A systematic review of models of care for polycystic ovary syndrome highlights the gap in the literature, especially in developing countries

Introduction The aim of the study was to identify available polycystic ovary syndrome (PCOS) models of care (MoCs) and describe their characteristics and alignment with the international PCOS guideline. Methods Ovid MEDLINE, All EBM, PsycINFO, Embase, and CINAHL were searched from inception until 11 July 2022. Any study with a description of a PCOS MoC was included. Non-evidence-based guidelines, abstracts, study protocols, and clinical trial registrations were excluded. We also excluded MoCs delivered in research settings to minimize care bias. Meta-analysis was not performed due to heterogeneity across MoCs. We describe and evaluate each MoC based on the recommendations made by the international evidence-based guideline for assessing and managing PCOS. Results Of 3,671 articles, six articles describing five MoCs were included in our systematic review. All MoCs described a multidisciplinary approach, including an endocrinologist, dietitian, gynecologist, psychologist, dermatologist, etc. Three MoCs described all aspects of PCOS care aligned with the international guideline recommendations. These include providing education on long-term risks, lifestyle interventions, screening and management of emotional well-being, cardiometabolic diseases, and the dermatological and reproductive elements of PCOS. Three MoCs evaluated patients’ and healthcare professionals’ satisfaction, with generally positive findings. Only one MoC explored the impact of their service on patients’ health outcomes and showed improvement in BMI. Conclusion There is limited literature describing PCOS MoCs in routine practice. Future research should explore developing cost-effective co-created multidisciplinary PCOS MoCs globally. This may be facilitated by the exchange of best practices between institutions with an established MoC and those who are interested in setting one up. Systematic review registration https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=346539, identifier CRD42022346539.


Introduction
Polycystic ovary syndrome (PCOS) is one of the most common endocrinopathies among women of reproductive age, with a prevalence of 8%-13%, depending on the phenotype and the diagnostic criteria used (1). The diagnostic features of the disease are clinical and/or biochemical hyperandrogenism, oligo/ anovulation, and a polycystic morphologic appearance of the ovaries (2,3). PCOS was originally perceived as a reproductive disorder. However, mounting evidence suggests that PCOS is also a metabolic condition associated with overweight/obesity (4,5), type 2 diabetes mellitus (T2DM) (6,7), fatty liver disease (8)(9)(10), and cardiovascular disease (CVD) (11,12). It also has a significant psychological burden that is more than just a consequence of the physical symptoms of PCOS (13)(14)(15)(16). Kempegowda et al. proposed an 'iceberg phenomenon' to highlight the neglected and overlooked impact on various aspects of women's and individuals; with PCOS, alongside potential reproductive dysfunction ( Figure 1) (17).
Several studies have shown that women and individuals with PCOS often have a significant delay in diagnosis and are dissatisfied with their diagnostic experience, information provision, and the management of their PCOS (18)(19)(20)(21). Qualitative research has shown that women and individuals with PCOS often felt they were not taken seriously by their doctors (22) and that care fell short of their expectations due to limited evidence-based treatment options (23). The International PCOS Guideline (24) recommends patientcentric models of care (MoC) that meet the needs of women and individuals with PCOS across the complexity of clinical features.
An MoC is generally conceptualized as an overarching provision of care that is codesigned with end-users, may be shaped by a theoretical basis, and aligns with evidence-based practice and defined standards (25,26). A holistic, best-practice PCOS MoC would entail access to primary care, endocrinologists, gynecologists, dermatologists, dieticians, and psychologists as required to educate women and individuals with PCOS about their condition and its long-term consequences, address cardiometabolic, reproductive, and dermatological issues, and provide lifestyle interventions and psychological and emotional support ( Figure 2) (24). In the US and Australia, some MoCs have been implemented in accordance with the international guidelines. The involvement of a psychologist and cognitive-behavioral therapy in PCOS resulted in greater weight loss, improved quality of life, and reduced depression and anxiety (27). However, there is no literature comparing findings across MoCs to advance best practices that can be shared and adopted in other places where women and individuals with PCOS are managed. Furthermore, it is also important to consider the settings for the MoCs, as they can be significantly different in low-and middle-income countries (LMICs) due to resource and healthcare system constraints.

