Do antipsychotics prevent postoperative delirium? A systematic review and meta-analysis.

OBJECTIVE: To summarize the effect of antipsychotics for preventing postoperative delirium. DESIGN: We conducted a literature search using Medline, EMBASE, Cochrane Central Register of Controlled Trials, Cochrane Database of Systematic Reviews and clinicaltrials.gov. We included randomized controlled trials of adults undergoing surgery who were given antipsychotics to prevent postoperative delirium. Quality was assessed via the Cochrane risk of bias tool. Random-effects meta-analysis and meta-regression were conducted. Q-statistics and I(2) were used for assessment of heterogeneity. The main outcome was delirium incidence using validated definitions. RESULTS: A total of 1710 subjects were included, with a mean age ranging from 60.7 to 86.4 years. Antipsychotics reduced the incidence of postoperative delirium with the global effect-size estimate (weighted odds ratio) using the random effects model of 0.44 (95% confidence interval: 0.28-0.70; N = 6; Q-value: 16, p-value 0.0005; I(2) = 69%). Significant heterogeneity existed with the pooled global effect of delirium incidence; however, meta-regression allowed us to test both treatment-level and patient-level explanations for significant between-study variance. Baseline risk for delirium was found to be a significant contributor to study heterogeneity, and meta-regression suggested that antipsychotic type and dosage were two of the several treatment-level factors that also may have led to heterogeneity. Our analysis implied the presence of a breakeven baseline level of delirium risk below which preventive treatment with antipsychotics might prove ineffective. CONCLUSIONS: Within the limits of few randomized controlled trials, antipsychotics appeared to reduce the incidence of postoperative delirium in several surgical settings, predominantly orthopedic and for those at higher risk for delirium.

Sociotechnical challenges and progress in using social media for health.

Social media tools that connect patients, caregivers, and health providers offer great potential for helping people access health advice, receive and give social support, manage or cope with chronic conditions, and make day-to-day health decisions. These systems have seen widespread adoption, but often fail to support the goals as fully as designers and users would like. Through Ackerman's lens of the "sociotechnical gap" and computer supported cooperative work (CSCW) as a science of the artificial, we review contemporary sociotechnical challenges and progress for using social media to support health. These challenges include a tension between privacy and sharing, policy information credibility, accessibility, and tailoring in social spaces. Those studying, building, deploying, and using social media systems to further health goals will benefit from approaching this work by borrowing from Ackerman's framing of CSCW. In particular, this requires acknowledgment that technical systems will not fully meet our social goals, and then adopting design and educational approaches that are appropriate to fill this gap, building less-nuanced systems as partial solutions and tools for advancing our understanding, and by working with the CSCW research community to develop and pursue key lines of inquiry.

Application of language processing techniques to capture the use of nursing clinical terms from narrative statements: report of a pilot study.

OBJECTIVE: The authors piloted the use of the "General Architecture for Text Engineering" (GATE) program in an analysis of writings from the nursing literature to determine if this standard language processing technique could be used to capture the use of complex nursing terms. This work was undertaken as an initial step in evaluating if widely-available natural language processing methods could be applied to narrative nursing notes in a way that a nursing diagnosis could be identified and extracted from a narrative text. METHODS: For purposes of the pilot study, the complex nursing term "powerlessness", which is identified as a NANDA-I nursing diagnosis, was selected as the test case. A PubMed search was performed on the term "powerlessness" limited to articles in the nursing literature that contained abstracts, resulting in 232 articles published between 1981 to 2010 meeting the criteria. Three-sentence extracts from each abstract were analyzed by applying GATE to identify noun and adjective roots occurring in close proximity to the index word, and then identifying if these proximal words reflected the standardized defining characteristics, adjectives and qualifiers of the diagnostic term. RESULTS: The analysis resulted identification 2,174 unique terms. While a few terms coincided with the NANDA-I defining characteristics of "powerlessness", most of the established defining characteristics were not reflected in the use of the term. CONCLUSIONS: Machine language processing techniques are promising in identifying meanings and contextual use of words related to nursing concepts, but the use of such words in published papers does not represent definitions found in standard nursing nomenclature. Nursing writers use terms that are also understood outside the disciplinary domain, making standardization and coding particularly challenging. Future research in Nursing should apply the techniques described to clinical reports and to evaluate the match between clinical usage and standardized meanings.

How can we know whether short term trends in a hospital's HSMR are significant?

The Hospital Standardized Mortality Ratio (HSMR) has been chosen by CIHI as its primary mortality measure. The indirect standardization used in the calculation of HSMR does not allow for valid comparison between hospitals but it does invite the assessment of quarterly trends in hospital mortality. However, statistical methods for assessing HSMR trends are not well-developed. In 2007 one large hospital in our health authority had four consecutive quarters of apparently increasing HSMR. As a result, we needed to assess the significance of this trend which, if it were to continue into the next quarter, would lead to an HSMR that significantly exceeded 100. We explored four methods to assess statistical significance of time trends in HSMR data: the WINPEPI "Describe" module, the CUSUM representation of Observed-Expected differences, the Variable Life Adjusted Display (VLAD) plots with CUSUM overlays, and the Change Point Analysis using Monte Carlo simulation.

