Skip to main content

Table 4 LOS variation attributable to mental illness, for each of the five chronic medical diseases

From: Hospital length of stay variation and comorbidity of mental illness: a retrospective study of five common chronic medical conditions

Chronic Disease (number of patients with MI)

Mean LOS in days in scenarios of 1. Without MI (SD) and 2. With MI (SD)

Type of MIa (number of patients with the type of MI)

Mean LOS (SD) in days for patients with this type of MI

Adjusted LOS variationb: % increase in mean LOS with MI (95% CI)

Difference in Bed-days’ usec

Per patient

in 5 years

Cancer (n= 100)

Scenario 1:

8.6 (9.3)

Scenario 2**: 12.8 (11.8)

F10–F19 Mental and behavioural disorders due to psychoactive substance use

(n = 48)

9.5 (7.4)

14.6% (–14.3%–53.2%)

NA

NA

All other subtypes

(n = 52)

15.8 (14.1)

97.0% (49.9%–159%)

7.8

404

COPD( n= 320)

Scenario 1:

4.3 (4.6)

Scenario 2*:

4.9 (4.6)

F40–F48 Neurotic, stress-related and somatoform disorders (n = 102)

6.2 (5.2)

64.2% (39.4%–93.4%)

2.4

248

All other subtypes

(n = 218)

4.3 (4.1)

8.2% (–4.7%–22.9%)

NA

NA

Diabetes (n = 90)

Scenario 1:

5.6 (9.3)

Scenario 2**:

8.7 (8.5)

All (n = 90)

8.7 (8.5)

68.0% (29.6%–118%)

3.5

317

IHD (n = 250)

Scenario 1:

3.2 (4.1)

Scenario 2**:

6.2 (9.6)

All (n = 250)

6.2 (9.6)

90.5% (71.5%–112%)

2.9

735

Stroke (n= 286)

Scenario 1:

7.9 (10.8)

Scenario 2**: 16.7 (26.7)

F10–F19 Mental and behavioural disorders due to psychoactive substance use

(n = 151)

12.5 (28.1)

36.3% (16.2%–59.9%)

3.3

505

All other subtypes

(n = 135)

21.3 (24.4)

109% (78%–146%)

11.1

1502

  1. MI Mental illness, SD Standard deviation, CI Confidence Interval, NA Not applicable
  2. *The variable differs significantly between scenario 1 and 2 at P < 0.05
  3. **The variable differs significantly between scenario 1 and 2 at P < 0.01
  4. aPatients with MI were divided into subtypes if more than 10 admissions were coded with the subtype (see Table 3), and if there was a significant difference in length of stay (LOS) between patients with this subtype of MI and patients with other subtypes of MI. Otherwise, results are presented for all patients with MI grouped together as the case for diabetes and IHD
  5. bPercentage increases in LOS, in relation to patients without MI, were estimated after adjusting for the variables: hospital, financial year of patient separation, presence of a Charlson Comorbidity, patient age, patient gender, and socio-economic status (SES) by geographical region
  6. cThese hypothetical difference in bed-days’ use is calculated with the assumption of same average LOS for patients of chronic medical conditions, between the scenarios of with and without a comorbidity of MI. Difference in bed-days use were calculated based on both MI subtype (if any) and total MI (if no subtype). NA refers to scenarios where hypothetical bed-days’ use was not calculated since no significant increase in adjusted LOS existed