- Email: [email protected]
Factors Associated with Unwanted Pregnancy among Adolescents in Russia
Accepted Manuscript Factors associated with unwanted pregnancy among adolescents in Russia O.V. Panova, MD, A.M. Kulikov, MD, professor, A. Berchtold, PhD, J.C. Suris, MD, MPH, PhD PII:
S1083-3188(16)30011-0
DOI:
10.1016/j.jpag.2016.04.004
Reference:
PEDADO 1985
To appear in:
Journal of Pediatric and Adolescent Gynecology
Received Date: 6 April 2015 Revised Date:
31 March 2016
Accepted Date: 5 April 2016
Please cite this article as: Panova OV, Kulikov AM, Berchtold A, Suris JC, Factors associated with unwanted pregnancy among adolescents in Russia, Journal of Pediatric and Adolescent Gynecology (2016), doi: 10.1016/j.jpag.2016.04.004. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Factors associated with unwanted pregnancy among adolescents in Russia
Panova O.V., MD 1
Berchtold A., PhD 2 Suris J.C., MD, MPH, PhD 3 1
RI PT
Kulikov A.M., MD, professor 1
Department of Pediatrics and Child Cardiology, North-Western State Medical University
SC
named after I.I. Mechnikov, Saint-Petersburg, Russia
Institute of Social Sciences & LIVES, University of Lausanne, Lausanne, Switzerland
3
Institute of Social and Preventive Medicine, Lausanne University Hospital, Lausanne,
Switzerland All correspondence: Oksana Panova [email protected]
TE D
+7 921 400 79 98
M AN U
2
AC C
EP
Dunaiskiy prosp. 34/16, apt.530, Saint Petersburg, Russia
ACCEPTED MANUSCRIPT Abstract Study objective: to identify social and behavioral factors associated with unintended pregnancy among adolescents in Russia.
RI PT
Design: cross-sectional. Setting and participants: The investigation group consisted of 145 adolescents aged 15-19 years admitted to a youth friendly clinic for induced abortion. The control group consisted of
SC
77 sexually active never pregnant adolescent girls of similar age admitted for other reasons. Interventions: Both groups completed a self-administered questionnaire about their
M AN U
occupation, educational level of their mother, family structure and socioeconomic status, relationship with their mother, sexual behavior and substance use.
Main Outcome Measures: factors associated with unintended pregnancy. Results: In the bivariate analysis pregnant adolescents tended to have a lower educational level, to live in a non-intact family, to prefer stronger alcohol drinks, and to have a younger
TE D
age at sexual debut and more frequent sexual intercourse. In the log-linear model six factors appeared to be directly linked to pregnancy: mother’s lower education, alcohol consumption,
EP
non intact family structure, low academic grades, higher frequency of sexual intercourse and lower age at sexual debut.
AC C
Conclusion: Although the issue of unwanted pregnancy and possible ways to prevent it should be kept in mind in the psychosocial assessment of any adolescent, the results of our study suggest that some adolescents are at increased risk of unwanted pregnancy. This group includes adolescent girls with low educational attainment, living in a disrupted family and reporting other risk behaviors. These adolescents may specially benefit from specific preventive programs in order to reduce the number of adverse outcomes. Key words: adolescents, unwanted pregnancy, associated factors.
ACCEPTED MANUSCRIPT
Introduction Adolescent pregnancy and abortion remains an important issue for many countries
their children and influences the whole family and the society 1,2,3
RI PT
including Russia as it has social, behavioral and health adverse effects on adolescent mothers,
Teenage pregnancy and abortion rates in the Russian Federation are among the highest in Europe 4,5 in spite of the gradual decrease seen during the last decade 6. In 2009 nearly
SC
100.000 abortions were performed in Russia among adolescents 7, which represents a rate of 21.9 per 1000 adolescents aged 15-19 years old 6. Adolescent pregnancies in the Russian
M AN U
Federation are mostly unwanted and in more than 50% of the cases the pregnancy is interrupted 5. In order to detect youths at risk and to elaborate preventive programs against unwanted adolescent pregnancies, numerous studies have been conducted in the United States and Europe to determine the risk factors for adolescent pregnancy 2, 3, 8, 9 that can be classified
TE D
at the sociodemographic, family or individual level 1.
A number of studies indicated that disrupted family structure with absence of the father was an important family level risk factor for adolescent pregnancy 1, 8, 10 while the presence of
EP
the father in the household was determined as a key protective factor 11. Similarly, a close relationship and connectedness of the family was shown to be a protective factor against teen
AC C
pregnancy 2, 10 specially the relationship with the mother12. Low socioeconomic status of the family 3, 10 and low educational attainment of parents 1, 2 have also been considered important family risk factors.
At the individual-level, both substance use and risky sexual behavior, have been associated with adolescent pregnancy. There is evidence indicating that smoking and alcohol 8 and drug use 10 increase the probability of becoming pregnant among adolescents. Other studies have shown that adolescents experiencing early sexual debut 8, 13, with high frequency
ACCEPTED MANUSCRIPT of sexual intercourse 11 and greater number of sexual partners 14 also have a higher likelihood of becoming pregnant. Low academic performance and educational expectations of adolescents are also considered to be strong individual risk factors associated with adolescent pregnancy 1, 11, 15.
RI PT
Although adolescent pregnancy remains an important problem in Russia due to its high prevalence, only a few studies have been conducted in the domain of the sexual and
reproductive behavior of adolescents that may influence the fertility rate 16, 17. Understanding
SC
the key factors contributing to unwanted adolescent pregnancy can be used in the
development of preventive programs to reduce the number of unwanted pregnancies among
M AN U
adolescents.
The objective of our research was to identify social and behavioral factors associated with the first unwanted pregnancy among adolescent females attending a youth-friendly clinic in St Petersburg, Russia.
