Polycystic ovary syndrome is associated with weight‐loss attempts and perception of overweight independent of BMI: a population‐based cohort study

Up to 70% of women with polycystic ovary syndrome (PCOS) have pre‐obesity or obesity. The aim of this study was to investigate whether women with PCOS have more weight‐loss attempts than women without PCOS, regardless of BMI. Moreover, women's weight perceptions in relation to previous weight‐loss attempts were evaluated.


INTRODUCTION
Polycystic ovary syndrome (PCOS) is the most common endocrine disorder in reproductive-aged women, with an estimated prevalence ranging from 8% to 18% [1,2]. PCOS is a multimorbid syndrome characterized by hyperandrogenism, as well as various metabolic, reproductive, and psychological impairments [3,4]. Furthermore, women with PCOS are burdened with excess weight; the prevalence of obesity has been shown to be 50% to 70% [5], with weight gain being higher compared with control individuals from childhood onward, around the time when the second adiposity rebound occurs at the age of 5 years [6]. Indeed, a clinical [7] and genetic link [8] between obesity and PCOS exists. Although there are still unknown factors, the current understanding is that women with PCOS are biologically susceptible to weight gain [9] and are more affected by adverse lifestyles than women without PCOS [10].
It is estimated that 40% of adults in the general population have tried to lose weight at some point during the past 5 years [11], although history of weight-loss attempts has been found to be associated with higher body mass index (BMI) and type 2 diabetes mellitus (T2DM) risk factors [12]. In PCOS, weight loss of 5% to 10% and weight management are the primary treatment modalities according to the international PCOS guideline [13] because weight loss has been shown to improve clinical features and metabolic and reproductive outcomes [14]. A majority of previously conducted weight-loss studies in women with PCOS have been lifestyle interventions, although new data on medical and surgical weight management are also emerging [15]. Weight loss is not a simple task, even in a supervised context. So far, none of the lifestyle interventions has proven to be better than others; the success rate in achieving weight loss has varied widely and the dropout rates have been high [16]. However, most weight-loss attempts are selfinitiated and executed without any practical support from health professionals, predisposing women with PCOS to weight-loss challenges [17]. To date, there has been no prior general populationbased research comparing the prevalence of multiple weight-loss attempts between women with and without PCOS.
It is assumed that the recognition of having overweight is a prerequisite for successful weight loss or management. Although the perception of having overweight has been shown to be associated with weight-loss attempts [18], it also increases the risk of greater weight gain and disordered eating [19,20]. Women with PCOS are prone to psychological distress [21], which is partly explained by body-image distress [22] related to the physical manifestations of PCOS, including acne, hirsutism (H), and obesity [23]. The association between weight perception and weight-loss attempts in women with PCOS has not been investigated previously, to our knowledge.
The present population-based cohort study aimed to determine whether women with PCOS are more likely to have experienced more multiple weight-loss attempts at ages 31 and 46 years compared with women without PCOS. The second objective was to evaluate weight perception among women with PCOS in relation to weight-loss attempts.

Study population
The study population was derived from the Northern Finland Birth Cohort 1966 (NFBC1966), which includes individuals from the two northernmost provinces of Finland whose expected year of birth was Study Importance What is already known?
• Commonly, all women with polycystic ovary syndrome (PCOS), regardless of BMI, receive information about the risk of weight gain; however, support for weight management is limited.
• Women with PCOS are more likely to be using both healthy and detrimental weight-management practices compared with women without PCOS. Also, body-image distress and a perceived susceptibility to weight issues are common in women with PCOS.
• Prevalence of weight-loss attempts during a life course has not been investigated, to our knowledge, in women with PCOS.
What does this study add?
• Multiple weight-loss attempts were more common in women with PCOS compared with women without PCOS at ages 31 and 46 years. Perception of having overweight was a strong predictor of weight-loss attempts.
• Women with PCOS perceived themselves as having overweight significantly more often than women without PCOS, even when their BMI was in the normal range.
• At age 46 years, PCOS was an independent predictor of multiple weight-loss attempts and a perception of having overweight.
How might these results change the direction of research or the focus of clinical practice?
• The results challenge clinicians to evaluate how they discuss weight-related matters with women with PCOS, especially when they are at a normal weight.
• Emphasis should be placed on supporting women with PCOS to reduce inefficient weight-loss attempts and stress.
• A weight-neutral approach to lifestyle interventions should be a key area of future research. 1966 (total n = 12,231; women n = 5889). Since birth, information about the participants has been recorded regularly; the follow-up data collection points have been at ages 1, 14, 31, and 46 years. The part regarding female health was merged from age 31 years onward; therefore, this study used the two most recent data collection points.
The detailed cohort description and follow-up protocols have been previously published [24,25]. This study was approved by the Ethics Committee of the Northern Ostrobothnia Hospital District in Oulu, Finland (94/2011). All participants provided informed consent for the study. The anonymity of the participants was ensured by encrypting their personal identity information and by using identification codes.
Postal questionnaires regarding health, behavior, and social background were recorded (Figure 1)

