Responding to patient safety incidents: the "seven pillars".

BACKGROUND: Although acknowledged to be an ethical imperative for providers, disclosure following patient safety incidents remains the exception. The appropriate response to a patient safety incident and the disclosure of medical errors are neither easy nor obvious. An inadequate response to patient harm or an inappropriate disclosure may frustrate practitioners, dent their professional reputation, and alienate patients. METHODS: The authors have presented a descriptive study on the comprehensive process for responding to patient safety incidents, including the disclosure of medical errors adopted at a large, urban tertiary care centre in the United States. RESULTS: In the first two years post-implementation, the "seven pillars" process has led to more than 2,000 incident reports annually, prompted more than 100 investigations with root cause analysis, translated into close to 200 system improvements and served as the foundation of almost 106 disclosure conversations and 20 full disclosures of inappropriate or unreasonable care causing harm to patients. CONCLUSIONS: Adopting a policy of transparency represents a major shift in organisational focus and may take several years to implement. In our experience, the ability to rapidly learn from, respond to, and modify practices based on investigation to improve the safety and quality of patient care is grounded in transparency.

Changes in the distribution of body mass index of white US men, 1890-2000.

OBJECTIVE: The study aimed to describe changes in the distribution of body mass index (BMI) among white non-Hispanic US men aged 40-69 years throughout the 20th century. SUBJECTS: The subjects were 12 312 randomly drawn Union Army veterans examined between 1890 and 1900, and 4059 NHANES (National Health and Nutrition Examination Survey) participants examined between 1976 and 2000. METHOD: The study compared descriptive statistics of the age- and year-specific distributions of BMI. RESULTS: Between 1890 and 2000, median BMI of men aged 50-59 years increased by 5.7 kg/m(2) (25%), while the standard deviation almost doubled. In this age group, the current distribution of BMI is less right-skewed than in the earlier cohort. Obesity prevalence increased from 3.4% to 35%. In 1890-1894, median BMI declined with age, but by 2000 the age pattern had been reversed. The average annual growth rate of median BMI was lowest between 1900 and 1976 and has been rising to 0.5% per annum between 1988 and 2000. CONCLUSIONS: The increase in median BMI accounts for 75% of the rise in obesity prevalence between 1890 and 2000. The remainder must be attributed to changes in other features of the distribution, most notably the increased variance of BMI.