Abstract
Background: The effects of dietary protein on bone health are controversial.
Objective: We examined the relation between protein intake with fracture and bone mineral density (BMD) within the Women's Health Initiative (WHI).
Design: This prospective analysis included 144,580 women aged 50–79 y at baseline in the WHI clinical trials (CTs) and observational study (OS) that recruited participants in 1993–1998 with follow-up through 2011. Self-reported clinical fractures were collected semiannually through the original end of the trials (WHI CTs) and annually (WHI OS) by questionnaires. Hip fracture was adjudicated by a central review of radiology reports. BMDs for total body, hip, and spine were measured at baseline and 3 and 6 y in 9062 women at 3 WHI clinics by using dual-energy X-ray absorptiometry. Protein intake was assessed via food-frequency questionnaire and calibrated by using biomarkers of energy and protein intakes. Associations between protein intake and fracture were estimated by using Cox proportional hazards regression, and the relation between protein intake and BMD was estimated by using linear regression.
Results: Median biomarker-calibrated protein intake was 15% of energy intake. Per 20% increase in calibrated protein intake (percentage of energy), there was no significant association with total fracture (HR: 0.99; 95% CI: 0.97, 1.02) or hip fracture (HR: 0.91; 95% CI: 0.84, 1.00), but there was an inverse association with forearm fracture (HR: 0.93; 95% CI: 0.88, 0.98). Each 20% increase in calibrated protein intake was associated with a significantly higher BMD for total body (mean 3-y change: 0.003 g/cm2; 95% CI: 0.001, 0.005 g/cm2) and hip (mean 3-y change: 0.002 g/cm2; 95% CI: 0.001, 0.004 g/cm2).
Conclusions: Higher biomarker-calibrated protein intake within the range of usual intake was inversely associated with forearm fracture and was associated with better maintenance of total and hip BMDs. These data suggest higher protein intake is not detrimental to bone health in postmenopausal women. The WHI program was registered at clinicaltrials.gov as NCT00000611.
INTRODUCTION
The effects of dietary protein intake on bone health are controversial. A supply of protein is required for bone maintenance, and low protein intake has adverse effects on bone health (1). However, high protein intake increases urinary calcium to counteract the acidifying amino acids released after protein digestion, and there has been debate over whether the source of the calcium is bone or increased intestinal absorption (2, 3). If increased intestinal absorption of calcium is the source, higher protein intake may be detrimental to bone if calcium intake is low (4). A systematic review of dietary protein, bone mineral density (BMD)4, and fracture risk studies reported a positive association between protein intake and BMD and an inverse association with bone resorption markers, but there was no significant association between protein and fracture risk (5). Another systematic review of health effects of protein intake in healthy adults suggested that previous studies that evaluated the association between protein and bone health were often weakened by limited information about the quality of the dietary assessment methods, use of measures that did not include total energy intake, or lack of distinction between animal and vegetable protein sources (6). Studies that have a large number of events with sufficient follow-up are needed to discern whether the positive association observed with BMD translates into a long-term benefit as measured by lower rates of fracture.
An approach for statistically correcting for the measurement error by using biomarkers for total energy and protein has been developed by investigators from the Women's Health Initiative (WHI) (7). This approach provides an opportunity to examine associations between diet and health outcomes while taking into account the measurement error that attends self-reported protein intake. This prospective analysis examines the role of biomarker-calibrated protein intake in bone health measured by the change in BMD (total, hip, and spine) and incidence of fracture (any, hip, spine, and forearm) in postmenopausal women in the WHI.
SUBJECTS AND METHODS
Study population
The WHI includes an observational study (OS; n = 93,676) and clinical trials (CTs; n = 68,132) of postmenopausal hormone therapy, dietary modification (DM), and calcium and vitamin D supplementation. As previously described, women aged 50–79 y were recruited between 1 October 1993 and 31 December 1998 at 40 clinical centers in the United States (8). This analysis included 144,580 women with follow-up through 2011 in women enrolled in the OS and CT components who reported plausible energy intakes on a food-frequency questionnaire (FFQ) (600–5000 cal/d) and had complete data for model covariates. All protocols were approved by institutional review boards at participating institutions, and all women documented a willingness to participate via signed informed consent forms.
