Background: Unnecessary admissions of acute cases have major impacts on hospital efficiency and organization. This study is aimed to identify percentage of unnecessary admissions from a hospital emergency department and the reasons why, as well as to quantify the unnecessary hospital days of care generated by these admissions. Methods: It has been analyzed the appropriateness of 622 admissions made in 2002, selected at random, all of the hospital stays generated by the inappropriate admissions and a representative sample of the appropriate admissions of a second-level hospital in Asturias. The review tool was the Appropriateness Evaluation Protocol. A descriptive analysis, a bivariate analysis and a multivariate logic regression analysis were made. Results: A total of 63 admissions (10.1%) were judged inappropriate. The main cause of inappropriateness were admissions for performing diagnostic tests and/or treatments, which could be carried out on an outpatient basis. These unnecessary admissions generated 78.2% of unnecessary stays, and the appropriate admissions generated 24.8% of unnecessary stays. Referrals to hospital emergency rooms from physicians outside of the hospital proper heightened the risk of unnecessary admissions (OR:4.50, 95% CI: 1.59-12.76), daytime admissions (OR: 13.97, 95% CI: 1.86-104.76) or evening admissions (OR: 7.70, 95% CI: 1.01-58.72), admissions to cardiology wards (OR: 3.93, 95% CI: 1.22-12.70) and neurology wards (OR: 5.86, 95% CI: 1.88-18.30), the experience of prior admissions having lowered the risk of unnecessary admission (OR: 0.34, 95% CI: 0.18-0.65). Conclusions: Unnecessary admissions generate three times more inappropriate stays than the necessary admissions. Hospital organization-related problems are the main cause of inappropriate admissions.