Current status of adolescent girls' reproductive health in Almaty

According to the United Nations Committee on the Elimination of Discrimination against Women (2018), the Kazakhstan population of children and adolescents aged 0-18 years is estimated to be 5.6 million, including 2.7 million girls [1]. Considerable research attention has been devoted to various aspects of adolescent girls’ reproductive health. N.A. Buralkina, E.V. Uvarova (2010), A.A. Baranov et al. (2010, 2014), G. B. Dicke and A.V. Erofeyeva (2013) showed that modern girls have a high risk of developing gynecological diseases and are prone to changes in reproductive and sexual behavior [2–5]. Z.B. Makhkanova et al. (2016), T.A. Nangus et al. (2016), N.V. Artymuk, N.A. Apikhtin (2017) reported a high prevalence of inflammatory diseases of external genital organs and menstrual cycle disorders among adolescents’ gynecological diseases [6–8]. E.A. Mateykovich (2018) showed that menstrual disorders are a leading cause of adolescents’ gynecological diseases (60.7%), including dysmenorrhea (19.4%), oligomenorrhea (13.6%), irregular menstrual cycles (10.3%), amenorrhea (8.7%), hyperpolymenorrhea (6.2%), abnormal uterine bleeding (AUB) during puberty (2.5%) [9]. Kazakhstan researchers U.A. Abdukayumova et al. (2011) reported that the number of unintended pregnancies among 15–17-year-old girls in the Republic exceed 8,000, and the number of registered abortions in this population is more than 4,000 [10].

Behavior indicators that negatively affect the formation of girls’ reproductive health also include early onset of sexual activity, HIV infection, sexually transmitted diseases, use of tobacco, alcohol, drugs, malnutrition, physical inactivity, relationships in the family, with peers, and stress factors [11–16].

This study aimed to estimate the burden of gynecological diseases and determine the impact of behavioral indicators on adolescent girls’ reproductive health in Almaty using logistic regression.

Materials and Methods

The study comprised two cohorts of female adolescents: 1) girls aged 10 - 17 years, who sought medical care at the Almaty City Center for Human Reproduction from 2013 to 2017; 2) schoolgirls aged 11, 13 and 15 interviewed in September-October 2017 with the HBSC International Program questionnaire.

The prevalence and variety of gynecological diseases were estimated using absolute numbers and rates (per 1000 girls). To determine the trend of the incidence of gynecological morbidity in adolescent girls, the time series were leveled using quadratic approximation, the coefficient of determination (R-Squared) and its error were calculated.

To identify behavioral indicators that form the reproductive potential of school-age girls, a social survey was conducted with the standard HBSC International Questionnaire with specified age intervals (11, 13, 15 years) [17]. A total of 667 girls were interviewed in the single-step cross-sectional study including girls aged 11 (n=173; 25.9%), 13 (n=172; 25.8), and 15 (n= 322; 48.3%). Reproductive behavior of the studied population was assessed using 23 behavioral indicators included in the questionnaire. The cohort was formed by random sampling from 8 secondary schools in Almaty. The sampling error was 5%. The main limitations in the study participants selection were as follows: 1) girls whose parents did not sign informed consent were not interviewed; 2) after discussion with the parents and the administration of the schools, questions regarding sexual behavior, drug and alcohol use, smoking and domestic violence were excluded from the questionnaire; 3) when extrapolating to the whole country, it was taken into account that the regions of Kazakhstan differ in their socio-economic development, as well as moral, spiritual, and religious attitudes toward the formation of young people’s life. The study was approved by the Local Ethics Committee of Kazakhstan Medical University KSPH, protocol number IRB - А084 dated August 31, 2017.

Statistical analysis

The hypothesis of a possible relation between behavioral indicators and the reproductive health of school-age girls was tested using binary logistic regression. Logistic regression was used to eliminate the influence of interfering factors (confounders) that may influence the results of the analysis and to take into account the mutual influence of all factors included in the analysis, which will allow evaluating the direct relationship with the resulting variable [18, 19]. In the model, the menstrual dysfunction of adolescent girls was the dependent variable, which, according to most researchers, makes up one of the largest proportions among gynecological diseases of the studied population and is an integral indicator of the reproductive system [5, 8, 9, 11].

Independent variables were participants’ responses acquired by a questionnaire survey, the so-called behavioral indicators. A quantitative measure of the impact of the determinants of behavior in binary logistic regression was the odds ratio (OR). Uncorrected values ​​of the OR were calculated by univariate regression, which was performed separately for each independent predictor, during which the independent influence of each factor under study was evaluated separately, without taking into account the mutual influence of the variables. The corrected OR values, on the contrary, taking into account the mutual influence of the studied factors, were calculated as a result of multiple regression analysis, where all independent variables were entered using forced-entry procedure. The 95% confidence intervals (CI) for the rates of negative and positive assessments of the quality of medical care were calculated using the Wilson method [19]. In forming the ideas, strategy, program, and the study design, we were guided by the principles and methodology of scientific research, outlined by E. DePoy and L.N. Gitlin (2016) [20]. Statistical analysis was performed using the SPSS version 22.0 and Statistica version 6.0 software packages.

