Risk of polycystic ovary syndrome: a population-based analysis of sociodemographic factors, healthcare access, health behaviors, and health status

Abstract

Background

Polycystic ovarian syndrome (PCOS) is the most prevalent endocrine concern among women of reproductive age. In Saudi Arabia, there is a lack of evidence to identify who is at higher risk of PCOS and what the potential risk factors are. Thus, this study aimed to investigate the associations of PCOS risk with demographic and socioeconomic characteristics, access to healthcare, health behaviors, and health status.

Methods

This cross-sectional study was conducted among women in all different regions of Saudi Arabia to assess PCOS risk and related factors. Ethical approval was obtained, and data collectors distributed anonymous, self-administered questionnaires through social media platforms, with informed consent from participants. Sociodemographic characteristics, health behaviors, and perceived stress were measured, with stress assessed using the Arabic version of Cohen’s Perceived Stress Scale. Data management and analyses included statistical description, bivariate analysis, and multinomial logistic regression analyses using SPSS, with significance set at p < 0.05.

Results

The majority were younger than 30 years old, single, educated, urban residents, employed or students, and non-smokers. Most participants reported no chronic illnesses, with an average stress level of 19.71 (± 6.68). Concerning the risk of PCOS, 41.3% were at low risk, 33.3% were at suspected risk, 2.9% were at high risk, and 22.5% were diagnosed with PCOS. Factors associated with PCOS risk included age, region of residence, income, weight status, smoking status, presence of chronic conditions, medication and herbal remedy use, and perceived stress. Adjusted findings indicated that younger age, lower income, and higher stress levels were linked to an increased risk of PCOS, while chronic conditions were significantly associated with PCOS diagnosis rates.

Conclusion

The study suggested the need for tailored interventions addressing lifestyle, stress, and comorbid disease management to reduce the risk of PCOS and improve women’s health outcomes.

Introduction

Polycystic ovarian syndrome (PCOS) is considered a common women’s health issue, which is characterized by endocrinal abnormalities and accompanied by many metabolic disorders [1, 2]. It was first described comprehensively by Stein and Leventhal in 1935 [3]. It is affecting approximately one out of ten women in reproductive ages [4]. While the global prevalence of PCOS is well-documented, its incidence in Saudi Arabia has been less clear. A recent cross-sectional study conducted in the Western region of Saudi Arabia estimated the incidence of PCOS to be about 31.8% [5].

Empirical studies have shown the profound impact of PCOS on women’s health [6,7,8,9]. Research indicated that even during the fetal period, the disease can cause hyperandrogenism and insulin resistance, leading to numerous health issues, such as menstrual irregularities, infertility, and metabolic complications [6]. A meta-analysis showed that women with PCOS were at greater risk of developing metabolic syndromes, including hyperglycemia, insulin resistance, hypertension, central obesity, and dyslipidemia [7]. Additionally, women with PCOS were more likely to experience gestational diabetes, preeclampsia, preterm delivery, and miscarriages during pregnancy [8, 9]. Despite these negative impacts on women’s health, a substantial percentage of PCOS cases remained unrecognized [10]. A retrospective study in southern Australia estimated that 68% of the PCOS cases were undiagnosed [11], and this failure to identify the disease could lead to poor management and severe complications, such as glucose intolerance, hyperlipidemia, type 2 diabetes, insulin resistance, and cardiovascular diseases [11,12,13].

Previous literature showed that some risk factors, such as obesity, family history of diabetes, family history of infertility, depression, and lack of physical exercise, could be associated with an elevated risk of PCOS [14,15,16].

Considering each country’s unique social and cultural characteristics, studies exploring the risk factors of PCOS among women in Saudi Arabia are scarce. This may result in an incomplete understanding of the condition and hinder the development of tailored prevention and treatment strategies. To optimize public health outcomes, this study investigated the potential influence of demographic and socioeconomic factors, access to healthcare, health behaviors, and health status on PCOS risk in Saudi Arabia.

Method

Study design

A cross-sectional population-based study was conducted among the 13 administrative regions of Saudi Arabia using a convenient sampling technique.

