Confidential: For Review Only Weight change across adulthood in relation to all cause and cause specific mortality: a prospective cohort study of U.S. adults Journal: BMJ Manuscript ID BMJ-2019-050369 Article Type: Research BMJ Journal: BMJ Date Submitted by the Author: 23-Apr-2019 Complete List of Authors: Chen, Chen; Huazhong University of Science and Technology, School of Public Health Ye, Yi; Huazhong University of Science and Technology, School of Public Health Zhang, Yanbo; Huazhong University of Science and Technology, School of Public Health Pan, Xiong-Fei; Huazhong University of Science and Technology, School of Public Health Pan, An; Huazhong University of Science and Technology, School of Public Health Keywords: Cohort study, Mortality, Obesity, Weight change https://mc.manuscriptcentral.com/bmj BMJ
60
Embed
Confidential: For Review Only - BMJ · Confidential: For Review Only Weight change across adulthood in relation to all cause and cause specific mortality: a prospective cohort study
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Confidential: For Review OnlyWeight change across adulthood in relation to all cause and cause specific mortality: a prospective cohort study of U.S.
adults
Journal: BMJ
Manuscript ID BMJ-2019-050369
Article Type: Research
BMJ Journal: BMJ
Date Submitted by the Author: 23-Apr-2019
Complete List of Authors: Chen, Chen; Huazhong University of Science and Technology, School of Public HealthYe, Yi; Huazhong University of Science and Technology, School of Public HealthZhang, Yanbo; Huazhong University of Science and Technology, School of Public HealthPan, Xiong-Fei; Huazhong University of Science and Technology, School of Public HealthPan, An; Huazhong University of Science and Technology, School of Public Health
Fig 1 Associations between weight change patterns across adulthood and risk of all cause mortality stratified by baseline age, smoking status and sex in NHANES 1988-2014.
Risk estimates were adjusted for baseline age (not adjusted in subgroup analysis by age), sex (not adjusted in subgroup analysis by sex), race/ethnicity, education level, household income level, marital status, drinking and smoking status (not adjusted in subgroup analysis by smoking status) and family history of diabetes and heart attack. P for interaction was <0.001, 0.95, 0.01 by baseline age, sex and smoking status respectively in age 25 years to ten years before baseline. P for interaction was <0.001, 0.49 and 0.03 by baseline age, sex and smoking status respectively in age 25 years to baseline. P for interaction was <0.001, 0.53 and 0.01 by baseline age, sex and smoking status respectively in ten years period before baseline.
Fig 2 Dose response association between weight change across adulthood and risk of all cause mortality.
Associations were examined by multivariable Cox regression based on restricted cubic splines. Dash line represents estimates of hazard ratios and dotted line represents 95% confidence intervals. Risk estimates were adjusted for baseline age, sex, race/ethnicity, education level, household income level, marital status, drinking and smoking status and family history of diabetes and heart attack. For weight change from age 25 years to ten years before baseline or to baseline, weight at age 25 and baseline height were also adjusted. For weight change from ten years before baseline to baseline, weight at 10 years ago and baseline height were adjusted. P for overall association <0.001 in all three periods. P for non-linear association <0.001 in all three periods.