Outcome ascertainment
Fracture
Total fractures were defined as all reported clinical fractures other than those of the ribs, sternum, skull or face, fingers, toes, and cervical vertebrae. Self-reported clinical fractures were collected by questionnaires semiannually through the original end of the CTs and annually thereafter; for the OS, fractures were reported annually. Participants were asked the following question: “Since (last reporting date), has a doctor told you that you had a broken, fractured, or crushed bone?” If the answer was “yes,” women were asked to answer the question, “Which bone did you break, fracture, or crush?” by designating 1) hip, 2) upper leg (not hip), 3) pelvis, 4) knee (patella), 5) lower leg or ankle, 6) foot (not toe), 7) tailbone (coccyx), 8) spine or back (vertebra), 9) lower arm or wrist (forearm), 10) hand (not finger), 11) elbow, 12) upper arm or shoulder, or 13) other (specify). Additional questions were asked regarding whether fractures were diagnosed or treated during an overnight hospital stay and whether an X-ray or imaging scan (MRI) was taken at the same medical facility where fractures were treated. Hip fracture was adjudicated by a central review of radiology reports. Other fracture outcomes (spine, forearm, and any fractures) were centrally adjudicated during the CTs and self-reported otherwise. On average, the agreement between self-reported fracture and medical records was >70% for single-site fractures, with a higher agreement for hip and forearm fractures compared with fractures at other sites (9).
Measurement of BMD
BMDs at the hip, posterior-anterior spine, and total body were measured at baseline and 3 and 6 y at 3 clinical centers (Pittsburgh, PA; Birmingham, AL; and Phoenix and Tucson, AZ) in 9062 women by using dual X-ray absorptiometry with a Hologic QDR densitometer (Hologic Inc). Standard protocols for positioning and analysis were used by trained technicians, and an ongoing quality assurance program was conducted.
Protein exposure
WHI FFQ
All WHI women completed the FFQ at baseline. The self-administered FFQ included 122 items for individual foods and food groups, 19 adjustment items, and summary questions (10). Protein intake was characterized as total intake (g), as a percentage of total kilocalorie intake (percentage of kcal), and relative to body weight (g/kg).
Calibrated protein estimation
As previously described (7), the WHI Nutritional Biomarkers Study was conducted in 2004–2005 to further assess the measurement properties of the FFQ by using objective biomarkers of total energy expenditure (equivalent to energy intake in weight-stable persons) and protein intake. A total of 544 women from the Dietary Modification trial participated in a doubly labeled water protocol to estimate total energy expenditure over a 2-wk period and a urinary nitrogen protocol to estimate protein consumption over a 24-h period to be compared with concurrent self-reported dietary intake data. These results showed that FFQ total energy was considerably underestimated, and protein was modestly underestimated, whereas the percentage of energy from protein was overestimated. Calibration equations were developed separately for energy, protein (g), and the percentage of energy from protein by using a linear regression of log-biomarker estimates on corresponding log-FFQ estimates, BMI (in kg/m2), age, and other participant characteristics.
Potential confounders
Information on all covariates was obtained by self-report at baseline. Baseline questionnaires ascertained information on race-ethnicity, history of fracture, and current and past smoking. Information was collected by self-report of several physician-diagnosed conditions. BMI was calculated from measured weight divided by height squared. Self-reported leisure physical activity was summarized as metabolic equivalent tasks (11). Dietary intake of calcium was measured by using a semiquantitative FFQ, and total calcium intake was defined as the sum of calcium from diet and supplements.
Energy intake was estimated from the FFQ and calibrated by using regression equations (7). Dietary supplement use was assessed by using an inventory-type questionnaire in which study staff recorded nutrients from participants’ bottles brought to a clinic visit. Smoking status was classified as current, past, or never. Postmenopausal hormone therapy was categorized as current, past, or never use of any estrogen with or without progestin.
Statistical analysis
Characteristics of women by quintile of calibrated protein intake (ie, calibrated percentage of calories from protein) at baseline were compared by using chi-square tests (for categorical variables) or ANOVA (for continuous variables). HRs for fracture per 20% difference in calibrated protein intake were computed from Cox proportional hazards survival models for each fracture outcome. SEs were estimated from a bootstrap procedure (1000 replicates), whereby the nutrient intake–calibration equations were refitted for each bootstrap sample.
