Assessment of factors related to poly cystic ovarian syndrome – A comparative and correlational study

Abstract

Polycystic ovarian syndrome (PCOS) is a common endocrine disorder that primarily affects women of reproductive age. It is particularly prevalent among adolescent females who receive an insufficient diagnosis despite having potentially adverse consequences. The use of PCOS screening questionnaires has the potential to aid in the early detection of symptoms. The goal of this study is to observe if a self-administered questionnaire may be useful for a clear cognizance of the associated conditions like mental stress and menstrual characteristics correlated to polycystic ovary syndrome. In this study, we selected women within an age group of 17–40 with and without PCOS based on the modified Rotterdam criteria to fill out a self-administrated questionnaire based on the signs and symptoms of PCOS majorly focusing on mental stress and menstrual characteristics. SPSS software, univariate analyses were employed to elucidate the associations among the components of PCOS, demographic factors, and lifestyle characteristics, hence providing insights into the interrelationships among those variables. 64 women with PCOS and 141 women without PCOS participated in the present study. The present study revealed PCOS is greatly influenced by age at menarche (p-value= .043), typical cycle length (p-value = .000) mental health problems during menstruation (p-value = .032), and body mass index (p-value = .001). Multivariate hierarchical logistic regression analysis showed only 2 variables BMI (a-OR 1.156,95% CI (1.067–1.242), p-value = .000), and typical cycle length (a-OR 2.278, 95% CI (1.079–4.809), p-value = .003) were significant. The present study showed that BMI and menstrual cycle length were most closely associated with the incidence of PCOS, which is important in diagnosing and treating the condition. Considering the high incidence of PCOS among women of reproductive age and its potential for significant health implications, it would be prudent to incorporate inquiries regarding mental health concerns and menstrual patterns into routine medical assessments for this demographic analysis. This approach aims to ascertain whether additional diagnostic evaluations and screenings for PCOS are warranted.

Keywords:

• Binary logistic regression
• polycystic ovarian syndrome
• menstruation
• mental stress

Introduction

Polycystic ovarian syndrome (PCOS) is a prevalent endocrinopathy linked to obesity and infertility. The occurrence of infertility is about 50% in PCOS women, while obesity rates range from 30 to 60% [1]. It affects 15–20% of women of reproductive age based on demographics and diagnosis. The pathophysiology of this complex disease remains unclear, but complex interactions between genetic, metabolic, and environmental factors are assumed to be its root cause. It is characterized by higher levels of androgens, trouble ovulating, and the shape of the ovaries with many cysts. PCOS is linked to health problems like irregular periods and infertility, as well as metabolic problems like insulin resistance, diabetes, the risk of heart disease, and mental problems [2]. According to Rotterdam’s criteria, it has been estimated that PCOS prevalence in India is 11.34% [3].

Apart from infertility, menstrual dysfunctions, hirsutism, and obesity are all the major symptoms of PCOS, and they are all associated with increased levels of psychological stress [4]. Several studies found that PCOS women are more likely to experience emotional problems than women who do not have PCOS. They are at a much higher risk for developing depressive behavior (as described by the Diagnostic and Statistical Manual IV) [5–13]. Research that has been done on PCOS women found that they were more likely to suffer from depression and other challenges related to their emotional state. Prevalence estimates for depression ranged from 35% (N = 35) and 40% (N = 60), respectively, as compared to a control group estimate of 10% [14]. In another study, the Beck Depression Inventory (BDI) [15] showed that 23.9% and 25.2% of women with PCOS scored in the mild-to-moderate and clinically relevant ranges of sadness, respectively. The average BDI score for this group was 12.7, much higher than the results for a normative sample [16]. Women with PCOS are also more likely to withdraw from social life than women without PCOS. According to previous studies, the higher lifetime incidence of depressive episodes, social phobia, eating disorders, and suicide attempts was seven times more common in the PCOS group than in the control group [17]. Depression and anxiety often coexist in women, and research suggests that women with PCOS may be at a higher risk for clinically severe anxiety. Depression and anxiety both amplify the negative effects of PCOS on quality of life [18].

