Managing manifest diseases, but not health risks, saved PepsiCo money over seven years

Estimation of the water footprint of kiwifruit: in the areas transferred from hazelnut to kiwi

Agriculture is the largest consumer of freshwater and plays a critical role in addressing global water scarcity. While numerous studies have focused on the water footprint (WF) of various agricultural products, little attention has been paid to changing cropping patterns and their impact on WF. Here, we investigate the impact of conversion from hazelnut fields to kiwi orchards on green, blue, and gray WF between 2010 and 2021 in Ordu, Turkey. Our results show a total increase of 803,901 tons WF for all green, blue, and gray WF. Compared to the previous situation, changing the agricultural product and growing kiwifruit on previously established hazelnut fields increases green WF by 372,106 tons and blue WF by 334,167 tons. Thus, the change of cultivation pattern could significantly contribute to the water scarcity in the area, and at the same time, the increase in WF. Although kiwi cultivation might be advantageous economically, this economic benefit might be an ecological disadvantage as kiwi production is highly dependent on limited blue water resources. Therefore, it is suggested to further promote the rain-fed product, the hazelnut.

Preliminary results of the implementation of the program to reduce morbidity and prevent mortality from diseases of the circulatory system in workers of locomotive crews

Corporate health programs are a common measure for the primary and secondary prevention of chronic non-communicable diseases.

Aim. To study the first implementation results of a Targeted comprehensive program to reduce morbidity and prevent mortality from circulatory system diseases and early cancer detection in employees of JSC "Russian Railways" for the period from 2019 to 2023.

Material and methods. The study used a survey of employees of locomotive crews (RLC), which was conducted twice: in the summer of 2018 and February-March 2021. The survey was conducted using a specially designed questionnaire that takes into account the health status of drivers and their assistants, production, and non-production risk factors. In 2018, 10476 questionnaires were collected (>7% of employees), and in 2021 — 14403 questionnaires (>10% of employees). The age structure of railways has not changed, which made it possible to analyze the frequency of occurrence of risk factors in dynamics.

Results. In general, the mention of the RLC of the interfering effect of the noise factor, uncomfortable temperature, and undesirable odors in the driver’s cabin decreased for JSC "Russian Railways". The number of smokers on the South-Eastern Railway significantly increased during the study period. The number of people consuming insufficient amounts of vegetables and fruits has increased on the Far Eastern, West Siberian, Krasnoyarsk, and Volga railways. The frequency of workers’ meals at fast food restaurants has increased on the Southeastern Railway. The number of people with a good commitment to the basic principles of a healthy lifestyle has increased on the East Siberian, Trans-Baikal, West Siberian, Kuibyshev, Oktyabrskaya, Sverdlovsk, North Caucasian, and South Ural railways.

Conclusion. The conducted research has shown the effectiveness of the initial stage of the implementation of the corporate program to reduce morbidity and prevent mortality from diseases of the circulatory system in RLC. The heterogeneity of the results for different railways was revealed. 

Error and Timeliness Analysis for Using Machine Learning to Predict Asthma Hospital Visits: Retrospective Cohort Study

Background

Asthma hospital visits, including emergency department visits and inpatient stays, are a significant burden on health care. To leverage preventive care more effectively in managing asthma, we previously employed machine learning and data from the University of Washington Medicine (UWM) to build the world’s most accurate model to forecast which asthma patients will have asthma hospital visits during the following 12 months.

Objective

Currently, two questions remain regarding our model’s performance. First, for a patient who will have asthma hospital visits in the future, how far in advance can our model make an initial identification of risk? Second, if our model erroneously predicts a patient to have asthma hospital visits at the UWM during the following 12 months, how likely will the patient have ≥1 asthma hospital visit somewhere else or ≥1 surrogate indicator of a poor outcome? This work aims to answer these two questions.

Methods

Our patient cohort included every adult asthma patient who received care at the UWM between 2011 and 2018. Using the UWM data, our model made predictions on the asthma patients in 2018. For every such patient with ≥1 asthma hospital visit at the UWM in 2019, we computed the number of days in advance that our model gave an initial warning. For every such patient erroneously predicted to have ≥1 asthma hospital visit at the UWM in 2019, we used PreManage and the UWM data to check whether the patient had ≥1 asthma hospital visit outside of the UWM in 2019 or any surrogate indicators of poor outcomes. Such surrogate indicators included a prescription for systemic corticosteroids during the following 12 months, any type of visit for asthma exacerbation during the following 12 months, and asthma hospital visits between 13 and 24 months later.

Results

Among the 218 asthma patients in 2018 with asthma hospital visits at the UWM in 2019, 61.9% (135/218) were given initial warnings of such visits ≥3 months ahead by our model and 84.4% (184/218) were given initial warnings ≥1 day ahead. Among the 1310 asthma patients in 2018 who were erroneously predicted to have asthma hospital visits at the UWM in 2019, 29.01% (380/1310) had asthma hospital visits outside of the UWM in 2019 or surrogate indicators of poor outcomes.

Conclusions

Our model gave timely risk warnings for most asthma patients with poor outcomes. We found that 29.01% (380/1310) of asthma patients for whom our model gave false-positive predictions had asthma hospital visits somewhere else during the following 12 months or surrogate indicators of poor outcomes, and thus were reasonable candidates for preventive interventions. There is still significant room for improving our model to give more accurate and more timely risk warnings.

International Registered Report Identifier (IRRID)

RR2-10.2196/5039

Adolescents Facing Transmedia Learning: Reflections on What They Can Do, What They Think and What They Feel

The integration of new media literacies and, consequently, strategies such as transmedia learning in the teaching–learning processes has been a topic of interest among various types of national and international institutions and governments. In this sense, the current article deals with the abilities, thoughts and expectations that Italian students in classical high schools have in order to face these new formative changes. For this purpose, a mixed methods approach (qualitative and quantitative) was designed and applied in the context of a classical high school in Cagliari (Italy): a questionnaire on digital skills (N = 128), a set of semi-structured interviews (N = 17) and two focus groups (N = 14). The results obtained allow us to verify that, from the point of view of skills, adolescents are prepared to take on the challenges of transmedia learning (navigation, information management), although their collaboration skills need to be strengthened. On the other hand, from the cognitive and affective points of view, they are positive and enthusiastic about these new possibilities: greater interaction, flexibility, engagement and variety of resources and learning strategies.

A Roadmap for Boosting Model Generalizability for Predicting Hospital Encounters for Asthma

In the United States, ~9% of people have asthma. Each year, asthma incurs high health care cost and many hospital encounters covering 1.8 million emergency room visits and 439,000 hospitalizations. A small percentage of patients with asthma use most health care resources. To improve outcomes and cut resource use, many health care systems use predictive models to prospectively find high-risk patients and enroll them in care management for preventive care. For maximal benefit from costly care management with limited service capacity, only patients at the highest risk should be enrolled. However, prior models built by others miss >50% of true highest-risk patients and mislabel many low-risk patients as high risk, leading to suboptimal care and wasted resources. To address this issue, 3 site-specific models were recently built to predict hospital encounters for asthma, gaining up to >11% better performance. However, these models do not generalize well across sites and patient subgroups, creating 2 gaps before translating these models into clinical use. This paper points out these 2 gaps and outlines 2 corresponding solutions: (1) a new machine learning technique to create cross-site generalizable predictive models to accurately find high-risk patients and (2) a new machine learning technique to automatically raise model performance for poorly performing subgroups while maintaining model performance on other subgroups. This gives a roadmap for future research.

New Media Literacies and Transmedia Learning… Do We Really Have the Conditions to Make the Leap? An Analysis from the Context of Two Italian licei classici

In recent years, the integration of new media literacies and, consequently, strategies such as transmedia learning in teaching–learning processes has been a topic of interest among various types of national and international institutions and governments. In this sense, the current article deals with the abilities and habits of Italian students of licei classici (Italian classical high schools) to cope with these new formative contexts that are arising. For this purpose, different quantitative instruments (from the field of attitudes, digital skills and multitasking, and corresponding to the transmedia sphere) were administered to 400 students (N = 400). The results show that most young people have access to devices and that they prefer the mobile ones when consuming or creating content on the net. Moreover, although they are inclined towards transmedia practices, they have some difficulties in becoming creative agents collaborating and fully participating in digital citizenship.