Table 1 Baseline characteristics of study participants aged 40 or over in NHANES 1988-2014 according to patterns of weight change from age 25 years to baseline* Weight change patterns from age 25 years to baseline†
Characteristics Stable normal Maximum overweight Loss Gain Stable obese Total
No. of participants, n (%) 8529 (29.1) 12 955 (37.7) 590 (1.4) 9956 (26.9) 1750 (4.9) 33 780Age, mean (95% CI), years 56.8 (56.2 to 57.4) 57.8 (57.4 to 58.3) 59.9 (58.3 to 61.5) 56.9 (56.5 to 57.3) 53.5 (52.8 to 54.2) 57.1 (56.7 to 57.5)Women 4799 (63.3) 5523 (43.6) 201 (31.4) 5736 (56.9) 898 (47.8) 17 157 (52.9)BMI, mean (95% CI), kg/m2
At age 25 years 20.8 (20.7 to 20.8) 23.1 (23.1 to 23.2) 32.9 (32.5 to 33.3) 24.3 (24.2 to 24.3) 34.7 (34.4 to 35.0) 23.4 (23.3 to 23.5)At 10 years prior baseline 22.2 (22.1 to 22.2) 26.0 (25.9 to 26.1) 30.2 (29.6 to 30.9) 30.1 (29.9 to 30.2) 37.7 (37.1 to 38.3) 26.6 (26.5 to 26.7)Maximum 24.6 (24.5 to 24.7) 29.4 (29.2 to 29.4) 35.1 (34.5 to 35.7) 36.4 (36.2 to 36.5) 43.6 (43.0 to 44.2) 30.6 (30.5 to 30.8)At survey 22.2 (22.2 to 22.3) 27.1 (27.0 to 27.1) 26.4 (26.1 to 26.7) 34.5 (34.3 to 34.6) 39.3 (38.8 to 39.9) 28.2 (28.1 to 28.4)
Waist circumference, mean (95% CI), cm 83.6 (83.3 to 83.8) 96.9 (96.7 to 97.1) 97.1 (95.7 to 98.6) 111.9 (111.6 to 112.3) 121.8 (120.8 to 122.8) 98.3 (97.9 to 98.6)Race
Table 2 Hazard ratios (95% CI) of all cause and cause specific mortality with weight change patterns across adulthood among 36 812 participants in NHANES 1988-2014*
Weight change patterns†
Cause of death Stable normal Maximum overweight Loss Gain Stable obeseFrom age 25 years to ten years before baseline All cause
No. of deaths/Person-years 3094/133 393 3849/127 257 160/3724 1837/54 404 524/16 220Age-adjusted mortality rate‡ 26.4 (25.5 to 27.3) 27.8 (26.9 to 28.6) 38.3 (32.2 to 44.4) 30.2 (28.8 to 31.6) 39.1 (35.6 to 42.5)Model 1 1.00 1.05 (0.98 to 1.12) 1.42 (1.06 to 1.92) 1.30 (1.18 to 1.42) 1.85 (1.63 to 2.09)Model 2 1.00 1.08 (1.01 to 1.15) 1.28 (0.92 to 1.79) 1.33 (1.22 to 1.45) 1.93 (1.72 to 2.16)
Cardiovascular diseasesNo. of deaths 745 1096 44 515 162Model 1 1.00 1.18 (1.04 to 1.35) 1.24 (0.84 to 1.83) 1.48 (1.24 to 1.76) 2.45 (1.89 to 3.16)Model 2 1.00 1.21 (1.06 to 1.39) 1.10 (0.74 to 1.64) 1.52 (1.28 to 1.81) 2.52 (1.97 to 3.23)
CancerNo. of deaths 789 872 27 380 90Model 1 1.00 0.96 (0.83 to 1.12) 1.10 (0.60 to 2.02) 1.20 (0.97 to 1.49) 1.07 (0.81 to 1.43)Model 2 1.00 1.01 (0.88 to 1.17) 0.98 (0.51 to 1.88) 1.26 (1.02 to 1.56) 1.16 (0.87 to 1.55)
Other causesNo. of deaths 1560 1881 89 942 272Model 1 1.00 1.03 (0.94 to 1.12) 1.69 (1.12 to 2.55) 1.27 (1.13 to 1.43) 2.03 (1.67 to 2.49)Model 2 1.00 1.05 (0.96 to 1.14) 1.53 (0.99 to 2.38) 1.29 (1.15 to 1.44) 2.08 (1.71 to 2.54)
From age 25 years to baselineAll cause
No. of deaths/Person-years 2677/87 892 3915/131 968 264/4923 2302/98 464 423/15 156Age-adjusted mortality rate‡ 29.1 (28.0 to 30.2) 27.8 (26.9 to 28.7) 45.2 (39.5 to 50.8) 26.0 (24.9 to 27.2) 36.3 (32.6 to 40.1)Model 1 1.00 0.91 (0.84 to 0.99) 1.38 (1.10 to 1.73) 1.08 (0.99 to 1.17) 1.68 (1.48 to 1.92)Model 2 1.00 0.94 (0.86 to 1.03) 1.21 (0.93 to 1.58) 1.11 (1.01 to 1.21) 1.81 (1.59 to 2.05)
Cardiovascular diseasesNo. of deaths 687 1071 80 629 125Model 1 1.00 0.99 (0.84 to 1.17) 1.35 (0.95 to 1.91) 1.19 (1.02 to 1.38) 2.13 (1.61 to 2.82)Model 2 1.00 1.01 (0.86 to 1.20) 1.18 (0.83 to 1.69) 1.21 (1.05 to 1.40) 2.24 (1.71 to 2.94)
CancerNo. of deaths 604 919 47 545 73Model 1 1.00 0.92 (0.79 to 1.07) 1.04 (0.67 to 1.62) 1.05 (0.92 to 1.21) 1.05 (0.75 to 1.48)Model 2 1.00 0.97 (0.83 to 1.14) 0.91 (0.57 to 1.46) 1.11 (0.96 to 1.28) 1.18 (0.84 to 1.66)
No. of deaths 1386 1925 137 1128 225Model 1 1.00 0.87 (0.80 to 0.96) 1.57 (1.18 to 2.08) 1.05 (0.94 to 1.17) 1.84 (1.48 to 2.28)Model 2 1.00 0.89 (0.82 to 0.98) 1.38 (1.00 to 1.90) 1.07 (0.96 to 1.20) 1.94 (1.56 to 2.40)
