Table 2.
Meta‐analysis findings for updated and new comparisons completed for the CSDD, GDS‐15, and MADRS.
| Tool, cut‐off | Study IDa (first author, year) | Total sample size (n) | Total depression (n) | Sensitivity | Specificity |
|---|---|---|---|---|---|
| CSDD, Best reported (n = 12) | |||||
| da Gloria Portugal et al., 2011 | 71 | 51 | 0.80 | 0.65 | |
| Huynh et al., 2022 | 46 | 10 | 0.70 | 0.92 | |
| Jeon et al., 2014 | 46 | 13 | 0.69 | 0.58 | |
| Knapskog et al., 2011 | 55 | 13 | 0.92 | 0.52 | |
| Korner et al., 2006 | 51 | 38 | 0.95 | 0.92 | |
| Leontjevas et al., 2012 | 101 | 18 | 0.94 | 0.49 | |
| Lim et al., 2012 | 121 | 59 | 0.87 | 0.98 | |
| Maixner et al., 1995 | 115 | 23 | 0.65 | 0.85 | |
| Mougias et al., 2017 | 136 | 25 | 0.88 | 0.79 | |
| Porta‐Etessam et al., 2011 | 1239 | 67 | 0.57 | 0.83 | |
| Vida et al., 1993 | 34 | 10 | 0.90 | 0.75 | |
| Wongpakaran et al., 2013 | 35 | 13 | 0.92 | 0.95 | |
| Pooled estimate: Sensitivity = 0.83 (95% CI = 0.74–0.90, I 2 = 71.71%, p < 0.001), Specificity = 0.81 (95% CI = 0.69–0.89, I 2 = 91.14%, p < 0.001) | |||||
| CSDD, ≥6 (n = 5) | |||||
| Knapskog et al., 2011 | 55 | 13 | 0.85 | 0.59 | |
| Leontjevas et al., 2012 | 101 | 18 | 1.00 | 0.43 | |
| Lim et al., 2012 | 121 | 59 | 0.91 | 0.96 | |
| Mougias et al., 2017 | 136 | 25 | 0.88 | 0.79 | |
| Vida et al., 1993 | 34 | 10 | 0.90 | 0.67 | |
| Pooled estimate: Sensitivity = 0.90 (95% CI = 0.82–0.95, I 2 = 0%, p = 0.89), Specificity = 0.74 (95% CI = 0.50–0.89, I 2 = 93.17%, p < 0.001) | |||||
| CSDD, ≥8 (n = 5) | |||||
| Huynh et al., 2022 | 46 | 10 | 0.90 | 0.69 | |
| Knapskog et al., 2011 | 55 | 13 | 0.62 | 0.74 | |
| Leontjevas et al., 2012 | 101 | 18 | 0.83 | 0.58 | |
| Lim et al., 2012 | 121 | 59 | 0.82 | 1.00 | |
| Vida et al., 1993 | 34 | 10 | 0.80 | 0.83 | |
| Pooled estimate: Sensitivity = 0.79 (95% CI = 0.69–0.87, I 2 = 0%, p = 0.44), Specificity = 0.81 (95% CI = 0.60–0.93, I 2 = 89.12%, p < 0.001) | |||||
| CSDD, ≥12 (n = 4) | |||||
| da Gloria Portugal et al., 2011 | 71 | 51 | 0.78 | 0.69 | |
| Huynh et al., 2022 | 46 | 10 | 0.70 | 0.89 | |
| Porta‐Etessam et al., 2011 | 1239 | 67 | 0.56 | 0.83 | |
| Vida et al., 1993 | 34 | 10 | 0.40 | 0.96 | |
| Pooled estimate: Sensitivity = 0.61 (95% CI = 0.42–0.77, I 2 = 66.10%, p = 0.03), Specificity = 0.83 (95% CI = 0.76–0.88, I 2 = 50.83%, p = 0.11) | |||||
| GDS‐15, Best reported (n = 3) | |||||
| Burke et al., 1991 | 72 | 10 | 0.60 | 0.63 | |
| Korner et al., 2006 | 47 | 36 | 0.81 | 0.73 | |
| Li et al., 2015 | 45 | 13 | 0.38 | 0.88 | |
| Pooled estimate: Sensitivity = 0.65 (95% CI = 0.40–0.83, I 2 = 58.47%), Specificity = 0.72 (95% CI = 0.55–0.85, I 2 = 52.54%) | |||||
| MADRS, Best reported (n = 3) | |||||
| da Gloria Portugal et al., 2011 | 71 | 51 | 0.75 | 0.75 | |
| Knapskog et al., 2011 | 55 | 13 | 0.85 | 0.67 | |
| Leontjevas et al., 2012 | 101 | 18 | 0.78 | 0.66 | |
| Pooled estimate: Sensitivity = 0.77 (95% CI = 0.67–0.85, I 2 = 0%), Specificity = 0.68 (95% CI = 0.60–0.75, I 2 = 0%) | |||||
Abbreviations: CI, confidence interval; CSDD, Cornell Scale for Depression in Dementia; GDS, Geriatric Depression Scale‐15 item; MADRS, Montgomery Asberg Depression Rating Scale; n, number of participants.
a
Complete references available for each study available in Supporting Information Material.