Geographic disparities in telemedicine mental health use by applying three way ANOVA on Medicaid claims population data

BACKGROUND

Utilization of telemedicine care for vulnerable and low income populations, especially individuals with mental health conditions, is not well understood. The goal is to describe the utilization and regional disparities of telehealth care by mental health status in Texas. Texas Medicaid claims data were analyzed from September 1, 2012, to August 31, 2018 for Medicaid patients enrolled due to a disability.

METHODS

We analyzed the growth in telemedicine care based on urban, suburban, and rural, and mental health status. We used t-tests to test for differences in sociodemographic characteristics across patients and performed a three-way Analyses of Variance (ANOVA) to evaluate whether the growth rates from 2013 to 2018 were different based on geography and patient type. We then estimated patient level multivariable ordinary least square regression models to estimate the relationship between the use of telemedicine and patient characteristics in 2013 and separately in 2018. Outcome was a binary variable of telemedicine use or not. Independent variables of interest include geography, age, gender, race, ethnicity, plan type, Medicare eligibility, diagnosed mental health condition, and ECI score.

RESULTS

Overall, Medicaid patients with a telemedicine visit grew at 81%, with rural patients growing the fastest (181%). Patients with a telemedicine visit for a mental health condition grew by 77%. Telemedicine patients with mental health diagnoses tended to have 2 to 3 more visits per year compared to non-telemedicine patients with mental health diagnoses. In 2013, multivariable regressions display that urban and suburban patients, those that had a mental health diagnosis were more likely to use telemedicine, while patients that were younger, women, Hispanics, and those dual eligible were less likely to use telemedicine. By 2018, urban and suburban patients were less likely to use telemedicine.

CONCLUSIONS

Growth in telemedicine care was strong in urban and rural areas between 2013 and 2018 even before the COVID-19 pandemic. Those with a mental health condition who received telemedicine care had a higher number of total mental health visits compared to those without telemedicine care. These findings hold across all geographic groups and suggest that mental health telemedicine visits did not substitute for face-to-face mental health visits.

Racial and Ethnic Disparities in Hepatocellular Carcinoma Treatment Receipt in the United States: A Systematic Review and Meta-Analysis

BACKGROUND

Racial and ethnic disparities in hepatocellular carcinoma (HCC) prognosis exist, partly related to differential failures along the cancer care continuum. We characterized racial and ethnic disparities in treatment receipt among patients with HCC in the United States.

METHODS

We searched Medline, Embase, and CINAHL databases to identify studies published between January 2012 and March 2022 reporting HCC treatment receipt among adult patients with HCC, stratified by race or ethnicity. We calculated pooled odds ratios for HCC treatment using random effects models.

RESULTS

We identified 15 studies with 320,686 patients (65.8% White, 13.9% Black, 10.4% Asian, and 8.5% Hispanic). Overall, 33.2% of HCC patients underwent any treatment, and 22.7% underwent curative treatment. Compared with White patients, Black patients had lower odds of any treatment (OR 0.67, 95% CI 0.55-0.81) and curative treatment (OR 0.74, 95% CI 0.71-0.78). Similarly, Hispanic patients had lower pooled odds of curative treatment (OR 0.79, 95% CI 0.73-0.84).

CONCLUSIONS

There were significant racial and ethnic disparities in HCC treatment receipt, with Black patients having lower odds of receiving any and curative treatment while Hispanic patients having lower odds of curative treatment.

IMPACT

Racial and ethnic differences in treatment receipt serve as an intervention target to reduce disparities in HCC prognosis.

Case study on communicating with research ethics committees about minimizing risk through software: an application for record linkage in secondary data analysis

Objective

In retrospective secondary data analysis studies, researchers often seek waiver of consent from institutional Review Boards (IRB) and minimize risk by utilizing complex software. Yet, little is known about the perspectives of IRB experts on these approaches. To facilitate effective communication about risk mitigation strategies using software, we conducted two studies with IRB experts to co-create appropriate language when describing a software to IRBs.

Materials and Methods

We conducted structured focus groups with IRB experts to solicit ideas on questions regarding benefits, risks, and informational needs. Based on these results, we developed a template IRB application and template responses for a generic study using privacy-enhancing software. We then conducted a three-round Delphi study to refine the template IRB application and the template responses based on expert panel feedback. To facilitate participants' deliberation, we shared the revisions and a summary of participants' feedback during each Delphi round.

Results

11 experts in two focus groups generated 13 ideas on risks, benefits, and informational needs. 17 experts participated in the Delphi study with 13 completing all rounds. Most agreed that privacy-enhancing software will minimize risk, but regardless all secondary data studies have an inherent risk of unexpected disclosures. The majority (84.6%) noted that subjects in retrospective secondary data studies experience no greater risks than the risks experienced in ordinary life in the modern digital society. Hence, all retrospective data-only studies with no contact with subjects would be minimal risk studies.

Conclusion

First, we found fundamental disagreements in how some IRB experts view risks in secondary data research. Such disagreements are consequential because they can affect determination outcomes and might suggest IRBs at different institutions might come to different conclusions regarding similar study protocols. Second, the highest ranked risks and benefits of privacy-enhancing software in our study were societal rather than individual. The highest ranked benefits were facilitating more research and promoting responsible data governance practices. The highest ranked risks were risk of invalid results from systematic user error or erroneous algorithms. These societal considerations are typically more characteristic of public health ethics as opposed to the bioethical approach of research ethics, possibly reflecting the difficulty applying a bioethical approach (eg, informed consent) in secondary data studies. Finally, the development of privacy-enhancing technology for secondary data research depends on effective communication and collaboration between the privacy experts and technology developers. Privacy is a complex issue that requires a holistic approach that is best addressed through privacy-by-design principles. Privacy expert participation is important yet often neglected in this design process. This study suggests best practice strategies for engaging the privacy community through co-developing companion documents for software through participatory design to facilitate transparency and communication. In this case study, the final template IRB application and responses we released with the open-source software can be easily adapted by researchers to better communicate with their IRB when using the software. This can help increase responsible data governance practices when many software developers are not research ethics experts.

Determinants of Persistent, Multi-Year, Frequent Emergency Department Use Among Children and Young Adults in Three US States

OBJECTIVE

This study examines the factors associated with persistent, multi-year, and frequent emergency department (ED) use among children and young adults.

METHODS

We conducted a retrospective secondary analysis using the 2012-2017 Healthcare Cost and Utilization Project State Emergency Department Databases for children and young adults aged 0-19 who visited any ED in Florida, Massachusetts, and New York. We estimated the association between persistent frequent ED use and individuals' characteristics using multivariable logistic regression models.

RESULTS

Among 1.3 million patients with 1.8 million ED visits in 2012, 2.9% (37,558) exhibited frequent ED use (≥4 visits in 2012) and accounted for 10.2% (181,138) of all ED visits. Longitudinal follow-up of frequent ED users indicated that 15.4% (5770) remained frequent users periodically over the next 1 or 2 years, while 2.2% (831) exhibited persistent frequent use over the next 3-5 years. Over the 6-year study period, persistent frequent users had 31,551 ED visits at an average of 38.0 (standard deviation = 16.2) visits. Persistent frequent ED use was associated with higher intensity of ED use in 2012, public health insurance coverage, inconsistent health insurance coverage over time, residence in non-metropolitan and lower-income areas, multimorbidity, and more ED visits for less medically urgent conditions.

CONCLUSIONS

Clinicians and policymakers should consider the diverse characteristics and needs of pediatric persistent frequent ED users compared to broader definitions of frequent users when designing and implementing interventions to improve health outcomes and contain ED visit costs.

Data Sharing in a Decentralized Public Health System: Lessons From COVID-19 Syndromic Surveillance

The COVID-19 pandemic revealed that data sharing challenges persist across public health information systems. We examine the specific challenges in sharing syndromic surveillance data between state, local, and federal partners. These challenges are complicated by US federalism, which decentralizes public health response and creates friction between different government units. The current policies restrict federal access to state and local syndromic surveillance data without each jurisdiction's consent. These policies frustrate legitimate federal governmental interests and are contrary to ethical guidelines for public health data sharing. Nevertheless, state and local public health agencies must continue to play a central role as there are important risks in interpreting syndromic surveillance data without understanding local contexts. Policies establishing a collaborative framework will be needed to support data sharing between federal, state, and local partners. A collaborative framework would be enhanced by a governance group with robust state and local involvement and policy guardrails to ensure the use of data is appropriate. These policy and relational challenges must be addressed to actualize a truly national public health information system.

Actual and perceived E-cigarettes behaviors among a national sample of U.S. college students

Objective: Electronic cigarette use represents an important college health concern. This investigation assessed demographic and behavioral correlates associated with actual and perceived e-cigarette use among a national sample of American college students. Methods: Respondents (n = 19,861) comprised college students from over 40 distinct American higher education institutions. Multivariable logistic regression assessed whether (1) alcohol, tobacco or marijuana use were associated with e-cigarette use; (2) perceived peer use of alcohol, tobacco or marijuana were associated with perceived e-cigarette use. Results: Approximately 5% of survey respondents self-reported using e-cigarettes within the past month. More than 7 out of every 10 respondents, however, thought the 'typical student' was an e-cigarette user. As perceptions of typical student substance use increased, respondents were far more likely to contend the typical student used e-cigarettes. Discussion: In addition to adopting smoke-free campus policies, university officials should disseminate accurate information regarding e-cigarette behaviors of students.

