Search

Summary Background Children exposed to poverty and family adversities including domestic violence, parental mental ill health and parental alcohol misuse may experience poor outcomes across the life course. However, the complex interrelationships between these exposures in childhood are unclear. We therefore assessed the clustering of trajectories of household poverty and family adversities and their impacts on adolescent health outcomes. Methods We used longitudinal data from the UK Millennium Cohort study on 11564 children followed to age 14 years. Family adversities included parent reported domestic violence and abuse, poor mental health and frequent alcohol use. We used a group-based multi-trajectory cluster model to identify trajectories of poverty and family adversity for children. We assessed associations of these trajectories with child physical, mental and behavioural outcomes at age 14 years using multivariable logistic regression, adjusting for confounders. Findings Six trajectories were identified: low poverty and family adversity (43·2%), persistent parental alcohol use (7·7%), persistent domestic violence and abuse (3·4%), persistent poor parental mental health (11·9%), persistent poverty (22·6%) and persistent poverty and poor parental mental health (11·1%). Compared with children exposed to low poverty and adversity, children in the persistent adversity trajectory groups experienced worse outcomes; those exposed to persistent poor parental mental health and poverty were particularly at increased risk of socioemotional behavioural problems (adjusted odds ratio 6·4; 95% CI 5·0 – 8·3), cognitive disability (aOR 2·1; CI 1·5 – 2·8), drug experimentation (aOR 2·8; CI 1·8 – 4·2) and obesity (aOR 1·8; CI 1·3 – 2·5). Interpretation In a contemporary UK cohort, persistent poverty and/or persistent poor parental mental health affects over four in ten children. The combination of both affects one in ten children and is strongly associated with adverse child outcomes, particularly poor child mental health. Funding The National Institute for Health Research (NIHR) Policy Research Programme, NIHR Applied Research Collaboration South London (ARC South London) at King's College Hospital NHS Foundation Trust and the Medical Research Council (MRC).

Objective In this study, we aimed to inform clinical practice by identifying distinct subgroups of U.S. veteran psychiatry inpatients on their alcohol and drug use severity, posttraumatic stress disorder (PTSD) symptoms, and suicidal ideation over time. Method Participants were 406 patients with co-occurring substance use and mental health disorders. A parallel latent growth trajectory model was used to characterize participants' symptom severity across 15 months posttreatment intake. Results Four distinct classes were identified: 47% "normative improvement," 32% "high PTSD," 11% "high drug use," and 9% "high alcohol use." Eighty percent of the sample had reduced their drinking and drug intake by half from baseline to 3 months, and those levels remained stable from 3 to 15 months. The High PTSD, High Drug Use, and High Alcohol Use classes all reported levels of PTSD symptomatology at baseline consistent with a clinical diagnosis, and symptom levels remained high and stable across all 15 months. The Normative Improvement class showed declining drug and alcohol intake and was the only class exhibiting reductions in PTSD symptomatology over time. High substance use classes showed initial declines in suicidal ideation, then an increase from 9 to 15 months. Conclusions The reduction in frequency of drinking and drug use for 80% of the sample was substantial and supports the potential efficacy of current treatment approaches. However, the high and stable levels of PTSD for more than 50% of the sample, as well as the reemergence of suicidal ideation in a sizable subgroup, underscore the difficulty in finding and linking patients to effective interventions to decrease symptomatology over time.

In order to deepen understanding of team processes in dynamic organizational contexts, we suggest that analyses employing techniques to identify and analyze team member interaction patterns and trajectories are necessary. After presenting a brief review of interaction data coding and reliability requirements, we first review examples of two approaches used in the identification and analysis of interaction patterns in teams: lag sequential analysis and T-pattern analysis. We then describe and discuss three statistical techniques used to analyze team interaction trajectories: random coefficient modeling, latent growth modeling, and discontinuous growth analysis. We close by suggesting several ways in which these techniques could be applied to data analysis in order to expand our knowledge of team interaction, processes, and outcomes in complex and dynamic settings.