From ten years before baseline to baseline All cause
No. of deaths/Person-years 2362/81 771 3924/131 762 852/16 428 1137/58 251 1652/56 972Age-adjusted mortality rate‡ 28.7 (27.5 to 29.9) 27.8 (26.9 to 28.7) 40.5 (37.7 to 43.4) 25.3 (23.7 to 26.9) 29.7 (28.2 to 31.2)Model 1 1.00 0.93 (0.85 to 1.02) 1.46 (1.27 to 1.68) 1.07 (0.95 to 1.20) 1.27 (1.15 to 1.40)Model 2 1.00 0.96 (0.88 to 1.04) 1.44 (1.28 to 1.61) 1.09 (0.95 to 1.25) 1.33 (1.22 to 1.45)
Cardiovascular diseasesNo. of deaths 605 1058 270 318 455Model 1 1.00 0.92 (0.79 to 1.08) 1.68 (1.33 to 2.13) 1.13 (0.93 to 1.37) 1.40 (1.15 to 1.69)Model 2 1.00 0.95 (0.80 to 1.11) 1.66 (1.32 to 2.09) 1.15 (0.94 to 1.41) 1.45 (1.22 to 1.73)
CancerNo. of deaths 567 892 138 273 354Model 1 1.00 0.91 (0.76 to 1.10) 1.13 (0.85 to 1.51) 0.95 (0.75 to 1.21) 1.14 (0.95 to 1.35)Model 2 1.00 0.96 (0.80 to 1.15) 1.13 (0.86 to 1.48) 0.99 (0.78 to 1.27) 1.24 (1.03 to 1.48)
Other causesNo. of deaths 1190 1974 444 546 843Model 1 1.00 0.94 (0.85 to 1.04) 1.51 (1.25 to 1.82) 1.11 (0.95 to 1.29) 1.28 (1.13 to 1.46)Model 2 1.00 0.96 (0.87 to 1.07) 1.46 (1.23 to 1.74) 1.13 (0.96 to 1.33) 1.33 (1.17 to 1.50)
*Complex survey designs are considered for all estimates. A total of 3382, 3032, 2230 participants were excluded in analyses from age 25 years to ten years before baseline, age 25 years to baseline, ten years before baseline to baseline respectively due to missing values of BMI at both times. Model 1 were adjusted for baseline age, sex, race/ethnicity. Model 2 were additionally adjusted for additionally adjusted for education level, household income level, marital status, drinking and smoking status (never, ever, current) and family history of diabetes and heart attack.†Weight change patterns were defined by BMI at two time points. Stable normal participants had BMI <25.0 kg/m2 at both times, maximum overweight participants had BMI in 25.0-29.9 kg/m2 at either time but not ≥30.0 kg/m2 at the other time, weight loss participants had BMI ≥30.0 kg/m2 at younger age while <30.0 kg/m2 later, weight gain participants had BMI <30.0 kg/m2 at younger age while ≥30.0 kg/m2 later, and stable obese participants had BMI ≥30.0 kg/m2 at both times. ‡Mortality rates per 1000 person-years, directly standardized to the age distribution of the entire study population
Table 3 Hazard ratios of all cause and cause specific mortality with absolute weight change groups across adulthood in 36 812 participants in NHANES 1988-2014*
Weight change in three life periodsCause of death Weight loss ≥2.5 kg Weight loss <2.5
or gain <2.5 kg 2.5 kg ≤ Weight gain <10.0 kg 10 kg ≤ Weight gain <20 kg Weight gain ≥20 kg
From age 25 years to ten years before baselineAll cause
No. of deaths 762 2152 2661 2191 1698 Age-adjusted mortality rate† 37.2 (34.5 to 39.8) 30.0 (28.7 to 31.3) 26.2 (25.2 to 27.2) 26.2 (25.1 to 27.3) 31.1 (29.6 to 32.7) Model 1 1.30 (1.11 to 1.53) 1.00 0.90 (0.82 to 0.99) 0.95 (0.86 to 1.05) 1.17 (1.05 to 1.30) Model 2 1.07 (0.92 to 1.25) 1.00 0.97 (0.87 to 1.07) 1.06 (0.95 to 1.18) 1.28 (1.15 to 1.42)
