Interpretability is becoming increasingly important for predictive model analysis. Unfortunately, as remarked by many authors, there is still no consensus regarding this notion. The goal of this ...paper is to propose the definition of a score that allows for quickly comparing interpretable algorithms. This definition consists of three terms, each one being quantitatively measured with a simple formula: predictivity, stability and simplicity. While predictivity has been extensively studied to measure the accuracy of predictive algorithms, stability is based on the Dice-Sorensen index for comparing two rule sets generated by an algorithm using two independent samples. The simplicity is based on the sum of the lengths of the rules derived from the predictive model. The proposed score is a weighted sum of the three terms mentioned above. We use this score to compare the interpretability of a set of rule-based algorithms and tree-based algorithms for the regression case and for the classification case.
Although guidelines recommend in-person counseling before BRCA1/BRCA2 gene testing, genetic counseling is increasingly offered by telephone. As genomic testing becomes more common, evaluating ...alternative delivery approaches becomes increasingly salient. We tested whether telephone delivery of BRCA1/2 genetic counseling was noninferior to in-person delivery.
Participants (women age 21 to 85 years who did not have newly diagnosed or metastatic cancer and lived within a study site catchment area) were randomly assigned to usual care (UC; n = 334) or telephone counseling (TC; n = 335). UC participants received in-person pre- and post-test counseling; TC participants completed all counseling by telephone. Primary outcomes were knowledge, satisfaction, decision conflict, distress, and quality of life; secondary outcomes were equivalence of BRCA1/2 test uptake and costs of delivering TC versus UC.
TC was noninferior to UC on all primary outcomes. At 2 weeks after pretest counseling, knowledge (d = 0.03; lower bound of 97.5% CI, -0.61), perceived stress (d = -0.12; upper bound of 97.5% CI, 0.21), and satisfaction (d = -0.16; lower bound of 97.5% CI, -0.70) had group differences and confidence intervals that did not cross their 1-point noninferiority limits. Decision conflict (d = 1.1; upper bound of 97.5% CI, 3.3) and cancer distress (d = -1.6; upper bound of 97.5% CI, 0.27) did not cross their 4-point noninferiority limit. Results were comparable at 3 months. TC was not equivalent to UC on BRCA1/2 test uptake (UC, 90.1%; TC, 84.2%). TC yielded cost savings of $114 per patient.
Genetic counseling can be effectively and efficiently delivered via telephone to increase access and decrease costs.
In observational studies, control of confounding can be done in the design and analysis phases. Using examples from large health care database studies, this article provides the clinicians with an ...overview of standard methods in the analysis phase, such as stratification, standardization, multivariable regression analysis and propensity score (PS) methods, together with the more advanced high-dimensional propensity score (HD-PS) method. We describe the progression from simple stratification confined to the inclusion of a few potential confounders to complex modeling procedures such as the HD-PS approach by which hundreds of potential confounders are extracted from large health care databases. Stratification and standardization assist in the understanding of the data at a detailed level, while accounting for potential confounders. Incorporating several potential confounders in the analysis typically implies the choice between multivariable analysis and PS methods. Although PS methods have gained remarkable popularity in recent years, there is an ongoing discussion on the advantages and disadvantages of PS methods as compared to those of multivariable analysis. Furthermore, the HD-PS method, despite its generous inclusion of potential confounders, is also associated with potential pitfalls. All methods are dependent on the assumption of no unknown, unmeasured and residual confounding and suffer from the difficulty of identifying true confounders. Even in large health care databases, insufficient or poor data may contribute to these challenges. The trend in data collection is to compile more fine-grained data on lifestyle and severity of diseases, based on self-reporting and modern technologies. This will surely improve our ability to incorporate relevant confounders or their proxies. However, despite a remarkable development of methods that account for confounding and new data opportunities, confounding will remain a serious issue. Considering the advantages and disadvantages of different methods, we emphasize the importance of the clinical input and of the interplay between clinicians and analysts to ensure a proper analysis.
Background
To determine long‐term quality‐of‐life (QOL) trajectories among breast cancer survivors aged 65+ (older) evaluating the effects of personality and social support.
