(106) Association between antibiotics use and encephalopathies-related hospitalization among people living with dementia- a self-controlled case series study

Publication Number: 767

Background: Both encephalopathies (presented as seizure, delirium, or psychosis) and infections are common in people living with dementia. Therefore, there is urgent need to detangle the associations between infection, antibiotics and encephalopathies among dementia patients.

Objectives: To assess the encephalopathies related hospitalization rates among patients with dementia during antibiotics exposed compared to non-antibiotics exposed periods.

Methods: The Self-Controlled Case Series (SCCS) analysis was applied in this study to assess the encephalopathies related hospitalization rate during antibiotics exposed as compared to non-exposed periods, which was estimated by incidence rate ratio (IRR). Risk periods were defined as follows: 8–14 days pre- antibiotics, 7 days pre-antibiotics, the antibiotics exposed period, 1 to 7 days after antibiotics, 7 days post- antibiotics and 8–28 days post- antibiotics. The IRRs were estimated in each risk period. Data was retrieved from Clinical Data Analysis and Reporting System (CDARS) managed by the HK Hospital Authority, a statutory body that manages all public hospitals in HK. Any patients aged 65 or above who received a diagnosis of any types of dementia were identified using ICD9 codes. Patients who had a first hospitalization with a diagnosis of encephalopathies using ICD9 codes and at least one prescription of antibiotics were included. Encephalopathies include either seizure and myoclonus related (Type 1), or psychosis related (Type 2) hospitalization. Conditional Poisson regression model was used to estimate the IRR and the 95% confidence interval. Seasonal effect was adjusted in a 3-months band.
This study is supported by the Hong Kong Research Grants Council Early Career Scheme, No. 27105922.

Results: We found that for the seizure outcome induced by any antibiotics, all IRRs for all the risk periods are statistically significant. The IRR for 1-7 days pre-antibiotics is high (16.13, 95%CI: 13.96-18.65). The IRRs became lower during antibiotic exposure (7.34, 95%CI:6.35-8.49) and post-antibiotics periods (1.50, 95%CI:1.11-2.04 and 2.06, 95%CI:1.67- 2.53). For psychosis outcome induced by any antibiotics, all IRRs for all the risk periods are also statistically significant. The IRR for 1-7 days pre-antibiotics is high (14.51, 95%CI: 13.60-15.47). The IRRs became lower during antibiotic exposure (6.21, 95%CI: 5.80-6.65) and post-antibiotics periods (1.39, 95%CI:1.21-1.60 and 1.66, 95%CI:1.51- 1.83).

Conclusions: Since the risk of incident seizure and psychosis among dementia patients is high shortly before exposure to the antibiotics, the indications for which the antibiotics are prescribed may be associated with increased risk of seizure and psychosis.