TE D
Methods
This study was part of a larger research on abortion among adolescents conducted in Saint-Petersburg in 2010-2012. The investigation group consisted of 145 adolescents aged 15-
EP
19 years, admitted to Juventa clinic for induced abortion. The Juventa clinic was created in 1993 and became the largest youth friendly clinic in Russia. A network of more than 20 youth
AC C
friendly clinics, including Juventa, exists in Saint-Petersburg, but most cases of unwanted adolescent pregnancy are referred to Juventa clinic. The control group consisted of 77 sexually active never pregnant adolescent girls of the same age admitted to Juventa for other reasons. The exclusion criteria were the will of the adolescent to continue the pregnancy or having had a previous pregnancy. In the investigation group 17 adolescents were excluded from the study: 10 adolescents wanted to continue the pregnancy, 4 did not want to participate
ACCEPTED MANUSCRIPT and 3 experienced repeated pregnancy. Three adolescents in the control group refused to participate. Both groups completed an anonymous self-administered paper-and-pencil questionnaire in a separate room in the clinical settings. It included items concerning their occupation, the
RI PT
educational level of their mother, the family structure, the socioeconomic status of the family, the relationship with their mother, their sexual behavior and substance use.
School status was defined by a variable taking four modalities: working (not in school
SC
anymore), mandatory school, secondary professional school, and high school. Self-reported academic grades were evaluated with the question, “What is your average academic grade?”
M AN U
The response options were dichotomized into “average or better” and “below average”. The educational level of the mother was categorized as unknown, secondary mandatory/professional school or high school. Family type was dichotomized as living with both parents or living with only one parent or other family member. Relationship with the
TE D
mother was dichotomized as achieving mutual understanding (friendly or striving for mutual understanding) or difficult (with regular conflicts or with low connectedness). Socioeconomic status of the family was considered to be low if an adolescent gave a negative answer on the
EP
question “Does your family have enough money for everyday necessities?” Smoking was dichotomized into smokers and non-smokers. Alcohol consumption at present
AC C
time was also considered, and among adolescents reporting drinking alcohol, we distinguished between those preferring soft alcohol drinks (beer, wine) and those consuming strong alcohols (vodka for instance). Adolescents were also asked whether they had ever used illegal drugs. Participants from both groups reported the age at first sexual intercourse which was subsequently dichotomized as “before 15” and “15 and older”. We also inquired about the lifetime number of sexual partners which was dichotomized as 2 or less and more than 2.
ACCEPTED MANUSCRIPT Every adolescent was asked about the frequency of sexual intercourse which was divided in two categories (once a week or less/ more than once a week). The study was performed in accordance with the ethical standards on human
their informed consent to participate in the research.
Statistical analyses
RI PT
experimentation and was approved by the local ethics committee. All the adolescents gave
SC
In a first step, we performed a bivariate analysis comparing the two groups of
adolescents (pregnant and control). We used the Chi-square test to compare categorical
M AN U
variables and the Student’s t test for continuous variables.
To provide a broader picture of the factors possibly associated with the group variable, all factors significant at the bivariate level were then introduced in a log-linear analysis18. This non-directional method allowed us to consider all possible associations between variables
TE D
including associations between potential factors. Preliminary computations showed that associations of order higher than two were mostly non-significant, so we included only bivariate relations and the main effect of each variable into the starting model. We performed
EP
then a selection of the associations between variables using the Bayesian Information Criterion (BIC)19, the goal being to keep only necessary relations. BIC allowed us to take into
AC C
account both the quality of fit and the complexity of the model, leading to a parsimonious final model. Results are presented as a diagram showing all remaining significant associations in the final model. Finally, all variables directly associated with the dependent variable were included in a logistic regression. The significance level was set at 0.05 for all analyses. R 2.12.2 (The R Foundation for Statistical Computing) was used for all computations. Results
ACCEPTED MANUSCRIPT Table 1 describes the comparison between groups. Overall, adolescents in the pregnancy group were significantly less likely to attend high school, and more likely to have low grades and a less educated mother. They were also less likely to live in an intact family, but no differences were observed for family income or
RI PT
relationship with the mother. Girls in the pregnant group were also significantly more likely to report other risk behaviors such as smoking or drinking strong alcohol, but interestingly pregnant girls were also more likely to not drink alcohol at all. No difference was found
SC
regarding age or the use of illegal drugs.
Girls with unwanted pregnancy were significantly more likely to have had an early
M AN U
sexual debut and reported more frequent sexual intercourse. However, there was no significant difference in the lifetime number of sexual partners between groups. Log-linear model
Figure 1 shows the structure of association of the final model. Six factors that were
TE D
significant at the bivariate level were found to be directly associated with unwanted pregnancy in the final model: mother’s educations, academic grades, alcohol consumption, age at sexual debut and frequency of sexual intercourse.
EP
Logistic regression
Table 2 presents the results. Among the 6 factors directly associated with the group
AC C
membership in the log-linear model, five remained significant in the logistic regression, the only exception being progress in studies. Living with only one parent or other relatives, a mother with an unknown or a secondary school level of education, a preference for low alcohol drinks (as compared to no alcohol consumption), an onset of sexual relations before age 15, and having sexual intercourse at least twice weekly were all factors associated with a higher probability of being in the pregnant group. It must be noted that the preference for strong alcohol drinks was not significant.
ACCEPTED MANUSCRIPT
Discussion Associated factors The study describes the main characteristics of adolescents experiencing unwanted
RI PT
pregnancy in Saint-Petersburg based on a self-administered questionnaire. The results show that almost the same types of the family and individual risk factors described in the literature influence the probability of pregnancy among adolescents in Saint-Petersburg.
SC
School status and academic grades of the adolescents were found to be associated with pregnancy. Positive relations between education level and general health are well known.
