Prescribing Associated with High-Risk Opioid Exposures Among Non-cancer Chronic Users of Opioid Analgesics: a Social Network Analysis.

BACKGROUND: The continued rise in fatalities from opioid analgesics despite a steady decline in the number of individual prescriptions directing ≥ 90 morphine milligram equivalents (MME)/day may be explained by patient exposures to redundant prescriptions from multiple prescribers. OBJECTIVES: We evaluated prescribers' specialty and social network characteristics associated with high-risk opioid exposures resulting from single-prescriber high-daily dose prescriptions or multi-prescriber discoordination. DESIGN: Retrospective cohort study. PARTICIPANTS: A cohort of prescribers with opioid analgesic prescription claims for non-cancer chronic opioid users in an Illinois Medicaid managed care program in 2015-2016. MAIN MEASURES: Per prescriber rates of single-prescriber high-daily-dose prescriptions or multi-prescriber discoordination. KEY RESULTS: For 2280 beneficiaries, 36,798 opioid prescription claims were submitted by 3532 prescribers. Compared to 3% of prescriptions (involving 6% of prescribers and 7% of beneficiaries) that directed ≥ 90 MME/day, discoordination accounted for a greater share of high-risk exposures-13% of prescriptions (involving 23% of prescribers and 24% of beneficiaries). The following specialties were at highest risk of discoordinated prescribing compared to internal medicine: dental (incident rate ratio (95% confidence interval) 5.9 (4.6, 7.5)), emergency medicine (4.7 (3.8, 5.8)), and surgical subspecialties (4.2 (3.0, 5.8)). Social network analysis identified 2 small interconnected prescriber communities of high-volume pain management specialists, and 3 sparsely connected groups of predominantly low-volume primary care or emergency medicine clinicians. Using multivariate models, we found that the sparsely connected sociometric positions were a risk factor for high-risk exposures. CONCLUSION: Low-volume prescribers in the social network's periphery were at greater risk of intended or discoordinated prescribing than interconnected high-volume prescribers. Interventions addressing discoordination among low-volume opioid prescribers in non-integrated practices should be a priority. Demands for enhanced functionality and integration of Prescription Drug Monitoring Programs or referrals to specialized multidisciplinary pain management centers are potential policy implications.

Predicting out of intensive care unit cardiopulmonary arrest or death using electronic medical record data.

BACKGROUND: Accurate, timely and automated identification of patients at high risk for severe clinical deterioration using readily available clinical information in the electronic medical record (EMR) could inform health systems to target scarce resources and save lives. METHODS: We identified 7,466 patients admitted to a large, public, urban academic hospital between May 2009 and March 2010. An automated clinical prediction model for out of intensive care unit (ICU) cardiopulmonary arrest and unexpected death was created in the derivation sample (50% randomly selected from total cohort) using multivariable logistic regression. The automated model was then validated in the remaining 50% from the total cohort (validation sample). The primary outcome was a composite of resuscitation events, and death (RED). RED included cardiopulmonary arrest, acute respiratory compromise and unexpected death. Predictors were measured using data from the previous 24 hours. Candidate variables included vital signs, laboratory data, physician orders, medications, floor assignment, and the Modified Early Warning Score (MEWS), among other treatment variables. RESULTS: RED rates were 1.2% of patient-days for the total cohort. Fourteen variables were independent predictors of RED and included age, oxygenation, diastolic blood pressure, arterial blood gas and laboratory values, emergent orders, and assignment to a high risk floor. The automated model had excellent discrimination (c-statistic=0.85) and calibration and was more sensitive (51.6% and 42.2%) and specific (94.3% and 91.3%) than the MEWS alone. The automated model predicted RED 15.9 hours before they occurred and earlier than Rapid Response Team (RRT) activation (5.7 hours prior to an event, p=0.003) CONCLUSION: An automated model harnessing EMR data offers great potential for identifying RED and was superior to both a prior risk model and the human judgment-driven RRT.

The association between community crime and childhood asthma prevalence in Chicago.

BACKGROUND: Little attention has been given to exposure to crime as a possible socioenvironmental contributor to variability in urban childhood asthma prevalence. OBJECTIVE: To determine the association of violent crime, property crime, and drug abuse violations with childhood asthma prevalence in Chicago. METHODS: In 2003-2005, the Chicago Initiative to Raise Asthma Health Equity conducted an asthma screening survey of children in grades K to 8 attending Chicago public and Catholic schools. Crime data were obtained from the Chicago Police Department. In addition to simple regression analysis, multilevel logistic regression analysis was performed to estimate the effect of criminal activity on neighborhood asthma variance. RESULTS: Of the surveys returned, 45,371 (93%) were geocoded into 247 neighborhoods. Neighborhoods were divided into quartile groups by mean asthma prevalence (9%, 12%, 17%, and 22%). Criminal activity (annual incidence per 100,000 people) was significantly higher (P < .001) in neighborhoods with a high asthma prevalence, especially drug abuse violations, which increased more than 6-fold (461 vs 2,921), and violent crimes, which increased more than 3-fold (448 vs 1,566). After adjusting for community race/ethnicity, only violent crime continued to be significantly associated with the neighborhood asthma prevalence (odds ratio, 1.27; 95% confidence interval, 1.04-1.55, P < .05). When considered alongside sociodemographic and individual characteristics, violence continued to contribute significantly (P < .05), explaining 15% of neighborhood variation in childhood asthma. CONCLUSIONS: Evidence suggests an association between violent crime and childhood asthma prevalence in Chicago. A deeper understanding of the mechanisms that underlie this association may lend insight into potential interventions to address urban asthma.