Cardiovascular diseases No. of deaths 211 565 703 618 465 Model 1 1.17 (0.96 to 1.43) 1.00 0.82 (0.70 to 0.96) 0.85 (0.72 to 1.00) 1.11 (0.96 to 1.30) Model 2 0.90 (0.75 to 1.10) 1.00 0.88 (0.74 to 1.05) 0.96 (0.80 to 1.15) 1.26 (1.08 to 1.47)Cancer No. of deaths 170 510 662 453 363 Model 1 1.34 (1.05 to 1.72) 1.00 0.94 (0.79 to 1.11) 0.96 (0.77 to 1.11) 1.14 (0.91 to 1.44) Model 2 1.19 (0.92 to 1.53) 1.00 0.99 (0.83 to 1.19) 1.04 (0.83 to 1.31) 1.21 (0.96 to 1.53)Other causes No. of deaths 381 1077 1296 1120 870 Model 1 1.37 (1.10 to 1.72) 1.00 0.95 (0.83 to 1.09) 1.02 (0.91 to 1.15) 1.23 (1.09 to 1.40) Model 2 1.12 (0.89 to 1.42) 1.00 1.02 (0.88 to 1.18) 1.14 (1.01 to 1.29) 1.35 (1.19 to 1.53)From age 25 years to baselineAll cause No. of deaths 1801 1059 2027 2417 2277 Age-adjusted mortality rate† 44.4 (42.2 to 46.6) 29.1 (27.3 to 30.9) 26.5 (25.4 to 27.7) 25.6 (24.6 to 26.6) 25.7 (24.5 to 26.8) Model 1 1.44 (1.26 to 1.64) 1.00 0.97 (0.86 to 1.09) 0.91 (0.82 to 1.02) 1.09 (0.98 to 1.21) Model 2 1.21 (1.06 to 1.38) 1.00 1.01 (0.90 to 1.13) 0.99 (0.89 to 1.10) 1.16 (1.05 to 1.28)Cardiovascular disease No. of deaths 506 296 563 622 605 Model 1 1.18 (0.96 to 1.46) 1.00 0.88 (0.72 to 1.08) 0.84 (0.71 to 0.99) 1.01 (0.83 to 1.22) Model 2 0.95 (0.75 to 1.19) 1.00 0.92 (0.76 to 1.13) 0.92 (0.79 to 1.08) 1.11 (0.92 to 1.34)Cancer
No. of deaths 394 210 475 572 537Model 1 1.67 (1.35 to 2.07) 1.00 1.08 (0.89 to 1.31) 1.03 (0.85 to 1.26) 1.16 (0.98 to 1.37)Model 2 1.48 (1.17 to 1.86) 1.00 1.12 (0.93 to 1.34) 1.10 (0.89 to 1.36) 1.20 (1.01 to 1.42)
Other causes No. of deaths 901 553 989 1223 1135 Model 1 1.50 (1.25 to 1.79) 1.00 0.98 (0.83 to 1.15) 0.92 (0.79 to 1.08) 1.13 (0.98 to 1.30) Model 2 1.26 (1.03 to 1.53) 1.00 1.02 (0.87 to 1.19) 1.00 (0.86 to 1.16) 1.20 (1.06 to 1.37)From ten years before baseline to baselineAll cause No. of deaths 3853 2205 2184 1115 570 Age-adjusted mortality rate† 38.5 (37.2 to 39.7) 25.5 (24.4 to 26.6) 23.6 (22.6 to 24.6) 25.4 (23.7 to 27.1) 28.5 (25.3 to 31.6) Model 1 1.49 (1.36 to 1.64) 1.00 0.99 (0.91 to 1.07) 1.14 (1.04 to 1.25) 1.68 (1.43 to 1.97) Model 2 1.30 (1.17 to 1.44) 1.00 1.01 (0.93 to 1.09) 1.12 (1.02 to 1.23) 1.57 (1.31 to 1.87)Cardiovascular disease No. of deaths 1106 613 567 267 153 Model 1 1.37 (1.12 to 1.67) 1.00 1.03 (0.86 to 1.23) 1.02 (0.80 to 1.31) 1.82 (1.35 to 2.44) Model 2 1.16 (0.94 to 1.44) 1.00 1.04 (0.88 to 1.23) 0.99 (0.78 to 1.25) 1.70 (1.25 to 2.31)Cancer No. of deaths 773 478 562 281 130 Model 1 1.45 (1.24 to 1.71) 1.00 1.02 (0.86 to 1.21) 1.07 (0.90 to 1.29) 1.36 (1.01 to 1.82) Model 2 1.29 (1.09 to 1.53) 1.00 1.04 (0.87 to 1.25) 1.06 (0.88 to 1.29) 1.24 (0.93 to 1.66)Other causes No. of deaths 1974 1114 1055 567 287 Model 1 1.57 (1.41 to 1.75) 1.00 0.95 (0.84 to 1.07) 1.24 (1.09 to 1.42) 1.80 (1.41 to 2.30) Model 2 1.38 (1.22 to 1.54) 1.00 0.97 (0.86 to 1.09) 1.23 (1.07 to 1.41) 1.70 (1.32 to 2.19)
*Complex survey designs were taken into account for all estimates. A total of 3382, 3032, 2230 participants were excluded in analyses from age 25 years to ten years before baseline, age 25 years to baseline, ten years before baseline to baseline respectively due to missing values of BMI at both times. Model 1 were adjusted for baseline age, sex and race/ethnicity. In analyses from age 25 years to ten years before baseline and age 25 years to baseline, model 2 were adjusted for baseline age, sex, race/ethnicity, height, weight at age 25, education level, household income level, marital status, drinking and smoking status (never, ever, current) and family history of diabetes and heart attack. In analysis from ten years before baseline to baseline, model 2 were adjusted for baseline age, sex, race/ethnicity, height, weight at ten years ago, education level, household income level, marital status, drinking and smoking status (never, ever, current) and family history of diabetes and heart attack. †Mortality rates per 1000 person-years, directly standardized to the age distribution of the entire study population
Fig 1 Associations between weight change patterns across adulthood and risk of all cause mortality stratified by baseline age, smoking status and sex in NHANES 1988-2014.
