Prevalence and Characteristics of Diagnostic Error in Pediatric Critical Care: A Multicenter Study.

OBJECTIVES: Effective interventions to prevent diagnostic error among critically ill children should be informed by diagnostic error prevalence and etiologies. We aimed to determine the prevalence and characteristics of diagnostic errors and identify factors associated with error in patients admitted to the PICU. DESIGN: Multicenter retrospective cohort study using structured medical record review by trained clinicians using the Revised Safer Dx instrument to identify diagnostic error (defined as missed opportunities in diagnosis). Cases with potential errors were further reviewed by four pediatric intensivists who made final consensus determinations of diagnostic error occurrence. Demographic, clinical, clinician, and encounter data were also collected. SETTING: Four academic tertiary-referral PICUs. PATIENTS: Eight hundred eighty-two randomly selected patients 0-18 years old who were nonelectively admitted to participating PICUs. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: Of 882 patient admissions, 13 (1.5%) had a diagnostic error up to 7 days after PICU admission. Infections (46%) and respiratory conditions (23%) were the most common missed diagnoses. One diagnostic error caused harm with a prolonged hospital stay. Common missed diagnostic opportunities included failure to consider the diagnosis despite a suggestive history (69%) and failure to broaden diagnostic testing (69%). Unadjusted analysis identified more diagnostic errors in patients with atypical presentations (23.1% vs 3.6%, p = 0.011), neurologic chief complaints (46.2% vs 18.8%, p = 0.024), admitting intensivists greater than or equal to 45 years old (92.3% vs 65.1%, p = 0.042), admitting intensivists with more service weeks/year (mean 12.8 vs 10.9 wk, p = 0.031), and diagnostic uncertainty on admission (77% vs 25.1%, p < 0.001). Generalized linear mixed models determined that atypical presentation (odds ratio [OR] 4.58; 95% CI, 0.94-17.1) and diagnostic uncertainty on admission (OR 9.67; 95% CI, 2.86-44.0) were significantly associated with diagnostic error. CONCLUSIONS: Among critically ill children, 1.5% had a diagnostic error up to 7 days after PICU admission. Diagnostic errors were associated with atypical presentations and diagnostic uncertainty on admission, suggesting possible targets for intervention.

Impact of subspecialty consultations on diagnosis in the pediatric intensive care unit.

OBJECTIVES: Intensivists and subspecialists often collaborate in diagnosing patients in the pediatric intensive care unit (PICU). Our objectives were to characterize critically ill children for whom subspecialty consultations were requested, describe consultation characteristics, and determine consultations' impact on PICU diagnosis. METHODS: We performed a retrospective study using chart review in a single tertiary referral PICU including children admitted for acute illness. We collected data on patients with and without subspecialty consultations within the first three days of PICU admission and determined changes in PICU clinicians' diagnostic evaluation or treatment after consultations. RESULTS: PICU clinicians requested 152 subspecialty consultations for 87 of 101 (86%) patients. Consultations were requested equally for assistance in diagnosis (65%) and treatment (66%). Eighteen of 87 (21%) patients with consultations had a change in diagnosis from PICU admission to discharge, 11 (61%) attributed to subspecialty input. Thirty-nine (45%) patients with consultations had additional imaging and/or laboratory testing and 48 (55%) had medication changes and/or a procedure performed immediately after consultation. CONCLUSIONS: Subspecialty consultations were requested during a majority of PICU admissions. Consultations can influence the diagnosis and treatment of critically ill children. Future research should investigate PICU interdisciplinary collaborations, which are essential for teamwork in diagnosis.

Comparative effectiveness of five fecal immunochemical tests using colonoscopy as the gold standard: study protocol.