Ethnic disparities in asthma morbidity in Chicago.

RATIONALE: The role of ethnicity and socioeconomic status in explaining variations in asthma morbidity is unclear. OBJECTIVES: To describe the magnitude of ethnic disparities in asthma morbidity in Chicago and to determine whether differences in socioeconomic status explain these disparities. METHODS: We conducted a survey of 561 school-age children and 353 young adults with asthma and measured their self-reported ethnicity, socioeconomic status (using 11 variables), and asthma morbidity (symptom frequency, asthma-specific quality of life, and frequency of severe asthma exacerbations). MEASUREMENTS AND MAIN RESULTS: White children and adults had better asthma-specific quality of life and fewer severe asthma exacerbations compared to black and Hispanic children and adults. White children also had fewer days with asthma symptoms, but among adults there were no ethnic differences in the frequency of asthma symptoms. Socioeconomic status explained a large portion of the ethnic disparities in asthma quality of life but explained little of the disparities in other aspects of asthma morbidity. CONCLUSIONS: There are large disparities across ethnic groups in Chicago in asthma quality of life and in the frequency of severe exacerbations. Differences in socioeconomic status do not fully explain these disparities.

The protective effect of community factors on childhood asthma.

BACKGROUND: Asthma burden in the US is not evenly distributed. Although asthma prevalence varies widely across urban neighborhoods, little attention has been paid to the community as a key contributor. OBJECTIVE: To determine the effect of positive socio-environmental community factors on childhood asthma prevalence in Chicago. METHODS: From 2003 to 2005, an asthma screening survey was conducted among children attending Chicago Public/Catholic schools from kindergarten through eighth grade. One hundred five schools participated, yielding a stratified representation of 4 race-income groups. Positive community factors, such as social capital, economic potential, and community amenities, were assessed by using the Metro Chicago Information Center's Community Vitality Index. RESULTS: Of the surveys returned, 45,177 (92%) were geocoded into 287 neighborhoods. Neighborhoods were divided into quartile groups by asthma prevalence (mean, 8%, 12%, 17%, 25%). Community vitality (54% vs 44%; P < .0001) and economic potential (64% vs 38%; P < .0001) were significantly higher in neighborhoods with low asthma prevalence. Neighborhood interaction (36% vs 73%; P < .0001) and stability (40% vs 53%; P < .0001) were significantly higher in neighborhoods with high asthma prevalence. Overall, positive factors explained 21% of asthma variation. Childhood asthma increased as the black population increased in a community (P < .0001). Accordingly, race/ethnicity was controlled. In black neighborhoods, these factors remained significantly higher in neighborhoods with low asthma prevalence. When considered alongside socio-demographic/individual characteristics, overall community vitality as well as social capital continued to contribute significantly to asthma variation. CONCLUSION: Asthma prevalence in Chicago is strongly associated with socio-environmental factors thought to enrich a community. A deeper understanding of this impact may lend insight into interventions to reduce childhood asthma.

Geographic variability in childhood asthma prevalence in Chicago.

BACKGROUND: Childhood asthma prevalence has been shown to be higher in urban communities overall without an understanding of differences by neighborhood. OBJECTIVE: To characterize the geographic variability of childhood asthma prevalence among neighborhoods in Chicago. METHODS: Asthma screening was conducted among children attending 105 Chicago schools as part of the Chicago Initiative to Raise Asthma Health Equity. Additional child information included age, sex, race/ethnicity, and household members with asthma. Surveys were geocoded and linked with neighborhoods. Neighborhood information on race, education, and income was based on 2000 census data. Bivariate and multilevel analyses were performed. RESULTS: Of the 48,917 surveys, 41,255 (84.3%) were geocoded into 287 neighborhoods. Asthma prevalence among all children was 12.9%. Asthma rates varied among neighborhoods from 0% to 44% (interquartile range, 8% to 24%). Asthma prevalence (mean, SD, range) in predominantly black neighborhoods (19.9, +/-7, 4% to 44%) was higher than in predominantly white neighborhoods (11.4, +/-4.7, 2% to 30%) and predominantly Hispanic neighborhoods (12.1, +/-6.8, 0% to 29%). Although sex, age, household members with asthma, and neighborhood income significantly affected asthma prevalence, they did not explain the differences seen between neighborhoods. Race explained a significant proportion (about 80%) but not all of this variation. CONCLUSION: Childhood asthma prevalence varies widely by neighborhood within this urban environment. Adjacent areas in Chicago were identified with significantly different asthma prevalence. A better understanding of the effect of neighborhood characteristics may lend insight into potential interventions to reduce childhood asthma.