M AN U
There is good evidence that people with higher level of education have lower level of mortality, are less likely to suffer from acute and chronic diseases and are more likely to follow healthy life styles. These positive outcomes also include a lower probability of risky behaviors20. The power of education needs to be considered in improving the situation
TE D
regarding adolescent pregnancy.
Family settings were also shown to be associated with unwanted pregnancy. Living in a disrupted family with the absence of father may lead an adolescent girl to seek for
EP
interpersonal relations with males outside home. On the other hand a single parent in the family may have a lower probability to spend more time with an adolescent and have lower
AC C
monitoring over her activities and relations outside home. Contrary to some research 2, 10 we did not observe a significant difference in connectedness with their mother between the two groups, though there was a tendency for more complicated relations among adolescents who were pregnant. The majority of adolescent girls from both groups described their relations with mothers as achieving mutual understanding. This implies that good connectedness to mothers among adolescents does not necessarily reduce the possibility of unwanted pregnancy.
ACCEPTED MANUSCRIPT The relation between alcohol consumption and adolescent pregnancy was quite complex. If the consumption of strong alcohol drinks was clearly associated with a higher probability of being in the pregnant group, the same result was obtained for non-drinking girls. There is evidence that pregnancy is a strong motive for adolescents to stop smoking and
RI PT
drinking alcohol10. We hypothesize that a fraction of the pregnant girls did stop consuming alcohol after the beginning of their pregnancy for the sake of their baby.
There is evidence that adolescents exposed to alcohol are more likely to have sexual
SC
relations21, 22. In our research tobacco smoking was also a risk factor for unwanted pregnancy, but only indirectly as revealed by the log-linear analysis. We hypothesize that tobacco
M AN U
smoking can be generally associated with other types of risky behavior such as alcohol consumption, using illegal drugs and unprotected sexual intercourse but does not directly influence the behavior of adolescents including sexual behavior as it was shown for alcohol consumption. Contrary to the literature 10 we did not see a large difference in illegal drug use
TE D
among pregnant and non-pregnant adolescents. However some adolescents may not want to disclose this fact.
As shown in previous research 8, 11 sexual behavior is a strong factor influencing the
EP
teenage pregnancy like in many countries. In this respect adolescent girls with early sexual debut and high frequency of sexual intercourse may particularly benefit from the sexual
AC C
educational programmes.
There is an assumption that personal deprivation may be associated with living in a deprived area or community 23,24 where unstable employment and low aiming at education may influence personal goals of an adolescent. Family settings and attitudes may influence an adolescent even more. Mother’s education occurred to be another factor associated with adolescent pregnancy found in our research. We hypothesize that low educational level of
ACCEPTED MANUSCRIPT mothers may reflect a low value of education in families that influence the personal goals of adolescent girls. The results of our study highlight the importance of the educational level of mothers and children that were negatively associated with unwanted pregnancy among adolescents. In this
RI PT
respect, general educational level of the family and of the child should be taken into account that can be considered as an important tool to improve the situation Study limitations
SC
When interpreting the results of the study, a number of limitations should be taken into account. First of all, we performed a cross-sectional study, which cannot provide evidence of
M AN U
causal relationship. Second, there were some gaps in the data: a few adolescents did not have information about their mothers. These data were put into the final comparison table as unknown and the comparison of the two groups was made including this category. Third, a clinical sample was used and it may not represent all pregnant adolescents, even though it was
TE D
taken from the largest youth clinic of Saint-Petersburg which is also the largest center for abortions among adolescents. Finally, due to the relatively small size of the sample we may have a low statistical power.
EP
Implications of the study
Although the issue of unwanted pregnancy and possible ways to prevent it should be
AC C
kept in mind in the psychosocial assessment of any adolescent, the results of our study suggest that some adolescents are at increased risk of unwanted pregnancy. This group includes adolescent girls with low educational attainment, living in a disrupted family and reporting other risk behaviors. These adolescents may specially benefit from specific preventive programs in order to reduce the number of adverse outcomes.
ACCEPTED MANUSCRIPT
References 1. Coyne CA, D'Onofrio BM: Some (but not much) progress toward understanding teenage
42:113-152
RI PT
childbearing: a review of research from the past decade. Adv Child Dev Behav 2012;
2. Santelli JS, Melnikas AJ: Teen fertility in transition: recent and historic trends in the United States. Annu Rev Public Health 2010; 31:371-83
SC
3. Santos MI, Rosário F: A score for assessing the risk of first-time adolescent pregnancy. Fam Pract 2011; 28(5):482-8
M AN U
4. Avery L, Lazdane G: What do we know about sexual and reproductive health of adolescents in Europe? Eur J Contracept Reprod Health Care 2010; 15 Suppl 2:S54-66 5. Singh S, Darroch JE: Adolescent pregnancy and childbearing: levels and trends in developed countries. Fam Plann Perspect 2000; 32(1):14-23
TE D
6. The TransMonEE (Transformative Monitoring for Enhanced Equity) database – Monitoring Situation of the Women and Children in countries of Central Eastern Europe and Commonwealth of independent States (CEECIS). UNICEF Regional Office for
EP
CEECIS countries. Available at: http://www.transmonee.org/ (accessed 15.03.2015). 7. Eurostat database: http://ec.europa.eu/eurostat/data/database/ (accessed 15.03.2015)
AC C
8. Imamura M, Tucker J, Hannaford P, et al: Factors associated with teenage pregnancy in the European Union countries: a systematic review. Eur J Public Health 2007; 17(6):630-6.