Objective
The aim of the study was to describe the characteristics of available MoCs for PCOS, their alignment with international guidelines, and their evaluation of outcomes.

Eligibility criteria, information sources, and search strategy
This systematic review was registered on PROSPERO (CRD42022346539). Studies describing MoCs that have more than one specialty in their PCOS management were identified using a search strategy created using MEDLINE limited to English language and human studies. The search strategy was then adapted to different electronic databases. Ovid MEDLINE, All EBM, PsycINFO, Embase, and CINAHL were searched from inception until 11 July 2022. We also included articles identified by experts (CT) that might be relevant to the study. A full search strategy can be found in Supplementary 1. Studies were included if they described models of care for PCOS. Any study reported in English with a detailed description of a PCOS MoC was included. Non-evidence-based guidelines, abstracts, study protocols, and clinical trial registrations were excluded. We also excluded MoCs delivered in research settings to minimize care bias. Detailed reasons for exclusion can be found in Supplementary 2.

Study selection
The process for study selection is summarized in Figure 3. Titles and abstracts were independently screened by two reviewers (EM and MD) utilizing Covidence software (Covidence systematic review software, Veritas Health Innovation, Melbourne, Australia). Following title and abstract screening, EM and MD screened full texts against the eligibility criteria. Conflicts were resolved following a discussion between the two reviewers and, if needed, by a senior reviewer (PK).

Data extraction
The researchers (EM, MD, and PK) developed the data extraction template in partnership with the PCOS GDG members A B FIGURE 2 Detailed description of the best-practice PCOS MoC aligned with the international evidence-based guideline for the assessment and management of polycystic ovary syndrome (2018) (24). (A) outlines 10 recommendations arranged in alphabetical order. The red arrow is the pointer for the starting point of reading the hexagon at cardiometabolic screening. (B) outlines the descriptions for the components of MoC (CT, JB, and MT) to ensure relevance. Data extracted included the service name, a detailed description of the MoC and the service, management, and evaluation.

Assessment of risk of bias
Risk of bias assessments were done using the Monash Centre for Health Research and Implementation (MCHRI) evidence synthesis critical appraisal templates, adapted from the relevant Cochrane critical appraisal tool(s) for mixed-method studies and crosssectional studies (28). For each study, external and internal validity was assessed to determine the overall risk of bias for that study.
The findings of this review are reported based on the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines (29). Data are summarized in Table 1 with a narrative synthesis. Meta-analysis was not performed due to heterogeneity across MoCs. We describe and evaluate each MoC based on the recommendations made by the international evidencebased guideline for assessing and managing PCOS (24).

Principal findings
A total of 3,671 articles were identified through title and abstract screening. Of these, 51 articles underwent full-text screening, of which six articles describing five MoCs are included in this report ( Figure 3). Bekx, Connor, and Allen (30) and Geier, Bekx, and  (39). MoC E described the process of the models of care, including retrospective chart analysis of profiles of women attending the clinic (35). A summary of these MoCs is included in Table 1.

Characteristics and composition of PCOS MoCs
All the MoCs included had a multidisciplinary approach, but their compositions varied. MoC A was one of the first published MoCs for women and individuals with PCOS (30,31). Starting in 2005, it had a team of two pediatric endocrinologists, a pediatric gynecologist, a reproductive endocrinologist, a nutritionist, and a psychologist. MoC B, established in 2012, had a general practitioner (GP), a women's health nurse, a dietician, and a women's health worker (32). Set up in 2012, MoC C included pediatric endocrinologists, gynecologists/adolescent medicine specialists, psychologists, nutritionists, and exercise physiologists (33). A dermatologist was added to the MoC two years later following patient feedback. MoC D, set up in 2017, was an integrated public multidisciplinary service that comprised specialties including endocrinology, dermatology, health coaching, and dietetics (34). PRISMA chart describing the selection process for our systematic review.  Patients were referred to each specialist clinic when required. MoC E described a one-stop MoC involving a gynecologist, infertility specialist, dermatologist, psychiatrist, nutritionist, yoga expert, and counselor; Women were managed in the clinic regularly (once monthly) (35). Detailed descriptions and characteristics of the MoCs are presented in Table 1 and Figure 4.

