Length of stay in a hospital emergency room due to asthma and chronic obstructive pulmonary disease: Implications for air pollution studies

Citation
S. Perez-hoyos et al., Length of stay in a hospital emergency room due to asthma and chronic obstructive pulmonary disease: Implications for air pollution studies, EUR J EPID, 16(5), 2000, pp. 455-463
Citations number
20
Categorie Soggetti
Envirnomentale Medicine & Public Health","Medical Research General Topics
Journal title
EUROPEAN JOURNAL OF EPIDEMIOLOGY
ISSN journal
03932990 → ACNP
Volume
16
Issue
5
Year of publication
2000
Pages
455 - 463
Database
ISI
SICI code
0393-2990(200005)16:5<455:LOSIAH>2.0.ZU;2-1
Abstract
The aim of this paper is to model the length of stay following hospital eme rgency due to asthma and chronic obstructive pulmonary disease (COPD) as we ll as those factors relating to its duration. Clinical records of all those hospital emergencies for asthma and COPD in people aged over 14 at the Hos pital Clinico Universitario of Valencia (HCUV) during years 1993-1995 were reviewed. Data were analysed by entry and discharge time, sex, age and dest ination. Parametric survival models (exponential, Weibull and log-normal) w ere used for each diagnostic. Percentiles and relative percentiles were cal culated; 1369 asthma emergencies and 2734 cases COPD were observed. The mea n length of stay was 5.21 hours for asthma and 6.32 for COPD log-normal mod el was the best parametric model. Lengths of stay are greater for those hos pital admitted. Stays, in cases of asthma, were shorter during months of Au gust and November, with no differences for other variables. Also a shorter duration of stay for the month of August was found. Women are seen to stay for longer periods than men. Time modelling of emergency admission for asth ma and COPD permits its use in the determining of the dates of incidence fo r episodes of asthma and COPD based on hospital admissions data, as seeing a median of delay in emergency rooms greater than 6 hours. For studies on a ir pollution impact these results imply that all those admissions during th e early hours of the morning should be categorised as events of the previou s day.