Medicare Administrative Data Research Articles

Voluntary Disenrollment from Medicare Advantage Plans Is Higher Among Racial and Ethnic Minorities

AbstractResearch ObjectiveAbout one‐tenth of Medicare Advantage (MA) enrollees voluntarily disenroll from their health plans annually. Voluntary disenrollment is related to beneficiaries’ negative experiences with their plan, disrupts continuity of care, and conflicts with goals to reduce Medicare costs. Research on racial and ethnic differences in voluntary disenrollment is important, as it may inform efforts to reduce racial and ethnic disparities in care. The few prior studies of this topic have had significant limitations, including reliance on administrative data on race and ethnicity that are often missing or inaccurate, failure to control for potential confounding variables such as dual eligibility for Medicare and Medicaid, and use of data that is more than 20 years old. Our investigation of racial and ethnic differences in rates of disenrollment from MA plans overcomes these limitations.Study DesignDisenrollment data came from Medicare’s enrollment system. Social Security Administration data on race and ethnicity were augmented with surname, address, and other Medicare administrative data to calculate probabilities of membership in different racial and ethnic groups. We summarized differences in rates of disenrollment across racial and ethnic groups (Asian or Pacific Islander [API], black, Hispanic, and white) using four types of logistic regression models: adjusted and unadjusted models estimating overall differences and adjusted and unadjusted models estimating within‐plan differences. Unadjusted overall models included only racial and ethnic group probabilities as predictors. Adjusted overall models added age, gender, dual eligibility, disability, and state of residence as control variables. For each model, we calculated the difference from whites for each racial and ethnic minority group. We also estimated differences in disenrollment rates that are attributable to within‐plan effects. Between‐plan differences were estimated by subtracting within‐plan differences from overall differences.Population Studied16 770 319 Medicare beneficiaries enrolled in 736 MA plans nationwide in 2015.Principal FindingsAdjusted rates of disenrollment were significantly (P < .001) higher for Hispanic (+1.2 percentage points), black (+1.2 percentage points), and API beneficiaries (+2.4 percentage points) than for whites. Within states, all three racial and ethnic minority groups tended to be concentrated in higher‐disenrollment plans. More than 70% of the adjusted overall difference between API and white beneficiaries reflected API beneficiaries’ tendency to disenroll from the same plans at markedly higher rates than white beneficiaries.ConclusionsWe identified higher rates of voluntary disenrollment from MA plans for racial and ethnic minorities than for whites, even after controlling for factors such as dual eligibility and disability, which relate to both disenrollment and race and ethnicity. These differences exist partly because racial and ethnic minorities are more concentrated in plans with higher overall disenrollment rates and partly because of higher within‐plan disenrollment rates, particularly among API beneficiaries.Implications for Policy or PracticeRacial and ethnic minorities may require additional—or different—assistance with plan selection than what whites need, as well as improved access to low‐disenrollment plans. Getting to the root of higher voluntary disenrollment rates for racial and ethnic minorities could improve the MA marketplace as well as the care of racial and ethnic minority beneficiaries.Primary Funding SourceCenters for Medicare and Medicaid Services.

Inpatient rehabilitation facilities’ hospital readmission rates for medicare beneficiaries treated following a stroke

ABSTRACT Background Stroke is the leading cause for admission to the nearly 1,200 Inpatient Rehabilitation Facilities (IRFs) nationally in the US. For many patients, post-acute care is an important component of their rehabilitation. Several quality measures have been publicly reported for post-acute care providers, including hospital readmissions. However, to date none have focused on specific medical conditions, limiting the usability for patients and quality improvement. Objective To assess hospital readmission rates for Medicare patients receiving inpatient rehabilitation following stroke and to identify risk factors in order to evaluate the feasibility of a stroke-specific hospital readmission measure. Methods Observational study analyzing national Medicare inpatient claims and administrative data to assess hospital readmissions. Using logistic regression, we calculated unadjusted and risk-standardized readmission rates, which adjusted for patient characteristics, including type of stroke and admission function, to capture stroke severity. Results Our national study included 116,073 fee-for-service Medicare beneficiary discharged from IRFs in 2013–2014 following stroke from 1,162 IRFs nationally. The observed hospital readmission rate among IRF patients following stroke was 11.6% and varied by patients’ admission motor function. Patients with greater functional dependence had higher readmission rates on average. Lower admission function, hemorrhagic and other stroke types (relative to ischemic) were significantly associated with higher odds of hospital readmission. Conclusion Results suggest it is feasible to assess hospital readmission rates among a stroke-cohort treated in IRFs. Stroke-focused quality measures would be useful to patients in selecting a provider and for providers in evaluating their stroke rehabilitation program outcomes. Secondary results suggest that admission function (FIM) capture stroke severity, a limitation with other claims-based stroke measures.