Services provided in the MoC
Three clinics-MoC C, MoC D, and MoC E-reported all aspects of PCOS care in line with the international guidelines. All except MoC A had clear information about the criteria they used for diagnosing PCOS.

Cardiometabolic disease
All MoCs described some form of cardiometabolic screening, but the content varied. MoC A (30, 31) screened for anthropomorphic characteristics including height, weight, and body mass index (BMI). They also monitored trends in BMI over time to define successful weight loss or gain. A 2-hour oral glucose tolerance test including insulin levels and lipid profile was used to screen for dysglycemia, hyperinsulinemia, and dyslipidemia, respectively. MoC B (32) had all screening done by MoC A with the addition of glycated hemoglobin (HbA1c) and blood pressure measurements. MoC C (33) evaluated BMI, blood pressure, lipid profile, and HbA1c. MoC D (34) included screening for long-term complications. However, the individual components of how this was done were not included in the study. MoC E (35) included BMI screening, waist-hip ratio, ultrasound to assess for non-alcoholic fatty liver disease, and screening for metabolic syndrome, including a 2-hour oral glucose tolerance test, insulin, and lipid profile.

Lifestyle
All MoCs provided lifestyle interventions, mostly goal setting and education. In MoC A (30,31), the health psychologist focused on lifestyle changes and helped women and individuals with PCOS identify any barriers that might exist and possible solutions. The nutritionist helped provide education on the role of insulin, meal planning, goal setting, and exercise. In MoC B (32), patients were encouraged to set their own lifestyle goals, which included reducing portion sizes and increasing daily walks. Patients were then asked to

Dermatology
All MoCs except MoC B (32) described either screening or treatment for dermatological issues associated with PCOS. MoC A (30) described screening for hirsutism and acne. The screening tools used were not specified in the study. MoC C (33) measured hirsutism using the modified Ferriman-Gallwey (mFG) score. Screening for the presence and severity of acne was done during a physical examination. The presence and absence of acanthosis nigricans, androgenic alopecia, and hidradenitis suppurative were also noted. As for treatment, MoC C (33) used spironolactone, topical treatments, antibiotics, and isotretinoin to manage hirsutism and acne. The dermal clinic integrated into MoC D (34) used a medical-grade laser for hirsutism. MoC E (35) had a dermatologist within the MoC to address acne, oily skin, acanthosis nigricans, and/or hirsutism; however, no specific treatments were described.

