Accounting for informative observation process in transition models of binary longitudinal outcome: Application to medical record data

When extracting medical record data to form a retrospective cohort, investigators typically focus on a pre-specified study window, and select subjects who had hospital visits during that study window. However, such data extraction may suffer from an informative observation process, since sicker patients may have hospital visits more frequently. For example, Consecutive Pregnancy Study is a retrospective cohort study of women with multiple pregnancies in 23 Utah hospitals from 2003 to 2010, where the interest is to understand the risk factors of recurrent pregnancy outcomes, such as preterm birth. The observation process is informative in the sense that, women with adverse pregnancy outcomes may be less likely/willing/able to endure subsequent pregnancies. We proposed a three-part joint model with shared random effects structure to address this analytic complication. Particularly, a first-order transition model is used to model the longitudinal binary outcome; a gamma regression model is assumed for the inter-pregnancy intervals; a continuation ratio model specifies the probability of continuing with more births in the future. We note that the latter two parts give rise to a parametric cure-rate survival model. The performance of the proposed method was examined in extensive simulation studies, with both correctly and mis-specified models. The analyses of Consecutive Pregnancy Study data further demonstrate the inadequacies of fitting the transition model alone ignoring the informative observation process.

Measuring hand movement for suturing skill assessment: A simulation-based study

BACKGROUND

To maximize patient safety, surgical skills education is increasingly adopting simulation-based curricula for formative skills assessment and training. However, many standardized assessment tools rely on human raters for performance assessment, which is resource-intensive and subjective. Simulators that provide automated and objective metrics from sensor data can address this limitation. We present an instrumented bench suturing simulator, patterned after the clock face radial suturing model from the Fundamentals of Vascular Surgery, for automated and objective assessment of open suturing skills.

METHODS

For this study, 97 participants (35 attending surgeons, 32 residents, and 30 novices) were recruited at national vascular conferences. Automated hand motion metrics, especially focusing on rotational motion analysis, were developed from the inertial measurement unit attached to participants' hands, and the proposed suite of metrics was used to differentiate between the skill levels of the 3 groups.

RESULTS

Attendings' and residents' performances were found to be significantly different from novices for all metrics. Moreover, most of our novel metrics could successfully distinguish between finer skill differences between attending and resident groups. In contrast, traditional operative skill metrics, such as time and path length, were unable to distinguish attendings from residents.

CONCLUSION

This study provides evidence for the effectiveness of rotational motion analysis in assessing suturing skills. The suite of inertial measurement unit-based hand motion metrics introduced in this study allows for the incorporation of hand movement data for suturing skill assessment.

Needle Angle Matters: An Investigation of the Effect of Needle Angle on Hemodialysis Cannulation Skill

Key Points

Three metrics that quantify cannulation skill on the basis of needle angle are introduced. All three needle angle metrics were demonstrated to be useful in predicting cannulation outcomes on the simulator.

Background

Cannulation is critical for maintaining a functional vascular access for patients on hemodialysis. However, relatively little is known about the quantitative aspects of needle insertion dynamics required for skilled cannulation. In this article, we introduce three kinds of metrics that quantify needle insertion angle—recognized as important for safe and effective cannulation—during cannulation on an instrumented simulator for skill assessment. Three questions were examined in this study: (1 ) Are simulator-based needle angle metrics related to cannulation success? (2 ) Are needle angle metrics related to simulated blood flashback quality? and (3 ) Can needle angle metrics be used to distinguish between high and low skill levels?

Methods

Fifty-one cannulators with varying degrees of clinical experience performed cannulation on the instrumented simulator. Each participant cannulated 16 times on different fistulas with varying geometries. During each trial, needle angle along with other sensor data was obtained through a motion sensor placed inside the needle. Data analysis was conducted by relating needle angle over time with our previously validated simulator-based cannulation outcome metrics.

Results

The results revealed that all three types of needle angle metrics were useful in predicting the probability of cannulation success. In addition, they were also correlated with flashback quality metrics. Furthermore, these metrics successfully distinguished between high and low performers regardless of whether they were classified using subjective ratings or objective scores. These results indicate that needle insertion angle is an important component of cannulation skill.

Conclusions

The simulator-based metrics for needle insertion angle presented in this work measure a key aspect of skilled cannulation. As such, if implemented in a structured way, these metrics could lead to competency-based skill assessment and training for cannulation in the future. Raising the bar of cannulation skill of our clinicians can have a tangible effect on patient outcomes.