Beyond Black and White: Mapping Misclassification of Medicare Beneficiaries Race and Ethnicity.

The Centers for Medicare and Medicaid Services administrative data contains two variables that are used for research and evaluation of health disparities: the enrollment database (EDB) beneficiary race code and the Research Triangle Institute (RTI) race code. The objective of this article is to examine state-level variation in racial/ethnic misclassification of EDB and RTI race codes compared with self-reported data collected during home health care. The study population included 4,231,370 Medicare beneficiaries who utilized home health care services in 2015. We found substantial variation between states in Medicare administrative data misclassification of self-identified Hispanic, Asian American/Pacific Islander, and American Indian/Alaska Native beneficiaries. Caution should be used when interpreting state-level health care disparities and minority health outcomes based on existing race variables contained in Medicare data sets. Self-reported race/ethnicity data collected during routine care of Medicare beneficiaries may be used to improve the accuracy of minority health and health disparities reporting and research.

1183 Board-certified Sleep Medicine Physicians See A Greater Proportion Of Complex Sleep Patients Than Non-specialist Providers

Abstract Introduction Despite a growing literature regarding the impact of board-certification in sleep medicine, little is known about the complexity of patients seen by board-certified sleep medicine physicians (BCSMPs) relative to non-specialists. To address this gap, the purpose of the current study was to evaluate the differences in sleep complaints among Medicare beneficiaries seen by BCSMPs relative to individuals seen by non-specialists. Methods Our data source was a random 5% sample of Medicare administrative claims data from 2006-2013. Sleep disorders were operationalized using International Classification of Diseases, Ninth Revision, Clinical Modification codes. Descriptive analyses were performed to estimate the number of sleep disorder diagnoses patients received by provider status. BCSMPs were identified using a cross-matching procedure based on National Provider Identifier (NPI). Results A total of 57,209 Medicare beneficiaries received a sleep disorder diagnosis between 2006-2013. Of these, only 2.2% were seen by BCSMPs. Relative to beneficiaries seen by non-specialists, those seen by BCSMPs were more likely to be diagnosed with more than one sleep disorder (p<0.001). Specifically, 91.0% of individuals seen by non-specialists received only one sleep disorder diagnosis, whereas 75.9% of individuals seen by BCSMPs received only one sleep disorder diagnosis. Among beneficiaries seen by non-specialists, the most common sleep disorders were insomnia (48.2%; n=26,967), obstructive sleep apnea (OSA; 31.4%; n=17,554), and restless legs syndrome (8.7%; n=4,871). Among those seen by BCSMPs, the most common sleep disorders were OSA, (70.4%; n=901), sleep apnea with hypersomnia (16.5%; n=211), and insomnia (11.7%; n=150). Conclusion BCSMPs see more complex sleep patients than do non-specialists. These results suggest the possibility that more complex patients are referred for sleep specialty care. Further, these results demonstrate the value of board certification in sleep medicine in caring for complex sleep patients. Support This research was supported by an AASM Strategic Research Award from the AASM Foundation to the University of Maryland, Baltimore (PI: EMW).