Using a Constraint-Based Method to Identify Chronic Disease Patients Who Are Apt to Obtain Care Mostly Within a Given Health Care System: Retrospective Cohort Study

BACKGROUND

For several major chronic diseases including asthma, chronic obstructive pulmonary disease, chronic kidney disease, and diabetes, a state-of-the-art method to avert poor outcomes is to use predictive models to identify future high-cost patients for preemptive care management interventions. Frequently, an American patient obtains care from multiple health care systems, each managed by a distinct institution. As the patient's medical data are spread across these health care systems, none has complete medical data for the patient. The task of building models to predict an individual patient's cost is currently thought to be impractical with incomplete data, which limits the use of care management to improve outcomes. Recently, we developed a constraint-based method to identify patients who are apt to obtain care mostly within a given health care system. Our method was shown to work well for the cohort of all adult patients at the University of Washington Medicine for a 6-month follow-up period. It is unknown how well our method works for patients with various chronic diseases and over follow-up periods of different lengths, and subsequently, whether it is reasonable to perform this predictive modeling task on the subset of patients pinpointed by our method.

OBJECTIVE

To understand our method's potential to enable this predictive modeling task on incomplete medical data, this study assesses our method's performance at the University of Washington Medicine on 5 subgroups of adult patients with major chronic diseases and over follow-up periods of 2 different lengths.

METHODS

We used University of Washington Medicine data for all adult patients who obtained care at the University of Washington Medicine in 2018 and PreManage data containing usage information from all hospitals in Washington state in 2019. We evaluated our method's performance over the follow-up periods of 6 months and 12 months on 5 patient subgroups separately-asthma, chronic kidney disease, type 1 diabetes, type 2 diabetes, and chronic obstructive pulmonary disease.

RESULTS

Our method identified 21.81% (3194/14,644) of University of Washington Medicine adult patients with asthma. Around 66.75% (797/1194) and 67.13% (1997/2975) of their emergency department visits and inpatient stays took place within the University of Washington Medicine system in the subsequent 6 months and in the subsequent 12 months, respectively, approximately double the corresponding percentage for all University of Washington Medicine adult patients with asthma. The performance for adult patients with chronic kidney disease, adult patients with chronic obstructive pulmonary disease, adult patients with type 1 diabetes, and adult patients with type 2 diabetes was reasonably similar to that for adult patients with asthma.

CONCLUSIONS

For each of the 5 chronic diseases most relevant to care management, our method can pinpoint a reasonably large subset of patients who are apt to obtain care mostly within the University of Washington Medicine system. This opens the door to building models to predict an individual patient's cost on incomplete data, which was formerly deemed impractical.

INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID)

RR2-10.2196/13783.

Time to death and its associated factors among infants in sub-Saharan Africa using the recent demographic and health surveys: shared frailty survival analysis

BACKGROUND

Globally, approximately 4.1 million infants died, accounting for 75% of all under-five deaths. In sub-Saharan Africa (SSA), infant mortality was 52.7/1000 live births in 2018 This study aimed to assess the pooled estimate of infant mortality rate (IMR), time to death, and its associated factors in SSA using the recent demographic and health survey dataset between 2010 and 2018.

METHODS

Data were retrieved from the standard demographic and health survey datasets among 33 SSA countries. A total of 93,765 samples were included. The data were cleaned using Microsoft Excel and STATA software. Data analysis was done using R and STATA software. Parametric shared frailty survival analysis was employed. Statistical significance was declared as a two-side P-value < 0.05.

RESULTS

The pooled estimate of IMR in SSA was 51 per 1000 live births (95% Confidence Interval (CI): 46.65-55.21). The pooled estimate of the IMR was 53 in Central, 44 in Eastern, 44 in Southern, and 57 in Western Africa per 1000 live births. The cumulative survival probability at the end of 1 year was 56%. Multiple births (Adjusted Hazard ratio (AHR) = 2.68, 95% CI: 2.54-2.82), low birth weight infants (AHR = 1.28, 95% CI: 1.22-1.34), teenage pregnancy (AHR = 1.19, 95 CI: 1.10-1.29), preceding birth interval <  18 months (AHR = 3.27, 95% CI: 3.10-3.45), birth order ≥ four (AHR = 1.14, 95% CI:1.10-1.19), home delivery (AHR = 1.08, 95% CI: 1.04-1.13), and unimproved water source (AHR = 1.07, 95% CI: 1.01-1.13), female sex (AHR = 0.86, 95% CI: 0.83-0.89), immediately breastfeed (AHR = 0.24, 95% CI: 0.23-0.25), and educated mother (AHR = 0.88, 95% CI: 0.82-0. 95) and educated father (AHR = 0.90, 95% CI: 0.85-0.96) were statistically significant factors for infant mortality.

CONCLUSION

Significant number of infants died in SSA. The most common cause of infant death is a preventable bio-demographic factor. To reduce infant mortality in the region, policymakers and other stakeholders should pay attention to preventable bio-demographic risk factors, enhance women education and improved water sources.

Ranking Rule-Based Automatic Explanations for Machine Learning Predictions on Asthma Hospital Encounters in Patients With Asthma: Retrospective Cohort Study

Background

Asthma hospital encounters impose a heavy burden on the health care system. To improve preventive care and outcomes for patients with asthma, we recently developed a black-box machine learning model to predict whether a patient with asthma will have one or more asthma hospital encounters in the succeeding 12 months. Our model is more accurate than previous models. However, black-box machine learning models do not explain their predictions, which forms a barrier to widespread clinical adoption. To solve this issue, we previously developed a method to automatically provide rule-based explanations for the model’s predictions and to suggest tailored interventions without sacrificing model performance. For an average patient correctly predicted by our model to have future asthma hospital encounters, our explanation method generated over 5000 rule-based explanations, if any. However, the user of the automated explanation function, often a busy clinician, will want to quickly obtain the most useful information for a patient by viewing only the top few explanations. Therefore, a methodology is required to appropriately rank the explanations generated for a patient. However, this is currently an open problem.

Objective

The aim of this study is to develop a method to appropriately rank the rule-based explanations that our automated explanation method generates for a patient.

Methods

We developed a ranking method that struck a balance among multiple factors. Through a secondary analysis of 82,888 data instances of adults with asthma from the University of Washington Medicine between 2011 and 2018, we demonstrated our ranking method on the test case of predicting asthma hospital encounters in patients with asthma.

Results

For each patient predicted to have asthma hospital encounters in the succeeding 12 months, the top few explanations returned by our ranking method typically have high quality and low redundancy. Many top-ranked explanations provide useful insights on the various aspects of the patient’s situation, which cannot be easily obtained by viewing the patient’s data in the current electronic health record system.

Conclusions

The explanation ranking module is an essential component of the automated explanation function, and it addresses the interpretability issue that deters the widespread adoption of machine learning predictive models in clinical practice. In the next few years, we plan to test our explanation ranking method on predictive modeling problems addressing other diseases as well as on data from other health care systems.

International Registered Report Identifier (IRRID)

RR2-10.2196/5039

Return on Investment of Workplace Wellness Programs for Chronic Disease Prevention: A Systematic Review

CONTEXT

Individuals with noncommunicable diseases account for a disproportionate share of medical expenditures, absenteeism, and presenteeism. Therefore, employers are increasingly looking to worksite wellness programs as a cost-containment strategy. Previous reviews examining whether worksite wellness programs deliver a positive return on investment have shown mixed results, possibly because the more optimistic findings come from studies with poorer methodologic quality. The purpose of this systematic review is to critically revisit and update this literature to explore that hypothesis.

EVIDENCE ACQUISITION

A total of 4 databases were systematically searched for studies published before June 2019. Included studies were economic evaluations of worksite wellness programs that were based in the U.S., that lasted for at least 4 weeks, and that were with at least 1 behavior change component targeting 1 of the 4 primary modifiable behaviors for chronic disease: physical activity, healthy diet, tobacco use, and harmful consumption of alcohol. Methodologic quality was assessed using Consensus for Health Economic Criteria guidelines and the risk for selection bias associated with the study design. Data extraction (September 2019-February 2020) was followed by a narrative synthesis of worksite wellness programs characteristics and return on investment estimates.

EVIDENCE SYNTHESIS

A total of 25 relevant studies were identified. After conducting a quality and bias assessment, only 2 of the 25 studies were found to have both high methodologic rigor and lower risk for selection bias. These studies found no evidence of a positive return on investment in the short term.

CONCLUSIONS

The highest-quality studies do not support the hypothesis that worksite wellness programs deliver a positive return on investment within the first few years of initiation.