Women with PCOS often have heightened insulin resistance, inflammation, endocrine disorders, and depression which are linked and cause obesity and infertility [7]. Most PCOS patients have at least mild psychological distress, and the combination of emotional sadness and obesity significantly reduces their quality of life [19]. A study conducted in the Iranian population found that hirsutism, one of the major symptoms of PCOS, has the greatest effect on quality-of-life metrics related to the health of the PCOS patients [20]. Studies showed that changes in appearance, notably obesity, and hirsutism, can lower a person’s quality of life when they have PCOS [21]. Also, Oligo/amenorrhea, characterized by infrequent or absent menstrual cycles, represents the prevailing form of irregular menstrual patterns, and could potentially serve as a diagnostic marker for PCOS [21]. Women with PCOS and a high body mass index had the worst modified Ferriman Gallwey (mFG) scores. In this, the nine androgen-sensitive areas evaluated by the mFG score are the upper lip, jawline, chest, upper and lower back, upper and lower abdomen, upper arm, and thigh. An increase in mFG scores may hurt the self-esteem of women with PCOS [22]. Stress markers in the saliva and measures of women’s internal emotional state were analyzed in a study of PCOS. In this study, they found that there was no difference in stress levels as measured by salivary biomarkers between PCOS women and healthy controls, but that stress levels as judged by questionnaires were significantly higher in women with PCOS. The main moderators of subjective stress in PCOS appear to be body mass index, hirsutism, and age [23].

Thus, given the high prevalence of mental stress among PCOS patients, our objective is to evaluate the self-reported symptoms associated with PCOS and most importantly the mental stress-related symptoms, lifestyle habits, and menstrual characteristics that are correlated with PCOS through hierarchical binary logistic regression analysis.

Materials and methods

Study design and participants

The participants were selected based on a questionnaire circulated among female students and staff members from VIT University, Vellore, India. Female candidates with and without PCOS and between the age of 17–40 were enrolled in the present study. The classification of the respondents into control women and PCOS women was by the modified Rotterdam criteria [24] including any two of the following three features, (a) oligo-ovulation or anovulation, (b) clinical or biochemical signs of hyperandrogenism, (c) polycystic ovaries on the ultrasound. The control group had regular menstrual periods with no clinical or biochemical hyperandrogenism or history of PCOS or any other gynecological conditions.

Inclusion criteria: Women diagnosed with PCOS through ultrasound and hormonal tests by registered medical practitioners were considered for inclusion criteria. Also, participants undergoing any kind of treatment for PCOS conditions are considered. The age group of women considered for the survey was between 17 and 40 years.

Exclusion criteria: those who got pregnant during the study, and those who could not submit medical records are excluded from the analysis in the PCOS group. Women who had PCOS before the period of study are excluded since the study is checking the current BMI status, mental stress-related issues, and menstrual characteristics.

Ethical consideration: The data were obtained from VIT University, Vellore, India via legitimate methods, and with sufficient authorization. Informed consent was taken from each participant taking part in the study. There is no risk to participants and the dataset used is duly acknowledged and cited wherever needed.

Survey

The participants were requested to complete a questionnaire designed by the authors, based on some previous comparable studies [25,26]. The questionnaire was created using the Google Docs platform, and a hyperlink was disseminated to the participants for their access. The questionnaire included 22 questions related to information about the demographic variables like their age, height, weight, ethnicity, marital status, and current BMI (Body mass index) if available. The responses are collected and stored in a Microsoft Excel sheet.

Menstrual characteristics evaluation

Participants were asked questions concerning their menstrual patterns like, age at menarche, cycle length in days, flow duration in days, and menstrual regularity. Questions related to “Typical bleeding length” were with answers ranging from “1 to 7 d” and “ symptoms felt during periods”. Cycle length was examined by asking, “What was the average number of days from the start of one period to the start of another?” and the answers ranged from “28 to 40 d”. The type of menstrual blood flow was examined by responding to answers ranging from “mild to heavy”.

Evaluation of mental issues during periods

Questions were asked of the respondents describing their mental state during their periods. The questions answered were whether they had feelings of depression, anxiety, sleep problems or poor sleep quality, eating disorders, poor body image, or social phobia during their periods.

Evaluation of lifestyle-related factors in normal and PCOS women

Questions related to food habits, fast food consumption, and lifestyle habits like smoking, alcohol, drug usage, and exercise were asked of the respondents.

Statistical analysis

All analyses were conducted using SPSS software, version 22 (SPSS Inc., Chicago, IL, USA). A p-value ≤ .05 is considered statistically significant for all the statistical tests. Assumptions of normality were confirmed. The categorical variables are presented as counts (percentages). We compared the baseline characteristics among women with and without PCOS using the chi-square test for categorical variables. We first calculated the crude association, expressed as B or OR. We subsequently adjusted for variables that showed statistically significant differences between the groups. We performed further analysis by adjusting these variables to determine which variable affects the most. To find out the separate and simultaneous effects of various factors including menstrual characteristics on polycystic ovary syndrome as the dependent variable, with other predictors of PCOS, a logistic regression with backward stepwise regression analysis was used and the variables that remained in the final model were presented. Odd’s ratio (OR), is calculated to check the association between the variables, with corresponding 95% confidence intervals, which were used to estimate the precision of the OR.