Procedures

Ethical approval was obtained from the institutional review board (IRB) (IRB log No. 23–0975) by the Declaration of Helsinki. To ensure a representative sample encompassing the entire Saudi Arabian population, trained data collectors distributed the questionnaire via several social media platforms, such as WhatsApp, Telegram, and X platform. After reviewing the study’s objectives, participants voluntarily agreed to participate and provided informed consent to complete the anonymous, self-administered questionnaire. Inclusion criteria for this study included adult women aged 18 years and older living in Saudi Arabia who provided informed consent. Women who aged 50 years and older served as a reference group in the statistical analyses. This broader age range facilitates a comprehensive examination of PCOS and its implications across different age groups. Our study excluded participants aged under 18 years old. Also, exclusion criteria included women who did not complete the questionnaire, withdrew consent, or provided unreliable responses during data analysis. This approach ensured a focused and reliable assessment of the risk factors associated with PCOS.

Measures

The independent variables were demographic and socioeconomic factors, access to health care, health behaviors, health status and medication use, and perceived stress. The demographic and socioeconomic factors were age, sex, marital status, education, region of residence, family income, and nationality. Access to health care was measured through participants’ medical insurance status, categorized as follows: those with medical insurance, non-insured individuals who use public hospitals and clinics, and those who use private hospitals and clinics (out-of-pocket). This classification allows us to evaluate how different insurance statuses impact access to healthcare services among the study population. Health behaviors were assessed by asking about lifestyle choices impacting health, including smoking, diet, and physical activity. Health status and medication use were evaluated using various factors, including weight status, presence of chronic physical and psychological illnesses, and using medications and herbal remedies. For weight status, participants’ body mass index (BMI) was calculated using their reported weight in kilograms and height in centimeters, following the standard BMI formula. Based on the calculated BMI, participants were categorized as underweight (BMI < 18.5), normal weight (BMI 18.5 to 24.9), overweight (BMI 25 to 29.9), and obese (BMI ≥ 30). The presence of chronic illnesses was determined by self-reported diagnosis. Additionally, women were asked whether they use medications or herbal remedies. If yes, they had to mention the name of the medication or herbal remedies used. Perceived stress was assessed using Cohen’s Perceived Stress Scale (PSS), which measures how individuals perceive their life as stressful [17,18,19]. This scale could be further used to screen for individuals at risk of psychological health issues [18,19,20]. We utilized the Arabic version of the PSS (10 items), which has shown acceptable internal consistency (Cronbach’s alpha coefficient = 0.67) and strong validity [21]. Each item scored on a 5-point Likert-type scale ranging from 0 to 4, with a maximum score of 40 (the highest stress level) [21].

The primary outcome of interest was the risk of PCOS. Women were categorized into four groups: low risk, suspected risk, high risk, and women diagnosed with PCOS (self-reported diagnosis). For those who were not previously diagnosed with PCOS, the risk of PCOS was evaluated using Haq’s Scale for Clinical Evaluation of PCOS [22]. The scale’s items aligned with the Rotterdam criteria to facilitate self-reporting PCOS symptoms. Although this tool cannot be used for definite diagnosis as it does not include Polycystic Ovaries ultrasound, it allows researchers to publicly capture a comprehensive view of PCOS symptoms through screening for women who were not diagnosed with PCOS. Items were translated to Arabic using back-forward translation. Before the survey was distributed, its content validity, face validity, and reliability were tested using a pilot sample.

Women were categorized based on their scale scores and any prior PCOS diagnosis into four groups: low risk (scores less than four), suspected risk (scores ranged from 4 to 8), high risk (scores greater than eight), and those diagnosed with PCOS.

The English language version of the whole survey is provided in supplementary file A.

Statistical analysis

Means with standard deviations (SDs) were reported for continuous variables, and frequencies with percentages were reported for categorical data. The reliability of Haq’s Scale for Clinical Evaluation of PCOS was tested using Cronbach’s alpha. The unadjusted associations of women’s characteristics with the risk of PCOS were analyzed using chi-square test and Fisher’s exact test. Bivariate associations between perceived stress and sample characteristics were examined using an independent t-test and one-way analysis of variance (ANOVA). The adjusted associations of independent variables with the risk of PCOS were tested using multinomial logistic regression. P values < 0.05 were considered to indicate statistical significance. The statistical analysis was performed using the Statistical Package for Social Sciences (SPSS).

Results

Sample size

The response rate to the distributed survey was 95.9%. A total of 1,144 women were invited to participate in the study. Only 4% of the women (n = 47) declined to participate. Among the women who expressed initial interest, eligibility was further assessed based on age criteria, and 29 participants were excluded because they were under 18 years old. The final sample comprised 1,068 women who met the study’s eligibility criteria. Figure 1 depicts the flowchart outlining the participant recruitment process.