Parental language and asthma among urban Hispanic children.

BACKGROUND: Many Hispanics in the United States have limited English proficiency and prefer communicating in Spanish. Language barriers are known to adversely affect health care quality and outcomes. OBJECTIVE: We explored the relationship between parent language preference in a Hispanic population and the likelihood that a child with symptoms receives a diagnosis of asthma. METHODS: We conducted a school-based survey in 105 Chicago public and Catholic schools. Our sample included 14,177 Hispanic children 6 to 12 years of age with a parent who completed an asthma survey. Outcomes of diagnosed asthma and possible asthma (asthma symptoms without diagnosis) were assessed by using the Brief Pediatric Asthma Screen Plus instrument. RESULTS: Overall, 12.0% of children had diagnosed asthma, and 12.7% had possible asthma. Parents of children at risk who completed the survey in English reported higher rates of asthma diagnosis compared with parents who completed it in Spanish (55.2% vs 36.3%, P < .001). Predictors of asthma diagnosis were child sex, parental language preference, parental asthma status, and other household members with asthma. CONCLUSIONS: Parental language preference might be an important characteristic associated with childhood asthma diagnosis. Whether language itself is the key factor or the fact that language is a surrogate for other attributes of acculturation needs to be explored. CLINICAL IMPLICATIONS: Our findings suggest that estimates of asthma among Hispanic schoolchildren might be low because of underdiagnosis among children whose parents prefer communicating in Spanish.

A 6-item brief measure for assessing perceived control of asthma in culturally diverse patients.

BACKGROUND: A concise yet accurate measurement tool is needed for use in research and practice in asthma self-management perceptions across culturally diverse patient populations. OBJECTIVES: To evaluate the psychometric properties of the 11-item Perceived Control of Asthma Questionnaire (PCAQ) and to derive a brief, psychometrically sound, and culturally sensitive measure using item response theory. METHODS: The PCAQ was administered as one of a battery of measures to 375 adults with asthma as part of an ongoing larger project studying asthma disparities. Analyses of differential item functioning (DIF) were conducted to detect the effects of sex, race/ethnicity, and health literacy on psychometric properties. RESULTS: Forty-eight percent of the sample was non-Hispanic white and 44% was African American. The mean +/- SD age was 43.7 +/- 13.7 years. The derived 6-item version, with 5 DIF items not scored, correlated highly with its full version (r = 0.903; P < .001). CONCLUSIONS: The 6-item PCAQ short form has the potential to maintain scale integrity while reducing administration time and lessening survey fatigue in studies using multiple questionnaires. DIF analyses also enabled us to understand the unique aspects of perceived asthma control in demographic groups most affected by asthma.

Asthma burden in a citywide, diverse sample of elementary schoolchildren in Chicago.

OBJECTIVE: The purposes of this study are to describe and develop preliminary models of the burden of diagnosed asthma and symptoms of possible undiagnosed asthma in a large, citywide, ethnically and socioeconomically diverse sample of Chicago elementary schoolchildren. We hypothesized that considering possible asthma would give a more complete picture of race/ethnic disparities in pediatric asthma. METHODS: We studied 35,583 students aged 6 to 12 years attending Chicago Public and Archdiocese elementary schools for the Chicago Initiative to Raise Asthma Health Equity (CHIRAH) study. The full enrollments of 105 schools were surveyed for asthma and possible undiagnosed asthma by the Brief Pediatric Asthma Screen Plus (BPAS+) respiratory symptoms. The child had to be 6 to 12 years old, the valid age range for the BPAS+. Questionnaires included the BPAS+, basic demographic information, and household asthma information; they were sent home with each schoolchild for completion by the parent and returned to school for collection and scoring. RESULTS: Overall, 13.9% of students had diagnosed asthma. For children aged 6 to 12 years, rates of diagnosed asthma varied from 13.1% to 14.5%, whereas the rates of possible undiagnosed asthma varied from 14.8% to 10.9%. The rate of diagnosed asthma was 21.2% for African Americans, 9.7% for whites, 11.8% for Hispanics, with similar rates of possible undiagnosed asthma. By multinomial logistic regression, African Americans were more than twice as likely and Hispanics were 1.57 times more likely than whites to have diagnosed asthma at all school district income levels and controlling for other household members with asthma, type of school, age of the child, gender, and language preference. The odds of African Americans being diagnosed with asthma rather than having possible asthma were 76% higher and for Hispanics were 46% higher compared with whites, at all school district income levels and controlling for other household members with asthma, type of school, age of the child, gender, and language preference. CONCLUSIONS: Our study confirms national disparities in diagnosed asthma by race/ethnicity. Respiratory symptoms consistent with possible undiagnosed asthma increase the total potential burden of asthma overall to more than one-quarter of the school enrollees. Among students with respiratory symptoms, African Americans, Hispanics (controlling for language), and families where another person has asthma are more likely to have diagnosed rather than possible asthma. Improved knowledge about asthma, recognition of symptoms, and access to high-quality care are necessary to ascertain how much of the possible undiagnosed asthma represents additional cases of asthma requiring treatment.