Matching With Doses in an Observational Study of a Media Campaign Against Drug Abuse

Multivariate matching with doses of treatment differs from the treatment-control matching in three ways. First, pairs must not only balance covariates, but also must differ markedly in dose. Second, any two subjects may be paired, so that the matching is nonbipartite, and different algorithms are required. Finally, a propensity score with doses must be used in place of the conventional propensity score. We illustrate multivariate matching with doses using pilot data from a media campaign against drug abuse. The media campaign is intended to change attitudes and intentions related to illegal drugs, and the evaluation compares stated intentions among ostensibly comparable teens who reported markedly different exposures to the media campaign.

Healthcare utilization after a first hospitalization for COPD: a new approach of State Sequence Analysis based on the '6W' multidimensional model of care trajectories

Background

Published methods to describe and visualize Care Trajectories (CTs) as patterns of healthcare use are very sparse, often incomplete, and not intuitive for non-experts.

Our objectives are to propose a typology of CTs one year after a first hospitalization for Chronic Obstructive Pulmonary Disease (COPD), and describe CT types and compare patients’ characteristics for each CT type.

Methods

This is an observational cohort study extracted from Quebec’s medico-administrative data of patients aged 40 to 84 years hospitalized for COPD in 2013 (index date). The cohort included patients hospitalized for the first time over a 3-year period before the index date and who survived over the follow-up period. The CTs consisted of sequences of healthcare use (e.g. ED-hospital-home-GP-respiratory therapists, etc.) over a one-year period. The main variable was a CT typology, which was generated by a ‘tailored’ multidimensional State Sequence Analysis, based on the “6W” model of Care Trajectories. Three dimensions were considered: the care setting (“where”), the reason for consultation (“why”), and the speciality of care providers (“which”). Patients were grouped into specific CT types, which were compared in terms of care use attributes and patients’ characteristics using the usual descriptive statistics.

Results

The 2581 patients were grouped into five distinct and homogeneous CT types: Type 1 (n = 1351, 52.3%) and Type 2 (n = 748, 29.0%) with low healthcare and moderate healthcare use respectively; Type 3 (n = 216, 8.4%) with high healthcare use, mainly for respiratory reasons, with the highest number of urgent in-hospital days, seen by pulmonologists and respiratory therapists at primary care settings; Type 4 (n = 100, 3.9%) with high healthcare use, mainly cardiovascular, high ED visits, and mostly seen by nurses in community-based primary care; Type 5 (n = 166, 6.4%) with high healthcare use, high ED visits and non-urgent hospitalisations, and with consultations at outpatient clinics and primary care settings, mainly for other reasons than respiratory or cardiovascular. Patients in the 3 highest utilization CT types were older, and had more comorbidities and more severe condition at index hospitalization.

Conclusions

The proposed method allows for a better representation of the sequences of healthcare use in the real world, supporting data-driven decision making.

Fast and optimal algorithm for case-control matching using registry data: application on the antibiotics use of colorectal cancer patients

Background

In case-control studies most algorithms allow the controls to be sampled several times, which is not always optimal. If many controls are available and adjustment for several covariates is necessary, matching without replacement might increase statistical efficiency. Comparing similar units when having observational data is of utter importance, since confounding and selection bias is present. The aim was twofold, firstly to create a method that accommodates the option that a control is not resampled, and second, to display several scenarios that identify changes of Odds Ratios (ORs) while increasing the balance of the matched sample.

Methods

The algorithm was derived in an iterative way starting from the pre-processing steps to derive the data until its application in a study to investigate the risk of antibiotics on colorectal cancer in the INTEGO registry (Flanders, Belgium). Different scenarios were developed to investigate the fluctuation of ORs using the combination of exact and varying variables with or without replacement of controls. To achieve balance in the population, we introduced the Comorbidity Index (CI) variable, which is the sum of chronic diseases as a means to have comparable units for drawing valid associations.

Results

This algorithm is fast and optimal. We simulated data and demonstrated that the run-time of matching even with millions of patients is minimal. Optimal, since the closest controls is always captured (using the appropriate ordering and by creating some auxiliary variables), and in the scenario that a case has only one control, we assure that this control will be matched to this case, thus maximizing the cases to be used in the analysis. In total, 72 different scenarios were displayed indicating the fluctuation of ORs, and revealing patterns, especially a drop when balancing the population.