Education on long-term risk
MoCs C and D were the only MoCs that reported education on long-term risks (33,34). MoC C (33) set up a group education session where women and individuals with PCOS were taught by endocrinologists and gynecologists on the pathophysiology and medical treatment of PCOS. Due to the COVID-19 pandemic, they introduced recorded content for these sessions. MoC D (34) educated women and individuals with PCOS regarding clinical features, diagnosis, complications, and management of PCOS via a group session or printed fact sheets during the first appointment. MoC E (35) counseled women on the condition and the need for integrated multidisciplinary management following the diagnosis of PCOS.
Emotional well-being and reproductive screening and/or management Three clinics described the provision of screening for emotional well-being and reproductive health. It was unclear whether MoC A provided emotional and reproductive screening. However, we note that both included health psychologists and a pediatric gynecologist in the clinic. At MoC B (32), emotional distress screening was undertaken with the Kessler Psychological Distress Scale (37), which is a global measure of distress encompassing anxiety and depression items. A psychologist in MoC C (33) evaluated all patients for mental health symptoms, appetite self-regulation, and emotional eating. In MoC D (34), all women and individuals with PCOS were screened using a modified PCOS questionnaire (PCOSQ) (36) and the Hospital Anxiety and Depression Scale (HADS) (38) to evaluate their quality of life and emotional Graphical representation of the five models of care (A-F) for polycystic ovary syndrome included in this systematic review. Green represents the services that were provided in a model of care. Yellow represents that the element of MoC was either not reported or unavailable in their MoC.
distress, respectively. In MoC B (32), reported infertility treatment included lifestyle intervention, metformin prescription, and/or referral to a specialist. MoC C (33) included screening for endometrial hyperplasia and discussion regarding infertility issues, whereas MoC D (34) included family planning discussion. Women in MoC E (35) were screened for obvious anxiety and/or depression by counselors and addressed by a psychologist or psychiatrist.

MoC evaluation
MoC evaluation data were organized into three categories: patient outcomes, health professional outcomes, and other outcomes. MoCs B (32) and MoC D (34) were the only studies that reported their MoCs evaluation. MoC B evaluated outcomes from all three categories while MoC D only evaluated patient and other outcomes. Evaluation of patient outcomes was available for MoC A which investigated the impact of their service on BMI. No evaluation outcomes were available for MoC C.

Patient health outcomes and satisfaction
MoC A (31) evaluated patients' health outcomes, including improvements in body weight seen in 36% (n = 13/36). Having access to both a psychologist and a dietician was superior for improving weight compared to seeing either alone. MoC B (32) conducted semi-structured interviews and focus groups with women and individuals with PCOS to assess their satisfaction with the clinic. Women and individuals with PCOS found it helpful to have access to this clinic, and they found the staff knowledgeable. Patients also found collaboration with a dietician helpful and valuable in goal setting but suggested more tailored plans and ongoing supervision, indicating insufficiency in what was provided. Overall, 80% (n = 12/15) of patients in MoC D (34) were satisfied with the service. Further semi-structured interviews with women and individuals with PCOS revealed that MoC D covered their multifaceted needs and was effective in providing care and communication. Women and individuals with PCOS also reported a positive impact of this clinic on medical management, symptom severity, their understanding of PCOS, confidence in managing PCOS, and emotional well-being. Suggestions from the interviews included improvements in efficiency, patient communication, resource provision, infrastructure, and awareness of service availability. Patients also suggested more resources to promote self-management. For MoC E (35), telephone feedback was obtained from 155 women who attended the clinic. One year following clinic attendance, 83.8% reported medication adherence, and 52.3% and 46.5% adhered to exercise and dietary interventions, respectively. Sixty-eight percent of women were convinced that a multidisciplinary clinic was helpful in weight reduction and psychological well-being.

Health professional satisfaction
Health professionals' satisfaction was investigated by Boyle et al. (2017) in MoC B (32). A survey of service providers found high levels of job satisfaction and professional investment. The service providers saw the absence of a psychologist as a particular problem. Barriers and enablers to clinic sustainability and service delivery were also discussed. Key barriers to sustainability included issues that may arise due to a lack of cover during leave, administrative support, funding, high staff turnover, and system issues. The increased demand for the service, although a strong reason to continue expanding the clinic, was cited as a barrier due to the lack of service providers' availability.

Risk of bias in the included studies
Five studies were deemed to have a low risk of bias by the reviewers. One study [MoC A (30)] had a moderate risk of bias due to inadequate information on case selection. Furthermore, the inclusion and exclusion criteria for the study were not described. The detailed risk of bias assessment for each included study is presented in Supplementary 3.