Women contributed follow-up time until the occurrence of fracture, death, or end of follow-up, whichever came first. Models were stratified on the WHI component participation (ie, CT treatment arm; OS) and were adjusted for age, race-ethnicity, BMI, general health, physical activity, history of fracture at age ≥55 y, history of parental fracture, current smoking, hormone therapy use, corticosteroid use, glucocorticoid use, treated diabetes, and rheumatoid arthritis. For the survival modeling, the proportional hazards assumption was evaluated by examining plots of the baseline hazard as a function of the exposure variables of interest as well as by testing an interaction term of protein intake by the log follow-up time.
Linear regressions were used to assess the association of baseline BMD with protein intake as well as baseline, follow-up, and annualized changes in BMD according to protein intake. Mean BMDs by protein intake are presented with SEs estimated from a bootstrap procedure, as previously described.
The analysis was conducted in the combined CT and OS cohorts. As published by Howard et al (12), the Dietary Modification intervention significantly increased self-reported total dietary protein in the active intervention group. There was no effect of the DM intervention on fracture incidence, a small decrease in bone density, and an interaction between DM and hormone therapy intervention, whereby women assigned to the active intervention in both trials had a greater reduction in the occurrence of fracture (13). Thus, we first examined associations between calibrated protein intake and fracture and bone density within the CTs and OS separately and tested for an interaction (all P > 0.05) before combining cohorts. All analyses were conducted with SAS statistical software (version 9.3; SAS Institute Inc).
Analyses to examine the effect modification by key variables (age, BMI, race-ethnicity, and calcium intake) were conducted to determine whether associations between protein use and fracture or the change in BMD were apparent in key subgroups of women. Statistical tests for interactions were conducted for each of these variables to determine whether any stratum-specific differences were strong enough to interpret as potentially important.
RESULTS
Median calibrated protein intake was 15% of energy intake. Women who consumed a lower proportion of their calories from protein were more likely to be older, obese, nonwhite, have a personal, but not family, history of fracture, engage in less physical activity, be current smokers, report a lower health status, and have a history of arthritis (all P < 0.0001) (Table 1). The annualized incidence of clinical fracture was 2.6% for any fracture, 0.21% for hip fracture, 0.30% for spinal fracture, and 0.50% for forearm fracture.
TABLE 1.
Baseline characteristics by calibrated protein intake (calibrated percentage of calories from protein) in the WHI (n = 144,580)1
| Characteristic | Quintile 1 (<13.3%) | Quintile 3 (14.2–14.8%) | Quintile 5 (≥15.6%) |
| Age (y) | 66.0 ± 7.22 | 63.7 ± 6.9 | 59.6 ± 6.4 |
| BMI | |||
| Underweight (<18.5 kg/m2) | 262 (0.9) | 200 (0.7) | 332 (1.1) |
| Normal (18.5–24.9 kg/m2) | 8250 (28.7) | 9656 (33.7) | 12,357 (42.5) |
| Overweight (25.0–29.9 kg/m2) | 9481 (33.0) | 10,057 (35.0) | 10,242 (35.3) |
| Obese (≥30 kg/m2) | 10,780 (37.5) | 8781 (30.6) | 6121 (21.1) |
| Ethnicity | |||
| White | 22,283 (77.2) | 24,308 (84.7) | 24,725 (85.1) |
| Black | 4149 (14.4) | 2128 (7.4) | 1731 (6.0) |
| Hispanic | 1110 (3.9) | 981 (3.4) | 1189 (4.1) |
| American Indian | 157 (0.5) | 99 (0.3) | 116 (0.4) |
| Asian/Pacific Islander | 676 (2.3) | 781 (2.7) | 902 (3.1) |
| Unknown | 458 (1.6) | 397 (1.4) | 389 (1.3) |