Relating process and outcome metrics for meaningful and interpretable cannulation skill assessment: A machine learning paradigm

BACKGROUND AND OBJECTIVES

The quality of healthcare delivery depends directly on the skills of clinicians. For patients on hemodialysis, medical errors or injuries caused during cannulation can lead to adverse outcomes, including potential death. To promote objective skill assessment and effective training, we present a machine learning approach, which utilizes a highly-sensorized cannulation simulator and a set of objective process and outcome metrics.

METHODS

In this study, 52 clinicians were recruited to perform a set of pre-defined cannulation tasks on the simulator. Based on data collected by sensors during their task performance, the feature space was then constructed based on force, motion, and infrared sensor data. Following this, three machine learning models- support vector machine (SVM), support vector regression (SVR), and elastic net (EN)- were constructed to relate the feature space to objective outcome metrics. Our models utilize classification based on the conventional skill classification labels as well as a new method that represents skill on a continuum.

RESULTS

With less than 5% of trials misplaced by two classes, the SVM model was effective in predicting skill based on the feature space. In addition, the SVR model effectively places both skill and outcome on a fine-grained continuum (versus discrete divisions) that is representative of reality. As importantly, the elastic net model enabled the identification of a set of process metrics that highly impact outcomes of the cannulation task, including smoothness of motion, needle angles, and pinch forces.

CONCLUSIONS

The proposed cannulation simulator, paired with machine learning assessment, demonstrates definite advantages over current cannulation training practices. The methods presented here can be adopted to drastically increase the effectiveness of skill assessment and training, thereby potentially improving clinical outcomes of hemodialysis treatment.

Simulator-Based Metrics for Quantifying Vascular Palpation Skill for Cannulation

Palpation is essential for accurate diagnosis and treatment in many clinical examinations and procedures. Specifically, vascular palpation is used to diagnose cardiovascular health issues and identify anatomical landmarks in the peripheral vascular system. However, little attention has been given to quantifying what comprises skilled vascular palpation; therefore, this study aims to objectively quantify the differences between high performer (HP), mid performer (MP), and low performer (LP) behavior towards understanding vascular palpation skills. Eleven HPs, twenty-five MPs, and ten LPs completed sixteen trials on our simulator under various conditions. There were four fistulas, two skin thicknesses, and two motor vibration intensities. Finger force and location data were recorded for each trial on the simulator. We examined three types of palpation metrics: time, force, and location. All three types of metrics demonstrated statistically significant differences between HP and LP palpation behavior. Therefore, these metrics could be used for structured and standardized palpation skills training in the future, potentially improving patient outcomes.

Measuring Cannulation Skills for Hemodialysis: Objective Versus Subjective Assessment

Lack of cannulation skill during hemodialysis treatments results in poor clinical outcomes due to infiltration and other cannulation-related trauma. Unfortunately, training of patient care technicians and nurses, specifically on the "technical" aspects of cannulation, has traditionally not received much attention. Simulators have been successfully deployed in many medical specialties for assessment and training of clinical skills. However, simulators have not been as widely used in nursing, especially in the context of training clinical personnel in the dialysis unit. We designed a state-of-the-art simulator for quantifying skill for hemodialysis cannulation. In this study, 52 nurses and patient care technicians with varying levels of clinical experience performed 16 cannulations on the simulator with different fistula properties. We formulated a composite metric for objectively measuring overall success of cannulation and compared this metric with subjective assessment by experts. In addition, we examined if years of clinical experience correlated with objective and subjective scores for cannulation skill. Results indicated that, while subjective and objective metrics generally correlated with each other, the objective metric was more precise and better suited for quantifying cannulation skill. Further, the simulator-based objective metric provides several advantages over subjective ratings, including providing fine-grained assessment of skill, consistency in measurement unaffected by subjective biases, and basing assessment on a more complete evaluation of performance. Years of clinical experience, however, demonstrated little correlation with either method of skill assessment. The methods presented for cannulation skill assessment in this study, if widely applied, could result in improved cannulation skill among our PCTs and nurses, which could positively impact patient outcomes in a tangible way.

Motion Smoothness Metrics for Cannulation Skill Assessment: What Factors Matter?