9. Xie H, Cairns BD, Cairns RB: Predicting Teen Motherhood and Teen Fatherhood: Individual Characteristics and Peer Affiliations. Soc Dev 2001; 10: 488–511 10. Black AY, Fleming NA, Rome ES: Pregnancy in adolescents. Adolesc Med State Art Rev 2012; 23(1):123-38
ACCEPTED MANUSCRIPT 11. Mmari K, Blum RW: Risk and protective factors that affect adolescent reproductive health in developing countries: a structured literature review. Glob Public Health 2009; 4(4):350-66 12. Miller BC: Family influences on adolescent sexual and contraceptive behavior. J Sex
RI PT
Res 2002; 39(1):22-6
13. Lion KC, Prata N, Stewart C: Adolescent childbearing in Nicaragua: a quantitative assessment of associated factors. Int Perspect Sex Reprod Health 2009; 35(2):91-6
SC
14. Planned Parenthood Federation of America, Inc. "Pregnancy and Childbearing Among U.S. Teens." PlannedParenthood.org. (accessed 13.03.2015)
M AN U
http://www.plannedparenthood.org/files/PPFA/pregnancy_and_childbearing.pdf. 15. De Almeida MC, Aquino EM: The role of educational level in the intergenerational pattern of adolescent pregnancy in Brazil. Int Perspect Sex Reprod Health 2009;13:139– 146
TE D
16. Nizamov IG, Chechulina OV: A sociological study of factors affecting reproductive health of female teenagers and young women. [Article in Russian]. Probl Sotsialnoi Gig Zdravookhranenniiai Istor Med 2003; (3):10-2
EP
17. Perminova SG, Sotnikova EI: Sexual and reproductive behavior as seen in the results of a randomized study [Article in Russian]. Akush Ginekol (Mosk) 1993; (3):23-6.
AC C
18. Agresti A: Categorical Data Analysis, (2nd ed), 2002, Wiley-Interscience 19. Kass RE, Raftery AE: Bayes Factors. J Am Stat Assoc 1995; 90(430): 773-795 20. Cutler DM, Lleras-Muney A: Education and Health: Evaluating Theories and Evidence. NBER Working Paper No. 12352. Cambridge, MA: National Bureau of Economic Research, 2006 21. Bellis MA, Morleo M, Tocque K, et al: Contributions of alcohol use to teenage pregnancy: An initial examination of geographical and evidence based associations.
ACCEPTED MANUSCRIPT North West Public Health Observatory, Centre for Public Health, Liverpool John Moores University, Liverpool. (2009). (accessed 09.03.2015): http://www.cph.org.uk/showPublication.aspx?pubid=579 22. Cook PA, Harkins C, Morleo M, et al: Contributions of Alcohol Use to Teenage
RI PT
Pregnancy and Sexually Transmitted Infection Rates. North West Public Health
Observatory, Centre for Public Health, Liverpool John Moores University, Liverpool, 2010. (accessed 09.03.2015): http://www.cph.org.uk/showPublication.aspx?pubid=709
Chicago: University of Chicago Press, 1987
SC
23. Wilson WJ: The truly disadvantaged: The inner city, the underclass and public policy.
M AN U
24. Smith DM, Elander J: Effects of area and family deprivation on risk factors for teenage
AC C
EP
TE D
pregnancy among 13 – 15-year-old girls. Psychol Health Med 2006; 11(4): 399-410
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Table 1 Comparison between groups.
ACCEPTED MANUSCRIPT
Figure 1 Structure of the final log-linear model. Variables directly associated with the
AC C
EP
TE D
M AN U
SC
RI PT
Group variable appear in bold.
ACCEPTED MANUSCRIPT Table 2 Logistic regression model corresponding to the log-linear analysis for predicting the
AC C
EP
TE D
M AN U
SC
RI PT
probability of being in the pregnant group.
ACCEPTED MANUSCRIPT
Variable
Categories
Age (mean ± SE) School status
Pregnant %
Control %
(n=145)
(n=77)
17.12 ± 0.09
17.13 ± 0.14
0.943
5.5%
0.0%
<0.001
35.2%
26.0%
Working
school High school Low
Family type
Living with both parents
Mother’s education
Unknown
mother
13.1%
36.4% 27.3%
0.011
36.1%
67.5%
<0.001
6.9%
1.3%
0.001
Secondary school
66.2%
48.1%
High school
26.9%
50.6%
6.9%
2.6%
78.6%
88.3%
14.5%
9.1%
4.9%
6.5%
0.610
Unknown
TE D
Relation with the
37.7%
44.8%
M AN U
Academic degree
46.2%
SC
Secondary professional
RI PT
Mandatory school
p
Achieving mutual
0.178
EP
understanding Contentious
AC C
Family income
Low
Tobacco smoking
Smoke regularly
82.5%
46.8%
0.024
Alcohol consumption
No consumption
24.1%
5.2%
<0.001
Soft alcohol drinks (beer,
30.3%
68.8%
Strong alcohol drinks
45.5%
26.0%
Yes
15.3%
21.6%
wine)
Ever used illegal drugs
0.242
ACCEPTED MANUSCRIPT
Age at sexual debut
Before 15
37.8%
11.7%
<0.001
Number of sexual
More than 2
39.2%
34.7%
0.515
More than once a week
66.2%
38.7%
<0.001
partners (lifetime) Frequency of sexual
AC C
EP
TE D
M AN U
SC
RI PT
intercourse
ACCEPTED MANUSCRIPT
Academic degree
Family type
Odds ratio
95% CI
p
Low
1.31
0.61 – 2.86
0.494
Average and above
REF
Living with both parents
REF
Living with one parent or
2.85
1.38 - 6.07
0.005
9.42
1.41 – 189.34
0.049
4.59
2.09 – 10.70
<0.001
0.06
0.01 – 0.20
<0.001
Strong alcohol drinks
0.38
0.09 – 1.29
0.144
Before 15
5.66
2.27 – 15.81
<0.001
15 or after
REF
Once a week or less
REF
More than once a week
3.94
1.87 – 8.70
<0.001
other relatives Mother’s education
Unknown Secondary school High school No consumption
consumption
M AN U
Preferences in alcohol
Low alcohol drinks (beer,
Age at sexual debut
TE D
wine)
AC C
intercourse
EP
Frequency of sexual
RI PT
Categories
SC
Variable
REF
REF
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
S1083-3188(16)30011-0
DOI:
10.1016/j.jpag.2016.04.004
Reference:
PEDADO 1985
To appear in:
Journal of Pediatric and Adolescent Gynecology
Received Date: 6 April 2015 Revised Date:
31 March 2016
Accepted Date: 5 April 2016
Please cite this article as: Panova OV, Kulikov AM, Berchtold A, Suris JC, Factors associated with unwanted pregnancy among adolescents in Russia, Journal of Pediatric and Adolescent Gynecology (2016), doi: 10.1016/j.jpag.2016.04.004. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Factors associated with unwanted pregnancy among adolescents in Russia
Panova O.V., MD 1
Berchtold A., PhD 2 Suris J.C., MD, MPH, PhD 3 1
RI PT
Kulikov A.M., MD, professor 1
Department of Pediatrics and Child Cardiology, North-Western State Medical University
SC
named after I.I. Mechnikov, Saint-Petersburg, Russia
Institute of Social Sciences & LIVES, University of Lausanne, Lausanne, Switzerland
3
Institute of Social and Preventive Medicine, Lausanne University Hospital, Lausanne,
Switzerland All correspondence: Oksana Panova [email protected]
TE D
+7 921 400 79 98
M AN U
2
AC C
EP
Dunaiskiy prosp. 34/16, apt.530, Saint Petersburg, Russia
ACCEPTED MANUSCRIPT Abstract Study objective: to identify social and behavioral factors associated with unintended pregnancy among adolescents in Russia.