Most of the existing single object trackers track the target in a unitary local search window, making them particularly vulnerable to challenging factors such as heavy occlusions and out-of-view movements. Despite the attempts to further incorporate global search, prevailing mechanisms that cooperate local and global search are relatively static, thus are still sub-optimal for improving tracking performance. By further studying the local and global search results, we raise a question: can we allow more dynamics for cooperating both results? In this paper, we propose to introduce more dynamics by devising a dynamic attention-guided multi-trajectory tracking strategy. In particular, we construct dynamic appearance model that contains multiple target templates, each of which provides its own attention for locating the target in the new frame. Guided by different attention, we maintain diversified tracking results for the target to build multi-trajectory tracking history, allowing more candidates to represent the true target trajectory. After spanning the whole sequence, we introduce a multi-trajectory selection network to find the best trajectory that delivers improved tracking performance. Extensive experimental results show that our proposed tracking strategy achieves compelling performance on various large-scale tracking benchmarks. The project page of this paper can be found at https://sites.google.com/view/mt-track/., Comment: Accepted by IEEE T-CSVT 2021

Total CO2 emissions from the cement industry approximately correspond to 8% of the global CO2 emissions. In this chapter, we compare and contrast the co-evolutionary trajectories of eco-cement technology and policy in the Netherlands, China and Japan. We reveal that the transition processes away from Ordinary Portland Cement (OPC) to eco-cement follow different conjunctural causation patterns and temporalities due to “multiple stream determinants” emanating from varying degrees of (i) intervention capacities and capabilities of actors in politico-administrative and sociocultural domains, (ii) socio-technical alignment of the policy stream to techno-economic and scientific possibilities, and (iii) the nature of market demand with regard to both traditional and alternative cements. A common target in each context is to reduce the contribution of this basic industry in overall carbon dioxide (CO2) emissions, with circular economy policies playing a significantly increasing role in each trajectory.

OBJECTIVE: To examine the hypothesis that PTSD with delayed expression in some cases occurs without subthreshold PTSD symptoms above background levels bridging the gap between the traumatic exposure(s) and the clinical diagnosis.METHODS: We performed systematic searches of peer-reviewed papers in English referenced in Pubmed, Embase, or PsycINFO and ascertained 34 prospective studies of PTSD symptom trajectories identified by latent class growth statistical modeling. Studies with delayed and low-stable trajectories provided appropriate data for this study. We computed the difference between the delayed trajectory PTSD symptom sumscore and the low-stable PTSD sumscore at the observed points in time after the traumatic event(s).RESULTS: In 29 study populations, the latent class growth analyses displayed delayed trajectories, and in these, we identified 110 data points (% PTSD sumscore difference/months since traumatic exposure). The median PTSD symptom sumscore was 25% higher during the initial 6 months among individuals in the delayed trajectory compared to those in low-stable trajectory. From this level, the difference widened and reached a plateau of 40-50% higher. The variation was large, and the baseline participation rate and loss to follow-up were exceeding 25% in the majority of the studies. Heterogeneity of populations, measures, and analyses precluded formal meta-analysis.CONCLUSION: Delayed PTSD is preceded by PTSD symptoms during the first year in most cases. Still, few individuals may experience an asymptomatic delay. The results underpin the rationale for monitoring PTSD symptoms and may inform forensic assessments in that delayed PTSD without symptoms bridging the traumatic event is rare.

One of the challenging problems in robot navigation is efficient and safe planning in a highly dynamic environment, where the robot is required to understand pedestrian patterns in the environment, such as train station. The rapid movement of pedestrians makes the robot more difficult to solve the collision problem. In this paper, we propose a navigation probability map to solve the pedestrians&rsquo, rapid movement problem based on the influencer recognition model (IRM). The influencer recognition model (IRM) is a data-driven model to infer a distribution over possible causes of pedestrian&rsquo, s turning. With this model, we can obtain a navigation probability map by analyzing the changes in the effective pedestrian trajectory. Finally, we combined navigation probability map and artificial potential field (APF) method to propose a robot navigation method and verified it on our data-set, which is an unobstructed, overlooked pedestrians&rsquo, data-set collected by us.

With the rapid development of the Internet of Things (IoT), massive trajectories have been generated. Trajectory data is beneficial for many urban applications. This demo presents a holistic trajectory data management system based on distributed platforms, such as Spark and HBase, namely JUST-Traj. It provides a variety of indexes to efficiently support spatio-temporal queries and analyses on massive trajectories. Additionally, it provides a convenient SQL engine to execute all operations (storage, queries, analyses) through a SQL-like statement. Finally, we design a web portal for developers and demonstrate different operations in the portal.