Medical training simulators have the potential to provide remote and automated assessment of skill vital for medical training. Consequently, there is a need to develop "smart" training devices with robust metrics that can quantify clinical skills for effective training and self-assessment. Recently, metrics that quantify motion smoothness such as log dimensionless jerk (LDLJ) and spectral arc length (SPARC) are increasingly being applied in medical simulators. However, two key questions remain about the efficacy of such metrics: how do these metrics relate to clinical skill, and how to best compute these metrics from sensor data and relate them with similar metrics? This study addresses these questions in the context of hemodialysis cannulation by enrolling 52 clinicians who performed cannulation in a simulated arteriovenous (AV) fistula. For clinical skill, results demonstrate that the objective outcome metric flash ratio (FR), developed to measure the quality of task completion, outperformed traditional skill indicator metrics (years of experience and global rating sheet scores). For computing motion smoothness metrics for skill assessment, we observed that the lowest amount of smoothing could result in unreliable metrics. Furthermore, the relative efficacy of motion smoothness metrics when compared with other process metrics in correlating with skill was similar for FR, the most accurate measure of skill. These results provide guidance for the computation and use of motion-based metrics for clinical skill assessment, including utilizing objective outcome metrics as ideal measures for quantifying skill.

Practical issues in using generalized estimating equations for inference on transitions in longitudinal data: What is being estimated?

Generalized estimating equations (GEEs) are commonly used to estimate transition models. When the Markov assumption does not hold but first-order transition probabilities are still of interest, the transition inference is sensitive to the choice of working correlation. In this paper, we consider a random process transition model as the true underlying data generating mechanism, which characterizes subject heterogeneity and complex dependence structure of the outcome process in a very flexible way. We formally define two types of transition probabilities at the population level: "naive transition probabilities" that average across all the transitions and "population-average transition probabilities" that average the subject-specific transition probabilities. Through asymptotic bias calculations and finite-sample simulations, we demonstrate that the unstructured working correlation provides unbiased estimators of the population-average transition probabilities while the independence working correlation provides unbiased estimators of the naive transition probabilities. For population-average transition estimation, we demonstrate that the sandwich estimator fails for unstructured GEE and recommend the use of either jackknife or bootstrap variance estimates. The proposed method is motivated by and applied to the NEXT Generation Health Study, where the interest is in estimating the population-average transition probabilities of alcohol use in adolescents.

Prospective Associations of Actual and Perceived Descriptive Norms with Drinking Among Emerging Adults

BACKGROUND

Descriptive norms are commonly associated with participant drinking. However, study participants may incorrectly perceive that their peers drink about the same amount as they do, which would bias estimates of drinking homogeneity.

OBJECTIVES

This research examined the magnitude of associations between emerging adults' reports of their own drinking and peer drinking measured the previous year by measures of (1) participants' perceptions of friends' drinking; and (2) actual drinking reported by nominated peers.

METHODS

The data are from annual surveys conducted in 2014 and 2015, Waves 4 and 5 (the first 2 years after high school) of 7 annual assessments as part of the NEXT Generation Health Study (n = 323). Associations of participant alcohol use with perceived friend use (five closest, closest male, and closest female friends), and with actual peer use. Logistic regression analyses estimated the magnitudes of prospective associations between each measure of peer drinking at W4 and participant drinking at W5.

Cell Phone Use While Driving: Prospective Association with Emerging Adult Use

Secondary task engagement such as cell phone use while driving is a common behavior among adolescents and emerging adults. Texting and other distracting cell phone use in this population contributes to the high rate of fatal car crashes. Peer engagement in similar risky driving behaviors, such as texting, could socially influence driver phone use behavior. The present study investigates the prospective association between peer and emerging adult texting while driving the first year after high school. Surveys were conducted with a national sample of emerging adults and their nominated peers. Binomial logistic regression analyses, adjusting for gender, race/ethnicity, parental education, and family affluence, showed that participants (n=212) with peers (n=675) who reported frequently texting while driving, were significantly more likely to text while driving the following year (odds ratio, 3.01; 95% CI, 1.19-7.59; P=0.05). The findings are consistent with the idea that peer texting behavior influences the prevalence of texting while driving among emerging adults.

Do Young Drivers Become Safer After Being Involved in a Collision?