BACKGROUND: There are nearly 50,000 colorectal cancer (CRC) deaths in the United States each year. CRC is curable if detected in its early stages. Fecal immunochemical tests (FITs) can detect precursor lesions and many can be analyzed at the point-of-care (POC) in physician offices. However, there are few data to guide test selection. Broader use of FITs could make CRC screening more accessible, especially in resource-poor settings. METHODS: A total of 3600 racially and ethnically diverse individuals aged 50 to 85 years having either a screening or surveillance colonoscopy will be recruited. Each participant will complete five FITs on a single stool sample. Test characteristics for each FIT for advanced colorectal neoplasia (ACN) will be calculated using colonoscopy as the gold standard. RESULTS: We have complete data from a total of 2990 individuals. Thirty percent are Latino and 5.3% are black/African American. We will present full results once the study is completed. CONCLUSIONS: Our focus in this study is how well FITs detect ACN, using colonoscopy as the gold standard. Four of the five FITs being used are POC tests. Although FITs have been shown to have acceptable performance, there is little data to guide which ones have the best test characteristics and colonoscopy is the main CRC screening test used in the United States. Use of FITs will allow broader segments of the population to access CRC screening because these tests require no preparation, are inexpensive, and can be collected in the privacy of one's home. Increasing CRC screening uptake will reduce the burden of advanced adenomas and colorectal cancer.

DMPK mRNA Expression in Human Brain Tissue Throughout the Lifespan.

OBJECTIVE: Myotonic dystrophy is a multisystem disorder caused by a trinucleotide repeat expansion on the myotonic dystrophy protein kinase (DMPK) gene. To determine whether wildtype DMPK expression patterns vary as a function of age, we analyzed DMPK expression in the brain from 99 donors ranging from 5 postconceptional weeks to 80 years old. METHODS: We used the BrainSpan messenger RNA sequencing and the Yale Microarray data sets, which included brain tissue samples from 42 and 57 donors, respectively. Collectively, donors ranged in age from 5 postconceptional weeks to 80 years old. DMPK expression was normalized for each donor across regions available in both data sets. Restricted cubic spline linear regression models were used to analyze the effects of log-transformed age and sex on normalized DMPK expression data. RESULTS: Age was a statistically significant predictor of normalized DMPK expression pattern in the human brain in the BrainSpan (p < 0.005) and Yale data sets (p < 0.005). Sex was not a significant predictor. Across both data sets, normalized wildtype DMPK expression steadily increases during fetal development, peaks around birth, and then declines to reach a nadir around age 10. CONCLUSIONS: Peak expression of DMPK coincides with a time of dynamic brain development. Abnormal brain DMPK expression due to myotonic dystrophy may have implications for early brain development.

Dissemination of a telehealth cardiovascular risk service: The CVRS live protocol.

BACKGROUND: Medical clinics are increasingly hiring clinical pharmacists to improve management of cardiovascular disease (CVD). However, the limited number of clinical pharmacists employed in a clinic may not impact the large number of complex patients needing the services. We have developed a remote telehealth service provided by clinical pharmacists to complement CVD services provided by on-site clinical pharmacists and aid sites without a clinical pharmacist. This cardiovascular risk service (CVRS) has been studied in two NIH-funded trials, however, we identified barriers to optimal intervention implementation. The purpose of this study is to examine how to implement the CVRS into medical offices and see if the intervention will be sustained. METHODS: This is a 5-year, pragmatic, cluster-randomized clinical trial in 13 primary care clinics across the US. We randomized clinics to receive CVRS or usual care and will enroll 325 patient subjects and 288 key stakeholder subjects. We have obtained access to the electronic medical records (EMRs) of all study clinics to recruit subjects and provide the pharmacist intervention. The intervention is staggered so that after 12 months, the usual care sites will receive the intervention for 12 months. Follow-up will be accomplished though medical record abstraction at baseline, 12 months, 24 months, and 36 months. CONCLUSIONS: This study will enroll subjects through 2021 and results will be available in 2024. This study will provide unique information on how the CVRS provided by remote clinical pharmacists can be effectively implemented in medical offices, many of which already employ on-site clinical pharmacists. CLINICAL TRIAL REGISTRATION INFORMATION: NCT03660631: http://clinicaltrials.gov/ct2/show/NCT03660631.