While interstitial lung disease (ILD) is the leading cause of morbidity and mortality in systemic sclerosis (SSc), there remains a paucity of predictive markers to assess disease progression. We previously demonstrated that adipose tissue metabolism and adipokine homeostasis is dysregulated in SSc. The present study was undertaken to determine the association and predictive ability of the novel adipokine C1q/tumor necrosis factor-related protein 9 (CTRP9) for SSc-associated ILD.We performed a retrospective longitudinal study utilizing the Northwestern Scleroderma Program Patient Registry and Biorepository. Serum levels of CTRP9 were measured in 110 SSc patients at baseline, and demographic, clinical, and pulmonary function test data were collected in 12-month intervals to 48 months. Longitudinal trajectory of forced vital capacity percent predicted (FVC%) was used as a primary outcome measure. We utilized a mixed model to compare trajectories of lung function by CTRP9 groups and performed latent trajectory analysis to accommodate for heterogeneity.In cross-sectional analysis, elevated circulating CTRP9 was associated with significantly lower FVC% at baseline (72% ± 17 versus 80% ± 18; P = 0.02) and 48 months (68 ± 19 versus 84 ± 18; P = 0.001). In mixed model analysis, high CTRP9 was associated with worse lung function but not with a different trajectory (P = 0.23). In contrast, low CTRP9 identified patients with stability of lung disease with reasonable accuracy (sensitivity 73%). Latent trajectory analysis confirmed the association of lower CTRP9 with higher FVC%.Higher circulating CTRP9 associated with worse pulmonary function, while low CTRP9 identified patients with lung disease stability over time. These findings suggest that CTRP9 may be a potential biomarker in SSc-associated ILD.

Objective To investigate whether epileptiform discharge burden can identify those at risk for delayed cerebral ischemia (DCI) after subarachnoid hemorrhage (SAH). Methods Retrospective analysis of 113 moderate to severe grade SAH patients who had continuous EEG (cEEG) recordings during their hospitalization. We calculated the burden of epileptiform discharges (ED), measured as number of ED per hour. Results We find that many SAH patients have an increase in ED burden during the first 3–10 days following rupture, the major risk period for DCI. However, those who develop DCI have a significantly higher hourly burden from days 3.5–6 after SAH vs. those who do not. ED burden is higher in DCI patients when assessed in relation to the onset of DCI (area under the receiver operator curve 0.72). Finally, specific trends of ED burden over time, assessed by group-based trajectory analysis, also help stratify DCI risk. Conclusions These results suggest that ED burden is a useful parameter for identifying those at higher risk of developing DCI after SAH. The higher burden rate associated with DCI supports the theory of metabolic supply-demand mismatch which contributes to this complication. Significance ED burden is a novel biomarker for predicting those at high risk of DCI.

Maintenance of systemic homeostasis by kidney requires the coordinated response of diverse cell types. The use of single-cell RNA sequencing (scRNAseq) for patient tissue samples remains fraught with difficulties with cell isolation, purity, and experimental bias. The ability to characterize immune and parenchymal cells during transplant rejection will be invaluable in defining transplant pathology where tissue availability is restricted to needle biopsy fragments. Herein, we present feasibility data for multiplexing approach for droplet scRNAseq (Mux-Seq). Mux-Seq has the potential to minimize experimental batch bias and variation even with very small sample input. In this first proof-of-concept study for this approach, explant tissues from six normal and two transplant recipients after multiple early post-transplant rejection episodes leading to nephrectomy due to aggressive antibody mediated rejection, were pooled for Mux-Seq. A computational tool, Demuxlet was applied for demultiplexing the individual cells from the pooled experiment. Each sample was also applied individually in a single microfluidic run (singleplex) to correlate results with the pooled data from the same sample. Our applied protocol demonstrated that data from Mux-Seq correlated highly with singleplex (Pearson coefficient 0.982) sequencing results, with the ability to identify many known and novel kidney cell types including different infiltrating immune cells. Trajectory analysis of proximal tubule and endothelial cells demonstrated separation between healthy and injured kidney from transplant explant suggesting evolving stages of cell- specific differentiation in alloimmune injury. This study provides the technical groundwork for understanding the pathogenesis of alloimmune injury and host tissue response in transplant rejection and normal human kidney and provides a protocol for optimized processing precious and low input human kidney biopsy tissue for larger scale studies.

A considerable body of gambling-related research has addressed the task of segmenting a sample population of gamblers into homogenous sub-groups. Typically, “static” features are used as model inputs for cluster analysis, where variables are aggregated for each individual over a specified period of time; for example, the total amount wagered per gambler over the course of a study period. Engineering features in this way fails to capture the intricacies of a gambler’s behavior over time. Recent works have begun to address this limitation by using time-series data as model inputs and by employing trajectory analysis. While these methods incorporate the changeability in a single variable over time (e.g., the number of bets made over time) they fail to encapsulate interactions between multiple variables and the sequentiality of a gambler’s behavior over time. With this study we attempt to move beyond the use of static aggregated variables and traditionally formatted time-series data. Instead, we pre-process our data so that model inputs reflect a sequence of payment events (i.e., a categorical sequence) per gambler (e.g., deposits, withdrawals, declines, etc.). We will use these sequences to model gamblers’ behavior and extract common patterns via cluster analysis.