Economic Evaluation of Blood Pressure Monitoring Techniques in Patients With Hypertension: A Systematic Review

Importance

Blood pressure monitoring is critical to the timely diagnosis and treatment of hypertension. At-home self-monitoring techniques are highly effective in managing high blood pressure; however, evidence regarding the cost-effectiveness of at-home self-monitoring compared with traditional monitoring in clinical settings remains unclear.

Objective

To identify and synthesize published research examining the cost-effectiveness of at-home blood pressure self-monitoring relative to monitoring in a clinical setting among patients with hypertension.

Evidence Review

A systematic literature search of 5 databases (PubMed, MEDLINE, Embase, EconLit, and CINAHL) followed by a backward citation search was conducted in September 2022. Full-text, peer-reviewed articles in English including patients with high blood pressure (systolic blood pressure ≥130 mm Hg and diastolic blood pressure ≥80 mm Hg) at baseline were included. Data from studies comparing at-home self-monitoring with clinical-setting monitoring alternatives were extracted, and the outcomes of interest included incremental cost-effectiveness and cost-utility ratios. Non-peer-reviewed studies or studies with pregnant women and children were excluded. To ensure accuracy and reliability, 2 authors independently evaluated all articles for eligibility and extracted relevant data from the selected articles.

Findings

Of 1607 articles identified from 5 databases, 16 studies met the inclusion criteria. Most studies were conducted in the US (6 [40%]) and in the UK (6 [40%]), and almost all studies (14 [90%]) used a health care insurance system perspective to determine costs. Nearly half the studies used quality-adjusted life-years gained and cost per 1-mm Hg reduction in blood pressure as outcomes. Overall, at-home blood pressure monitoring (HBPM) was found to be more cost-effective than monitoring in a clinical setting, particularly over a minimum 10-year time horizon. Among studies comparing HBPM alone vs 24-hour ambulatory blood pressure monitoring (ABPM) or HBPM combined with additional support or team-based care, the latter were found to be more cost-effective.

Conclusions and Relevance

In this systematic review, at-home blood pressure self-monitoring, particularly using automatic 24-hour continuous blood pressure measurements or combined with additional support or team-based care, demonstrated the potential to be cost-effective long-term compared with care in the physical clinical setting and could thus be prioritized for patients with hypertension from a cost-effectiveness standpoint.

Using motor vehicle crash records for injury surveillance and research in agriculture and forestry

PROBLEM

Fatal injuries in the agriculture, forestry, and fishing sector (AgFF) outweigh those across all sectors in the United States. Transportation-related injuries are among the top contributors to these fatal events. However, traditional occupational injury surveillance systems may not completely capture crashes involving farm vehicles and logging trucks, specifically nonfatal events.

METHODS

The study aimed to develop an integrated database of AgFF-related motor-vehicle crashes for the southwest (Arkansas, Louisiana, New Mexico, Oklahoma, and Texas) and to use these data to conduct surveillance and research. Lessons learned during the pursuit of these aims were cataloged. Activities centered around the conduct of traditional statistical and geospatial analyses of structured data fields and natural language processing of free-text crash narratives.

RESULTS

The structured crash data in each state include fields that allowed farm vehicles or equipment and logging trucks to be identified. The variable definitions and coding were not consistent across states but could be harmonized. All states recorded data fields pertaining to person, vehicle, and crash/environmental factors. Structured data supported the construction of crash severity models and geospatial analyses. Law enforcement provided additional details on crash causation in free-text narratives. Crash narratives contained sufficient text to support viable machine learning models for farm vehicle or equipment crashes, but not for logging truck narratives.

DISCUSSION

Crash records can help to fill research and surveillance gaps in AgFF in the southwest region. This supports traffic safety's evolution to the current Safe System paradigm. There is a conceptual linkage between the Safe System and Total Worker Health approaches, providing a bridge between traffic safety and occupational health.

PRACTICAL APPLICATIONS

Despite limitations, crash records can be an important component of injury surveillance for events involving AgFF vehicles. They also can be used to inform the selection and evaluation of traffic countermeasures and behavioral interventions.

Real-World Adherence and Effectiveness of Remote Patient Monitoring Among Medicaid Patients With Diabetes: Retrospective Cohort Study

BACKGROUND

The prevalence of diabetes in the United States is high and increasing, and it is also the most expensive chronic condition in the United States. Self-monitoring of blood glucose or continuous glucose monitoring are potential solutions, but there are barriers to their use. Remote patient monitoring (RPM) with appropriate support has the potential to provide solutions.

OBJECTIVE

We aim to investigate the adherence of Medicaid patients with diabetes to daily RPM protocols, the relationship between adherence and changes in blood glucose levels, and the impact of daily testing time on blood glucose changes.

METHODS

This retrospective cohort study analyzed real-world data from an RPM company that provides services to Texas Medicaid patients with diabetes. Overall, 180 days of blood glucose data from an RPM company were collected to assess transmission rates and blood glucose changes, after the first 30 days of data were excluded due to startup effects. Patients were separated into adherent and nonadherent cohorts, where adherent patients transmitted data on at least 120 of the 150 days. z tests and t tests were performed to compare transmission rates and blood glucose changes between 2 cohorts. In addition, we analyzed blood glucose changes based on their testing time-between 1 AM and 10 AM, 10 AM and 6 PM, and 6 PM and 1 AM.

RESULTS

Mean patient age was 70.5 (SD 11.8) years, with 66.8% (n=255) of them being female, 91.9% (n=351) urban, and 89% (n=340) from south Texas (n=382). The adherent cohort (n=186, 48.7%) had a mean transmission rate of 82.8% before the adherence call and 91.1% after. The nonadherent cohort (n=196, 51.3%) had a mean transmission rate of 45.9% before and 60.2% after. The mean blood glucose levels of the adherent cohort decreased by an average of 9 mg/dL (P=.002) over 5 months. We also found that variability of blood glucose level of the adherent cohort improved 3 mg/dL (P=.03) over the 5-month period. Both cohorts had the majority of their transmissions between 1 AM and 10 AM, with 70.5% and 53.2% for the adherent and nonadherent cohorts, respectively. The adherent cohort had decreasing mean blood glucose levels over 5 months, with the largest decrease during the 6 PM to 1 AM time period (30.9 mg/dL). Variability of blood glucose improved only for those tested from 10 AM to 6 PM, with improvements of 6.9 mg/dL (P=.02). Those in the nonadherent cohort did not report significant changes.

CONCLUSIONS

RPM can help manage diabetes in Medicaid clients by improving adherence rates and glycemic control. Adherence calls helped improve adherence rates, but some patients still faced challenges in transmitting blood glucose levels. Nonetheless, RPM has the potential to reduce the risk of adverse outcomes associated with diabetes.

Adult Cancer-Related Emergency Department Utilization: An Analysis of Trends and Outcomes From Emergency Departments in Maryland and New York

PURPOSE

To explore emergency department (ED) visits by adults with cancer and to estimate associations between inpatient admissions through the ED and mortality with sociodemographic and clinical factors within this cohort.

METHODS

We conducted a retrospective, pooled, cross-sectional analysis of the Healthcare Cost and Utilization State Emergency Department Databases and State Inpatient Databases for Maryland and New York from January 2013 to December 2017. We examined inpatient admissions through the ED and mortality using frequencies. Among patients with cancer, multivariable regressions were used to estimate sociodemographic and clinical factors associated with inpatient admissions and outpatient ED and inpatient mortality overall.

RESULTS

Among 22.7 million adult ED users, 1.3 million (5.7%) had at least one cancer-related diagnosis. ED visit rates per 100,000 population increased annually throughout the study period for patients with cancer and were 9.9% higher in 2017 compared with 2013 (2013: 303.5; 2017: 333.6). Having at least one inpatient admission (68.7% v 20.5%; P < .001) and inpatient or ED mortality (6.5% v 1.0%; P < .001) were higher among ED users with cancer compared with those without. Among patients with cancer, being uninsured (adjusted odds ratio, 0.52; 95% CI, 0.44 to 0.62) compared with having Medicare coverage and non-Hispanic Black (adjusted odds ratio, 0.86; 95% CI, 0.80 to 0.92) compared with non-Hispanic White were associated with decreased odds of inpatient admissions. In contrast, patients with cancer without health insurance, non-Hispanic Black patients, and residents of nonlarge metropolitan areas and of areas with lower household incomes had increased odds of mortality.

CONCLUSION

High inpatient admissions through the ED and mortality among adult patients with cancer, coupled with an increase in cancer-related ED visit rates and observed disparities in outcomes, highlight the need to improve access to oncologic services to contain ED use and improve care for patients with cancer.

Healthcare Costs for Medicare Patients With Hepatocellular Carcinoma in the United States

BACKGROUND & AIMS

Hepatocellular carcinoma (HCC) has an increasing mortality in the United States and is a leading cause of morbidity and mortality in patients with cirrhosis. We aimed to estimate the financial burden related to HCC in a large nationally representative United States cohort.