RI PT
Design: cross-sectional. Setting and participants: The investigation group consisted of 145 adolescents aged 15-19 years admitted to a youth friendly clinic for induced abortion. The control group consisted of
SC
77 sexually active never pregnant adolescent girls of similar age admitted for other reasons. Interventions: Both groups completed a self-administered questionnaire about their
M AN U
occupation, educational level of their mother, family structure and socioeconomic status, relationship with their mother, sexual behavior and substance use.
Main Outcome Measures: factors associated with unintended pregnancy. Results: In the bivariate analysis pregnant adolescents tended to have a lower educational level, to live in a non-intact family, to prefer stronger alcohol drinks, and to have a younger
TE D
age at sexual debut and more frequent sexual intercourse. In the log-linear model six factors appeared to be directly linked to pregnancy: mother’s lower education, alcohol consumption,
EP
non intact family structure, low academic grades, higher frequency of sexual intercourse and lower age at sexual debut.
AC C
Conclusion: Although the issue of unwanted pregnancy and possible ways to prevent it should be kept in mind in the psychosocial assessment of any adolescent, the results of our study suggest that some adolescents are at increased risk of unwanted pregnancy. This group includes adolescent girls with low educational attainment, living in a disrupted family and reporting other risk behaviors. These adolescents may specially benefit from specific preventive programs in order to reduce the number of adverse outcomes. Key words: adolescents, unwanted pregnancy, associated factors.
ACCEPTED MANUSCRIPT
Introduction Adolescent pregnancy and abortion remains an important issue for many countries
their children and influences the whole family and the society 1,2,3
RI PT
including Russia as it has social, behavioral and health adverse effects on adolescent mothers,
Teenage pregnancy and abortion rates in the Russian Federation are among the highest in Europe 4,5 in spite of the gradual decrease seen during the last decade 6. In 2009 nearly
SC
100.000 abortions were performed in Russia among adolescents 7, which represents a rate of 21.9 per 1000 adolescents aged 15-19 years old 6. Adolescent pregnancies in the Russian
M AN U
Federation are mostly unwanted and in more than 50% of the cases the pregnancy is interrupted 5. In order to detect youths at risk and to elaborate preventive programs against unwanted adolescent pregnancies, numerous studies have been conducted in the United States and Europe to determine the risk factors for adolescent pregnancy 2, 3, 8, 9 that can be classified
TE D
at the sociodemographic, family or individual level 1.
A number of studies indicated that disrupted family structure with absence of the father was an important family level risk factor for adolescent pregnancy 1, 8, 10 while the presence of
EP
the father in the household was determined as a key protective factor 11. Similarly, a close relationship and connectedness of the family was shown to be a protective factor against teen
AC C
pregnancy 2, 10 specially the relationship with the mother12. Low socioeconomic status of the family 3, 10 and low educational attainment of parents 1, 2 have also been considered important family risk factors.
At the individual-level, both substance use and risky sexual behavior, have been associated with adolescent pregnancy. There is evidence indicating that smoking and alcohol 8 and drug use 10 increase the probability of becoming pregnant among adolescents. Other studies have shown that adolescents experiencing early sexual debut 8, 13, with high frequency
ACCEPTED MANUSCRIPT of sexual intercourse 11 and greater number of sexual partners 14 also have a higher likelihood of becoming pregnant. Low academic performance and educational expectations of adolescents are also considered to be strong individual risk factors associated with adolescent pregnancy 1, 11, 15.
RI PT
Although adolescent pregnancy remains an important problem in Russia due to its high prevalence, only a few studies have been conducted in the domain of the sexual and
reproductive behavior of adolescents that may influence the fertility rate 16, 17. Understanding
SC
the key factors contributing to unwanted adolescent pregnancy can be used in the
development of preventive programs to reduce the number of unwanted pregnancies among
M AN U
adolescents.
The objective of our research was to identify social and behavioral factors associated with the first unwanted pregnancy among adolescent females attending a youth-friendly clinic in St Petersburg, Russia.
