Results on mining NHANES data: a case study in evidence-based medicine

Assessing Rural Health Coalitions Using the Public Health Logic Model: A Systematic Review

CONTEXT

Rural communities face unique challenges including fewer healthcare providers and restricted access to nutritious foods, likely leading to poor health outcomes. Community health coalitions are groups of local organizations partnering to address local health needs. Employing such coalitions is one strategy for implementing policy-system-environment changes for improving rural health. However, their success is variable without standardized evaluation. In this review, rural community health coalitions were retrospectively assessed using the W.K. Kellogg Foundation Logic Model. Community health coalition-reported pathways through this model were explored using market basket analysis.

EVIDENCE ACQUISITION

During Spring 2018, PubMed, Web of Science, ScienceDirect, CINAHL, and PsycINFO were searched for (coalition) AND (rural) AND (health) AND (effectiveness OR impact OR outcome OR logic model). Full-text, peer-reviewed, English articles meeting PICOS criteria (Population, rural communities; Intervention, presence of a community health coalition; Comparator, the coalition over time; Outcomes, logic model pathways) were reviewed. During Summer and Fall 2018, coalition-reported pathways were categorized according to logic model inputs and resources; internal and external activities; outputs; short-, medium-, and long-term outcomes; and impact. Market basket analysis was conducted during Winter 2018.

EVIDENCE SYNTHESIS

The 10 most frequently reported pathway items were partner diversity; organizational structures; implementing pilot studies, programs, and interventions; funding; community engagement and outreach; university partners; holding regular meetings; having working groups and subcommittees; operating under or partnering with a regional research initiative; and conducting a community health and needs assessment. Half of community health coalitions reported 4 or more of the following: funding; partner diversity; university partners; organizational structures; community engagement and outreach; and implementing pilot studies, programs, and interventions.

CONCLUSIONS

Many rural community health coalitions reported inputs and capacity building; few impacted health. Recommending common early phase logic model pathways may facilitate downstream success.

Predicting and Weighting the Factors Affecting Workers' Hearing Loss Based on Audiometric Data Using C5 Algorithm

Introduction:

With the extensively spread of industrialization in the world, noise exposure is becoming more prevalent in the industrial settings. The most important and definite harmful effects of sound include hearing loss, both permanent and temporary.

Objective:

This study was designed aimed to use the C5 algorithm to determine the weight of factors affecting the workers’ hearing loss based on the audiometric data.

Methods:

This cross-sectional, descriptive, analytical study was conducted in 2018 in a mining industry in southeastern Iran. In this study, workers were divided into three exposed groups with different sound pressure levels (one control group and two case groups). Audiometry was conducted for each group of 50 persons; hence, the total number of subjects was 150. The stages of this study include: 1) selecting factors (predictive) to check and weigh them; 2) conducting the audiometry for both ears; 3) calculating the permanent hearing loss in each ear and permanent hearing loss of both ears; 4) classifying the types of hearing loss; and 5) investigating and determining the weight of factors affecting the hearing loss and their classification based on the C5 algorithm and determining the error and accuracy rate of each model. To assess and determine the factors affecting the hearing loss of workers, the C5 algorithm and IBM SPSS Modeler 18.0 were used. SPSS V.18 was used to analyze the linear regression and paired t-test tests, too.

Results:

The results showed that in the first model (SPL <70 dBA), the 8KHz frequency with the weight of 31% had the highest effect, the factors of work experience and the frequency of 250Hz each with the weight of 3%, had the least effect, and the accuracy of the model was 100%. In the second model (SPL 70–80 dBA) the frequency of 8KHz with the weight of 21% had the highest effect, the frequency of 250Hz and the working experience each had the lowest effect with the weight of 7% and the accuracy of the model was calculated as 100%. In the third model (SPL >85 dBA), the 4KHz frequency with the weight of 31% had the highest effect, and the work experience with a weight of 1% had the lowest effect, and the accuracy of the model was 94%. In the fourth model, the 4KHz frequency with the weight of 22% had the highest effect and 250Hz and age each with the weight of 8% had the lowest effects; the accuracy of this model was calculated to be 99.05%.

