Intake o f Even -chain Saturated Fatty Acids and Risk of · 16 6 Professor, Departments of...

Confidential: For Review O

nly

Intake of Even-chain Saturated Fatty Acids and Risk of

Coronary Heart Disease in two large prospective cohort studies of U.S. Men and Women

Journal: BMJ

Manuscript ID BMJ.2016.033680

Article Type: Research

BMJ Journal: BMJ

Date Submitted by the Author: 27-May-2016

Complete List of Authors: Zong, Geng; Harvard School of Public Health, Department of nutrition Li, Yanping; HSPH Wanders, Anne; Unilever Research and Development Vlaardingen Alssema, Marjan; Unilever Research and Development Vlaardingen Zock, Peter; Unilever Research and Development Vlaardingen Hu, Frank; HSPH Willett, Walter; Harvard School of Public Health Sun, Qi; Harvard School of Public Health, Nutrition

Keywords: saturated fatty acids, palmitic acid, stearic acid, coronary heart disease

https://mc.manuscriptcentral.com/bmj

BMJ


nly

1

Title: Intake of Even-chain Saturated Fatty Acids and Risk of Coronary Heart Disease in two 1

large prospective cohort studies of U.S. Men and Women 2

First author: Zong 3

Short title: Saturated fatty acids and coronary heart diseases 4

Authors: Geng Zong,1 Yanping Li,

2 Anne J. Wanders,

3 Marjan Alssema,

4 Peter L. Zock,

5 Walter 5

C.Willett,6 Frank B. Hu,

7 and Qi Sun.

8 6

Affiliations: 7

1 Postdoctoral research fellow, Department of Nutrition, Harvard T.H. Chan School of Public 8

Health, 665 Huntington Ave., Boston, MA 02115, USA; 2

Research Scientist, Department of 9

Nutrition, Harvard T.H. Chan School of Public Health, 665 Huntington Ave., Boston, MA 10

02115, USA; 3

Research scientist, Unilever Research and Development, 3133AT, Vlaardingen, 11

The Netherlands; 4

Research scientist, Unilever Research and Development, 3133AT, 12

Vlaardingen, The Netherlands; 3

Department of Epidemiology, Harvard T.H. Chan School of 13

Public Health, Boston, Massachusetts;

14

5 Lead Scientists, Unilever Research and Development, 3133AT, Vlaardingen, The Netherlands;

15

6 Professor, Departments of Nutrition and Epidemiology, Harvard T.H. Chan School of Public 16

Health, Channing Division of Network Medicine, Department of Medicine, Brigham and 17

Women’s Hospital and Harvard Medical School, 665 Huntington Ave., Boston, MA 02115, 18

USA; 7

Professor, Departments of Nutrition and Epidemiology, Harvard T.H. Chan School of 19

Public Health, Channing Division of Network Medicine, Department of Medicine, Brigham and 20

Page 1 of 33


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nly

2


USA; 8 Assistant Professor, Departments of Nutrition, Harvard T.H. Chan School of Public 22

Health, Channing Division of Network Medicine, Department of Medicine, Brigham and 23


USA. 25

Correspondence to: Q Sun, [email protected] 26

Contributors: QS and FBH obtained funding from the National Institutes of Health. GZ and QS 27

designed this study. QS and FBH were involved in data collection. GZ and QS provided 28

statistical expertise. GZ analysed the data and wrote the first draft of the manuscript. All authors 29

contributed to the interpretation of the results and critical revision of the manuscript for 30

important intellectual content and approved the final version of the manuscript. GZ and QS are 31

the guarantors of this investigation. 32

Funding: This study was funded by research grants UM1 CA186107, R01 HL034594, R01 33

HL35464, R01 HL60712, UM1 CA167552 from the National Institutes of Health. Qi Sun was 34

supported by a career development award, R00-HL098459, from the National Heart, Lung, and 35

Blood Institute. 36

Competing interests: All authors have completed the Unified Competing Interest form 37

at www.icmje.org/coi_disclosure.pdf (available on request from the corresponding author). Geng 38

Zong is supported by a restricted grant from Unilever. Anne Wanders, Marjan Alssema, and 39

Peter L. Zock are employees of Unilever Research of Development. Unilever is a producer of 40

Page 2 of 33


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nly

3

food consumer products. No other relationships or activities that could appear to have influenced 41

the submitted work 42

Ethical approval: The study protocols were approved by the institutional review boards of the 43

Brigham and Women's Hospital and the Harvard School of Public Health. The completion of the 44

self-administered questionnaire was considered to imply informed consent. 45

The sponsors have no role in: the study design; the collection, analysis, or interpretation of data; 46

the writing of the report; or in the decision to submit the article for publication. 47

Data sharing: No additional data available. 48

Page 3 of 33


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nly

4

Abstract 49

Objectives: This study aimed to investigate associations between individual SFA intakes and 50

coronary heart disease (CHD) risk, given the differential effects of saturated fatty acids(SFAs) on 51

blood lipids observed in intervention studies. 52

Design: Prospective longitudinal cohort study. 53

Setting: Health professionals in the United States. 54

Participants: 73,147 women in the Nurses’ Health Study and 42,635 men in the Health 55

Professionals Follow-up Study (1986-2012), who were free of major chronic diseases at 56

baseline. 57

Main outcome measure: Incident CHD identified by Medical records or National Death Index. 58

Results: Comparing the highest to the lowest SFA quintiles, hazard ratios(95% confidence 59

interval, HRs[95%CIs]) of CHD were 1.18(1.09, 1.28; Ptrend<0.001) for summed 12:0-18:0, 60

1.07(0.99, 1.15; Ptrend=0.05) for 12:0, 1.13(1.05, 1.22; Ptrend<0.001) for 14:0, 1.17(1.08, 1.26; 61

Ptrend<0.001) for 16:0, and 1.18(1.09, 1.28; Ptrend<0.001) for 18:0, after adjustment for lifestyle 62

factors and body mass index. HRs(95%CIs) of CHD for iso-caloric replacement of 1% energy of 63

12:0-18:0 were 0.92(0.89, 0.95; P<0.001) by polyunsaturated fatty acids(PUFA), 0.95(0.90, 64

1.00; P=0.07) by monounsaturated fatty acids(MUFA), 0.96(0.93, 0.99; P=0.01) by carbohydrate 65

from whole grains(CHO-WG), and 0.95(0.91, 0.99; P=0.02) by protein from plants(PRO-PL). 66

Replacement of 16:0, the most abundant SFA, by PUFA, MUFA, CHO-WG, or PRO-PL was 67

more strongly associated with CHD risk, with HRs(95%CIs) being 0.88(0.81, 0.95; P=0.002), 68

Page 4 of 33


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nly

5

0.90(0.83, 1.01; P=0.08), 0.92(0.85, 0.99; P=0.03), and 0.91(0.84, 0.99; P=0.03), respectively. 69

For other SFA, replacing 18:0 with PUFA was associated with a lower CHD risk (HRs[95CIs]): 70

0.92[0.84, 1.01], P=0.07). 71

Conclusions: Intakes of major SFA, especially 16:0, were associated with increased CHD risk. 72

Due to high correlations among individual SFA, dietary recommendations for CHD prevention 73

should keep focusing on replacing total SFA by healthy macronutrients. 74

Page 5 of 33


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nly

6

What this paper adds 75

What is already known on this subject 76

1. Saturated fat intake is a risk factor for coronary heart disease, and replacing saturated fat with 77

unsaturated fat or whole grains is associated with lower disease risk; 78

2. Evidences from intervention studies suggests major saturated fatty acids, including lauric 79

acid (12:0), mystic acid (14:0), palmitic acid (16:0), and stearic acid (18:0), could have 80

different effects on blood lipids; 81

3. The association between individual saturated fatty acids intake and risk of coronary heart 82

disease in free-living population has been investigated in only one prospective cohort study 83

of women. 84

What this study adds 85

4. Intakes of lauric acid (12:0), mystic acid (14:0), palmitic acid (16:0), and stearic acid (18:0) 86

are all associated with increased risk of coronary heart disease, and their intakes highly 87

correlated. 88

5. Palmitic acid is the most abundant saturated fatty acid in the diet, and replacing it with 89

polyunsaturated fatty acids, monounsaturated fatty acids carbohydrates from whole grains, 90

and protein from plants was more strongly associated with coronary heart disease risk. 91

6. Dietary recommendation should remain focus on reducing total fat for the prevention of 92

coronary heart disease, given the shared food sources and high correlations in the diet of 93

individual saturated fatty acids. 94

Page 6 of 33


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nly

7

Introduction 95

Prevailing dietary guidelines recommend keeping total saturated fat intake below 10% of total 96

energy for the prevention of cardiovascular disease, a leading cause of death worldwide.1, 2

In 97

large-scale prospective studies that explicitly examined effects of replacing saturated fatty acids 98

(SFA) with alternative macronutrients such as polyunsaturated fatty acids (PUFA) or 99

carbohydrates from whole grains (CHO-WG), higher intake of saturated fat has been associated 100

with an elevated risk of coronary heart disease (CHD).3, 4

There is evidence suggesting that 101

major dietary even-chain SFA, including lauric acid (12:0), mystic acid (14:0), palmitic acid 102