METHODS

We used the Surveillance, Epidemiology, and End Results program (SEER)-Medicare database to identify 4525 adult patients who were diagnosed with HCC between 2011 and 2015. We generated a 1:1 propensity score-matched cohort of patients with cirrhosis but no HCC as a comparator group to define incremental HCC-specific costs beyond costs related to underlying cirrhosis. Our main outcomes were patient liabilities and Medicare payments in the first year after HCC diagnosis.

RESULTS

Compared with patients with cirrhosis, those with HCC had higher incremental patient liabilities (median +$7166; interquartile range, $2401-$16,099) and Medicare payments (+$50,110; interquartile range, $142,42-$136,239; P < .001 for both) in the first year after diagnosis. Patients with HCC had significantly higher inpatient, outpatient, and physician service costs compared with the matched cohort with cirrhosis (P < .001 for all). Patients with early-stage HCC had lower incremental patient liabilities (median, $4195 vs $8238; P < .001) and Medicare payments (median, $28,207 vs $59,509; P < .001) than those with larger tumor burden. In multivariable median regression analysis, incremental patient liabilities and Medicare payments were significantly associated with the National Cancer Institute comorbidity index, nonalcoholic fatty liver disease etiology, presence of ascites, and presence of hepatic encephalopathy.

CONCLUSIONS

Patients with HCC suffer from cancer-related financial burden, highlighting a need for policy interventions and financial support systems.

Financial Toxicities Persist for Cancer Survivors Irrespective of Current Cancer Status: An Analysis of Medical Expenditure Panel Survey

This study estimates the out-of-pocket (OOP) expenditures for different cancer types among survivors with current vs no current cancer condition and across sex, which is understudied in the literature. This is a cross-sectional study of Medical Expenditure Panel Survey data for 2009-2018 where the primary outcome was the average per year OOP expenditure incurred by cancer survivors. Of 189 285 respondents, 15 010 (7.93%) were cancer survivors; among them, 46.28% had a current cancer condition. Average per year OOP expenditure for female survivors with a current condition of breast cancer ($1730), lung cancer ($1679), colon cancer ($1595), melanoma ($1783), non-Hodgkin lymphoma ($1656), nonmelanoma/other skin cancer (NMSC, $2118) and two or more cancers ($2310) were significantly higher than that of women with no history of cancer ($853, all P < .05). Similarly, average per year OOP expenditure for male survivors with a current condition of prostate cancer ($1457), lung cancer ($1131), colon cancer ($1471), melanoma ($1474), non-Hodgkin's lymphoma ($1653), NMSC ($1789), and bladder cancer ($2157) were significantly higher compared with the men with no history of cancer ($621, all P < .05). These differences persisted in survivors with no current cancer condition for breast cancer among women; prostate, lung, colon, and bladder cancer among men; and melanoma, NMSC, and two or more cancers among both sexes. OOP expenditure varied across cancer types and by sex for survivors with and without a current cancer condition. These findings highlight the need for targeted interventions for cancer survivors.

A Study of Publicly Available Resources Addressing Legal Data-Sharing Barriers: Systematic Assessment

Background

United States data protection laws vary depending on the data type and its context. Data projects involving social determinants of health often concern different data protection laws, making them difficult to navigate.

Objective

We systematically aggregated and assessed useful online resources to help navigate the data-sharing landscape.

Methods

We included publicly available resources that discussed legal data-sharing issues with some health relevance and published between 2010 and 2019. We conducted an iterative search with a common string pattern using a general-purpose search engine that targeted 24 different sectors identified by Data Across Sectors for Health. We scored each online resource for its depth of legal and data-sharing discussions and value for addressing legal barriers.

Results

Out of 3710 total search hits, 2721 unique URLs were reviewed for scope, 322 received full-text review, and 154 were selected for final coding. Legal agreements, consent, and agency guidance were the most widely covered legal topics, with HIPAA (The Health Insurance Portability and Accountability Act), Family Educational Rights and Privacy Act (FERPA), Title 42 of the Code of Federal Regulations Part 2 being the top 3 federal laws discussed. Clinical health care was the most prominent sector with a mention in 73 resources.

Conclusions

This is the first systematic study of publicly available resources on legal data-sharing issues. We found existing gaps where resources covering certain laws or applications may be needed. The volume of resources we found is an indicator that real and perceived legal issues are a substantial barrier to efforts in leveraging data from different sectors to promote health.

Racial, Ethnic, and Socioeconomic Disparities in Treatment Delay Among Patients With Hepatocellular Carcinoma in the United States

BACKGROUND & AIMS

Failures have been reported across the cancer care continuum in patients with hepatocellular carcinoma (HCC); however, the impact of treatment delays on outcomes has not been well-characterized. We described the prevalence of treatment delays in a racially and ethnically diverse cohort of patients and its association with overall survival.

METHODS

Using the Surveillance, Epidemiology, and End Results-Medicare database, we identified patients diagnosed with HCC between 2001 and 2015. We performed multivariable logistic regression analysis to identify factors associated with treatment delay (ie, receipt of HCC-directed therapy >3 months after diagnosis). Cox proportional hazards regression analysis with a 5-month landmark was used to characterize the association between treatment delay and overall survival, accounting for immortal time bias.

RESULTS

Of 8450 patients with treatment within 12 months of HCC diagnosis, 1205 (14.3%) experienced treatment delays. The proportion with treatment delays ranged from 6.8% of patients undergoing surgical resection to 21.6% of those undergoing liver transplantation. In multivariable analysis, Black patients (odds ratio, 1.96; 95% confidence interval [CI], 1.21-3.15) and those living in high poverty neighborhoods (odds ratio, 1.55; 95% CI, 1.25-1.92) were more likely to experience treatment delays than white patients and those living in low poverty neighborhoods, respectively. Treatment delay was independently associated with worse survival (hazard ratio 1.15, 95% CI, 1.05-1.25).

CONCLUSIONS

Nearly 1 in 7 patients with HCC experience treatment delays, with higher odds in Black patients and those living in high poverty neighborhoods. Treatment delays are associated with worse survival, highlighting a need for interventions to improve time-to-treatment.

Predictors of Pediatric Frequent Emergency Department Use Among 7.6 million Pediatric Patients in New York

OBJECTIVE

This study examines the characteristics and factors associated with frequent emergency department (ED) utilization among the pediatric population.

METHODS

We conducted a pooled cross-sectional secondary analysis using the Healthcare Cost and Utilization Project State Emergency and Inpatient Databases on ED visits to all hospitals in New York from 2011 to 2016 by patients aged 0 to 21. We used multivariable logistic and negative binomial regressions to investigate the predictors of multiple ED visits in the pediatric population.

RESULTS

Overall, our study included 7.6 million pediatric patients who accounted for more than 12 million ED visits. Of those, 6.2% of patients were frequent ED users (≥4 visits/year), accounting for 20.8% of all ED visits (5.4 ED visits/year on average). The strongest predictors of frequent ED use were having at least one ED visit related to asthma (aOR = 8.37 [95% CI: 6.34-11.04]), mental health disorders (aOR = 9.67 [95% CI: 8.60-10.89]), or multiple comorbidities compared to none. Larger shares of ED visits for not-emergent conditions were also associated with frequent ED use (aOR = 6.63 [95% CI = 5.08-8.65]). Being covered by Medicaid compared to private (aOR = 0.45 [95% CI: 0.42-0.47]) or no insurance (aOR = 0.41 [95% CI: 0.38-0.44]) were further associated with frequent ED use. The results from the negative binomial regression yielded consistent findings.

CONCLUSIONS

Pediatric patients who exhibit increased ED use are more medically complex and have increased healthcare needs that are inextricably tied to social determinants of health. Better integrated health systems should emphasize connecting vulnerable patients to appropriate social and primary care services outside of emergency settings.

Correction: Adherence to Telemonitoring Therapy for Medicaid Patients With Hypertension: Case Study

[This corrects the article DOI: 10.2196/29018.].

A Systematic Review of Patient-Perceived Barriers and Facilitators to the Adoption and Use of Remote Health Technology to Manage Diabetes and Cardiovascular Disease among Disproportionately Affected Populations

As noncontact health interventions have become critical during the Covid-19 pandemic, our study aimed to systematically review the published literature for barriers and facilitators influencing the adoption and use of remote health intervention and technology, as perceived by adult patients with diabetes or cardiovascular diseases (CVD) belonging to groups that are socially/economically marginalized and/or medically under-resourced. We searched Medline, Embase, CINAHL, and PsychINFO for peer-reviewed articles published from 2010 to 2018. We employed content analysis to analyze qualitative patient feedback from the included studies. We adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines. A total of 42 studies met the inclusion criteria. The design of the remote health technology used was the most frequently mentioned facilitator and barrier to remote health technology adoption and use. Our results should draw the attention of technology developers to the usability and feasibility of remote technology among populations that are socially/economically marginalized and/or medically under-resourced.

The Effect of the Medicaid Expansion on Frequent Emergency Department Use in New York

BACKGROUND

There is limited evidence on the effect of the Affordable Care Act (ACA) on frequent emergency department (ED) use.

