TABLE 3.
Stratified meta‐analysis of salt/sodium intake and risk of NAFLD
| Characteristics | Number of studies | Summary OR (95% CI) | p‐for‐difference | p‐for‐heterogeneity | I2 (%) |
|---|---|---|---|---|---|
| Overall | 7 | 1.60 (1.19–2.15) | ‐ | <.001 | 96.74 |
| Region | |||||
| Asia | 4 | 1.46 (1.09–1.95) | .597 | <.001 | 95.20 |
| America/Europe | 3 | 1.79 (0.88–3.62) | <.001 | 96.80 | |
| Study design | |||||
| Cohort/Case–control | 2 | 1.34 (1.15–1.56) | .285 | .421 | 0.00 |
| Cross‐sectional | 5 | 1.68 (1.14–2.47) | <.001 | 97.80 | |
| Risk of bias | |||||
| Low | 4 | 1.81 (1.24–2.65) | .031 | <.001 | 96.30 |
| High/Very high | 3 | 1.19 (1.13–1.24) | .383 | 0.00 | |
| Methods of NAFLD ascertainment | |||||
| Fibroscan/Ultrasonography | 4 | 1.19 (1.14–1.25) | .023 | .392 | 0.00 |
| Fatty liver index (FLI) | 3 | 2.02 (1.29–3.17) | <.001 | 96.70 | |
| Exposure | |||||
| Dietary instruments | 5 | 1.23 (1.15–1.32) | .005 | .31 | 16.20 |
| Urinary sodium excretion | 2 | 2.48 (1.52–4.06) | <.001 | 96.00 | |
Abbreviations: CI, confidence interval; NAFLD, nonalcoholic fatty liver disease; OR, odds ratio.
Bold values indicates the significant p‐values (p<0.05).