(16:0), and stearic acid (18:0), confer heterogeneous effects on blood lipids.5 Meta-analysis of 103

clinical trials found that stearic acid (18:0) raised total cholesterol, LDL cholesterol, and 104

total/HDL cholesterol ratio less than lauric acid (12:0), myristic acid (14:0), and palmitic acid 105

(16:0) did, using carbohydrate as a common reference.5 106

Prospective cohort studies on the association between different SFA and risk of CHD events are 107

scarce. We have previously shown that long-chain SFA (12:0-18:0) were positively associated 108

with 14-year risk of CHD in women from the Nurses’ Health Study (NHS), whereas 4:0-10:0 109

were not.6 Whether the same observation holds true in men has not been examined. In the 110

current study, we updated the analyses on associations between individual SFA and risk of CHD 111

in NHS with an additional 18 years of follow-up, and combined these with data from a cohort of 112

men in the Health Professionals Follow-up Study (HPFS). More importantly, we estimated 113

CHD risk when individual SFA are replaced by beneficial macronutrients that may facilitate 114

Page 7 of 33


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nly

8

CHD prevention, including PUFA, monounsaturated fatty acids (MUFAs), CHO-WG, and 115

protein from plants (PRO-PL) 3, 4, 7

. 116

Method 117

Study Population 118

The NHS included 121,700 female nurses aged 30-55 years in 1976,8 and the HPFS included 119

51,529 male health professionals aged 40-75 years in 1986.9 Information on medical history, 120

lifestyle, potential risk factors, and disease diagnosis was collected at baseline through a self-121

administered questionnaire and updated every 2 years using similar questionnaires in both 122

cohorts. Study protocol was approved by the institutional review boards of Brigham and 123

Women’s Hospital and Harvard T.H. Chan School of Public Health. Return of self-administered 124

questionnaires was considered informed consent. 125

The current analysis was conducted among 81,757 women and 51,530 men who completed 126

a food frequency questionnaire (FFQ) at study baseline (1984 for NHS and 1986 for HPFS). 127

Participants were excluded if: they reported physician-diagnosed cancer, diabetes, or 128

cardiovascular disease at study baseline(n=7,749 for NHS, and n=7,596 for HPFS); their total 129

energy intake was deemed implausible (<600 or >3,500 kcal/day in NHS, and <800 or >4,200 130

kcal/day in HPFS) or had missing individual SFA data (n=98 in NHS, and n=147 in HPFS); or 131

they answered baseline questionnaire only or had missing age at baseline (n=763 in NHS, and 132

n=1,152 in HPFS). The final sample consisted of 73,147 women and 42,635 men with complete 133

information. 134

Page 8 of 33


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nly

9

Ascertainment of diet 135

In 1980, NHS participants completed a 61-item FFQ on their usual intake of foods and beverages 136

in the past year. In 1984, 1986, and every 4 years thereafter through 2010, NHS participants were 137

mailed an expanded FFQ to assess and update dietary information. The same FFQ has been sent 138

to HPFS participants every 4 years from 1986 to 2010. This study used 1984 of NHS and 1986 139

of HPFS as baseline for more comprehensive estimates of individual SFA intake and the 140

comparability of two cohorts. 141

Participants were asked how often, on average, they had consumed specific foods during the past 142

year, with nine responses ranging from “never” to “≥6 times per day” based on a specified 143

standard portion size. They were also asked for the types of fat, oil, and margarine used during 144

cooking and at the table. Intakes of even-chain SFA with carbon chain length between 4 and 18 145

were calculated by multiplying frequency of consuming each food item by its fatty acid 146

composition, and then summing up values from all foods. SFA compositions of food were based 147

on the U.S. Department of Agriculture and Harvard University food-composition database, 148

which is updated over time to reflect nutrient profile of new food items. CHO-WG was estimated 149

based on a list of grain-based foods, including whole grain bread, cereal, oatmeal, and added 150

bran.3, 10

To better represent long-term or habitual intake and to minimize within-person 151

variation, we calculated cumulative averages of SFA intake from all available dietary 152

questionnaires throughout the follow-up period.11

We stopped updating diet information after 153

participants reported a diagnosis of diabetes, stroke, CHD, or cancer, to minimize the possibility 154

of reverse causation bias. Cumulative means of dietary variables before diagnosis of these 155

Page 9 of 33


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nly

10

diseases were then carried forward to represent diet for later follow-up. Missing values of 156

individual SFA were replaced with cumulative means of prior assessments. Fat intake from FFQs 157

was validated against multiple-week diet records.12-14

For example, the correlation coefficient 158

was 0.58 between FFQ and dietary record assessments for total SFA, 0.48 for PUFA, and 0.58 159

for MUFA.15

Fatty acid biomarker were also used to validate FFQ data. For example, the 160

correlation between PUFA in FFQ and PUFA in subcutaneous fat was 0.43-0.50.16

161

Ascertainment of Coronary Heart Diseases 162

In this study, total CHD included nonfatal myocardial infarction (MI) and fatal CHD.17

To 163

ascertain nonfatal MI, we first obtained permission of access to medical records from 164

participants who reported having a physician-diagnosed heart disease in follow-up 165

questionnaires. Medical records were reviewed by study physicians who were blinded to 166

exposure status, and telephone interviews were performed when medical records were not 167

provided. Nonfatal MI was confirmed using the World Health Organization criteria of typical 168

symptoms plus either elevated enzymes or diagnostic ECG changes.18

Deaths were identified by 169

reports from next of kin and U.S. postal authorities, or by searching the National Death Index. 170

More than 98% of deaths can be identified using these approaches.17

Fatal CHD was confirmed 171

by a review of hospital records or autopsy reports if CHD was listed as the underlying cause of 172

death and if evidence of previous CHD was available from medical records. Probable fatal CHD 173

was assigned to cases with CHD listed as the underlying cause of death on the death certificate 174

but no medical records concerning the death were available and no prior reports of CHD was 175

Page 10 of 33


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nly

11

indicated. For this analysis, confirmed (n=5250) and probable (n=1,785) fatal CHD cases were 176

included to maximize statistical power.19

177

Statistical analysis 178

To describe the trend of individual SFA intake over time, we calculated aged-adjusted individual 179

SFA intakes during follow-up without generating cumulative average, carry-over of missing 180

values, or stopping updating diet after diagnose of chronic diseases. 181

Person-years of follow-up were calculated from the return date of baseline questionnaire to 182

the date when participants were diagnosed with CHD, the date of death, or the end of follow-up, 183

whichever came first. Hazard ratios (HRs) and 95% confidence intervals (95%CIs) of incident 184

CHD according to quintiles of individual SFA intake were estimated by using time-dependent 185

Cox proportional hazards regression model in each cohort with follow-up duration as the 186

timescale. Analyses were conducted in the two cohorts separately, and then results were pooled 187

using a fixed-effect model, when heterogeneity was >0.05. Cox regression analysis was stratified 188

jointly by age in months and calendar year to better control for confounding by age, calendar 189

time, and any possible two-way interactions between them as much as possible. Multivariate 190

model was adjusted for ethnicity, family history of MI, body mass index (BMI), cigarette 191

smoking, alcohol intake, physical activity, multivitamin use, menopausal status and 192

postmenopausal hormone use (for women), current aspirin use, baseline hypertension, baseline 193

hypercholesterolemia, and total energy intake. A test for linear trend was performed by modeling 194

median values of individual SFA as continuous variables. We estimated HRs (95%CIs) of CHD 195

for iso-calorically replacing individual SFA with PUFA, MUFA, CHO-WG, or PRO-PL in an 196

Page 11 of 33


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nly

12

energy-density model. The model controlled for abovementioned covariates, intakes of cereal 197

fiber, fruits and vegetables, and all energy contributing nutrients including trans fat, PUFA, 198

MUFA, carbohydrate from non-wholegrain sources, CHO-WG, protein from non-plant sources, 199

PRO-PL, and the sum of SFA except the one at issue. By leaving one specific SFA out from the 200

model, regression coefficients of PUFA, MUFA, CHO-WG, and PRO-PL can be interpreted as 201

estimated effects of iso-calorically substituting these nutrients for that SFA while overall intake 202

of other SFA was hold as constant. We tested proportional hazard assumption by including 203

interaction terms between individual SFA and duration of follow-up in the model, and results did 204

not suggest that the assumption was violated (P>0.05). 205

To test whether associations of individual SFA may be modified by age (<65 years, 206

≥65years), BMI (<25kg/m2, ≥25kg/m

2), physical activity (<18, ≥18 metabolic equivalent of 207

task/week), and smoking status (current smoking or not), we performed substitution analyses 208

stratified by these variables, using a likelihood-ratio test to evaluate the significance of 209

interaction terms. We performed two sensitivity analyses to examine the robustness of our 210

findings: 1) controlling for baseline BMI instead of updated BMI, as obesity could be potential 211

mediator between SFA and CHD risk, or 2) adjusting for hypertension, hypercholesterolemia, 212

and diabetes diagnosed during follow-up. Statistical analyses were performed by using SAS 9.4 213