Specifics of corporate health programs for railway workers

Preventive workplace programs are one of the optimal organizational models for the prevention of noncommunicable diseases in the workingage population. Corporate health programs allow to effectively influence the lifestyle of employees, which makes it possible to reduce human resource risks due to morbidity and increase labor efficiency. First, programs for the prevention of cardiovascular diseases are being implemented. A number of researchers report that implementing prevention programs in the workplace can reduce the number of people with bad habits. The effectiveness of preventive workplace programs largely depends on the mechanisms of their implementation. A feature of railway companies is the presence of a large staff of employees of various specialties. Many factors affecting health are, in one way or another, related to the workflow, since most railway companies operate continuously. Low health literacy of railway workers on health protection and disease prevention was noted. Measures such as financial incentives, preventive counseling, the creation of personalized health profiles and the availability of healthy food in the workplace have been shown to be effective. The review also discusses Russian corporate preventive workplace programs.

Health And Economic Outcomes Up To Three Years After A Workplace Wellness Program: A Randomized Controlled Trial

Workplace wellness programs aim to improve employee health and lower health care spending. Recent randomized studies have found modest short-run effects on health behaviors, but longer-run effects remain poorly understood. We analyzed a clustered randomized trial of a workplace wellness program implemented at a large multisite US employer. Twenty-five randomly selected treatment worksites received the program, with five of the worksites added at the trial's midpoint, and 135 randomly selected control worksites did not. The program included modules on nutrition, physical activity, and stress reduction, implemented by registered dietitians. The effects of program availability and participation were assessed. At the end of three years, employees at the treatment worksites had better self-reported health behaviors, including a higher rate of actively managing their weight. No significant differences were found in self-reported health; clinical markers of health; health care spending or use; or absenteeism, tenure, or job performance. Improvements in health behaviors after three years were similar to those at eighteen months, but the longer follow-up did not yield detectable improvements in clinical, economic, or employment outcomes.

Using Computational Methods to Improve Integrated Disease Management for Asthma and Chronic Obstructive Pulmonary Disease: Protocol for a Secondary Analysis

Background

Asthma and chronic obstructive pulmonary disease (COPD) impose a heavy burden on health care. Approximately one-fourth of patients with asthma and patients with COPD are prone to exacerbations, which can be greatly reduced by preventive care via integrated disease management that has a limited service capacity. To do this well, a predictive model for proneness to exacerbation is required, but no such model exists. It would be suboptimal to build such models using the current model building approach for asthma and COPD, which has 2 gaps due to rarely factoring in temporal features showing early health changes and general directions. First, existing models for other asthma and COPD outcomes rarely use more advanced temporal features, such as the slope of the number of days to albuterol refill, and are inaccurate. Second, existing models seldom show the reason a patient is deemed high risk and the potential interventions to reduce the risk, making already occupied clinicians expend more time on chart review and overlook suitable interventions. Regular automatic explanation methods cannot deal with temporal data and address this issue well.

Objective

To enable more patients with asthma and patients with COPD to obtain suitable and timely care to avoid exacerbations, we aim to implement comprehensible computational methods to accurately predict proneness to exacerbation and recommend customized interventions.

Methods

We will use temporal features to accurately predict proneness to exacerbation, automatically find modifiable temporal risk factors for every high-risk patient, and assess the impact of actionable warnings on clinicians’ decisions to use integrated disease management to prevent proneness to exacerbation.

Results

We have obtained most of the clinical and administrative data of patients with asthma from 3 prominent American health care systems. We are retrieving other clinical and administrative data, mostly of patients with COPD, needed for the study. We intend to complete the study in 6 years.

Conclusions

Our results will help make asthma and COPD care more proactive, effective, and efficient, improving outcomes and saving resources.

International Registered Report Identifier (IRRID)

PRR1-10.2196/27065

Forecasting Future Asthma Hospital Encounters of Patients With Asthma in an Academic Health Care System: Predictive Model Development and Secondary Analysis Study

Background

Asthma affects a large proportion of the population and leads to many hospital encounters involving both hospitalizations and emergency department visits every year. To lower the number of such encounters, many health care systems and health plans deploy predictive models to prospectively identify patients at high risk and offer them care management services for preventive care. However, the previous models do not have sufficient accuracy for serving this purpose well. Embracing the modeling strategy of examining many candidate features, we built a new machine learning model to forecast future asthma hospital encounters of patients with asthma at Intermountain Healthcare, a nonacademic health care system. This model is more accurate than the previously published models. However, it is unclear how well our modeling strategy generalizes to academic health care systems, whose patient composition differs from that of Intermountain Healthcare.

Objective

This study aims to evaluate the generalizability of our modeling strategy to the University of Washington Medicine (UWM), an academic health care system.

Methods

All adult patients with asthma who visited UWM facilities between 2011 and 2018 served as the patient cohort. We considered 234 candidate features. Through a secondary analysis of 82,888 UWM data instances from 2011 to 2018, we built a machine learning model to forecast asthma hospital encounters of patients with asthma in the subsequent 12 months.

Results

Our UWM model yielded an area under the receiver operating characteristic curve (AUC) of 0.902. When placing the cutoff point for making binary classification at the top 10% (1464/14,644) of patients with asthma with the largest forecasted risk, our UWM model yielded an accuracy of 90.6% (13,268/14,644), a sensitivity of 70.2% (153/218), and a specificity of 90.91% (13,115/14,426).

Conclusions

Our modeling strategy showed excellent generalizability to the UWM, leading to a model with an AUC that is higher than all of the AUCs previously reported in the literature for forecasting asthma hospital encounters. After further optimization, our model could be used to facilitate the efficient and effective allocation of asthma care management resources to improve outcomes.

International Registered Report Identifier (IRRID)

RR2-10.2196/resprot.5039

Generalizability of an Automatic Explanation Method for Machine Learning Prediction Results on Asthma-Related Hospital Visits in Patients With Asthma: Quantitative Analysis

Background

Asthma exerts a substantial burden on patients and health care systems. To facilitate preventive care for asthma management and improve patient outcomes, we recently developed two machine learning models, one on Intermountain Healthcare data and the other on Kaiser Permanente Southern California (KPSC) data, to forecast asthma-related hospital visits, including emergency department visits and hospitalizations, in the succeeding 12 months among patients with asthma. As is typical for machine learning approaches, these two models do not explain their forecasting results. To address the interpretability issue of black-box models, we designed an automatic method to offer rule format explanations for the forecasting results of any machine learning model on imbalanced tabular data and to suggest customized interventions with no accuracy loss. Our method worked well for explaining the forecasting results of our Intermountain Healthcare model, but its generalizability to other health care systems remains unknown.

Objective

The objective of this study is to evaluate the generalizability of our automatic explanation method to KPSC for forecasting asthma-related hospital visits.

Methods

Through a secondary analysis of 987,506 data instances from 2012 to 2017 at KPSC, we used our method to explain the forecasting results of our KPSC model and to suggest customized interventions. The patient cohort covered a random sample of 70% of patients with asthma who had a KPSC health plan for any period between 2015 and 2018.

Results

Our method explained the forecasting results for 97.57% (2204/2259) of the patients with asthma who were correctly forecasted to undergo asthma-related hospital visits in the succeeding 12 months.

Conclusions

For forecasting asthma-related hospital visits, our automatic explanation method exhibited an acceptable generalizability to KPSC.

International Registered Report Identifier (IRRID)

RR2-10.2196/resprot.5039

Sorption as a rapidly response for oil spill accidents: A material and mechanistic approach

Accidents involving oil transportation has increase due to directly connection with the elevation of global energy demand. The environmental losses are tremendous and brings huge economic issues to remediate the spilled oil. This report presents an up-to-date review on an overall aspects of oil spill remediation techniques, the fundamentals and advantages of sorption, the most applied materials through diverse types of oil spill sites and oils with variety features, highlight to natural materials and future prospective. As the environment preservation progressively becomes a major social concern issue, the achievement of a worldwide distribution process aligned with environmental legislation and economic viability is crucial to the oil industry. For this, a specific preparation considering several scenarios must be carried out regarding minimization of oil spillages. Since the sorbent materials are decisive for sorption, it was approached the main sorbents: natural, graphenic, nano, polymeric and waste materials, and future trends.

Relationship between nurse burnout, patient and organizational outcomes: Systematic review

BACKGROUND

Burnout, characterized by emotional exhaustion, depersonalization, and decreased personal accomplishments, poses a significant burden on individual nurses' health and mental wellbeing. As growing evidence highlights the adverse consequences of burnout for clinicians, patients, and organizations, it is imperative to examine nurse burnout in the healthcare system.

