Two points on a 100-point diet score separated women newly diagnosed with PCOS from those without it. What a Mecca study of 300 women can and cannot say

Polycystic ovary syndrome is the most common hormonal condition in women of reproductive age, with global estimates of roughly 9 to 12 percent. It is diagnosed when two of three signs are present: irregular or absent ovulation, higher androgen levels (on a blood test or as signs such as excess hair growth), and many small follicles on an ovarian ultrasound. Diet comes up in almost every conversation about it, usually as single nutrients or single foods. A case-control study published on 30 October in Frontiers in Nutrition looked instead at the whole plate.

The team recruited 300 women aged 20 to 35 at Umm Al-Qura University Hospital in Mecca, Saudi Arabia. Half had been diagnosed with PCOS within the previous three months; the other half came from orthopaedic, ear-nose-throat and pre-surgery clinics, had regular cycles, and had blood tests and ultrasounds to rule PCOS out. Women on metformin, GLP-1 drugs, hormonal contraceptives or a special diet in the past three months were excluded. Trained interviewers took each woman through a 152-item food questionnaire validated for Saudi adults, covering the past year, and the answers were scored on the Diet Quality Index-International: up to 20 points for variety, 40 for adequacy (enough fruit, vegetables, grains, fibre, protein, calcium, iron and vitamin C), 30 for moderation (fat, saturated fat, cholesterol, sodium and 'empty calorie' foods) and 10 for balance between macronutrients and fats.

The women with PCOS scored a median of 68, against 70 for the others. The difference came from adequacy and moderation; variety and balance were the same. Women with PCOS reported a median of 430 g of fruit a day against 500 g, 300 g of vegetables against 362 g, and 565 g of low-fat dairy against 670 g. When everyone was split into thirds by score, the top third (above 70.5) held 35 women with PCOS and 65 without; the bottom third (66 or less) held the reverse. After allowing for age, BMI, smoking, fibre, physical activity and total calories, a top-third diet went with 70% lower odds of having PCOS (odds ratio 0.30). Sweets, sugary drinks, refined grains and red meat did not differ significantly between the groups.

Two details deserve a second look. First, the women with PCOS got slightly more of their calories from unsaturated fats, the kind usually called healthy (13.0% against 12.2% from monounsaturated fat), which the authors suggest may reflect which foods the fat came from rather than how much. Second, the whole sample scored well on variety and poorly on moderation, a median of about 7 out of 30 in both groups, so 'higher quality' here mostly means a bit more fruit, vegetables and low-fat dairy within diets that scored poorly on the fat, salt and empty-calorie limits for everyone. The authors list the limits plainly: diet and diagnosis were measured at the same time, so the study cannot show which came first; food questionnaires depend on memory; the women came from one hospital; PCOS types were not analysed separately; and no hormone or insulin tests were linked to diet scores. The best use of the finding is as one more reason the advice usually given in PCOS clinics, more fruit and vegetables and fewer energy-dense, nutrient-poor foods, is worth taking seriously, not as proof that diet prevents the condition.