Definition of PCOS
At age 31 years, the postal questionnaire included two questions screening for PCOS symptoms: oligomenorrhea (OA; "Is your menstrual cycle often [more than twice a year] longer than 35 days?") and H ("Do you have bothersome excessive body hair?"). Women who reported OA and H fulfilled the National Institutes of Health (NIH) criteria for PCOS, which were the only available PCOS criteria during 1997 [26]. At age 46 years, the postal questionnaire included the question "Have you ever been diagnosed as having polycystic ovaries (PCO) and/or polycystic ovary syndrome (PCOS) during your life?" to which 4.9% (n = 181) of the women replied "yes" (considered as women with PCOS diagnosis). The total group with PCOS included women reporting both OA and H at age 31 years or a self-reported PCOS diagnosis at age 46 years (n = 278). Validation of the group with PCOS has been previously performed [7,27,28]. The women classified as not having PCOS did not report any PCOS symptoms at age 31 years and answered "no" to the question regarding a PCOS diagnosis at age 46 years (n = 1560). Pregnant women and women using hormonal contraceptives (n = 1488) at age 31 years and women who did not give permission to use their data were excluded.

Weight-loss attempts
The participants were asked about their lifelong attempts to lose weight with the same question at ages 31 and 46 years: "Have you ever attempted to lose weight in a serious manner?" The answer options were as follows: 1) No, never; 2) Yes, once; and 3) Yes, multiple times.
Answer options 1 and 2 were merged to make a distinct difference between participants with occasional dieting and those with serious attempts to lose weight. Thus, the participants were classified into one of the following two categories at ages 31 and 46 years separately: 1) zero to one weight-loss attempt; and 2) multiple weight-loss attempts.

Weight perception
The participants were asked about their perception of their weight with the same question at ages 31 and 46 years: "How do you perceive your weight?" The answer options were as follows: 1) Significantly overweight; 2) Slightly overweight; 3) Just the right weight; and 4) Slightly or significantly underweight. As the focus was on the perception of having overweight, answer options 1 and 2 were merged, and the participants with the perception of having underweight were excluded (9 participants at age 31 years and 23 participants at age 46 years). Therefore, the final classification for weight perception was as follows: 1) perception of having overweight; and 2) perception of the right weight.

Anthropometry
At ages 31 and 46 years, the clinical examinations included the measurement of weight and height. Weight (kilograms) was measured with a digital scale that was calibrated regularly, and height (centimeters) was measured twice with a standard calibrated stadiometer. BMI (kilograms per meters squared) was calculated by using the measured weight and height, and, in case of missing measured data, self-reported values were used. There was no statistically significant difference observed among the measured and self-reported BMI values [7]. The participants were categorized into BMI groups based on the World Health Organization criteria for normal weight (BMI = 18.5-24.9), preobesity (BMI = 25.0-29.9), and obesity (BMI ≥ 30.0). The participants with BMI < 18.5 were excluded from the study. Waist circumference was measured from the midway point between the iliac crest and the lowest ribs. The cutoff value of >80 cm was based on the International Diabetes Federation's definition [29].

Self-reported diagnosis of T2DM
The participants were asked at age 46 years whether they had ever been diagnosed or treated by a doctor for T2DM.

Psychological distress
Psychological distress was assessed using the Hopkins Symptoms Checklist-25 (HSCL-25) at ages 31 and 46 years. The total score cutoff point of ≥1.75 indicates a psychiatric disorder.

Sociodemographic factors
Education was determined with several questions at ages 31 and 46 years. The variable was classified into three groups according to education level: 1) basic; 2) secondary; and 3) tertiary. The participants were asked about their marital status, both at ages 31 and 46 years. Two groups were formed: participants who were married, cohabiting, or had a registered partnership (only age 46 years); and participants who did not have a partner or who were widowed or divorced.