OBJECTIVES

To estimate the effect of the ACA Medicaid expansion on frequent ED use in New York.

METHODS

We used data from the Healthcare Cost and Utilization Project State Emergency Department Databases and State Inpatient Databases from 2011 to 2016. A consistent and unique patient identifier enabled us to identify ED visits by the same patient across different facilities within the state for each calendar year. Multivariate logistic regressions were used to quantify the policy's effect on frequent ED use (≥ 4 ED visits/year). We included in-state residents 18 to 64 years of age who were covered by Medicaid, private insurance, or were uninsured. Sensitivity analyses were conducted using alternative definitions of frequent use. To validate the findings, a falsification analysis was also conducted using only the 3 pre-expansion years.

RESULTS

Our study included 14.3 million ED patients with 23.8 million ED visits from 2011 to 2016. Frequent users (7.2%) accounted for 26.6% of all ED visits. The likelihood of frequent ED use declined by 4% among Medicaid beneficiaries (adjusted odds ratio [AOR] 0.96, 95% confidence intervals (CI) 0.95-0.97) and by 12% for the uninsured (AOR 0.88, 95% CI 0.86-0.89) in the post-expansion period, compared with the pre-expansion period. Private insurance enrollees were 9% more likely to exhibit frequent use in the post-expansion period (AOR 1.09, 95% CI 1.08-1.11). The sensitivity analyses yielded results similar to those of the main model. The falsification analyses revealed small and insignificant year-to-year changes in the 3 pre-expansion years.

CONCLUSION

The likelihood of frequent ED use decreased 3 years after New York implemented the ACA Medicaid expansion, particularly for Medicaid beneficiaries and the uninsured, highlighting the importance of expanding health insurance and provisions tailored at high-need populations.

Adherence to Telemonitoring Therapy for Medicaid Patients With Hypertension: Case Study

Background

Almost 50% of the adults in the United States have hypertension. Although clinical trials indicate that home blood pressure monitoring can be effective in managing hypertension, the reported results might not materialize in practice because of patient adherence problems.

Objective

The aims of this study are to characterize the adherence of Medicaid patients with hypertension to daily telemonitoring, identify the impacts of adherence reminder calls, and investigate associations with blood pressure control.

Methods

This study targeted Medicaid patients with hypertension from the state of Texas. A total of 180 days of blood pressure and pulse data in 2016-2018 from a telemonitoring company were analyzed for mean transmission rate and mean blood pressure change. The first 30 days of data were excluded because of startup effects. The protocols required the patients to transmit readings by a specified time daily. Patients not transmitting their readings received an adherence reminder call to troubleshoot problems and encourage transmission. The patients were classified into adherent and nonadherent cohorts; adherent patients were those who transmitted data on at least 80% of the days.

Results

The mean patient age was 73.2 (SD 11.7) years. Of the 823 patients, 536 (65.1%) were women, and 660 (80.2%) were urban residents. The adherent cohort (475/823, 57.7%) had mean transmission rates of 74.9% before the adherence reminder call and 91.3% after the call, whereas the nonadherent cohort (348/823, 42.3%) had mean transmission rates of 39% and 58% before and after the call, respectively. From month 1 to month 5, the transmission rates dropped by 1.9% and 10.2% for the adherent and nonadherent cohorts, respectively. The systolic and diastolic blood pressure values improved by an average of 2.2 and 0.7 mm Hg (P<.001 and P=.004), respectively, for the adherent cohort during the study period, whereas only the systolic blood pressure value improved by an average of 1.6 mm Hg (P=.02) for the nonadherent cohort.

Conclusions

Although we found that patients can achieve high levels of adherence, many experience adherence problems. Although adherence reminder calls help, they may not be sufficient. Telemonitoring lowered blood pressure, as has been observed in clinical trials. Furthermore, blood pressure control was positively associated with adherence.

Health services utilization, out-of-pocket expenditure, and underinsurance among insured non-elderly cancer survivors in the United States, 2011-2015

BACKGROUND

High out-of-pocket (OOP) expenditure and inadequate insurance coverage may adversely affect cancer survivors. We aimed to characterize the extent and correlates of healthcare utilization, OOP expenditures, and underinsurance among insured cancer survivors.

METHODS

We used 2011-2015 Medical Expenditure Panel Survey data to identify a nationally representative sample of insured non-elderly adult (age 18-64 years) cancer survivors. We used negative binomial, two-part (logistic and Generalized Linear Model with log link and gamma distribution), and logistic regression models to quantify healthcare utilization, OOP expenditures, and underinsurance, respectively, and identified sociodemographic correlates for each outcome.

RESULTS

We identified 2738 insured non-elderly cancer survivors. Adjusted average utilization of ambulatory, non-ambulatory, prescription medication, and dental services was 14.4, 0.51, 24.9, and 1.4 events per person per year, respectively. Higher ambulatory and dental services utilization were observed in older adults, females, non-Hispanic Whites, survivors with a college degree and high income, compared to their counterparts. Nearly all (97.7%) survivors had some OOP expenditures, with a mean adjusted OOP expenditure of $1552 per person per year. Adjusted mean OOP expenditures for ambulatory, non-ambulatory, prescription medication, dental, and other health services were $653, $161, $428, $194, and $83, respectively. Sociodemographic variations in service-specific OOP expenditures were generally consistent with respective utilization patterns. Overall, 8.8% of the survivors were underinsured.

CONCLUSION

Many insured non-elderly cancer survivors allocate a substantial portion of their OOP expenditure for healthcare-related services and experience financial vulnerability, resulting in nearly 8.8% of the survivors being underinsured. Utilization of healthcare services varies across sociodemographic groups.

Identifying and prioritizing benefits and risks of using privacy-enhancing software through participatory design: a nominal group technique study with patients living with chronic conditions

OBJECTIVE

While patients often contribute data for research, they want researchers to protect their data. As part of a participatory design of privacy-enhancing software, this study explored patients' perceptions of privacy protection in research using their healthcare data.

MATERIALS AND METHODS

We conducted 4 focus groups with 27 patients on privacy-enhancing software using the nominal group technique. We provided participants with an open source software prototype to demonstrate privacy-enhancing features and elicit privacy concerns. Participants generated ideas on benefits, risks, and needed additional information. Following a thematic analysis of the results, we deployed an online questionnaire to identify consensus across all 4 groups. Participants were asked to rank-order benefits and risks. Themes around "needed additional information" were rated by perceived importance on a 5-point Likert scale.

RESULTS

Participants considered "allowance for minimum disclosure" and "comprehensive privacy protection that is not currently available" as the most important benefits when using the privacy-enhancing prototype software. The most concerning perceived risks were "additional checks needed beyond the software to ensure privacy protection" and the "potential of misuse by authorized users." Participants indicated a desire for additional information with 6 of the 11 themes receiving a median participant rating of "very necessary" and rated "information on the data custodian" as "essential."

CONCLUSIONS

Patients recognize not only the benefits of privacy-enhancing software, but also inherent risks. Patients desire information about how their data are used and protected. Effective patient engagement, communication, and transparency in research may improve patients' comfort levels, alleviate patients' concerns, and thus promote ethical research.

The effect of expanded insurance coverage under the Affordable Care Act on emergency department utilization in New York

BACKGROUND

One of the proposed benefits of expanding insurance coverage under the Affordable Care Act (ACA) was a reduction in emergency department (ED) utilization for non-urgent visits related to lack of health insurance coverage and access to primary care providers. The objective of this study was to estimate the effect of the 2014 ACA implementation on ED use in New York.

METHODS

We used the Healthcare Cost and Utilization Project State Emergency Department and State Inpatient Databases for all outpatient and all inpatient visits for patients admitted through an ED from 2011 to 2016. We focused on in-state residents aged 18 to 64, who were covered under Medicaid, private insurance, or were uninsured prior to the 2014 expansion. We estimated the effect of the expanded insurance coverage on average monthly ED visits volumes and visits per 1000 residents (rates) using interrupted time-series regression analyses.

RESULTS

After ACA implementation, overall average monthly ED visits increased by around 3.0%, both in volume (9362; 95% Confidence Intervals [CI]: 1681-17,522) and in rates (0.80, 95% CI:0.12-1.49). Medicaid covered ED visits volume increased by 23,972 visits (95% CI: 16,240 -31,704) while ED visits by the uninsured declined by 13,297 (95% CI:-15,856 - -10,737), and by 1453 (95% CI:-4027-1121) for the privately insured. Medicaid ED visits rates per 1000 residents increased by 0.77 (95% CI:-1.96-3.51) and by 2.18 (95% CI:-0.55-4.92) for those remaining uninsured, while private insurance visits rates decreased by 0.48 (95% CI:-0.79 - -0.18). We observed increases in primary-care treatable ED visits and in visits related to mental health and alcohol disorders, substance use, diabetes, and hypertension. All estimated changes in monthly ED visits after the expansion were statistically significant, except for ED visit rates among Medicaid beneficiaries.

CONCLUSION

Net ED visits by adults 18 to 64 years of age increased in New York after the implementation of the ACA. Large increases in ED use by Medicaid beneficiaries were partially offset by reductions among the uninsured and those with private coverage. Our results suggest that efforts to expand health insurance coverage only will be unlikely to reverse the increase in ED use.