Factors Associated With Diagnostic Error on Admission to a PICU: A Pilot Study.

OBJECTIVES: Diagnostic errors can harm critically ill children. However, we know little about their prevalence in PICUs and factors associated with error. The objective of this pilot study was to determine feasibility of record review to identify patient, provider, and work system factors associated with diagnostic errors during the first 12 hours after PICU admission. DESIGN: Pilot retrospective cohort study with structured record review using a structured tool (Safer Dx instrument) to identify diagnostic error. SETTING: Academic tertiary referral PICU. PATIENTS: Patients 0-17 years old admitted nonelectively to the PICU. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: Four of 50 patients (8%) had diagnostic errors in the first 12 hours after admission. The Safer Dx instrument helped identify delayed diagnoses of chronic ear infection, increased intracranial pressure (two cases), and Bartonella encephalitis. We calculated that 610 PICU admissions are needed to achieve 80% power (α = 0.05) to detect significant associations with error. CONCLUSIONS: Our pilot study found four patients with diagnostic error out of 50 children admitted nonelectively to a PICU. Retrospective record review using a structured tool to identify diagnostic errors is feasible in this population. Pilot data are being used to inform a larger and more definitive multicenter study.

Fourth Ventricle Enlargement in Chiari Malformation Type I.

OBJECTIVE: How Chiari malformation type I (CM-I) affects posterior fossa brain structures and produces various symptoms remains unclear. The fourth ventricle is surrounded by critical structures required for normal function. The foramen of Magendie can be obstructed in CM-I; therefore, fourth ventricle changes may occur. To test this hypothesis, we assessed fourth ventricle volume in CM-I compared with healthy controls. METHODS: Using our database from 2007-2016, we studied 72 patients with CM-I and 30 age-matched healthy control subjects. Fourth and lateral ventricle volumes and posterior fossa volumes (PFV) were assessed and correlated with clinical signs and symptoms. Statistical analysis was performed. RESULTS: Patients with CM-I had larger fourth ventricle volumes compared with control subjects (1.31 vs. 0.95 mL; P = 0.012). There were no differences in lateral ventricle volume or PFV. CM-I fourth ventricle volume was associated with tonsillar descent (P = 0.030). CM-I fourth ventricle volume variance was larger than healthy controls (F71,29 = 8.33; P < 0.0001). Patients with CM-I with severe signs and symptoms had a significantly larger fourth ventricle than patients with CM-I with mild signs and symptoms (1.565 vs. 1.015 mL; P = 0.0002). CONCLUSIONS: The fourth ventricle can be enlarged in CM-I independent of lateral ventricle size and is associated with greater tonsillar descent. Most importantly, fourth ventricle enlargement was associated with a worse clinical and radiographic presentation independent of PFV. Fourth ventricle enlargement can affect critical structures and may be a mechanism contributing to symptoms unexplained by tonsil descent. Fourth ventricle enlargement is a useful adjunct in assessing CM-I.

Practical and statistical challenges in driving research.

Driving is an integral aspect of many modern societies, and motor vehicle safety is an important public health issue. With advances in sensor technology, more and more driving data are being collected by researchers, insurers, and automobile companies, which has increased the need and opportunities for statisticians to be involved in driving research. This report discusses several practical and statistical challenges in driver-level studies, including the process of defining meaningful driving metrics, issues related to "Big Data" aspects of driving research, and the principle of reproducible research.

Black Clouds vs Random Variation in Hospital Admissions.

