Of the 3318 participants enrolled in the HCS, 584 participants with existing CVD at baseline were excluded. Twenty-one participants with no CVD were also excluded (10 underweight with missing metabolic health data, and 11 underweight with non-missing metabolic health data). Of the remaining 2713 participants, 400 had missing metabolic health data, leaving 2313 participants with complete metabolic health and BMI data allowing classification into one of the six phenotypes. A further 448 participants with incomplete confounding variable data were excluded, leaving 1865 people with complete phenotype and confounding variable data included in the current analysis (Fig. 1).
Baseline characteristics of study population
Sociodemographic and baseline characteristics of the total study population according to metabolic health and BMI phenotype are reported in Table 1. Approximately 23.4% and 12.9% of the overall participants were classified as having MHOW and MHO respectively. Of the participants with obesity, 38% were classified as metabolically healthy.
Table 1
Baseline Characteristics of participants in the six obesity phenotypes.
| Normal Weight | Overweight | Obese | |
| Metabolically Healthy (N = 326) | Metabolically Unhealthy (N = 167) | Metabolically Healthy (N = 541) | Metabolically Unhealthy (N = 495) | Metabolically Healthy (N = 298 ) | Metabolically Unhealthy (N = 486) | Total (N = 2313) |
Male (%) | 33.5 | 31.7 | 52.4 | 52.6 | 41.7 | 44.7 | 45.3 |
Age(years), Mean, SD | 64.7 (7.5) | 67.6 (7.5) | 64.2 (6.7) | 67.0 (7.6) | 63.3 (6.3) | 65.5 (6.7) | 65.3 (7.1) |
BMI (kg/m2) Mean, SD | 23.0 (1.5) | 23.1 (1.5) | 27.3 (1.4) | 27.6 (1.4) | 33.4 (3.5) | 34.6 (4.4) | 28.8 (5.0) |
SBP (mmHg) Mean, SD | 129.4 (18.6) | 137.7 (20.6) | 134.5 (18.3) | 140.4 (20.1) | 133.7 (16.7) | 137.0 (16.9) | 135.7 (18.7) |
DBP (mmHg) Mean, SD | 76.5 (9.4) | 77.8 (9.0) | 80.0 (9.5) | 79.9 (10.1) | 79.7 (9.2) | 79.4 (10.3) | 79.2 (9.8) |
Trig (mmol/L) Mean, SD | 1.0 (0.5) | 1.2 (0.7) | 1.2 (0.6) | 1.5 (0.9) | 1.4 (0.9) | 1.9 (1.5) | 1.4 (1.0) |
HDL-chol (mmol/L) Mean, SD | 1.6 (0.4) | 1.5 (0.4) | 1.4 (0.4) | 1.3 (0.3) | 1.4 (0.4) | 1.2 (0.3) | 1.4 (0.4) |
Fglu (mmHg) Mean, SD | 4.7 (0.6) | 4.9 (1.1) | 4.8 (0.6) | 5.2 (1.3) | 4.9 (0.6) | 5.7 (1.7) | 5.1 (1.2) |
HHI 1 (%) | 49.3 | 67.1 | 49.2 | 60.3 | 47.5 | 67.9 | 56.5 |
HHI 2 (%) | 50.7 | 32.9 | 50.8 | 39.7 | 52.6 | 32.1 | 43.5 |
HLE 1 (%) | 43.8 | 47.9 | 39.7 | 45.8 | 46.2 | 53.1 | 45.8 |
HLE 2 (%) | 56.3 | 52.1 | 60.3 | 54.2 | 53.8 | 46.9 | 54.2 |
SS Never 1 (%) | 60.8 | 58.7 | 54.9 | 54.5 | 52.7 | 57.6 | 56.2 |
SS Ever 2 (%) | 30.2 | 33.5 | 36.3 | 40.0 | 38.8 | 35.8 | 36.2 |
SS Now 3 (%) | 9.0 | 7.8 | 8.9 | 5.5 | 8.5 | 6.6 | 7.6 |
ETOH Mean, SD | 1.8 (2.1) | 1.4 (2.2) | 2.1 (2.5) | 2.1 (2.9) | 2.1 (2.7) | 2.0 (3.0) | 2.0 (2.7) |
PA Mean, SD | 8438.7(2916.4) | 7638.1 (3047.9) | 7386.6 (3110.8) | 6997.5 (3241.5) | 6396.1 (2825.7) | 5836.9 (3004.3) | 7032.1 (3158.9) |
Categorical variables are presented in percentage (%). |
Continuous variables are presented in mean and standard deviation (SD). |
BMI = Body Mass Index (kg/m2). |
SBP = Systolic Blood Pressure (mmHg). |
DBP = Diastolic Blood Pressure (mmHg). |
Trig = Triglycerides (mm/L). |
HDL-Chol = High density lipoprotein cholesterol (mm/L). |
Fglu = Fasting Glucose (mm/L). |
HHI = Household income before tax, 1: $5000 to $39,999 per year, 2: $40,000 + per year. |
HLE = Highest Level of Education, 1: Primary and Secondary, 2: Tertiary and University. |
SS = Smoking status: 1: Never, 2: Ever, 3: Now. |
ETOH = Alcohol use: mean daily alcohol use. |
PA = mean daily steps. |
Compared to the healthy normal weight group, the MHOW or MHO groups were younger, had higher a proportion of males, higher household income and level of education, fewer never smokers, and lower daily step counts. However, participants categorised as metabolically healthy and having overweight or obesity were younger, more educated with higher household income, were less likely to smoke, and more physically active than the unhealthy group (Table 1).
