Skip to main content

Table 4 Predicting disagreement or false reporting of diabetes

From: Comparison of self-reports and biomedical measurements on hypertension and diabetes among older adults in China

 

Sensitivity

Specificity

Correct Reporting

False Negative VS Correct Reporting

False Positive VS Correct Reporting

Diabetes

 Education (Illiterate = 0)

  Primary education

1.136

0.959

1.057

0.930

1.029

(1.009)

(−0.216)

(0.681)

(− 0.814)

(0.149)

  Secondary and above

1.436*

0.745

1.124

0.811+

1.310

(2.423)

(−1.337)

(1.177)

(−1.916)

(1.233)

  Hukou (Urban = 0)

0.616***

1.420*

1.186*

0.869

0.752+

(−4.147)

(2.050)

(2.118)

(−1.577)

(−1.669)

 Drink (None = 0)

   < 3 days a /month

0.942

0.545*

0.948

0.892

1.873*

(−0.261)

(−2.311)

(−0.376)

(−0.690)

(2.389)

  Once or 2 to 3 days a/week

0.918

0.589+

0.996

0.864

1.732*

(−0.373)

(−1.938)

(− 0.028)

(− 0.886)

(2.012)

  4 to 6 days a/week or daily

0.516**

0.787

0.872

1.113

1.319

(−3.224)

(−0.866)

(−1.139)

(0.818)

(1.000)

  Twice a day or above

0.433**

0.886

1.046

0.910

1.200

(−2.976)

(−0.346)

(0.294)

(−0.564)

(0.521)

  Number of cigarettes/day

0.940

1.122

1.040

0.976

0.897

(−1.292)

(1.577)

(1.268)

(−0.726)

(−1.488)

  Sex (female = 0)

1.005

1.337

1.117

0.930

0.752

(0.042)

(1.543)

(1.315)

(−0.780)

(−1.516)

 Age (40–49 = 0)

  50–59

1.542*

0.892

0.765*

1.371**

1.072

(2.394)

(−0.494)

(−2.505)

(2.659)

(0.304)

  60–69

1.819***

0.629*

0.614***

1.676***

1.456+

(3.423)

(−2.074)

(−4.707)

(4.477)

(1.686)

  70–79

1.430+

0.521*

0.521***

1.967***

1.745*

(1.801)

(−2.491)

(−5.446)

(5.102)

(2.133)

   > =80

1.105

0.440+

0.456***

2.228***

2.068

(0.308)

(−1.820)

(−3.986)

(3.735)

(1.617)

  Marriage (Not = 0)

1.277

0.907

1.059

0.918

1.089

(1.629)

(−0.431)

(0.600)

(−0.830)

(0.376)

N

1770

10,419

12,189

12,189

12,189

  1. Note: + p < 0.10; * p < 0.05; ** p < 0.01; *** p < 0.001; odds ratio estimates, logistic and multinomial logistic models, with z scores in parentheses