Comparison with existing literature
To date, little progress has been made towards establishing evidence-based PCOS MoCs. Existing MoCs vary considerably in the breadth of multidisciplinary features, with few covering all recommended aspects of care (cardiology, reproduction, dermatology, emotional well-being, lifestyle, and long-term risk). Moreover, it is important to note that some of these studies were not designed to evaluate their MoC, which accounted for the lack of details in each reported MoC. Lack of progress could be because models exist but are not published, health system constraints hinder development (funding, health policy), or there is a lack of knowhow about development. Good MoCs for PCOS may exist; however, without their publication, the opportunity to share best practices is lost. We also noted the lack of systematic reporting and evaluation of MoCs in PCOS, and here we have established a structure for capturing and reporting MoC characteristics to support future work. Future research should concentrate on the evaluation of routine MoCs with a focus on patients' experiences and satisfaction. This would enable the sharing of best practices in the care of women and individuals with PCOS.
The lack of progress in PCOS MoC evaluation in the literature is surprising considering the high prevalence of PCOS as a chronic condition. A systematic review of chronic disease MoCs reported that >90% of their included MoCs (n = 75/77) reported a positive impact on healthcare practices and outcomes (40). A recent systematic review of models of care in primary care for chronic diseases showed improved outcomes for physiological measures of disease, risk behavior, quality of life, health status, patient satisfaction, functional status, and knowledge levels with a model of care (41). Interestingly, the number of studies included was substantially higher in this review in 2014 compared to their previous study in 2006. This shows a growing interest in models of care, especially for complex chronic conditions. However, this is not seen in our review of MoCs in PCOS (in both developed and LMICs). There is a need for and an apparent benefit from multidisciplinary, dedicated one-stop clinics covering all aspects of PCOS, such as the MoCs by Tay (33). This is also in line with the study by , where people expressed the need for more PCOS-centric clinics (20). As the management of PCOS is largely individualized due to heterogeneity and a broad range of clinical features, having access to multiple disciplines is important (24). However, our results show that the integration of different disciplines varied considerably, yet four out of the five MoCs showed positive patient and/or healthcare professionals' satisfaction.
Our systematic review showed that none of the peer-reviewed MoCs is optimized in line with our suggested MoC structure for women and individuals with PCOS. Despite having all the services for women and individuals with PCOS described by Tay  There is a clear need for greater awareness and priority for this condition. This also impacts access to treatment options for PCOSrelated symptoms, such as expensive laser hair removal and electrolysis for hirsutism (24,43,45). Adequate dermatology management should be provided, as hormonal manipulation with contraceptive pills is not always effective and acne can cause significant mental health issues. Women and individuals with PCOS should be educated about subfertility due to anovulation and, more importantly, referred to a fertility specialist when indicated. As PCOS is also recognized as a metabolic condition, women and individuals with PCOS should be regularly screened for cardiovascular risks and informed of its long-term consequences. Because PCOS is also associated with endometrial cancer, education and public awareness regarding weight loss and progesterone use to reduce endometrial cancer risk are of paramount importance. Emotional well-being screening and appropriate referral are also important for women and individuals with PCOS due to the high prevalence of anxiety, depression, and reduced quality of life that go beyond the physical manifestations of PCOS. All of these would improve self-management strategies for women and individuals with PCOS when coupled with lifestyle interventions that can be provided by healthcare professionals, namely nutritionists, exercise physiologists, and lifestyle coaches. Moreover, it is important to ensure race, culture, and tradition are also factored in when designing an MoC, as these have been shown to influence the differential services received by women and individuals with PCOS (46)(47)(48)(49)(50). Other factors that need to be considered include distance to the health facility, affordability that could impact continuity of care, availability of essential treatments, diagnostic tools, trained staff, and coordination of care across public, private, and alternative healthcare providers. These are especially important in LMICs, as shown by the qualitative synthesis from a best-framework synthesis on models of care in LMICs by Lall et al. (51). Moreover, it is also important to consider PCOS MoCs at different life stages, considering the difficulty of diagnosing PCOS in adolescents. These make it vital to involve women, individuals with PCOS, and their families in co-designing services (49).
Many studies have shown that women and individuals with PCOS are generally dissatisfied with their diagnosis experience, the information provided, and the management of their PCOS (18,19,21,22,52). Patient satisfaction is also an important aspect of healthcare, as it has been shown to affect clinical outcomes and patient retention. Furthermore, patient satisfaction also affects the time and efficacy of healthcare delivery, which is often used as a proxy for the quality of healthcare (53). In addition, healthcare professional satisfaction is key to ensuring the productivity and sustainability of the service (54). In this context, surprisingly few studies focused on PCOS MoC, and most studies did not assess patients' and healthcare professionals' satisfaction. We have described satisfaction assessments for two MoCs (32, 34) with positive results. Our findings are like those of two studies describing an MoC based at the Royal Berkshire Hospital, UK, which were not included in this systematic review due to being published as conference abstracts without details of evaluation methodologies (55). An audit was conducted on their MoC to assess the adequacy of investigations and the efficacy of treatment for women and individuals with PCOS attending multidisciplinary clinics. Their patient satisfaction survey showed that 62 out of 63 women found the clinic useful and were happy with the results. They also reported high satisfaction and improved clinical outcomes such as weight loss, menstruation patterns, hirsutism, and physical activity levels (55,56). A further seven studies that might include PCOS MoC were also excluded from this systematic review because they were abstracts. Hebbar et al. investigated the prevalence of anxiety and body dysmorphia in women and individuals with PCOS attending PCOS specialist clinics in the UK and India (57). The components of their MoC were not described in the abstract (57). Abudu et al. also studied the patient characteristics and subjective improvements in acne for women and individuals attending multidisciplinary PCOS clinics without a description of specialists in the multidisciplinary team (58). Other studies excluded three studies that described either group counseling, self-management, and/or support services for women and individuals with PCOS (59-61).

