Search

Most statistical and machine learning models used for binary data modeling and classification assume that the data are balanced. However, this assumption can lead to poor predictive performance and bias in parameter estimation when there is an imbalance in the data due to the threshold election for the binary classification. To address this challenge, several authors suggest using asymmetric link functions in binary regression, instead of the traditional symmetric functions such as logit or probit, aiming to highlight characteristics that would help the classification task. Therefore, this study aims to introduce new classification functions based on the Lomax distribution (and its variations; including power and reverse versions). The proposed Bayesian functions have proven asymmetry and were implemented in a Stan program into the R workflow. Additionally, these functions showed promising results in real-world data applications, outperforming classical link functions in terms of metrics. For instance, in the first example, comparing the reverse power double Lomax (RPDLomax) with the logit link showed that, regardless of the data imbalance, the RPDLomax model assigns effectively lower mean posterior predictive probabilities to failure and higher probabilities to success (21.4% and 63.7%, respectively), unlike Logistic regression, which does not clearly distinguish between the mean posterior predictive probabilities for these two classes (36.0% and 39.5% for failure and success, respectively). That is, the proposed asymmetric Lomax approach is a competitive model for differentiating binary data classification in imbalanced tasks against the Logistic approach., Competing Interests: The authors have declared that no competing interests exist., (Copyright: © 2024 Reis et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.)

Background: There is growing evidence of the anti-inflammatory effect of the anti-diabetic drug metformin and its use to reduce pain. However, we currently lack studies investigating whether metformin is associated with a reduction in chronic back pain prevalence when considering physical activity levels, body mass index (BMI), and age., Objective: To investigate whether use of metformin is associated with lower levels of reporting of chronic back pain in a large cohort with type 2 diabetes when stratified for physical activity, BMI, and age., Methods: This is a cross-sectional study of 21,889 participants with type 2 diabetes who were drawn from the UK Biobank database. We investigated whether people using metformin reported a higher prevalence of chronic low back pain than those who did not. Type 2 diabetes, chronic back pain, and metformin were self-reported. Participants were stratified according to their physical activity level (low, moderate and high), BMI (normal, overweight, and obese), and age (40 to <50; 50 to < 60; and ≥60 years). Logistic regression models were built for each physical activity level, BMI and age category to investigate the prevalence of chronic back pain amongst those using and not using metformin., Results: Participants who were using metformin and who had low levels of physical activity [OR 0.87, 95%CI 0.78 to 0.96] or who were obese [OR 0.90, 95%CI 0.86 to 0.98] or older [OR 0.85, 95%CI 0.78 to 0.93] had lower odds of reporting chronic back pain than their counterparts., Conclusion: The anti-diabetic drug metformin might reduce prevalence of chronic low back pain in people who are older, overweight, or less active. These findings should be confirmed in studies using a longitudinal design., Competing Interests: The authors have declared no competing interests exist., (Copyright: © 2023 Carvalho-e-Silva et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.)

Learning techniques involve unraveling regression structures, which aim to analyze in a probabilistic frame the associations across variables of interest. Thus, analyzing fraction and/or proportion data may not be adequate with standard regression procedures, since the linear regression models generally assume that the dependent (outcome) variable is normally distributed. In this manner, we propose a statistical model called unit-Lindley regression model, for the purpose of Statistical Process Control (SPC). As a result, a new control chart tool was proposed, which targets the water monitoring dynamic, as well as the monitoring of relative humidity, per minute, of Copiapó city, located in Atacama Desert (one of the driest non-polar places on Earth), north of Chile. Our results show that variables such as wind speed, 24-hour temperature variation, and solar radiation are useful to describe the amount of relative humidity in the air. Additionally, Information Visualization (InfoVis) tools help to understand the time seasonality of the water particle phenomenon of the region in near real-time analysis. The developed methodology also helps to label unusual events, such as Camanchaca, and other water monitoring-related events., Competing Interests: The authors have declared that no competing interests exist.