0538 What Happens After Prescription of Insomnia Medication Among Older Adults?

Abstract Introduction Despite consensus recommendations regarding need for caution and careful management, sedative hypnotic insomnia therapies remain commonly prescribed among older adults. Further, sleep medications are often prescribed in the absence of a thorough sleep history or evaluation. However, little is known about delivery of sleep-related care following prescription of insomnia medications. Thus, the purpose of this study was to characterize the course of sleep-related care following a prescription fill for insomnia medication among older adults. Methods Our data source was a random 5% sample of Medicare administrative claims data from 2006-2013. Insomnia medications were identified by searching the Part D prescription drug claims and included FDA-approved insomnia-related medication classes and drugs. Sleep disorders were operationalized using International Classification of Diseases, Ninth Revision, Clinical Modification codes. Descriptive analyses were performed to estimate the number of insomnia medication users who received sleep disorder diagnoses. Results A total of 33,252 Medicare beneficiaries without prior history of sleep disorders received at least one FDA-approved insomnia medication fill between 2006-2013. Of these, 43.2% (n=14,354) eventually received a sleep disorder diagnosis. Among those receiving a sleep diagnosis after being prescribed insomnia medication, insomnia was the most common disorder (71.0%; n=10,198). Further, 15.0% (n=2,149) of individuals who were prescribed insomnia medication received an obstructive sleep apnea diagnosis, 6.6% (n=947) were diagnosed with sleep disturbances, and 5.9% (n=845) were diagnosed with restless legs syndrome. Of those who received a sleep disorder diagnosis, 95.1% (n=13,649) were diagnosed with one sleep disorder, 4.5% (n=639) were diagnosed with two sleep disorders, and 0.4% (n=66) were diagnosed with three or more sleep disorders. Conclusion Fewer than half of Medicare beneficiaries prescribed FDA-approved sedative hypnotic insomnia medications ever received a formal sleep-related diagnosis. These results suggest that clinicians prescribe sedative hypnotics without thoroughly evaluating sleep complaints. Support This research was supported by an AASM Strategic Research Award from the AASM Foundation to the University of Maryland, Baltimore (PI: EMW).

1178 Predictors Of Being Seen By A Board-certified Sleep Medicine Provider

Abstract Introduction Although several studies have evaluated the impact of board-certification in sleep medicine regarding obstructive sleep apnea treatment outcomes, no studies to date have identified predictive factors to determine which patients are evaluated by board-certified sleep medicine providers (BCSMP) in the clinical practice. Thus, the purpose of this study was to identify predictors of being seen by a BCSMP, relative to non-sleep specialist providers. Methods Our data source was a random 5% sample of Medicare administrative claims data from 2006-2013. Sleep disorder diagnoses such as insomnia, obstructive sleep apnea, restless legs syndrome, hypersomnias, and parasomnias, as well as medical comorbidities including cardiovascular, cerebrovascular, mood, pulmonary, and neurological disorders, were operationalized using International Classification of Diseases, Ninth Revision, Clinical Modification codes. Demographic data were obtained from the claims. BCSMP were identified using a novel cross-matching approach based on National Provider Identifier (NPI). Results A total of 57,209 Medicare beneficiaries received a sleep disorder diagnosis between 2006-2013, but only 1,279 (2.2%) of these individuals were ever seen by a BCSMP. Within a multivariate logistic regression model, male gender, asthma, and heart failure were significantly associated with being seen by a BCSMP. Additionally, BCSMP were more likely to evaluate patients with two or more sleep diagnoses. Conclusion Complexity of sleep disorders and cardiovascular and lung comorbidities were predictive of being seen by a BCSMP. These results demonstrate the importance of BCSMPs in caring for complex sleep medicine patients. Support This research was supported by an AASM Strategic Research Award from the AASM Foundation to the University of Maryland, Baltimore (PI: EMW).

The impact of the use of opioids among older breast cancer survivors and adverse events.

12098 Background: Older adults and cancer survivors are underrepresented in the literature underpinning recent opioid prescribing guidelines. As prevention of unnecessary persistent opioid use and inadvertent opioid-related harms gains importance in clinical practice, it is necessary to fully capture the risks of opioid related adverse events among patients with cancer pain. The objective of this study was to determine the association between opioid use after cancer diagnosis and comprehensive opioid-related adverse events among older adult breast cancer survivors. Methods: We conducted a retrospective cohort study using Surveillance, Epidemiology, and End Results tumor registry data linked with Medicare administrative claims data from 2007-2016 of women with newly diagnosed non-metastatic breast cancer. The study observation period was the year following a patient’s end of active cancer treatment. The primary exposure was a daily measure of opioid exposure based on Part D prescription claims. The primary outcomes were daily indicators of all-cause hospitalization, substance use event and a composite of other opioid-related adverse events (infections, gastrointestinal events, falls/fractures, cardiovascular events) and each component of the composite adverse event. We estimated the association of current opioid use and the immediate risk of an outcome event the following day using modified Poisson generalized estimating equation models. We adjusted for patient demographics, cancer characteristics and cancer treatments received. Results: We found that opioid exposure more than doubled the immediate risk of all-cause hospitalization (aRR = 2.77; 95%CI = 2.57, 2.99; p < 0.001) and having a composite adverse event (aRR = 2.50; 95%CI = 2.18, 2.87; p < 0.001) and dramatically increases the immediate risk of a substance use event (aRR = 14.26; 95%CI = 7.11, 28.59; p < 0.001). We find consistent results when looking at individual components of the composite adverse event measure. Conclusions: Older adult breast cancer survivors with continued prescription opioid use in the year after completing active cancer treatment experienced an immediate increased risk of all-cause hospitalization, substance use events, and myriad opioid-related adverse effects.