Emergency Department Utilization by Adolescents Experiencing Homelessness in Massachusetts

BACKGROUND

Adolescents who experience homelessness rely heavily on emergency departments (EDs) for their health care.

OBJECTIVES

This study estimates the relationship between homelessness and ED use and identifies the sociodemographic, clinical, visit-level, and contextual factors associated with multiple ED visits among adolescents experiencing homelessness in Massachusetts.

RESEARCH DESIGN

We used the Healthcare Cost and Utilization Project State Emergency Department Databases on all outpatient ED visits in Massachusetts from 2011 to 2016. We included all adolescents who were 11-21 years old. We estimated the association between homelessness and ED utilization and investigated predictors of multiple ED visits among adolescents who experience homelessness using multivariate logistic and negative binomial regressions.

RESULTS

Our study included 1,196,036 adolescents, of whom about 0.8% experienced homelessness and this subset of adolescents accounted for 2.2% of all ED visits. Compared with those with stable housing, adolescents who were homeless were mostly covered through Medicaid (P<0.001), diagnosed with 1 or more comorbidities (P<0.001), and visited the ED at least once for reasons related to mental health; substance and alcohol use; pregnancy; respiratory distress; urinary and sexually transmitted infections; and skin and subcutaneous tissue diseases (P<0.001). Homeless experience was associated with multiple ED visits (incidence rate ratio=1.18; 95% confidence intervals, 1.16-1.19) and frequent ED use (4 or more ED visits) (adjusted odds ratio=2.21; 95% confidence interval, 2.06-2.37). Factors related to clinical complexity and Medicaid compared with lack of coverage were also significant predictors of elevated ED utilization within the cohort experiencing homelessness.

CONCLUSIONS

Adolescents who experience homelessness exhibit higher ED use compared with those with stable housing, particularly those with aggravated comorbidities and chronic conditions. Health policy interventions to integrate health care, housing, and social services are essential to transition adolescents experiencing homelessness to more appropriate community-based care.

Communicating With Patients About Software for Enhancing Privacy in Secondary Database Research Involving Record Linkage: Delphi Study

BACKGROUND

There is substantial prior research on the perspectives of patients on the use of health information for research. Numerous communication barriers challenge transparency between researchers and data participants in secondary database research (eg, waiver of informed consent and knowledge gaps). Individual concerns and misconceptions challenge the trust in researchers among patients despite efforts to protect data. Technical software used to protect research data can further complicate the public's understanding of research. For example, MiNDFIRL (Minimum Necessary Disclosure For Interactive Record Linkage) is a prototype software that can be used to enhance the confidentiality of data sets by restricting disclosures of identifying information during the record linkage process. However, software, such as MiNDFIRL, which is used to protect data, must overcome the aforementioned communication barriers. One proposed solution is the creation of an interactive web-based frequently asked question (FAQ) template that can be adapted and used to communicate research issues to data subjects.

OBJECTIVE

This study aims to improve communication with patients and transparency about how complex software, such as MiNDFIRL, is used to enhance privacy in secondary database studies to maintain the public's trust in researchers.

METHODS

A Delphi technique with 3 rounds of the survey was used to develop the FAQ document to communicate privacy issues related to a generic secondary database study using the MiNDFIRL software. The Delphi panel consisted of 38 patients with chronic health conditions. We revised the FAQ between Delphi rounds and provided participants with a summary of the feedback. We adopted a conservative consensus threshold of less than 10% negative feedback per FAQ section.

RESULTS

We developed a consensus language for 21 of the 24 FAQ sections. Participant feedback demonstrated preference differences (eg, brevity vs comprehensiveness). We adapted the final FAQ into an interactive web-based format that 94% (31/33) of the participants found helpful or very helpful. The template FAQ and MiNDFIRL source code are available on GitHub. The results indicate the following patient communication considerations: patients have diverse and varied preferences; the tone is important but challenging; and patients want information on security, identifiers, and final disposition of information.

CONCLUSIONS

The findings of this study provide insights into what research-related information is useful to patients and how researchers can communicate such information. These findings align with the current understanding of health literacy and its challenges. Communication is essential to transparency and ethical data use, yet it is exceedingly challenging. Developing FAQ template language to accompany a complex software may enable researchers to provide greater transparency when informed consent is not possible.

Crash narrative classification: Identifying agricultural crashes using machine learning with curated keywords

OBJECTIVE

Traditionally, structured or coded data fields from a crash report are the basis for identifying crashes involving different types of vehicles, such as farm equipment. However, using only the structured data can lead to misclassification of vehicle or crash type. The objective of the current article is to examine the use of machine learning methods for identifying agricultural crashes based on the crash narrative and to transfer the application of models to different settings (e.g., future years of data, other states).

METHODS

Different data representations (e.g., bag-of-words [BoW], bag-of-keywords [BoK]) and document classification algorithms (e.g., support vector machine [SVM], multinomial naïve Bayes classifier [MNB]) were explored using Texas and Louisiana crash narratives across different time periods.

RESULTS

The BoK-support vector classifier (SVC), BoK-MNB, and BoW-SVC models trained with Texas data were better predictive models than the baseline rule-based algorithm on the future year test data, with F1 scores of 0.88, 0.89, 0.85 vs. 0.84. The BoK-MNB trained with Louisiana data performed the closest to the baseline rule-based algorithm on the future year test data (F1 scores, 0.91 baseline rule-based algorithm vs. 0.89 BoK-MNB). The BoK-SVC and BoK-MNB models trained with Texas and Louisiana data were better productive models for Texas future year test data with F1 scores 0.89 and 0.90 vs. 0.84. The BoK-MNB model trained with both states' data was a better predictive model for the Louisiana future year test data, F1 score 0.94 vs. 0.91.

CONCLUSIONS

The findings of this study support that machine learning methodologies can potentially reduce the amount of human power required to develop key word lists and manually review narratives.

US Privacy Laws Go Against Public Preferences: Impeding Public Health and Research (Preprint)

Background

Reaping the benefits from massive volumes of data collected in all sectors to improve population health, inform personalized medicine, and transform biomedical research requires the delicate balance between the benefits and risks of using individual-level data. There is a patchwork of US data protection laws that vary depending on the type of data, who is using it, and their intended purpose. Differences in these laws challenge big data projects using data from different sources. The decisions to permit or restrict data uses are determined by elected officials; therefore, constituent input is critical to finding the right balance between individual privacy and public benefits.

Objective

This study explores the US public’s preferences for using identifiable data for different purposes without their consent.

Methods

We measured data use preferences of a nationally representative sample of 504 US adults by conducting a web-based survey in February 2020. The survey used a choice-based conjoint analysis. We selected choice-based conjoint attributes and levels based on 5 US data protection laws (Health Insurance Portability and Accountability Act, Family Educational Rights and Privacy Act, Privacy Act of 1974, Federal Trade Commission Act, and the Common Rule). There were 72 different combinations of attribute levels, representing different data use scenarios. Participants were given 12 pairs of data use scenarios and were asked to choose the scenario they were the most comfortable with. We then simulated the population preferences by using the hierarchical Bayes regression model using the ChoiceModelR package in R.

Results

Participants strongly preferred data reuse for public health and research than for profit-driven, marketing, or crime-detection activities. Participants also strongly preferred data use by universities or nonprofit organizations over data use by businesses and governments. Participants were fairly indifferent about the different types of data used (health, education, government, or economic data).

Conclusions

Our results show a notable incongruence between public preferences and current US data protection laws. Our findings appear to show that the US public favors data uses promoting social benefits over those promoting individual or organizational interests. This study provides strong support for continued efforts to provide safe access to useful data sets for research and public health. Policy makers should consider more robust public health and research data use exceptions to align laws with public preferences. In addition, policy makers who revise laws to enable data use for research and public health should consider more comprehensive protection mechanisms, including transparent use of data and accountability.

Abstract 4340: Out-of-pocket expenditure, probability of underinsurance and health services utilization among non-elderly cancer survivors

Background: Financial burden among cancer survivors by type of health services utilized is an under-explored topic. We estimated health services utilization, out-of-pocket (OOP) expenditure, and underinsurance for a large cohort of cancer survivors. Methods: We used Medical Expenditure Panel Survey (MEPS) Household component (2011-2015) and accounted for MEPS survey design. Underinsurance was defined as OOP≥5% of family income (FI) for FI&lt;200% of federal poverty level (FPL) or ≥10% of FI for FI≥200% of FPL, for full year insured individuals. We identified cancer survivors using Clinical Classifications Software (CCS) codes and excluded uninsured or intermittently insured (n=2738). OOP, underinsurance and utilization were estimated using two-part regression (logistic and GLM), logistic regression, and negative binomial regression models respectively. Average Adjusted Prediction (AAP) and Average Marginal Effect (AME) measures were used to quantify all estimates. Predictors in the models were: age, sex, race/ethnicity, marital status, education, income (excluded in underinsurance model), insurance status, number of comorbid conditions, health status and census region (all categorical variables). Service categories analyzed were: ambulatory (office based + outpatient), non-ambulatory (inpatient + ER), prescription medication, dental care and other services. Results: The category-wise services utilization and OOP analysis revealed that patients aged 60-64 years utilize (1.59 vs. 1.16 events, p&lt;0.01) and spend more ($239 vs. $146, p&lt;0.05) on dental services vs. those aged 18-49 years; females utilize (16 vs. 12 events, p&lt;0.01) and spend more ($768 vs. $575, p&lt;0.05) on ambulatory services vs. males; Blacks and Asians utilize (21 and 21 vs. 26 events, p&lt;0.01 and p&lt;0.05 respectively) and spend less ($298 and $288 vs. $479, p&lt;0.01 for both) on prescriptions vs. Whites; and patients with a college degree utilize (16 vs. 12 events, p&lt;0.01) and spend more ($827 vs. $561, p&lt;0.01) on ambulatory services vs. those with less education. Medicare-only insurance was associated with higher prescription OOP. Underinsurance was significantly higher in older patients and lower among Blacks and married patients (Table 1). Conclusion: Subgroups of non-elderly cancer survivors utilize several health services at significantly higher levels than their counterparts, resulting in underinsurance and high OOP cost.