OBJECTIVE

The purpose of this review is to systematically and critically appraise the current literature to examine the associations between nurse burnout and patient and hospital organizational outcomes.

DESIGN AND DATA SOURCES

A systematic review following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses was conducted. PubMed, CINAHL, PsychInfo, Scopus, and Embase were the search engines used. The inclusion criteria were any primary studies examining burnout among nurses working in hospitals as an independent variable, in peer-reviewed journals, and written in English. The search was performed from October 2018 to January 2019 and updated in January and October 2020.

RESULTS

A total of 20 studies were included in the review. The organizational-related outcomes associated with nurse burnout were (1) patient safety, (2) quality of care, (3) nurses' organizational commitment, (4) nurse productivity, and (5) patient satisfaction. For these themes, nurse burnout was consistently inversely associated with outcome measures.

CONCLUSIONS

Nurse burnout is an occupational hazard affecting nurses, patients, organizations, and society at large. Nurse burnout is associated with worsening safety and quality of care, decreased patient satisfaction, and nurses' organizational commitment and productivity. Traditionally, burnout is viewed as an individual issue. However, reframing burnout as an organizational and collective phenomenon affords the broader perspective necessary to address nurse burnout. Tweetable abstract: Not only nurse burnout associated w/ worsening safety & quality of care, but also w/ nurses' organizational commitment and productivity. Reframing burnout, as an organizational & collective phenomenon is necessary.

A Scoping Review of Economic Evaluations of Workplace Wellness Programs

OBJECTIVE

Debates about the effectiveness of workplace wellness programs (WWPs) call for a review of the evidence for return on investment (ROI) of WWPs. We examined literature on the heterogeneity in methods used in the ROI of WWPs to show how this heterogeneity may affect conclusions and inferences about ROI.

METHODS

We conducted a scoping review using systematic review methods and adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. We reviewed PubMed, EconLit, Proquest Central, and Scopus databases for published articles. We included articles that (1) were published before December 20, 2019, when our last search was conducted, and (2) met our inclusion criteria that were based on target population, target intervention, evaluation method, and ROI as the main outcome.

RESULTS

We identified 47 peer-reviewed articles from the selected databases that met our inclusion criteria. We explored the effect of study characteristics on ROI estimates. Thirty-one articles had ROI measures. Studies with costs of presenteeism had the lowest ROI estimates compared with other cost combinations associated with health care and absenteeism. Studies with components of disease management produced higher ROI than programs with components of wellness. We found a positive relationship between ROI and program length and a negative relationship between ROI and conflict of interest. Evaluations in small companies (≤500 employees) were associated with lower ROI estimates than evaluations in large companies (>500 employees). Studies with lower reporting quality scores, including studies that were missing information on statistical inference, had lower ROI estimates. Higher methodologic quality was associated with lower ROI estimates.

CONCLUSION

This review provides recommendations that can improve the methodologic quality of studies to validate the ROI and public health effects of WWPs.

Automatically Explaining Machine Learning Prediction Results on Asthma Hospital Visits in Patients With Asthma: Secondary Analysis

Background

Asthma is a major chronic disease that poses a heavy burden on health care. To facilitate the allocation of care management resources aimed at improving outcomes for high-risk patients with asthma, we recently built a machine learning model to predict asthma hospital visits in the subsequent year in patients with asthma. Our model is more accurate than previous models. However, like most machine learning models, it offers no explanation of its prediction results. This creates a barrier for use in care management, where interpretability is desired.

Objective

This study aims to develop a method to automatically explain the prediction results of the model and recommend tailored interventions without lowering the performance measures of the model.

Methods

Our data were imbalanced, with only a small portion of data instances linking to future asthma hospital visits. To handle imbalanced data, we extended our previous method of automatically offering rule-formed explanations for the prediction results of any machine learning model on tabular data without lowering the model’s performance measures. In a secondary analysis of the 334,564 data instances from Intermountain Healthcare between 2005 and 2018 used to form our model, we employed the extended method to automatically explain the prediction results of our model and recommend tailored interventions. The patient cohort consisted of all patients with asthma who received care at Intermountain Healthcare between 2005 and 2018, and resided in Utah or Idaho as recorded at the visit.

Results

Our method explained the prediction results for 89.7% (391/436) of the patients with asthma who, per our model’s correct prediction, were likely to incur asthma hospital visits in the subsequent year.

Conclusions

This study is the first to demonstrate the feasibility of automatically offering rule-formed explanations for the prediction results of any machine learning model on imbalanced tabular data without lowering the performance measures of the model. After further improvement, our asthma outcome prediction model coupled with the automatic explanation function could be used by clinicians to guide the allocation of limited asthma care management resources and the identification of appropriate interventions.

Eight-Year Health Risks Trend Analysis of a Comprehensive Workplace Health Promotion Program

Research has shown that workplace health promotion (WHP) efforts can positively affect employees' health risk accumulation. However, earlier literature has provided insights of health risk changes in the short-term. This prospective longitudinal quasi-experimental study investigated trends in health risks of a comprehensive, eight-year WHP program (n = 523-651). Health risk data were collected from health risk assessments in 2010-2011, 2013-2014, and 2016-2017, applying both a questionnaire and biometric screenings. Health risk changes were investigated for three different time-periods, 2010-2013, 2014-2017, and 2010-2017, using descriptive analyses, t-tests, and the Wilcoxon Signed Rank and McNemar's test, where appropriate. Overall health risk transitions were assessed according to low-, moderate-, and high-risk categories. Trend analyses observed 50-60% prevalence for low-, 30-35% for moderate-, and 9-11% high-risk levels across the eight years. In the overall health risk transitions of the three time-periods, 66-73% of participants stayed at the same risk level, 13-15% of participants improved, and 12-21% had deteriorated risk level across the three intervention periods. Our findings appear to indicate that the multiyear WHP program was effective in maintaining low and moderate risk levels, but fell short of reducing the total number of health risks at the population level.

Developing a Predictive Model for Asthma-Related Hospital Encounters in Patients With Asthma in a Large, Integrated Health Care System: Secondary Analysis

Background

Asthma causes numerous hospital encounters annually, including emergency department visits and hospitalizations. To improve patient outcomes and reduce the number of these encounters, predictive models are widely used to prospectively pinpoint high-risk patients with asthma for preventive care via care management. However, previous models do not have adequate accuracy to achieve this goal well. Adopting the modeling guideline for checking extensive candidate features, we recently constructed a machine learning model on Intermountain Healthcare data to predict asthma-related hospital encounters in patients with asthma. Although this model is more accurate than the previous models, whether our modeling guideline is generalizable to other health care systems remains unknown.

Objective

This study aims to assess the generalizability of our modeling guideline to Kaiser Permanente Southern California (KPSC).

Methods

The patient cohort included a random sample of 70.00% (397,858/568,369) of patients with asthma who were enrolled in a KPSC health plan for any duration between 2015 and 2018. We produced a machine learning model via a secondary analysis of 987,506 KPSC data instances from 2012 to 2017 and by checking 337 candidate features to project asthma-related hospital encounters in the following 12-month period in patients with asthma.

Results

Our model reached an area under the receiver operating characteristic curve of 0.820. When the cutoff point for binary classification was placed at the top 10.00% (20,474/204,744) of patients with asthma having the largest predicted risk, our model achieved an accuracy of 90.08% (184,435/204,744), a sensitivity of 51.90% (2259/4353), and a specificity of 90.91% (182,176/200,391).

Conclusions

Our modeling guideline exhibited acceptable generalizability to KPSC and resulted in a model that is more accurate than those formerly built by others. After further enhancement, our model could be used to guide asthma care management.

International Registered Report Identifier (IRRID)

RR2-10.2196/resprot.5039

Translating CDSMP to the Workplace: Results of the Live Healthy Work Healthy Program

PURPOSE

Report the results of a randomized, controlled trial of Live Healthy, Work Healthy (LHWH), a worksite translation of the Chronic Disease Self-Management Program (CDSMP).

DESIGN

14 worksites were randomly assigned to LHWH, standard CDSMP (usual care) or no-intervention (control) group.

SETTING

The diverse set of work organizations centered around a rural community in SE US.

SUBJECTS

411 participants completed baseline data with 359 being included in the final analyses.