BACKGROUND AND OBJECTIVES: Physicians often accuse their peers of being "black clouds" if they repeatedly have more than the average number of hospital admissions while on call. Our purpose was to determine whether the black-cloud phenomenon is real or explainable by random variation. METHODS: We analyzed hospital admissions to the University of Iowa family medicine service from July 1, 2010 to June 30, 2015. Analyses were stratified by peer group (eg, night shift attending physicians, day shift senior residents). We analyzed admission numbers to find evidence of black-cloud physicians (those with significantly more admissions than their peers) and white-cloud physicians (those with significantly fewer admissions). The statistical significance of whether there were actual differences across physicians was tested with mixed-effects negative binomial regression. RESULTS: The 5-year study included 96 physicians and 6,194 admissions. The number of daytime admissions ranged from 0 to 10 (mean 2.17, SD 1.63). Night admissions ranged from 0 to 11 (mean 1.23, SD 1.22). Admissions increased from 1,016 in the first year to 1,523 in the fifth year. We found 18 white-cloud and 16 black-cloud physicians in simple regression models that did not control for this upward trend. After including study year and other potential confounding variables in the regression models, there were no significant associations between physicians and admission numbers and therefore no true black or white clouds. CONCLUSIONS: In this study, apparent black-cloud and white-cloud physicians could be explained by random variation in hospital admissions. However, this randomness incorporated a wide range in workload among physicians, with potential impact on resident education at the low end and patient safety at the high end.

Effect of Trinucleotide Repeats in the Huntington's Gene on Intelligence.

BACKGROUND: Huntington's Disease (HD) is caused by an abnormality in the HTT gene. This gene includes trinucleotide repeats ranging from 10 to 35, and when expanded beyond 39, causes HD. We previously reported that CAG repeats in the normal range had a direct and beneficial effect on brain development with higher repeats being associated with higher cognitive function. The current study now expands this line of inquiry to evaluate the effects of CAG repeat throughout the entire spectrum of repeats from 15 to 58. METHODS: We evaluated brain function in children ages 6-18 years old. DNA samples were processed to quantify the number of CAG repeats within HTT. Linear regression was used to determine if number of CAG repeats predicted measures of brain function. FINDINGS: The number of repeats in HTT, had a non-linear effect on a measure of general intelligence with an inverted U shape pattern. Increasing repeat length was associated with higher GAI scores up until roughly 40-41 repeats. After this peak, increasing repeat length was associated with declining GAI scores. INTERPRETATION: HTT may confer an advantage or a disadvantage depending upon the repeat length, playing a key role in the determination of intelligence, or causing a uniquely human brain disease.

Sigmoidal mixed models for longitudinal data.

Linear mixed models are widely used to analyze longitudinal cognitive data. Often, however, the trajectory of cognitive function is nonlinear. For example, some participants may experience cognitive decline that accelerates as death approaches. Polynomial regression and piecewise linear models are common approaches used to characterize nonlinear trajectories, although both have assumptions that may not correspond with the actual trajectories. An alternative is to use a flexible sigmoidal mixed model based on the logistic family of curves. We describe a general class of such a model, which has up to five parameters, representing (1) final level, (2) rate of decline, (3) midpoint of decline, (4) initial level before decline, and (5) asymmetry. Focusing on a four-parameter symmetric sub-class of the model, with random effects on two of the parameters, we demonstrate that a likelihood approach to fitting this model produces accurate estimates of mean levels across time, even in the case of model misspecification. We also illustrate the method on deceased participants who had completed at least 5 years of annual cognitive testing and annual assessment of body mass. We show that departures from a stable body can modify the trajectory curves and anticipate cognitive decline.

Reducing Accelerometer Data from Instrumented Vehicles.