TE D
Methods
This study was part of a larger research on abortion among adolescents conducted in Saint-Petersburg in 2010-2012. The investigation group consisted of 145 adolescents aged 15-
EP
19 years, admitted to Juventa clinic for induced abortion. The Juventa clinic was created in 1993 and became the largest youth friendly clinic in Russia. A network of more than 20 youth
AC C
friendly clinics, including Juventa, exists in Saint-Petersburg, but most cases of unwanted adolescent pregnancy are referred to Juventa clinic. The control group consisted of 77 sexually active never pregnant adolescent girls of the same age admitted to Juventa for other reasons. The exclusion criteria were the will of the adolescent to continue the pregnancy or having had a previous pregnancy. In the investigation group 17 adolescents were excluded from the study: 10 adolescents wanted to continue the pregnancy, 4 did not want to participate
ACCEPTED MANUSCRIPT and 3 experienced repeated pregnancy. Three adolescents in the control group refused to participate. Both groups completed an anonymous self-administered paper-and-pencil questionnaire in a separate room in the clinical settings. It included items concerning their occupation, the
RI PT
educational level of their mother, the family structure, the socioeconomic status of the family, the relationship with their mother, their sexual behavior and substance use.
School status was defined by a variable taking four modalities: working (not in school
SC
anymore), mandatory school, secondary professional school, and high school. Self-reported academic grades were evaluated with the question, “What is your average academic grade?”
M AN U
The response options were dichotomized into “average or better” and “below average”. The educational level of the mother was categorized as unknown, secondary mandatory/professional school or high school. Family type was dichotomized as living with both parents or living with only one parent or other family member. Relationship with the
TE D
mother was dichotomized as achieving mutual understanding (friendly or striving for mutual understanding) or difficult (with regular conflicts or with low connectedness). Socioeconomic status of the family was considered to be low if an adolescent gave a negative answer on the
EP
question “Does your family have enough money for everyday necessities?” Smoking was dichotomized into smokers and non-smokers. Alcohol consumption at present
AC C
time was also considered, and among adolescents reporting drinking alcohol, we distinguished between those preferring soft alcohol drinks (beer, wine) and those consuming strong alcohols (vodka for instance). Adolescents were also asked whether they had ever used illegal drugs. Participants from both groups reported the age at first sexual intercourse which was subsequently dichotomized as “before 15” and “15 and older”. We also inquired about the lifetime number of sexual partners which was dichotomized as 2 or less and more than 2.
ACCEPTED MANUSCRIPT Every adolescent was asked about the frequency of sexual intercourse which was divided in two categories (once a week or less/ more than once a week). The study was performed in accordance with the ethical standards on human
their informed consent to participate in the research.
Statistical analyses
RI PT
experimentation and was approved by the local ethics committee. All the adolescents gave
SC
In a first step, we performed a bivariate analysis comparing the two groups of
adolescents (pregnant and control). We used the Chi-square test to compare categorical
M AN U
variables and the Student’s t test for continuous variables.
To provide a broader picture of the factors possibly associated with the group variable, all factors significant at the bivariate level were then introduced in a log-linear analysis18. This non-directional method allowed us to consider all possible associations between variables
TE D
including associations between potential factors. Preliminary computations showed that associations of order higher than two were mostly non-significant, so we included only bivariate relations and the main effect of each variable into the starting model. We performed
EP
then a selection of the associations between variables using the Bayesian Information Criterion (BIC)19, the goal being to keep only necessary relations. BIC allowed us to take into
AC C
account both the quality of fit and the complexity of the model, leading to a parsimonious final model. Results are presented as a diagram showing all remaining significant associations in the final model. Finally, all variables directly associated with the dependent variable were included in a logistic regression. The significance level was set at 0.05 for all analyses. R 2.12.2 (The R Foundation for Statistical Computing) was used for all computations. Results
ACCEPTED MANUSCRIPT Table 1 describes the comparison between groups. Overall, adolescents in the pregnancy group were significantly less likely to attend high school, and more likely to have low grades and a less educated mother. They were also less likely to live in an intact family, but no differences were observed for family income or
RI PT
relationship with the mother. Girls in the pregnant group were also significantly more likely to report other risk behaviors such as smoking or drinking strong alcohol, but interestingly pregnant girls were also more likely to not drink alcohol at all. No difference was found
SC
regarding age or the use of illegal drugs.
Girls with unwanted pregnancy were significantly more likely to have had an early
M AN U
sexual debut and reported more frequent sexual intercourse. However, there was no significant difference in the lifetime number of sexual partners between groups. Log-linear model
Figure 1 shows the structure of association of the final model. Six factors that were
TE D
significant at the bivariate level were found to be directly associated with unwanted pregnancy in the final model: mother’s educations, academic grades, alcohol consumption, age at sexual debut and frequency of sexual intercourse.
EP
Logistic regression
Table 2 presents the results. Among the 6 factors directly associated with the group
AC C
membership in the log-linear model, five remained significant in the logistic regression, the only exception being progress in studies. Living with only one parent or other relatives, a mother with an unknown or a secondary school level of education, a preference for low alcohol drinks (as compared to no alcohol consumption), an onset of sexual relations before age 15, and having sexual intercourse at least twice weekly were all factors associated with a higher probability of being in the pregnant group. It must be noted that the preference for strong alcohol drinks was not significant.
ACCEPTED MANUSCRIPT
Discussion Associated factors The study describes the main characteristics of adolescents experiencing unwanted
RI PT
pregnancy in Saint-Petersburg based on a self-administered questionnaire. The results show that almost the same types of the family and individual risk factors described in the literature influence the probability of pregnancy among adolescents in Saint-Petersburg.
SC
School status and academic grades of the adolescents were found to be associated with pregnancy. Positive relations between education level and general health are well known.
M AN U
There is good evidence that people with higher level of education have lower level of mortality, are less likely to suffer from acute and chronic diseases and are more likely to follow healthy life styles. These positive outcomes also include a lower probability of risky behaviors20. The power of education needs to be considered in improving the situation
TE D
regarding adolescent pregnancy.