Fatal passenger vehicle crashes with at least 1 driver younger than 15 years: a fatality analysis reporting system study.

CONTEXT: A small number of fatalities continue to occur due to motor vehicle crashes on highways in which at least 1 passenger vehicle (automobile, van, or small truck) is driven by a child younger than 15 years. PURPOSE: The purpose of this study was to extend previous work suggesting that such crashes occur frequently in the Southern states and have relatively high rates in rural areas in the South and Great Plains. METHODS: This study utilizes data for the 5-year period 1999-2003 from the National Highway Transportation Safety Administration's online Fatality Analysis Reporting System. All cases were identified in which at least 1 conventional passenger vehicle in a fatal crash was being driven by a child younger than 15 years. FINDINGS: During the 5-year period, 350 fatal crashes occurred with at least 1 driver younger than 15 years involved. Twenty-one of these drivers were licensed (11) or driving with a learner's permit (10). A total of 987 individuals in 419 vehicles were involved in these crashes, and 402 deaths resulted (1.16 deaths/crash). These crashes occurred primarily in Texas, Florida, Arkansas, and Arizona, but the highest rates per 100,000 children were found in North and South Dakota and predominantly in a band of Intermountain and Plains states. There was a strong correlation between crash rates and several measures of rurality. CONCLUSIONS: Crashes involving young, largely unlicensed, drivers account for about 70 deaths yearly.

Cost of child lead poisoning to taxpayers in Mahoning County, Ohio.

Lead poisoning in children imposes both immediate and long-term financial burdens on taxpayers. The District Board of Health of Mahoning County, Ohio, quantified some of the direct costs to taxpayers of providing medical care and public health services to the 279 children diagnosed with lead poisoning in the county in 2002, using methods described by Katrina Korfmacher at the University of Rochester. The Board of Health also attempted to quantify the longer-term costs of special education and juvenile justice services attributable to lead exposure. The realization that lead poisoning costs local government on the order of 0.5 million dollars each year has mobilized community leaders in education and juvenile justice to demand more aggressive action against rental property owners who fail to remediate lead hazards.

A cohort study of family practice residency graduates in a predominantly rural state: initial practice site selection and trajectories of practice movement.

Limited information is available on long-term patterns of practice location choice among family physicians, although these patterns will significantly affect the quantity, quality, and continuity of health care provided in rural areas. From 1992 through 2000, five biennial surveys were administered to graduates of three family practice residency programs. Graduates were asked the location of their current and previous practice site locations, specific practice information, and global satisfaction with family practice. Identifiable practice sites were assigned a U.S. Department of Agriculture rural-urban continuum code based on the county in which the practice was located. The use of rural-urban continuum codes as a measure of rurality was believed to be particularly applicable to the small rural and frontier counties typical of the Great Plains region. For purposes of this study, practice locations were considered rural if they were in rural-urban continuum codes 7, 8, or 9. Of 593 graduates, 514 (87%) returned at least one questionnaire. Overall, about one-third of graduates had chosen their first practice sites in counties with rural-urban continuum codes of 7, 8, or 9. While most graduates remained in their initial practice site, almost half moved at least once. Those who left nonmetropolitan practices tended to move to less rural locations, though a few moved against this gradient. Moves away from the initial practice site occurred after a median of 3 years, and the likelihood of a first move decreased rapidly 5 to 6 years after graduation. Much remains to be learned about decisions to enter or leave practice in rural-urban continuum code 7, 8, and 9 counties.

Fatal passenger vehicle crashes 1999 to 2004 with drivers under age 15: the impact in Texas and other southern and southwestern states.

Texas has more fatal crashes involving unlicensed drivers under age 15 than does any other US state. Numbers and rates of such crashes are also above the national mean in many southern and Southwest states. Data on fatal passenger vehicle crashes from 1999 through 2004 were obtained from the online Fatality Analysis Reporting System (FARS). During the study period, there were 51 fatal passenger vehicle crashes in Texas in which drivers were under age 15. These crashes accounted for 12.3% of the US total. Nine southern states, including Texas, together accounted for 44% of all fatal crashes involving drivers under 15. Unlicensed crash rates per million inhabitants were higher in Texas than in other states with comparable populations but were much lower than those in other southern, southwest, and north central states. While Texas has recently improved its compliance with proposed graduated licensing models, state law explicitly prohibits police from stopping drivers based solely on age-related probable cause. This restriction may be a major barrier to effective detection and interdiction of under-age unlicensed driving. Because of the relatively high number of fatal crashes involving drivers under age 15 occurring in Texas, preventive efforts targeted to this state could modestly reduce the national burden of deaths due to very young unlicensed drivers. Expanding these efforts to other southern and southwest states could further reduce numbers and rates of such crashes. Expanded use of graduated licensing and increased public awareness are likely to prove effective tools in this public health effort.