Outcome events for CVD, CVD-related death and all-cause mortality
After a median follow up of 9.7 years for participants across all obesity phenotypes, 283 developed CVD. There were 58 CVD related deaths and 277 deaths from all causes.
Mhow And Mho And Risk Of Cvd
Bayesian survival analysis
Adjusted complete case uninformative prior
In a multivariable Bayesian model of complete cases and uninformative prior, the adjusted hazard ratio (HR) for CVD in MHOW was 2.48 (95% credible interval: 1.31–5.05) and in MHO was 2.51 (95% credible interval: 1.23–5.40). The HRs for CVD were similarly increased in the remaining metabolically unhealthy phenotypes with even higher HRs observed for those classified as overweight/obese and metabolically unhealthy; MUNW (HR = 2.41, 95% credible interval: 1.09–5.41), MUOW (HR = 3.76, 95% credible interval: 2.06–7.53), MUO (HR = 3.65, credible interval: 1.93–7.33). The reference group was MHNW for all the obesity phenotypes (Table 2). In all models the Rhat was < 1.05 and pd was > 95%.
Table 2: Bayesian analysis results with median hazard ratio for risk of developing Cardiovascular disease in older Australians from The Hunter Community Study stratified by BMI and metabolic health.
|
|
Normal Weight
|
Overweight
|
Obese
|
Number of participants
|
Total = 2313
|
Metabolically Healthy (N=326)
|
Metabolically Unhealthy (N=167)
|
Metabolically Healthy(N=541)
|
Metabolically Unhealthy (N=495)
|
Metabolically Healthy (N=298)
|
Metabolically Unhealthy (N=486)
|
Incident CVD (n)
|
Total = 283
|
19
|
18
|
56
|
84
|
31
|
75
|
Unadjusted Complete cases
N=2313
|
Median HR
95% Credible Interval
Pd
Rhat
|
1
|
1.93
1.04-3.72
97.80%
1.000
|
1.89
1.15-3.20
99.25%
1.000
|
3.22
1.98-5.36
100%
1.001
|
1.92
1.05-3.27
98.70%
1.000
|
2.90
1.77-4.83
100%
1.001
|
Unadjusted
Multiple Imputation#
N=2713
|
Median HR
95% Credible Interval
Pd
Rhat
|
1
|
1.74
0.90-3.41
95.04%
1.064
|
1.70
1.01-2.98
97.91%
1.088
|
2.85
1.73-4.82
100%
1.079
|
1.80
1.00-3.25
97.95%
1.064
|
2.70
1.63-4.55
100%
1.072
|
Adjusted Complete cases uninformed prior N=1865
|
Median HR
95% Credible Interval
pd
Rhat
|
1
|
2.41
1.09-5.41
98.82%
1.001
|
2.48
1.31-5.05
99.88%
1.001
|
3.76
2.06-7.53
100%
1.000
|
2.51
1.23-5.40
99.60%
1.001
|
3.65
1.93-7.33
100%
1.001
|
Adjusted Multiple imputation* uninformed prior
N=2713
|
Median HR
95% Credible Interval
Pd
Rhat
|
1
|
1.63
0.85-3.14
92.81%
1.065
|
1.58
0.94-2.75
95.84%
1.088
|
2.40
1.46-4.03
99.99%
1.075
|
1.83
1.03-3.32
98.30%
1.069
|
2.52
1.52-4.22
100%
1.063
|
Adjusted Complete cases informative prior
N=1865
|
Median HR
95% Credible Interval
pd
Rhat
|
1
|
1.67
0.94-2.87
96.10%
1.001
|
1.52
0.96-2.36
96.58%
1.001
|
2.58
1.71-4.01
100%
1.001
|
1.86
1.14-3.08
99.28%
1.001
|
2.44
1.56-3.81
100%
1.001
|
Adjusted Multiple imputation* informative prior
N=2713
|
Median HR
95% Credible Interval
pd
Rhat
|
1
|
1.67
0.92-3.04
95.17%
1.046
|
1.58
0.97-2.57
96.96%
1.063
|
2.40
1.54-3.84
100%
1.056
|
1.83
1.08-3.14
98.79%
1.049
|
2.52
1.58-4.01
100%
1.040
|
Multivariate model is adjusted for age, gender, alcohol consumption, average daily steps, household income, highest level of education, smoking status.