In on-road driving behavior studies, vehicle acceleration is sampled at high frequencies and then reduced to meaningful metrics over short driving segments. We examined road test data from 65 subjects driving over a common route, as well as driving in naturalistic situations using their own vehicle. We isolated 24-second segments, then reduced the accelerometer data via two methods: 1) standard deviation (SD) within a segment, and 2) re-centering parameter from a time series model previously developed for driving simulator data. We analyzed the data via random effects models to ascertain the intraclass correlations (ICC's) of the metrics. With and without adjusting for speed, the ICC of SD within a segment tended to be much greater than the ICC of the re-centering parameter for the segment (range: 0-30% vs. 0-1%). Also, ICC's from the naturalistic driving data tended to be greater than the fixed-route data (range: 0-27% vs. 0-9%), which could reflect individuals exhibiting their more usual driving behavior in naturalistic environments. Findings illustrate the challenges of identifying meaningful driving metrics and comparing these across different epochs, road segments and research platforms.

Intradural pathology and pathophysiology associated with Chiari I malformation in children and adults with and without syringomyelia.

OBJECTIVE The pathophysiology underlying tonsillar herniation and CSF obstruction in Chiari malformation Type I (CM-I) is unclear, and the cause of CM-I-associated syringomyelia is not well understood. A better understanding of this pathophysiology is important for an improved treatment strategy. Therefore, the authors sought to identify, characterize, and examine the intradural pathology and CSF flow pathophysiology in the posterior fossa and at the level of the foramen magnum that occurs in the setting of CM-I. They determined the incidence of these intradural findings and assessed differences across age, with the degree of tonsillar herniation, and in the presence and absence of syringomyelia. METHODS A prospective database initiated in March 2003 recorded all intraoperative findings during surgical treatment of children and adults with CM-I with or without syringomyelia. A total of 389 surgeries for CM-I were performed in 379 patients between March 2003 and June 2016. A total of 109 surgeries were performed in 109 patients with CM-I (without osseoligamentous abnormalities) in whom both a posterior fossa extradural and intradural decompression with duraplasty was performed (first-time intradural procedures). Using a surgical microscope, intradural pathology and obstruction of CSF channels were identified and assessed. Student t-tests and Fisher's exact tests compared groups in a series of univariate analyses, followed by multivariate logistic regression. RESULTS The following intradural pathological entities were observed (prevalence noted in parentheses). These include those that did not obstruct CSF flow channels: opacified arachnoid (33.0%), thickened arachnoid (3.7%), ischemic and gliotic tonsils (40.4%), tonsillar cysts (0.9%), and inferior descent of the fourth ventricle and cervicomedullary junction (CMJ) (78.0%). The following intradural pathological entities were observed to obstruct CSF flow channels: medialized tonsils (100%), tonsil overlying and obstructing the foramen of Magendie (21.1%), intertonsillar and tonsil to CMJ arachnoid adhesions (85.3%), vermian posterior inferior cerebellar artery branches obstructing the foramen of Magendie (43.1%), and arachnoid veils or webs obstructing or occluding the foramen of Magendie (52.3%). Arachnoid veils varied in type and were observed in 59.5% of patients with CM-I who had syringomyelia, which was significantly greater than the 33.3% of patients with CM-I without syringomyelia who had an arachnoid veil (p = 0.018). The presence of CM-I with an arachnoid veil had 3.22 times the odds (p = 0.013, 95% CI 1.29-8.07, by multivariate logistic regression) of being associated with syringomyelia, adjusting for tonsillar herniation. The inferior descent of the fourth ventricle and CMJ occurred with a greater degree of tonsillar herniation (p < 0.001) and correlated with a cervicomedullary kink or buckle on preoperative MRI. CONCLUSIONS Intradural pathology associated with CM-I with or without syringomyelia exists in many forms, is more prevalent than previously recognized in patients of all ages, and may play a role in the pathophysiology of CM-I tonsillar herniation. Arachnoid veils appear to partially obstruct CSF flow, are significantly more prevalent in cases of CM-I with syringomyelia, and therefore may play a role in the pathophysiology of CM-I-associated syringomyelia.

Longitudinal decline of driving safety in Parkinson disease.

