Fall prediction according to nurses' clinical judgment: differences between medical, surgical, and geriatric wards.

OBJECTIVES: To assess the value of nurses' clinical judgment (NCJ) in predicting hospital inpatient falls. DESIGN: Prospective multicenter study. SETTING: Six Belgian hospitals. PARTICIPANTS: Two thousand four hundred seventy participants (mean age 67.6 ± 18.3; female, 55.7%) on four surgical (n = 812, 32.9%), eight geriatric (n = 666, 27.0%), and four general medical wards (n = 992, 40.1%) were included upon admission. All participants were hospitalized for at least 48 hours. MEASUREMENTS: Within 24 hours after admission, nurses gave their judgment on the question "Do you think your patient is at high risk for falling?" Nurses were not trained in assessing fall risk. Falls were documented on a standardized incident report form. RESULTS: During hospitalization, 143 (5.8%) participants experienced one or more falls, accounting for 202 falls and corresponding to an overall rate of 7.9 falls per 1,000 patient days. NCJ of participant's risk of falling had high sensitivity (78-92%) with high negative predictive value (94-100%) but low positive predictive value (4-17%). Although false-negative rates were low (8-22%) for all departments and age groups, false-positive rates were high (55-74%), except on surgical and general medical wards and in participants younger than 75. CONCLUSION: This analysis, based on multicenter data and a large sample size, suggests that NCJ can be recommended on surgical and general medical wards and in individuals younger than 75, but on geriatric wards and in participants aged 75 and older, NCJ overestimates risk of falling and is thus not recommended because expensive comprehensive fall-prevention measures would be implemented in a large number of individuals who do not need it.

Content validity and internal consistency of the Dutch translation of the Safety Attitudes Questionnaire: an observational study.

BACKGROUND: Patient safety is fundamental to healthcare quality. Attention has recently focused on the patient safety culture of an organisation and its impact on patient outcomes. A strong safety climate appears to be an essential condition for safe patient care in the hospital. A number of instruments are used to measure this patient safety climate or culture. The Safety Attitudes Questionnaire is a validated, widely used instrument to investigate multiple dimensions of safety climate at the clinical level in a variety of inpatient and outpatient settings. OBJECTIVES: The purpose of this study is to explore the face- and content validity and the internal consistency of the Safety Attitudes Questionnaire in a large Belgian academic medical center. METHOD: The translation into Dutch was done by three researchers. A panel of fifteen Dutch speaking experts evaluated the translation and its content validity. Content validity was quantified by the content validity index (CVI) and a modified kappa index. Face validity was evaluated by two nurses and two physicians who assessed the Dutch version of the SAQ. A cross-sectional design was used to test internal consistency of the SAQ items by calculating Cronbach's alpha and corrected item-total correlations. RESULTS: Twenty-three of the 33 SAQ items showed excellent and seven items showed good content validity. One item had a fair kappa value (item 20) and two items had a low content validity index (items 15 and 16). The average CVI of the total scale was 0.83 and ranged from 0.55 to 0.97 for the six subscales. The face-validity was good with no fundamental remarks given. The SAQ's overall Cronbach's alpha was 0.9 and changed minimally when removing items. The item-total correlations ranged from 0.10 to 0.63, no single items were strongly correlated with the sum of the other items. CONCLUSION: We conclude that in this study the Dutch version of the Safety Attitudes Questionnaire showed acceptable to good psychometric properties. In line with previous evidence, this instrument seems to be an acceptable to adequate tool to evaluate the safety climate.