As more variables are analyzed, regression models can get more complex. As a result, we decided to utilize a hierarchical logistic regression (HLR) methodology to determine which factors in our comprehensive model significantly and statistically contributed to PCOS. Hierarchical models incorporate the variability seen at each level of the hierarchy, enabling the examination of cluster effects at distinct levels within the model [27]. A multilevel analysis was done using a multivariate HLR model to analyze the impact of important variables on PCOS. The model with the lowest Akaike Information Criterion (AIC) score gave the best results. After considering all other variables included in the model, we identified the variables that significantly contribute to the explained variance in PCOS by using hierarchical regression.

Results

Based on the questionnaire among the 205 responses, 64 candidates were diagnosed with PCOS as defined by the inclusion criteria. The remaining 141 candidates were women who never had any symptoms associated with PCOS or other gynecological conditions. Briefly, 31.2% of the individuals were diagnosed with PCOS. Participants had an average age of 26.85 ± 6.2 years, a mean BMI of 23.8 ± 4.7, and a mean age of menarche of 13.09 ± 1.7. Table 1 shows a univariate analysis of the connection between demographic variables.

Table 1. Comparison between the socio-demographic status of PCOS and healthy controls.

Variables		N (%)	Non-PCOS	PCOS	p-value
Age			27.07(6.8)	26.3(4.5)	.926
BMI			22.7(4.01)	26.36(4.13)	.001
Marital status	Single	141(68.8)	92	49	.105
	Married	64(31.2)	49	15

Table 2 shows the connection between PCOS and the non-PCOS (control) group based on menstrual characteristics and mental issues using the Chi-square test.

Table 2. Comparison between the menstrual characteristics status and mental health issues of PCOS and healthy controls.

Variables		N (%)	Non-PCOS (%)	PCOS (%)	p-value
Age of Menarche			13.11(1.3)	13.06(2.3)	.043
Typical Cycle Length	<21 d	3(1.5)	3(2.1)	0(0.0)	.000
	<35 d	51(24.9)	31(22)	20(31.3)
	>35 d	16(7.8)	2(1.4)	14(21.9)
	28–30 d	135(65.9)	105(74.5)	30(46.9)
Typical Bleeding Length	>7 d	2(1.0)	2(1.4)	0(0.0)	.205
	1–3 d	79(38.5)	59(41.8)	20(31.3)
	2–7 d	28(13.7)	16(11.3)	12(18.8)
	2–5 d	96(46.8)	64(45.4)	32(50.0)
Typical Menstrual Flow	Heavy	9(4.4)	5(3.5)	4(6.3)	.143
	Mild	20(9.8)	12(8.5)	8(12.5)
	Moderate	114(55.6)	86(61.0)	28(43.8)
	Moderately Heavy	62(30.2)	38(27.0)	24(37.5)
Mental Health Issues during periods	Anxiety	27	22(81.5)	5(18.5)	.032
	Depression	17	14(82.4)	3(17.6)
	Sleep Disturbances	16	11(68.8)	5(31.3)
	Nil	26	18(69.2)	8(30.8)

In Table 3 univariate analysis was used to find the connection between PCOS and the control group based on lifestyle characteristics using the Chi-square test.

Table 3. Univariate analyses are used to show the connection between PCOS and the control group based on lifestyle characteristics using the Chi-square test.

Variables		N (%)	Non-PCOS	PCOS	p-value
Food Consumption	Vegetarian	66(32.2)	24(37.5)	42(29.8)	.273
	Non-vegetarian	139(67.8)	40(62.5)	99(70.2)	.273
Weekly fast-food consumption	1–2 times	131(63.9)	42(65.6)	89(63.1)	.953
	3–4 times	28(13.7)	9(14.1)	19(13.5)
	5–6 times	5(2.4)	1(1.6)	4(2.8)
	>7 times	9(4.4)	2(3.1)	7(5.0)
	NA	32(15.6)	10(15.6)	22(15.6)
Monthly Exercise Habit	Every day	29(14.1)	11(17.2)	18(12.8)	.757
	Every Other day	34(16.6)	10(15.6)	24(17.0)
	Less often	111(54.1)	34(53.1)	77(54.6)
	Never	4(2.0)	0(0.0)	4(2.8)
	Once per week	13(6.3)	4(6.3)	9(6.4)
	Several times per month	14(6.8)	5(7.8)	9(6.4)

Odds ratios are also computed for the two variables of marital status and food intake type, displayed in Table 4.

