McAllen And El Paso Revisited: Medicare Variations Not Always Reflected In The Under-Sixty-Five Population

Toward Better Structure and Constraint to Mine Negative Sequential Patterns

Nonoccurring behavior (NOB) studies have attracted the growing attention of scholars as a crucial part of behavioral science. As an effective method to discover both NOB and occurring behaviors (OB), negative sequential pattern (NSP) mining is successfully used in analyzing medical treatment and abnormal behavior patterns. At this time, NSP mining is still an active and challenging research domain. Most of the algorithms are inefficient in practice. Briefly, the key weaknesses of NSP mining are: 1) an inefficient positive sequential pattern (PSP) mining process, 2) a strict constraint of negative containment, and 3) the lack of an effective Negative Sequential Candidate (NSC) generation method. To address these weaknesses, we propose a highly efficient algorithm with improved techniques, named sc-NSP, to mine NSP efficiently. We first propose an improved PrefixSpan algorithm in the PSP mining process, which connects to a bitmap storage structure instead of the original structure. Second, sc-NSP loosens the frequency constraint and exploits the NSC generation method of positive and negative sequential patterns mining (PNSP) (a classic NSP mining method). Furthermore, a novel pruning strategy is designed to reduce the computational complexity of sc-NSP. Finally, sc-NSP obtains the support of NSC by using the most efficient bitwise-based calculation operation. Theoretical analyses show that sc-NSP performs particularly well on data sets with a large number of elements and items in sequence. Comparison and extensive experiments along with case studies on health data show that sc-NSP is 10 times more efficient than other state-of-the-art methods, and the number of NSPs obtained is 5 times greater than other methods.

Geographical Variation in Health Spending Across the US Among Privately Insured Individuals and Enrollees in Medicaid and Medicare

IMPORTANCE

Little is known about small-area variations in health care spending and utilization across the 3 major funders of health care in the US: Medicare, Medicaid, and private insurers.

OBJECTIVE

To measure regional health spending and utilization across Medicare, Medicaid, and the privately insured; to observe whether there are regions that are simultaneously low spending for all 3 payers; and to determine what factors are correlated with regional spending and utilization by payer.

DESIGN, SETTING, AND PARTICIPANTS

Observational cross-sectional analysis of the US health system in 2016 and 2017 for 241 of 306 hospital referral regions (HRRs) and 2 states. Participants include individuals with employer-sponsored coverage from Aetna, Humana, or UnitedHealth; individuals with Medicaid fee-for-service coverage in 2016 and 2017; and individuals with Medicare coverage. The analysis was carried out from January 2020 to May 2022.

MAIN OUTCOMES AND MEASURES

Spending per beneficiary and inpatient days per beneficiary by payer and overall.

RESULTS

The data include 25 381 167 individuals with employer-sponsored coverage, 69 891 299 with Medicaid coverage in 2016 and 2017, and 26 711 426 individuals with Medicare fee-for-service coverage. The percentage of enrollees who identified as female was 54.1% in the Medicaid program, 56.2% in the Medicare program, and 50.4% in private insurance. The mean (SD) age was 26.9 (21.8) years for Medicaid and 75.0 (7.9) years for Medicare enrollees; for private insurance enrollees, just age brackets were reported: 18 to 24 years (15.9%), 25 to 34 years (24.2%), 35 to 44 years (21.3%), 45 to 54 years (20.8%), and 55 to 64 years (17.8%). In 2017, the mean (SD) HRR-level spending per beneficiary was $4441 ($710) for private insurance, $10 281 ($1294) for Medicare, and $6127 ($1428) for Medicaid. Across HRRs, the correlation coefficients and 95% CIs were 0.020 (-0.106 to 0.146; P = .76) for private insurance and Medicare spending, 0.213 (0.090 to 0.330; P < .001) for private insurance and Medicaid, and 0.162 (0.037 to 0.282; P < .01) for Medicare and Medicaid. Just 3 HRRs (Boulder, Colorado; Bloomington, Illinois; and Olympia, Washington) were in the lowest spending quintile for all 3 insurance programs; 4 HRRs were in the highest (The Bronx, New York; Manhattan, New York; White Plains, New York; and Dallas, Texas). By contrast, the correlation coefficients and 95% CIs for utilization, measured in hospital days, were 0.465 (0.361 to 0.559; P < .001) for private insurance and Medicare, 0.527 (0.429 to 0.612; P < .001) for private insurance and Medicaid, and 0.278 (0.157 to 0.390; P < .001) for Medicare and Medicaid.

CONCLUSIONS AND RELEVANCE

These findings suggest that payer-specific factors are correlated with health spending variation among Medicare beneficiaries, Medicaid beneficiaries, and the commercially insured and that payer-specific policies will be necessary to improve efficiency in the US health sector.

Characteristics Associated with Feeding Tube Placement: Retrospective Cohort Study of Texas Nursing Home Residents with Advanced Dementia

OBJECTIVES

To investigate resident-level, provider-type, nursing home (NH), and regional factors associated with feeding tube (FT) placement in advanced dementia.

DESIGN

Retrospective cohort study.

SETTING AND PARTICIPANTS

NH residents in Texas with dementia diagnosis and severe cognitive impairment (N = 20,582).

METHODS

This study used 2011-2016 Texas Medicare data to identify NH residents with a stay of at least 120 days who had a diagnosis of dementia on Long Term Care Minimum Data Set (MDS) evaluation and severe cognitive impairment on clinical score. Multivariable repeated measures analyses were conducted to identify associations between FT placement and resident-level, provider-type, NH, and regional factors.

RESULTS

The prevalence of FT placement in advanced dementia in Texas between 2011 and 2016 ranged from 12.5% to 16.1% with a nonlinear trend. At the resident level, the prevalence of FT decreased with age [age > 85 years, prevalence ratio (PR) 0.60, 95% confidence interval (CI) 0.52-0.69] and increased among residents who are black (2.74, 95% CI 2.48-3.03) or Hispanic (PR 1.91, 95% CI 1.71-2.13). Residents cared for by a nurse practitioner or physician assistant were less likely to have an FT (PR 0.90, 95% CI 0.85-0.96). No facility characteristics were associated with prevalence of FT placement in advanced dementia. There were regional differences in FT placement with the highest use areas on the Texas-Mexico border and in South and East Texas (Harlingen border area, PR 4.26, 95% CI 3.69-4.86; San Antonio border area, PR 3.93, 95% CI 3.04-4.93; Houston, PR 2.17, 95% CI 1.87-2.50), and in metro areas (PR 1.36, 95% CI 1.22-1.50).

CONCLUSIONS AND IMPLICATIONS

Regional, race, and ethnic variations in prevalence of FT use among NH residents suggest opportunities for clinicians and policy makers to improve the quality of end-of-life care by especially considering other palliative care measures for minorities living in border towns.

Spending On Postacute Care After Hospitalization In Commercial Insurance And Medicare Around Age Sixty-Five

Postacute care costs are the primary determinant of episode spending around hospitalization. Yet there is little evidence that greater spending on postacute care improves readmission rates or functional recovery. Recent Medicare payment reform evaluations have suggested that postacute care spending is responsive to episode-based incentives. However, it remains unknown whether Medicare payment policies are responsible for excess postacute care spending, compared with that of commercial payers. In a population-based, statewide collaborative of Michigan hospitals, we used regression discontinuity design among propensity-weighted, age-adjusted cohorts to compare postacute care spending between patients with commercial insurance and those with Medicare around age sixty-five. Spending was 68-230 percent greater among fee-for-service Medicare beneficiaries than among similar commercially insured people across varied medical and surgical conditions. Despite greater spending, there were no differences in readmission rates. These findings suggest that postacute care utilization is highly sensitive to payer influence, and there may be an opportunity for additional savings in Medicare without sacrificing quality.

Behavioral Risk Factors and Regional Variation in Cardiovascular Health Care and Death

INTRODUCTION

Reducing the burden of death from cardiovascular disease includes risk factor reduction and medical interventions.

METHODS

This was an observational analysis at the hospital service area (HSA) level, to examine regional variation and relationships between behavioral risks, health services utilization, and cardiovascular disease mortality (the outcome of interest). HSA-level prevalence of cardiovascular disease behavioral risks (smoking, poor diet, physical inactivity) were calculated from the Behavioral Risk Factor Surveillance System; HSA-level rates of stress tests, diagnostic cardiac catheterization, and revascularization from a statewide multi-payer claims data set from Maine in 2013 (with 606,260 patients aged ≥35 years), and deaths from state death certificate data. Analyses were done in 2016.

RESULTS

There were marked differences across 32 Maine HSAs in behavioral risks: smoking (12.4%-28.6%); poor diet (43.6%-73.0%); and physical inactivity (16.4%-37.9%). After adjustment for behavioral risks, rates of utilization varied by HSA: stress tests (28.2-62.4 per 1,000 person-years, coefficient of variation=17.5); diagnostic cardiac catheterization (10.0-19.8 per 1,000 person-years, coefficient of variation=17.3); and revascularization (4.6-6.2 per 1,000 person-years; coefficient of variation=9.1). Strong HSA-level associations between behavioral risk factors and cardiovascular disease mortality were observed: smoking (R²=0.52); poor diet (R²=0.38); and physical inactivity (R²=0.35), and no association between revascularization and cardiovascular disease mortality after adjustment for behavioral risk factors (R²=0.02). HSA-level behavioral risk factors were also strongly associated with all-cause mortality: smoking (R²=0.57); poor diet (R²=0.49); and physical inactivity (R²=0.46).

CONCLUSIONS

There is substantial regional variation in behavioral risks and cardiac utilization. Behavioral risk factors are associated with cardiovascular disease mortality regionally, whereas revascularization is not. Efforts to reduce cardiovascular disease mortality in populations should focus on prevention efforts targeting modifiable risk factors.

Diabetes Screening and Prevention in a High-Risk, Medically Isolated Border Community

Introduction

A project in a Texas border community setting, Prevention Organized against Diabetes and Dialysis with Education and Resources (POD²ER), offered diabetes prevention information, screening, and medical referrals. The setting was a large, longstanding flea market that functions as a shopping mall for low-income people. The priority population included medically underserved urban and rural Mexican Americans. Components of the program addressed those with diabetes, prediabetes, and accompanying relatives and friends.

Background

People living in the Lower Rio Grande Valley (LRGV) face challenges of high rates of type 2 diabetes, lack of knowledge about prevention, and inadequate access to medical care. Recent statistics from actual community-wide screenings indicate a high diabetes prevalence, 30.7% among adults in the LRGV compared with 12.3% nationwide.

Methods

A diverse team composed of public health faculty, students, a physician, a community health worker, and community volunteers conceived and developed the project with a focus on cultural and economic congruence and a user-friendly atmosphere. The program provided screening for prediabetes and diabetes with a hemoglobin A1c test. Screening was offered to those who were at least 25 years of age and not pregnant. When results indicated diabetes, a test for kidney damage was offered (urinary albumin-to-creatinine ratio). A medical appointment at a community clinic within a week was provided to those who tested positive for diabetes and lacked a medical home. Health education modules addressed all family members.

Discussion

The project was successful in recruiting 2,332 high-risk people in 26 months in a community setting, providing clinic referrals to those without a doctor, introducing them to treatment, and providing diabetes prevention information to all project participants. Implications for research and practice are highlighted.

Conclusion

This study shows that a regular access point in a place frequented by large numbers of medically marginalized people in a program designed to eliminate cultural and economic barriers can succeed in providing a hard-to-reach community with diabetes prevention services. Aspects of this program can serve as a model for other service provision for similar populations and settings.

Geographic Variation in Quality of Care for Commercially Insured Patients

BACKGROUND

Extensive evidence documents geographic variation in spending, but limited research assesses geographic variation in quality, particularly among commercially insured enrollees.

OBJECTIVE

To measure geographic variation in quality measures, correlation among measures, and correlation between measures and spending for commercially insured enrollees.

DATA SOURCE

Administrative claims from the 2007-2009 Truven MarketScan database.

METHODS

We calculated variation in, and correlations among, 10 quality measures across 306 Hospital Referral Regions (HRRs), adjusting for beneficiary traits and sample size differences. Further, we created a quality index and correlated it with spending.

RESULTS

The coefficient of variation of HRR-level performance ranged from 0.04 to 0.38. Correlations among quality measures generally ranged from 0.2 to 0.5. Quality was modestly positively related to spending.

CONCLUSION

Quality varied across HRRs and there was only a modest geographic "quality footprint."

In-office magnetic resonance imaging (MRI) equipment ownership and MRI volume among medicare patients in orthopedic practices

BACKGROUND

Concerns have been raised about physician ownership of onsite advanced imaging equipment as allowed under Stark laws by the in-office ancillary service exception (IOASE).

METHODS

A web-based survey of orthopedic practices in the United States was used to assign a first date of onsite MRI capacity acquisition (if any) to specific orthopedic practices. Medicare claims data for 2006-2010 was obtained for providers in orthopedic practices acquiring onsite MRI capacity and in matched orthopedic practices without an onsite MRI over the same period of time. Multivariate regression was used to estimate the change in provider Medicare MRI volume one year before and one year after the onsite MRI acquisition year for providers in MRI practices compared to providers in propensity-score matched non-MRI practices.

RESULTS

In all of the MRI volume change models estimated, the association between onsite MRI acquisition and the change in provider Medicare MRI volume (one-year post-onsite-MRI-acquisition less one year pre-acquisition) was consistently small and not statistically significant. This lack of association was robust to changes in model specification in terms of types of MRI exams considered, specific covariates included in the multivariate model, or the process used to confirm individual provider affiliation with study practices in study years.

CONCLUSIONS

Our analysis of Medicare claims data provides no empirical support for the proposition that acquisition of onsite MRI capacity within an orthopedic surgery practice induces an increase in the rate of MRI use for Medicare patients among practice providers, relative to physicians in practices without MRI capacity over the same time period.

No association between Centers for Medicare and Medicaid services payments and volume of Medicare beneficiaries or per-capita health care costs for each state

The Centers for Medicare and Medicaid Services recently published data on Medicare payments to physicians for 2012. We investigated regional variations in payments to gastroenterologists and evaluated whether payments correlated with the number of Medicare patients in each state. We found that the mean payment per gastroenterologist in each state ranged from $35,293 in Minnesota to $175,028 in Mississippi. Adjusted per-physician payments ranged from $11 per patient in Hawaii to $62 per patient in Washington, DC. There was no correlation between the mean per-physician payment and the mean number of Medicare patients per physician (r = 0.09), there also was no correlation between the mean per-physician payment and the overall mean per-capita health care costs for each state (r = -0.22). There was a 5.6-fold difference between the states with the lowest and highest adjusted Medicare payments to gastroenterologists. Therefore, the Centers for Medicare and Medicaid Services payments do not appear to be associated with the volume of Medicare beneficiaries or overall per-capita health care costs for each state.

The use of sequential pattern mining to predict next prescribed medications

BACKGROUND

Therapy for certain medical conditions occurs in a stepwise fashion, where one medication is recommended as initial therapy and other medications follow. Sequential pattern mining is a data mining technique used to identify patterns of ordered events.

OBJECTIVE

To determine whether sequential pattern mining is effective for identifying temporal relationships between medications and accurately predicting the next medication likely to be prescribed for a patient.

DESIGN

We obtained claims data from Blue Cross Blue Shield of Texas for patients prescribed at least one diabetes medication between 2008 and 2011, and divided these into a training set (90% of patients) and test set (10% of patients). We applied the CSPADE algorithm to mine sequential patterns of diabetes medication prescriptions both at the drug class and generic drug level and ranked them by the support statistic. We then evaluated the accuracy of predictions made for which diabetes medication a patient was likely to be prescribed next.

RESULTS

We identified 161,497 patients who had been prescribed at least one diabetes medication. We were able to mine stepwise patterns of pharmacological therapy that were consistent with guidelines. Within three attempts, we were able to predict the medication prescribed for 90.0% of patients when making predictions by drug class, and for 64.1% when making predictions at the generic drug level. These results were stable under 10-fold cross validation, ranging from 89.1%-90.5% at the drug class level and 63.5-64.9% at the generic drug level. Using 1 or 2 items in the patient's medication history led to more accurate predictions than not using any history, but using the entire history was sometimes worse.

CONCLUSION

Sequential pattern mining is an effective technique to identify temporal relationships between medications and can be used to predict next steps in a patient's medication regimen. Accurate predictions can be made without using the patient's entire medication history.

Medicare and private spending trends from 2008 to 2012 diverge in Texas

The recent relatively slow growth in health care spending masks significant differences among payers, clinical settings, and geographic areas. To better understand the spending slowdown, we focus on 2008-2012 trends in Texas among Medicare fee-for-service beneficiaries and enrollees in Blue Cross Blue Shield of Texas (BCBSTX). Spending per person for Medicare grew only 1.5% per year on average, compared with 5.2% for BCBSTX. In Medicare, utilization rates were relatively flat, while prices grew more slowly than input prices. In BCBSTX, spending growth was driven by increases in negotiated prices, in particular hospital prices. We find that geographic variation declined sharply in Medicare, due to drops in spending on post-acute care in two notoriously high-spending regions but rose slightly in BCBSTX. The aggregate spending trends mask two divergent stories: spending growth in Medicare is very slow, but price increases continue to drive unsustainable spending growth among the privately insured.

The Relationship between Commercial Health Care Prices and Medicare Spending and Utilization

OBJECTIVE

To explore the relationship between commercial health care prices and Medicare spending/utilization across U.S. regions.

DATA SOURCES

Claims from large employers and Medicare Parts A/B/D over 2007-2009.

STUDY DESIGN

We compared prices paid by commercial health plans to Medicare spending and utilization, adjusted for beneficiary health and the cost of care, across 301 hospital referral regions.

PRINCIPAL FINDINGS

A 10 percent lower commercial price (around the average level) is associated with 3.0 percent higher Medicare spending per member per year, and 4.3 percent more specialist visits (p < .01).

CONCLUSIONS

Commercial health care prices are negatively associated with Medicare spending across regions. Providers may respond to low commercial prices by shifting service volume into Medicare. Further investigation is needed to establish causality.

Changes in Geographic Variation in the Use of Percutaneous Coronary Intervention for Stable Ischemic Heart Disease After Publication of the Clinical Outcomes Utilizing Revascularization and Aggressive Drug Evaluation (COURAGE) Trial

Background—

Clinical uncertainty is cited as a cause of geographic variation. However, little is known about the effect of comparative effectiveness research on variation. We examined whether geographic variation in the use of percutaneous coronary intervention (PCI) for stable ischemic heart disease (SIHD) declined after publication of the Clinical Outcomes Utilizing Revascularization and Aggressive Drug Evaluation (COURAGE) trial.