(SAS Institute, Cary, NC). All P values were 2-sided, with statistical significance defined 214

as P<0.05. 215

Patient involvement 216

Page 12 of 33


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nly

13

No patients were involved in setting the research question or the outcome measures, nor were 217

they involved in developing plans for recruitment, design, or implementation of the study. No 218

patients were asked to advise on interpretation or writing up of results. There are no plans to 219

disseminate the results of the research to study participants or the relevant patient community. 220

Results 221

At baseline, participants with higher SFA intake were younger, were more likely to be 222

Caucasians and non-smokers, had lower alcohol consumption, engaged in less physical activity, 223

and had lower prevalence of hypercholesterolemia and hypertension. They were less likely to 224

take multivitamin supplements, and had higher BMI (Table 1). They also had higher intakes of 225

total energy, PUFA, MUFA, trans fat, and protein, and lower intakes of carbohydrate, PRO-PL, 226

cereal fiber, and fruits and vegetables. 227

Proportions of energy from SFA decreased gradually from 11.3% in NHS and 10.4% in 228

HPFS at baseline to 9.1% and 9.0% in 1998, and slightly increased again to 9.9% and 9.8% in 229

2010, respectively (Figure 1). Throughout the follow-up period, mean energy contribution of 230

16:0 ranged from 5.2% (in 1998) to 6.4% (in 1984), followed by 18:0 (2.4-3.0% energy) and 231

14:0 (0.7-1.0% energy), and 12:0 (0.2%-0.3%), with 4:0-10:0 contributing to only 0.4-0.6% of 232

total energy. Individual SFAs in the diet were highly correlated, and Spearman correlation 233

coefficients ranged from 0.38 between 4:0-10:0 and 18:0 in NHS to 0.93 between 16:0 and 18:0 234

in NHS (all P<0.001; Supplementary table 1). 235

Page 13 of 33


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nly

14

During 2.76 million person-years of follow-up, 7035 incident CHD cases were identified 236

(including 2687 fatal CHD and 4348 nonfatal MI). As shown in Table 2, the sum of 4:0-10:0 237

was positively associated with CHD risk in age-adjusted model (P=0.009), which was attenuated 238

after multivariate adjustment (P=0.30). Intakes of 12:0, 14:0, 16:0, 18:0, and the sum of 12:0-239

18:0 were all associated with a higher risk of CHD in age- and multivariate-adjusted model (all P 240

for trend<0.05). Pooled HRs (95%CIs) of CHD comparing the highest to the lowest quintiles 241

were 1.07 (0.99, 1.15; P for trend=0.05) for 12:0, 1.13 (1.05, 1.22; P for trend<0.001) for 14:0, 242

1.17 (1.08, 1.26; P for trend<0.001) for 16:0, 1.18 (1.09, 1.28; P for trend<0.001) for 18:0, and 243

1.24 (1.08, 1.43; P for trend=0.002) for 12:0-18:0. 244

We further estimated CHD risk when PUFA, MUFA, CHO-WG, and PRO-PL are used to 245

replace 12:0, 14:0, 16:0, and 18:0. Because of low abundance and high correlations of 12:0 and 246

14:0, we combined them in the analysis. As shown in Table 3, iso-calorically replacing 1% 247

energy from 16:0 with PUFA, MUFA, CHO-WG, or PRO-PL was most strongly associated with 248

lower CHD risk, and pooled HRs (95%CIs) of CHD were 0.88 (0.81, 0.95; P=0.002), 0.92 (0.83, 249

1.01; P=0.08), 0.92 (0.85, 0.99; P=0.03), and 0.91 (0.84, 0.99; P=0.03) respectively. In addition, 250

replacing 18:0 with PUFA was marginally associated with lower risk of CHD (HR[9%CI]: 251

0.92[0.84, 1.01]; P=0.07). After major SFA were pooled together, estimated CHD risk was lower 252

and more robust when 12:0-18:0 was replaced by PUFA, MUFA , CHO-WG, and PRO-PL. We 253

did not find heterogeneity when pooling results of the two cohorts. 254

Supplementary table 2 presented substitution analysis stratified by age, BMI, physical activity, 255

and smoking status, and all P values for interaction were >0.05. Associations between SFA and 256

Page 14 of 33


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nly

15

risk of CHD were similar when baseline BMI was adjusted as a covariate (Supplementary 257

table 3), but were slightly attenuated when incident hypertension, hypercholesterolemia, and 258

diabetes diagnosed during follow-up were further adjusted (Supplementary table 4). 259

260

Page 15 of 33


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nly

16

Discussion 261

In the current study of two large ongoing prospective cohorts, dietary intakes of major SFAs 262

(12:0-18:0) were positively associated with CHD risk, whereas 4:0-10:0 was not. We found that 263

replacing total 12:0-18:0 intake with PUFA, MUFA, CHO-WG, or PRO-PL was associated with 264

lower CHD risk, and the association was more pronounced for 16:0, the most abundant SFA, 265

than that for other even-chain SFA. 266

Limited observational studies have been conducted to examine associations of individual 267

SFAs with CHD risk. In the Seven Countries Study, middle-aged men from populations with 268

higher average intakes of 12:0, 14:0, 16:0, or 18:0 had higher serum cholesterol and increased 269

CHD mortality, but it is not clear whether these associations were independent of other lifestyle 270

and dietary factors.20

A cross-sectional study among 993 Costa Ricans reported intakes of 12:0, 271

14:0, 16:0, and 18:0 were all positively associated with odds of having nonfatal acute MI, after 272

adjusting for intakes of unsaturated fat, trans fat, and proteins.21

Our previous study in NHS was 273

the first to prospectively analyze associations of individual SFAs and CHD risk. In this early 274

analysis, 12:0-14:0, 16:0, and 18:0 were all associated with increased CHD risk, which 275

collectively showed a modest positive association with CHD after adjusting for MUFA, trans 276

fat, protein, fiber, cholesterol, and total energy.6 By extending the follow-up duration with 19 277

more years and additionally including a cohort of male participants, the current study confirmed 278

existing findings from observational studies.6, 20, 21

279

Page 16 of 33


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nly

17

Fatty acids with carbon chain length between 12 and 18 are the major dietary SFA in diet 280

of the U.S. population.22

Our findingof positive associations between these fatty acids and CHD 281

risk is consistent with those focusing on replacement of total saturated fat. For example, a meta-282

analysis of 8 intervention studies found that substituting PUFA for SFA was associated with 283

19% lower risk of CHD with a median follow-up of 4.5 years.23

By pooling data from 11 284

prospective cohort studies, Jakobsen et al reported that replacing total saturated fat with PUFA 285

was inversely associated with CHD morbidity and mortality.4 Our recent analysis in NHS and 286

HPFS also found a lower risk of CHD when saturated fat was replaced by PUFA, MUFA, or 287

CHO-WG.3 By linking both total saturated fat and individual SFA with CHD risk, these 288

findings consistently support the benefit of lowering consumption of SFA for CHD prevention, 289

as recommended by existing dietary guidelines. 290

Estimated reduction in CHD risk was larger when PUFA, MUFA, CHO-WG, and PRO-PL 291

were used to replace 16:0 than when they replaced other SFA. This might be explained by the 292

fact that 16:0 is more abundant in the U.S. diet than other SFA. Our results are also in line with 293

findings from intervention studies showing a stronger effect of 16:0 on blood lipids than to 294

18:0.5 However, the present data should be interpreted with caution given the strong correlation 295

between 16:0 and 18:0. In U.S. population, dairy products and meats are the primary food 296

sources of SFA.24

Dairy foods contain over 40% fat as 16:0, followed by 18:0 (15-21% fat) and 297

14:0 (14-15% fat).25

Meat and meat products(such as beef, porc, chicken, lamb) contains more 298

than half of fat as 16:0 and over 30% as 18:0 on average.25

Because the ood sources of 16:0 and 299

18:0 overlap, their intakes in the current study are highly correlated, which limits the ability to 300

differentiate independent associations of individual SFAs with CHD. Given the similarities in 301

Page 17 of 33


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nly

18

SFA composition of meat and dairy products, it is not likely that saturated fat from these sources 302

might have distinctive effects on health. Instead, observed different associations of saturated fat 303

from meat and dairy products with CVD risk may be attributed to other components, such as 304

calcium and vitamin D in dairy products.26

305

Unlike observational studies, controlled intervention studies have been able to specifically 306

address differences in effects of individual SFA, In intervention studies, substantial differences 307

in intakes of individual fatty acid can be achieved by choosing different dietary fat sources for 308

the experimental diets.5 For example, coconut oil can be used to design a high 12:0 diet as it 309

contains nearly 50% saturated fat as 12:0.27

Fractionated palm oil is rich in 16:0 (88% of total 310

fat),27

and thus used in 16:0 intervention studies.28

For high 18:0 experimental diets, modified 311

plant oils (e.g., by fully hydrogenation), cocoa butter or shea butter are often used.29-32