Methods
Older women ...(N = 1280) newly examined with invasive, nonmetastatic breast cancer completed baseline assessments. Follow‐up data were collected 6 and 12 months later and then annually for up to 7 years (median 4.5 years). Quality of life was assessed using EORTC‐QLQ‐C30 emotional, physical, and cognitive scales. Optimism (Life Orientation Test), Coping (Brief COPE), and social support (Medical Outcomes Study) were assessed at baseline. Group‐based trajectory modeling identified QOL trajectories; multinomial regression evaluated effects of predictors on trajectory groups. Age, education, systemic therapy, comorbidity, and reported precancer function (SF‐12) were considered as controlling variables.
Results
Three trajectories were identified for each QOL domain: “maintained high,” “phase shift” (lower but parallel scores to “maintained high” group), and “accelerated decline” (lowest baseline scores and steepest decline). Accelerated decline in emotional, physical, and cognitive function was seen in 6.9%, 31.8%, and 7.6% of older survivors, respectively. Maladaptive coping and lower social support increased adjusted odds of being in the accelerated decline group for all QOL domains; lower optimism was only related to decline in emotional function. Chemotherapy was related to physical and cognitive but not emotional function trajectories.
Conclusions
Personality and social resources affect the course of long‐term emotional well‐being of older breast cancer survivors; treatment is more important for physical and cognitive than emotional function. Early identification of those vulnerable to deterioration could facilitate clinical and psychological support.
Purpose
Breast cancer patients aged 65+ (“older”) vary in frailty status. We tested whether a deficits accumulation frailty index predicted long-term mortality.
Methods
Older patients (
n
= 1280) ...with non-metastatic, invasive breast cancer were recruited from 78 Alliance sites from 2004 to 2011, with follow-up to 2015. Frailty categories (robust, pre-frail, and frail) were based on 35 baseline illness and function items. Cox proportional hazards and competing risk models were used to calculate all-cause and breast cancer-specific mortality for up to 7 years, respectively. Potential covariates included demographic, psychosocial, and clinical factors, diagnosis year, and care setting.
Results
Patients were 65–91 years old. Most (76.6%) were robust; 18.3% were pre-frail, and 5.1% frail. Robust patients tended to receive more chemotherapy ± hormonal therapy (vs. hormonal) than pre-frail or frail patients (45% vs. 37 and 36%,
p
= 0.06), and had the highest adherence to hormonal therapy. The adjusted hazard ratios for all-cause mortality (
n
= 209 deaths) were 1.7 (95% CI 1.2–2.4) and 2.4 (95% CI 1.5–4.0) for pre-frail and frail versus robust women, respectively, with an absolute mortality difference of 23.5%. The adjusted hazard of breast cancer death (
n
−99) was 3.1 (95% CI 1.6–5.8) times higher for frail versus robust patients (absolute difference of 14%). Treatment differences did not account for the relationships between frailty and mortality.
Conclusions
Most older breast cancer patients are robust and could consider chemotherapy where otherwise indicated. Patients who are frail or pre-frail have elevated long-term all-cause and breast cancer mortality. Frailty indices could be useful for treatment decision-making and care planning with older patients.
The choice of the factorization rank of a matrix is critical, e.g., in dimensionality reduction, filtering, clustering, deconvolution, etc., because selecting a rank that is too high amounts to ...adjusting the noise, while selecting a rank that is too low results in the oversimplification of the signal. Numerous methods for selecting the factorization rank of a non-negative matrix have been proposed. One of them is the cophenetic correlation coefficient (ccc), widely used in data science to evaluate the number of clusters in a hierarchical clustering. In previous work, it was shown that ccc performs better than other methods for rank selection in non-negative matrix factorization (NMF) when the underlying structure of the matrix consists of orthogonal clusters. In this article, we show that using the ratio of ccc to the approximation error significantly improves the accuracy of the rank selection. We also propose a new criterion, concordance, which, like ccc, benefits from the stochastic nature of NMF; its accuracy is also improved by using its ratio-to-error form. Using real and simulated data, we show that concordance, with a CUSUM-based automatic detection algorithm for its original or ratio-to-error forms, significantly outperforms ccc. It is important to note that the new criterion works for a broader class of matrices, where the underlying clusters are not assumed to be orthogonal.