In this paper, we propose a hierarchical statistical model for a single repairable system subject to several failure modes (competing risks). The paper describes how complex engineered systems may be modelled hierarchically by use of Bayesian methods. It is also assumed that repairs are minimal and each failure mode has a power-law intensity. Our proposed model generalizes another one already presented in the literature and continues the study initiated by us in another published paper. Some properties of the new model are discussed. We conduct statistical inference under an objective Bayesian framework. A simulation study is carried out to investigate the efficiency of the proposed methods. Finally, our methodology is illustrated by two practical situations currently addressed in a project under development arising from a partnership between Petrobras and six research institutes., Competing Interests: The authors have declared that no competing interests exist.

In this paper, from the practical point of view, we focus on modeling traumatic brain injury data considering different stages of hospitalization, related to patients' survival rates following traumatic brain injury caused by traffic accidents. From the statistical point of view, the primary objective is related to overcoming the limited number of traumatic brain injury patients available for studying by considering different estimation methods to obtain improved estimators of the model parameters, which can be recommended to be used in the presence of small samples. To have a general methodology, at least in principle, we consider the very flexible Generalized Gamma distribution. We compare various estimation methods using extensive numerical simulations. The results reveal that the penalized maximum likelihood estimators have the smallest mean square errors and biases, proving to be the most efficient method among the investigated ones, mainly to be used in the presence of small samples. The Simulated Annealing technique is used to avoid numerical problems during the optimization process, as well as the need for good initial values. Overall, we considered an amount of three real data sets related to traumatic brain injury caused by traffic accidents to demonstrate that the Generalized Gamma distribution is a simple alternative to be used in this type of applications for different occurrence rates and risks, and in the presence of small samples., Competing Interests: The authors have declared that no competing interests exist.

Background: Regular physical activity participation is known to promote better mobility and coordination. Although previous research has established that lack of physical activity participation may increase the risk of developing low back pain, the role of physical activity as a predictor of recovery among symptomatic individuals remains unclear., Objective: To evaluate whether: (i) the level of physical activity participation before an acute episode of low back pain predicts recovery within 12 months following the acute pain episode (i.e. index episode); and (ii) participants return to their pre-pain level of physical activity participation 12 months from the index episode., Design: This study used longitudinal data from the TRIGGERS case-crossover study., Setting: This study was conducted through over-the-phone interviews to participants that presented to 300 primary care clinics with a new episode of acute low back pain., Participants: This study included 999 consecutive patients, aged 18 years or older. Follow-up assessment was completed at 12 months following the index episode., Methods: This is a cohort study where consecutive patients, aged 18 years or older, that presented with a new episode of acute low back pain were recruited between October 2011 and November 2012., Main Outcome Measurements: Self-reported level of physical activity participation., Results: A total of 830 participants completed the study. When comparing participants who reported pain at 12 months follow-up with those without pain, all participants reported similar levels of physical activity participation one week before (p = 0.449), one week after (p = 0.812) and 12 months after the index episode (p = 0.233). The level of physical activity participation before the index episode was not a reliable predictor of presence of pain at either 3 or 12 months follow-up (OR 0.99; 95% CI 0.993 to 1.003; p = 0.523 and OR 1; 95% CI 0.992 to 1.008; p = 0.923, respectively)., Conclusion: Physical activity participation did not predict recovery from the pain episode. Also all participants returned to their pre-pain level of physical activity participation after 12 months., Competing Interests: The authors have declared that no competing interests exist.