Based on 312

a pooled analysis of intervention studies applying various sources of SFA , Mensink et al found 313

that 12:0, 14:0, and 16:0, but not 18:0 increased LDL cholesterol compared with carbohydrates.5 314

However, we did not find that replacing 12:0-14:0 with alternative macronutrients was 315

associated with CHD risk, which may be explained by a much lower abundance of these fatty 316

acids in the diet than 16:0 and 18:0, or by potential over adjustment due to inclusion of other 317

SFA, especially 16:0, in the model. 318

The strengths of our study include its large sample size and long follow-up duration, the 319

repeated measurements of diet exposure, lifestyle factors, and health status, and the use of 320

cumulative average to reflect long-term diet.11

However, several limitations should be 321

discussed. First, because of the observational design of this study, we cannot fully exclude 322

Page 18 of 33


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nly

19

potential residual confounding, although a series of demographic, lifestyle, medical history, and 323

dietary factors were adjusted for. Second, our study populations exclusively consisted of health 324

professionals with relatively homogeneous socioeconomic status. This renders the study less 325

susceptible for confounding by socioeconomic factors, but may limit the generalizability of 326

findings to other populations. Third, measurement errors for individual SFAs, especially the 327

ones with relatively low intake levels, such as 4:0-10:0, are inevitable. Because such 328

measurement errors were not related to outcome ascertainment, they are more likely to be non-329

differential and tend to attenuate true associations toward the null. Lastly, the collinearity among 330

individual SFAs in the diet make it difficult to tease out associations for individual SFA. 331

In summary, our data suggest that intakes of major dietary SFA, including 12:0, 14:0, 16:0, 332

and 18:0, are associated with higher CHD risk. Although lower CHD risk was most pronounced 333

for 16:0 being iso-calorically replaced, an advice to reduce total saturated fat seems more 334

practical for CHD prevention, given the high correlations and overlapping dietary sources of 335

SFA in the population. More data from prospective studies in populations with different diet 336

patterns as well as varying food sources of SFA are warranted to replicate our findings. 337

338

Page 19 of 33


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nly

20

Reference List 339

340

1. Mozaffarian D, Benjamin EJ, Go AS, et al. Heart disease and stroke statistics--2015 update: a report from 341

the American Heart Association. Circulation 2015; 131: e29-322. 342

2. Astrup A, Dyerberg J, Elwood P, et al. The role of reducing intakes of saturated fat in the prevention of 343

cardiovascular disease: where does the evidence stand in 2010? The American journal of clinical nutrition 2011; 93: 344

684-8. 345

3. Li Y, Hruby A, Bernstein AM, et al. Saturated Fats Compared With Unsaturated Fats and Sources of 346

Carbohydrates in Relation to Risk of Coronary Heart DiseaseA Prospective Cohort Study. Journal of the American 347

College of Cardiology 2015; 66: 1538-48. 348

4. Jakobsen MU, O'Reilly EJ, Heitmann BL, et al. Major types of dietary fat and risk of coronary heart 349

disease: a pooled analysis of 11 cohort studies. The American journal of clinical nutrition 2009; 89: 1425-32. 350

5. Mensink RP, Zock PL, Kester AD, Katan MB. Effects of dietary fatty acids and carbohydrates on the ratio 351

of serum total to HDL cholesterol and on serum lipids and apolipoproteins: a meta-analysis of 60 controlled trials. 352

The American journal of clinical nutrition 2003; 77: 1146-55. 353

6. Hu FB, Stampfer MJ, Manson JE, et al. Dietary saturated fats and their food sources in relation to the risk 354

of coronary heart disease in women. The American journal of clinical nutrition 1999; 70: 1001-8. 355

7. Bernstein AM, Sun Q, Hu FB, Stampfer MJ, Manson JE, Willett WC. Major dietary protein sources and 356

risk of coronary heart disease in women. Circulation 2010; 122: 876-83. 357

8. Hu FB, Manson JE, Stampfer MJ, et al. Diet, lifestyle, and the risk of type 2 diabetes mellitus in women. 358

The New England journal of medicine 2001; 345: 790-7. 359

9. van Dam RM, Willett WC, Rimm EB, Stampfer MJ, Hu FB. Dietary fat and meat intake in relation to risk 360

of type 2 diabetes in men. Diabetes care 2002; 25: 417-24. 361

10. Jensen MK, Koh-Banerjee P, Hu FB, et al. Intakes of whole grains, bran, and germ and the risk of coronary 362

heart disease in men. The American journal of clinical nutrition 2004; 80: 1492-9. 363

11. Hu FB, Stampfer MJ, Rimm E, et al. Dietary fat and coronary heart disease: a comparison of approaches 364

for adjusting for total energy intake and modeling repeated dietary measurements. American journal of epidemiology 365

1999; 149: 531-40. 366

12. Salvini S, Hunter DJ, Sampson L, et al. Food-based validation of a dietary questionnaire: the effects of 367

week-to-week variation in food consumption. International journal of epidemiology 1989; 18: 858-67. 368

13. Feskanich D, Rimm EB, Giovannucci EL, et al. Reproducibility and validity of food intake measurements 369

from a semiquantitative food frequency questionnaire. Journal of the American Dietetic Association 1993; 93: 790-370

6. 371

14. Garland M, Sacks FM, Colditz GA, et al. The relation between dietary intake and adipose tissue 372

composition of selected fatty acids in US women. The American journal of clinical nutrition 1998; 67: 25-30. 373

15. Willett W. Nutritional epidemiology. 3rd ed. Oxford: Oxford University Press; 2013. 374

16. Hunter DJ, Rimm EB, Sacks FM, et al. Comparison of measures of fatty acid intake by subcutaneous fat 375

aspirate, food frequency questionnaire, and diet records in a free-living population of US men. American journal of 376

epidemiology 1992; 135: 418-27. 377

17. Chiuve SE, Sampson L, Willett WC. The association between a nutritional quality index and risk of chronic 378

disease. American journal of preventive medicine 2011; 40: 505-13. 379

18. Rose GA. Cardiovascular survey methods. 2nd ed. Geneva Albany, N.Y.: World Health Organization;WHO 380

Publications Centre distributor; 1982. 381

19. Malik VS, Chiuve SE, Campos H, et al. Circulating Very-Long-Chain Saturated Fatty Acids and Incident 382

Coronary Heart Disease in US Men and Women. Circulation 2015; 132: 260-8. 383

20. Kromhout D, Menotti A, Bloemberg B, et al. Dietary saturated and trans fatty acids and cholesterol and 25-384

year mortality from coronary heart disease: the Seven Countries Study. Preventive medicine 1995; 24: 308-15. 385

21. Kabagambe EK, Baylin A, Siles X, Campos H. Individual saturated fatty acids and nonfatal acute 386

myocardial infarction in Costa Rica. European journal of clinical nutrition 2003; 57: 1447-57. 387

22. U.S. Department of Agriculture, Agricultural Research Service. 2014. Nutrient Intakes from Food and 388

Beverages: Mean Amounts Consumed per Individual, by Gender and Age, What We Eat in America, NHANES 389

2011-2012. 390

23. Mozaffarian D, Micha R, Wallace S. Effects on coronary heart disease of increasing polyunsaturated fat in 391

Page 20 of 33


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nly

21

place of saturated fat: a systematic review and meta-analysis of randomized controlled trials. PLoS Med 2010; 7: 392

e1000252. 393

24. O'Neil CE, Keast DR, Fulgoni VL, Nicklas TA. Food sources of energy and nutrients among adults in the 394

US: NHANES 2003-2006. Nutrients 2012; 4: 2097-120. 395

25. U.S. Department of Agriculture, Nutrient Data Laboratory. Fat and Fatty Acid Content of Selected Foods 396

Containing Trans-Fatty Acids. http://www.ars.usda.gov/SP2UserFiles/Place/80400525/Data/Classics/trans_fa.pdf. 397

Accessed February, 2016 398

26. de Oliveira Otto MC, Mozaffarian D, Kromhout D, et al. Dietary intake of saturated fat by food source and 399

incident cardiovascular disease: the Multi-Ethnic Study of Atherosclerosis. The American journal of clinical 400

nutrition 2012; 96: 397-404. 401

27. Vannice G, Rasmussen H. Position of the academy of nutrition and dietetics: dietary fatty acids for healthy 402

adults. J Acad Nutr Diet 2014; 114: 136-53. 403

28. Fattore E, Bosetti C, Brighenti F, Agostoni C, Fattore G. Palm oil and blood lipid-related markers of 404

cardiovascular disease: a systematic review and meta-analysis of dietary intervention trials. The American journal of 405

clinical nutrition 2014; 99: 1331-50. 406

29. Thijssen MA, Mensink RP. Small differences in the effects of stearic acid, oleic acid, and linoleic acid on 407

the serum lipoprotein profile of humans. The American journal of clinical nutrition 2005; 82: 510-6. 408

