the Institutional Review Board (IRB) of Johns Hopkins Bloomberg School of
Unintended pregnancy in Shanghai youth
Public Health as well as the Shanghai Institute for Planned Parenthood Research. All participants were provided with full information on the purpose of the study, and informed consent was obtained in accordance with local IRB protocols.
Measures UP was defined for females as a pregnancy that was either unwanted or mistimed22,23 and for males as unintentionally getting their partner pregnant. A self-report questionnaire collected the history of first and subsequent pregnancies. Females were asked, ‘Just before you learned of this [first/second/nth] pregnancy, did you want to get pregnant?’, while males were asked, ‘Just before the [first/second/nth] time you got a girl pregnant, did you want her to get pregnant?’ Responses of ‘Yes, but not until later’ and ‘No’ were coded as UP, while ‘Yes, right then’ was coded as not UP. Time to the first UP was defined as the age difference in years between sexual debut and the event. Pregnancy outcome was coded as ‘live birth’, ‘abortion’, ‘miscarriage’ or ‘stillbirth’. Reports of ‘currently pregnant’ were coded as censored. Independent variables for analysis were selected according to our ecological risk and protective framework, which had three levels (self, family and extrafamilial factors) and several domains (neighbourhood, family, school and individual),24 with a focus on experiences during early and middle adolescence. They were grouped into five clusters: 1. Demographics included age, sex, residence history and employment status. 2. Family characteristics included family living arrangements and maternal education level. 3. School characteristics included school climate and sex education on pregnancy in school. The school climate indicator was developed from the total score of a four-item scale (Cronbach’s alpha = 0.70), which was dichotomised by its mean to indicate supportive versus non-supportive school climate. 4. Neighbourhood health services referred to perceived access to contraceptive services for minors in a neighbourhood. 5. Sexual behaviours included contraceptive use during first sexual intercourse, age at sexual debut and ever having cohabited. Furthermore, employment status, family living arrangements and whether respondents had ever cohabited were recorded both for the time of survey and for each age.
Data analysis The Stata/SE 12 statistical package (College Station, TX, USA) was used for data analysis. First, weighted descriptive analyses were used to estimate the prevalence of UP. Second, time to the first UP was analysed using weighted survival analysis. Third, parameter hazard models with Weibull distribution were used to conduct bivariate and multivariate analyses to identify factors associated with the hazard of UP, as preliminary analysis rejected the assumption of constant hazard over time for the observed data.25 Factors that were statistically significant (P < 0.05) in the bivariate analyses for either sex were entered into multivariate analyses. We found the overall effects of sex (main effect plus all the interaction effects) in the final model
Journal of Paediatrics and Child Health 49 (2013) 912–918 © 2013 The Authors Journal of Paediatrics and Child Health © 2013 Paediatrics and Child Health Division (Royal Australasian College of Physicians)
913
Unintended pregnancy in Shanghai youth
H He et al.
Table 1 Distribution of unintended pregnancy, demographic characteristics and family structure by sex for sexually active never-married youth, Shanghai, 2006
Total Pregnancy Never Intended only Unintended Demographics Age (years) 15–18 19–21 22–24 Residence history Always in urban area Others Working (at time of survey) Yes No Family characteristics Family living arrangements (at time of survey) Mostly with parents Sometimes with parents Not with parents Maternal education Primary school or less Junior secondary/high school Senior secondary/high school or above School experience School climate Unsupportive Supportive Sex education on pregnancy Yes No Neighbourhood health services Perceived access to contraceptive services for minors Yes No Don’t know Sexual behaviours Contraceptive use at first intercourse Yes No Age of sexual debut (years) 18–24
doi:10.1111/jpc.12426
ORIGINAL ARTICLE
Prevalence of unintended pregnancy and its associated factors among sexually active never-married youth in Shanghai Huan He and Robert W Blum Department of Population, Family and Reproductive Health, Bloomberg School of Public Health, Johns Hopkins University, Baltimore, Maryland, United States
Aims: This study was conducted to determine the prevalence and the associated factors for unintended pregnancy among never-married sexually active youth (15–24 years old) in Shanghai. Methods: Using a probability sample (n = 765, including 502 males and 263 females) from 2006, we estimated prevalence of unintended pregnancy involvement for females and males by demographic factors and its associations with contextual and behavioural factors, guided by a social-ecological framework. Results: Weighted proportion analyses estimated that 13.8% of females and 17.7% of males had been involved in an unintended pregnancy, and 99.0% and 90.5% of such pregnancies were reported to have ended in abortion by females and males, respectively. Controlling for sexual behaviour factors and residence history, multivariate hazard analyses with Weibull functions found increased hazard among older female youth (19–21 vs. 15–18 years old) but decreased hazard among older male youth (15–18 vs. 19–21/22–24 years old) as compared with the younger groups. Parallel analyses of unintended pregnancy’s associations with contextual factors by sex also suggested that female unintended pregnancy involvement was more likely to be associated with family factors (family living arrangements and maternal education), while unintended pregnancy involvement among males was only associated with school factors (school climate and sex education on pregnancy) and perceived neighbourhood contraceptive services access. Conclusions: Sexually active and never-married youth are at considerable risk of unintended pregnancy and abortion in Shanghai. Age- and sex-specific strategies will be needed if China is going to be successful in addressing unintended pregnancy. Key words:
abortion; contraception; sex; Shanghai; unintended pregnancy; youth.
What is already known on this topic
What this paper adds
1 As a result of the social changes in sexual values in the past 40 years, never-married Chinese youth are likely to report a higher sexual activity level than previously. 2 Never married youth are disadvantaged in obtaining contraceptive service under China’s ‘One Child’ family planning policy. 3 Studies in other regions of the world have found that unintended pregnancy has caused huge societal, familial and individual burdens.
1 The relatively high prevalence of pregnancy involvement and subsequent abortion among never-married youth in Shanghai is confirmed in a probability sample, which shows higher pregnancy involvement reported by males as compared with that reported by females, though not statistically significantly so. 2 Hazard of involvement in unintended pregnancy is increased among older female youth (19–21 vs. 15–18 years old), but in younger male youth (15–18 vs. 19–21/22–24 years old), as compared with their counterparts. 3 Compared with other areas of the world, there is a distinctly different pattern of involvement in unintended pregnancy, suggesting that in Shanghai, female youth are more sensitive to family influence, while males are more sensitive to school and neighbourhood factors.
Correspondence: Dr Huan He, Department of Population, Family and Reproductive Health, Johns Hopkins Bloomberg School of Public Health, 615 N. Wolfe Street E4527, Baltimore, MD 21205, USA. Fax: +1 410 955 2303; email: [email protected] Conflict of interest: The authors have no conflict of interest related to the study or the present manuscript. Accepted for publication 30 August 2013.
912
Since the 1980s, China has undergone tremendous changes related to social norms of sexual behaviour. In this process of ‘opening up’ to Western culture,1 the high value attached to women’s premarital virginity has decreased,2,3 and the societal disapproval of premarital sex has weakened.2–5 Therefore, it has become more commonplace6 for youth to initiate intercourse prior to marriage2,7 and for never-married women to obtain an abortion.8
Journal of Paediatrics and Child Health 49 (2013) 912–918 © 2013 The Authors Journal of Paediatrics and Child Health © 2013 Paediatrics and Child Health Division (Royal Australasian College of Physicians)
H He et al.
However, never-married youth are disadvantaged in the current Chinese family policy schema because publicly funded contraceptive services are mainly targeted at married couples.6 Given inadequate sex education9 and the ‘One Child Policy’ restrictions, unmarried pregnant youth are expected to get married soon or obtain an abortion.10,11 Thus, Chinese female youth might be at increased risk of unintended pregnancy (UP). As a consequence of its rising prevalence, increased medical risk for maternal and child health, and social costs,12–15 UP is an emerging public health concern for youth in China, especially in coastal metropolitan areas, such as Shanghai.16 However, there is very little information on UP prevalence and its associated factors among Chinese youth. There is still no population-level estimation of youth UP; neither has there been adequate analysis of its associated risk and protective factors. Most Chinese studies related to UP have been limited to selected subpopulations, such as college students11,17 and women attending premarital health checks16 or seeking clinic-based abortions.18 Few studies have explored individual-level factors, such as early sexual debut and rates of contraceptive use,11 rather than contextual-level factors, such as school and neighbourhood factors. Learning more about the associated factors of UP in a social-ecological framework19 will contribute to developing school- or community-based interventions with multifaceted strategies, which tend to be more effective than the usual ‘oneshot’ and merely individual-targeted programmes.20 Using a representative sample in Shanghai interviewed in 2006, this study aims to estimate the UP prevalence for never-married Chinese youth by sex and demographics and then assess its associated factors guided by a modified Bronfenbrenner’s model.21
Methods Data This study is based on data from the Three Asian City Youth Study.7 A cohort of 6299 Shanghainese youth aged 15–24 were recruited using a multistage probability sampling method from both household residential areas and group living facilities (GLFs) in Shanghai in 2006. In the household survey, dwellings with residents aged 15–24 were organised into three levels: from largest to smallest, district, ward and residential group. Districts were randomly selected by economic stratum; then wards were randomly sampled, and then residential groups and dwellings. The GLF samples were from five universities, with a random sample from each, and from all identified informal facilities. Sampling weights were generated to represent the 2.6 million youth population (both household residents and migrants) in Shanghai. One thousand thirty-two youth in the total sample were sexually active. Our study sample was restricted to those who were both sexually active and never married (n = 765, with 502 males and 263 females). The general individual and family background information was collected through face-to-face interviews, while the sensitive questions, such as those about sexual behaviours and pregnancy history, were collected by computer-assisted selfinterview. The study protocol was approved by the Institutional Review Board (IRB) of Johns Hopkins Bloomberg School of
Unintended pregnancy in Shanghai youth
Public Health as well as the Shanghai Institute for Planned Parenthood Research. All participants were provided with full information on the purpose of the study, and informed consent was obtained in accordance with local IRB protocols.
