BACKGROUND: Referral pattern is a potential confounding factor when waiting-list performance is reported across hospitals or periods. A common concern is the ability to accurately estimate proportions of patients undergoing surgery in the recommended time without considering emergency caseload. In this study, the relation between emergency referrals and the rate of elective admissions to hospital within the recommended time was estimated. DESIGN: A prospective cohort study. SETTING: An acute care hospital in Kingston, Ont. PATIENTS: Between 1994 and 1999, 1,173 consecutive patients accepted for elective vascular surgery. MAIN OUTCOME MEASURES: The proportion of patients who underwent surgery within the recommended time, and time to surgery. STUDY VARIABLES: The weekly number of emergency cases, enrolment periods, urgency and type of surgery. RESULTS: Overall, the proportion of patients who underwent surgery within recommended time was 0.45, (95% confidence interval [CI], 0.42-0.48). Adjusted for enrolment period, urgency and type of surgery, the estimated proportion was 0.57, (95% CI, 0.49-0.64). Compared with surgery for peripheral vascular disease, the odds of the procedure being done within the recommended time were 34% lower for aortic abdominal aneurysm repair and 41% lower for carotid endarterectomy. After adjustment for the case-mix and access attributes, the rate of elective admission within recommended time was on average 30% lower for weeks in which there were 1 to 2 emergency cases (rate ratio [RR] = 0.70, [95% CI, 0.53- 0.93]), and 39% lower for weeks with 3 or more emergency cases (RR = 0.61 [95% CI, 0.53-0.83]), relative to weeks with no emergency cases. CONCLUSIONS: When there is an increase in the number of emergency cases, a lower proportion of patients undergo elective surgery within the recommended time. Thus, when performance of surgical servces is evaluated, the probability of patients undergoing elective surgery on time should be adjusted relative to the number of emergency referrals.
BACKGROUND: Referral pattern is a potential confounding factor when waiting-list performance is reported across hospitals or periods. A common concern is the ability to accurately estimate proportions of patients undergoing surgery in the recommended time without considering emergency caseload. In this study, the relation between emergency referrals and the rate of elective admissions to hospital within the recommended time was estimated. DESIGN: A prospective cohort study. SETTING: An acute care hospital in Kingston, Ont. PATIENTS: Between 1994 and 1999, 1,173 consecutive patients accepted for elective vascular surgery. MAIN OUTCOME MEASURES: The proportion of patients who underwent surgery within the recommended time, and time to surgery. STUDY VARIABLES: The weekly number of emergency cases, enrolment periods, urgency and type of surgery. RESULTS: Overall, the proportion of patients who underwent surgery within recommended time was 0.45, (95% confidence interval [CI], 0.42-0.48). Adjusted for enrolment period, urgency and type of surgery, the estimated proportion was 0.57, (95% CI, 0.49-0.64). Compared with surgery for peripheral vascular disease, the odds of the procedure being done within the recommended time were 34% lower for aortic abdominal aneurysm repair and 41% lower for carotid endarterectomy. After adjustment for the case-mix and access attributes, the rate of elective admission within recommended time was on average 30% lower for weeks in which there were 1 to 2 emergency cases (rate ratio [RR] = 0.70, [95% CI, 0.53- 0.93]), and 39% lower for weeks with 3 or more emergency cases (RR = 0.61 [95% CI, 0.53-0.83]), relative to weeks with no emergency cases. CONCLUSIONS: When there is an increase in the number of emergency cases, a lower proportion of patients undergo elective surgery within the recommended time. Thus, when performance of surgical servces is evaluated, the probability of patients undergoing elective surgery on time should be adjusted relative to the number of emergency referrals.
Authors: Boris G Sobolev; Guy Fradet; Robert Hayden; Lisa Kuramoto; Adrian R Levy; Mark J FitzGerald Journal: BMC Health Serv Res Date: 2008-09-19 Impact factor: 2.655