Generic escitalopram initiation and substitution among Medicare beneficiaries: A new user cohort study.

ObjectivesTo examine patterns of generic escitalopram initiation and substitution among Medicare beneficiaries.MethodsThis retrospective new user cohort used a 5% random sample of 2013–2015 Medicare administrative claims data. Fee-for-service Medicare beneficiaries continuously enrolled in Parts A, B, and D during a 6-month washout period prior to their initial generic or brand oral escitalopram prescriptions were included (n = 12,351). The primary outcomes were generic escitalopram treatment initiation, and among brand escitalopram initiators, generic substitution within 12 months. Patient demographics, health service utilization, and prescription level factors were measured and assessed.ResultsAmong all escitalopram initiators, about 88.2% Medicare beneficiaries initiated generic escitalopram. Beneficiaries who were younger age, male, residing in non-Northeast regions or urban area, in the Part D plan deductible benefit phase, and filling prescriptions at community/retail pharmacies were more likely to initiate generic treatment. Among brand escitalopram initiators (n = 1,464), about 20.7% switched to generic escitalopram, 31.2% switched to another alternative antidepressant, 25.1% discontinued treatment, and 8.7% were lost to follow up or passed away within 12 months after brand initiation. Factors associated with generic escitalopram substitution included region (Midwest vs. Northeast, adjusted hazard ratio (HR) = 1.46, 95% CI = 1.04–2.05), pre-index hospitalization (HR = 1.31; 95% CI = 1.16–1.48) and lower escitalopram average daily dosage (HR = 0.97; 95% CI = 0.95–0.99).ConclusionsIn 2013–2015, almost 90% Medicare beneficiaries initiated generic escitalopram treatment. Among brand escitalopram initiators, about 1 in 5 patients switched to generic escitalopram within 1 year, as compared to 1 in 4 or 1 in 3 who discontinued current or switched to alternative treatment, respectively. Medicare beneficiary’s geographic region was independently associated with generic escitalopram initiation and substitution. Findings from this study not only provide up-to-date evidence in generic escitalopram use patterns among Medicare population, but also can guide educational and practice interventions to further increase generic escitalopram use.

Events Associated With Changes in Reliance on the Veterans Health Administration Among Medicare-eligible Veterans.

We can learn something about how Veterans value the Veterans Health Administration (VHA) versus community providers by observing Veterans' choices between VHA and Medicare providers after they turn 65. For a cohort of Veterans who were newly age-eligible for Medicare, we estimated the change in VHA reliance (VHA outpatient visits divided by total VHA and Medicare visits) associated with specific events: receiving a life-threatening diagnosis, having a Medicare-paid hospitalization, or moving further from the VHA. A longitudinal cohort study of VHA and Medicare administrative data. A total of 5932 VHA users who completed a health survey in 1999 and became age-eligible for Medicare from 1998 to 2000 were followed through 2016. More Veterans chose to rely on the VHA than Medicare (64% vs. 36.%). For a VHA-reliant Veteran, a Medicare-paid hospital stay was associated with a decrease of 7.8 percentage points (pps) (P<0.001) in VHA reliance in the subsequent 12 months, but by 36 months reliance increased to near prehospitalization levels (-1.5 pps; P=0.138). Moving further from the VHA, or receiving a diagnosis of cancer, heart failure, or renal failure had no significant association with subsequent VHA reliance; however, a diagnosis of dementia was associated with a decrease in VHA reliance (-8.6 pps; P=0.026). A significant majority of newly Medicare-eligible VHA users voted with their feet in favor of sustaining the VHA as a provider of comprehensive medical care for Veterans. These VHA-reliant Veterans maintained their reliance even after receiving a life-threatening diagnosis, and after experiencing Medicare-provided hospital care.

Comorbidity Assessment Is Uneven Across Veterans Health Administration and Medicare for the Same Patient: Implications for Risk Adjustment.