30. Aro A, Jauhiainen M, Partanen R, Salminen I, Mutanen M. Stearic acid, trans fatty acids, and dairy fat: 409

effects on serum and lipoprotein lipids, apolipoproteins, lipoprotein(a), and lipid transfer proteins in healthy 410

subjects. The American journal of clinical nutrition 1997; 65: 1419-26. 411

31. Zock PL, Katan MB. Hydrogenation alternatives: effects of trans fatty acids and stearic acid versus linoleic 412

acid on serum lipids and lipoproteins in humans. J Lipid Res 1992; 33: 399-410. 413

32. Dougherty RM, Allman MA, Iacono JM. Effects of diets containing high or low amounts of stearic acid on 414

plasma lipoprotein fractions and fecal fatty acid excretion of men. The American journal of clinical nutrition 1995; 415

61: 1120-8. 416

Page 21 of 33


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review Only22

Table 1. Age-standardized baseline characteristics in the Nurse’s Health Study (1984) and the Health Professionals’ Follow-up study (1986)

according to quintiles of total saturated fat intake

Quintiles of total saturated fat intake as % energy

Nurse’s Health Study Health Professionals’ Follow-up Study

Q1

(n=14629)

Q2

(n=14630)

Q3

(n=14629)

Q4

(n=14630)

Q5

(n=14629)

Q1

(n=8527)

Q2

(n=8527)

Q3

(n=8527)

Q4

(n=8527)

Q5

(n=8527)

Total saturated fat,% energy* 8.2±1.1 10.1±0.4 11.3±0.3 12.6±0.4 15.0±1.6 6.9±1.2 9.1±0.4 10.4±0.4 11.7±0.4 14.3±1.7

Age, years 51.7±7.1 50.6±7.2 49.8±7.2 49.5±7.1 49.4±7.0 54.2±9.6 53.5±9.7 52.8±9.6 52.5±9.4 52.7±9.4

Caucasians, % 96 98 98 98 99 93 93 94 96 96

Family history of MI, % 39 39 39 39 39 35 35 33 32 31

Alcohol intake, g/day 9.8±15.3 7.5±11.5 6.6±10.0 5.8±9.1 5.1±8.3 14.1±19.3 13.2±16.3 11.5±14.6 10.0±13.0 8.4±11.9

Current smoking, % 22 22 22 25 31 6 7 9 11 14

Multivitamin use,% 42 38 37 34 33 48 44 41 39 36

Any use of postmenopausal hormone, % 23 23 22 21 20

Current use of Asprin, % 64 67 68 69 67 26 28 28 27 26

Hypertension, % 22 21 20 20 20 21 21 19 19 18

Hypercholesterolemia, % 10 8 7 6 6 16 11 10 8 7

Physical activity, MET/week 17.8±25.2 15.0±22.7 13.4±18.3 12.8±17.9 11.9±18.9 28.4±35.9 23.2±31.6 20.4±29.0 18.6±26.2 16.5±23.4

BMI, kg/m2 24.2±4.2 24.7±4.4 25.0±4.6 25.3±4.8 25.3±5.1 24.1±4.7 24.7±4.7 25.0±4.9 25.3±5.0 25.5±5.1

Dietary factors

Total energy, kcal 1688±530 1751±520 1770±526 1770±527 1757±549 1893±601 1957±589 2021±610 2038±617 2081±668

Cereal fiber, gram/day 5.0±3.0 4.4±2.3 4.1±2.1 3.8±1.9 3.2±1.7 7.8±5.4 6.4±3.8 5.7±3.2 5.1±2.9 4.3±2.6

Fruit and vegetables, servings/day 6.3±3.0 5.6±2.5 5.2±2.3 4.8±2.1 4.2±2.0 6.9±3.4 5.8±2.6 5.3±2.4 4.8±2.2 4.2±2.1

Wholegrain carbohydrate,% energy 3.4±3.4 2.6±2.3 2.3±2.0 1.9±1.7 1.5±1.5 4.9±4.3 3.4±2.8 2.8±2.3 2.3±2.0 1.8±1.7

Non-wholegrain carbohydrate,% energy 49.7±8.3 46.2±6.3 44.0±5.9 42.1±5.7 38.6±6.3 49.5±8.7 45.5±6.9 43.8±6.5 41.8±6.1 38.6±6.4

Plant protein,% energy 5.3±1.4 5.0±1.0 4.7±0.9 4.5±0.8 4.2±0.8 5.9±1.5 5.2±1.04 4.9±0.9) 4.6±0.9 4.2±0.9

Non-plant protein,% energy 11.7±3.9 12.6±3.3 13.1±3.2 13.5±3.2 14.2±3.5 12.4±4.0 13.3±3.5 13.5±3.3) 14.0±3.3 14.6±3.4

PUFA,% energy 6.2±1.8 6.6±1.7 6.8±1.7 6.8±1.7 6.6±1.8 5.5±1.6 6.0±1.5 6.1±1.5 6.1±1.5 6.0±1.6

MUFA,% energy 10.0±1.9 11.9±1.6 12.8±1.7 13.7±1.7 14.8±1.9 9.1±2.0 11.3±1.7 12.5±1.8 13.5±1.8 14.9±2.0

Trans fat,% energy 1.6±0.6 1.9±0.6 2.0±0.6 2.1±0.6 2.1±0.6 0.8±0.4 1.2±0.4 1.3±0.4 1.4±0.5 1.6±0.5

4:0,% energy 0.15±0.06 0.18±0.06 0.21±0.07 0.24±0.08 0.33±0.14 0.10±0.05 0.14±0.06 0.16±0.07 0.19±0.08 0.27±0.13

6:0,% energy 0.08±0.03 0.10±0.04 0.11±0.05 0.13±0.05 0.20±0.09 0.06±0.03 0.08±0.03 0.09±0.04 0.11±0.05 0.16±0.08

8:0,% energy 0.05±0.02 0.06±0.02 0.06±0.03 0.07±0.03 0.09±0.05 0.04±0.02 0.05±0.02 0.05±0.02 0.06±0.03 0.08±0.05

10:0,% energy 0.11±0.04 0.14±0.04 0.16±0.05 0.19±0.06 0.27±0.10 0.09±0.04 0.12±0.04 0.14±0.04 0.16±0.05 0.23±0.10

12:0,% energy 0.15±0.09 0.19±0.11 0.22±0.13 0.26±0.15 0.36±0.21 0.12±0.08 0.16±0.10 0.19±0.12 0.22±0.13 0.31±0.28

14:0,% energy 0.70±0.17 0.89±0.15 1.01±0.17 1.16±0.20 1.53±0.36 0.54±0.16 0.74±0.15 0.87±0.16 1.01±0.18 1.35±0.36

16:0,% energy 4.84±0.66 5.90±0.34 6.54±0.35 7.18±0.40 8.34±0.85 4.16±0.72 5.40±0.34 6.12±0.34 6.81±0.40 8.03±0.87

Page 22 of 33


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review Only23

Quintiles of total saturated fat intake as % energy

Nurse’s Health Study Health Professionals’ Follow-up Study

Q1

(n=14629)

Q2

(n=14630)

Q3

(n=14629)

Q4

(n=14630)

Q5

(n=14629)

Q1

(n=8527)

Q2

(n=8527)

Q3

(n=8527)

Q4

(n=8527)

Q5

(n=8527)

18:0,% energy 2.13±0.36 2.68±0.22 3.00±0.24 3.33±0.27 3.90±0.51 1.75±0.40 2.41±0.25 2.80±0.27 3.18±0.32 3.83±0.59

Abbreviations: MET, metabolic equivalent tasks; BMI, body mass index; PUFA, polyunsaturated fatty acids; MUFA, monounsaturated fatty acids.

*, Values are means(SD) or percentages and are standardized to the age distribution of the study population.

Page 23 of 33


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nly

24

Figure 1. Aged-adjusted intake (% energy) of individual saturated fatty acids over time in the

NHS (1984–2010) and HPFS (1986–2010).