Conclusions

We created an optimal and computationally efficient algorithm to derive a matched case-control sample with and without replacement of controls. The code and the functions are publicly available as an open source in an R package. Finally, we emphasize the importance of displaying several scenarios and assess the difference of ORs while using an index to balance population in observational data.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12874-021-01256-3.

Trajectories of vision in older people: the role of age and social position

Visual impairment becomes more prevalent with age. Rather than a uniform decline in vision with age, the strength and direction of change varies between people. This study applies an analytical method that posits multiple trajectories in eyesight, allowing for a more specific description of developmental course of this health outcome and its relationship with social position. The analysis uses the responses of 2956 respondents, aged 60 years and over, followed over 8 years (five observations) as part of the English longitudinal study of ageing. At each observation respondents self-reported their general vision. Optimal matching (sequence analysis), hierarchical clustering, and multinomial logistic regression were used to describe the sequential data, produce a typology of vision trajectories, and examine the socio-demographic characteristics associated with different trajectories. Eight distinctive clusters of trajectories were identified. The probability of reporting different types of vision trajectory varies with a change in age; however, the magnitude of the age effect is associated with social position. Visual impairment in older people is an increasingly relevant area for policy focus, with the rapid growth and diversity of the older population. Identifying factors underpinning the patterning of changes in visual function is essential for developing evidence-based policy, which both meets the needs of those most at risk and increases cost-effectiveness of public health interventions.

Repeated randomization and matching in multi-arm trials

Cluster randomized trials with relatively few clusters have been widely used in recent years for evaluation of health-care strategies. The balance match weighted (BMW) design, introduced in Xu and Kalbfleisch (2010, Biometrics 66, 813-823), applies the optimal full matching with constraints technique to a prospective randomized design with the aim of minimizing the mean squared error (MSE) of the treatment effect estimator. This is accomplished through consideration of M independent randomizations of the experimental units and then selecting the one which provides the most balance evaluated by matching on the estimated propensity scores. Often in practice, clinical trials may involve more than two treatment arms and multiple treatment options need to be evaluated. Therefore, we consider extensions of the BMW propensity score matching method to allow for studies with three or more arms. In this article, we propose three approaches to extend the BMW design to clinical trials with more than two arms and evaluate the property of the extended design in simulation studies.

Explaining patterns in the school-to-work transition: An analysis using optimal matching

This paper studies the school to work transition in the UK with the aim of achieving a richer understanding of individuals' trajectories in the five years after reaching school leaving age. By applying the technique of 'optimal matching' on data from 1991 to 2008, we group individuals' trajectories post-16, and identify a small number of distinct transition patterns. Our results suggest that while 9 out of 10 young people have generally positive experiences post-16, the remaining individuals exhibit a variety of histories that might warrant policy attention. We assess the extent to which characteristics at age 16 can predict which type of trajectory a young person will follow. Our analysis shows that, despite the apparent heterogeneity, virtually all at-risk trajectories are associated with a relatively small set of key 'risk factors': early pregnancy; low educational attainment and self-confidence; and disadvantaged family background. These characteristics are known to be strongly correlated across individuals and raise concerns about the degree of socio-economic polarisation in the transition from school to work.

Evaluating long-term effects of a psychiatric treatment using instrumental variable and matching approaches

Evaluating treatment effects in non-randomized studies is challenging due to the potential unmeasured confounding and complex form of observed confounding. Propensity score based approaches, such as matching or weighting, are commonly used to handle observed confounding variables. The instrumental variable (IV) method is known to guard against unmeasured confounding if a good instrument can be identified. We propose to combine both methods to estimate the long-term treatment effect in a longitudinal psychiatric study. The NIMH collaborative Multi-site Treatment study of children with Attention-deficit/hyperactivity disorder (ADHD) compared different treatment strategies for children diagnosed with ADHD (known as MTA study). The first 14 months is a randomized study and the participants are allowed to choose their desired treatment strategies afterwards. Follow-up measurements are taken at 24, 36 and 72 months. Randomization is often considered as a good instrument since it is not associated with any covariate, observed or unobserved. We first apply a randomization based IV method to estimate the self-selected medication effect on outcome at the end of 72 months. However this approach yields results with huge standard errors due to randomization's weak relationship with later treatment selection. We then consider the self-selection right after the randomization as an instrument, because it is associated with later treatment selection and it is unlikely to affect the outcome directly given the five-year time lapse. To better control the confounding due to observed factors, propensity score matching is used to create a subpopulation with comparable covariate distributions across different self-selected treatments. Using MTA data, matching-enhanced IV estimation yields the most sensible result, while other estimation strategies tend to imply a spurious significant effect. Also, our simulation study shows that the matching-enhanced IV estimation outperforms non-matched methods in terms of relative bias.