As drivers age, their risk of being involved in a car collision decreases. The present study investigated if this trend is due, in part, to some risky drivers having a collision early in their driving lives and subsequently reducing their risky driving after that negative experience. Accelerometers and video cameras were installed in the vehicles of 16- to 17-year-old drivers ( N = 254), allowing coders to measure the number of g-force events (i.e., events in which a threshold acceleration level was exceeded) per 1,000 miles and the number of collisions. Among the 41 participants who experienced a severe collision, the rate of g-force events dropped significantly in the 1st month after the collision, remained unchanged for the 2nd month, and increased significantly in the 3rd month. There were no changes in the rate of g-force events at comparable time points for the drivers not involved in a collision. Being involved in a collision led to a decrease in risky driving, but this may have been a temporary effect.

Variability in measures of health and health behavior among emerging adults 1 year after high school according to college status

OBJECTIVE

To examine changes in health behaviors among US emerging adults 1 year after high school.

PARTICIPANTS

The national sample of participants (N = 1,927), including those attending 4-year college/university (n = 884), 2-year colleges/technical schools (n = 588), and no college (n = 455), participated in annual spring surveys 2013-2014.

METHODS

Health behaviors were assessed the last year of high school and first year of college; differences by college status controlling for previous-year values were estimated using regression analyses.

RESULTS

Relative to 4-year college attendees, those attending technical school/community college were less likely to binge drink (odds ratio [OR] = 0.57, confidence interval [CI] = 0.38-0.86) but more likely to speed (OR = 1.26, CI = 1.0-2.84), consume sodas (OR = 1.57, CI = 1.0-2.47), and report lower family satisfaction (p < .01), with marginally more physical and depressive symptoms. College nonattendees reported more DWI (driving while intoxicated; OR = 1.60, CI = 1.05-2.47), soda drinking (OR = 2.51, CI = 1.76-3.59), oversleeping (OR = 4.78, CI = 3.65-8.63), and less family satisfaction (p < .04).

CONCLUSIONS

Health risk behaviors among emerging adults varied by college status.

Cluster adjusted regression for displaced subject data (CARDS): Marginal inference under potentially informative temporal cluster size profiles

Ignorance of the mechanisms responsible for the availability of information presents an unusual problem for analysts. It is often the case that the availability of information is dependent on the outcome. In the analysis of cluster data we say that a condition for informative cluster size (ICS) exists when the inference drawn from analysis of hypothetical balanced data varies from that of inference drawn on observed data. Much work has been done in order to address the analysis of clustered data with informative cluster size; examples include Inverse Probability Weighting (IPW), Cluster Weighted Generalized Estimating Equations (CWGEE), and Doubly Weighted Generalized Estimating Equations (DWGEE). When cluster size changes with time, i.e., the data set possess temporally varying cluster sizes (TVCS), these methods may produce biased inference for the underlying marginal distribution of interest. We propose a new marginalization that may be appropriate for addressing clustered longitudinal data with TVCS. The principal motivation for our present work is to analyze the periodontal data collected by Beck et al. (1997, Journal of Periodontal Research 6, 497-505). Longitudinal periodontal data often exhibits both ICS and TVCS as the number of teeth possessed by participants at the onset of study is not constant and teeth as well as individuals may be displaced throughout the study.

Multiplex molecular testing for management of infectious gastroenteritis in a hospital setting: a comparative diagnostic and clinical utility study

Laboratory diagnosis and clinical management of inpatients with diarrhoea is complex and time consuming. Tests are often requested sequentially and undertaken in different laboratories. This causes prolonged unnecessary presumptive isolation of patients, because most cases are non-infectious. A molecular multiplex test (Luminex(®) Gastrointestinal Pathogen Panel (GPP)) was compared with conventional testing over 8 months to determine diagnostic accuracy, turnaround times, laboratory costs, use of isolation facilities and user acceptability. A total of 262 (12%) patients had a pathogen detected by conventional methods compared with 483 (22.1%) by GPP. Most additional cases were detected in patients developing symptoms in the first 4 days of admission. Additional cases were detected because of presumed improved diagnostic sensitivity but also because clinicians had not requested the correct pathogen. Turnaround time (41.8 h) was faster than bacterial culture (66.5 h) and parasite investigation (66.5 h) but slower than conventional testing for Clostridium difficile (17.3 h) and viruses (27 h). The test could allow simplified requesting by clinicians and a consolidated laboratory workflow, reducing the overall number of specimens received by the laboratory. A total of 154 isolation days were saved at an estimated cost of £30 800. Consumables and labour were estimated at £150 641 compared with £63 431 for conventional testing. Multiplex molecular testing using a panel of targets allowed enhanced detection and a consolidated laboratory workflow. This is likely to be of greater benefit to cases that present within the first 4 days of hospital admission.

