[Pneumonia in the elderly: results of quality improvement program for a geriatric department in Lower Saxony 2006-2009].

BACKGROUND: Is the time to diagnosis and to start antibiotic therapy a major factor contributing to the outcome in geriatric patients?
METHODS: In 2006, a program for documentation and improvement for diagnostic and therapeutic procedures in patients with community-acquired pneumonia was introduced in German hospitals. The analyses were performed centrally by an independent board.
SETTING: Tertiary geriatric department with 70 beds for acute care and rehabilitation in a hospital, also including a department of neurology and neurological rehabilitation.
RESULTS: In Lower Saxony (LS), 81,853 patients were treated between 2006-2009 in our geriatric department (GD). In LS, 55.3% of the population was male, while 45.2% of the patients in the GD were male (p=0.063). Throughout an age of 79 years, the distribution was equal; however in the age groups 80-89 years (LS vs GD: 32.3 vs 47.6%) and >90 years (LS vs GD: 10.2 vs 15.5%, p<0.001) there were a higher proportion of male patients in the GD. The proportion of male nursing home patients was 46.8% vs 24.3%, hospital or rehabilitation unit 6.2% vs 40.5%, and status of confinement to bed was 47% vs 35.1% (LS vs GD, p<0.001). Delirium caused by pneumonia occurred in 24.4% vs 9.3% and a status of chronic delirium (dementia) was assessed by 75.6% vs. 90.7% of all cases (LS vs GD, p=0.021). The distribution about the risk classes 1/2/3 of the CRB-65 score was 14.9/76.9/8.2% in LS and 3.6/89.3/7.1% in the GD (p=0.011). The time to starting antibiotics (no therapy, <4, 4-8, and >8 h) was 2.2/83.0/7.6/7.2% in LS and 15.4/47.4/10.3/26.9% in GD patients (p<0.001). Overall mortality rates did not differ significantly (LS 14.6% vs GD 11.9%, p=0.53).
CONCLUSIONS: Patients in the GD were older and more functionally dependent. The distribution of the risk index CRB-65 shows that these patients were at higher risk, were more often cognitively impaired (not caused by pneumonia), and time to starting antibiotics was longer. However, none of these differences had an influence on total mortality. The results are limited by the number of patients, potential differences of the treatment groups, and the quality of data in general as a result of a quality improvement program.