This case–control study was carried out on a total population of 969 women (453 patients and 516 controls) aged 19 to 80 (mean age: 44.8 ± 10.8). Comparing the characteristics between case and control was previously shown [30]. Four dietary patterns were derived using the FA. Four major nutrient patterns were extracted through factor analysis which could explain 66.49% of participants’ overall intakes. The factor loadings of each nutrient across the main nutrient patterns are reported in Table 1. The first nutrient pattern, associated with higher values of animal protein, retinol, riboflavin, pantothenic acid, cobalamin, and calcium was characterized as “dairy, eggs, and fatty fish”. This pattern has described 21.53% of overall nutrient intakes. The second nutrient pattern, full of SFA, MUFA, PUFA, and TFA and correlated with a lower intake of niacin, was named “animal protein”. 17.57% of all nutrient intakes were explained by this nutrient pattern. Higher intakes of fiber, vitamin C, and potassium have been reported in the third nutrient pattern. So, it was characterized as “vegetables” and includes 16.25% of overall nutrient intakes. Nutrient pattern 4, correlated with higher amounts of vegetable protein, alpha-tocopherol, and magnesium was considered a “nuts and seeds” nutrient pattern.

The general characteristics of participants across tertiles of nutrient patterns have been described in Table 2. Educational level was higher in subjects in the highest tertile versus the lowest tertile of the first nutrient pattern (P < 0.01). Women in the third tertile of nutrient pattern 2 were younger (P = 0.005) and had lower BMI levels (P = 0.007) in comparison with the first tertile. On the other hand, the mean age (P = 0.007) and physical activity (P =  < 0.01) of participants were higher in the highest tertile compared with the lowest tertile of the nutrient pattern 3. People in the highest tertiles of nutrient pattern 4 were also more physically active (P < 0.01) and had higher educational levels (P = 0.03).

The energy-adjusted dietary intakes of participants across tertiles of nutrient patterns have been shown in Table 3. Consumption of animal protein, carbohydrate, saturated fatty acids (SFAs), fiber, cholesterol, retinol, beta-carotene, calcium, magnesium, iron, zinc, selenium, potassium, vitamins B1, B2, B5, B6, B12, folate and vitamin C were significantly higher in the third tertile compared with the first tertile of nutrient pattern 1 (P < 0.01). However, significantly lower intakes of niacin, monounsaturated fatty acid (MUFA), polyunsaturated fatty acids (PUFAs), and trans fatty acids (TFAs) have been observed in highest versus lowest tertile of this nutrient pattern. Intakes of vegetable protein, cholesterol, retinol, cobalamin, alpha-tocopherol, SFA, mono-unsaturated fatty acids (MUFAs), PUFAs, and TFAs were significantly higher in the third tertile compared with the first tertile of nutrient pattern 2. While the overall consumption of animal protein, carbohydrate, fiber, beta carotene, thiamin, riboflavin, niacin, pantothenic acid, folate, vitamin C, calcium, magnesium, iron, zinc, selenium and potassium significantly reduced in highest tertile in comparison with the lowest tertile of the second nutrient pattern. The highest tertile of nutrient pattern 3, compared to the lowest tertile, was significantly associated with higher intakes of carbohydrate, fiber, beta carotene, vitamins B1, B2, B5, B12, C, and folate, alpha-tocopherol, calcium, magnesium, iron, and potassium, and lower intakes of cholesterol, retinol, niacin, zinc, selenium, SFA, PUFA, MUFAs, and TFAs. Higher, statistically significant consumption of animal protein, vegetable protein, carbohydrate, fiber, cholesterol, beta carotene, thiamin, riboflavin, pantothenic acid, folate, cobalamin, vitamin C, alpha-tocopherol, calcium, magnesium, iron, zinc, selenium, PUFAs, and potassium was observed in the highest tertile of nutrient pattern 4 compared to the lowest tertile. However, the third tertile of this nutrient pattern compared to the first tertile was significantly associated with lower intakes of niacin, SFA, MUFA, and TFA.

Crude and adjusted Odds Ratios and 95% CIs of breast cancer across tertiles of each nutrient pattern have been shown in Table 4. No significant association was observed between the first nutrient pattern and the risk of breast cancer either in crude or adjusted models. A significant direct relationship has been shown between the second nutrient pattern and the risk of breast cancer after controlling for age and energy intake in model A (OR = 1.41, 95% CI: 1.03,1.95, P = 0.03). However, no increase in the risk of breast cancer was observed by adherence to the second pattern after further adjustments for physical activity, educational level, family history of breast cancer, smoking, marriage status, and BMI in model B. Adherence to the third nutrient pattern has been associated with a reduction in the risk of breast cancer after adjusting for age and energy intake in the first model (OR = 0.69, 95% CI: 0.50, 0.95, P = 0.02). This relationship was also significant after further adjustments for other potential covariates in model B (OR = 0.70, 95% CI: 0.50, 0.97, P = 0.03). The relationship between nutrient pattern 4 and the risk of breast cancer was not significant in crude or adjusted models. There was no significant interaction effect between nutrient patterns and menopausal status for risk of BC in whole population after controlling for confounders.

Table 5 shows the stratified association between nutrient pattern and the risk of breast cancer among pre and post-menopause subjects. Adhering to the third nutrient pattern was indirectly associated with breast cancer risk in crude and adjusted models among premenopausal women (OR = 0.59, 95%CI: 0.39–0.89, P = 0.01). However, the relationship between other nutrient patterns and the odds of breast cancer was not significant. On the other hand, A significant inverse relationship was estimated between the first nutrient pattern and the risk of breast cancer after adjusting for age and total energy intake in the first model among post-menopausal women. This association was not significant in the model B after further controls for physical activity, BMI, marital status, education, previous history of breast cancer, smoking, and parity. Moreover, following the second nutrient pattern (animal protein pattern) was correlated with an increase in the risk of breast cancer in crude and partial adjustment models for age and energy intake. However, this relationship was no longer significant in the fully adjusted model among this population. No significant association was observed between other nutrient patterns and the risk of breast cancer among post-menopausal participants.