Fig. 1

Flowchart of the study sample

Characteristics

The sample characteristics, including sociodemographic factors, health behavior, access to health care, and health status, were presented in Table 1. A total of 1068 women were enrolled in this study, 91.9% of whom were Saudi citizens. More than half of the participants were less than 30 years old (57.2%), single (50.2%), had a bachelor’s degree (72.4%), lived in an urban area (92.8%), had enough income or enough income with savings (81%), and were either employed or students (69.5%). For access to health care, 33.5% had medical insurance, 33.9% of women were non-insured but used public hospitals or clinics, and 32.6% were non-insured and used private hospitals or clinics (out-of-pocket expenditures). For health behaviors, 93.4% were nonsmokers, 78.1% did not follow a healthy diet, and 43.4% did not exercise. For weight status (based on BMI), 40% of participants had normal weight, 29.2% were overweight, 21.9% were obese, and most women had no chronic physical or psychological illnesses (75.1% and 88.4%, respectively). Most women did not use medications or herbal remedies (76.7% and 84.8%, respectively).

Table 1 Sociodemographic factors, health behaviors, access to health care, and health status of the included women (N = 1068)

Figure 2 displays women’s most commonly reported chronic conditions, classified into physical and psychological categories. The prevalence of anxiety and depression among adult females was approximately 19% and 13%, respectively, making them the most common mental health problems in the studied population. Further psychological disorders that were identified included eating disorders (3%) and obsessive-compulsive disorder (5%). For physical conditions, cardiovascular conditions accounted for 34% of the reported conditions, while endocrine and hormonal conditions constituted 16%. Moreover, 16% reported bone pain, 13% had headaches or migraines, 8% reported sleep disorders, and only 4% had skin conditions.

Fig. 2

The most commonly reported physical and physiological chronic illness in adult women

Medications and herbal remedies use

Figure 3 shows the most used medications reported by participants and provides insights into their underlying health profiles. The most frequently used medications were multivitamin supplements, antihypertensive/cardiac medications, oral antidiabetic medications, or metformin. Additionally, a small proportion reported the use of antidepressants, hypothyroidism, and iron supplements. To gain further insight into the self-reported health practices of women at risk of PCOS, an analysis of the most used herbal remedies within the study sample was conducted (Fig. 4). A significant percentage of women preferred herbal remedies over conventional medicines (n = 212), highlighting the preference for herbal remedies within the study population. Several herbs emerged as particularly prevalent among participants. These included cumin, anise, chamomile, ginger, marjoram, and purslane. Other herbal remedies were cinnamon, fennel, Indian kudzu, and senna leaves. Furthermore, a wide range of herbal remedies, including coffee husk, peppermint, rosemary, moringa, fenugreek, palm dates, matcha, thyme, lavender, turmeric, sumac, chia seeds, flaxseeds, cloves, frankincense, safflower, olive oil, cardamom, hibiscus, wild blackberry leaves, black seeds, and sesame oil, were reported to a lesser extent.

Fig. 3

The most commonly used medications were reported by women in our sample

Fig. 4

The most common herbal remedies used by women in our sample

Unadjusted association between PCOS risk and sample characteristics

The reliability of Haq’s Scale for Clinical Evaluation of PCOS was acceptable (Cronbach’s alpha = 0.716). The bivariate associations of PCOS risk with sample characteristics (sociodemographic factors, health behaviors, access to health care, and health status) are shown in Table 2. Overall, 41.3% were at low risk of PCOS, 33.3% were at suspected risk of PCOS, 2.9% were at high risk of PCOS, and 22.5% were diagnosed with PCOS. Factors that were significantly associated with PCOS risk included age, marital status, occupation, region of residence, income status, weight status, smoking status, presence of physical and physiological conditions, medication use, use of herbal remedies, and perceived stress.

Table 2 Unadjusted association of PCOS risk with sociodemographic factors, health behaviors, access to health care, and health status of the included women (N = 1068)

Perceived stress and sample characteristics

Table 3 demonstrates perceived stress and its association with the sample characteristics. The average perceived stress of women was 19.71 (± 6.68). The variables significantly associated with perceived stress were age, marital status, educational level, occupation, income status, nationality, smoking status, diet, physical activity, and the presence of any physical or psychological illness. For age groups, the highest perceived stress score was found for girls younger than 23 years (21.28 ± 6.27), followed by women aged 23 to 30 years (20.85 ± 6.98). Single females had the highest average score for perceived stress (21.18 ± 6.61) compared to married (18.03 ± 6.33) and divorced/widowed women (19.55 ± 6.96). Additionally, the average scores of perceived stress for women who had a bachelor’s degree (19.9 ± 6.77), were unemployed and looking for a job (20.05 ± 6.65), had a low income (22.57 ± 6.13), were non-Saudi (21.47 ± 6.58), were a past smoker (22.81 ± 7.90), had not followed a healthy diet (20.05 ± 7.90), had no exercise or physical activity (20.26 ± 6.52), and reported having a physical illness (20.44 ± 7.13) and psychological illness (24.99 ± 5.52) were greater than those of their counterparts.