Family settings were also shown to be associated with unwanted pregnancy. Living in a disrupted family with the absence of father may lead an adolescent girl to seek for
EP
interpersonal relations with males outside home. On the other hand a single parent in the family may have a lower probability to spend more time with an adolescent and have lower
AC C
monitoring over her activities and relations outside home. Contrary to some research 2, 10 we did not observe a significant difference in connectedness with their mother between the two groups, though there was a tendency for more complicated relations among adolescents who were pregnant. The majority of adolescent girls from both groups described their relations with mothers as achieving mutual understanding. This implies that good connectedness to mothers among adolescents does not necessarily reduce the possibility of unwanted pregnancy.
ACCEPTED MANUSCRIPT The relation between alcohol consumption and adolescent pregnancy was quite complex. If the consumption of strong alcohol drinks was clearly associated with a higher probability of being in the pregnant group, the same result was obtained for non-drinking girls. There is evidence that pregnancy is a strong motive for adolescents to stop smoking and
RI PT
drinking alcohol10. We hypothesize that a fraction of the pregnant girls did stop consuming alcohol after the beginning of their pregnancy for the sake of their baby.
There is evidence that adolescents exposed to alcohol are more likely to have sexual
SC
relations21, 22. In our research tobacco smoking was also a risk factor for unwanted pregnancy, but only indirectly as revealed by the log-linear analysis. We hypothesize that tobacco
M AN U
smoking can be generally associated with other types of risky behavior such as alcohol consumption, using illegal drugs and unprotected sexual intercourse but does not directly influence the behavior of adolescents including sexual behavior as it was shown for alcohol consumption. Contrary to the literature 10 we did not see a large difference in illegal drug use
TE D
among pregnant and non-pregnant adolescents. However some adolescents may not want to disclose this fact.
As shown in previous research 8, 11 sexual behavior is a strong factor influencing the
EP
teenage pregnancy like in many countries. In this respect adolescent girls with early sexual debut and high frequency of sexual intercourse may particularly benefit from the sexual
AC C
educational programmes.
There is an assumption that personal deprivation may be associated with living in a deprived area or community 23,24 where unstable employment and low aiming at education may influence personal goals of an adolescent. Family settings and attitudes may influence an adolescent even more. Mother’s education occurred to be another factor associated with adolescent pregnancy found in our research. We hypothesize that low educational level of
ACCEPTED MANUSCRIPT mothers may reflect a low value of education in families that influence the personal goals of adolescent girls. The results of our study highlight the importance of the educational level of mothers and children that were negatively associated with unwanted pregnancy among adolescents. In this
RI PT
respect, general educational level of the family and of the child should be taken into account that can be considered as an important tool to improve the situation Study limitations
SC
When interpreting the results of the study, a number of limitations should be taken into account. First of all, we performed a cross-sectional study, which cannot provide evidence of
M AN U
causal relationship. Second, there were some gaps in the data: a few adolescents did not have information about their mothers. These data were put into the final comparison table as unknown and the comparison of the two groups was made including this category. Third, a clinical sample was used and it may not represent all pregnant adolescents, even though it was
TE D
taken from the largest youth clinic of Saint-Petersburg which is also the largest center for abortions among adolescents. Finally, due to the relatively small size of the sample we may have a low statistical power.
EP
Implications of the study
Although the issue of unwanted pregnancy and possible ways to prevent it should be
AC C
kept in mind in the psychosocial assessment of any adolescent, the results of our study suggest that some adolescents are at increased risk of unwanted pregnancy. This group includes adolescent girls with low educational attainment, living in a disrupted family and reporting other risk behaviors. These adolescents may specially benefit from specific preventive programs in order to reduce the number of adverse outcomes.
ACCEPTED MANUSCRIPT
References 1. Coyne CA, D'Onofrio BM: Some (but not much) progress toward understanding teenage
42:113-152
RI PT
childbearing: a review of research from the past decade. Adv Child Dev Behav 2012;
2. Santelli JS, Melnikas AJ: Teen fertility in transition: recent and historic trends in the United States. Annu Rev Public Health 2010; 31:371-83
SC
3. Santos MI, Rosário F: A score for assessing the risk of first-time adolescent pregnancy. Fam Pract 2011; 28(5):482-8
M AN U
4. Avery L, Lazdane G: What do we know about sexual and reproductive health of adolescents in Europe? Eur J Contracept Reprod Health Care 2010; 15 Suppl 2:S54-66 5. Singh S, Darroch JE: Adolescent pregnancy and childbearing: levels and trends in developed countries. Fam Plann Perspect 2000; 32(1):14-23
TE D
6. The TransMonEE (Transformative Monitoring for Enhanced Equity) database – Monitoring Situation of the Women and Children in countries of Central Eastern Europe and Commonwealth of independent States (CEECIS). UNICEF Regional Office for
EP
CEECIS countries. Available at: http://www.transmonee.org/ (accessed 15.03.2015). 7. Eurostat database: http://ec.europa.eu/eurostat/data/database/ (accessed 15.03.2015)
AC C
8. Imamura M, Tucker J, Hannaford P, et al: Factors associated with teenage pregnancy in the European Union countries: a systematic review. Eur J Public Health 2007; 17(6):630-6.