Measures UP was defined for females as a pregnancy that was either unwanted or mistimed22,23 and for males as unintentionally getting their partner pregnant. A self-report questionnaire collected the history of first and subsequent pregnancies. Females were asked, ‘Just before you learned of this [first/second/nth] pregnancy, did you want to get pregnant?’, while males were asked, ‘Just before the [first/second/nth] time you got a girl pregnant, did you want her to get pregnant?’ Responses of ‘Yes, but not until later’ and ‘No’ were coded as UP, while ‘Yes, right then’ was coded as not UP. Time to the first UP was defined as the age difference in years between sexual debut and the event. Pregnancy outcome was coded as ‘live birth’, ‘abortion’, ‘miscarriage’ or ‘stillbirth’. Reports of ‘currently pregnant’ were coded as censored. Independent variables for analysis were selected according to our ecological risk and protective framework, which had three levels (self, family and extrafamilial factors) and several domains (neighbourhood, family, school and individual),24 with a focus on experiences during early and middle adolescence. They were grouped into five clusters: 1. Demographics included age, sex, residence history and employment status. 2. Family characteristics included family living arrangements and maternal education level. 3. School characteristics included school climate and sex education on pregnancy in school. The school climate indicator was developed from the total score of a four-item scale (Cronbach’s alpha = 0.70), which was dichotomised by its mean to indicate supportive versus non-supportive school climate. 4. Neighbourhood health services referred to perceived access to contraceptive services for minors in a neighbourhood. 5. Sexual behaviours included contraceptive use during first sexual intercourse, age at sexual debut and ever having cohabited. Furthermore, employment status, family living arrangements and whether respondents had ever cohabited were recorded both for the time of survey and for each age.
Data analysis The Stata/SE 12 statistical package (College Station, TX, USA) was used for data analysis. First, weighted descriptive analyses were used to estimate the prevalence of UP. Second, time to the first UP was analysed using weighted survival analysis. Third, parameter hazard models with Weibull distribution were used to conduct bivariate and multivariate analyses to identify factors associated with the hazard of UP, as preliminary analysis rejected the assumption of constant hazard over time for the observed data.25 Factors that were statistically significant (P < 0.05) in the bivariate analyses for either sex were entered into multivariate analyses. We found the overall effects of sex (main effect plus all the interaction effects) in the final model
Journal of Paediatrics and Child Health 49 (2013) 912–918 © 2013 The Authors Journal of Paediatrics and Child Health © 2013 Paediatrics and Child Health Division (Royal Australasian College of Physicians)
913
Unintended pregnancy in Shanghai youth
H He et al.
Table 1 Distribution of unintended pregnancy, demographic characteristics and family structure by sex for sexually active never-married youth, Shanghai, 2006
Total Pregnancy Never Intended only Unintended Demographics Age (years) 15–18 19–21 22–24 Residence history Always in urban area Others Working (at time of survey) Yes No Family characteristics Family living arrangements (at time of survey) Mostly with parents Sometimes with parents Not with parents Maternal education Primary school or less Junior secondary/high school Senior secondary/high school or above School experience School climate Unsupportive Supportive Sex education on pregnancy Yes No Neighbourhood health services Perceived access to contraceptive services for minors Yes No Don’t know Sexual behaviours Contraceptive use at first intercourse Yes No Age of sexual debut (years) 18–24
