Skip to main content

Table 3 Propensity Scores model selected

From: Effectiveness of crisis resolution home treatment for the management of acute psychiatric crises in Southern Switzerland: a natural experiment based on geography

Covariates

Dependent variable: Intervention/control group membership

Coefficient (SEa)

95% confidence interval

Female gender

0.837**

(0.271;1.403)

(0.289)

Living alone

-0.623*

(-1.215;-0.032)

(0.302)

Compulsory admission

-0.721*

(-1.413;-0.029)

(0.353)

Disorders of personality and behaviour in adults (F6)

1.338**

(0.561;2.115)

(0.397)

Number of previous hospitalizations

-0.028

(-0.077;0.022)

(0.025)

Constant

-0.557*

(-1.082;-0.032)

(0.268)

Number of observations (n)

237

 

LR χ2 (5)

34.59***

 

Bayesian Information Criterion (BIC)

315.707

 

Predicted probabilitiesb

CPCc

Prd(CPC): mean (SDe)

0.663 (0.156)

 

Pr(CRHT): mean (SD)

0.337 (0.156)

 

CRHTf

Pr(CPC): mean (SD)

0.521 (0.188)

 

Pr(CRHT): mean (SD)

0.479 (0.188)

 
  1. aSE Standard Error
  2. b Predicted probabilities for each treatment level were computed and summarized conditional to each treatment level
  3. cCPC Cantonal Psychiatric Clinic
  4. dPr Probability
  5. eSD Standard Deviation
  6. fCRHT Crisis Resolution Home Treatment
  7. *p-value < 0.05, ** p-value < 0.01, *** p-value < 0.001