Background and Objective: Approximately half of the population will experience either low back pain or neck pain, at some point in their lives. Previous studies suggest that people with diabetes are more likely to present with chronic somatic pain, including shoulder, knee and spinal pain. This study aimed to systematically review and appraise the literature to explore the magnitude as well as the nature of the association between diabetes and back, neck, or spinal (back and neck) pain., Databases and Data Treatment: A systematic search was performed using the Medline, CINAHL, EMBASE, and Web of Science electronic databases. Studies which assessed the association between diabetes and back or neck pain outcomes, in participants older than 18 years of age were included. Two independent reviewers extracted data on the incidence of pain and reported associations., Results: Eight studies were included in the meta-analyses. Meta-analyses showed that people with diabetes are more likely to report low back pain [5 studies; n: 131,431; odds ratio (OR): 1.35; 95% Confidence Interval (CI): 1.20 to 1.52; p<0.001] and neck pain (2 studies; n: 6,560; OR: 1.24; 95% CI: 1.05 to 1.47; p = 0.01) compared to those without diabetes. Results from one longitudinal cohort study suggested that diabetes is not associated with the risk of developing future neck, low back or spinal pain., Conclusions: Diabetes is associated with low back and neck individually, and spinal pain. The longitudinal analysis showed no association between the conditions. Our results suggest that diabetes co-exists with back pain; however, a direct causal link between diabetes and back pain was not established., Systematic Review Registration: PROSPERO registration CRD42016050738., Competing Interests: The authors have declared that no competing interests exist.

Objective: To investigate the chronic low back pain and coronary heart disease relationship, after adjusting for relevant confounders, including genetics., Methods: In a cross-sectional design, 2148 twins were recruited from the Murcia Twin Registry, Spain. The exposure was chronic LBP and the outcomes were myocardial infarction and other coronary heart diseases-lifetime and in the last 2 years-based on standardized health-related questionnaires. First, logistic regression analysis investigated associations of the total sample followed by a matched co-twin control analyses, with all complete twin pairs discordant for chronic LBP utilised, separated for zygosity-dizygotic (DZ) and monozygotic (MZ) pairs, which adjusted for shared familial factors, including genetics., Results: Chronic LBP pain is associated with lifetime myocardial infarction [odds ratio (OR) = 2.69, 95% confidence interval (CI) = 1.35-5.36], other coronary heart diseases over a lifetime (OR = 2.58, 95% CI: 1.69-3.93) and in the last two years (OR = 2.19, 95% CI: 1.33-3.60), while there was a borderline association with myocardial infarction in the last 2 years (OR = 2.64, 95% CI: 0.98-7.12). Although the magnitude of the association remained or increased in the co-twin control analyses, none reached statistical significance., Conclusion: Chronic LBP is associated with a higher prevalence of myocardial infarction and coronary heart disease. It is possible that this association remains even when controlling for genetics and early shared environment, although this should be investigated with larger samples of twins discordant for LBP.

Background: The management of spinal stenosis by surgery has increased rapidly in the past two decades, however, there is still controversy regarding the efficacy of surgery for this condition. Our aim was to investigate the efficacy and comparative effectiveness of surgery in the management of patients with lumbar spinal stenosis., Methods: Electronic searches were performed on MEDLINE, EMBASE, AMED, CINAHL, Web of Science, LILACS and Cochrane Library from inception to November 2014. Hand searches were conducted on included articles and relevant reviews. We included randomised controlled trials evaluating surgery compared to no treatment, placebo/sham, or to another surgical technique in patients with lumbar spinal stenosis. Primary outcome measures were pain, disability, recovery and quality of life. The PEDro scale was used for risk of bias assessment. Data were pooled with a random-effects model, and the GRADE approach was used to summarise conclusions., Results: Nineteen published reports (17 trials) were included. No trials were identified comparing surgery to no treatment or placebo/sham. Pooling revealed that decompression plus fusion is not superior to decompression alone for pain (mean difference -3.7, 95% confidence interval -15.6 to 8.1), disability (mean difference 9.8, 95% confidence interval -9.4 to 28.9), or walking ability (risk ratio 0.9, 95% confidence interval 0.4 to 1.9). Interspinous process spacer devices are slightly more effective than decompression plus fusion for disability (mean difference 5.7, 95% confidence interval 1.3 to 10.0), but they resulted in significantly higher reoperation rates when compared to decompression alone (28% v 7%, P < 0.001). There are no differences in the effectiveness between other surgical techniques for our main outcomes., Conclusions: The relative efficacy of various surgical options for treatment of spinal stenosis remains uncertain. Decompression plus fusion is not more effective than decompression alone. Interspinous process spacer devices result in higher reoperation rates than bony decompression.