| Family history of fracture | 9834 (34.2) | 10,657 (37.1) | 11,162 (38.4) |
| History of fracture (at age ≥55 y) | 4473 (15.5) | 3778 (13.2) | 2498 (8.6) |
| Calibrated energy intake (kcal)3 | 2122 ± 233 | 2143 ± 214 | 2143 ± 171 |
| Physical activity (METs/wk) | 9.9 ± 12.6 | 12.6 ± 13.6 | 15.0 ± 14.9 |
| Smoking | |||
| Never | 12,116 (42.1) | 15,368 (53.6) | 15,538 (53.5) |
| Past | 10,050 (34.9) | 12,544 (43.7) | 13,364 (46.0) |
| Current | 6607 (23.0) | 782 (2.7) | 150 (0.5) |
| Hormone use | |||
| Never | 14,687 (51.0) | 12,244 (42.7) | 10,903 (37.5) |
| Past | 5083 (17.7) | 4519 (15.7) | 4092 (14.1) |
| Current | 8981 (31.2) | 11,902 (41.5) | 14,033 (48.3) |
| Corticosteroid use | 299 (1.0) | 227 (0.8) | 200 (0.7) |
| Glucocorticoid use | 294 (1.0) | 215 (0.8) | 199 (0.7) |
| General health status | |||
| Excellent/very good | 14,309 (49.7) | 16,891 (58.9) | 19,226 (66.2) |
| Good | 10,868 (37.8) | 9466 (33.0) | 7942 (27.3) |
| Fair/poor | 3596 (12.5) | 2337 (8.1) | 1884 (6.5) |
| Medical history | |||
| Arthritis | 15,199 (52.8) | 13,823 (48.2) | 11,718 (40.3) |
| Rheumatoid arthritis | 1791 (6.2) | 1412 (4.9) | 1155 (4.0) |
| Diabetes (treated with pills or shots) | 1176 (4.1) | 1287 (4.5) | 1299 (4.5) |
P < 0.0001 for all baseline characteristics across quintiles of calibrated protein intake. P-value testing did not include participants who were not randomly assigned to the trial. MET, metabolic task hours; WHI, Women's Health Initiative.
Mean ± SD (all such values).
All values are geometric means ± SDs, because calibrated energy was back-transformed from the log scale.
Women who consumed 20% higher calibrated protein intake (percentage of energy) were 7% less likely to have a forearm fracture (95% CI: 2%, 12%), but there were no significant associations with any, hip, or spinal fractures (Table 2). When associations by quintiles and in other units (g/d and g · kg body weight−1 · d−1) were examined, associations differed in magnitude but remained consistent in the overall directionality (data not shown).
TABLE 2.
Risk of fracture per 20% increase in daily biomarker-calibrated protein in the WHI (n = 144,580)1
| Fracture site | No. of events | HR (95% CI) |
| Any fracture | 36,166 | 0.99 (0.97, 1.02) |
| Hip | 3286 | 0.91 (0.84, 1.00) |
| Spine | 4836 | 1.05 (0.98, 1.13) |
| Forearm | 7800 | 0.93 (0.88, 0.98) |
HRs were derived from Cox proportional hazard regression models adjusted for age, BMI, race-ethnicity, calibrated energy intake, general health, physical activity, history of fracture at age ≥55 y, history of parental fracture, current smoking, corticosteroid use, glucocorticoid use, treated diabetes, rheumatoid arthritis, and hormone use. WHI, Women's Health Initiative.
The directionality of associations by site were similar for BMD compared with fracture (Table 3). An increase in calibrated protein intake was associated with a significantly higher BMD (Table 3). Women who consumed 20% higher protein showed more positive changes in total BMD (0.004 g/cm2; 95% CI: 0.001, 0.007 g/cm2) after 6 y follow-up (Table 3). There were no longitudinal associations between protein intake and spine BMD. There were no significant interactions by race-ethnicity or calcium intake (data not shown). There was a significant interaction in the association between calibrated protein intake and risk of any fracture by BMI (Table 4). The strongest inverse associations between calibrated protein intake and any fracture risk was in women who had lower BMI [HR for women with BMI of 18.5 was 0.95 (95% CI: 0.90, 1.00) compared with 1.02 (95% CI: 0.98, 1.07) in women with BMI of 35); Table 4]. For BMD, there were no significant tests for interaction, and none of the subgroup analyses were significant (Table 5).
TABLE 3.