Compare comorbidity identification in Medicare and Veterans Health Administration (VA) data for the purposes of risk adjustment. Analysis of Medicare and VA datasets for dually-enrolled Veterans receiving care in both settings, fiscal years 2010-2014. A retrospective analysis of administrative data for a national sample of cancer decedents. Comorbidities were evaluated using Elixhauser and Charlson coding algorithms. Clinical comorbidities were more likely to be recorded in Medicare than in VA datasets. Of 42 comorbidities, 36 (86%) were recorded at a different frequency. For example, congestive heart failure was recorded for 22.0% of patients in Medicare data and for 11.3% of patients in VA data (P<0.001). There are large differences in comorbidity assessment across VA and Medicare administrative data for the same patient, posing challenges for risk adjustment.

Risk Factors for Long-term Mortality and Patterns of End-of-Life Care Among Medicare Sepsis Survivors Discharged to Home Health Care

Despite a growing recognition of the increased mortality risk among sepsis survivors, little is known about the patterns of end-of-life care among this population. To describe patterns of end-of-life care among a national sample of sepsis survivors and identify factors associated with long-term mortality risk and hospice use. This cohort study assessed sepsis survivors who were Medicare fee-for-service beneficiaries discharged to home health care using national Medicare administrative, claims, and home health assessment data from 2013 to 2014. The initial and final primary analyses were conducted in July 2017 and from July to August 2019, respectively. Sepsis hospital discharge and 1 or more home health assessments within 1 week. Outcomes were 1-year mortality among all sepsis survivors and hospitalization in the last 30 days of life, death in an acute care hospital, and hospice use among decedents. Multivariate logistic regression was used to identify factors associated with 1-year mortality and hospice use. Among 87 581 sepsis survivors who were Medicare fee-for-service beneficiaries discharged to home health care, 49 323 (56.3%) were aged 75 years or older, 69 499 (79.4%) were non-Hispanic white, and 48 472 (55.3%) were female. Among the total survivors, 24 423 (27.9%) people died within 1 year of discharge, with a median (interquartile range) survival time of 119 (51-220) days. Among these decedents, 16 684 (68.2%) were hospitalized in the last 30 days of life, 6560 (26.8%) died in an acute care hospital, and 12 573 (51.4%) were enrolled in hospice, with 5729 (45.6%) using hospice for 7 or fewer days. Several factors were associated with 1-year mortality, including a cancer diagnosis (odds ratio [OR], 3.66; 95% CI, 3.50-3.83; P < .001), multiple dependencies of activities of daily living or instrumental activities of daily living (OR, 2.80; 95% CI, 2.57-3.05; P < .001), and an overall poor health status (OR, 2.21; 95% CI, 2.01-2.44; P < .001) documented on home health assessment. Among the decedents, cancer was associated with hospice use (OR, 2.25; 95% CI, 2.11-2.41; P < .001), patients aged 85 years or older (OR, 1.49; 95% CI, 1.37-1.61; P < .001), and living in an assisted living setting (OR, 1.93; 95% CI, 1.69-2.19; P < .001). The findings of this study suggest that death within 1 year after sepsis discharge may be common among Medicare beneficiaries discharged to home health care. Although 1 in 2 decedents used hospice, aggressive care near the end of life and late hospice referral were common. Readily identifiable risk factors suggest opportunities to target efforts to improve palliative and end-of-life care among high-risk sepsis survivors.

Severity of Traumatic Brain Injury in Older Adults and Risk of Ischemic Stroke and Depression.

Risk of ischemic stroke and depression is elevated among older adults following traumatic brain injury (TBI), yet little is known about how the severity of TBI influences risk. Thus, our objective was to assess the association between severity of the index TBI and risk of ischemic stroke and depression in a sample of older adults treated for TBI. Retrospective cohort study. R Adams Cowley Shock Trauma Center. Adults 65 years and older treated for TBI between 2006 and 2010 who survived to hospital discharge and could be linked to their Medicare administrative claims data with continuous enrollment for at least 6 months pre-TBI and 12 months post-TBI. First dates of ischemic stroke and depression available in Medicare claims were used to exclude individuals with a history. Next, we separately assessed the association between TBI severity and time to first stroke and depression using Cox proportional hazards models. Among 132 patients without preexisting history of stroke, high TBI severity was associated with increased risk of stroke compared with low TBI severity (adjusted hazard ratio 6.68, 95% confidence interval 2.49-17.94). Among 163 patients without preexisting history of depression, high TBI severity was not significantly associated with increased risk of depression compared with low TBI severity (adjusted hazard ratio 1.90, 95% confidence interval 0.94-3.84). In this group of older adults with TBI, higher TBI severity was associated with increased risk of ischemic stroke, but not depression. These results suggest that increased monitoring of older adults with moderate-severe TBI for stroke may be warranted.