Risk estimates were adjusted for baseline age (not adjusted in subgroup analysis by age), sex (not adjusted in subgroup analysis by sex), race/ethnicity, education level, household income level, marital status,
drinking and smoking status (not adjusted in subgroup analysis by smoking status) and family history of diabetes and heart attack. P for interaction was <0.001, 0.95, 0.01 by baseline age, sex and smoking status
respectively in age 25 years to ten years before baseline. P for interaction was <0.001, 0.49 and 0.03 by baseline age, sex and smoking status respectively in age 25 years to baseline. P for interaction was <0.001,
0.53 and 0.01 by baseline age, sex and smoking status respectively in ten years period before baseline.
Fig 2 Dose response association between weight change across adulthood and risk of all cause mortality.
Associations were examined by multivariable Cox regression based on restricted cubic splines. Dash line represents estimates of hazard ratios and dotted line represents 95% confidence intervals. Risk estimates
were adjusted for baseline age, sex, race/ethnicity, education level, household income level, marital status, drinking and smoking status and family history of diabetes and heart attack. For weight change from age 25 years to ten years before baseline or to baseline, weight at age 25 and baseline height were also adjusted. For weight change from ten years before baseline to baseline, weight at 10 years ago and baseline height
were adjusted. P for overall association <0.001 in all three periods. P for non-linear association <0.001 in all three periods.
Table S1 Weight change patterns across adulthood in NHANES 1988-2014
Table S2 Characteristics of 36 812 study participants aged 40 years or over in NHANES1988-2014 according to weight change from age 25 years to ten years before
baseline
Table S3 Characteristics of 36 812 study participants aged 40 years or over in NHANES1988-2014 according to weight change in the ten year period before baseline
Table S4 Hazard ratios of all cause mortality with obesity status at three age points among 36 812 participants in NHANES 1988-2014
Table S5 Hazard ratios of all cause mortality with weight change patterns stratified by study period
Table S6 Hazard ratios of all cause and cause specific mortality with weight change patterns across adulthood in 36 812 participants in NHANES 1988-2014 with further
adjustment of disease status
Table S7 Hazard ratios of all cause mortality with weight change patterns among 36 812 stratified by age at death in NHANES 1988-2014
Table S8 Hazard ratios of all cause mortality with weight change patterns after excluding participants who were underweight in NHANES 1988-2014
Table S9 Hazard ratios of all cause and cause specific mortality with weight change patterns across adulthood among 31 077 participants after excluding deaths within the
first three years of follow-up in NHANES 1988-2014
Table S10 Hazard ratios of all cause and cause specific mortality with absolute weight change groups across adulthood in 36 812 participants in NHANES 1988-2014 with
Table S1 Weight change patterns across adulthood in NHANES 1988-2014From age 25 yearsto ten years before baseline From age 25 years to baseline From ten years before baseline
to baselineAbsolute weight change, median (IQR), kg 6.80 (1.81, 13.6) 11.70 (4.10, 21.07) 3.67 (-1.61, 10.13)Stable Normal
Table S2 Baseline characteristics of study participants aged 40 or over in NHANES1988-2014 according to patterns of weight change from age 25 years to ten years before baseline a
Weight change patterns bCharacteristics Stable normal Maximum overweight Loss Gain Stable obese Total
a Complex survey designs are considered for all estimates; percentages may not add up to 100% due to missing values. b From age 25 years to ten years before baseline, stable normal: BMI25 < 25 kg/m2 and BMI10prior < 25 kg/m2; maximum overweight: either BMI25 or BMI10prior was 25.0-29.9 kg/m2 and the other BMI not ≥30.0 kg/m2; loss: BMI25 ≥ 30 kg/m2 and 25 ≤ BMI10prior < 30 kg/m2; gain: 25 ≤ BMI25 < 30 kg/m2 and BMI10prior≥ 30 kg/m2; stable obese: BMI25 ≥ 30 kg/m2 and BMI10prior ≥ 30 kg/m2.
Table S3 Baseline characteristics of study participants aged 40 or over in NHANES1988-2014 according to weight change patterns in ten years period before baseline a
Weight change patterns bCharacteristics Stable normal Maximum overweight Loss Gain Stable obese Total
Cancer 1059 (14.0) 1729 (13.6) 286 (14.7) 551 (11.5) 776 (12.5) 4401 (13.2)a Complex survey design factors were taken into account for all estimates; percentages may not add up to 100% due to missing values.b From ten years before baseline to baseline, stable normal: BMI10prior < 25 kg/m2 and BMIbaseline < 25 kg/m2; max overweight: either BMI10prior or BMIbaseline was 25.0-29.9 kg/m2 and the other BMI not ≥30.0 kg/m2; losing: BMI10prior ≥ 30 kg/m2 and 25 ≤ BMIbaseline < 30 kg/m2; gaining: 25 ≤ BMI10prior < 30 kg/m2 and BMIbaseline ≥ 30 kg/m2; stable obese: BMI10prior ≥ 30 kg/m2 and BMIbaseline ≥ 30 kg/m2.
