Skip to main content

Table 3 Logistic regression for the factors predicting response1 to treatment of depression during the first 6 months of follow-up in the Ostrobothnian Depression Study. Logistic regression was conducted for the total sample and separately for patients with or without heavy alcohol use (AUD)2.

From: Predictors of short-term response and the role of heavy alcohol use in treatment of depression

 

All patients (n = 150)

Patients without AUD (n = 100)

Patients with AUD (n = 50)

Nagelkerke R Square 0.250

Nagelkerke R Square = 0.257

Nagelkerke R Square = 0.582

OR (95% CI)

p

OR (95% CI)

p

OR (95%CI)

p

Age

1.02 (0.98–1.05)

0.333

1.00 (0.96–1.04)

0.933

1.06 (0.99–1.14)

0.093

Gender3

0.51 (0.23–1.12)

0.094

0.56 (0.18–1.74)

0.320

0.41 (0.09–1.94)

0.263

Baseline MADRS

1.10 (1.02–1.18)

0.008

1.12 (1.01–1.25)

0.027

1.11 (0.98–1.25)

0.093

Baseline AUDIT-C4

1.16 (1.03–1.31)

0.013

1.11 (0.89–1.38)

0.371

1.54 (1.04–2.27)

0.029

Dose of antidepressants5

0.99 (0.97–1.01)

0.172

0.98 (0.95-1.00)

0.066

1.01 (0.98–1.05)

0.493

Use of antipsychotics6

0.17 (0.07–0.44)

< 0.001

0.20 (0.06–0.67)

0.010

0.12 (0.02–0.60)

0.011

Any anxiety disorder

0.50 (0.22–1.14)

0.099

0.40 (0.15–1.08)

0.070

1.32 (0.30–5.87)

0.712

  1. 1 Positive response was defined as minimum of 50% reduction in Montgomery-Åsberg Depression Rating Scale (MADRS) score
  2. 2Groups formed according to alcohol use at baseline according to the Alcohol Use Disorders Identification Test, AUDIT: non-AUD group when baseline AUDIT ≤ 10 and AUD group when baseline AUDIT > 10
  3. 3Men compared to women
  4. 4Alcohol Use Disorders Identification Test, first three questions eliciting quantity-frequency
  5. 5Cumulative dose in fluoxetine equivalents. The OR < 1 indicates a greater likelihood for response in patients with a lower dose of antidepressant
  6. 6The OR < 1 indicates a greater likelihood for response in patients not on antipsychotic medication