Table 3 Risk analysis of different dietary patterns and all-cause mortality in individuals with dyslipidemia.

From: Dietary patterns and risk of all-cause mortality in individuals with dyslipidemia based on a prospective cohort in Guizhou China

Dietary patterns

Number of cases

Incidence density (No. per 1,000 person-years)

  

HR(95%CI)

   

Model 1

P

Model 2

P

Model 3

P

Equilibrium pattern

        

Low

155

7.346

1.000(ref)

 

1.000(ref)

 

1.000(ref)

 

Medium

123

5.772

0.783(0.618,0.992)

0.043

0.846(0.668,1.073)

0.167

0.857(0.675,1.087)

0.204

High

87

4.017

0.536(0.412,0.697)

< 0.001

0.676(0.520,0.880)

0.004

0.740(0.565,0.969)

0.029

Ptrend

  

0.618(0.503,0.759)

< 0.001

0.740(0.603,0.909)

0.004

0.794(0.643,0.981)

0.032

Oil and salt high loadings pattern

        

Low

113

5.327

1.000(ref)

 

1.000(ref)

 

1.000(ref)

 

Medium

133

6.240

1.168(0.909,1.500)

0.226

1.140(0.887,1.465)

0.306

1.173(0.910,1.512)

0.219

High

119

5.525

1.019(0.788,1.320)

0.883

0.951(0.735,1.232)

0.704

0.973(0.747,1.268)

0.839

Ptrend

  

0.992(0.841,1.170)

0.920

0.947(0.802,1.117)

0.515

0.957(0.809,1.134)

0.613

Grain-tuber and vegetables high loadings pattern

        

Low

138

6.520

1.000(ref)

 

1.000(ref)

 

1.000(ref)

 

Medium

104

4.835

0.730(0.565,0.942)

0.016

0.960(0.741,1.242)

0.755

0.982(0.755,1.277)

0.890

High

123

5.750

0.878(0.689,1.120)

0.295

1.302(1.012,1.675)

0.040

1.338(1.031,1.737)

0.029

Ptrend

  

0.931(0.804,1.079)

0.344

1.170(1.008,1.359)

0.039

1.189(1.019,1.388)

0.028

Aquatic products and poultry high loadings pattern

        

Low

139

6.533

1.000(ref)

 

1.000(ref)

 

1.000(ref)

 

Medium

135

6.322

0.967(0.763,1.225)

0.780

0.970(0.765,1.231)

0.804

0.988(0.778,1.255)

0.924

High

91

4.245

0.642(0.492,0.837)

0.001

0.782(0.597,1.023)

0.072

0.757(0.575,0.997)

0.047

Ptrend

  

0.556(0.396,0.782)

< 0.001

0.722(0.509,1.024)

0.068

0.694(0.486,0.990)

0.044

  1. Model 1: Adjusting none of variables; Model 2: Adjusting gender and age; Model 3: Adjusting total activity, LDL-C, TC, marital status, daily smoking, dangerous drinking, sleep, hypertension, diabetes, CVD, COPD, asthma, obesity, educational levels and taking lipid-lowering medications on the basis of model 2.
  2. The P value of trend tests was calculated by including the median score of each pattern tertile as a continuous variable in the models.
Back to article page