Table S4 Hazard ratios of all cause mortality with adiposity status at different age points among 36 812 participants in NHANES 1988-2014 aStandard BMI groupsUnderweight Normal weight Overweight Obese
a Complex survey designs are considered for all estimates. A total of 2897, 2094, 761, 2324 participants were excluded due to missing values of adiposity indicator at specific age point respectively. Model 1 were adjusted for baseline age, sex, race, education level, marital status, income level, drinking status, smoking status (never, ever, current) and family history of chronic metabolic diseases. Model 2 were additionally adjusted for baseline history of diabetes, hypertension, cardiovascular diseases, cancer and chronic obstructive pulmonary diseases based on model 1.
Table S5 Hazard ratios of all cause mortality with weight change patterns stratified by study period aWeight change patterns bStratified by study period Stable Normal Maximum overweight Loss Gain Stable obese
From age 25 years to ten years before baselineNHANES 1988-2014
a Complex survey designs are considered for all estimates. A total of 3382, 3032, 2230 participants were excluded in analyses from age 25 years to ten years before baseline, age 25 years to baseline, ten years before baseline to baseline respectively due to missing values of BMI at both times. Risk estimates were adjusted for baseline age, sex,
race/ethnicity, education level, household income level, marital status, drinking and smoking status (never, ever, current) and family history of diabetes and heart attack.b Weight change patterns were defined by BMI at two time points. Stable normal participants had BMI <25.0 kg/m2 at both times, maximum overweight participants had BMI in 25.0-29.9 kg/m2 at either time but not ≥30.0 kg/m2 at the other time, weight loss participants had BMI ≥30.0 kg/m2 at younger age while <30.0 kg/m2 later, weight gain participants had BMI <30.0 kg/m2 at younger age while ≥30.0 kg/m2 later, and stable obese participants had BMI ≥30.0 kg/m2 at both times.
Table S6 Hazard ratios of all cause and cause specific mortality with weight change patterns across adulthood in 36 812 participants in NHANES 1988-2014 with further adjustment of disease status a
Weight change patternsCause of death Stable normal Maximum overweight Loss Gain Stable obeseFrom age 25 years to ten years before baselineAll cause
a For all cause mortality, risk estimates were adjusted for baseline age, sex, race/ethnicity, education level, household income level, marital status, drinking and smoking status (never, ever, current), family history of diabetes and heart attack and baseline history of diabetes, hypertension, cardiovascular diseases, cancer and chronic obstructive pulmonary diseases. For cardiovascular disease mortality, risk estimates were adjusted for baseline age, sex, race/ethnicity, education level, household income level, marital status, drinking and smoking status (never, ever, current), family history of diabetes and heart attack and baseline history of diabetes, hypertension, cancer and chronic obstructive pulmonary diseases. For cancer mortality, risk estimates were adjusted for baseline age, sex, race/ethnicity, education level, household income level, marital status, drinking and smoking status (never, ever, current), family history of diabetes and heart attack and baseline history of diabetes, hypertension, cardiovascular diseases and chronic obstructive pulmonary diseases. For other-causes mortality, risk estimates were adjusted for baseline age, sex, race/ethnicity, education level, household income level, marital status, drinking and smoking status (never, ever, current), family history of diabetes and heart attack and baseline history of cardiovascular diseases and cancer.
Table S7 Hazard ratios of all cause mortality with weight change patterns among 36 812 stratified by age at death aWeight change patternsStable Normal Maximum overweight Loss Gain Stable obese P for interaction
From age 25 years to ten years before baseline <.0001Stratified by death age groupDead before 65y
Dead between 66y and 75y 395 685 175 238 396No. of deaths 1.00 0.97 (0.81, 1.17) 1.15 (0.88, 1.51) 0.87 (0.66, 1.15) 0.94 (0.80, 1.10)Multivariable-adjusted HR (95% CI)
Dead between 76y-85yNo. of deaths 769 1338 307 390 586Multivariable-adjusted HR (95% CI) 1.00 0.95 (0.82, 1.10) 1.37 (1.12, 1.66) 1.11 (0.94, 1.32) 1.11 (0.97, 1.27)
Dead after 85yNo. of deaths 891 1441 272 254 386Multivariable-adjusted HR (95% CI) 1.00 0.96 (0.83, 1.09) 1.40 (1.11, 1.77) 1.00 (0.77, 1.32) 1.26 (1.03, 1.53)
a Complex survey designs are considered for all estimates. Risk estimates were adjusted for baseline age, sex, race/ethnicity, education level, household income level, marital status, drinking and smoking status (never, ever, current) and family history of diabetes and heart attack.