OBJECTIVE: To longitudinally assess and predict on-road driving safety in Parkinson disease (PD). METHODS: Drivers with PD (n = 67) and healthy controls (n = 110) drove a standardized route in an instrumented vehicle and were invited to return 2 years later. A professional driving expert reviewed drive data and videos to score safety errors. RESULTS: At baseline, drivers with PD performed worse on visual, cognitive, and motor tests, and committed more road safety errors compared to controls (median PD 38.0 vs controls 30.5; p < 0.001). A smaller proportion of drivers with PD returned for repeat testing (42.8% vs 62.7%; p < 0.01). At baseline, returnees with PD made fewer errors than nonreturnees with PD (median 34.5 vs 40.0; p < 0.05) and performed similar to control returnees (median 33). Baseline global cognitive performance of returnees with PD was better than that of nonreturnees with PD, but worse than for control returnees (p < 0.05). After 2 years, returnees with PD showed greater cognitive decline and larger increase in error counts than control returnees (median increase PD 13.5 vs controls 3.0; p < 0.001). Driving error count increase in the returnees with PD was predicted by greater error count and worse visual acuity at baseline, and by greater interval worsening of global cognition, Unified Parkinson's Disease Rating Scale activities of daily living score, executive functions, visual processing speed, and attention. CONCLUSIONS: Despite drop out of the more impaired drivers within the PD cohort, returning drivers with PD, who drove like controls without PD at baseline, showed many more driving safety errors than controls after 2 years. Driving decline in PD was predicted by baseline driving performance and deterioration of cognitive, visual, and functional abnormalities on follow-up.

CONTEXTUALIZING NATURALISTIC DRIVING DATA IN A RURAL STATE AMONG DRIVERS WITH AND WITHOUT OBSTRUCTIVE SLEEP APNEA.

In naturalistic studies, Global Positioning System (GPS) data and date/time stamps can link driver exposure to specific environments (e.g., road types, speed limits, night driving, etc.), providing valuable context for analyzing critical events, such as crashes, near crashes, and breaches of accelerometer limits. In previous work, we showed how to automate this contextualization, using GPS data obtained at 1 Hz and merging this with Geographic Information Systems (GIS) databases maintained by the Iowa Department of Transportation (DOT). Here we further demonstrate our methods by analyzing data from 80 drivers with obstructive sleep apnea (OSA) and 48 controls, and comparing the two groups with respect to several factors of interest. The majority of comparisons found no difference between groups, suggesting similar patterns of exposures to driving environments in OSA and control drivers. However, OSA drivers appeared to spend slightly more time on roads with annual traffic counts of 500-10,000 and less time driving on wider highways, during twilight, and on roads with 10,000-25,000 annual traffic counts.

Using kinematic driving data to detect sleep apnea treatment adherence.

People spend a significant amount of time behind the wheel of a car. Recent advances in data collection facilitate continuously monitoring this behavior. Previous work demonstrates the importance of this data in driving safety but does not extended beyond the driving domain. One potential extension of this data is to identify driver states related to health conditions such as obstructive sleep apnea (OSA). We collected driving data and medication adherence from a sample of 75 OSA patients over 3.5 months. We converted speed and acceleration behaviors to symbols using symbolic aggregate approximation and converted these symbols to pattern frequencies using a sliding window. The resulting frequency data was matched with treatment adherence information. A random forest model was trained on the data and evaluated using a held-aside test dataset. The random forest model detects lapses in treatment adherence. An assessment of variable importance suggests that the important patterns of driving in classification correspond to route decisions and patterns that may be associated with drowsy driving. The success of this approach suggests driving data may be valuable for evaluating new treatments, analyzing side effects of medications, and that the approach may benefit other drowsiness detection algorithms.

Sex-specific effects of the Huntington gene on normal neurodevelopment.