INTERVENTION

LHWH had been adapted to fit the unique characteristics of work organizations. This translated program consists of 15 sessions over 8 weeks and was facilitated by trained lay leaders.

MEASURES

The primary outcomes including health risk, patient-provider communication, quality of life, medical adherence and work performance were collected pretest, posttest (6 mos.) and follow-up (12 mos.).

ANALYSIS

Analyses were conducted using latent change score models in a structural equation modeling framework.

RESULTS

79% of participants reported at least one chronic condition with an average of 2.7 chronic conditions reported. Results indicated that LHWH program demonstrated positive changes in a most outcomes including significant exercise (uΔ = 0.89, p < .01), chronic disease self-efficacy (uΔ = 0.63, p < .05), fatigue (uΔ = -1.45, p < .05), stress (uΔ = -0.98, p < .01) and mentally unhealthy days (uΔ = -3.47, p < .001).

CONCLUSIONS

The translation of LHWH is an effective, low cost, embeddable program that has the potential to improve the health and work life of employees.

Testing the Generalizability of an Automated Method for Explaining Machine Learning Predictions on Asthma Patients' Asthma Hospital Visits to an Academic Healthcare System

Asthma puts a tremendous overhead on healthcare. To enable effective preventive care to improve outcomes in managing asthma, we recently created two machine learning models, one using University of Washington Medicine data and the other using Intermountain Healthcare data, to predict asthma hospital visits in the next 12 months in asthma patients. As is common in machine learning, neither model supplies explanations for its predictions. To tackle this interpretability issue of black-box models, we developed an automated method to produce rule-style explanations for any machine learning model's predictions made on imbalanced tabular data and to recommend customized interventions without lowering the prediction accuracy. Our method exhibited good performance in explaining our Intermountain Healthcare model's predictions. Yet, it stays unknown how well our method generalizes to academic healthcare systems, whose patient composition differs from that of Intermountain Healthcare. This study evaluates our automated explaining method's generalizability to the academic healthcare system University of Washington Medicine on predicting asthma hospital visits. We did a secondary analysis on 82,888 University of Washington Medicine data instances of asthmatic adults between 2011 and 2018, using our method to explain our University of Washington Medicine model's predictions and to recommend customized interventions. Our results showed that for predicting asthma hospital visits, our automated explaining method had satisfactory generalizability to University of Washington Medicine. In particular, our method explained the predictions for 87.6% of the asthma patients whom our University of Washington Medicine model accurately predicted to experience asthma hospital visits in the next 12 months.

Ten Modifiable Health Risk Factors and Employees' Medical Costs-An Update

PURPOSE

To estimate the relationship between employees' health risks and health-care costs to inform health promotion program design.

DESIGN

An observational study of person-level health-care claims and health risk assessment (HRA) data that used regression models to estimate the relationship between 10 modifiable risk factors and subsequent year 1 health-care costs.

SETTING

United States.

PARTICIPANTS

The sample included active, full-time, adult employees continuously enrolled in employer-sponsored health insurance plans contributing to IBM MarketScan Research Databases who completed an HRA. Study criteria were met by 135 219 employees from 11 employers.

MEASURES

Ten modifiable risk factors and individual sociodemographic and health characteristics were included in the models as independent variables. Five settings of health-care costs were outcomes in addition to total expenditures.

ANALYSIS

After building the analytic file, we estimated generalized linear models and conducted postestimation bootstrapping.

RESULTS

Health-care costs were significantly higher for employees at higher risk for blood glucose, obesity, stress, depression, and physical inactivity (all at P < .0001) than for those at lower risk. Similar cost differentials were found when specific health-care services were examined.

CONCLUSION

Employers may achieve cost savings in the short run by implementing comprehensive health promotion programs that focus on decreasing multiple health risks.

Developing a Model to Predict Hospital Encounters for Asthma in Asthmatic Patients: Secondary Analysis

Background

As a major chronic disease, asthma causes many emergency department (ED) visits and hospitalizations each year. Predictive modeling is a key technology to prospectively identify high-risk asthmatic patients and enroll them in care management for preventive care to reduce future hospital encounters, including inpatient stays and ED visits. However, existing models for predicting hospital encounters in asthmatic patients are inaccurate. Usually, they miss over half of the patients who will incur future hospital encounters and incorrectly classify many others who will not. This makes it difficult to match the limited resources of care management to the patients who will incur future hospital encounters, increasing health care costs and degrading patient outcomes.

Objective

The goal of this study was to develop a more accurate model for predicting hospital encounters in asthmatic patients.

Methods

Secondary analysis of 334,564 data instances from Intermountain Healthcare from 2005 to 2018 was conducted to build a machine learning classification model to predict the hospital encounters for asthma in the following year in asthmatic patients. The patient cohort included all asthmatic patients who resided in Utah or Idaho and visited Intermountain Healthcare facilities during 2005 to 2018. A total of 235 candidate features were considered for model building.

Results

The model achieved an area under the receiver operating characteristic curve of 0.859 (95% CI 0.846-0.871). When the cutoff threshold for conducting binary classification was set at the top 10.00% (1926/19,256) of asthmatic patients with the highest predicted risk, the model reached an accuracy of 90.31% (17,391/19,256; 95% CI 89.86-90.70), a sensitivity of 53.7% (436/812; 95% CI 50.12-57.18), and a specificity of 91.93% (16,955/18,444; 95% CI 91.54-92.31). To steer future research on this topic, we pinpointed several potential improvements to our model.

Conclusions

Our model improves the state of the art for predicting hospital encounters for asthma in asthmatic patients. After further refinement, the model could be integrated into a decision support tool to guide asthma care management allocation.

International Registered Report Identifier (IRRID)

RR2-10.2196/resprot.5039

Showers, Culture, and Conflict Resolution: A Qualitative Study of Employees' Perceptions of Workplace Wellness Opportunities

OBJECTIVE

Research on employee opinions of workplace wellness programs is limited.

METHODS

At a large academic medical center in Boston, we conducted 12 focus groups on employee perceptions of wellness programs. We analyzed data using the immersion-crystallization approach. Participant mean age (N = 109) was 41 years; 89% were female; 54% were white.

RESULTS

Employees cited prominent barriers to program participation: limited availability; time and marketing; disparities in access; and workplace culture. Encouraging supportive, interpersonal relationships among employees and perceived institutional support for wellness may improve workplace culture and improve participation. Employees suggested changes to physical space, including onsite showers and recommended that a centralized wellness program could create and market initiatives such as competitions and incentives.

CONCLUSION

Employees sought measures to address serious constraints on time and space, sometimes toxic interpersonal relationships, and poor communication, aspects of workplaces not typically addressed by wellness efforts.

Anteceding factors predicting absenteeism and presenteeism in urban area in Malaysia

Background

Organization productivity is strongly linked to employees’ socioeconomic characteristics and health which is marked by absenteeism and presenteeism. This study aims to identify anteceding factors predicting employees’ absenteeism and presenteeism by income, physical and mental health.

Methods

An online health survey was conducted between May to July 2017 among employees from 47 private companies located in urban Malaysia. A total of 5235 respondents completed the 20-min online employee health survey on a voluntary basis. Chi-Square or Fisher’s exact tests were used to determine association between income with demographic and categorical factors of absenteeism and presenteeism. Multivariate linear regression was used to identify factors predicting absenteeism and presenteeism.

Results

More than one third of respondents’ monthly income were less than RM4,000 (35.4%), 29.6% between RM4,000-RM7,999 and 35.0% earned RM8,000 and above. The mean age was 33.8 years (sd ± 8.8) and 49.1% were married. A majority were degree holders (74.4%) and 43.6% were very concerned about their financial status. Mean years of working was 6.2 years (sd ± 6.9) with 68.9% satisfied with their job. More than half reported good general physical health (54.5%) (p = 0.065) and mental health (53.5%) (p = 0.019). The mean hours of sleep were 6.4 h (sd ± 1.1) with 63.2% reporting being unwell due to stress for the past 12 months. Mean work time missed due to ill-health (absenteeism) was 3.1% (sd ± 9.1), 2.8% (sd ± 9.1) and 1.8% (sd ± 6.5) among employees whose monthly income was less than RM4,000, RM4,000-RM7,999 and over RM8,000 respectively (p = 0.0066). Mean impairment while working due to ill-health (presenteeism) was 28.2% (sd ± 25.3), 24.9% (sd ± 25.5) and 20.3% (sd ± 22.9) among employees whose monthly income was less than RM4,000, RM4,000-RM7,999 and over RM8,000 respectively (p < 0.0001). Factors that predict both absenteeism and presenteeism were income, general physical health, sleep length and being unwell due to stress.