BACKGROUND Human metapneumovirus (HMPVi) and parainfluenza virus (PIVi) infections are common community-acquired infections in lung transplant recipients (LTRs), but data are extremely limited. METHODS A retrospective study including all LTRs at the Johns Hopkins Hospital during July 2010-June 2019 with positive HMPV and PIV polymerase chain reaction respiratory specimens was performed. RESULTS Thirty-one HMPV- and 53 PIV-infected LTRs were identified. LTRs with HMPVi and PIVi had similar baseline characteristics, infection parameters, treatment allocation, and allograft function outcomes. Among entire cohort, 31.6% had chronic allograft dysfunction (CLAD) stage progression within 1 y postinfections (29.2% versus 35.5% for PIVi versus HMPVi, respectively, P = 0.56). In forced expiratory volume in 1 s percent (FEV1%) trajectory analysis showed steadily decline of FEV1 across time among CLAD stage progressors from both viruses. FEV1% decline ≥10% at 90 d had adjusted hazard ratio for CLAD stage progression of 18.4 (4.98-67.76) and 4.6 (1.36-15.34) for PIVi and HMPVi, respectively. PIVi caused higher donor-specific antigen development (11.8% versus 3.2%, P = 0.18) and 1-y mortality (9.4% versus 0%, P = 0.11), compared with HMPVi, even though the results were not statistically significant. Ribavirin did not show protective effect, and mycophenolate discontinuation during infection did not increase risk of CLAD stage progression. CONCLUSIONS One-third of HMPV- and PIV-infected LTRs developed CLAD stage progression within 1 y. The lack of early lung function recovery may predict long-term CLAD progression.

BACKGROUND AND AIMS: Employment during and following treatment for alcohol use disorder (AUD) is important for the individual's health and well-being and for reducing the societal costs associated with benefit payments. Nonetheless, this is an under-researched topic. This study aimed to identify trajectories of labour force participation among people enrolled in AUD treatment and describe the characteristics of those following contrasting pathways. DESIGN: Using trajectory analysis, we modelled employment trajectory groups among AUD patients during the year of treatment entry and the 4 subsequent years, applying Norwegian longitudinal register data. PARTICIPANTS: Patients in Norway who entered treatment with AUD as the primary diagnosis during 2009 and 2010 (9000 patients, age 20-61 years). MEASUREMENTS: The outcome variable 'labour force attachment' was measured as being in full-time employment, partly employed, on temporary welfare benefits or on permanent disability pension. Predictors were age, gender, education and comorbid mental health and drug use disorders. FINDINGS: We distinguished six employment trajectories among AUD patients: 15.8% were on permanent disability pension throughout, 8.7% exited the labour force on permanent disability pension during the observation period, 32.1% had a medium attachment throughout follow-up, and 9.2% had a decreasing attachment; 23.3% had a high labour force attachment throughout, and 10.9% experienced increasing attachment. High attachment throughout was negatively associated with being female (P

Asymptomatic patients (AP) travel through neighborhoods in communities. The mobility dynamics of the AP makes it hard to tag them with specific interests. The lack of efficient monitoring systems can enable the AP to infect several vulnerable people in the communities. This article studied the monitoring of AP through their mobility and trajectory towards reducing the stress of socio-economic complications in the case of pandemics. Mobility and Trajectory based Technique for Monitoring Asymptomatic Patients (MTT-MAP) was established. The time-ordered spatial and temporal trajectory records of the AP were captured through their activities. A grid-based index data structure was designed based on network topology, graph theory and trajectory analysis to cater for the continuous monitoring of the AP over time. Also, concurrent object localisation and recognition, branch and bound, and multi-object instance strategies were adopted. The MTT-MAP has shown efficient when experimented with GeoLife dataset and can be integrated with state-of-the-art patients monitoring systems.

Background Critical care nurses have a burnout rate among the highest of any nursing field. Nurse burnout may impact care quality. Few studies have considered how temporal patterns may influence outcomes. Objective To test a longitudinal model of burnout clusters and associations with patient and clinician outcomes. Methods An observational study analyzed data from annual employee surveys and administrative data on patient outcomes at 111 Veterans Health Administration intensive care units from 2013 through 2017. Site-level burnout rates among critical care nurses were calculated from survey responses about emotional exhaustion and depersonalization. Latent trajectory analysis was applied to identify clusters of facilities with similar burnout patterns over 5 years. Regression analysis was used to analyze patient and employee outcomes by burnout cluster and organizational context measures. Outcomes of interest included patient outcomes (30-day standardized mortality rate and observed minus expected length of stay) for 2016 and 2017 and clinician outcomes (intention to leave and employee satisfaction) from 2013 through 2017. Results Longitudinal analysis revealed 3 burnout clusters among the 111 sites: low (n = 37), medium (n = 68), and high (n = 6) burnout. Compared with sites in the low-burnout cluster, those in the high-burnout cluster had longer patient stays, higher employee turnover intention, and lower employee satisfaction in bivariate models but not in multivariate models. Conclusions In this multiyear, multisite study, critical care nurse burnout was associated with key clinician and patient outcomes. Efforts to address burnout among nurses may improve patient and employee outcomes.