Change in mean BMD per 20% increase in the calibrated percentage of calories from protein in the WHI BMD Cohort1
| BMD site | n | BMD |
| g/cm2 | ||
| Total body | ||
| Baseline | 9062 | 0.009 (0.004, 0.016)2 |
| 3-y – baseline Δ | 7440 | 0.003 (0.001, 0.005) |
| 6-y – baseline Δ | 6522 | 0.004 (0.001, 0.007) |
| Hip | ||
| Baseline | 9062 | 0.010 (0.005, 0.017) |
| 3-y – baseline Δ | 7489 | 0.002 (0.001, 0.004) |
| 6-y – baseline Δ | 6553 | 0.003 (0.000, 0.005) |
| Spine | ||
| Baseline | 9062 | 0.014 (0.006, 0.023) |
| 3-y – baseline Δ | 7499 | 0.003 (0.000, 0.006) |
| 6-y – baseline Δ | 6457 | 0.003 (0.000, 0.008) |
Means were estimated from linear regression models adjusted for age, BMI, race-ethnicity, calibrated energy intake, general health, physical activity, history of fracture at age ≥55 y, history of parental fracture, current smoking, corticosteroid use, glucocorticoid use, treated diabetes, rheumatoid arthritis, and hormone use. BMD, bone mineral density; WHI, Women's Health Initiative.
Mean; 95% CI in parentheses (all such values).
TABLE 4.
Risk of any and hip fracture per 20% increase in calibrated protein intake (percentage of kcal) in the WHI at subgroup levels1
| Outcome | Any fracture | P-interaction | Hip fracture | P-interaction |
| Overall | 0.99 (0.97, 1.02) | — | 0.91 (0.84, 1.00) | — |
| Age at baseline | 0.106 | 0.429 | ||
| 55 y | 1.02 (0.96, 1.07) | 0.90 (0.75, 1.05) | ||
| 65 y | 0.99 (0.96, 1.01) | 0.91 (0.82, 0.99) | ||
| 75 y | 0.96 (0.91, 1.02) | 0.92 (0.83, 1.02) | ||
| BMI | 0.035 | 0.191 | ||
| 18.5 kg/m2 | 0.95 (0.90, 1.00) | 0.87 (0.74, 1.00) | ||
| 25.0 kg/m2 | 0.98 (0.95, 1.00) | 0.91 (0.83, 0.99) | ||
| 30.0 kg/m2 | 1.00 (0.97, 1.03) | 0.94 (0.85, 1.05) | ||
| 35.0 kg/m2 | 1.02 (0.98, 1.07) | 0.97 (0.83, 1.15) |
All values are HRs; 95% CIs in parentheses. HRs were derived from Cox proportional hazard regression models adjusted for age, BMI, race-ethnicity, calibrated energy intake, income, general health, physical activity, history of fracture at age ≥55 y, history of parental fracture, current smoking, corticosteroid use, glucocorticoid use, treated diabetes, rheumatoid arthritis, and hormone use and calculated at the subgroup point of interest. WHI, Women's Health Initiative.
TABLE 5.
Three-year change in total body and hip BMD per 20% increase in calibrated protein intake (percentage of kcal) in the WHI BMD Cohort at subgroup levels1
| Outcome | Total body | P-interaction | Hip | P-interaction |
| g/cm2 | g/cm2 | |||
| Overall | 0.003 (0.001, 0.005) | — | 0.002 (0.001, 0.005) | — |
| Age at baseline | 0.334 | 0.154 | ||
| 55 y | 0.002 (0.000, 0.005) | 0.002 (−0.001, 0.004) | ||
| 65 y | 0.003 (0.001, 0.005) | 0.003 (0.001, 0.005) | ||
| 75 y | 0.003 (0.000, 0.007) | 0.004 (0.001, 0.007) | ||
| BMI | 0.496 | 0.118 | ||
| 18.5 kg/m2 | 0.002 (0.000, 0.007) | 0.000 (−0.002, 0.004) | ||
| 25.0 kg/m2 | 0.003 (0.001, 0.005) | 0.002 (0.000, 0.004) | ||
| 30.0 kg/m2 | 0.003 (0.001, 0.005) | 0.003 (0.001, 0.005) | ||
| 35.0 kg/m2 | 0.003 (0.000, 0.006) | 0.004 (0.001, 0.007) |
All values are means; 95% CIs in parentheses. Estimates were derived from linear regression models adjusted for age, BMI, race-ethnicity, calibrated energy intake, income, general health, physical activity, history of fracture at age ≥55 y, history of parental fracture, current smoking, corticosteroid use, glucocorticoid use, treated diabetes, rheumatoid arthritis, and hormone use and calculated at the subgroup point of interest. BMD, bone mineral density; WHI, Women's Health Initiative.