# Multiple imputation for missing metabolic measures.
* Multiple imputation for missing metabolic measures and missing confounding variables.
Adjusted complete case informative prior
In a multivariable Bayesian model of complete cases and informative prior, the adjusted HR for CVD in MHOW was 1.52 (95% credible interval: 0.96–2.36) and in MHO was 1.86 (95% credible interval: 1.14–3.08). The HRs for CVD were similarly increased in the remaining metabolically unhealthy phenotypes with even higher HRs observed for those classified as overweight/obese and metabolically unhealthy; MUNW (HR = 1.67, 95% credible interval: 0.94–2.87), MUOW (HR = 2.58, 95% credible interval: 1.71–4.01), MUO (HR = 2.44, credible interval: 1.56–3.81). The reference group was MHNW for all the obesity phenotypes (Table 2). The informed prior adjusted effects remained similar when missing data were imputed. In all models the Rhat was < 1.05 and pd was > 95%.
The risks of CVD in different obesity phenotypes were not affected by gender. Kaplan-Meier survival curves showed the highest survival in the MHNW group, followed by MHOW and MHO groups. Survival in the MUNW was lower than in the MHOW and MHO groups but better than in the MUOW and MUO groups, which had the lowest survival probability (Fig. 2).
Mhow And Mho And Risk Of Cvd Mortality
Bayesian survival analysis
Adjusted multiple imputed data with uninformative prior
In a multivariable Bayesian model of imputed cases and uninformative prior, the adjusted HR for CVD mortality in MHOW was 5.65 (95% credible interval: 0.68–94.78) and in MHO was 17.51 (95% credible interval: 2.15-285.05). The remaining metabolically unhealthy phenotypes had increased HR for CVD mortality with even higher risks observed for those classified as overweight/obese and metabolically unhealthy; MUNW (HR = 16.70, 95% credible interval: 2.08-277.02), MUOW (HR = 7.95, 95% credible interval: 1.04-120.49), MUO (HR = 13.86, credible interval: 1.91-219.31). The reference group was MHNW for all the obesity phenotypes (Table 3a). In all models the Rhat was < 1.05 and pd was > 95%.
Table 3
a: Bayesian analysis results with median hazard ratio for risk of Cardiovascular disease mortality in older Australians from The Hunter Community Study stratified by BMI and metabolic health.
| | Normal Weight | Overweight | Obese |
| | Metabolically Healthy | Metabolically Unhealthy | Metabolically Healthy | Metabolically Unhealthy | Metabolically Healthy | Metabolically Unhealthy |
Number participants | Total = 2713 | 326 | 541 | 298 | 167 | 495 | 486 |
CVD related death | Total: n = 58 | 1 | 7 | 9 | 8 | 14 | 15 |
Adjusted Multiple Imputation* uninformative prior N = 2713 | Median HR 95% Credible Interval pd Rhat | 1 | 16.70 2.08-277.02 99.94% 1.001 | 5.65 0.68–97.78 96.24% 1.001 | 7.95 1.04-120.49 99.20% 1.001 | 17.51 2.15-285.05 99.96% 1.002 | 13.86 1.91-219.31 99.94% 1.002 |
Adjusted Multiple Imputation* informative prior N = 2713 | Median HR 95% Credible Interval pd Rhat | 1 | 4.64 1.80-12.18 99.93% 1.001 | 1.63 0.62–4.23 84.08% 1.001 | 2.38 1.01–5.68 97.71% 1.001 | 4.68 1.81–12.02 99.90% 1.002 | 4.01 1.70–4.22 99.94% 1.004 |
Table 3
b: Bayesian analysis results with median hazard ratio for risk of All-cause mortality in older Australians from The Hunter Community Study stratified by BMI and metabolic health.