Highlights • Tumor-specific smoking-induced differentially expressed genes are different in male and female patients. • A smoking-related tumor-specific Treg subset is defined, highly expressed ADAM12. • ADAM12+ CTLA4+ Tregs actively communicated with exhausted T cells than other Tregs., Tobacco smoking is the major cause of non-small-cell-lung cancer (NSCLC). However, it is barely known how smoking impact the tumor immune environment (TIME) of lung cancer. We integrated single-cell RNA-seq and bulk RNA-seq data from several studies to systematically study the impact of smoking on T cells in treatment naïve NSCLC patients. We defined a set of smoking-induced differentially expressed genes (SIDEGs) in different cells in TIME.. Specifically, we defined a smoking-related tumor-specific Treg subset, ADAM12+ CTLA4+ Tregs according to the trajectory analysis and highly express genes in cell adhesion pathways and lipid metabolism. Using independent datasets from treatment naïve patients, we found that the fraction of ADAM12+ CTLA4+ Tregs are significantly increased in patients with smoking history. Moreover, the fraction of ADAM12+ CTLA4+ Tregs are positively correlated with the fraction of exhausted T cells. Additionally, we reconstructed the spatial organization of the tumor immune microenvironment and found that ADAM12+ CTLA4+ Tregs more actively communicate with LAYN+CD8+ exhausted T cells compared with ADAM12−CTLA4+ Tregs. Our data demonstrate that smoking induced a unique subset of tumor-specific activated Tregs which interact with exhausted T cells in the TIME. Our findings not only explained how smoking impact the TIME but also provide new targets and biomarkers for precision immunotherapy of lung cancer.

Background Independence is related to the aging process. Loss of independence is defined as the inability to make decisions and participate in activities of daily living (ADLs). Independence is related to physical, psychological, biological, and socioeconomic factors. An enhanced understanding of older people’s independence trajectories and associated risk factors would enable the develop early intervention strategies. Methods Independence trajectory analysis was performed on patients identified in the Unité de Prévention de Suivi et d’Analyse du Vieillissement (UPSAV) database. UPSAV cohort is a prospective observational study. Participants were 221 community-dwelling persons aged ≥75 years followed for 24 months between July 2011–November 2013 and benefits from a prevention strategy. Data were collected prospectively using a questionnaire. Independence was assessed using the “Functional Autonomy Measurement System (Système de Mesure de l’Autonomie Fonctionnelle (SMAF))”. Group-based trajectory modeling (GBTM) was performed to identify independence trajectories, and the results were compared with those of k-means and hierarchical ascending classifications. A multinomial logistic regression was performed to identify predictive factors of the independence trajectory. Results Three distinct trajectories of independence were identified including a “Stable functional autonomy (SFA) trajectory” (53% of patients), a “Stable then decline functional autonomy decline (SDFA) trajectory” (33% of patients) and a “Constantly functional autonomy decline (CFAD) trajectory” (14% of patients). Not being a member of an association, and previous fall were significantly associated of a SDFA trajectory (P P Conclusions Community-living older persons exhibit distinct independence trajectories and the predictive factors. The evidence from this study suggests that the prevention and screening for the loss of independence of the older adults should be anticipated to maintaining autonomy.

Background and Objectives: This study aimed to assess the effect of neck stabilization exercise on respiratory function in stroke patients through longitudinal observation and determine whether there is a difference in its effect based on the side of paralysis in the patients. It is difficult to observe the amount of change observed in individuals and groups as most intergroup comparison studies only use mean values. To address these shortcomings, this study adopted a hierarchical linear model (HLM) in our trajectory analysis. Materials and Methods: We conducted neck stabilization training three times a week for four weeks in a single group of 21 stroke patients. To evaluate respiratory function, their forced vital capacity (FVC), forced expiratory volume in the first second (FEV1), forced expiration ratio (FEV1/FVC), and peak cough flow (PCF) were measured. Data analysis was performed using HLM 8.0. Results: A significant increase was found in the respiratory function after neck stabilization training (p < 0.05). While neck stabilization training overall was longitudinally effective, the growth rate of respiratory function in left-sided paralytic patients was less than the whole group value. Conversely, the growth rate of respiratory function in right-sided paralytic patients was greater than the whole group value. Conclusions: This study demonstrated that neck stabilization training is longitudinally effective in improving respiratory function in stroke patients. Additionally, the growth rate of respiratory function was greater in patients with right side paralysis than in patients with left side paralysis.