Practice patterns of board-certified sleep medicine providers: a national analysis among older adult Medicare beneficiaries.

To examine the proportion of Medicare beneficiaries with sleep disorders who were evaluated by board-certified sleep medicine providers (BCSMPs). Using a random 5% sample of Medicare administrative claims data (2007-2011), BCSMPs were identified by employing a novel cross-matching approach based on National Provider Identifiers available within the Medicare database. Sleep disorders were included based partially on the International Classification of Sleep Disorders, Third Edition (insomnia, sleep-related breathing disorders, hypersomnias, circadian rhythm sleep-wake disorders, parasomnias, and restless legs syndrome), and operationalized as International Classification of Diseases, Ninth Revision, Clinical Modification diagnostic codes. The proportion of beneficiaries with each disorder who were seen by BCSMPs and nonspecialists was computed. Among older adult Medicare beneficiaries with sleep disorders, the most common sleep disorder was insomnia (n = 65,033), and the least common sleep disorder was narcolepsy (n = 784). Individuals with central sleep apnea (n = 1,561) were most likely to be treated by a BCSMP (63.9% of beneficiaries with central sleep apnea), and individuals diagnosed with insomnia were least likely to be treated by a BCSMP (16.4% of beneficiaries with insomnia). Most BCSMPs treated beneficiaries with obstructive sleep apnea (84.9% of BCSMPs) and insomnia (75.8% of BCSMPs). BCSMPs are involved in the care of a substantial proportion of Medicare beneficiaries with sleep disorders.

Comparison of alpha-spending plans for near real-time monitoring for Guillain-Barré Syndrome after influenza vaccination during the 2010/11 influenza season

BackgroundNear real-time surveillance of the influenza vaccine, which is administered to a large proportion of the US population every year, is essential to ensure safety of the vaccine. For efficient near real-time surveillance, it is key to select appropriate parameters such as monitoring start date, number of interim tests and a scheme for spending a pre-defined total alpha across the entire influenza season. Guillain-Barré Syndrome, shown to be associated with the 1976 influenza vaccine, is used to evaluate how choices of these parameters can affect whether or not a signal is detected and the time to signal. FDA has been monitoring for the risk of GBS after influenza vaccination for every influenza season since 2008. MethodsUsing Medicare administrative data and the Updating Sequential Probability Ratio Test methodology to account for claims delay, we evaluated a number of different alpha-spending plans by varying several parameters. ResultsFor relative risks of 5 or greater, almost all alpha-spending plans have 100% power; however, for relative risks of 1.5 or lower, the constant and O’Brien-Fleming plans have increasingly more power. For RRs of 1.5 and greater, the Pocock plan signals earliest but would not signal at a RR of 1.25, as observed in prior influenza seasons. There were no remarkable differences across the different plans in regards to monitoring start dates defined by the number of vaccinations; reducing the number of interim tests improves performance only marginally. ConclusionsA constant alpha-spending plan appears to be robust, in terms of power and time to detect a signal, across a range of these parameters, including alternate monitoring start dates based on either cumulative vaccinations or GBS claims observed, frequency of monitoring, hypothetical relative risks, and vaccine uptake patterns.

Health Care Reimbursement, Service Utilization, and Outcomes among Medicare Beneficiaries with Multiple Myeloma Receiving Autologous Hematopoietic Cell Transplantation in Inpatient and Outpatient Settings.

Autologous hematopoietic stem cell transplantation (auto-HCT) is a complex procedure that can be performed in both inpatient (IP) and outpatient (OP) care settings. We examined reimbursement, service utilization, and patient financial responsibility among Medicare beneficiaries with multiple myeloma who underwent auto-HCT in the IP and OP settings using a merged dataset of the Center for International Blood and Marrow Transplant Research observational database and Centers for Medicare & Medicaid Services Medicare administrative claims data. Selection criteria included first auto-HCT, time from diagnosis to auto-HCT <18 months, and continuous enrollment in Medicare Parts A and B for 30 days before HCT index claims and 100 days post-HCT or until death. Total reimbursement and patient responsibility were adjusted for patient and disease characteristics using a weighted generalized linear model. The final cohort comprised 1640 patients, 1445 (88%) who received IP-HCT and 195 (12%) who received OP-HCT. The adjusted total mean reimbursement was higher for IP-HCT compared with OP-HCT ($82,368 [95% CI, $77,643 to $87,381] versus $46,824 [95% CI, $43,567-$50,325]; P < .0001). Adjusted total mean patient responsibility was $4736 for IP-HCT (95% CI, $4731 to $5133) and $6944 for OP-HCT (95% CI, $6296 to $7658) (P < .0001). Within 100 days post-HCT, 107 of the 195 OP-HCT recipients (55%) had at least 1 subsequent admission, compared with 348 of the 1445 IP-HCT recipients (24%). Reimbursement, service utilization, and financial responsibility varied by HCT setting. As the number of Medicare beneficiaries who undergo auto-HCT increases, coverage policy needs to consider how location of services leads to variations in the financial burden for both hospital systems and patients.