| | Normal Weight | Overweight | Obese |
| | Metabolically Healthy | Metabolically Unhealthy | Metabolically Healthy | Metabolically Unhealthy | Metabolically Healthy | Metabolically Unhealthy |
Adjusted Multiple Imputation* uninformative prior | Median HR 95% Credible Interval pd Rhat | 1 | 1.11 0.85–3.14 64.50% 1.002 | 0.83 0.94–2.75 76.97% 1.003 | 0.94 1.46–4.03 60.75% 1.004 | 1.21 1.03–3.32 75.37% 1.008 | 1.30 1.52–4.22 86.77% 1.010 |
Adjusted Multiple Imputation* informative prior | Median HR 95% Credible Interval Pd Rhat | 1 | 1.27 0.73–2.20 80.10% 1.002 | 0.92 0.59–1.49 63.21% 1.002 | 1.05 0.68–1.68 58.59% 1.003 | 1.34 0.81–2.25 87.19% 1.007 | 1.45 0.95–2.28 95.30% 1.010 |
Multivariate model is adjusted for age, gender, alcohol consumption, average daily steps, household income, highest level of education, smoking status. |
* Multiple imputation for missing metabolic measures and missing confounding variables. |
Adjusted multiple imputed data with informative prior
In a multivariable Bayesian model of imputed cases and informative prior, the adjusted HR for CVD mortality in MHOW was 1.63 (95% credible interval: 0.62–4.23) and in MHO was 4.68 (95% credible interval: 1.81–12.02). The remaining metabolically unhealthy phenotypes had increased HR for CVD mortality with even higher risks observed for those classified as overweight/obese and metabolically unhealthy; MUNW (HR = 4.64, 95% credible interval: 1.80-12.18), MUOW (HR = 2.38, 95% credible interval: 1.01–5.68), MUO (HR = 4.01, credible interval: 1.70–4.22). The reference group was MHNW for all the obesity phenotypes (Table 3a). The uninformative prior adjusted effects on the imputed data were attenuated when the informative prior was applied to the adjusted imputed data, with increased HR for CVD mortality observed for MHO and all metabolically unhealthy groups. In all models the Rhat was < 1.05 and pd was > 95% except for MHOW.
Mhow And Mho And Risk Of All-cause Mortality
Bayesian survival analysis
Adjusted multiple imputed data with uninformative prior
In a multivariable Bayesian model of imputed cases and uninformed prior, the adjusted HR for all-cause mortality in MHOW was 0.83 (95% credible interval: 0.52–1.38) and in MHO was 1.21 (95% credible interval: 0.71–2.11). The remaining metabolically unhealthy phenotypes had similarly increased HR for all-cause mortality with the exception of those classified as MUOW. MUNW (HR = 1.11, 95% credible interval: 0.62–1.99), MUOW (HR = 0.94, 95% credible interval: 0.60–1.50), MUO (HR = 1.30, credible interval: 0.81–2.05). The reference group was MHNW for all the obesity phenotypes (Table 3b). In all models Rhat was < 1.05 and pd was < 95%.
Adjusted multiple imputed data with informative prior
In a multivariable Bayesian model of imputed cases and informative prior, the adjusted HR for all-cause mortality in MHOW was 0.92 (95% credible interval: 0.59–1.49) and in MHO was 1.34 (95% credible interval: 0.81–2.25). HRs for all-cause mortality was similarly increased in the remaining metabolically unhealthy phenotypes; MUNW (HR = 1.27, 95% credible interval: 0.73–2.20), MUOW (HR = 1.05, 95% credible interval: 0.68–1.68), MUO (HR = 1.45, credible interval: 0.95–2.28). The reference group was MHNW for all the obesity phenotypes (Table 3b).
The uninformed prior adjusted effects were slightly strengthened when an informative prior was applied to the adjusted imputed data, with an increased HR for all-cause mortality observed for MHO and all metabolically unhealthy groups. In all models the Rhat was < 1.05 and pd was < 95% in all groups except MUO.
Cox proportional hazard regression:
The frequentist approach conducted for all the three outcome categories showed similar results to the Bayesian analysis (Supplementary Tables 1 and 3).
Sensitivity analysis using competing risk analysis with death as a competing risk did not change the results for risk of CVD, CVD mortality or all-cause mortality (Supplementary Table 2).