Lifestyle
Smoking status was determined at ages 31 and 46 years by forming three different groups: 1) nonsmokers; 2) former or occasional T A B L E 1 Characteristics of the study population (n = 1838)

Laboratory methods
Serum testosterone and sex hormone binding globulin (SHBG) were assayed, as previously described [27].

Weight-loss attempts
The prevalence of participants with multiple weight-loss attempts during the life course was higher in women with PCOS compared with women without PCOS at ages 31 (47% vs. 34%, p < 0.001) and 46 years (63% vs. 47%, p < 0.001; Figure 2). Among participants with pre-obesity or obesity, multiple weight-loss attempts were significantly more prevalent in women with PCOS compared with those without PCOS at age 46 years (PCOS 79% vs. CON 66%, p = 0.003).
A similar but nonsignificant trend was also seen at age 31 years.
Binary logistic regression analyses were used to assess the association between PCOS and multiple weight-loss attempts (Tables 2 and   3

Weight perception
Weight perception was strongly associated with weight-loss attempts (Tables 2 and 3 Binary logistic regression analyses were used to assess the independent risk of PCOS in the perception of having overweight (Table 4). In the univariable analysis, PCOS was associated with the perception of having  (Table 4). At age 46 years, PCOS was independently associated with a perception of having overweight when adjusted for BMI, psychological distress, FAI, and physical activity (Table 4).

Dropout analysis
The dropout rate was higher in women with PCOS than in women without PCOS. Regarding the question of weight-loss attempts, the dropout rates were 31% for those with PCOS and 12% for those without PCOS (p < 0.001) and, in terms of weight perception, 34% for those with PCOS and 14% for those without PCOS (p < 0.001). The perception of having overweight was higher in women with PCOS in the follow-up group compared with the dropout group (p = 0.006).