Table 4. The calculated odds ratio for marital status and food intake type.

Variable	Value	95% confidence interval
		Lower	Upper
Odds Ratio for marital status (Married/Single)	1.740	0.887	3.415
Odds Ratio for Type of food consumption (Non-vegetarian/Vegetarian)	1.414	0.760	2.633

Among the variables, typical cycle length, age of menarche, mental health issues during the periods, and BMI have the most significant impact on the incidence of PCOS. Only four types of mental health concerns are included in Table 2 as having a significant number. They are depression, anxiety, and sleep disturbances. Around 13% of the respondents have not felt any mental health concerns. Results for Hierarchical multivariate logistic regression analysis are given in Table 5.

Table 5. Factors associated with PCOS using hierarchical multivariate logistic regression (n = 205).

Characteristics	Step 1		Step 2
	Adjusted OR (95% CI)	p-value	Adjusted OR (95% CI)	p-value
Age	NS		NS
BMI	1.156(1.067–1.242)	.000	1.160(1.072–1.244)	.000
Marital status	NS		NS
Age of Menarche	NS		NS
Typical Cycle Length	2.278(1.079–4.809)	.003	2.225(1.066–4.641)	.033
Typical Bleeding Length	NS		NS
Typical Menstrual Flow	NS		NS
Mental Health Issues	0.452(0.125–1.631)	.659	NS
Food consumption (Veg/Non veg)	NS		NS
Monthly Exercise Habits	NS		NS
AIC value	303.004		310.406

Discussion

In our study, the overall prevalence of anxiety conditions in PCOS women was 18.5%, depression was 17.6% and sleep disturbances were 31.3%. Consistent with previous studies done on PCOS and mental issues [9, 22, 28–34], our study found that the incidence of PCOS is primarily influenced by factors such as the presence of mental health issues during menstruation, the typical cycle length, and the individual’s body mass index (BMI). We could not find any significant correlation with variables like lifestyle factors like food consumption, exercise habits, age, and marital status. The other menstrual characteristics like, age at menarche, typical bleeding length, and typical menstrual flow did not show any significant correlation in PCOS women compared to non-PCOS women.

In our study, the respondents were asked to answer about their mental state during periods, and among that, most of the PCOS people were found to have mental health-related problems like anxiety, depression, and sleep disturbances. A systematic review and meta-analysis on anxiety and depression in PCOS concluded that women with PCOS have modestly elevated anxiety and depression, which is consistent with the findings of the current study [11,12, 35]. The causes of the increased prevalence of depression and anxiety in PCOS are complicated. Emotional distress in women with PCOS may have psychological and/or biological origins [36]. The clinical characteristics of PCOS, including obesity, insulin resistance, hyperandrogenism, inflammation, and infertility, may be related to the mechanism driving the increasing prevalence of mental stress in PCOS [28]. However, recent studies have focused more on the shared clinical aspects of PCOS and mental stress compared to the underlying molecular mechanisms that contribute to psychological symptoms in PCOS [9, 29]. Many researchers have proposed molecular reasons linking PCOS symptoms to anxiety and depression, including hyperandrogenism and insulin resistance, but the results have not been consistent [12, 37].

The exact effect of sleep disruptions on the health of women with PCOS is unknown; nevertheless, both PCOS and sleep disturbances have been linked to long-term declines in cardiometabolic health and an increased risk of type 2 diabetes [38]. Compared to healthy women PCOS women were more likely to experience difficulty getting sleep and experience severe tiredness throughout the day [39]. Studies found that poor sleep can lead to increased stress, and contribute to a depressed mood state. Also, been associated with increased cholesterol levels, high blood pressure, and weight gain [40,41]. The findings from the current study also strengthen the previous research studies.

The BMI of the PCOS women was found to be 26.36 which was higher compared to normal women. Studies conducted on the effect of an increase in the body mass index of PCOS patients have found that increased body weight may influence the self-esteem of PCOS patients [22, 30]. The present study found that, among PCOS people BMI is significant and is correlated with the incidence of PCOS and may have an influence on the self-confidence of PCOS women. PCOS is correlated with a wide range of adverse health consequences in women [42–51]. Women with PCOS not only have worse health outcomes but also have inferior physical and mental quality of life indicators compared to age-matched controls [52,53]. The decrease in the mean BMI and age at menarche indicated the presence of PCOS symptoms. Women with menstrual irregularities were more likely to be overweight and to have a larger waist-to-hip ratio than PCOS patients with normal menstrual cycles. Oligo/amenorrhea was the most prevalent form of an abnormal menstrual cycle and may be indicative of PCOS. In PCOS patients menstrual cycle disruption was associated with greater levels of IR, the LH/FSH ratio, and androgens. Overall, PCOS patients have a higher incidence of infertility, which has direct effects on their physical, social, and emotional quality of life [32,33, 54]. The evaluation of the quality of life could therefore contribute vital information to evaluating treatment efficacy in clinical trials and natural history studies of PCOS.