Table 1:Probability of underinsurance by sociodemographicsFactorLevelProb. of Underinsurance %)page range18-49 [ref.]750-5990.16960-64130.000race/ethnicitywhite [ref.]10black50.001hispanic100.961asian/others60.135marital statusnor married [ref.]16married60.000

Citation Format: Mohammad Karim, Robert L. Ohsfeldt, Amit G. Singal, Hye-Chung Kum. Out-of-pocket expenditure, probability of underinsurance and health services utilization among non-elderly cancer survivors [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 4340.

Post-acute care referral in United States of America: a multiregional study of factors associated with referral destination in a cohort of patients with coronary artery bypass graft or valve replacement

Background

The use of post-acute care (PAC) for cardiovascular conditions is highly variable across geographical regions. Although PAC benefits include lower readmission rates, better clinical outcomes, and lower mortality, referral patterns vary widely, raising concerns about substandard care and inflated costs. The objective of this study is to identify factors associated with PAC referral decisions at acute care discharge.

Methods

This study is a retrospective Electronic Health Records (EHR) based review of a cohort of patients with coronary artery bypass graft (CABG) and valve replacement (VR). EHR records were extracted from the Cerner Health-Facts Data warehouse and covered 49 hospitals in the United States of America (U.S.) from January 2010 to December 2015. Multinomial logistic regression was used to identify associations of 29 variables comprising patient characteristics, hospital profiles, and patient conditions at discharge.

Results

The cohort had 14,224 patients with mean age 63.5 years, with 10,234 (71.9%) male and 11,946 (84%) Caucasian, with 5827 (40.96%) being discharged to home without additional care (Home), 5226 (36.74%) to home health care (HHC), 1721 (12.10%) to skilled nursing facilities (SNF), 1168 (8.22%) to inpatient rehabilitation facilities (IRF), 164 (1.15%) to long term care hospitals (LTCH), and 118 (0.83%) to other locations. Census division, hospital size, teaching hospital status, gender, age, marital status, length of stay, and Charlson comorbidity index were identified as highly significant variables (p- values < 0.001) that influence the PAC referral decision. Overall model accuracy was 62.6%, and multiclass Area Under the Curve (AUC) values were for Home: 0.72; HHC: 0.72; SNF: 0.58; IRF: 0.53; LTCH: 0.52, and others: 0.46.

Conclusions

Census location of the acute care hospital was highly associated with PAC referral practices, as was hospital capacity, with larger hospitals referring patients to PAC at a greater rate than smaller hospitals. Race and gender were also statistically significant, with Asians, Hispanics, and Native Americans being less likely to be referred to PAC compared to Caucasians, and female patients being more likely to be referred than males. Additional analysis indicated that PAC referral practices are also influenced by the mix of PAC services offered in each region.

The Effects of Chronic Disease on Ambulatory Care-Sensitive Hospitalizations for Children or Youth

Considerable research has focused on hospitalizations for ambulatory care–sensitive conditions (ACSHs), but little of that research has focused on the role played by chronic disease in ACSHs involving children or youth (C/Y). This research investigates, for C/Y, the effects of chronic disease on the likelihood of an ACSH. The database included 699 473 hospital discharges for individuals under 18 in Texas between 2011 and 2015. Effects of chronic disease, individual, and contextual factors on the likelihood of a discharge involving an ACSH were estimated using logistic regression. Contrary to the results for adults, the presence of chronic diseases or a complex chronic disease among children or youth was protective, reducing the likelihood of an ACSH for a nonchronic condition. Results indicate that heightened ambulatory care received by C/Y with chronic diseases is largely protective. Two of more chronic conditions or at least one complex chronic condition significantly reduced the likelihood of an ACSH.

Characteristics and predictors of adult frequent emergency department users in the United States: A systematic literature review

STUDY OBJECTIVE

We conducted a systematic literature review to identify and to update patient characteristics and contextual factors for adult frequent emergency department users (FEDUs) compared with non-FEDU in an era where the US health care system underwent substantial changes.

METHODS

We searched MEDLINE, CINAHL, and EMBASE to identify all relevant articles after 2010 through July 2018 that describe FEDU. We included US studies on adult FEDU only and excluded studies on specific subgroups of FEDU. We included demographic, clinical, and health care utilization information, and two reviewers independently evaluated the studies using the Joanna Briggs Institute Critical Appraisal tool.

RESULTS

The 11 studies included in the review indicated that FEDU were 4% to 16% of total ED users but accounted for 14% to 47% of ED visits, with six to nine visits per year on average. The majority of FEDU were young or middle-aged adults, females, of low socioeconomic status and high school or less education, with public insurance, multiple primary care provider visits, and chronic conditions. Fair or poor self-perceived health status, unemployment, unmet needs from primary care providers (PCPs), mental health, and substance abuse were predictors of FEDU.

CONCLUSION

FEDUs are disproportionally sicker and are also heavy users of non-ED health care service providers. The limited data for non-ED health services use in facility-specific studies of FEDU may contribute to findings in such studies that complex and unmet needs from PCPs contributed to ED visits. This suggests the need for more comprehensive data analysis beyond a few sites that can inform systemic management approaches.

Self-Reported Quality, Health, and Cost-Related Outcomes of Care Coordination Among Patients with Complex Health Needs

Care coordination is an increasingly popular strategy to help patients with complex health conditions manage their diseases more effectively. The purpose of the current study was to assess associations between patient-reported receipt of care coordination and their experiences of health, health care quality, and cost-related outcomes. Regression analyses of data from 431 patients across 13 Texas providers indicate that patients who reported receiving care coordination had higher odds of perceiving having enough information about how to manage their conditions (OR 2.02, P < .01), having information about education and treatments available (OR 1.87, P < .01), and reporting better access to care (beta = .27, P < .01). Receipt of care coordination was not associated with patients' reports of how up-to-date their doctors were about care from other providers, patients' health-related functioning, or patients' intention to return to the emergency department. Patients' reasons for intending to return to the emergency department included the speed of care there relative to alternatives and satisfaction with the quality of care they had received previously in the emergency department. Results suggest that care coordination in usual practice may improve patient preventive care, but not some other health or cost-related outcomes.

A semi-automated tool for identifying agricultural roadway crashes in crash narratives

Objective: Crash reports contain precoded structured data fields and a crash narrative that can be a source of rich information not included in the structured data. The narrative can be useful for identifying vulnerable roadway users, such as agricultural workers. However, using the narratives often requires manual reviews that are time consuming and costly. The objective of this research was to develop a simple and relatively inexpensive, semi-automated tool for screening crash narratives and expediting the process of identifying crashes with specific characteristics, such as agricultural crashes. Methods: Crash records for Louisiana from 2010 to 2015 were obtained from the Louisiana Department of Transportation (LaDOTD). Records with narratives were extracted and stratified by vehicle type. The majority of analyses focused on a vehicle type of farm equipment (Type T). Two keyword lists, an inclusion list and an exclusion list, were created based on the published literature, subject-matter experts, and findings from a pilot project. Next, a semi-automated tool was developed in Microsoft Excel to identify agricultural crashes. Lastly, the tool's performance was assessed using a gold standard set of agricultural narratives identified through manual review. Results: The tool reduced the search space (e.g., number of narratives that need manual review) for narratives requiring manual review from 6.7 to 59.4% depending on the research question. Sensitivity was high, with 96.1% of agricultural crash narratives being correctly classified. Of the gold standard agricultural narratives, 58.3% included an equipment keyword and 72.8% included a farm equipment brand. Conclusion: This article provides information on how crash narratives can supplement structured crash data. It also provides an easy-to-implement method to facilitate incorporating narratives into safety research along with keyword lists for identifying agricultural crashes.

Hepatocellular Carcinoma Screening Is Associated With Increased Survival of Patients With Cirrhosis

BACKGROUND & AIMS

Hepatocellular carcinoma (HCC) screening of patients with cirrhosis is recommended by professional societies to increase detection of early stage tumors and survival, but is underused in clinical practice.

