Univariate analyses | Multivariate logistic regression | |||||
---|---|---|---|---|---|---|
No (n=1223) | Mild (n=493) | Moderate to severe (n=277) | p Value | OR (95% CI) | p Value | |
Sociodemographics | ||||||
Sex, male | 625 (51.1) | 235 (47.7)* | 159 (57.4)† | 0.04 | 1.45 (1.09 to 1.93) | 0.01 |
Age, years (OR per 10 years) | 50.8±12.1 | 49.8±12.4 | 49.3±12.8 | 0.12 | 0.86 (0.76 to 0.97) | 0.02 |
Clinical characteristics | ||||||
CAG repeats, number | 44.4±3.89 | 44.7±4.93 | 44.7±4.60 | 0.34 | ||
Duration of disease, year | 5.7 (3.3–9.6)* | 6.0 (3.5–9.8) | 6.3 (3.5–10.6)† | 0.05 | 1.01 (0.98 to 1.04) | 0.47 |
TMS, points | 34.9±19.7* | 34.6±19.6* | 38.4±20.7† | 0.02 | 1.01 (1.00 to 1.02) | 0.35 |
TFC, points | 9 (5–12)* | 9 (6–12)* | 7 (5–11)† | <0.001 | 0.97 (0.91 to 1.02) | 0.23 |
Psychiatric history | ||||||
Depression | 552 (45.1)* | 264 (53.5)† | 168 (60.6)† | <0.001 | 1.70 (1.26 to 2.28) | 0.001 |
OCBs | 94 (7.7)* | 65 (13.2)† | 43 (15.5)† | <0.001 | 1.48 (0.96 to 2.29) | 0.08 |
Psychosis | 64 (5.2)* | 41 (8.3)† | 33 (11.9)† | <0.001 | 1.70 (1.03 to 2.82) | 0.04 |
Suicide attempt | 83 (6.8)* | 32 (6.5)* | 39 (14.1)† | <0.001 | 1.71 (1.10 to 2.66) | 0.02 |
Psychotropic medication use | ||||||
Benzodiazepines | 206 (17.5) | 75 (15.5) | 57 (22.1) | 0.08 | ||
Antidepressants | 417 (35.7) | 168 (35.4) | 107 (40.1) | 0.38 | ||
Antipsychotics | 422 (35.8)* | 144 (29.9)† | 113 (43.0)‡ | 0.002 | 0.97 (0.71 to 1.32) | 0.85 |
Mood stabilisers/antiepileptics | 84 (6.9)* | 27 (5.5)* | 33 (12.0)† | 0.003 | 1.32 (0.83 to 2.10) | 0.24 |
Bold represents the values with p <0.05. *, † and ‡: Values in the same row with different superscript symbols are significantly different: p <0.05 (post hoc test). Univariate analyses: Data are presented as n (%), mean (±SD) or median (IQR) when appropriate.
p Values by χ^{2} tests for categorical data, by one-way between-groups analysis of variance for independent samples with normal distribution, or non-parametric Kruskal–Wallis test for continuous variables without normal distributions. Values in the same row with different superscript letters are significantly different (p<0.05).
Multivariate analyses: OR, 95% CI, and p value by binary logistic regression.
Logistic regression analysis (enter model) for ‘moderate to severe irritability/aggression’ (n=261) versus ‘no irritability/aggression’ (ie, reference: n=1174) (65 missings due to missing values (no start and stop date) for use of psychotropics). Covariates logistic regression analysis: All variables with p value <0.05 in the univariate analysis were entered (duration of disease, Total Motor Score (TMS), Total Functional Capacity (TFC), depression in past, obsessive/compulsive behaviours (OCBs) in past, psychosis in past, suicide attempt in past, use of antipsychotics, use of mood stabilisers/antiepileptics), while sex and age were forced into the multivariate model.
HD, Huntington's disease.