Table 3 Bivariate associations of perceived stress with the sample’s characteristics (N = 1068)

Adjusted association of PCOS risk with sample characteristics

The results of multivariate analysis of the associations between PCOS risk and sample characteristics are shown in Table 4. The sample characteristics significantly associated with PCOS risk were age, region of residence, income status, weight status, smoking status, presence of physical and physiological conditions, medication use, use of herbal remedies, and perceived stress. The risk of PCOS was lower in women with enough income (OR = 0.54; 95% CI = 0.32–0.92) than in women with low income. Similarly, women who were underweight (odds ratio [OR] = 0.34; 95% CI = 0.18–0.68), had a normal weight (OR = 0.43; 95% CI = 0.27–0.69), or were overweight (OR = 0.53; 95% CI = 0.33–0.83) were at lower risk of PCOS than obese women. Women who were diagnosed with physical (OR = 1.59; 95% CI = 1.04–2.42) or psychological (OR = 1.71; 95% CI = 1.01–2.88) conditions were more likely to be at suspected risk of PCOS than women without these conditions. The likelihood of being at high risk of PCOS was significantly greater in patients who were past smokers (OR = 5.93; 95% CI = 1.31–26.91) and had any psychological illness (OR = 3.06; 95% CI = 1.06–8.84) than in their counterparts. The diagnosis of PCOS was significantly associated with younger age, region of residence, weight status, presence of any physical or psychological conditions, medication use, and herbal remedy use. Women aged less than 23 years were 6 times more likely to be diagnosed with PCOS than women aged 50 years and older. Women who lived in the eastern region were 60% less likely to be diagnosed with PCOS than those in the southern region (OR = 0.42; 95% CI = 0.19–0.91). Underweight women (OR = 0.37; 95% CI = 0.17–0.79), women with normal weight (OR = 0.29; 95% CI = 0.17–0.49), or overweight women (OR = 0.45; 95% CI = 0.27–0.74) were less likely to be diagnosed with PCOS than obese women. On the other hand, the presence of any physical or psychological conditions and the use of medications or herbal remedies were associated with an increased likelihood of a PCOS diagnosis. Women who reported having physical illness were two times more likely to be diagnosed with PCOS than women without physical chronic conditions (OR = 2.01; 95% CI = 1.28–3.16). Similarly, women who reported having a psychological illness were two times more likely to be diagnosed with PCOS than women without chronic psychological conditions (OR = 2.12; 95% CI = 1.19–3.16). Women who used medication were 1.67 more times to be diagnosed with PCOS (95% CI = 1.07–2.61), and women who used herbal remedies were 2.15 more times (95% CI = 1.34–3.43) to be diagnosed with PCOS than women who did not use medication or herbal remedies.

Table 4 Adjusted association of PCOS risk with sociodemographic factors, health behaviors, access to health care, and health status of the included women (N = 1068)

Discussion

This study provided valuable insights into the prevalence of polycystic ovary syndrome (PCOS) risk among a sample of Saudi women and its associations with various sociodemographic factors, health behaviors, access to healthcare, and health status.

The participants’ sociodemographic profile revealed that most women were Saudi citizens, predominantly younger than 50 years old, single or married with enough income, educated, resided in urban areas, and were employed or students. Access to healthcare varied, with a significant portion relying on public hospitals/clinics or medical insurance. While a positive finding was the high prevalence of non-smokers and a reported absence of physical and psychological illness in most participants, a significant portion of the women did not adhere to healthy dietary practices or engage in regular physical activity. This highlights the need for public health campaigns and interventions explicitly _targeting these behaviors to promote the adoption of a healthy lifestyle among women.

The study revealed that more than a third of the sample was at low risk, while a notable proportion of the participants were either at suspected risk (one-third) or diagnosed with PCOS (a quarter), and a tiny portion was at high risk. These results could refer to high awareness about PCOS as there were few individuals at high risk for PCOS who had not been diagnosed. However, the considerable percentage of diagnosed women underscores the importance of understanding the factors contributing to PCOS risk within the female population.