DISCUSSION
Data from this large, long-term study of postmenopausal women suggested that women who consumed more protein did not have a higher risk of fracture or lower BMD than do women who consumed less protein, irrespective of the bone site measured. Rather, a 20% higher protein intake was associated with 7% lower risk of forearm fracture (95% CI: 2%, 12%). Higher protein intake was also significantly associated with higher baseline BMD overall and at the hip and spine sites. Women who consumed greater protein intake were more likely to preserve BMD over time as well.
The inclusion of 36,166 fractures (including 3 286 hip fractures) over more than a decade of study provided us with the unique opportunity to substantially augment data on the relation between protein intake and fracture. Previous studies with fracture as the outcome in women aged >50 y reported inconsistent results, with some studies of higher protein intake reporting an increased risk of fracture (4, 14), whereas others studies showed a decreased risk (15, 16). The meta-analysis including 4 studies reported no significant effect for protein and fracture risk (RR :0.75; 95% CI: 0.47, 1.21) (5). However, because of the magnitude and duration of this study, the preponderance of evidence suggested that, if higher protein has any impact on fracture risk, it results in slightly reduced risk.
Studies of the association between protein intake and BMD also reported inconsistent results, with some studies that showed beneficial associations (17, 18), other studies that reported inconsistent associations (19), and other studies that found adverse associations (20). The systematic review including 61 studies reported a small beneficial association between total protein intake and BMD, estimating that the proportion of BMD attributable to protein was 1–2% (5). A weight-loss feeding study in middle-aged adults showed that a high-protein diet (1.4 g · kg−1 · d−1) with 3 dairy servings/d attenuated bone loss relative to a diet consistent with the current Recommended Dietary Allowance for protein (0.8 g · kg−1 · d−1) during both weight loss (4 mo) and the maintenance of weight loss (8 mo) (21).
The protein source (ie, animal or vegetable) may influence protein's effect on bone health. Studies that have investigated the role of protein source on bone health have been conducted primarily in postmenopausal women and reported disparate findings. In a cohort of adults aged ≥55 y, higher animal protein intake was associated with higher BMD, whereas vegetable protein intake was inversely correlated with BMD (22). Another study showed no overall association between protein intake and fracture risk but did see a trend toward increased fracture risk with increased intake of animal protein (23). A 2008 study in older women showed increased odds of osteoporosis for total protein but a decrease in odds with increased vegetable protein intake (24). An investigation of postmenopausal women in a large cohort study (the European Prospective Investigation into Cancer and Nutrition, Potsdam) showed an inverse association between increased animal protein and bone structure assessed by ultrasound but a positive association with higher vegetable-protein intake (25).
Because we lacked a biomarker for the protein source (animal compared with vegetable), we were unable to correct for the measurement error in self-reported intake by source. Because significant associations were only observed after we corrected for the measurement error in total protein intake, our analyses focused on total, rather than the type, of protein intake.
Limitations should be considered in interpreting our findings. The FFQ had considerable measurement error and, thus, may have substantially attenuated diet-disease associations (26). However, by using a biomarker of total protein intake, we were able to include a correction for the measurement error in self-reported diet. Also notable is that protein intake did not vary across the entire recommended range of 10–35% of energy intake. Thus, although these inferences applied to typical protein intake in the population, data were not available to evaluate lower and upper bounds of recommended ranges of intake. The study population was predominantly non-Hispanic white, and thus, our findings may not be generalizable to other racial-ethnic groups with differences in bone metabolism.
Strengths of the current study included the large sample size of postmenopausal women, which allowed us to examine associations between dietary intake and bone health over more than a decade of follow-up. The excellent follow-up of fracture incidence and longitudinal measures of BMD as measured by dual X-ray absorptiometry provided us with the opportunity to accurately and precisely detect changes in bone health over time. Data were collected on multiple exposures related to bone health in addition to biomarker-calibrated energy and protein intake, such as physical activity and smoking, and these factors were accounted for in the analysis.