METHODS

We conducted a retrospective cohort study of 13,714 patients diagnosed with HCC from 2003 through 2013 included in the Surveillance, Epidemiology, and End Results Program-Medicare database. We characterized receipt of HCC screening in the 3 years before HCC diagnosis using mutually exclusive categories (consistent vs inconsistent vs no screening) and the proportion of time covered with screening. Correlates for screening receipt were assessed using a multivariable 2-part regression model. We examined the association between screening receipt and early detection of tumors using multivariable logistic regression. We evaluated associations between screening receipt and overall survival using a Cox proportional hazards model, after adjustments for effects of lead-time bias and length-time bias on survival rate estimators.

RESULTS

Most patients with cirrhosis (51.1%) did not receive any screening in the 3 years before a diagnosis of HCC, and only 6.8% of patients underwent consistent annual screening. The proportion with consistent screening increased from 5.4% in 2003 to 2006 to 8.8% in 2011 to 2013 (P < .001). The mean proportion of time covered was 13.4% overall, which increased from 11.7% in 2003 to 2006 to 15.2% in 2011 to 2013. Receipt of consistent screening was associated with detection of early stage tumors (odds ratio, 1.98; 95% CI, 1.68-2.33) and a reduced risk of death after correction for lead-time bias (hazard ratio, 0.76; 95% CI, 0.70-0.83). Inconsistent screening was associated with a slightly smaller increase in early detection of HCC (odds ratio, 1.31; 95% CI, 1.20-1.43) and a reduced risk of death (hazard ratio, 0.86; 95% CI, 0.83-0.90). After correction for lead- and length-time biases, higher proportions of patients with consistent (23%; 95% CI, 21%-25%) and inconsistent screening (19%; 95% CI, 19%-20%) survived for 3 years compared with patients without screening (13%; 95% CI, 12%-14%).

CONCLUSIONS

In an analysis of the Surveillance, Epidemiology, and End Results Program-Medicare database, we found HCC screening to be underused for patients with cirrhosis. This contributes to detection of liver tumors at later stages and shorter times of survival. However, the proportion of patients screened for HCC has increased over time.

Assessing the Effect of Clinical Inertia on Diabetes Outcomes: a Modeling Approach

BACKGROUND

There are an increasing number of newer and better therapeutic options in the management of diabetes. However, a large proportion of diabetes patients still experience delays in intensification of treatment to achieve appropriate blood glucose targets-a phenomenon called clinical inertia. Despite the high prevalence of clinical inertia, previous research has not examined its long-term effects on diabetes-related health outcomes and mortality.

OBJECTIVE

We sought to examine the impact of clinical inertia on the incidence of diabetes-related complications and death. We also examined how the impact of clinical inertia would vary by the length of treatment delay and population characteristics.

DESIGN

We developed an agent-based model of diabetes and its complications. The model was parameterized and validated by data from health surveys, cohort studies, and trials.

SUBJECTS

We studied a simulated cohort of patients with diabetes in San Antonio, TX.

MAIN MEASURES

We examined 25-year incidences of diabetes-related complications, including retinopathy, neuropathy, nephropathy, and cardiovascular disease.

KEY RESULTS

One-year clinical inertia could increase the cumulative incidences of retinopathy, neuropathy, and nephropathy by 7%, 8%, and 18%, respectively. The effects of clinical inertia could be worse for populations who have a longer treatment delay, are aged 65 years or older, or are non-Hispanic whites.

CONCLUSION

Clinical inertia could result in a substantial increase in the incidence of diabetes-related complications and mortality. A validated agent-based model can be used to study the long-term effect of clinical inertia and, thus, inform clinicians and policymakers to design effective interventions.

An effective privacy enhanced interface to support record linkage decisions

IntroductionPrivacy enhanced technologies (PET) are those that measure and protect privacy by preventing unnecessary use of personal data without loss of the functionality of the information system. In practice, implementing such a system requires fine-grained access control so that access can be granted in smaller chunks of data. Objectives and ApproachIn record linkage, PET to date has mostly meant separation of identifiers and sensitive information to allow access to only the necessary part. Moving beyond this current norm, we have designed a privacy enhanced interface to support linkage that discloses only the needed information at the sub variable level, when needed to make good decisions, to reduce exposure of personally identifiable information (PII). The system allows for access to PII both at (1) cell level (e.g., only names of needed people are released) or (2) sub-cell level (e.g., only part of a name, suffix or characters, is released). ResultsIn a user study (N=104) where participants tried to link complicated situations (e.g. twins, Sr/Jr, change of last name) using the interface, we found that users given fully masked data, 0% of information disclosed, were still able to get 75% accuracy using supplemental visual markup. The markups depict data discrepancies such as swapped first and last names, transposed characters, different characters and missing data. More importantly, with this effective interface, we found that there were no statistical difference in accuracy of linkage (84%) or time taken between users with access to all data and those with access to only 30% of the data. We have released a tutorial where users can experience balancing between information disclosure and accuracy of results on sample data. Conclusion/ImplicationsPrivacy is a major public concern when PII is legitimately accessed to link data. Our study demonstrates that a well-designed privacy enhanced interface can significantly reduce exposure of PII to people when resolving ambiguous linkages without compromising linkage quality. This research points to a new direction for PET in record linkage beyond encryption.

Establishing an International Data Linkage Repository Workgroup Toward a Benchmarking Repository

IntroductionAccess to real data with diverse attributes is critical for effective development of any data analytic algorithm. Benchmarking data repositories have all been vital to the development of research communities focused on algorithm development. This work reports on the development of such a data repository for record linkage. Objectives and ApproachEstablishing a common benchmarking repository of real data can propel a field to the next level of rigor by facilitating comparison of different algorithms, understanding what type of algorithms work best under certain real data conditions and problem domains, promoting transparency and replicability of research, and creating incentives for proper citations for contributions. In addition, benchmarking repositories can bring together the diverse stakeholders (e.g., computer scientists, statisticians, data custodians, data users including social, behaviour, economic, and health (SBEH) scientists) that can advance the field more effectively than could researchers from any single discipline. ResultsIn Fall 2016, international leaders in record linkage formed a Data Linkage Repository workgroup (DLRep) to establish a benchmarking data repository for record linkage. The workgroup is working in collaboration with The Inter-university Consortium for Political and Social Research (ICPSR) to host the site data repository planned for release in Summer 2018. The repository for record linkage research will house various types of real data that require linking with metadata, unique handles for citations, proposed algorithms for evaluation criteria, and a platform for posting, sharing, and comparing results as well as citations of relevant papers. Some datasets will have the gold standard published that researchers can evaluate their results against. Other datasets will gather results to build the gold standard as a community. Conclusion/ImplicationsRecord linkage methodology is important to domains where data needs to be integrated from multiple sources, including diverse disciplines. Establishing an international interdisciplinary research community around a benchmark data linkage repository to validate and compare linkage algorithms is crucial to fully realizing the social benefits of data about people.

Post-acute Care for Children and Youth in Texas, 2011-2014

Little is known about services provided to children and youth (C/Y) discharged from an acute care facility. Recent research has provided a foundation for efforts to supplement or complement that early work. This research investigates post-acute care (PAC) in Texas. It focuses on what differentiates those discharges that receive PAC from those that do not and on what differentiates those C/Y who receive PAC in a health care facility from those who receive home health services. The results show that only 6.4% of discharges involving C/Y receive PAC and that many factors affected the 2 issues under investigation quite differently. These results clearly demonstrate the low prevalence of PAC use for C/Y and the clear preference of using PAC home health in this population.

A Position Statement on Population Data Science: The Science of Data about People

Information is increasingly digital, creating opportunities to respond to pressing issues about human populations using linked datasets that are large, complex, and diverse. The potential social and individual benefits that can come from data-intensive science are large, but raise challenges of balancing individual privacy and the public good, building appropriate socio-technical systems to support data-intensive science, and determining whether defining a new field of inquiry might help move those collective interests and activities forward. A combination of expert engagement, literature review, and iterative conversations led to our conclusion that defining the field of Population Data Science (challenge 3) will help address the other two challenges as well. We define Population Data Science succinctly as the science of data about people and note that it is related to but distinct from the fields of data science and informatics. A broader definition names four characteristics of: data use for positive impact on citizens and society; bringing together and analyzing data from multiple sources; finding population-level insights; and developing safe, privacy-sensitive and ethical infrastructure to support research. One implication of these characteristics is that few people possess all of the requisite knowledge and skills of Population Data Science, so this is by nature a multi-disciplinary field. Other implications include the need to advance various aspects of science, such as data linkage technology, various forms of analytics, and methods of public engagement. These implications are the beginnings of a research agenda for Population Data Science, which if approached as a collective field, can catalyze significant advances in our understanding of trends in society, health, and human behavior.

Posttraumatic Stress Disorder and Mobile Health: App Investigation and Scoping Literature Review

BACKGROUND

Posttraumatic stress disorder (PTSD) is a prevalent mental health issue among veterans. Access to PTSD treatment is influenced by geographic (ie, travel distance to facilities), temporal (ie, time delay between services), financial (ie, eligibility and cost of services), and cultural (ie, social stigma) barriers.

OBJECTIVE

The emergence of mobile health (mHealth) apps has the potential to bridge many of these access gaps by providing remote resources and monitoring that can offer discrete assistance to trauma survivors with PTSD and enhance patient-clinician relationships. In this study, we investigate the current mHealth capabilities relevant to PTSD.