Several sociodemographic factors were associated with the risk of PCOS. Age emerged as a significant determinant, with younger women being more susceptible to the condition. Specifically, women under 30 years exhibited a higher likelihood of being diagnosed with PCOS compared to those in older age groups. This finding aligns with the literature highlighting the greater prevalence of PCOS among younger women than among older women [23].

Region of residence and income status also exhibited significant associations with PCOS risk, emphasizing the multifactorial nature of this syndrome. For example, living in the eastern region of Saudi Arabia was associated with a 59% decrease in the risk of PCOS diagnoses as compared to living in the southern region. For other regions, there was no significant difference in the risk of PCOS after controlling for other variables (sociodemographic variables, health behavior, health status, and access to healthcare).

Although there was heterogeneity in access to health care, this discrepancy had no significant impact on the risk of PCOS or PCOS diagnosis.

In terms of healthy behaviors, smoking was significantly associated with PCOS. In contrast, dietary habits and physical activity did not correlate substantially with PCOS risk after adjusting for other independent variables such as obesity. Consistent with our results, a previous study conducted in the USA revealed that physical activity and diet were not associated with the risk of PCOS [24].

For health status, perceived stress, weight status, and the presence of any chronic physical and/or psychological conditions were significantly associated with the risk of PCOS after controlling for other independent variables. Higher perceived stress levels were significantly linked to an increased risk of PCOS. Compared to the average women, women with a high risk of PCOS had higher stress levels while women with a low risk of PCOS had lower levels of stress. As shown in a previous study, chronic stress can cause the generation of excessive harmful amounts of free radicals (reactive oxygen species) that could lead to oxidative stress, which is an imbalance in antioxidant capacity [25]. Murri et al. reported a significantly elevated level of the circulating marker of oxidative stress in women with PCOS compared to that in women without PCOS [26]. These consistent and interrelated findings demonstrated the intricate interplay between psychological factors and reproductive health outcomes. Therefore, understanding the risk factors associated with stress could be crucial for reducing the risk of PCOS. Thus, our analyses revealed that younger, single women with low education and low-income levels, non-Saudi nationality, a history of smoking, unhealthy lifestyle behaviors, and those diagnosed with physical and/or psychological illnesses experienced higher levels of stress compared to their counterparts. Therefore, stress management strategies should address these factors to reduce the risk of PCOS in women without the condition and to help manage the disease in those diagnosed with PCOS.

Further, women with existing chronic illnesses, whether physical or psychological, exhibited a 2- to 3-fold greater risk of being diagnosed with PCOS compared to women without any chronic conditions. This underscores the importance of addressing comorbidities in the management and treatment of PCOS. Regarding chronic conditions, anxiety, and depression emerged as the most prevalent mental health issues, while cardiovascular and metabolic disorders were the most common physical health concerns among women.

Interestingly, the study also revealed that the use of medications and herbal remedies significantly increased the likelihood of a PCOS diagnosis. The medications predominantly used included multivitamin supplements, antihypertensive/cardiac medications, and oral antidiabetic medications. Moreover, a substantial portion of the sample preferred herbal remedies.

A few limitations should be considered when interpreting the study’s findings. First, using a convenient sample may limit the generalizability of the results to all women in Saudi Arabia. Also, the data related to chronic diseases were based on self-reported diagnoses, which could be prone to recall bias. Furthermore, our study’s survey-based nature may not capture the full spectrum of PCOS symptoms, as it relied on women’s understanding and perception of their condition. Although Haq’s Scale for Clinical Evaluation is intended to provide an overview of potential risk factors rather than definitive susceptibility for those already diagnosed with PCOS, it is a useful screening instrument to categorize women based on reported symptoms that align with the Rotterdam criteria of PCOS diagnosis. Also, this research was a cross-sectional study, no causal relationship can be inferred between the risk of PCOS and the discussed risk factors. However, the multivariate levels of analyses revealed evidence of significant associations with the abovementioned risk factors.

Despite these limitations, our study has multiple strengths. First, this study provided a comprehensive overview of PCOS risk and examined its association with all potential factors, including sociodemographic variables, health behaviors, access to healthcare services, and health status. Second, the sample size was large and robust, with a high response rate of 95.9%, resulting in 1,068 women meeting the study’s eligibility criteria. Furthermore, there was variability in the sample’s characteristics, with a broad sociodemographic profile representing most women in Saudi Arabia.

Conclusion

The findings of this study underscore the complex interplay of various factors in determining women’s health outcomes, particularly about perceived stress and PCOS risk. The study provides valuable insights for healthcare professionals and policymakers in developing _targeted interventions and support strategies to mitigate these risks and improve women’s health and well-being.