The association between menstrual cycle irregularities and PCOS was found in several studies. Yan-min Ma et.al identified that oligo amenorrhea is a common menstrual difficulty found in PCOS women [32]. In another study, the severity of oligomenorrhea and IR is positively correlated with vaginal bleeding interval [33]. Also, there is a positive correlation between the timing of menarche and PCOS and diminished ovarian reserve has been identified [34]. Compared to control subjects, girls with PCOS might experience menarche at a considerably younger age, with examples being early menarche at or before the age of nine, primary amenorrhea, in which menarche did not occur by the age of sixteen, or four years after the commencement of thelarche [55]. Consistent with this evidence, the age at menarche and typical cycle length characteristics showed a significant effect on PCOS women as compared to the control population.

In detail, factors were incorporated into the HLR model in two stages, using backward stepwise binary logistic regression. This technique starts with a fully saturated model and progressively eliminates variables to arrive at the best-fitting reduced model at each stage [56]. To make sure that all the odd/risk factors discussed already take into consideration the influencing factor or bias in our variables, the adjusted odds ratio is employed in the results. As illustrated in Table 5, in step 2 of multivariate hierarchical logistic regression only 2 variables are significant, which are BMI and typical cycle length. But Step 1 is the model with the lowest AIC value and is the most suitable of our models, as shown in Table 5. This included BMI, a menstrual feature - the normal cycle length - and mental health difficulties during periods as PCOS predictors.

From the whole study, it is found that implementing a PCOS screening questionnaire that includes questions about various associated symptoms may aid in epidemiologic studies of PCOS [57–59]. It is found that asking women about their mental state and menstruation-related characteristics via a self-administrated questionnaire has high sensitivity and specificity in predicting the relationship between mental issues and PCOS. The previous studies that have already been done on the relationship between BMI and PCOS have found that there is a significant correlation between them [22, 42,43], In our study we found a positive correlation between these characteristics. A successful PCOS treatment that reduces the burden of symptoms and associated psychosocial stress should also have a significant impact on the health-related quality of life of women [35, 58, 60].

The accuracy of self-assessment in comparison to an evaluation conducted by qualified professionals has been a topic of interest. This study provides evidence that women possess the ability to effectively assess the presence of psychological issues within themselves. Also, we discovered that Self-screening of the symptoms and mental stress-related parameters helps the patients to identify and manage the condition. Self-administered questionnaires that ask women about their psychological health and menstrual patterns showed good sensitivity and intermediate specificity for regulating fluctuations in psychiatric symptoms across the menstrual cycle in PCOS patients.

Conclusion

According to the current study findings, a significant percentage of PCOS women in the present cohort suffer from a variety of mental illnesses like anxiety, depression, and sleep disturbances with prevalences of 18.5%, 17.6%, and 31.3% respectively. Various socio-demographic variables are linked to poor physical and psychological health. The typical menstrual cycle length and BMI were significantly correlated to PCOS incidence. The current study found that the prevalence of mental health-related problems in large epidemiological investigations can be carried out more easily because of the independence of simple, self-administered questionnaires from expert review. Implementing a PCOS screening tool that asks about mental state and menstrual-related characteristics may be helpful for epidemiological investigations of PCOS. From our findings, we recommend proper treatment, public awareness, and a healthy lifestyle to support the mental health of PCOS women. To truly understand the severity of the problem, more research is needed with the same variables used in the current study and a large cohort size in women with PCOS.

Authors contributions

Ramasamy Tamizhselvi designed the experiments. Aparna Eledath Kolasseri developed the questionnaire and collected the responses. Anjana Eledath Kolasseri participated in the design of the study and performed the statistical analysis. Jayanthi Sivaraman analyzed the results and helped to draft the manuscript. Aparna Eledath Kolasseri prepared the manuscript with contributions from all coauthors. All authors read and approved the final manuscript.

Acknowledgment

The authors would like to acknowledge the School of Biosciences and Technology, Vellore Institute of Technology for the resources.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The data that support the findings of this study are available on request from the corresponding author, Ramasamy Tamizhselvi. The data are not publicly available due to restrictions since they contain information that could compromise the privacy of research participants.

Additional information

Funding

The author(s) reported there is no funding associated with the work featured in this article.