Page 24 of 33


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nly

25

Table 2. Associations between saturated fatty acids and CHD risk in NHS (1984-2012) and HPFS (1986-2010).*

Quintiles of saturated fatty acid intake (% energy)

Q1 Q2 Q3 Q4 Q5 P trend

4:0~10:0

NHS Median 0.30 0.41 0.50 0.61 0.82

Case/person year 587/370588 579/370698 560/370466 583/370160 637/368728

Age adjusted 1 1.02 (0.91, 1.15) 0.99 (0.88, 1.11) 1.04 (0.93, 1.17) 1.13 (1.01, 1.27) 0.02

Multivariate1 1 1.07 (0.95, 1.20) 1.01 (0.90, 1.14) 1.06 (0.95, 1.20) 1.08 (0.96, 1.21) 0.26

HPFS Median 0.21 0.32 0.41 0.52 0.73

Case/person year 811/173021 766/173450 791/173160 819/173069 902/172314

Age adjusted 1 0.99 (0.89, 1.09) 1.00 (0.91, 1.11) 1.03 (0.93, 1.14) 1.06 (0.96, 1.16) 0.14

Multivariate1 1 1.00 (0.90, 1.10) 1.00 (0.91, 1.11) 1.02 (0.93, 1.13) 1.01 (0.92, 1.12) 0.68

Pooled† Age adjusted 1 1.00 (0.93, 1.08) 1.00 (0.93, 1.08) 1.03 (0.96, 1.11) 1.09 (1.01, 1.17) 0.009

Multivariate1 1 1.03 (0.95, 1.11) 1.01 (0.93, 1.09) 1.04 (0.96, 1.12) 1.04 (0.96, 1.12) 0.30

12:0

NHS Median 0.12 0.16 0.20 0.24 0.37

Case/person year 593/370058 556/370205 565/370484 572/370234 660/369659

Age adjusted 1 0.99 (0.88, 1.11) 1.04 (0.92, 1.16) 1.05 (0.94, 1.18) 1.23 (1.10, 1.38) <0.001

Multivariate1 1 1.01 (0.90, 1.14) 1.05 (0.93, 1.18) 1.03 (0.92, 1.16) 1.11 (0.99, 1.24) 0.07

HPFS Median 0.10 0.14 0.18 0.22 0.34

Case/person year 844/172454 789/173116 794/173261 781/173342 881/172840

Age adjusted 1 1.00 (0.90, 1.10) 1.02 (0.93, 1.13) 1.01 (0.91, 1.11) 1.10 (1.00, 1.21) 0.03

Multivariate1 1 0.99 (0.90, 1.10) 1.02 (0.92, 1.12) 1.00 (0.90, 1.10) 1.04 (0.95, 1.15) 0.33

Pooled Age adjusted 1 0.99 (0.92, 1.07) 1.03 (0.95, 1.11) 1.02 (0.95, 1.10) 1.15 (1.07, 1.24) <0.001

Multivariate1 1 1.00 (0.93, 1.08) 1.03 (0.95, 1.11) 1.01 (0.94, 1.09) 1.07 (0.99, 1.15) 0.05

14:0

NHS Median 0.63 0.79 0.91 1.05 1.33

Case/person year 596/370997 558/370854 534/370521 605/369940 653/368328

Age adjusted 1 1.02 (0.91, 1.15) 1.03 (0.91, 1.15) 1.17 (1.05, 1.31) 1.28 (1.14, 1.43) <0.001

Multivariate1 1 1.01 (0.89, 1.13) 0.99 (0.88, 1.12) 1.08 (0.97, 1.22) 1.10 (0.98, 1.23) 0.04

HPFS Median 0.50 0.68 0.80 0.95 1.22

Case/person year 775/173398 754/173399 809/173246 812/172832 939/172138

Age adjusted 1 1.06 (0.95, 1.17) 1.17 (1.06, 1.29) 1.17 (1.06, 1.30) 1.29 (1.17, 1.42) <0.001

Multivariate1 1 1.03 (0.93, 1.14) 1.12 (1.01, 1.24) 1.09 (0.98, 1.21) 1.15 (1.04, 1.27) 0.004


Multivariate1 1 1.02 (0.94, 1.10) 1.06 (0.98, 1.15) 1.09 (1.01, 1.17) 1.13 (1.05, 1.22) <0.001

16:0

NHS Median 4.68 5.49 6.04 6.61 7.55

Case/person year 586/370403 574/370902 569/370506 591/370144 626/368685

Age adjusted 1 1.10 (0.98, 1.24) 1.18 (1.05, 1.32) 1.29 (1.15, 1.45) 1.45 (1.29, 1.62) <0.001

Multivariate1 1 1.07 (0.95, 1.20) 1.08 (0.96, 1.21) 1.13 (1.01, 1.27) 1.12 (1.00, 1.27) 0.04

HPFS Median 4.22 5.21 5.87 6.52 7.52

Case/person year 772/173340 767/173368 786/173172 840/172897 924/172237

Age adjusted 1 1.07 (0.96, 1.18) 1.14 (1.03, 1.25) 1.26 (1.14, 1.39) 1.41 (1.28, 1.56) <0.001

Multivariate1 1 1.05 (0.95, 1.16) 1.08 (0.98, 1.20) 1.15 (1.04, 1.27) 1.21 (1.09, 1.33) <0.001

Page 25 of 33


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nly

26


Multivariate1 1 1.06 (0.98, 1.14) 1.08 (1.00, 1.17) 1.14 (1.06, 1.24) 1.17 (1.08, 1.26) <0.001

18:0

NHS Median 2.02 2.46 2.76 3.07 3.56

Case/person year 553/370679 557/370728 547/370516 634/370053 655/368663

Age adjusted 1 1.13 (1.00, 1.27) 1.17 (1.04, 1.32) 1.42 (1.27, 1.59) 1.53 (1.36, 1.71) <0.001

Multivariate1 1 1.09 (0.97, 1.23) 1.07 (0.95, 1.21) 1.22 (1.09, 1.38) 1.17 (1.04, 1.32) 0.003

HPFS Median 1.76 2.30 2.67 3.03 3.58

Case/person year 730/173367 767/173438 795/173142 865/172821 932/172244

Age adjusted 1 1.11 (1.00, 1.23) 1.20 (1.08, 1.32) 1.31 (1.18, 1.44) 1.42 (1.29, 1.57) <0.001

Multivariate1 1 1.08 (0.97, 1.19) 1.13 (1.02, 1.25) 1.19 (1.07, 1.32) 1.19 (1.08, 1.32) 0.004


Multivariate1 1 1.08 (1.00, 1.17) 1.11 (1.02, 1.20) 1.20 (1.11, 1.30) 1.18 (1.09, 1.28) <0.001

12:0~18:0

NHS Median 7.58 9.00 9.96 10.98 12.64

Case/person year 575/370561 572/370873 548/370588 595/370025 656/368593

Age adjusted 1 1.12 (0.99, 1.25) 1.15 (1.02, 1.29) 1.30 (1.16, 1.46) 1.50 (1.34, 1.68) <0.001

Multivariate1 1 1.08 (0.96, 1.21) 1.05 (0.94, 1.19) 1.14 (1.01, 1.28) 1.17 (1.04, 1.31) 0.01

HPFS Median 6.71 8.42 9.57 10.72 12.49

Case/person year 754/173363 767/173360 775/173230 870/172818 923/172242

Age adjusted 1 1.08 (0.98, 1.20) 1.16 (1.04, 1.28) 1.32 (1.19, 1.45) 1.41 (1.28, 1.55) <0.001

Multivariate1 1 1.05 (0.95, 1.17) 1.10 (0.99, 1.22) 1.20 (1.08, 1.32) 1.20 (1.08, 1.32) <0.001


Multivariate1 1 1.09 (1.00, 1.18) 1.11 (1.00, 1.22) 1.21 (1.08, 1.35) 1.24 (1.08, 1.43) 0.002

*, Further adjusted for ethnicity (Caucasian, African American, Asian, and other ethnicity), smoking status

(never, former, current (1–14, 15–24, or ≥25 cigarettes/day), or missing), alcohol intake (gram/day: 0, 0.1-4.9,

5.0-14.9, and >15.0 in women, 0, 0.1-4.9, 5.0-29.9, and >30.0 in men, or missing), family history of MI (yes/no),

menopausal status and post-menopausal hormone use (pre-menopause, post-menopause (never, former, or

current hormone use), or missing, for women), physical activity (<3, 3.0-8.9, 9.0-17.9, 18.0-26.9, ≥27.0 MET

(metabolic equivalent of task) hours/week, or missing), current aspirin use (yes/no), multivitamin use (yes/no),

baseline hypertension, baseline hypercholesterolemia, BMI (<23, 23-24.9, 25-29.9, 30-34.9, >35kg/m2, or

missing), and total energy intake.

†, Study estimates from two cohorts were pooled using a fixed effects model.

Page 26 of 33


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nly

27

Table 3. Multivariate adjusted hazard ratios of CHD by substituting 1% energy from individual saturated

fatty acids by of alternative nutrients, based on NHS (1984-2012) and HPFS (1986-2010).