9. Xie H, Cairns BD, Cairns RB: Predicting Teen Motherhood and Teen Fatherhood: Individual Characteristics and Peer Affiliations. Soc Dev 2001; 10: 488–511 10. Black AY, Fleming NA, Rome ES: Pregnancy in adolescents. Adolesc Med State Art Rev 2012; 23(1):123-38
ACCEPTED MANUSCRIPT 11. Mmari K, Blum RW: Risk and protective factors that affect adolescent reproductive health in developing countries: a structured literature review. Glob Public Health 2009; 4(4):350-66 12. Miller BC: Family influences on adolescent sexual and contraceptive behavior. J Sex
RI PT
Res 2002; 39(1):22-6
13. Lion KC, Prata N, Stewart C: Adolescent childbearing in Nicaragua: a quantitative assessment of associated factors. Int Perspect Sex Reprod Health 2009; 35(2):91-6
SC
14. Planned Parenthood Federation of America, Inc. "Pregnancy and Childbearing Among U.S. Teens." PlannedParenthood.org. (accessed 13.03.2015)
M AN U
http://www.plannedparenthood.org/files/PPFA/pregnancy_and_childbearing.pdf. 15. De Almeida MC, Aquino EM: The role of educational level in the intergenerational pattern of adolescent pregnancy in Brazil. Int Perspect Sex Reprod Health 2009;13:139– 146
TE D
16. Nizamov IG, Chechulina OV: A sociological study of factors affecting reproductive health of female teenagers and young women. [Article in Russian]. Probl Sotsialnoi Gig Zdravookhranenniiai Istor Med 2003; (3):10-2
EP
17. Perminova SG, Sotnikova EI: Sexual and reproductive behavior as seen in the results of a randomized study [Article in Russian]. Akush Ginekol (Mosk) 1993; (3):23-6.
AC C
18. Agresti A: Categorical Data Analysis, (2nd ed), 2002, Wiley-Interscience 19. Kass RE, Raftery AE: Bayes Factors. J Am Stat Assoc 1995; 90(430): 773-795 20. Cutler DM, Lleras-Muney A: Education and Health: Evaluating Theories and Evidence. NBER Working Paper No. 12352. Cambridge, MA: National Bureau of Economic Research, 2006 21. Bellis MA, Morleo M, Tocque K, et al: Contributions of alcohol use to teenage pregnancy: An initial examination of geographical and evidence based associations.
ACCEPTED MANUSCRIPT North West Public Health Observatory, Centre for Public Health, Liverpool John Moores University, Liverpool. (2009). (accessed 09.03.2015): http://www.cph.org.uk/showPublication.aspx?pubid=579 22. Cook PA, Harkins C, Morleo M, et al: Contributions of Alcohol Use to Teenage
RI PT
Pregnancy and Sexually Transmitted Infection Rates. North West Public Health
Observatory, Centre for Public Health, Liverpool John Moores University, Liverpool, 2010. (accessed 09.03.2015): http://www.cph.org.uk/showPublication.aspx?pubid=709
Chicago: University of Chicago Press, 1987
SC
23. Wilson WJ: The truly disadvantaged: The inner city, the underclass and public policy.
M AN U
24. Smith DM, Elander J: Effects of area and family deprivation on risk factors for teenage
AC C
EP
TE D
pregnancy among 13 – 15-year-old girls. Psychol Health Med 2006; 11(4): 399-410
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Table 1 Comparison between groups.
ACCEPTED MANUSCRIPT
Figure 1 Structure of the final log-linear model. Variables directly associated with the
AC C
EP
TE D
M AN U
SC
RI PT
Group variable appear in bold.
ACCEPTED MANUSCRIPT Table 2 Logistic regression model corresponding to the log-linear analysis for predicting the
AC C
EP
TE D
M AN U
SC
RI PT
probability of being in the pregnant group.
ACCEPTED MANUSCRIPT
Variable
Categories
Age (mean ± SE) School status
Pregnant %
Control %
(n=145)
(n=77)
17.12 ± 0.09
17.13 ± 0.14
0.943
5.5%
0.0%
<0.001
35.2%
26.0%
Working
school High school Low
Family type
Living with both parents
Mother’s education
Unknown
mother
13.1%
36.4% 27.3%
0.011
36.1%
67.5%
<0.001
6.9%
1.3%
0.001
Secondary school
66.2%
48.1%
High school
26.9%
50.6%
6.9%
2.6%
78.6%
88.3%
14.5%
9.1%
4.9%
6.5%
0.610
Unknown
TE D
Relation with the
37.7%
44.8%
M AN U
Academic degree
46.2%
SC
Secondary professional
RI PT
Mandatory school
p
Achieving mutual
0.178
EP
understanding Contentious
AC C
Family income
Low
Tobacco smoking
Smoke regularly
82.5%
46.8%
0.024
Alcohol consumption
No consumption
24.1%
5.2%
<0.001
Soft alcohol drinks (beer,
30.3%
68.8%
Strong alcohol drinks
45.5%
26.0%
Yes
15.3%
21.6%
wine)
Ever used illegal drugs
0.242
ACCEPTED MANUSCRIPT
Age at sexual debut
Before 15
37.8%
11.7%
<0.001
Number of sexual
More than 2
39.2%
34.7%
0.515
More than once a week
66.2%
38.7%
<0.001
partners (lifetime) Frequency of sexual
AC C
EP
TE D
M AN U
SC
RI PT
intercourse
ACCEPTED MANUSCRIPT
Academic degree
Family type
Odds ratio
95% CI
p
Low
1.31
0.61 – 2.86
0.494
Average and above
REF
Living with both parents
REF
Living with one parent or
2.85
1.38 - 6.07
0.005
9.42
1.41 – 189.34
0.049
4.59
2.09 – 10.70
<0.001
0.06
0.01 – 0.20
<0.001
Strong alcohol drinks
0.38
0.09 – 1.29
0.144
Before 15
5.66
2.27 – 15.81
<0.001
15 or after
REF
Once a week or less
REF
More than once a week
3.94
1.87 – 8.70
<0.001
other relatives Mother’s education
Unknown Secondary school High school No consumption
consumption
M AN U
Preferences in alcohol
Low alcohol drinks (beer,
Age at sexual debut
TE D
wine)
AC C
intercourse
EP
Frequency of sexual
RI PT
Categories
SC
Variable
REF
REF
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT