Geographic Variation In Costs Research Articles

Payer-Negotiated Price Variation and Relationship to Surgical Outcomes for the Most Common Cancers at NCI-Designated Cancer Centers.

Federal rules mandate that hospitals publish payer-specific negotiated prices for all services. Little is known about variation in payer-negotiated prices for surgical oncology services or their relationship to clinical outcomes. We assessed variation in payer-negotiated prices associated with surgical care for common cancers at National Cancer Institute (NCI)-designated cancer centers and determined the effect of increasing payer-negotiated prices on the odds of morbidity and mortality. A cross-sectional analysis of 63 NCI-designated cancer center websites was employed to assess variation in payer-negotiated prices. A retrospective cohort study of 15,013 Medicare beneficiaries undergoing surgery for colon, pancreas, or lung cancers at an NCI-designated cancer center between 2014 and 2018 was conducted to determine the relationship between payer-negotiated prices and clinical outcomes. The primary outcome was the effect of median payer-negotiated price on odds of a composite outcome of 30 days mortality and serious postoperative complications for each cancer cohort. Within-center prices differed by up to 48.8-fold, and between-center prices differed by up to 675-fold after accounting for geographic variation in costs of providing care. Among the 15,013 patients discharged from 20 different NCI-designated cancer centers, the effect of normalized median payer-negotiated price on the composite outcome was clinically negligible, but statistically significantly positive for colon [aOR 1.0094 (95% CI 1.0051-1.0138)], lung [aOR 1.0145 (1.0083-1.0206)], and pancreas [aOR 1.0080 (1.0040-1.0120)] cancer cohorts. Payer-negotiated prices are statistically significantly but not clinically meaningfully related to morbidity and mortality for the surgical treatment of common cancers. Higher payer-negotiated prices are likely due to factors other than clinical quality.

Who and Where? The Spatial Context of Racial and Ethnic Disparities in Economic Security among Older Adults

Many older adults experience challenging financial circumstances and do not have sufficient income to afford a basic budget in their home communities. Far higher proportions of racial and ethnic minority older adults live on incomes that fall short of what is needed to make ends meet relative to their White counterparts. We describe racial/ethnic disparities in late-life economic insecurity, which occurs when an older person lacks sufficient financial resources to cover necessary expenses in their home community. Although nationwide half of older singles are economically insecure, Massachusetts (62%), New York (65%), Vermont (57%), and Mississippi (57%) have the highest shares of older adults who experience economic insecurity. Compared to Whites, minority older adults have higher rates of economic insecurity in nearly every state, but racial/ethnic disparities are higher in some locations (Rhode Island, Massachusetts, Mississippi, Louisiana) and lower in others (Oregon, Arizona, Nevada, West Virginia). Disparities in economic insecurity reflect the precarious financial situations experienced by many older adults, rooted not only in risks and disadvantages accumulated over time, but also in the variable and uncertain social and economic contexts that accompany the aging experience. By situating older adults in their places of residence, we observe that the cost of remaining in community intersect with life-course experiences associated with social identities to produce disparities in economic security at older ages. The geographic variation in cost of living calls for context-specific assessment of economic security to evaluate the adequacy of economic resources and the associated risk of hardship.

Frailty and Dementia: Differences in Health Care Utilization and Costs

Frailty and dementia are associated with poor health outcomes and increased health care utilization. A more nuanced understanding of this dynamic may be useful in improving care and developing policies. This retrospective cohort study was conducted using 5% random sample of Medicare fee-for-service beneficiaries (n=1,132,367; mean age 76.2 years; 57.9% female) in 2014-2016. We compared average 1-year home time (number of days alive outside of the hospital and SNF), mean total cost per beneficiary, and number of incident ICU stays per 100 person-years (PY) across four groups: frailty and dementia, dementia alone, frailty alone or neither. Frailty and dementia were identified using validated claims-based algorithms. We also determined differences in costs per group across different regions within the United States. Beneficiaries with both frailty and dementia had a high 1-year mortality rate of 21.9% (vs. dementia alone [9.7%], frailty alone [9.4%] or neither [2.1%]), while having less home time (306 days; difference of 36 days, 31 days, and 53 days, respectively), and more incident ICU stays per 100 PY (29.9 vs 9.5, 25.8, and 5.6, respectively). Mean total costs for beneficiaries with both was $26,030 compared to other groups ($12,096, $24,693, and $9,029, respectively). Across the United States, range of costs varied the most for beneficiaries with both frailty and dementia ($13,244-31,987 vs $4,621-15,364, $20,090-30,965, and $7,672-10,450, respectively). Increase in health care utilization and wide geographic variation in costs associated with patients with frailty and dementia suggests room for improvement in health care delivery to improve outcomes of this group.