Data augmentation is needed to use Deep Learning methods for the typically small time series datasets. There is limited literature on the evaluation of the performance of the use of Generative ...Adversarial Networks for time series data augmentation. We describe and discuss the results of a pilot study that extends a recent evaluation study of two families of data augmentation methods for time series (i.e., transformation-based methods and pattern-mixing methods), and provide recommendations for future work in this important area of research.
Primary androgen-deprivation therapy (PADT) is often used to treat clinically localized prostate cancer, but its effects on cause-specific and overall mortality have not been established. Given the ...widespread use of PADT and the potential risks of serious adverse effects, accurate mortality data are needed to inform treatment decisions.
We conducted a retrospective cohort study using comprehensive utilization and cancer registry data from three integrated health plans. All men were newly diagnosed with clinically localized prostate cancer. Men who were diagnosed between 1995 and 2008, were not treated with curative intent therapy, and received follow-up through December 2010 were included in the study (n = 15,170). We examined all-cause and prostate cancer-specific mortality as our main outcomes. We used Cox proportional hazards models with and without propensity score analysis.
Overall, PADT was associated with neither a risk of all-cause mortality (hazard ratio HR, 1.04; 95% CI, 0.97 to 1.11) nor prostate-cancer-specific mortality (HR, 1.03; 95% CI, 0.89 to 1.19) after adjusting for all sociodemographic and clinical characteristics. PADT was associated with decreased risk of all-cause mortality but not prostate-cancer-specific mortality. PADT was associated with decreased risk of all-cause mortality only among the subgroup of men with a high risk of cancer progression (HR, 0.88; 95% CI, 0.78 to 0.97).
We found no mortality benefit from PADT compared with no PADT for most men with clinically localized prostate cancer who did not receive curative intent therapy. Men with higher-risk disease may derive a small clinical benefit from PADT. Our study provides the best available contemporary evidence on the lack of survival benefit from PADT for most men with clinically localized prostate cancer.
Abstract
Background
During, shortly after, and sometimes for years after hematopoietic stem cell transplant, a large proportion of hematological cancer patients undergoing transplant report ...significant physical and psychological symptoms and reduced health-related quality of life. To address these survivorship problems, we developed a low-burden, brief psychological intervention called expressive helping that includes two theory- and evidence-based components designed to work together synergistically: emotionally expressive writing and peer support writing. Building on evidence from a prior randomized control trial showing reductions in physical symptoms and distress in long-term transplant survivors with persistent survivorship problems, the Writing for Insight, Strength, and Ease (WISE) trial will evaluate the efficacy of expressive helping when used during transplant and in the early post-transplant period, when symptoms peak, and when intervention could prevent development of persistent symptoms.
Methods
WISE is a multi-site, two-arm randomized controlled efficacy trial. Adult hematological cancer patients scheduled for a hematopoietic stem cell transplant will complete baseline measures and then, after hospitalization but prior to transplant, they will be randomized to complete either expressive helping or a time and attention “neutral writing” task. Both expressive helping and neutral writing involve four brief writing sessions, beginning immediately after randomization and ending approximately 4 weeks after hospital discharge. Measures of symptom burden (primary outcome), distress, health-related quality of life, and fatigue (secondary outcomes) will be administered in seven assessments coinciding with medically relevant time points from baseline and to a year post-intervention.
Discussion
The steady and continuing increase in use of stem cell transplantation has created growing need for efficacious, accessible interventions to reduce the short- and long-term negative physical and psychosocial effects of this challenging but potentially life-saving treatment. Expressive helping is a psychological intervention that was designed to fill this gap. It has been shown to be efficacious in long-term transplant survivors but could have even greater impact if it is capable of reducing symptoms during and soon after transplant. The WISE study will evaluate these benefits in a rigorous randomized controlled trial.
Trial registration
Clinicaltrial.gov
NCT03800758
. Registered January 11, 2019