Table 4. Multiple linear regression modeling association between log-transformed length of hospitalization and category of encephalitis cause, Ontario, Canada, 2002–2013.
| Variable* | No. (%) cases | Mean length of hospitalization, d (median) | Exponentiated β-coefficient (95% CI) | t | p value |
|---|---|---|---|---|---|
|
Intercept
|
8.9 (8.0–9.8) |
41.0 |
<0.0001 |
||
| Encephalitis cause | |||||
| Viral | 1,788 (27.7) | 27.36 (14) | |||
| Bacterial | 152 (2.4) | 27.45 (19) | 1.2 (1.0–1.4) | 1.5 | 0.128 |
| Amebic/parasitic/fungal | 92 (1.4) | 43.17 (18) | 1.3 (1.0–1.6) | 2.1 | 0.038 |
| Immune mediated | 657 (10.2) | 26.37 (14) | 1.1 (1.0–1.2) | 1.8 | 0.0804 |
| Other | 466 (7.2) | 24.33 (11) | 0.9 (0.8–1.0) | −1.7 | 0.0817 |
| Unknown | 3,299 (51.0) | 19.79 (9) | 0.8 (0.7–0.8) | −7.8 | <0.0001 |
| Unable to classify |
9 (0.1) |
38.78 (16) |
2.1 (1.0–4.1) |
2.0 |
0.0448 |
| Sex | |||||
| M | 3,417 (52.8) | 23.07 (11) | |||
| F |
3,046 (47.1) |
23.82 (12) |
1.0 (1.0–1.1) |
0.9 |
0.3634 |
| Age group, y | |||||
| <1 | 173 (2.7) | 22.83 (16) | 1.0 (0.8–1.1) | −0.6 | 0.55 |
| 1–4 | 377 (5.8) | 12.28 (5) | 0.6 (0.5–0.7) | −8.6 | <0.0001 |
| 5–19 | 915 (14.2) | 14.54 (6) | 0.8 (0.8–0.9) | −8.4 | <0.0001 |
| 20–44 | 1,486 (23.0) | 23.27 (9) | 0.8 (0.8–0.9) | −3.6 | 0.0003 |
| 45–64 | 1,823 (28.2) | 25.02 (13) | |||
|
>65 |
1,689 (26.1) |
29.19 (17) |
1.1 (1.1–1.2) |
3.7 |
0.0003 |
| Immune status | |||||
| Immunocompetent | 5,525 (85.5) | 21.29 (10) | |||
| Immunocompromised |
938 (14.5) |
35.99 (19) |
1.3 (1.2–1.4) |
5.6 |
<0.0001 |
| Co-morbidity level† | |||||
| None | 3718 (57.5) | 14.05 (8) | |||
| Low | 564 (8.7) | 26.08 (14) | 1.7 (1.6–1.9) | 11.0 | <0.0001 |
| Moderate | 911 (14.1) | 26.13 (15) | 1.8 (1.7–2.0) | 14.8 | <0.0001 |
| High | 575 (8.9) | 39.00 (25) | 2.8 (2.6–3.1) | 21.2 | <0.0001 |
| Very high | 520 (8.1) | 55.28 (36) | 4.0 (3.6–4.4) | 27.0 | <0.0001 |
| Missing data | 175 (2.7) | 14.05 (41) | 4.4 (3.7–5.2) | 17.9 | <0.0001 |
*Season, year, and patient Local Health Integration Network were not found to cause a significant change (>20%) in the parameter estimates for the exposure variable (category of encephalitis cause) and were thus excluded from the mode. †Case mix grouping plus comorbidity levels are based on cumulative cost impact of comorbidities on patient stay, where "none" represents no impact and "very high” represents the greatest impact.