Table 2.
Studies evaluating postdiagnosis total dietary fat in relation to breast cancer and all-cause mortalitya
| Reference | Location | Design | Sample (n) | Dietary assessment method |
Contrast | RR/HR (95% CI) for breast cancer mortality |
RR/HR (95% CI) for all-cause mortality |
Covariates |
|---|---|---|---|---|---|---|---|---|
| Newman et al. 1986 (42) | Canada | Prospective cohort study (follow-up of cases from a case-control study) |
300 females with nonmetastatic disease Age: 3 5–74 years 87 deaths 73 breast cancer deaths |
Diet history, 24-hour recall Diet record |
RR of dying from breast cancer associated with greater than average daily total fat intake (77.7 g) near diagnosis |
0.99 (p = 0.963) | Not available | Weight near time of diagnosis |
| Nomura et al. 1991 (43) | United States (Hawaii) |
Prospective cohort study (follow-up of cases from a case-control study) |
182 Japanese 161 Caucasian Age: 45–74 years |
Diet history interview |
High versus low fat intake (g/wk) |
In Japanese women: 0.66 (0.25–1.76) In Caucasian women: 3.17 (1.17–8.55) |
Not available | Disease stage, menopausal status, obesity index, estrogen use |
| Ewertz et al. 1991 (20) | Denmark | Prospective cohort study |
2,445 patients with stage I—III invasive breast cancer Age: < 70 years 805 deaths |
FFQ | Fat consumption (quartiles) Quartile 4 versus 1 |
Not available |
0.96 (0.75–1.22) |
Tumor size, skin invasion, number of positive lymph nodes, grade, and stratified for residence |
| Rohan et al. 1993 (50) | Australia | Prospective cohort study (follow-up of cases from a case-control study) |
412 cases Age: 20–74 years 112 breast cancer deaths |
FFQ | Total fat (g) Quintile 5 versus 1 |
1.40 (0.66–2.96) | Not available | Energy, age at menarche, BMI |
| Holmes et al. 1999 (25) | United States |
Prospective cohort study |
1,982 women with invasive breast cancer 686 premenopausal 1,267 postmenopausal |
FFQ | Total fat (g) Quintile 5 versus 1 |
1.44 (1.01–2.04) | 1.51 (1.02–2.24) | Age, diet interval, calendar year of diagnosis, BMI, oral contraceptive use, menopausal status, HRT, smoking, age at first birth and parity, number of metastatic lymph nodes, tumor size, energy |
| Goodwin et al. 2003 (22) | Canada | Prospective cohort study |
477 women with surgically resected breast cancer Age <75 years 52 deaths 50 breast cancer deaths |
FFQ | Quadratic form of: Total fat (g/d) Quintile 1 versus 5 |
Not available | 2.9 (p = 0.10) 1.7 (p = 0.03) |
Age, energy, BMI, tumor and nodule stage, HRT, adjuvant chemotherapy |
| Total fat (% energy) Quintile 1 versus 5 |
||||||||
| Chlebowski et al. 2006 (15) |
United States |
RCT | 2,437 women with resected early-stage breast cancer receiving conventional cancer management Intervention: n = 975 Control: n = 1,462 Age: 48–79 years |
Intervention: low-fat diet postdiagno- sis (20% of energy from fat) Both groups had 30% of energy from fat at baseline |
Overall survival in intervention versus control |
Not available | 0.89 (0.65–1.21) | Systemic adjuvant therapy, nodal status, tumor size, mastectomy |
| Recurrence-free survival in intervention versus control |
0.71 (0.53–0.94) | |||||||
| Disease-free survival in intervention versus control |
0.81 (0.65–0.99) | |||||||
| Relapse-free survival in intervention versus control |
0.76 (0.60–0.98) | |||||||
| Pierce et al. 2007 (46) | United States |
RCT | 3,088 women previously treated for early breast cancer Age: 18–70 years Intervention: n = 1,537 155 deaths 127 breast cancer deaths Control: n = 1,551 160 deaths 135 breast cancer Deaths |
Intervention: Very high vegetable, fruit, and fiber and low-fat diet Control: 5-a-day dietary guidelines |
Intervention effects: Fat (% energy) Quartile 4 versus 1 |
Not available | 0.98 (0.64–1.49) | Anti-estrogen use, oophorectomy status, and stratification factors (tumor stage, clinic site, and age) |
| Beasley et al. 2011 (9) | United States |
Prospective cohort study |
4,441 women with history of invasive breast cancer and no history of recurrence Age: 29–70 years 525 breast cancer Deaths |
FFQ | Total fat (% energy) |
Age, state of residence, menopausal status, smoking, disease stage, alcohol, history of HRT, energy, treatment, BMI, physical activity |
||
| Quintile 5 versus 1 |
0.92 (0.53–1.60) | 1.05 (0.79–1.39) | ||||||
Abbreviations: BMI, body mass index; FFQ, food frequency questionnaire; HR, hazard ratio; HRT, hormone replacement therapy; RCT, randomized controlled trial; RR, rate ratio.