Observed-predicted length of stay for an acute psychiatric department, as an indicator of inpatient care inefficiencies. Retrospective case-series study.

Table 4 Multiple linear regression results¹. Optimum predicting function for logarithm of length of stay.

	Unstandardized Coefficients		Standardized Coefficients
	B	Std. Error	Beta	Sig
(Constant)	2.358	0.189		0.000
D1DIAG (Mood)	-0.085	0.086	-0.103	0.325
D2DIAG(Psychosis)	-0.179	0.102	-0.165	0.082
D3DIAG (Personality)	-0.238	0.092	-0.236	0.011
D4DIAG (Adjustment)²	-0.157	0.102	-0.141	0.125
Age	0.007	0.002	0.254	0.002
Gender (1: male, 2: female)	0.029	0.056	0.037	0.605
D1PR(Centre and Old Havana)	-0.017	0.082	-0.020	0.836
D2PR(Other municipalities in Havana)	-0.074	0.077	-0.093	0.339
D3PR (Havana Province)³	0.097	0.100	0.077	0.332
Marital Status	0.093	0.055	0.118	0.095
D0DR(None)	0.051	0.073	0.058	0.481
D1DR (Ant depressive drugs ≤ 150 mg)⁴	0.109	0.071	0.132	0.125
Electroconvulsive therapy	0.267	0.088	0.215	0.003
D0COM(None)	-0.030	0.070	-0.037	0.674
D1COM (Systemic diseases)⁵	0.010	0.076	0.012	0.891
Response to Treatment	0.230	0.064	0.246	0.000
Factor score 1⁶	0.069	0.031	0.174	0.025
Factor score 2	0.051	0.027	0.128	0.059
Factor score 3	0.080	0.026	0.203	0.003
Factor score 4	-0.019	0.026	-0.048	0.461
Factor score 5	-0.013	0.026	-0.032	0.626
Factor score 6	0.018	0.027	0.046	0.504
Factor score 7	0.0009	0.026	0.002	0.974
Factor score 8	0.029	0.027	0.074	0.282

1. Obtained with the first group of 185 histories after eliminating 2 outliers. 2. Dummies for diagnosis. Reference category is Rest of diagnosis. 3. Dummies for place of residence. Reference category is another province. 4. Dummies for drugs. Reference category is Antipsychotic agents 5. Dummies for co morbidities. Reference category is Other co morbidity 6. Factor scores for each of the 8 principal components. See Table 2. Model Summary: R = 0.611; R Square = 0.374

ISSN: 1472-6963