Table 3.
Univariate meta‐regression of study and patient characteristics on prevalence of alcohol dependence in non‐selective patients in the UK hospital system.
| Alcohol dependence in non‐selective patients | Prevalence estimates (n) | Patients (n) | Prevalence | LCI | UCI | τ 2 | I 2 (%) | ||
| 23 | 992 784 | 10.25 | 7.06 | 13.96 | 0.070 | 99.75 | |||
| Prevalence estimates (n) | Patients (n) | Beta | LCI | UCI | τ 2 | I 2 (%) | P‐value a | Adjusted R 2 (%) | |
| Is the study a conference abstract? (binary) | 23 | 992 784 | −1.39 | −11.80 | 9.03 | 0.004 | 96.21 | 0.79 | 0 |
| Is the alcohol diagnosis diagnostic assessment robust? (binary) | 23 | 993 784 | −1.09 | −9.30 | 7.13 | 0.004 | 95.22 | 0.79 | 0 |
| Mean age of patients (continuous) | 7 | 7496 | −0.80 | −1.36 | −0.24 | 0.000 | 0.00 | 0.02 | 100.00 |
| Proportion female (continuous) | 10 | 8588 | 0.03 | −0.29 | 3.50 | 0.002 | 46.80 | 0.83 | 0 |
| Total Newcastle–Ottawa scale quality score (continuous) | 23 | 992 784 | 0.31 | −2.14 | 2.77 | 0.004 | 95.00 | 0.79 | 0 |
| Year study conducted in (continuous) | 16 | 986 504 | 0.45 | −1.10 | 2.01 | 0.004 | 94.59 | 0.54 | 0 |
| Type of setting patients admitted to (categorical) | 23 | 992 784 | – | – | – | 0.001 | 64.09 | 0.009 | 60.84 |
| Nation study conducted in (categorical) | 23 | 992 784 | – | – | – | 0.004 | 95.38 | 0.40 | 2.01 |
a
Result from t‐test where the null hypothesis was no linear relationship between prevalence and each explanatory variable. LCI = lower confidence interval; UCI = upper confidence interval; τ2 = between‐study variance; adjusted R 2 = percentage of variation in prevalence explained by a particular covariate