DISCUSSION
This was the first study, to our knowledge, to investigate the prevalence of weight-loss attempts and the perception of having overweight in early and late adulthood in women with PCOS compared with women without the syndrome. The results revealed that women with PCOS reported multiple weight-loss attempts and a perception of having overweight more frequently than women without PCOS.
We also found that women with PCOS perceived themselves as having overweight significantly more often, even when their BMI was in the normal range. In the multivariable regression analysis, PCOS was independently associated with multiple weight-loss attempts and the perception of having overweight at age 46 years, but not at age 31 years.
In the present study, women with PCOS were more likely to have experienced multiple weight-loss attempts during their life course compared with women without PCOS. However, the syndrome was not independently associated with weight-loss attempts at age 31 years, although the other risk factors, high BMI, psychological distress, and perception of having overweight accumulated in women with PCOS and were likely to mediate the association. A previous population-based study found that women with PCOS engage in a wide range of both healthy and unhealthy weight-management practices more often than women without the syndrome in their early 30s, independent of BMI [31]. According to the present study, PCOS status contributed to multiple weight-loss attempts by a late fertile age, perhaps due to awareness of the high risk of developing obesity-related diseases [32]. However, it is notable that PCOS was associated with multiple weight-loss attempts independent of T2DM, even though T2DM was the strongest predictor at age 46 years.
T A B L E 2 Risk factors for multiple weight-loss attempts by age 31 years according to univariable and multivariable binary logistic regression analysis in the study population of 1255 women with (n = 188) and without (n = 1067) PCOS The total group with PCOS included women reporting both oligomenorrhea and hirsutism at age 31 years or self-reporting PCOS diagnosis at age 46 years.
In our study population, women with PCOS display significantly greater weight gain, BMI values, and prevalence of pre-obesity and obesity already since adolescence [6,7] despite the higher prevalence of weight-loss attempts, as shown in the present study.
According to previous evidence, there is no difference between women with and without PCOS in terms of their ability to lose weight [33]. However, our findings suggest that women with PCOS experience challenges with weight loss. Women with PCOS are repeatedly recommended to lose weight, although several barriers exist in the implementation: lifestyle management is not offered consistently and does not meet the needs of the affected women [17]. A qualitative report indicates that women with PCOS find weight loss to be a difficult goal, and they lack motivation due to the history of failed weight-loss attempts [34]. Consequently, previous insufficient weight-loss attempts can have detrimental effects on body image, self-esteem, eating [34], and future lifestyle modification outcomes [12]. Given the negative consequences of failed weight-loss attempts, the need for a weight-neutral approach has been brought up [34,35].
Women with PCOS have been shown to have higher rates of anxiety and depression than women without PCOS [21], and the results of the present study indicate that psychological distress is associated with previous multiple weight-loss attempts. Given that implementing effective lifestyle changes to achieve weight loss requires drive and commitment, psychological distress is a potential factor disrupting weight-loss attempts in women with PCOS [35]. Indeed, lifestyle interventions have reported high attrition rates, varying between 12% and 47% in women with PCOS [16], especially among participants with depressive symptoms [36]. Recent evidence indicates that participants with less psychological distress have a better chance of successful weight loss [37]. Indeed, there are several lifestyle interventions that have included cognitive behavioral therapy; the interventions have shown promising results for weight loss [38,39], which suggests that psychological distress should be considered in the treatment of PCOS.
In the present study, the perception of having overweight was shown to be more common in women with PCOS, even among participants with normal weight. In addition, PCOS was independently associated with the perception of having overweight when adjusted for all confounders at age 46 years. Having a high waist circumference at age 46 years was found to partly explain the perception of having overweight in participants with normal weight. A higher rate of perception of having overweight could also stem from fear of gaining weight; women with PCOS more often have a feeling of poor control over their weight and a perceived susceptibility to weight issues compared with control individuals without PCOS [32]. In the present study, PCOS was associated with the perception of having overweight as well as multiple weight-loss attempts regardless of pre-obesity or obesity at age 46 years. This finding is rather concerning, considering that dieting among individuals with normal weight often leads to long-term weight gain [40].
T A B L E 3 Risk factors for multiple weight-loss attempts by age 46 years according to univariable and multivariable binary logistic regression analysis in the study population of 1519 women with (n = 206) and without (n = 1313) PCOS The perception of having overweight was significantly associated with multiple weight-loss attempts in the present study. This is consistent with previous findings from the general population, which suggest that people who perceive themselves as having overweight are more likely to try to lose weight [18,41]. There is strong evidence indicating that the perception of having overweight is associated with disordered eating and future weight gain [41]. Indeed, a high prevalence of bodyimage distress has been previously shown to occur in women with PCOS, and it has been suggested to be associated with difficulties in adopting lifestyle changes, especially physical activity [35]. In addition, detrimental weight-management practices [31] and disordered eating behaviors [42] have already been shown to be common in women with PCOS. People with a perception of having overweight have been found to have poorer mental and physical health compared with those who did not perceive themselves as having overweight [19]; thus, experiencing stigma related to obesity can motivate individuals to use more extreme weight-loss methods [19,41]. Based on the findings of this study, the perception of having overweight is a potential risk factor for disrupting weight management in women with PCOS with both normal weight and pre-obesity or obesity.  Correspondingly, Jiskoot and colleagues (2022) found that worse body image at baseline was associated with a lower proportion to achieve ≥5% weight loss through a three-component lifestyle intervention [43]. On the other hand, the lifestyle intervention in question resulted in improvements in depression, body-image distress [38], and disordered eating behavior [44] in women with PCOS, independent of weight loss. It is notable that the three-component therapy included cognitive behavioral therapy. Given this result and the fact that an exclusive focus on weight may have negative consequences, future research should examine whether a weight-neutral approach to lifestyle treatment, also including psychological support, has longer-term benefits for women with PCOS.
A strength of this study is the unique population-based data set that is not hospital-derived and the high participation and response rates. The prevalence of weight-loss attempts in women with PCOS has not, to our knowledge, been previously investigated at a population level, with long follow-ups and controls for several confounders.
Moreover, weight perception is a novel approach to weight-loss studies in women with PCOS.
The self-reported diagnosis of PCOS can be considered a limitation. However, the validity of the self-reported diagnosis has been assessed in our previous studies [7,27,28]. Furthermore, the genetic architecture has been shown to be similar among self-reported PCOS diagnosis and PCOS diagnosis by NIH criteria [45]. Unfortunately, the question regarding weight-loss attempts did not include information about the age when the weight-loss attempts occurred or provide details on weight-loss methods. The leisure-time physical activity questionnaire used in this study has not been validated in our data set, which can be seen as a limitation. However, previously, the variable has shown concordance with accelerometer-measured physical activity levels [30]. The study sample included only Caucasian individuals; therefore, the results may not be applicable to other ethnic groups.