Conclusions

A combination of socioeconomic, physical and mental health factors predicted absenteeism and presenteeism with different strengths. Having insufficient income may lead to second jobs or working more hours which may affect their sleep, subjecting them to stressful condition and poor physical health. These findings demand holistic interventions from organizations and the government.

Using Temporal Features to Provide Data-Driven Clinical Early Warnings for Chronic Obstructive Pulmonary Disease and Asthma Care Management: Protocol for a Secondary Analysis

Background

Both chronic obstructive pulmonary disease (COPD) and asthma incur heavy health care burdens. To support tailored preventive care for these 2 diseases, predictive modeling is widely used to give warnings and to identify patients for care management. However, 3 gaps exist in current modeling methods owing to rarely factoring in temporal aspects showing trends and early health change: (1) existing models seldom use temporal features and often give late warnings, making care reactive. A health risk is often found at a relatively late stage of declining health, when the risk of a poor outcome is high and resolving the issue is difficult and costly. A typical model predicts patient outcomes in the next 12 months. This often does not warn early enough. If a patient will actually be hospitalized for COPD next week, intervening now could be too late to avoid the hospitalization. If temporal features were used, this patient could potentially be identified a few weeks earlier to institute preventive therapy; (2) existing models often miss many temporal features with high predictive power and have low accuracy. This makes care management enroll many patients not needing it and overlook over half of the patients needing it the most; (3) existing models often give no information on why a patient is at high risk nor about possible interventions to mitigate risk, causing busy care managers to spend more time reviewing charts and to miss suited interventions. Typical automatic explanation methods cannot handle longitudinal attributes and fully address these issues.

Objective

To fill these gaps so that more COPD and asthma patients will receive more appropriate and timely care, we will develop comprehensible data-driven methods to provide accurate early warnings of poor outcomes and to suggest tailored interventions, making care more proactive, efficient, and effective.

Methods

By conducting a secondary data analysis and surveys, the study will: (1) use temporal features to provide accurate early warnings of poor outcomes and assess the potential impact on prediction accuracy, risk warning timeliness, and outcomes; (2) automatically identify actionable temporal risk factors for each patient at high risk for future hospital use and assess the impact on prediction accuracy and outcomes; and (3) assess the impact of actionable information on clinicians’ acceptance of early warnings and on perceived care plan quality.

Results

We are obtaining clinical and administrative datasets from 3 leading health care systems’ enterprise data warehouses. We plan to start data analysis in 2020 and finish our study in 2025.

Conclusions

Techniques to be developed in this study can boost risk warning timeliness, model accuracy, and generalizability; improve patient finding for preventive care; help form tailored care plans; advance machine learning for many clinical applications; and be generalized for many other chronic diseases.

International Registered Report Identifier (IRRID)

PRR1-10.2196/13783

The impact of financial incentives on health and health care: Evidence from a large wellness program

Workplace wellness programs have become increasingly common in the United States, although there is not yet consensus regarding the ability of such programs to improve employees' health and reduce health care costs. In this paper, we study a program offered by a large U.S. employer that provides substantial financial incentives directly tied to employees' health. The program has a high participation rate among eligible employees, around 80%, and we analyze the data on the first 4 years of the program, linked to health care claims. We document robust improvements in employee health and a correlation between certain health improvements and reductions in health care cost. Despite the latter association, we cannot find direct evidence causally linking program participation to reduced health care costs, although it seems plausible that such a relationship will arise over longer horizons.

Identifying Patients Who Are Likely to Receive Most of Their Care From a Specific Health Care System: Demonstration via Secondary Analysis

Background

In the United States, health care is fragmented in numerous distinct health care systems including private, public, and federal organizations like private physician groups and academic medical centers. Many patients have their complete medical data scattered across these several health care systems, with no particular system having complete data on any of them. Several major data analysis tasks such as predictive modeling using historical data are considered impractical on incomplete data.

Objective

Our objective was to find a way to enable these analysis tasks for a health care system with incomplete data on many of its patients.

Methods

This study presents, to the best of our knowledge, the first method to use a geographic constraint to identify a reasonably large subset of patients who tend to receive most of their care from a given health care system. A data analysis task needing relatively complete data can be conducted on this subset of patients. We demonstrated our method using data from the University of Washington Medicine (UWM) and PreManage data covering the use of all hospitals in Washington State. We compared 10 candidate constraints to optimize the solution.

Results

For UWM, the best constraint is that the patient has a UWM primary care physician and lives within 5 miles of at least one UWM hospital. About 16.01% (55,707/348,054) of UWM patients satisfied this constraint. Around 69.38% (10,501/15,135) of their inpatient stays and emergency department visits occurred within UWM in the following 6 months, more than double the corresponding percentage for all UWM patients.

Conclusions

Our method can identify a reasonably large subset of patients who tend to receive most of their care from UWM. This enables several major analysis tasks on incomplete medical data that were previously deemed infeasible.

Pharmaceutical use according to participation in worksite wellness screening and health campaigns

This study evaluated whether participation in worksite wellness screening and health campaigns influences the number and cost (USD) of pharmacy medication claims. Analyses are based on 2531 workers employed all four academic years in a large school district in the western United States, 2010–11 through 2013–14. Mean and ratio comparisons were adjusted by age, sex, year, and baseline health. Approximately 84.2% of employees participated in wellness screening and 60.1% completed one or more health campaigns. Those completing wellness screening were 1.09 (95% CI 1.06–1.13) times more likely to file a claim. Mean total cost remained near $934 (SD = $3695) over the academic years, positively associated with years of wellness screening, suggesting increased awareness of the need for medication through screening. Women were 1.02 (95% CI 1.00–1.05) times more likely than men to participate in wellness screening and had greater total pharmacy cost ($990.6 [SD = $4023.7] vs. $777.9 [SD = $2580.5], p = 0.0104). Women were 1.38 (95% CI 1.32–1.44) times more likely to complete a health campaign. Mean number of pharmacy claims was lower (9.8 vs. 10.6, p = 0.0069) in those completing at least one health campaign, suggesting greater health orientation in women. Those completing at least one health campaign were 0.96 (95% CI 0.92–0.99) times as likely to have a total cost of medication above the median, 0.94 (95% CI 0.88–1.01) as likely to have a total cost of medication above the 75th percentile, and 0.84 (0.75–0.96) times as likely to have a total cost above the 90th percentile.

Cost Effectiveness of a Weight Management Program Implemented in the Worksite: Translation of Fuel Your Life

OBJECTIVE

Conduct a cost-effectiveness analysis of the Fuel Your Life (FYL) program dissemination.

METHODS

Employees were recruited from three workplaces randomly assigned to one of the conditions: telephone coaching, small group coaching, and self-study. Costs were collected prospectively during the efficacy trial. The main outcome measures of interest were weight loss and quality-adjusted life years (QALYs).

RESULTS

The phone condition was most costly ($601 to $589/employee) and the self-study condition was least costly ($145 to $143/employee). For weight loss, delivering FYL through the small group condition was no more effective, yet more expensive, than the self-study delivery. For QALYs, the group delivery of FYL was in an acceptable cost-effectiveness range ($22,400/QALY) relative to self-study (95% confidence interval [CI]: $10,600/QALY-dominated).

CONCLUSIONS

Prevention programs require adaptation at the local level and significantly affect the cost, effectiveness, and cost-effectiveness of the program.

Matching and Regression to the Mean in Difference-in-Differences Analysis

OBJECTIVE

To demonstrate regression to the mean bias introduced by matching on preperiod variables in difference-in-differences studies.

DATA SOURCES

Simulated data.

STUDY DESIGN

We performed a Monte Carlo simulation to estimate the effect of a placebo intervention on simulated longitudinal data for units in treatment and control groups using unmatched and matched difference-in-differences analyses. We varied the preperiod level and trend differences between the treatment and control groups, and the serial correlation of the matching variables. We assessed estimator bias as the mean absolute deviation of estimated program effects from the true value of zero.

PRINCIPAL FINDINGS

When preperiod outcome level is correlated with treatment assignment, an unmatched analysis is unbiased, but matching units on preperiod outcome levels produces biased estimates. The bias increases with greater preperiod level differences and weaker serial correlation in the outcome. This problem extends to matching on preperiod level of a time-varying covariate. When treatment assignment is correlated with preperiod trend only, the unmatched analysis is biased, and matching units on preperiod level or trend does not introduce additional bias.