Hazard ratio of CHD

NHS P HPFS P Pooled|| P

PUFA to replace *

12:0-14:0 1.01 (0.85, 1.20) 0.92 1.03 (0.90, 1.18) 0.68 1.02 (0.92, 1.14) 0.70

16:0 0.85 (0.74, 0.98) 0.02 0.89 (0.80, 0.99) 0.04 0.88 (0.81, 0.95) 0.002

18:0 0.83 (0.71, 0.97) 0.02 0.97 (0.87, 1.08) 0.57 0.92 (0.84, 1.01) 0.07

12:0-18:0 0.88 (0.83, 0.93) <0.001 0.95 (0.91, 0.99) 0.03 0.92 (0.89, 0.96) <0.001

MUFA to replace †

12:0-14:0 1.07 (0.91, 1.25) 0.44 1.08 (0.95, 1.22) 0.26 1.07 (0.97, 1.18) 0.17

16:0 0.90 (0.77, 1.05) 0.19 0.93 (0.82, 1.05) 0.23 0.92 (0.83, 1.01) 0.08

18:0 0.87 (0.73, 1.05) 0.14 1.02 (0.89, 1.16) 0.80 0.97 (0.87, 1.07) 0.52

12:0-18:0 0.91 (0.83, 0.99) 0.03 0.98 (0.91, 1.05) 0.56 0.95 (0.90, 1.00) 0.07

Wholegrain

carbohydrate to

replace ‡

12:0-14:0 1.10 (0.92, 1.31) 0.30 1.04 (0.91, 1.19) 0.56 1.06 (0.95, 1.18) 0.27

16:0 0.93 (0.82, 1.06) 0.28 0.91 (0.82, 1.00) 0.06 0.92 (0.85, 0.99) 0.03

18:0 0.90 (0.76, 1.06) 0.19 0.98 (0.87, 1.10) 0.73 0.95 (0.87, 1.05) 0.30

12:0-18:0 0.95 (0.90, 1.01) 0.10 0.96 (0.92, 1.00) 0.07 0.96 (0.93, 0.99) 0.01

Plant protein to

replace §

12:0-14:0 1.12 (0.93, 1.34) 0.23 1.02 (0.89, 1.18) 0.76 1.06 (0.95, 1.18) 0.33

16:0 0.95 (0.83, 1.09) 0.44 0.89 (0.80, 0.99) 0.03 0.91 (0.84, 0.99) 0.03

18:0 0.91 (0.76, 1.08) 0.28 0.96 (0.85, 1.08) 0.51 0.94 (0.86, 1.04) 0.25

12:0-18:0 0.96 (0.89, 1.04) 0.35 0.94 (0.89, 0.99) 0.03 0.95 (0.91, 0.99) 0.02

Abbreviations: PUFA, polyunsaturated fatty acids; MUFA, monounsaturated fatty acids.

Hazard ratios were adjusted for age, ethnicity (Caucasian, African American, Asian, and other ethnicity),

smoking status (never, former, current (1–14, 15–24, or ≥25 cigarettes/day), or missing), alcohol intake

(gram/day: 0, 0.1-4.9, 5.0-14.9, and >15.0 in women, 0, 0.1-4.9, 5.0-29.9, and >30.0 in men, or missing),

family history of MI (yes/no), menopausal status and post-menopausal hormone use (pre-menopause, post-

menopause (never, former, or current hormone use), or missing, for women), physical activity (<3, 3.0-8.9,

9.0-17.9, 18.0-26.9, ≥27.0 MET (metabolic equivalent of task) hours/week, or missing), current aspirin use

(yes/no), multivitamin use (yes/no), baseline hypertension, baseline hypercholesterolemia, BMI (<23, 23-

24.9, 25-29.9, 30-34.9, >35kg/m2, or missing), total energy intake, fruit and vegetable, percentage of energy

from trans fat, carbohydrate from non-wholegrain sources, protein from non-vegetable sources;

*, further adjusted for MUFA, wholegrain carbohydrate, vegetable protein, and the sum of other saturated

fatty acids;

†, further adjusted for PUFA, wholegrain carbohydrate, vegetable protein, and the sum of other saturated

fatty acids;

Page 27 of 33


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nly

28

‡, further adjusted for PUFA, MUFA, vegetable protein, and the sum of other saturated fatty acids;

§, further adjusted for PUFA, MUFA, wholegrain carbohydrate, and the sum of other saturated fatty acids.

||, Study estimates from two cohorts were pooled using a fixed effects model.

Page 28 of 33


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nly

29

S-Table 1. Spearman correlations among individual saturated fatty acids

NHS HPFS

12:0 14:0 16:0 18:0 12:0 14:0 16:0 18:0

4:0-10:0 0.82 0.91 0.46 0.38 0.84 0.90 0.48 0.41

12:0 0.82 0.44 0.45 0.81 0.46 0.43

14:0 0.70 0.65 0.75 0.67

16:0

0.93

0.92

All P values <0.001.

Page 29 of 33


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nly

30

S-table 2. Stratified analysis of associations between saturated fatty acids and CHD risk in NHS (1984-2012) and