Table S8 Hazard ratios of all cause mortality with weight change patterns after excluding participants who were underweight in NHANES 1988-2014 aWeight change patternsStable Normal Maximum overweight Loss Gain Stable obese
From age 25 years to ten years before baselineAll participants
a Complex survey designs are considered for all estimates. A total of 3382, 3032, 2230 participants were excluded in analyses from age 25 years to ten years before baseline, age 25 years to baseline, ten years before baseline to baseline respectively due to missing values of BMI at both times. Risk estimates were adjusted for baseline age, sex, race/ethnicity, education level, household income level, marital status, drinking and smoking status (never, ever, current) and family history of diabetes and heart attack.
Table S9 Hazard ratios of all cause and cause specific mortality with weight change patterns across adulthood among 31 077 participants after excluding deaths within three years of follow-up in NHANES 1988-2014 a
Weight change patternsCause of death b Stable normal Maximum overweight Loss Gain Stable obese
All cause
From age 25 years to ten years before baselineNo. of deaths 2597 3118 127 1462 404Model 1 1.00 1.05 (0.97, 1.13) 1.30 (0.89, 1.88) 1.32 (1.20, 1.45) 1.98 (1.74, 2.26)Model 2 1.00 1.02 (0.94, 1.10) 1.24 (0.84, 1.84) 1.17 (1.06, 1.29) 1.64 (1.41, 1.91)From age 25 years to baseline No. of deaths 2172 3187 188 1923 343Model 1 1.00 0.93 (0.84, 1.02) 1.19 (0.86, 1.63) 1.12 (1.02, 1.24) 1.90 (1.65, 2.18)Model 2 1.00 0.92 (0.83, 1.01) 1.10 (0.78, 1.54) 1.03 (0.93, 1.14) 1.58 (1.36, 1.85)From ten years before baseline to baselineNo. of deaths 1940 3172 634 982 1335Model 1 1.00 0.93 (0.84, 1.03) 1.42 (1.23, 1.63) 1.12 (0.96, 1.31) 1.34 (1.22, 1.48)Model 2 1.00 0.91 (0.81, 1.01) 1.23 (1.08, 1.41) 1.03 (0.88, 1.20) 1.16 (1.04, 1.28)
Cardiovascular diseases
From age 25 years to ten years before baselineNo. of deaths 623 879 39 406 117Model 1 1.00 1.16 (1.00, 1.35) 1.20 (0.79, 1.81) 1.50 (1.25, 1.81) 2.56 (1.93, 3.40)Model 2 1.00 1.11 (0.95, 1.30) 1.20 (0.78, 1.84) 1.30 (1.07, 1.59) 2.09 (1.55, 2.83)From age 25 years to baseline No. of deaths 552 859 54 529 99Model 1 1.00 0.98 (0.81, 1.18) 1.11 (0.73, 1.69) 1.21 (1.04, 1.41) 2.37 (1.77, 3.17)Model 2 1.00 0.95 (0.79, 1.15) 1.07 (0.69, 1.66) 1.08 (0.92, 1.26) 1.93 (1.43, 2.60)From ten years before baseline to baselineNo. of deaths 485 858 194 279 367Model 1 1.00 0.91 (0.75, 1.09) 1.57 (1.20, 2.04) 1.14 (0.92, 1.40) 1.45 (1.19, 1.77)Model 2 1.00 0.87 (0.72, 1.05) 1.36 (1.03, 1.80) 1.03 (0.84, 1.27) 1.22 (1.00, 1.50)
From age 25 years to ten years before baselineNo. of deaths 1328 1555 70 762 217Model 1 1.00 1.02 (0.92, 1.12) 1.48 (0.90, 2.43) 1.27 (1.12, 1.45) 2.16 (1.73, 2.69)Model 2 1.00 1.02 (0.92, 1.12) 1.44 (0.86, 2.42) 1.26 (1.10, 1.43) 2.08 (1.66, 2.60)From age 25 years to baselineNo. of deaths 1146 1595 105 954 184Model 1 1.00 0.88 (0.79, 0.98) 1.39 (0.96, 2.03) 1.09 (0.96, 1.23) 2.01 (1.59, 2.55)Model 2 1.00 0.88 (0.79, 0.99) 1.35 (0.90, 2.01) 1.07 (0.94, 1.21) 1.94 (1.54, 2.45)From ten years before baseline to baselineNo. of deaths 1004 1617 341 476 690Model 1 1.00 0.94 (0.83, 1.05) 1.47 (1.20, 1.80) 1.13 (0.96, 1.39) 1.33 (1.16, 1.53)Model 2 1.00 0.93 (0.83, 1.05) 1.44 (1.18, 1.76) 1.12 (0.93, 1.35) 1.30 (1.13, 1.50)
a Complex survey designs are considered for all estimates. Model 1 were adjusted for baseline age, sex, race/ethnicity, education level, household income level, marital status, drinking and smoking status (never, ever, current) and family history of diabetes and heart attack. b For all cause mortality, model 2 were additionally adjusted for baseline history of diabetes, hypertension, cardiovascular diseases, cancer and chronic obstructive pulmonary diseases. For CVD mortality, model 2 were additionally adjusted for baseline history of diabetes, hypertension, cancer and chronic obstructive pulmonary diseases. For cancer mortality, model 2 were additionally adjusted for baseline history of diabetes, hypertension, cardiovascular diseases and chronic obstructive pulmonary diseases. For other-causes mortality, model 2 were additionally adjusted for baseline history of cardiovascular diseases and cancer.