METHODS

This study consists of two parts: (1) a review of publicly available PTSD apps designed to determine the availability of PTSD apps, which includes more detailed information about three dominant apps and (2) a scoping literature review performed using a systematic method to determine app usage and efforts toward validation of such mHealth apps. App usage relates to how the end users (eg, clinicians and patients) are interacting with the app, whereas validation is testing performed to ensure the app's purpose and specifications are met.

RESULTS

The results suggest that though numerous apps have been developed to aid in the diagnosis and treatment of PTSD symptoms, few apps were designed to be integrated with clinical PTSD treatment, and minimal efforts have been made toward enhancing the usability and validation of PTSD apps.

CONCLUSIONS

These findings expose the need for studies relating to the human factors evaluation of such tools, with the ultimate goal of increasing access to treatment and widening the app adoption rate for patients with PTSD.

Post-acute care for children with special health care needs

BACKGROUND

Almost all studies of post-acute care (PAC) focus on older persons, frequently those suffering from chronic health problems. Some research is available on PAC for the pediatric population in general. However, very few studies focus on PAC services for children with special health care needs (SHCN).

OBJECTIVE

To investigate factors affecting the provision of PAC to children with SHCN.

METHODS

Pooled cross-sectional data from Texas Department of State Health Services hospital discharge database from 2011-2014 were analyzed. Publicly available algorithms identified chronic conditions, complex chronic conditions, and the principal problem leading to hospitalization. Analysis involved estimating two logistic regressions, with clustered robust standard errors, concerning the likelihood of receiving PAC and where that PAC was delivered. Models included patient characteristics and conditions, as well as hospital characteristics and location.

RESULTS

Only 5.8 percent of discharges for children with SHCN resulted in the provision of PAC. Two-thirds of PAC was provided in a health care facility (HCF). Severity of illness and the number of complex chronic conditions, though not the number of chronic problems, made PAC more likely. Patient demographics had no effect on PAC decisions. Hospital type and location also affected PAC decision-making.

CONCLUSIONS

PAC was provided to relatively few children with SHCN, which raises questions concerning the potential underutilization of PAC for children with SHCN. Also, the provision of most PAC in a HCF (66%) seems at odds with professional judgment and family preferences indicating that health care for children with SHCN is best provided in the home.

Does Maternity Care Coordination Influence Perinatal Health Care Utilization? Evidence from North Carolina

OBJECTIVE

To examine effects of maternity care coordination (MCC) on perinatal health care utilization among low-income women.

DATA SOURCES

North Carolina Center for Health Statistics Baby Love files that include birth certificates, maternity care coordination records, WIC records, and Medicaid claims.

STUDY DESIGN

Causal effects of MCC participation on health care outcomes were estimated in a sample of 7,124 singleton Medicaid-covered births using multiple linear regressions with inverse probability of treatment weighting (IPTW).

PRINCIPAL FINDINGS

Maternity care coordination recipients were more likely to receive first-trimester prenatal care (p < .01) and averaged three more prenatal visits and two additional primary care visits during pregnancy; they were also more likely to participate in WIC and to receive postpartum family planning services (p < .01). Medicaid expenditures were greater among mothers receiving MCC.

CONCLUSIONS

Maternity care coordination facilitates access to health care and supportive services among Medicaid-covered women. Increased maternal service utilization may increase expenditures in the short run; however, improved newborn health may reduce the need for costly neonatal care, and by implication the need for early intervention and other supports for at-risk children.

Incremental Interactive Record Linkage using Human Intelligence Tasks (HITs)

ABSTRACTObjectiveWhen analyzing population data, there is a need to link data about organizations. One challenge in linking organization level data is that unlike a person, there can be many definitions for an entity. For example, for hospitals, depending on the dataset, an entity might represent any one of the following similar but different semantic types:(1) physical units, (2) billing units (3) legal units, (5) licensed units, or (5) reporting units. How these different entities relate to each other can be complex such as one billing unit can span across many physical units or multiple billing units can exist for one physical unit. Thus, linking organization level data requires human involvement to sort through these issues in heterogeneous data sources to make informed decisions on the messy data. We design and evaluate a general framework for a hybrid Human-Machine process for ongoing integration and cleaning of hospital level data when no common identifiers exist such that we highlight the decisions that need human judgement and document and track the full processes to ensure reproducibility. Such ongoing integration is often called incremental record linkage (RL). ApproachAccurate linkage in big data requires well-defined tasks that need automatic or human processing. In the human computer interaction (HCI) field, Human Intelligence Tasks (HITs) are defined as micro tasks requiring human judgment and are often used in designing crowdsourcing systems. We designed HITs for linking organization level data and embed them into automatic deterministic linkage algorithms that supports interactive stepwise RL. The hybrid system is a framework for reproducible incremental RL. ResultsWe illustrate this framework by integrating four databases of hospitals in Texas from 2008 to 2014(N=664). The IDs used in the databases are the Texas Provider ID, the National Provider ID, the Medicare ID, and the Facility ID. We link the databases using provider name, including dba (i.e., doing business as), addresses, and phone numbers. Similarities in hospital names and addresses and the dynamic nature of hospital attributes over time make it impossible to build a fully automated linkage system for hospitals. Using our system to iteratively standardize and clean the data, we linked the hospitals with 100% precision using HITs that required confirming 79 approximate linkages and manually linking 28 hospitals. ConclusionEffective software that can support the interactive and iterative process of RL with well-designed HITs can streamline the linkage processes to support high quality replicable research using big data.

Barriers to Remote Health Interventions for Type 2 Diabetes: A Systematic Review and Proposed Classification Scheme

Background

Diabetes self-management involves adherence to healthy daily habits typically involving blood glucose monitoring, medication, exercise, and diet. To support self-management, some providers have begun testing remote interventions for monitoring and assisting patients between clinic visits. Although some studies have shown success, there are barriers to widespread adoption.

Objective

The objective of our study was to identify and classify barriers to adoption of remote health for management of type 2 diabetes.

Methods

The following 6 electronic databases were searched for articles published from 2010 to 2015: MEDLINE (Ovid), Embase (Ovid), CINAHL, Cochrane Central, Northern Light Life Sciences Conference Abstracts, and Scopus (Elsevier). The search identified studies involving remote technologies for type 2 diabetes self-management. Reviewers worked in teams of 2 to review and extract data from identified papers. Information collected included study characteristics, outcomes, dropout rates, technologies used, and barriers identified.

Results

A total of 53 publications on 41 studies met the specified criteria. Lack of data accuracy due to input bias (32%, 13/41), limitations on scalability (24%, 10/41), and technology illiteracy (24%, 10/41) were the most commonly cited barriers. Technology illiteracy was most prominent in low-income populations, whereas limitations on scalability were more prominent in mid-income populations. Barriers identified were applied to a conceptual model of successful remote health, which includes patient engagement, patient technology accessibility, quality of care, system technology cost, and provider productivity. In total, 40.5% (60/148) of identified barrier instances impeded patient engagement, which is manifest in the large dropout rates cited (up to 57%).

Conclusions

The barriers identified represent major challenges in the design of remote health interventions for diabetes. Breakthrough technologies and systems are needed to alleviate the barriers identified so far, particularly those associated with patient engagement. Monitoring devices that provide objective and reliable data streams on medication, exercise, diet, and glucose monitoring will be essential for widespread effectiveness. Additional work is needed to understand root causes of high dropout rates, and new interventions are needed to identify and assist those at the greatest risk of dropout. Finally, future studies must quantify costs and benefits to determine financial sustainability.

Predictors of Colorectal Cancer Screening: Does Rurality Play a Role?

PURPOSE

The purpose of this study was to explore the associations between sociodemographic factors such as residence, health care access, and colorectal cancer (CRC) screening among residents of Texas.

METHODS

Using the 2012 Behavioral Risk Factor Surveillance Survey, we performed logistic regression analyses to determine predictors of CRC screening among Texas residents, including rural versus urban differences. Our outcomes of interest were previous (1) CRC screening using any CRC test, (2) fecal occult blood test (FOBT), or (3) endoscopy, as well as up-to-date screening using (4) any CRC test, (5) FOBT, or (6) endoscopy. The independent variable of interest was rural versus urban residence; we controlled for other sociodemographic and health care access variables such as lack of health insurance.

RESULTS

Multivariate analysis showed that individuals who were residents of a rural/non-Metropolitan Statistical Area (MSA) location (OR = 0.70, 95% CI = 0.51-0.97) or a suburban county (OR = 0.61, 95% CI = 0.39-0.95) were less likely to report ever having any CRC screening compared to residents of a center city of an MSA. Residents of a rural/non-MSA location were less likely (OR = 0.49, 95% CI = 0.28-0.87) than residents of a center city of an MSA to be up-to-date using FOBT. There was decreased likelihood of ever being screened for CRC among the uninsured (OR = 0.43, 95% CI = 0.31-0.59).

CONCLUSIONS

Effective development and implementation of strategies to improve screening rates should aim at improving access to health care, taking into account demographic characteristics such as rural versus urban residence.