Table 3 Results of subgroups analyses of pooled RRs and 95% CIs.

From: Tooth loss is associated with increased risk of esophageal cancer: evidence from a meta-analysis with dose-response analysis

Subgroups

No. of studies

Heterogeneity

Effect model

Meta-analysis

I2(%)

p h

RR (95%CI)

p

Study design

Cohort

3

0

0.39

Fixed

1.02 (0.71–1.46)

0.91

Case-control

5

14.2

0.32

Fixed

1.48 (1.12–1.96)

<0.01

Cross-sectional

1

NA

NA

NA

1.45 (0.76–2.76)

0.26

Outcomes

ESCC

7

1.2

0.42

Fixed

1.21 (0.95–1.53)

0.13

Mixed

2

36.4

0.21

Fixed

1.66 (1.08–2.54)

0.02

Definition of reference group

None lost tooth

2

0

0.64

Fixed

1.01 (0.70–1.46)

0.94

No. of lost tooth

7

6.4

0.38

Fixed

1.46 (1.14–1.89)

<0.01

Effect estimation

Relative risk

1

NA

NA

NA

0.90 (0.48–1.68)

0.74

Hazard ratio

2

40.3

0.20

Fixed

1.09 (0.70–1.68)

0.71

Odds ratio

6

0

0.46

Fixed

1.47 (1.14–1.91)

<0.01

Adjustment

Smoking

8

0

0.46

Fixed

1.37 (1.10–1.70)

<0.01

Smoking and alcohol

7

10.4

0.35

Fixed

1.36 (1.08–1.71)

0.01

  1. No., number; ESCC, esophageal squamous cell carcinoma; RR, relative risk; CI, confidence interval.