CONCLUSIONS

Researchers should be aware of the threat of regression to the mean when constructing matched samples for difference-in-differences. We provide guidance on when to incorporate matching in this study design.

The Impact of Worksite Clinics on Teacher Health Care Utilization and Cost, Self-Reported Health Status, and Student Academic Achievement Growth in a Public School District

OBJECTIVE

The aim of this study was to examine the impact of worksite clinics on health care utilization and cost, self-reported health status, and student achievement growth in a public school district.

METHODS

We used insurance claims, health risk assessment, and student achievement growth data for active teachers during 2007 to 2015. A difference-in-differences approach was applied to measure the impact of worksite clinics.

RESULTS

Compared with using a community-based clinic as the usual source of primary care, using a worksite clinic was associated with significantly lower inpatient admissions (53 vs 31 per 1000 teacher years), annual health care cost ($5043 vs $4298 in 2016 US dollars, a difference of $62 per teacher per month), and annual absent work hours (63 vs 61). No significant differences were detected in self-reported health status or student achievement growth.

CONCLUSION

Worksite clinics reduce teacher health care cost and absenteeism.

Impact of a Translated Disease Self-Management Program on Employee Health and Productivity: Six-Month Findings from a Randomized Controlled Trial

Disease management is gaining importance in workplace health promotion given the aging workforce and rising chronic disease prevalence. The Chronic Disease Self-Management Program (CDSMP) is an effective intervention widely offered in diverse community settings; however, adoption remains low in workplace settings. As part of a larger NIH-funded randomized controlled trial, this study examines the effectiveness of a worksite-tailored version of CDSMP (wCDSMP [n = 72]) relative to CDSMP (‘Usual Care’ [n = 109]) to improve health and work performance among employees with one or more chronic conditions. Multiple-group latent-difference score models with sandwich estimators were fitted to identify changes from baseline to 6-month follow-up. Overall, participants were primarily female (87%), non-Hispanic white (62%), and obese (73%). On average, participants were age 48 (range: 23–72) and self-reported 3.25 chronic conditions (range: 1–16). The most commonly reported conditions were high cholesterol (45%), high blood pressure (45%), anxiety/emotional/mental health condition (26%), and diabetes (25%). Among wCDSMP participants, significant improvements were observed for physically unhealthy days (uΔ = −2.07, p = 0.018), fatigue (uΔ = −2.88, p = 0.002), sedentary behavior (uΔ = −4.49, p = 0.018), soda/sugar beverage consumption (uΔ = −0.78, p = 0.028), and fast food intake (uΔ = −0.76, p = 0.009) from baseline to follow-up. Significant improvements in patient–provider communication (uΔ = 0.46, p = 0.031) and mental work limitations (uΔ = −8.89, p = 0.010) were also observed from baseline to follow-up. Relative to Usual Care, wCDSMP participants reported significantly larger improvements in fatigue, physical activity, soda/sugar beverage consumption, and mental work limitations (p < 0.05). The translation of Usual Care (content and format) has potential to improve health among employees with chronic conditions and increase uptake in workplace settings.

Association Between Exercise Frequency and Health Care Costs Among Employees at a Large University and Academic Medical Center

OBJECTIVE

The aim of this study was to evaluate the relationship between exercise frequency and health care costs associated with medical and pharmacy claims among a 10-year employee cohort.

METHODS

The relationship between self-reported exercise (days/week) and health care costs was analyzed with negative binomial regression, using an integrated database involving 32,044 person-years and linking employee demographics, health risk appraisal information, and health insurance claims.

RESULTS

An association demonstrating exercise frequency lowering health care costs was present in most medical and prescription drug categories and was strongest among employees reporting 2 to 3 and 4 to 5 days/week of exercise. Increased exercise was associated with statistically significant reductions in endocrine disease costs and gastrointestinal prescription drug costs.

CONCLUSIONS

This cohort demonstrates lower health care costs in employee populations when exercise frequency is increased. Employers may lower modifiable risk factors for chronic disease and reduce health care costs by promoting exercise among their employee population.

Effect of Health Plan Financial Incentive Offering on Employees with Prediabetes

BACKGROUND

Prediabetes may be improved or reversed with lifestyle interventions. A worksite wellness program offering financial incentives for participation may be effective in improving the health of employees with prediabetes. We studied the effect of employee health plan financial incentives on health outcomes for employees with prediabetes.

METHODS

We conducted a retrospective cohort study using electronic medical record data from January 2008 to December 2012. Our study participants were employees with prediabetes and propensity-matched non-employees with prediabetes and commercial health insurance, all receiving care within one health system. Exposures included fixed annual financial incentives for program participation and later a premium discount divided between program participation and achievement of goals. We used longitudinal linear mixed models to assess yearly changes in glycosylated hemoglobin (HbA1c), weight, and low-density lipoprotein cholesterol in employees versus non-employees. We also compared outcomes of employees by ever- versus never- program participant status.

RESULTS

Our study population included 1005 employees and 1005 matched non-employees. The yearly reduction in HbA1c for employees versus matched non-employees did not differ in 2008-2010 but was greater in 2010-2012, when incentives were tied to program participation as well as achievement of goals (-0.10% vs -0.08 %, respectively; P for difference in change [DIC] = .01 from 2010 to 2012). Analyses from both periods showed that employees lost more weight per year than matched non-employees (-1.85 vs -0.21 lb [1 lb=0.45 kg] from 2008 to 2010; P for DIC < .001 and -2.35 vs -0.65 lb from 2010 to 2012; P for DIC < .001). Employees who participated in disease management lost more weight than those who did not (-2.14 vs 0.79 lb yearly before 2010 and -2.82 vs -0.91 after January 1, 2010, P for DIC < .01 and < .001, respectively).

CONCLUSION

A worksite wellness program offering health plan financial incentives for participation and outcomes was associated with improvements in weight and HbA1c.

A Roadmap for Optimizing Asthma Care Management via Computational Approaches

Asthma affects 9% of Americans and incurs US $56 billion in cost, 439,000 hospitalizations, and 1.8 million emergency room visits annually. A small fraction of asthma patients with high vulnerabilities, severe disease, or great barriers to care consume most health care costs and resources. An effective approach is urgently needed to identify high-risk patients and intervene to improve outcomes and to reduce costs and resource use. Care management is widely used to implement tailored care plans for this purpose, but it is expensive and has limited service capacity. To maximize benefit, we should enroll only patients anticipated to have the highest costs or worst prognosis. Effective care management requires correctly identifying high-risk patients, but current patient identification approaches have major limitations. This paper pinpoints these limitations and outlines multiple machine learning techniques to address them, providing a roadmap for future research.

Automating Construction of Machine Learning Models With Clinical Big Data: Proposal Rationale and Methods

Background

To improve health outcomes and cut health care costs, we often need to conduct prediction/classification using large clinical datasets (aka, clinical big data), for example, to identify high-risk patients for preventive interventions. Machine learning has been proposed as a key technology for doing this. Machine learning has won most data science competitions and could support many clinical activities, yet only 15% of hospitals use it for even limited purposes. Despite familiarity with data, health care researchers often lack machine learning expertise to directly use clinical big data, creating a hurdle in realizing value from their data. Health care researchers can work with data scientists with deep machine learning knowledge, but it takes time and effort for both parties to communicate effectively. Facing a shortage in the United States of data scientists and hiring competition from companies with deep pockets, health care systems have difficulty recruiting data scientists. Building and generalizing a machine learning model often requires hundreds to thousands of manual iterations by data scientists to select the following: (1) hyper-parameter values and complex algorithms that greatly affect model accuracy and (2) operators and periods for temporally aggregating clinical attributes (eg, whether a patient’s weight kept rising in the past year). This process becomes infeasible with limited budgets.

Objective

This study’s goal is to enable health care researchers to directly use clinical big data, make machine learning feasible with limited budgets and data scientist resources, and realize value from data.

Methods

This study will allow us to achieve the following: (1) finish developing the new software, Automated Machine Learning (Auto-ML), to automate model selection for machine learning with clinical big data and validate Auto-ML on seven benchmark modeling problems of clinical importance; (2) apply Auto-ML and novel methodology to two new modeling problems crucial for care management allocation and pilot one model with care managers; and (3) perform simulations to estimate the impact of adopting Auto-ML on US patient outcomes.

Results

We are currently writing Auto-ML’s design document. We intend to finish our study by around the year 2022.