A security vulnerabilities assessment tool for interim storage facilities of low-level radioactive wastes

Limited permanent low-level radioactive waste (LLRW) disposal capacity and correspondingly high disposal costs have resulted in the creation of numerous interim storage facilities for either decay-in-storage operations or longer term accumulation efforts. These facilities, which may be near the site of waste generation or in distal locations, often were not originally designed for the purpose of LLRW storage, particularly with regard to security. Facility security has become particularly important in light of the domestic terrorist acts of 2001, wherein LLRW, along with many other sources of radioactivity, became recognized commodities to those wishing to create disruption through the purposeful dissemination of radioactive materials. Since some LLRW materials may be in facilities that may exhibit varying degrees of security control sophistication, a security vulnerabilities assessment tool grounded in accepted criminal justice theory and security practice has been developed. The tool, which includes dedicated sections on general security, target hardening, criminalization benefits, and the presence of guardians, can be used by those not formally schooled in the security profession to assess the level of protection afforded to their respective facilities. The tool equips radiation safety practitioners with the ability to methodically and systematically assess the presence or relative status of various facility security aspects, many of which may not be considered by individuals from outside the security profession. For example, radiation safety professionals might not ordinarily consider facility lighting aspects, which is a staple for the security profession since it is widely known that crime disproportionately occurs more frequently at night or in poorly lit circumstances. Likewise, the means and associated time dimensions for detecting inventory discrepancies may not be commonly considered. The tool provides a simple means for radiation safety professionals to assess, and perhaps enhance in a reasonable fashion, the security of their interim storage operations. Aspects of the assessment tool can also be applied to other activities involving the protection of sources of radiation as well.

Weight Lifting and Aortic Dissection: More Evidence for a Connection

BACKGROUND/AIMS

In 2003, we reported on a small number of patients in whom acute aortic dissection appeared to be causally related to intense weight lifting. If additional cases could be identified, the phenomenon of weight lifting induced aortic dissection would be further substantiated. We now report a substantially larger number of cases in which aortic dissection is associated with intense physical exertion.

METHODS

Additional cases of acute aortic dissection occurring at the time of intense physical exertion were accumulated and analyzed. Cases were culled from retrospective review of a large university data base and from reports forwarded to our attention from around the country. We determined type of activity bringing on symptoms, age and sex of the patients, location of the dissection (ascending or descending aorta), aortic size, therapy, and survival.

RESULTS

We identified 31 patients in whom acute aortic dissection occurred in the context of severe physical exertion, predominantly weight lifting or similar activities. All patients except one were males. Mean age was 47.3 (range = 19-76). All except four dissections were in the ascending aorta. Only three patients (9.7%) had a family history of aortic disease. Mean aortic diameter on the initial imaging study was 4.63 cm. Twenty-six of the 31 cases were diagnosed ante-mortem and 5 post-mortem. Overall, 10 of the 31 patients (32.2%) died. Of 24 patients reaching surgical therapy, 20 (83.3%) survived.

CONCLUSION

Weight lifting related acute aortic dissection appears to be a real phenomenon, with increasing evidence for the association of extreme exertion with this catastrophic aortic event. Moderate aortic dilatation confers vulnerability to exertion-related aortic dissection. Individuals with known aortic dilatation should be cautioned to refrain from weight lifting or strenuous exertion. Routine echocardiographic screening of individuals engaging in heavy strength training should be considered, in order to prevent this tragic loss of life.

Viral load as a determinant for transmission of human papillomavirus type 16 from mother to child

Whilst genital papillomaviruses are commonly believed to be sexually transmitted, transmission of human papillomavirus type 16 (HPV-16) from mother to child at delivery has been described previously [Pakarian et al. (in press) British Journal of Obstetrics and Gynaecology]. In order to determine whether viral load in cervical/vaginal cells was an important determinant of transmission 15 pregnant women with HPV-16 infections were studied. Eight of these women had infants who were positive for HPV-16 DNA at genital and/or buccal sites. Viral load was estimated by laser densitometry of polymerase chain reaction (PCR) products. The eight mothers--four with a previous history of abnormal smears and two with previous genital warts--who transmitted infection to their infants had significantly higher viral loads (P < 0.05) than those who did not. It is concluded that viral load is an important, but not the sole, determinant for the transmission of HPV-16 from mother to infant.