Sepsis Survivors Transitioned to Home Health Care: Characteristics and Early Readmission Risk Factors

ObjectiveTo profile the characteristics of growing numbers of sepsis survivors receiving home healthcare (HHC) by type of sepsis before, during, and after a sepsis hospitalization and identify characteristics significantly associated with 7-day readmission. DesignCross-sectional descriptive study. Data sources included the Outcome and Assessment Information Set (OASIS) and Medicare administrative and claims data. Setting and ParticipantsNational sample of Medicare beneficiaries hospitalized for sepsis who were discharged to HHC between July 1, 2013 and June 30, 2014 (N = 165,228). MethodsWe used an indicator distinguishing among 3 types of sepsis: explicitly coded sepsis diagnosis without organ dysfunction; severe sepsis with organ dysfunction; and septic shock. We compared these subgroups' demographic, clinical and functional characteristics, comorbidities, risk factors for rehospitalization, characteristics of the index hospital stay, and predicted 7-day hospital readmission. ResultsThe majority (80.7%) had severe sepsis, 5.7% had septic shock, and 13.6% had sepsis without acute organ system dysfunction. The medical diagnoses recorded at HHC admission identified sepsis or blood infection only 7% of the time, potentially creating difficulty identifying the sepsis survivor in HHC. Among sepsis types, septic shock survivors had the greatest illness burden profile. This study describes 12 key variables, each of which individually raises the relative 7-day readmission risk by as much as 60%. Increased risk of 7-day rehospitalization was found among those with septic shock, 3 or more previous inpatient stays, index hospital length of stay of >8 days, dyspnea, >6 functional dependencies, and other risk factors. Conclusions and ImplicationsImplications for practice include using our findings to identify sepsis survivors who are at risk for early readmission. Assessment for these factors may profile the at-risk patient, thereby triggering the call for additional acute care intervention such as delayed discharge, or post-acute intervention such as early home visit and outpatient follow-up.

Older adult US Medicare beneficiaries with untreated obstructive sleep apnea are heavier users of health care than matched control patients.

To examine the effect of untreated obstructive sleep apnea (OSA) on health care utilization (HCU) and costs among a nationally representative sample of Medicare beneficiaries. Our data source was a random 5% sample of Medicare administrative claims data for years 2006-2013. OSA was operationalized as (1) receipt of one or more International Classification of Disease, Version 9, Clinical Modification diagnostic codes for OSA in combination with (2) initiation of OSA treatment with either continuous positive airway pressure or oral appliance (OA) therapy. First, HCU and costs were assessed during the 12 months prior to treatment initiation. Next, these HCU and costs were compared between beneficiaries with OSA and matched control patients without sleep-disordered breathing using generalized linear models. The final sample (n = 287,191) included 10,317 beneficiaries with OSA and 276,874 control patients. In fully adjusted models, during the year prior to OSA diagnosis and relative to matched control patients, beneficiaries with OSA demonstrated increased HCU and higher mean total annual costs ($19,566, 95% confidence interval [CI] $13,239, $25,894) as well as higher mean annual costs across all individual points of service. Inpatient care was associated with the highest incremental costs (ie, greater than control patients; $15,482, 95% CI $8,521, $22,443) and prescriptions were associated with the lowest incremental costs (ie, greater than control patients; $431, 95% CI $339, $522). In this randomly selected and nationally representative sample of Medicare beneficiaries and relative to matched control patients, individuals with untreated OSA demonstrated increased HCU and costs across all points of service.

RATES OF HOSPITALIZATION AND EMERGENCY DEPARTMENT VISITS AND THE QUALITY OF NURSING HOMES IN THE U.S.