In conclusion, data from this large cohort study of postmenopausal women provide evidence that protein intake in the upper range of typical consumption in the United States does not negatively affect bone mass in postmenopausal women. Additional studies in populations consuming protein in the upper end of the recommended range (25–35% of energy from protein) could inform whether higher protein intake contributes to better health outcomes in older women.
Acknowledgments
A short list of WHI investigators can be found in Appendix A. For a list of all of the investigators who have contributed to WHI science, please visit https://cleo.whi.org/researchers/SitePages/Write%20a%20Paper.aspx.
The authors’ responsibilities were as follows—JMB and AZL: designed the research; MLN, LFT, RJ, KCJ, AZL, and RLP: conducted the research; JCL and YH: analyzed data and provided statistical expertise; JMB, LS, and CBE: wrote the manuscript; JMB: had primary responsibility for the final content of the manuscript; and all authors: read and approved the final manuscript. None of the authors declared a conflict of interest.
Footnotes
Abbreviations used: BMD, bone mineral density; CT, clinical trial; DM, dietary modification; FFQ, food-frequency questionnaire; OS, observational study; WHI, Women's Health Initiative.
REFERENCES
- 1.Heaney RP, Layman DK. Amount and type of protein influences bone health. Am J Clin Nutr 2008;87:1567S–70S. [DOI] [PubMed] [Google Scholar]
- 2.Feskanich D, Willett WC, Stampfer MJ, Colditz GA. Protein consumption and bone fractures in women. Am J Epidemiol 1996;143:472–9. [DOI] [PubMed] [Google Scholar]
- 3.Kerstetter JE, O'Brien KO, Insogna KL. Dietary protein, calcium metabolism, and skeletal homeostasis revisited. Am J Clin Nutr 2003;78(suppl):584S–92S. [DOI] [PubMed] [Google Scholar]
- 4.Meyer HE, Pedersen JI, Loken EB, Tverdal A. Dietary factors and the incidence of hip fracture in middle-aged Norwegians. A prospective study. Am J Epidemiol 1997;145:117–23. [DOI] [PubMed] [Google Scholar]
- 5.Darling AL, Millward DJ, Torgerson DJ, Hewitt CE, Lanham-New SA. Dietary protein and bone health: a systematic review and meta-analysis. Am J Clin Nutr 2009;90:1674–92. [DOI] [PubMed] [Google Scholar]
- 6.Pedersen AN, Kondrup J, Borsheim E. Health effects of protein intake in healthy adults: a systematic literature review. Food & nutrition research 2013;57. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Neuhouser ML, Tinker L, Shaw PA, Schoeller D, Bingham SA, Horn LV, Beresford SA, Caan B, Thomson C, Satterfield M, et al. Use of recovery biomarkers to calibrate nutrient consumption self-reports in the Women's Health Initiative. Am J Epidemiol 2008;167:1247–59. [DOI] [PubMed] [Google Scholar]
- 8.Anderson GL. Implementation of the Women's Health Initiative study design. Ann Epidemiol 2003;13:S5–17. [DOI] [PubMed] [Google Scholar]
- 9.Chen Z, Kooperberg C, Pettinger MB, Bassford T, Cauley JA, LaCroix AZ, Lewis CE, Kipersztok S, Borne C, Jackson RD. Validity of self-report for fractures among a multiethnic cohort of postmenopausal women: results from the Women's Health Initiative observational study and clinical trials. Menopause 2004;11:264–74. [DOI] [PubMed] [Google Scholar]
- 10.Patterson RE, Kristal AR, Tinker LF, Carter RA, Bolton MP, Agurs-Collins T. Measurement characteristics of the Women's Health Initiative food frequency questionnaire. Ann Epidemiol 1999;9:178–87. [DOI] [PubMed] [Google Scholar]
- 11.Ainsworth BE, Haskell WL, Leon AS, Jacobs DR, Jr, Montoye HJ, Sallis JF, Paffenbarger RS., Jr Compendium of physical activities: classification of energy costs of human physical activities. Med Sci Sports Exerc 1993;25:71–80. [DOI] [PubMed] [Google Scholar]