Conclusions:

During investigating and determining the weight of the factors affecting hearing loss by the C5 algorithm, the high weight and effect of the 4KHz frequency were predicted in hearing loss changes. Considering the high accuracy obtained in this modeling, this algorithm is a suitable and powerful tool for predicting and modeling the hearing loss.

Stacked classifiers for individualized prediction of glycemic control following initiation of metformin therapy in type 2 diabetes

OBJECTIVE

Metformin is the preferred first-line medication for management of type 2 diabetes and prediabetes. However, over a third of patients experience primary or secondary therapeutic failure. We developed machine learning models to predict which patients initially prescribed metformin will achieve and maintain control of their blood glucose after one year of therapy.

MATERIALS AND METHODS

We performed a retrospective analysis of administrative claims data for 12,147 commercially-insured adults and Medicare Advantage beneficiaries with prediabetes or diabetes. Several machine learning models were trained using variables available at the time of metformin initiation to predict achievement and maintenance of hemoglobin A_1c (HbA_1c) < 7.0% after one year of therapy.

RESULTS

AUC performances based on five-fold cross-validation ranged from 0.58 to 0.75. The most influential variables driving the predictions were baseline HbA_1c, starting metformin dosage, and presence of diabetes with complications.

CONCLUSIONS

Machine learning models can effectively predict primary or secondary metformin treatment failure within one year. This information can help identify effective individualized treatment strategies. Most of the implemented models outperformed traditional logistic regression, highlighting the potential for applying machine learning to problems in medicine.

Machine Learning and Data Mining Methods in Diabetes Research

The remarkable advances in biotechnology and health sciences have led to a significant production of data, such as high throughput genetic data and clinical information, generated from large Electronic Health Records (EHRs). To this end, application of machine learning and data mining methods in biosciences is presently, more than ever before, vital and indispensable in efforts to transform intelligently all available information into valuable knowledge. Diabetes mellitus (DM) is defined as a group of metabolic disorders exerting significant pressure on human health worldwide. Extensive research in all aspects of diabetes (diagnosis, etiopathophysiology, therapy, etc.) has led to the generation of huge amounts of data. The aim of the present study is to conduct a systematic review of the applications of machine learning, data mining techniques and tools in the field of diabetes research with respect to a) Prediction and Diagnosis, b) Diabetic Complications, c) Genetic Background and Environment, and e) Health Care and Management with the first category appearing to be the most popular. A wide range of machine learning algorithms were employed. In general, 85% of those used were characterized by supervised learning approaches and 15% by unsupervised ones, and more specifically, association rules. Support vector machines (SVM) arise as the most successful and widely used algorithm. Concerning the type of data, clinical datasets were mainly used. The title applications in the selected articles project the usefulness of extracting valuable knowledge leading to new hypotheses targeting deeper understanding and further investigation in DM.

Integrating and mining diverse data in human immunological studies

Bioanalysts and immunologists can interrogate the immune system with a variety of high-throughput technologies such as gene expression, multiplex bead arrays and flow cytometry. Conceptually, these assays support systems immunology studies, in which phenomena can be measured and correlated across biological compartments. First, however, the resulting high-dimensional data must be combined in a consistent fashion that supports analysis of the data as an integrated whole. Next, analytical methods must be applied to the hundreds or thousands of readouts. We recommend the use of a four-part analytical pipeline, consisting of data integration, hypothesis generation, prediction and hypothesis testing, and validation. We describe a variety of established methods appropriate for these integrated datasets, and highlight their application to human immunological studies. Our goal is to provide bioanalysts, immunologists and data analysts with a valuable perspective with which to approach the multiassay high-dimensional datasets generated by contemporary immunological studies.