Automatic Identification System (AIS) data have been widely used in many fields, such as collision detection, navigation, and maritime traffic management. Similarity analysis is an important process for most AIS trajectory analysis topics. However, most traditional AIS trajectory similarity analysis methods calculate the distance between trajectory points, which requires complex and time-consuming calculations, often leading to substantial errors when processing AIS trajectory data characterized by substantial differences in length or uneven trajectory points. Therefore, we propose a cell-based similarity analysis method that combines the weight of the direction and k-neighborhood (WDN-SIM). This method quantifies the similarity between trajectories based on the degree of proximity and differences in motion direction. In terms of its effectiveness and efficiency, WDN-SIM outperformed seven traditional methods for trajectory similarity analysis. Particularly, WDN-SIM has a high robustness to noise and can distinguish the similarities between trajectories under complex situations, such as when there are opposing directions of motion, large differences in length, and uneven point distributions.

BackgroundUpdatable understanding of the onset and progression of individuals COVID-19 trajectories underpins pandemic mitigation efforts. In order to identify and characterize individual trajectories, we defined and validated ten COVID-19 phenotypes from linked electronic health records (EHR) on a nationwide scale using an extensible framework.MethodsCohort study of 56.6 million people in England alive on 23/01/2020, followed until 31/05/2021, using eight linked national datasets spanning COVID-19 testing, vaccination, primary & secondary care and death registrations data. We defined ten COVID-19 phenotypes reflecting clinically relevant stages of disease severity using a combination of international clinical terminologies (e.g. SNOMED-CT, ICD-10) and bespoke data fields; positive test, primary care diagnosis, hospitalisation, critical care (four phenotypes), and death (three phenotypes). Using these phenotypes, we constructed patient trajectories illustrating the transition frequency and duration between phenotypes. Analyses were stratified by pandemic waves and vaccination status.FindingsWe identified 3,469,528 infected individuals (6.1%) with 8,825,738 recorded COVID-19 phenotypes. Of these, 364,260 (11%) were hospitalised and 140,908 (4%) died. Of those hospitalised, 38,072 (10%) were admitted to intensive care (ICU), 54,026 (15%) received non-invasive ventilation and 21,404 (6%) invasive ventilation. Amongst hospitalised patients, first wave mortality (30%) was higher than the second (23%) in non-ICU settings, but remained unchanged for ICU patients. The highest mortality was for patients receiving critical care outside of ICU in wave 1 (51%). 13,083 (9%) COVID-19 related deaths occurred without diagnoses on the death certificate, but within 30 days of a positive test while 10,403 (7%) of cases were identified from mortality data alone with no prior phenotypes recorded. We observed longer patient trajectories in the second pandemic wave compared to the first.InterpretationOur analyses illustrate the wide spectrum of severity that COVID-19 displays and significant differences in incidence, survival and pathways across pandemic waves. We provide an adaptable framework to answer questions of clinical and policy relevance; new variant impact, booster dose efficacy and a way of maximising existing data to understand individuals progression through disease states.Research in ContextEvidence before the studyWe searched PubMed on October 14, 2021, for publications with the terms “COVID-19” or “SARS-CoV-2”, “severity”, and “electronic health records” or “EHR” without date or language restrictions. Multiple studies explore factors associated with severity of COVID-19 infection, and model predictions of outcome for hospitalised patients. However, most work to date focused on isolated facets of the healthcare system, such as primary or secondary care only, was conducted in subpopulations (e.g. hospitalised patients) of limited sample size, and often utilized dichotomised outcomes (e.g. mortality or hospitalisation) ignoring the full spectrum of disease. We identified no studies which comprehensively detailed severity of infections while describing disease severity across pandemic waves, vaccination status, and patient trajectories.Added value of this studyTo our knowledge, this is the first study providing a comprehensive view of COVID-19 across pandemic waves using national data and focusing on severity, vaccination, and patient trajectories. Drawing on linked electronic health record (EHR) data on a national scale (56.6 million people alive and registered with GP in England), we describe key demographic factors, frequency of comorbidities, impact of the two main waves in England, and effect of full vaccination on COVID-19 severities. Additionally, we identify and describe patient trajectory networks which illustrate the main transition pathways of COVID-19 patients in the healthcare system. Finally, we provide reproducible COVID-19 phenotyping algorithms reflecting clinically relevant stages of disease severity i.e. positive tests, primary care diagnoses, hospitalisation, critical care treatments (e.g. ventilatory support) and mortality.Implications of all the available evidenceThe COVID-19 phenotypes and trajectory analysis framework outlined produce a reproducible, extensible and repurposable means to generate national-scale data to support critical policy decision making. By modelling patient trajectories as a series of interactions with healthcare systems, and linking these to demographic and outcome data, we provide a means to identify and prioritise care pathways associated with adverse outcomes and highlight healthcare system ‘touch points’ which may act as tangible targets for intervention.