HPFS (1986-2010). *

Hazard ratio of CHD P

interaction Age

<65yrs P ≥65yrs P

PUFA to replace

12:0-14:0 0.91 (0.77, 1.08) 0.29 1.10 (0.96, 1.27) 0.17 0.84

16:0 0.96 (0.84, 1.10) 0.56 0.83 (0.74, 0.92) 0.000 0.81

18:0 0.92 (0.80, 1.07) 0.31 0.91 (0.81, 1.02) 0.11 0.85

12:0-18:0 0.93 (0.87, 0.98) 0.01 0.92 (0.88, 0.96) 0.000 0.83

MUFA to replace

12:0-14:0 0.96 (0.82, 1.12) 0.61 1.15 (1.01, 1.31) 0.03 0.46

16:0 1.01 (0.86, 1.18) 0.94 0.86 (0.76, 0.97) 0.02 0.41

18:0 0.97 (0.81, 1.15) 0.71 0.96 (0.84, 1.09) 0.52 0.47

12:0-18:0 0.95 (0.87, 1.05) 0.32 0.95 (0.88, 1.01) 0.10 0.38

Wholegrain carbohydrate

to replace

12:0-14:0 0.95 (0.80, 1.13) 0.57 1.15 (1.00, 1.32) 0.06 0.59

16:0 1.00 (0.88, 1.14) 0.96 0.86 (0.78, 0.95) 0.004 0.63

18:0 0.96 (0.82, 1.13) 0.63 0.94 (0.83, 1.06) 0.30 0.58

12:0-18:0 0.97 (0.91, 1.03) 0.28 0.95 (0.91, 0.99) 0.02 0.64

Plant protein to replace

12:0-14:0 0.98 (0.83, 1.18) 0.87 1.12 (0.97, 1.29) 0.12 0.30

16:0 1.04 (0.91, 1.19) 0.57 0.84 (0.76, 0.94) 0.001 0.27

18:0 0.99 (0.84, 1.17) 0.92 0.92 (0.81, 1.04) 0.16 0.30

12:0-18:0 1.00 (0.93, 1.08) 1.00 0.92 (0.88, 0.98) 0.01 0.27

BMI

<30kg/m2 P ≥30kg/m

2 P

PUFA to replace

12:0-14:0 1.04 (0.92, 1.18) 0.50 0.89 (0.68, 1.17) 0.41 0.66

16:0 0.88 (0.80, 0.97) 0.01 0.92 (0.74, 1.13) 0.43 0.68

18:0 0.93 (0.84, 1.03) 0.17 0.88 (0.71, 1.11) 0.29 0.65

12:0-18:0 0.94 (0.90, 0.97) 0.001 0.89 (0.82, 0.97) 0.01 0.67

MUFA to replace

12:0-14:0 1.09 (0.97, 1.22) 0.15 0.99 (0.77, 1.26) 0.92 0.73

16:0 0.91 (0.82, 1.02) 0.11 1.00 (0.78, 1.29) 0.98 0.77

18:0 0.98 (0.87, 1.10) 0.69 0.96 (0.74, 1.25) 0.76 0.71

12:0-18:0 0.96 (0.91, 1.02) 0.21 0.94 (0.82, 1.08) 0.36 0.78


to replace

12:0-14:0 1.09 (0.96, 1.23) 0.19 0.92 (0.70, 1.19) 0.51 0.64

16:0 0.92 (0.84, 1.00) 0.06 0.95 (0.77, 1.16) 0.61 0.65

18:0 0.96 (0.87, 1.07) 0.50 0.90 (0.71, 1.14) 0.40 0.64

12:0-18:0 0.97 (0.93, 1.01) 0.12 0.92 (0.84, 1.00) 0.06 0.65


12:0-14:0 1.07 (0.94, 1.21) 0.30 0.99 (0.75, 1.31) 0.96 0.31

16:0 0.90 (0.82, 0.99) 0.03 1.02 (0.83, 1.27) 0.83 0.30

18:0 0.94 (0.85, 1.05) 0.30 0.98 (0.77, 1.25) 0.87 0.32

12:0-18:0 0.95 (0.90, 1.00) 0.04 0.98 (0.88, 1.10) 0.77 0.31

Physical activity

<18 METs/wk P ≥18 METs/wk P

PUFA to replace

12:0-14:0 0.96 (0.84, 1.09) 0.54 1.16 (0.95, 1.42) 0.15 0.61

16:0 0.89 (0.80, 0.99) 0.03 0.82 (0.70, 0.95) 0.01 0.59

18:0 0.93 (0.83, 1.04) 0.20 0.93 (0.79, 1.09) 0.37 0.62

12:0-18:0 0.92 (0.88, 0.96) 0.000 0.93 (0.87, 0.99) 0.02 0.60

MUFA to replace

12:0-14:0 1.02 (0.91, 1.15) 0.73 1.19 (0.99, 1.44) 0.07 0.55

16:0 0.94 (0.83, 1.07) 0.34 0.83 (0.70, 0.99) 0.03 0.55

18:0 0.99 (0.86, 1.13) 0.84 0.96 (0.80, 1.16) 0.67 0.56

Page 30 of 33


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nly

31

12:0-18:0 0.96 (0.89, 1.03) 0.24 0.93 (0.85, 1.03) 0.15 0.52


to replace

12:0-14:0 1.00 (0.88, 1.14) 0.98 1.20 (0.98, 1.47) 0.08 0.32

16:0 0.93 (0.84, 1.03) 0.17 0.84 (0.73, 0.97) 0.02 0.32

18:0 0.96 (0.86, 1.09) 0.54 0.95 (0.80, 1.12) 0.55 0.32

12:0-18:0 0.96 (0.91, 1.00) 0.05 0.95 (0.90, 1.01) 0.10 0.32


12:0-14:0 0.99 (0.86, 1.13) 0.88 1.21 (0.98, 1.49) 0.07 0.53

16:0 0.92 (0.83, 1.02) 0.13 0.85 (0.74, 0.99) 0.04 0.53

18:0 0.95 (0.84, 1.08) 0.42 0.96 (0.81, 1.14) 0.62 0.52

12:0-18:0 0.94 (0.89, 1.00) 0.04 0.96 (0.89, 1.03) 0.25 0.54

Current smoking

No P Yes P

PUFA to replace

12:0-14:0 1.02 (0.90, 1.15) 0.75 1.04 (0.81, 1.33) 0.76 0.20

16:0 0.87 (0.80, 0.96) 0.004 0.91 (0.74, 1.13) 0.39 0.21

18:0 0.91 (0.82, 1.01) 0.07 1.02 (0.80, 1.31) 0.85 0.19

12:0-18:0 0.92 (0.88, 0.95) <0.001 0.99 (0.90, 1.09) 0.83 0.21

MUFA to replace

12:0-14:0 1.08 (0.96, 1.20) 0.20 0.99 (0.80, 1.24) 0.95 0.97

16:0 0.92 (0.82, 1.02) 0.11 0.88 (0.68, 1.14) 0.32 0.98

18:0 0.96 (0.86, 1.08) 0.49 1.00 (0.74, 1.34) 0.98 0.97

12:0-18:0 0.95 (0.89, 1.00) 0.07 0.96 (0.82, 1.12) 0.60 0.99


to replace

12:0-14:0 1.06 (0.94, 1.20) 0.35 1.03 (0.81, 1.32) 0.80 0.36

16:0 0.91 (0.84, 1.00) 0.04 0.92 (0.75, 1.12) 0.40 0.36

18:0 0.94 (0.85, 1.05) 0.27 1.02 (0.79, 1.32) 0.89 0.37

12:0-18:0 0.95 (0.92, 0.99) 0.01 0.98 (0.89, 1.09) 0.74 0.36


12:0-14:0 1.06 (0.93, 1.20) 0.39 1.04 (0.80, 1.35) 0.75 0.61

16:0 0.91 (0.83, 1.00) 0.04 0.92 (0.74, 1.15) 0.46 0.61

18:0 0.94 (0.84, 1.05) 0.25 1.03 (0.78, 1.35) 0.84 0.62

12:0-18:0 0.95 (0.90, 0.99) 0.02 0.99 (0.87, 1.13) 0.93 0.65


*, Other covariates adjusted were the same as the final model of table 3.

Page 31 of 33


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nly

32

S-table 3. Associations between individual saturated fatty acids and CHD risk in NHS (1984-2012) and HPFS (1986-

2010) adjusting for baseline BMI as a covariate.

Hazard ratio of CHD

NHS P HPFS P Pooled5 P

PUFA to replace

12:0-14:0 0.99 (0.83, 1.18) 0.94 1.01 (0.88, 1.16) 0.84 1.01 (0.90, 1.12) 0.91

16:0 0.87 (0.76, 0.99) 0.04 0.91 (0.82, 1.01) 0.07 0.89 (0.82, 0.97) 0.01

18:0 0.84 (0.71, 0.98) 0.03 0.99 (0.88, 1.10) 0.80 0.93 (0.85, 1.02) 0.14

12:0-18:0 0.88 (0.83, 0.94) <0.001 0.96 (0.92, 1.00) 0.07 0.93 (0.90, 0.96) <0.001

MUFA to replace

12:0-14:0 1.05 (0.90, 1.23) 0.54 1.06 (0.94, 1.20) 0.35 1.06 (0.96, 1.17) 0.27

16:0 0.92 (0.78, 1.07) 0.28 0.94 (0.83, 1.07) 0.34 0.93 (0.84, 1.03) 0.16

18:0 0.88 (0.73, 1.06) 0.17 1.04 (0.91, 1.18) 0.59 0.98 (0.88, 1.09) 0.73

12:0-18:0 0.91 (0.84, 1.00) 0.04 0.99 (0.92, 1.06) 0.81 0.96 (0.91, 1.01) 0.15

Wholegrain

carbohydrate to

replace

12:0-14:0 1.08 (0.91, 1.29) 0.38 1.03 (0.90, 1.18) 0.70 1.05 (0.94, 1.17) 0.40

16:0 0.95 (0.83, 1.08) 0.41 0.92 (0.83, 1.02) 0.11 0.93 (0.86, 1.01) 0.08

18:0 0.91 (0.77, 1.07) 0.24 1.00 (0.89, 1.12) 0.98 0.97 (0.88, 1.06) 0.50

12:0-18:0 0.96 (0.91, 1.02) 0.17 0.97 (0.93, 1.01) 0.18 0.97 (0.93, 1.00) 0.06

Plant protein to

replace

12:0-14:0 1.10 (0.91, 1.31) 0.32 1.01 (0.87, 1.16) 0.93 1.04 (0.93, 1.16) 0.50

16:0 0.96 (0.84, 1.10) 0.56 0.90 (0.81, 1.00) 0.06 0.92 (0.85, 1.00) 0.06

18:0 0.92 (0.77, 1.09) 0.33 0.98 (0.87, 1.10) 0.71 0.96 (0.87, 1.06) 0.39

12:0-18:0 0.97 (0.90, 1.04) 0.41 0.95 (0.90, 1.00) 0.07 0.96 (0.92, 1.00) 0.05



Page 32 of 33


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


nly

33

S-table 4. Associations between individual saturated fatty acids and CHD risk in NHS (1984-2012) and

HPFS (1986-2010) adjusting for incident hypercholesterolemia, hypertension and diabetes during the

follow-up.

Hazard ratio of CHD

NHS P HPFS P Pooled5 P

PUFA to replace

12:0-14:0 0.98 (0.83, 1.17) 0.85 1.02 (0.89, 1.18) 0.73 1.01 (0.91, 1.12) 0.88

16:0 0.88 (0.77, 1.01) 0.08 0.89 (0.80, 0.99) 0.03 0.89 (0.82, 0.96) 0.004

18:0 0.85 (0.72, 0.99) 0.04 0.98 (0.88, 1.09) 0.70 0.93 (0.85, 1.02) 0.14

12:0-18:0 0.89 (0.84, 0.95) <0.001 0.95 (0.91, 1.00) 0.03 0.93 (0.90, 0.96) <0.001

MUFA to replace

12:0-14:0 1.04 (0.89, 1.22) 0.61 1.07 (0.94, 1.21) 0.31 1.06 (0.96, 1.17) 0.27

16:0 0.94 (0.80, 1.10) 0.43 0.92 (0.81, 1.04) 0.17 0.92 (0.84, 1.02) 0.12

18:0 0.89 (0.74, 1.07) 0.23 1.03 (0.90, 1.17) 0.69 0.98 (0.88, 1.09) 0.72

12:0-18:0 0.93 (0.85, 1.02) 0.11 0.98 (0.92, 1.05) 0.60 0.96 (0.91, 1.02) 0.16

Wholegrain

carbohydrate to

replace

12:0-14:0 1.08 (0.90, 1.28) 0.41 1.04 (0.91, 1.19) 0.58 1.05 (0.95, 1.17) 0.34

16:0 0.97 (0.85, 1.10) 0.65 0.90 (0.82, 1.00) 0.05 0.93 (0.86, 1.01) 0.07

18:0 0.92 (0.78, 1.09) 0.35 0.99 (0.89, 1.11) 0.91 0.97 (0.88, 1.07) 0.53

12:0-18:0 0.98 (0.92, 1.03) 0.42 0.96 (0.92, 1.01) 0.10 0.97 (0.94, 1.00) 0.08

Plant protein to

replace

12:0-14:0 1.10 (0.92, 1.31) 0.31 1.02 (0.89, 1.18) 0.78 1.05 (0.94, 1.17) 0.40

16:0 0.99 (0.86, 1.13) 0.88 0.89 (0.80, 0.98) 0.02 0.92 (0.85, 1.00) 0.06

18:0 0.94 (0.79, 1.12) 0.49 0.97 (0.86, 1.10) 0.66 0.96 (0.87, 1.06) 0.45

12:0-18:0 0.99 (0.92, 1.07) 0.84 0.95 (0.90, 1.00) 0.04 0.96 (0.92, 1.00) 0.08



Page 33 of 33


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Intake o f Even -chain Saturated Fatty Acids and Risk of · 16 6 Professor, Departments of...

Documents

Transcript of Intake o f Even -chain Saturated Fatty Acids and Risk of · 16 6 Professor, Departments of...