Neonatal creatinemia trends as biomarker of subsequent cognitive outcome in extremely low birth weight neonates

BACKGROUND AND AIMS

Serum creatinine is traditionally used as a marker of renal function in neonates and relates to gestational age and disease severity in extremely low birth weight (ELBW) infants. Creatinine is commonly used as a biomarker for early morbidity, but we aim to compare postnatal creatinemia trends as a biomarker for subsequent cognitive outcome. We hypothesize that impaired microcirculation not only in the kidney, but also in general (i.e. brain development) can explain this possible link.

STUDY DESIGN AND OUTCOME MEASURES

A cohort of ELBW infants was analyzed by Bayley Scales of Infant Development (BSID-II) at the corrected age of 2years old. Besides other perinatal indicators, neonatal creatinemia trends of survivors (n=140) and BSID scores (n=96) are compared and analyzed using optimal matching analysis. Hierarchical clustering analysis is applied to identify createnimia trends.

RESULTS

Four different creatinemia trends were identified (persistently high, normal, low, high but normalizing). A low creatinemia trend is significantly associated with the lowest percentages of postnatal corticosteroids, NSAIDS and intraventricular hemorrhage (p=0.005, p=0.013 and p=0.041 respectively) compared to a normal or persistently high creatinemia trend and associated with the best cognitive outcome (+13 points compared to the mean creatinemia trend and +23 points compared to a persistently high creatinemia trend).

CONCLUSIONS

Creatinemia trends after birth are not only useful to predict renal function, but are also associated with cognitive outcome in extremely low birth weight infants. Neonates who have low creatinemia trends after birth, have the highest BSID scores at the age of two years old.

Applying sequence analysis to uncover 'real-world' clinical pathways from routinely collected data: a systematic review

OBJECTIVES

This systematic review aims to elucidate the methodological practices and reporting standards associated with sequence analysis (SA) for the identification of clinical pathways in real-world scenarios, using routinely collected data.

STUDY DESIGN AND SETTING

We conducted a methodological systematic review, searching five medical and health databases: MEDLINE, PsycINFO, CINAHL, EMBASE and Web of Science. The search encompassed articles from the inception of these databases up to February 28, 2023. The search strategy comprised two distinctive sets of search terms, specifically focused on sequence analysis and clinical pathways.

RESULTS

19 studies met the eligibility criteria for this systematic review. Nearly 60% of the included studies were published in or after 2021, with a significant proportion originating from Canada (n = 7) and France (n = 5). 90% of the studies adhered to the fundamental SA steps. The optimal matching (OM) method emerged as the most frequently employed dissimilarity measure (63%), while agglomerative hierarchical clustering using Ward's linkage was the preferred clustering algorithm (53%). However, it is imperative to underline that a majority of the studies inadequately reported key methodological decisions pertaining to SA.

CONCLUSION

This review underscores the necessity for enhanced transparency in reporting both data management procedures and key methodological choices within SA processes. The development of reporting guidelines and a robust appraisal tool tailored to assess the quality of SA would be invaluable for researchers in this field.

Clinical care pathways of patients with biliary tract cancer: A French nationwide longitudinal cohort study

BACKGROUND

Although the incidence of BTC is raising, national healthcare strategies to improve care lack. We aimed to explore patient clinical care pathways and strategies to improve biliary tract cancer (BTC) care.

METHODS

We analysed the French National Healthcare database of all BTC inpatients between January 1, 2017 and December 31, 2021. Multinomial logistic regression adjusted odds ratios (aOR) were used to identify healthcare organisation factors that influenced access to curative care both overall and in a longitudinal sensibility analysis using optimal matching and hierarchical ascending classification to detect a subgroup of curative-care patients with a high survival over a two-year period.