Background The aim of this nationwide cohort study was to examine the course of symptoms and trajectories of health-related quality of life (HR-QoL) and psychological distress during follow-up and to identify vulnerable patients. Methods Patients with pathological stage I–III colorectal cancer (CRC) between 2013 and 2018 were included. Baseline characteristics were collected from the Netherlands Cancer Registry, and patients completed the European Organisation for Research and Treatment of Cancer QLQ-C30/CR29, Hospital Anxiety and Depression Scale and low anterior resection syndrome (LARS) questionnaires at the baseline and subsequently at 3, 6, 12, 18 and 24 months. Latent class growth and multinomial logistic regression analyses were performed to outline 24-month trajectories in HR-QoL and distress and to identify predictive factors. Results : A total of 1535 patients with colon cancer or rectal cancer were included. Trajectory analysis of HR-QoL identified three patient classes: high HR-QoL (62.7%), improving HR-QoL (29.0%) and low HR-QoL (8.3%). The following patient groups were identified with having low distress (64.0%), moderate distress (26.9%) and high distress (9.1%). Around 13% of the total cohort had either persistent low HR-QoL or high psychological distress throughout follow-up. Patients belonging to this vulnerable group were significantly more likely to be female, to be younger aged, to have lower education, to have disease stage II–III or to have major LARS. Conclusions Although most patients treated for stage I–III CRC fared well, a small but significant proportion of around 13% did not recover during follow-up and reported low HR-QoL and/or high psychological distress levels throughout. This study's findings should be taken into account when organising and selecting patients for tailored follow-up.

Video-based trajectory analysis might be rather well discussed in sports, such as soccer or basketball, but in cycling, this is far less common. In this paper, a video processing pipeline to extract riding lines in cyclocross races is presented. The pipeline consists of a stepwise analysis process to extract riding behavior from a region (i.e., the fence) in a video camera feed. In the first step, the riders are identified by an Alphapose skeleton detector and tracked with a spatiotemporally aware pose tracker. Next, each detected pose is enriched with additional meta-information, such as rider modus (e.g., sitting on the saddle or standing on the pedals) and detected team (based on the worn jerseys). Finally, a post-processor brings all the information together and proposes ride lines with meta-information for the riders in the fence. The presented methodology can provide interesting insights, such as intra-athlete ride line clustering, anomaly detection, and detailed breakdowns of riding and running durations within the segment. Such detailed rider info can be very valuable for performance analysis, storytelling, and automatic summarization.

Background Supporting the ability of women living with HIV (WLWH) to avoid unintended pregnancy during the postpartum period decreases the number of new pediatric HIV infections, reduces pregnancy-related morbidity and mortality, and is a cost-effective strategy for the elimination of mother-to-child transmission. However, little is currently known about the contraceptive intentions and experiences of reinitiating family planning use among mothers living with HIV as they transition from pregnancy into postpartum. Study objectives To (1) understand the contraceptive trajectories of women living with HIV during pregnancy and postpartum in Cape Town, South Africa, and (2) identify factors shaping differing contraceptive trajectories during the postpartum period. Methods Thirty pregnant WLWH were interviewed during their eighth month of pregnancy and completed follow-up interviews at 6-8 weeks and 9-12 months postpartum (n = 81 total interviews). Interview topics included postpartum contraception intentions, contraceptive use, and experiences accessing family planning services. Trajectory analysis of contraceptive intentions was applied after initial thematic coding. Results While nearly half of women interviewed during pregnancy expressed an intention to utilize a non-injectable contraceptive option after childbirth (e.g. implant, IUD, sterilization, oral contraceptive pills), all women interviewed at one year postpartum had received at least one injection. Three main contraceptive trajectories were identified. (1) realization of contraceptive intentions postpartum; (2) unrealized contraceptive intentions postpartum; and (3) change in contraceptive intention over time. Provider influence, coordination of services, and low contraceptive inventory were identified as potential factors shaping the contraceptive trajectories of participants enrolled in the study. Conclusion Disparities between contraceptive method intentions articulated by WLWH during pregnancy and methods attained postpartum suggest that significant barriers remain for women who are unsatisfied with injectable contraception. Failing to provide postpartum mothers living with HIV their intended family planning method undermines efforts to prevent unintended pregnancy, a key pillar of elimination of mother-to-child transmission.