- 12.Howard BV, Manson JE, Stefanick ML, Beresford SA, Frank G, Jones B, Rodabough RJ, Snetselaar L, Thomson C, Tinker L, et al. Low-fat dietary pattern and weight change over 7 years: the Women's Health Initiative Dietary Modification Trial. JAMA 2006;295:39–49. [DOI] [PubMed] [Google Scholar]
- 13.McTiernan A, Wactawski-Wende J, Wu L, Rodabough RJ, Watts NB, Tylavsky F, Freeman R, Hendrix S, Jackson R. Women's Health Initiative I. Low-fat, increased fruit, vegetable, and grain dietary pattern, fractures, and bone mineral density: the Women's Health Initiative Dietary Modification Trial. Am J Clin Nutr 2009;89:1864–76. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Abelow BJ, Holford TR, Insogna KL. Cross-cultural association between dietary animal protein and hip fracture: a hypothesis. Calcif Tissue Int 1992;50:14–8. [DOI] [PubMed] [Google Scholar]
- 15.Munger RG, Cerhan JR, Chiu BC. Prospective study of dietary protein intake and risk of hip fracture in postmenopausal women. Am J Clin Nutr 1999;69:147–52. [DOI] [PubMed] [Google Scholar]
- 16.Wengreen HJ, Munger RG, West NA, Cutler DR, Corcoran CD, Zhang J, Sassano NE. Dietary protein intake and risk of osteoporotic hip fracture in elderly residents of Utah. J Bone Miner Res 2004;19:537–45. [DOI] [PubMed] [Google Scholar]
- 17.Hannan MT, Tucker KL, Dawson-Hughes B, Cupples LA, Felson DT, Kiel DP. Effect of dietary protein on bone loss in elderly men and women: the Framingham Osteoporosis Study. J Bone Miner Res 2000;15:2504–12. [DOI] [PubMed] [Google Scholar]
- 18.Ilich JZ, Brownbill RA, Tamborini L. Bone and nutrition in elderly women: protein, energy, and calcium as main determinants of bone mineral density. Eur J Clin Nutr 2003;57:554–65. [DOI] [PubMed] [Google Scholar]
- 19.Wang MC, Luz Villa M, Marcus R, Kelsey JL. Associations of vitamin C, calcium and protein with bone mass in postmenopausal Mexican American women. Osteoporos Int 1997;7:533–8. [DOI] [PubMed] [Google Scholar]
- 20.Metz JA, Anderson JJ, Gallagher PN., Jr Intakes of calcium, phosphorus, and protein, and physical-activity level are related to radial bone mass in young adult women. Am J Clin Nutr 1993;58:537–42. [DOI] [PubMed] [Google Scholar]
- 21.Thorpe MP, Jacobson EH, Layman DK, He X, Kris-Etherton PM, Evans EM. A diet high in protein, dairy, and calcium attenuates bone loss over twelve months of weight loss and maintenance relative to a conventional high-carbohydrate diet in adults. J Nutr 2008;138:1096–100. [DOI] [PubMed] [Google Scholar]
- 22.Promislow JH, Goodman-Gruen D, Slymen DJ, Barrett-Connor E. Protein consumption and bone mineral density in the elderly: the Rancho Bernardo Study. Am J Epidemiol 2002;155:636–44. [DOI] [PubMed] [Google Scholar]
- 23.Dargent-Molina P, Sabia S, Touvier M, Kesse E, Breart G, Clavel-Chapelon F, Boutron-Ruault MC. Proteins, dietary acid load, and calcium and risk of postmenopausal fractures in the E3N French women prospective study. J Bone Miner Res 2008;23:1915–22. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Kim J, Lim SY, Kim JH. Nutrient intake risk factors of osteoporosis in postmenopausal women. Asia Pac J Clin Nutr 2008;17:270–5. [PubMed] [Google Scholar]
- 25.Weikert C, Walter D, Hoffmann K, Kroke A, Bergmann MM, Boeing H. The relation between dietary protein, calcium and bone health in women: results from the EPIC-Potsdam cohort. Ann Nutr Metab 2005;49:312–8. [DOI] [PubMed] [Google Scholar]
- 26.Subar AF, Kipnis V, Troiano RP, Midthune D, Schoeller DA, Bingham S, Sharbaugh CO, Trabulsi J, Runswick S, Ballard-Barbash R, et al. Using intake biomarkers to evaluate the extent of dietary misreporting in a large sample of adults: the OPEN study. Am J Epidemiol 2003;158:1–13. [DOI] [PubMed] [Google Scholar]