RESULTS

A total of 19,825 new BTC patients and three clinical care pathways (CCP) were identified: 'Palliative care' (PC-CCP), 'Non-curative Care' (NCC-CCP) and 'Curative Care' (CC-CCP) involving 7669 (38.7%), 7721 (38.9%) and 4435 (22.4%) patients respectively. Out of 1200 centers involved in BTC treatment, 84%, 11% and 5% were of low- (<15 patients/year), medium- (15-30 patients/year) and high-volume (>30 patients/year) respectively. Among patient, tumor and hospital factors, BTC management in academic (aOR: 2.32; 95%CI: 1.98-2.71), private (2.51; 2.22-2.83), semi-private (2.25; 1.91-2.65) and in high- (2.09; 1.81-2.42) or medium-volume (1.49; 1.33-1.68) centers increased probability to CC-CCP. These results were maintained in a longitudinal cluster of 2363 (53%) CC-CCP patients presenting a higher two-year survival compared with the rest [96.4% (95.1; 97.6) vs. 38.8% (36.3; 41.4), log-rank p < 0.001].

CONCLUSIONS

Among factors subject to healthcare policy improvement, the volume and type of centers managing BTC strongly influenced access to curative care.

Dissecting the structure-property relationship of ceramic membrane with asymmetric multilayer structures for maximizing permselectivity

Robust ceramic membranes presented attractive features of easy cleaning and excellent stability compared to polymeric membranes. Nevertheless, their inherent relationships between the membrane microstructures and separation properties are not completely clear. In this work, we established a quantitative structure-property model using α-Al₂O₃ membrane on account of the theory of filtrated cake to predict the effects of membrane structure-controlled factors (i.e., α-Al₂O₃ particle size and layer thickness) on separation performances (i.e., solute rejection and water permeance). The simulation results show that membrane pore size mainly depends upon α-Al₂O₃ particle size rather than the layer thickness. When the microstructure of top layer in a double-layer asymmetric ceramic membrane is fixed, there exists optimum particle size and layer thickness that constitute the support layer to achieve maximum water permeance. For a triple-layer ceramic membrane, a similar matching relationship exists between top layer and intermediate layer, indicating that the intermediate layer has a vital role in determining water permeance. While the bottom layer has little effect on overall separation property. Finally, the upper-bound tradeoff relationship between permeance and selectivity is further established for the α-Al₂O₃ membrane. This study reveals the structure-property relationship of ceramic membrane and provides insights into performance enhancement.

Citizenship and education trajectories among children of immigrants: A transition-oriented sequence analysis

During recent decades, the educational outcomes of the children of immigrants have been extensively studied, with a growing emphasis on the heterogeneity of the so-called second generation. Yet, the impact of host country citizenship on children's educational outcomes has only received limited attention so far, although children of immigrants do not get automatic birthright citizenship in most European countries. Focusing on the Netherlands, this paper compares educational trajectories among citizen and non-citizen children of immigrants. Register data and sequence analysis are used to map and cluster the trajectories of a full cohort of second-generation students from the start of secondary school. We apply a variant of optimal matching focusing on sequences of transitions, which enables us to uncover different patterns of (im)mobility within a stratified school system better than the standard approach. Multinomial logistic regressions show that students who acquire Dutch citizenship are significantly more likely to follow upward trajectories, taking advantage of the system's flexibility and "back doors". Conversely, not having Dutch citizenship is associated with a higher risk of dropout and school interruptions. These findings are in line with our theoretical expectation that, during the naturalisation process, parents acquire or further develop important resources for navigating a complex educational system such as the Dutch one.

Impact of event encoding and dissimilarity measures on traffic crash characterization based on sequence of events

Crash sequence analysis has been shown in prior studies to be useful for characterizing crashes and identifying safety countermeasures. Sequence analysis is highly domain-specific, but its various techniques have not been evaluated for adaptation to crash sequences. This paper evaluates the impact of encoding and dissimilarity measures on crash sequence analysis and clustering. Sequence data of interstate highway, single-vehicle crashes in the United States, from 2016 to 2018, were studied. Two encoding schemes and five optimal matching based dissimilarity measures were compared by evaluating the sequence clustering results. The five dissimilarity measures were categorized into two groups based on correlations between dissimilarity matrices. The optimal dissimilarity measure and encoding scheme were identified based on the agreements with a benchmark crash categorization. The transition-rate-based, localized optimal matching dissimilarity and consolidated encoding scheme had the highest agreement with the benchmark. Evaluation results indicate that the selection of dissimilarity measure and encoding scheme determines the results of sequence clustering and crash characterization. A dissimilarity measure that considers the relationships between events and domain context tends to perform well in crash sequence clustering. An encoding scheme that consolidates similar events naturally takes domain context into consideration.