Conclusions

Auto-ML will generalize to various clinical prediction/classification problems. With minimal help from data scientists, health care researchers can use Auto-ML to quickly build high-quality models. This will boost wider use of machine learning in health care and improve patient outcomes.

Return on Investment: Evaluating the Evidence Regarding Financial Outcomes of Workplace Wellness Programs

Workplace wellness programs are expected to reduce employee healthcare costs, increase productivity, and provide a positive return on investment. A review of the literature from 2000 to 2016 was conducted to determine whether workplace wellness programs deliver a positive economic impact. Individual financial metrics and results varied; 6 of 7 studies reported a positive economic impact. Additional study is recommended because of the high-degree variability and lack of longitudinal data.

Strategies for Worksite Health Interventions to Employees with Elevated Risk of Chronic Diseases

Chronic disease rates have become more prevalent in the modern American workforce, which has negative implications for workplace productivity and healthcare costs. Offering workplace health interventions is recognized as an effective strategy to reduce chronic disease progression, absenteeism, and healthcare costs as well as improve population health. This review documents intervention and evaluation strategies used for health promotion programs delivered in workplaces. Using predetermined search terms in five online databases, we identified 1,131 published items from 1995 to 2014. Of these items, 27 peer-reviewed articles met the inclusion criteria; reporting data from completed United States-based workplace interventions that recruited at-risk employees based on their disease or disease-related risk factors. A content rubric was developed and used to catalogue these 27 published field studies. Selected workplace interventions targeted obesity (n = 13), cardiovascular diseases (n = 8), and diabetes (n = 6). Intervention strategies included instructional education/counseling (n = 20), workplace environmental change (n = 6), physical activity (n = 10), use of technology (n = 10), and incentives (n = 13). Self-reported data (n = 21), anthropometric measurements (n = 17), and laboratory tests (n = 14) were used most often in studies with outcome evaluation. This is the first literature review to focus on interventions for employees with elevated risk for chronic diseases. The review has the potential to inform future workplace health interventions by presenting strategies related to implementation and evaluation strategies in workplace settings. These strategies can help determine optimal worksite health programs based on the unique characteristics of work settings and the health risk factors of their employee populations.

Comparative Effectiveness of Wellness Programs: Impact of Incentives on Healthcare Costs for Obese Enrollees

INTRODUCTION

Employee wellness programs show mixed effectiveness results. This study examined the impact of an insurer's lifestyle modification program on healthcare costs of obese individuals.

METHODS

This nonrandomized comparative effectiveness study evaluated changes in healthcare costs for participants in two incentivized programs, an Internet-mediated pedometer-based walking program (WalkingSpree, n=7,594) and an in-person weight-loss program (Weight Watchers, n=5,764). The primary outcome was the change in total healthcare costs from the baseline year to the year after program participation. Data were collected from 2009 to 2011 and the analysis was done in 2014-2015.

RESULTS

After 1 year, unadjusted mean costs decreased in both programs, with larger decreases for Weight Watchers participants than WalkingSpree participants (-$1,055.39 vs -$577.10, p=0.019). This difference was driven by higher rates of women in Weight Watchers, higher baseline total costs among women, and a greater decrease in costs for women in Weight Watchers (-$1,037.60 vs -$388.50, p=0.014). After adjustment for baseline costs, there were no differences by program or gender.

CONCLUSIONS

Comparable cost reductions in both programs suggest that employers may want to offer more than one choice of incentivized wellness program with monitoring to meet the diverse needs of employees.

What's Mine is Yours: Evaluation of Shared Well-Being Among Married Couples and the Dyadic Influence on Individual Well-Being Change

OBJECTIVE

To evaluate the relationship between partner well-being and outcomes of chronically diseased individuals participating in an employer sponsored well-being improvement program.

METHODS

Using the Actor Partner Interdependence Model, we evaluated whether prior partner well-being was associated with well-being change among 2025 couples. Logistic regression models were then used to explore how spousal well-being risks relate to development and elimination of risks among program participants.

RESULTS

High well-being partners were associated with positive well-being change. Specifically, the partner effect for spouses' high well-being on disease management participants was a 1.5 point higher well-being in the following time period (P = 0.001) while the partner effect of participants' high well-being on spouses was nearly 1.1 points (P = 0.010).

CONCLUSIONS

Well-being within couples is interdependent, and partner well-being is an important predictor of individual well-being change.

Statistical analysis of a low cost method for multiple disease prediction

Early identification of individuals at risk for chronic diseases is of significant clinical value. Early detection provides the opportunity to slow the pace of a condition, and thus help individuals to improve or maintain their quality of life. Additionally, it can lessen the financial burden on health insurers and self-insured employers. As a solution to mitigate the rise in chronic conditions and related costs, an increasing number of employers have recently begun using wellness programs, which typically involve an annual health risk assessment. Unfortunately, these risk assessments have low detection capability, as they should be low-cost and hence rely on collecting relatively few basic biomarkers. Thus one may ask, how can we select a low-cost set of biomarkers that would be the most predictive of multiple chronic diseases? In this paper, we propose a statistical data-driven method to address this challenge by minimizing the number of biomarkers in the screening procedure while maximizing the predictive power over a broad spectrum of diseases. Our solution uses multi-task learning and group dimensionality reduction from machine learning and statistics. We provide empirical validation of the proposed solution using data from two different electronic medical records systems, with comparisons over a statistical benchmark.

And Why So Great a "No?": The Donor and Academic Communities' Failure to Confront Global Chronic Disease

Chronic diseases are the dominant issues for global public health in terms of mortality, morbidity, and cost, and they have been identified as such for >40 years. Despite their predominance, however, these diseases-cardiovascular disease (CVD), diabetes, cancer, pulmonary disease, mental health, and dementia-attract little attention in the public health curriculum and even less from the funding community. We explore the rationales that have perpetuated this inability or unwillingness to match need with effort. We examine 3 concepts that impede changing this relationship: 1) the traditional contextual view of public health that emerged, to be sure with great success, in the post-World War II era; 2) the failure of public health to transition to economic development as the goal of health assistance; and 3) the unwillingness of public health to confront social, political, and economic policies as the foci of upstream drivers of the public's health. We conclude with a discussion of the need for public health to expand its horizon and tear down the walls of the silos that inhibit the emergence of relevant global public health.

Perceived workplace health support is associated with employee productivity

PURPOSE

To examine the relationship between perceived workplace health support and employee productivity.

DESIGN

A quantitative cross-sectional study.

SETTING

Washington State agencies.

SUBJECTS

A total of 3528 employees from six state agencies were included in this analysis.

MEASURES

Perceived workplace health support was assessed by two questions that queried respondents on how often they felt supported by the workplace for healthy living and physical activity. The Work Productivity and Activity Impairment Questionnaire was used to measure health-related absenteeism and presenteeism in the past 7 days.

ANALYSIS

Multivariate linear regression was used to estimate the mean differences in productivity by levels of perceived health support.

RESULTS

Most participants were between 45 and 64 years of age and were predominantly non-Hispanic white. Presenteeism varied significantly by the level of perceived workplace health support, with those who felt least supported having higher presenteeism than those who felt most supported. The difference in presenteeism by perceived workplace support remained significant in models adjusting for sociodemographic and health characteristics (mean difference: 7.1% for support for healthy living, 95% confidence interval: 3.7%, 10.4%; 4.3% for support for physical activity, 95% confidence interval: 1.7%, 6.8%). Absenteeism was not associated with perceived workplace health support.

CONCLUSION

Higher perceived workplace health support is independently associated with higher work productivity. Employers may see productivity benefit from wellness programs through improved perceptions of workplace health support.

Promoting Healthy Workplaces by Building Cultures of Health and Applying Strategic Communications

OBJECTIVE

The aim of the study was to identify key success elements of employer-sponsored health promotion (wellness) programs.

METHODS

We conducted an updated literature review, held discussions with subject matter experts, and visited nine companies with exemplary programs to examine current best and promising practices in workplace health promotion programs.

RESULTS

Best practices include establishing a culture of health and using strategic communications. Key elements that contribute to a culture of health are leadership commitment, social and physical environmental support, and employee involvement. Strategic communications are designed to educate, motivate, market offerings, and build trust. They are tailored and targeted, multichanneled, bidirectional, with optimum timing, frequency, and placement.

CONCLUSIONS

Increased efforts are needed to disseminate lessons learned from employers who have built cultures of health and excellent communications strategies and apply these insights more broadly in workplace settings.