Health-related quality of life in women with polycystic ovary syndrome attending to a tertiary hospital in Southeastern Spain: a case-control study

This was a case-control study conducted from September 2014 to May 2016 at the Department of Obstetrics and Gynecology of the University Clinical Hospital “Virgen de la Arrixaca” in the Murcia Region (southeastern Spain). The study conception and design have been previously described elsewhere [33]. Women were excluded if they: were < 18 and > 40 years old, had endocrine disorders (e.g. Cushing’s syndrome, congenital adrenal hyperplasia, androgen-secreting tumors, hyperprolactinemia and hyper- and hypothyroidism) or were taking any hormonal medication (including contraception) during the 3 months prior to the study; were pregnant or lactating; had been exposed to oncological treatment; or had genitourinary prolapse. For both groups, women with PCOS and controls, gynecologists recruited consecutive women attending the clinic (total n = 307), and more than 95 % of the approached women fulfilling the study criteria agreed to participate (n = 14 declined and n = 23 were excluded). Those that declined to participate was due to a lack of time for filling out questionnaires. Women with PCOS (n = 117) were women attending the gynecology unit of the hospital, and included newly diagnosed cases as well as prevalent ones. Diagnosis of PCOS was established following the Rotterdam criteria [31] which included a complete medical history with a modified Ferriman-Galwey (mF-G) score [19], transvaginal ultrasound (TVUS) and serum sexual hormones. Diagnosis of PCOS required completion of at least two of the following three criteria: (i) hyperandrogenism either biochemical (total testosterone level ≥ 2.6 nmol/l) or clinical (mF-G score ≥ 12) [1] with or without acne or androgenic alopecia; (ii) oligo- and/or anovulation (menstrual cycles > 35 days or amenorrhea > 3 months); (iii) polycystic ovarian morphology (POM) on TVUS (≥12 follicles measuring 2–9 mm in diameter, mean of both ovaries) [15]. Possible phenotypic subtypes were: phenotype A (oliganovulation + hyperandrogenism + polycystic ovary morphology); phenotype B (oliganovulation + hyperandrogenism); phenotype C (hyperandrogenism + polycystic ovary morphology); and phenotype D (oliganovulation + polycystic ovary morphology) [22]. Finally, A, B and D phenotypes were reclassified as “anovulatory phenotypes” (n = 84) and phenotype C as “ovulatory phenotype” (n = 33) and evaluated separately in the current study.

The Short Form (SF)-12v2 Health Survey is a validated shorter version of the SF-36 generic questionnaire that encompasses 12 items, evaluating physical and mental health from the participant’s point of view (4 weeks recall period) [20, 30, 39, 40]. The questionnaire generates eight scales: physical functioning, role physical, bodily pain, general health, role emotional, vitality, social functioning and mental health. All raw scale scores were converted to a 0–100 scale, with higher scores representing higher levels of HRQoL. Additionally, the subscales were also transformed to normative-based scores according to the SF-12v2 recommendations to give a mean of 50 and a standard deviation of 10, using a representative sample of the 1998 US general population [20, 30, 40]. This transformation allows to obtain two summary measures: Physical and Mental Component Summary (PCS and MCS, respectively) that may be directly compared with other scales and scores. As the mean score is set to 50, scores ≥50 or < 50 indicate better or worse physical or mental health than the 1998 US general population. Scores bounds are set at 48 (0.2 SD) for a small effect on HRQoL, 45 (0.5 SD) for a moderate influence and, f ≤ 42 (0.8 SD) for a large effect on HRQoL [13, 14].

Descriptive statistics are presented using raw data. Continuous variables were compared using unpaired Student T tests, and categorical variables with chi-squared. Analysis of covariance was employed to calculate adjusted crude (0–100) and norm-based scales and component summaries differences between women with PCOS and controls. Multiple logistic regression was used to explore associations between women with PCOS and controls and norm-based scales and summary measures’ score with a cut-off of above/below 50 using odds ratios (OR) and 95% confidence intervals (95%CI). In both cases, several relevant covariates were considered (e.g. age, BMI, infertility problems, educational level, current employment, etc.) as potential confounders. When inclusion of a potential covariate resulted in a change in the β-coefficient of more than 10%, the variable was retained in the final models. These variables included factors previously related to PCOS in this or other studies, regardless of whether they had been previously described as predictors of PCOS. From previous publications [28, 29], we aimed to detect a difference of at least 3 points (with a standard deviation of about 7 points) in the global scores (PCS or MCS) between women with PCOS and controls. For an of alpha error of 0.05 and 80% statistical power to detect differences, a minimum of 90 women would be required in each group. All tests were two-tailed at 0.05 significance level. Analyses were conducted with IBM SPSS 25.0 (IBM Corporation, Armonk, New York, USA).