Strengths and limitations
The strength of this review includes applying clear definitions of an MoC, which enables the capture of studies aligned with international guidelines. We also established a system to report MoCs; it is important to note that there might be another system that exists for an "optimal MoC." Our key weakness is related to the limited number of MoCs described internationally, and we note the included MoCs are from two high-income countries-the US and Australia. Therefore, we are unable to generalize our findings to a wider population. Furthermore, due to the design of the included studies, not every component of the MoCs included is captured in our findings. This does not mean that they did not provide the service. Despite only a small number of included studies, this systematic review provides a structured evaluation of the current MoCs for PCOS internationally and further explores their effectiveness.

Conclusion and implications
There are a limited number of models of multidisciplinary care currently available for PCOS, with a scarcity of data, especially in low-and middle-income countries. Good MoCs may exist, but without their publications, the opportunities to share best practices are lost. Studies on MoC that evaluated patients' and healthcare professionals' satisfaction were generally positive. Future work focusing on MoC scale-up should include the development of a best-practice MoC framework, co-designed with women and individuals affected by PCOS across different countries. Alignment with the updated best practice in the 2023 guideline will be important, along with adaptation to the range of health systems and resource settings, and a need for ongoing evaluation and sharing of results to further develop evidence based on realworld experiences.

Data availability statement
The original contributions presented in the study are included in the article Supplementary Material. Further inquiries can be directed to the corresponding author.

Author contributions
Both EM and MD were involved in all stages of the study and have contributed equally to this work and share the first authorship. KM contributed to the design of the study and data analysis. CT, JB, and MT supervised the data extraction and finalizing of the articles included in the study. Members of the PCOS SEva working group provided substantial contributions to the conception and design of the work and were involved in discussions at all stages of the study. AM and HT were involved in scoping the clinical question and eligibility criteria (PICO), overseeing the review methodology in alignment with approved PCOS guideline evidence synthesis processes, and reviewing and editing the manuscript. PK conceptualized the study and supervised all stages of data collection, analysis, interpretation, and write-up of this study. All authors contributed to the article and approved the submitted version.