Natural killer (NK) cells are innate lymphoid cells that mediate anti-tumor responses and exhibit innate memory following stimulation with IL-12, IL-15, and IL-18, thereby differentiating into cytokine-induced memory-like (ML) NK cells. ML NK cells have well-described enhanced anti-tumor properties; however, the molecular mechanisms underlying their enhanced functionality are not well-understood. Initial reports of allogeneic donor ML NK cellular therapy for relapsed/refractory (rel/ref) acute myeloid leukemia (AML) demonstrated safety and a 47% CR/CRi rate (PMID32826231). In this setting, allogeneic ML NK cells are rejected after 3 weeks by recipient T cells, which precludes long-term evaluation of their biology. To address this limitation, we conducted a clinical trial for rel/ref AML patients that added adoptive transfer of same-donor ML NK cells on day +7 of a reduced-intensity conditioning (RIC) MHC-haploidentical HCT, followed by 4 doses of IL-15 (N-803) over 2 weeks (NCT02782546). Since the ML NK cells are from the HCT donor, they are not rejected, but remain MHC-haploidentical to the patient leukemia. Using samples from these patients, we profiled the single cell transcriptomes of NK cells using multidimensional CITE-seq, combining scRNAseq with a custom NK panel of antibodies. To identify donor ML NK cells in an unbiased fashion, we developed a CITE-seq ML NK classifier from in vitro differentiated paired conventional NK (cNK) and ML NK cells. This classifier was applied via transfer learning to CITE-seq analyzed samples from the donor (cNK cells) and patients at days +28 and +60. This approach identified 28-40% of NK cells as ML at Day +28 post-HCT. Only 1-6% of donor peripheral blood NK cells and 4-7% of NK cells in comparator leukemia patients at day +28 after conventional haplo-HCT alone were identified as ML NK cells (Fig 1A). These ML NK cells had a cell surface receptor profile analogous to a previously reported mass cytometry phenotype. Within the CITE-seq data, ML NK cells expressed a transcriptional profile consistent with enhanced functionality (GZMK, GZMA, GNLY), secreted proteins (LTB, CKLF), a distinct adhesome, and evidence of prior activation (MHC Class II and interferon-inducible genes). ML NK cells had a unique NK receptor repertoire including increased KIR2DL4, KLRC1(NKG2A), CD300A, NCAM1(CD56) , and CD2 with decreased expression of the inhibitory receptor KLRB1(CD161). Furthermore, ML NK cells upregulated HOPX, a transcription factor implicated in memory T cells and murine CMV adaptive NK cells. Additionally, ML NK cells downregulated transcription factors related to terminal maturation (ZEB2) and exhaustion (NR4A2). We next sought to identify changes during ML differentiation in patients post-HCT from day +28 to +60 post-HCT. Trajectory analysis identified a ML NK cell state distinct from cNK cells that was present at least 60 days post-HCT (Fig 1B). The ML transcriptional phenotype continued to modulate during late differentiation, including downregulation of GZMK and NCAM1, and upregulation of maturation related transcription factors, while maintaining high expression of HOPX. ML NK cells retained their enhanced functionality during in vivo differentiation, as patient ML NK cells had significantly increased IFNγ production compared to cNK cells after restimulation with leukemia targets or cytokines using mass cytometry (Fig. 2). Subsequently, we confirmed the ML CITE-seq profile in an independent clinical trial treating pediatric AML relapsed after allogenic HCT with same-donor ML NK cells (NCT03068819). In this setting, ML NK cells expressed a similar transcriptional signature and persisted for at least 2 months in the absence of exogenous cytokine support. Thus, ML NK cells possess a distinct transcriptional and surface proteomic profile and undergo in vivo differentiation while persisting within patients for at least 2 months. These findings reveal novel and unique aspects of the ML NK cell molecular program, as well as their prolonged functional persistence in vivo in patients, assisting in future clinical trial design. Figure 1 Figure 1. Disclosures Foltz: Kiadis: Patents & Royalties: TGFbeta expanded NK cells; EMD Millipore: Other: canine antibody licensing fees. Berrien-Elliott: Wugen: Consultancy, Patents & Royalties: 017001-PRO1, Research Funding. Bednarski: Horizon Pharmaceuticals: Membership on an entity's Board of Directors or advisory committees. Fehniger: Wugen: Consultancy, Current equity holder in publicly-traded company, Patents & Royalties: related to memory like NK cells, Research Funding; ImmunityBio: Research Funding; Kiadis: Other; Affimed: Research Funding; Compass Therapeutics: Research Funding; HCW Biologics: Research Funding; OrcaBio: Other; Indapta: Other.