Results

Evaluation of the prevalence of gynecological diseases in girls aged 10-17 in Almaty showed that AUB in puberty (32.0%; 95% CI 29.06; 34.94), pelvic inflammatory diseases (15.5%; 95% CI 13.15; 17.85) and dysmenorrhea (9.9%; 95% CI 7.94; 11.86) were the most common among different gynecological diseases. Moreover, gynecological diseases in girls included inflammatory diseases of the external genital organs (vulvovaginitis, vulvitis). The main causes of these diseases were a chronic focal infection, ENT diseases, urinary tract diseases, allergic disorders, and poor personal hygiene habits. The leading causes of the menstrual cycle disorders were AUB in puberty and dysmenorrhea. There were considerable fluctuations in incidence rates of AUB in puberty and dysmenorrhea in different years. For AUB puberty, there was a marked decrease in 2016, then a rise in 2017; the R2 coefficient of determination was 0.86, error was 7.78, growth rate was 172.34. The incidence rate of dysmenorrhea was stable in 2013–2015, then showed a sharp rise in 2016 and decline in 2017; the R2 coefficient of determination was 0.68, error was 25.58, growth rate was 55.56. The alignment of the dynamic series of the above-mentioned gynecological diseases revealed a tendency of the indicators to increase and decrease in the parabolic trend forms.

At the next stage, a binary logistic regression analysis was performed to estimate the impact of behavioral indicators on the girls’ reproductive health in the age groups of 11, 13 and 15 years. Of the 667 adolescent girls surveyed, 300 participants, who reported regular and irregular menstrual cycles (dependent variable), were included in the study (Table 1).

Table 2 presents the results of a logistic regression.

The results of regression analysis suggest that frequent (2-4 times a week) consumption of sweets, frequent bullying by peers, and complicated relationship with friends are associated with a 4.24, [95% CI (1, 57-11.44)], 3.52, [95% CI (1.37-9.05)], and 2.41[95% CI (1.07–5.42)] fold increased chance of having menstrual disorders, respectively. No other behavioral factors were found to be significantly associated with menstrual disorders.

Discussion

The findings of our study demonstrated that annual incidence rates of gynecological diseases among girls aged 10-17 in Almaty did not have a clear tendency to decrease and remained at a stable level. AUB in puberty, pelvic inflammatory diseases and dysmenorrhea were the most common gynecological diseases in girls. This may adversely affect the girls’ reproductive function and result in a significant medical and social issue for the public and health care in Almaty. We believe that this category of the population requires closer attention and special strategy on the part of the city health authorities and institutions, pediatric, adolescent and adult gynecologists to prepare the adolescent girls for their future roles as mothers. This issue of adolescent reproductive health was raised by the World Health Organization (2014), which stated that the health and well-being of children and adolescents should be given more attention [21].

Logistic regression modeling identified quantitative behavioral indicators associated with menstrual disorders in school-aged girls including frequent consumption of sweets (OR - 4.24), bullying by peers (OR - 3.52) and complicated relationship with friends (OR - 2.41).

These results are consistent with domestic and international literature addressing the problem of eating behavior among children and adolescents, issues of interpersonal relations, humiliation, violence, and bullying and cyber-bullying. Thus, according to the HBSC study report (2016), half of the schoolchildren aged 13-15, who constitute 150 million people, are characterized by sweets consumption and bullying by peers at and outside the school [17]. The most common behavioral indicator in school-age girls is the consumption of sweets and sugar-sweetened beverages. I. Huybrechts et al. (2017) also reported that European and Australian adolescents are characterized by high consumption of cakes, chocolate, and sugar-sweetened drinks [22].

It is of interest that M. Kherkheulidze et al. (2012) observed that among school children in Tbilisi (Georgia) daily consumption of sweets and snacks was significantly higher among girls (64.2%) than among boys (47.5%) [23]. One of the main trends is interpersonal relationships among transitional aged youth. The need to communicate with peers who cannot be replaced by parents arises in children very early and increases with age. From an early age, boys are prone to extensive communications, and the girls are to intensive communication; boys most often play in large groups, and girls prefer two or three peers. Worldwide, as estimated by UNICEF (2013), every third teenager aged 13–15 is bullied and engaged in physical fights [24]. Violence, extortion, insults, bullying, sexual abuse, cyber-bullying are serious problems among adolescents at school. E.N. Volkova, I.V. Volkova (2017) showed that every second teenager is a victim of cyber-bullying in the Russian school [25]. Similar data for China are reported by Chu XWet. al (2018) [26]. The problem of school violence must be addressed collectively using a coordinated approach with the participation of school representatives, children and their parents.

Conclusion

The prevalence of gynecological diseases among girls aged 10-17 and a range of risky behaviors among schoolgirls aged 11, 13, 15 in Almaty indicate the advisability of introducing a multifactorial approach to adolescent reproductive health services aimed at correcting several risk factors. The binary logistic regression allows us to quantify the significant behavioral indicators that affect the reproductive potential of school-age girls.