Table S10 Hazard ratios (95% CI) of all cause and cause specific mortality with weight change in three life periods in 36 812 participants in NHANES 1988-2014 with further adjustment of disease status a
Weight change in three life periodsCause of death b Weight loss ≥2.5 kg Weight loss <2.5 kg
or gain <2.5 kg 2.5 kg ≤ Weight gain < 10 kg 10 kg ≤ Weight gain <20 kg Weight gain ≥20 kg
From age 25 years to ten years before baselineAll cause
No. of deaths 3853 2205 2184 1115 570 Age-adjusted mortality rate c 38.5 (37.2, 39.7) 25.5 (24.4, 26.6) 23.6 (22.6, 24.6) 25.4 (23.7, 27.1) 28.5 (25.3, 31.6) HR (95% CI) 1.25 (1.12, 1.38) 1.00 1.00 (0.92, 1.08) 1.09 (0.98, 1.20) 1.37 (1.14, 1.66)Cardiovascular disease No. of deaths 1106 613 567 267 153 HR (95% CI) 1.14 (0.92, 1.41) 1.00 1.04 (0.88, 1.23) 0.97 (0.76, 1.22) 1.49 (1.09, 2.05)Cancer No. of deaths 773 478 562 281 130 HR (95% CI) 1.27 (1.07, 1.50) 1.00 1.04 (0.87, 1.26) 1.07 (0.88, 1.29) 1.22 (0.92, 1.62)Other causes No. of deaths 1974 1114 1055 567 287 HR (95% CI) 1.34 (1.19, 1.51) 1.00 0.96 (0.85, 1.08) 1.20 (1.04, 1.38) 1.59 (1.22, 2.08)a Complex survey designs were taken into account for all estimates. A total of 3382, 3032, 2230 participants were excluded in analyses from age 25 years to ten years before baseline, age 25 years to baseline, ten years before baseline to baseline respectively due to missing values of BMI at both times. In analyses from age 25 years to ten years before baseline and age 25 years to baseline, risk estimates were adjusted for baseline age, sex, race/ethnicity, height, weight at age 25, education level, household income level, marital status, drinking and smoking status (never, ever, current) and family history of diabetes and heart attack. In analysis from ten years before baseline to baseline, risk estimates were adjusted for baseline age, sex, race/ethnicity, height, weight at ten years ago, education level, household income level, marital status, drinking and smoking status (never, ever, current) and family history of diabetes and heart attack.b For all cause mortality, risk estimates were additionally adjusted for baseline history of diabetes, hypertension, cardiovascular diseases, cancer and chronic obstructive pulmonary diseases. For cardiovascular disease mortality, risk estimates were additionally adjusted for baseline history of diabetes, hypertension, cancer and chronic obstructive pulmonary diseases. For cancer mortality, risk estimates additionally adjusted for baseline history of diabetes, hypertension, cardiovascular diseases and chronic obstructive pulmonary diseases. For other-causes mortality, risk estimates additionally adjusted for baseline history of cardiovascular diseases and cancer based on model 2.
c Mortality rates per 1000 person-years, directly standardized to the age distribution of the entire study population.
Figure S2 Association between fine weight change patterns across adulthood and all cause mortality in NHANES 1988-2014. All risk estimates were calculated using multivariable Cox regression adjusted for baseline age, sex, race/ethnicity, education level, household income level, marital status, drinking and smoking status and family history of diabetes and heart attack.a Only two participants were in the group so that the effect estimates were not stable and were not shown.b Only three participants were in the group so that the effect estimates were not stable and were not shown.c Only five participants were in the group so that the effect estimates were not stable and were not shown.
Confidential: For Review OnlyFigure S3 Dose response association between weight change across adulthood and risk of cause specific mortality. Associations were examined by multivariable Cox regression based on restricted cubic splines. Dash line represents estimates of hazard ratios and dotted line represents 95% confidence intervals. Risk estimates were adjusted for baseline age, sex, race/ethnicity, education level, household income level, marital status, drinking and smoking status and family history of diabetes and heart attack. For weight change from age 25 years to ten years before baseline or to baseline, weight at age 25 and baseline height were also adjusted. For weight change from ten years before baseline to baseline, weight at 10 years ago and baseline height were adjusted.