CMS strives to reduce costs and improve care for nursing home (NH) residents by reducing acute care transfers. We used a national database of Medicare claims and the Minimum Data Set to build NH stays from July 2017 through June 2018 and identify dates of hospital admissions and emergency department visits without hospitalization (ED) among all residents. We calculated rates of 30-day re-hospitalization and ED among short-stay (rehabilitation) residents, and the number of hospitalizations or ED per long-stay resident day (LSRD), then examined associations with NH Five-Star ratings (data.medicare.gov) and other provider characteristics available from Medicare administrative data. We identified 1.79 million short-stays and 898,290 long-stays at 15,576 NHs. Nationally, the 30-day re-hospitalization rate is 22.6%, the short-stay ED rate is 12.0%, there was one hospitalization every 561 LSRD (1.8 per 1000 LSRD), and there was one ED every 617 LSRD (1.6 per 1000 LSRD). Median facility rates were 22.3% (IQR=17.8%, 27.1%) for 30-day re-hospitalizations, 12.0% (IQR=8.7%, 16.1%) for short-stay EDs, 1.6 hospitalizations per 1000 LSRD (IQR=1.1, 2.3), and 1.4 ED per 1000 LSRD (IQR =0.9, 2.2). Higher rates were strongly associated with lower Five-Star ratings, particularly staffing ratings, and larger, for-profit, non-hospital facilities; even after risk-adjustment. NH variation and associations with provider characteristics suggest it is possible to further reduce acute care transfers. CMS incorporated these measures into the Five-Star rating system, providing greater transparency for residents and possibly incentivizing NHs to improve through competition. Future research should monitor success or identify the need for other avenues to improve.

HOME HEALTH CARE AS A PREDICTOR OF SURVIVAL AMONG PEOPLE LIVING WITH ALZHEIMER’S DISEASE AND RELATED DEMENTIAS

Over the past 10 years, dementia care has been shifting to the community; however, there are significant regional and sociodemographic differences in the use of formal home health care services. Does the use of home health care improve survival and other outcomes among people diagnosed with dementia? The aim of the study was to determine the individual, societal, and health systems predictors of survival after a diagnosis of dementia. Using linked Medicare administrative, claims, and assessment data (N = 4,349,565); we found that home health care significantly reduced risk of death for males, but not females. This effect was strongest among older adults between the ages of 65 and 80. While men who were living alone or in a congregate/assisted living environment benefited the most (O.R. = 0.87) the effect was also strong for men living with a spouse or other caregiver (O.R. = 0.90). These findings suggest home health care may provide a survival advantage for men through the provision of rehabilitative and supportive services, as well as patient and family caregiver education. Further research is needed to understand if sex based survival differences are associated with the intensity of home health care services provided or social determinants of health.

RATES OF DISENROLLMENT FROM MEDICARE MANAGED CARE PLANS ARE HIGHER AMONG RACIAL, ETHNIC, AND LINGUISTIC MINORITIES

Voluntary disenrollment from Medicare managed care (Medicare Advantage; MA) plans is related to beneficiaries’ negative experiences with their plan, disrupts continuity of care, and conflicts with goals to reduce Medicare costs. Information on associated factors may help illuminate the dynamics that drive decisions to disenroll. We used data from 17,517,852 beneficiaries enrolled in 736 MA plans in 2015 to investigate differences in rates of disenrollment by race, ethnicity, and preferred language. Disenrollment data came from Medicare’s enrollment system. Social Security Administration data on race and ethnicity were augmented with surname, address, and other Medicare administrative data to calculate probabilities of membership in seven race/ethnicity/language-preference groups: White, Black, English-preferring Hispanic, Spanish-preferring Hispanic, Asian or Pacific Islander (API), American Indian or Alaska Native, and multiracial. We summarized disparities across groups using regression models with and without plan intercepts, controlling for gender, disability, and Medicaid eligibility. Adjusted rates of disenrollment were significantly higher for Spanish-preferring Hispanics (19.1%), Blacks (10.2%), and APIs (9.4%) than for Whites (7.7%), and significantly lower for English-preferring Hispanics (7.4%, p’s<0.001). Within-plan disparities accounted for only a small fraction of overall disparities, indicating that Spanish-preferring Hispanics, Blacks, and APIs tended to be enrolled in plans with higher disenrollment than plans in which Whites were enrolled, whereas English-preferring Hispanics tended to be enrolled in plans with lower disenrollment. These between-plan differences may indicate that high-minority-enrollment plans less effectively inform beneficiaries about the cost and coverage of care or that racial/ethnic/linguistic minorities more often select plans that raise rates or restrict coverage.