Methods and Results—

We examined changes in utilization and geographic variation in 67 hospital referral regions using the State Inpatient Databases. We compared age- and sex-adjusted rates of PCI for SIHD before (2006) and after (2008) publication of the COURAGE trial and compared those with contemporaneous changes in PCI volume for acute coronary syndrome. A total of 272 659 PCIs for SIHD from 526 hospitals were included in the analysis. After the publication of the COURAGE trial, PCI volume for SIHD declined by 25% ( P <0.001) and decreased by 12% for acute coronary syndrome ( P <0.001). This was predominantly attributable to changes in hospital referral regions with the highest levels of utilization pre-COURAGE trial (35% decline in the highest tertile versus 18% in the lowest). As measured by the systematic component of variation, there was substantial geographic variation in the use of PCI for SIHD preceding the publication of the COURAGE trial. Variation declined by 28% (0.53 versus 0.40) after publication, but geographic variation remained higher for SIHD than acute coronary syndrome (0.40 versus 0.17).

Conclusions—

There was a substantial decline in the use of and geographic variation in PCI for SIHD after the publication of the COURAGE trial. However, geographic variation in the use of PCI for SIHD remained high.

Saving Money or Just Saving Lives? Improving the Productivity of US Health Care Spending

There is growing concern over the rising share of the US economy devoted to health care spending. Fueled in part by demographic transitions, unchecked increases in entitlement spending will necessitate some combination of substantial tax increases, elimination of other public spending, or unsustainable public debt. This massive increase in health spending might be warranted if each dollar devoted to the health care sector yielded real health benefits, but this does not seem to be the case. Although we have seen remarkable gains in life expectancy and functioning over the past several decades, there is substantial variation in the health benefits associated with different types of spending. Some treatments, such as aspirin, beta blockers, and flu shots, produce a large health benefit per dollar spent. Other more expensive treatments, such as stents for cardiovascular disease, are high value for some patients but poor value for others. Finally, a large and expanding set of treatments, such as proton-beam therapy or robotic surgery, contributes to rapid increases in spending despite questionable health benefits. Moving resources toward more productive uses requires encouraging providers to deliver and patients to consume high-value care, a daunting task in the current political landscape. But widespread inefficiency also offers hope: Given the current distribution of resources in the US health care system, there is tremendous potential to improve the productivity of health care spending and the fiscal health of the United States.

Hospital ownership type and treatment choices

In the face of increasing health care costs, taxing not-for-profit hospitals may be seen as the right choice to increase government revenues if not-for-profit hospitals are not different from their for-profit counterparts. This study investigates how hospital ownership type affects treatment choices to show whether ownership type and teaching status are correlated with choosing a procedure as the treatment and how these choices relate to patient insurance type. Not-for-profit hospitals significantly differ from for-profits in terms of treatment choices of less profitable patients and all hospitals are more likely to accord the procedure when the patient is privately insured than uninsured though teaching government hospitals are the most likely to accord the procedures for all insurance types. Considering treatment choices, not-for-profit hospitals have different objectives than for-profit and government hospitals and in terms of profit-seeking behavior, not-for-profit hospitals seem to lie between for-profit and government hospitals.

Moving (realistically) from volume-based to value-based health care payment in the USA: starting with medicare payment policy

Employers and policy-makers in the USA are desperate to slow the rate at which health expenditures grow. Changing how most health care providers are reimbursed will be necessary to achieve this. Although both politically and practically daunting, massive restructuring or replacement of fee-for-service (FFS) reimbursement is what is most required. As the dominant reimbursement model in the USA, FFS payment to individual providers strongly encourages and financially rewards the quantity of care provided, regardless of its quality or necessity. Providing high quality, lower cost care with fewer complications and hospital re-admissions can even financial penalize providers. Unfortunately, physicians and other health providers respond rationally to existing financial incentives (translation: they do what they get paid to do and generally try to, or have to, minimize those activities and services for which they are not paid). Altering this reality and fostering the expansion of exemplary delivery models-such as the Mayo Clinic or Geisinger Health System-requires change in how providers behave. And changing behavior often starts with adjusting how providers are paid. Medicare is the programme and payer most capable of using payment reform to catalyze delivery system reform.