Huntington disease is a neurodegenerative disorder caused by a gene (HTT) with a unique feature of trinucleotide repeats ranging from 10 to 35 in healthy people; when expanded beyond 39 repeats, Huntington disease develops. Animal models demonstrate that HTT is vital to brain development; however, this has not been studied in humans. Moreover, evidence suggests that triplet repeat genes may have been vital in evolution of the human brain. Here we evaluate brain structure using magnetic resonance imaging and brain function using cognitive tests in a sample of school-aged children ages 6 to 18 years old. DNA samples were processed to quantify the number of CAG repeats within HTT. We find that the number of repeats in HTT, below disease threshold, confers advantageous changes in brain structure and general intelligence (IQ): the higher the number of repeats, the greater the change in brain structure, and the higher the IQ. The pattern of structural brain changes associated with HTT is strikingly different between males and females. HTT may confer an advantage or a disadvantage depending on the repeat length, playing a key role in either the evolution of a superior human brain or development of a uniquely human brain disease. © 2016 Wiley Periodicals, Inc.

Time-to-contact estimation errors among older drivers with useful field of view impairments.

Previous research indicates that useful field of view (UFOV) decline affects older driver performance. In particular, elderly drivers have difficulty estimating oncoming vehicle time-to-contact (TTC). The objective of this study was to evaluate how UFOV impairments affect TTC estimates in elderly drivers deciding when to make a left turn across oncoming traffic. TTC estimates were obtained from 64 middle-aged (n=17, age=46±6years) and older (n=37, age=75±6years) licensed drivers with a range of UFOV abilities using interactive scenarios in a fixed-base driving simulator. Each driver was situated in an intersection to turn left across oncoming traffic approaching and disappearing at differing distances (1.5, 3, or 5s) and speeds (45, 55, or 65mph). Drivers judged when each oncoming vehicle would collide with them if they were to turn left. Findings showed that TTC estimates across all drivers, on average, were most accurate for oncoming vehicles travelling at the highest velocities and least accurate for those travelling at the slowest velocities. Drivers with the worst UFOV scores had the least accurate TTC estimates, especially for slower oncoming vehicles. Results suggest age-related UFOV decline impairs older driver judgment of TTC with oncoming vehicles in safety-critical left-turn situations. Our results are compatible with national statistics on older driver crash proclivity at intersections.

Modeling Likert scale outcomes with trend-proportional odds with and without cluster data.

Likert scales are commonly used in epidemiological studies employing surveys. In this tutorial we demonstrate how the proportional odds model and the trend odds model can be applied simultaneously to data measured in Likert scales, allowing for random cluster effects. We use two datasets as examples: an epidemiological study on aging and cognition among community-dwelling Black persons, and a clustered large survey data from 28,882 students in 81 middle schools. The first example models the Likert outcome from the question: "People act as if they think you are dishonest". The trend-proportional odds model indicates that Black men have higher odds than Black women of reporting being perceived dishonest. The second example models the Likert outcome from the question: "How often have you been beaten up at school?". The trend-proportional odds model indicates that children with disability have a higher odds of severe violence than other children. For both examples, the cumulative odds ratio increases by more than 60% at the higher Likert levels.

LINKING GPS DATA TO GIS DATABASES IN NATURALISTIC STUDIES: EXAMPLES FROM DRIVERS WITH OBSTRUCTIVE SLEEP APNEA.

In naturalistic studies, it is vital to give appropriate context when analyzing driving behaviors. Such contextualization can help address the hypotheses that explore a) how drivers perform within specific types of environment (e.g., road types, speed limits, etc.), and b) how often drivers are exposed to such specific environments. In order to perform this contextualization in an automated fashion, we are using Global Positioning System (GPS) data obtained at 1 Hz and merging this with Geographic Information Systems (GIS) databases maintained by the Iowa Department of Transportation (DOT). In this paper, we demonstrate our methods of doing this based on data from 43 drivers with obstructive sleep apnea (OSA). We also use maps from GIS software to illustrate how information can be displayed at the individual drive or day level, and we provide examples of some of the challenges that still need to be addressed.