Frailty and Dementia: Differences in Health Care Utilization and Costs

Frailty is a progressive physical decline leading to higher morbidity and mortality in older adults. Previous studies have demonstrated shared mechanisms between insulin resistance, inflammation, and frailty. The purpose of this trial is to determine whether metformin prevents frailty in non-frail, community-dwelling older adults (≥65 years) with pre-diabetes, determined by 2-hour oral glucose tolerance test (OGTT). Frail individuals (Fried criteria) and those with renal impairment (glomerular filtration rate <45 mL/min) are excluded. Eligible participants are randomized to metformin or placebo and followed for two years. The primary outcome is frailty; secondary outcomes include physical function (short physical performance battery), systemic and skeletal muscle inflammation (plasma and muscle inflammatory markers), muscle insulin signaling (muscle biopsy), insulin sensitivity (insulin clamp), glucose tolerance (OGTT), and body composition (dual-energy x-ray absorptiometry) measurements. Participants are followed every 3 months for safety assessments, every 6 months for frailty assessment and OGTT, and every 12 months for muscle biopsy. Currently, 99 participants, including 53 (53.5%) male and 91 (91.9%) white, are active (54) or have completed the study (35). At baseline, mean age was 72.3 ± 5.5 years, body mass index was 30.7 ± 5.9 kg/m2, and Hemoglobin A1c was 5.73 ± 0.37%. Mean frailty score was 0.5 ± 0.6 and the proportion of non-frail and pre-frail participants were 58.6% (n = 58) and 41.5% (n = 41), respectively. Findings of this clinical trial may have future implications for the use of metformin in older adults with pre-diabetes in order to prevent the onset of frailty.

Geographic Variation in Costs of Transsphenoidal Pituitary Surgery in the United States

Geographic variations in health care costs have been reported for many surgical specialties. In this study, we sought to describe national and regional costs associated with transsphenoidal pituitary surgery (TPS). Data from the Truven-MarketScan 2010-2014 were analyzed. We examined overall total, hospital/facility, physician, and out-of-pocket payments in patients undergoing TPS including technique-specific costs. Mean payments were obtained after risk adjustment for patient-level and system-level confounders and estimated differences across regions. The estimated overall annual burden was $43 million/year in our cohort. The average overall total payment associated with TPS was $35,602.30, hospital/facility payment was $26,980.45, physician payment was $4685.95, and out-of-pocket payment was $2330.78. Overall total and hospital/facility costs were highest in the West and lowest in the South (both P < 0.001), whereas physician reimbursements were highest in the North-east and lowest in the South (P < 0.001). There were no differences in out-of-pocket expenses across regions. On a national level, there were significantly higher overall total and hospital/facility payments associated with endoscopic compared with microscopic procedures (both P<0.001); there were no significant differences in physician payments or out-of-pocket expenses between techniques. There were also significant within-region cost differences in overall total, hospital/facility, and physician payments in both techniques as well as in out-of-pocket expenses associated with microsurgery. There were no significant regional differences in out-of-pocket expenses associated with endoscopic surgery. Our results show significant geographic cost disparities associated with TPS. Understanding factors behind disparate costs is important for developing cost containment strategies.

Regional Differences in Cost and Length of Stay in Neonates with Hypoplastic Left Heart Syndrome.

Hypoplastic left heart syndrome (HLHS) is a highly resource-intensive diagnosis. Geographic variation in cost and length of stay (LOS) in HLHS is not well described. Neonates diagnosed with HLHS between 2000 and 2012 were identified using the Kids' Inpatient Database. Hospitalizations were stratified into two groups: (1) birth and (2) secondary. United States regional differences in hospital charges and LOS were compared using adjusted linear regression. Of 2431 birth hospitalizations, 449 neonates (18.5%) died while inpatient and mortality rates differed by region (p = 0.02). After birth, 40.5% (n = 985) of neonates were transferred; transfers were most common in the Midwest (p < 0.0001). Adjusted average LOS was shortest in the West and longest in the South (26.1 days; 95% CI 24.0, 35.1 vs. 34.9 days; 95% CI 31.8, 38.1). Average adjusted charges were lowest in the Northeast ($324,600; 95% CI $271,400, $377,900) and highest in the West ($400,500; 95% CI $346,700, $454,300, p = 0.05). Among 1895 secondary hospitalizations, 24.9% of neonates died as inpatients, and the average adjusted LOS was shortest in the West (26.8 days; 95% CI 23.9, 29.7) and longest in the South (38.5 days; 95% CI 34.4, 42.4). Average adjusted charges were lowest in the Northeast ($326,900; 95% CI $270,700, $383,100) and highest in the South ($505,900; 95% CI $450,200, $561,500, p < 0.0001). Significant geographic variations in mortality, LOS, and hospital charges exist in care of US HLHS neonates. Reducing variation in care should remain a priority in national quality efforts in congenital heart disease.

PHS29 - Geographic Variation in Opioid Related Outpatient Healthcare Utilization in the United States: A Retrospective Study of a Large Health System’s Claims Data

The abuse and dependence on opioids has been on the rise nationally. However, the impact of opioid use on healthcare utilization varies widely across the United States. The objective was to estimate the geographic variation in costs and payments associated with outpatient opioid-related visits in a large hospital claims database. The Vizient Health System Database was used to identify opioid-related outpatient patient visits between January 2014 and June 2017. This database includes detailed billing data across all payers for more than 400 hospitals across the US. Eligible visits had a primary diagnosis of opioid use or dependence defined by International Classification of Diseases, 9th and 10th Revision (ICD-9/10). Geography was defined the 9 US Census sub-regions. Multivariable regression models were performed to estimate the mean expected cost and payment per visit for each region controlling for patient demographics, comorbid conditions, hospital characteristics, and year. A total of 160,901 hospital outpatient visits were identified in 447 hospitals. The overall adjusted mean cost and payment for an opioid visit was $533 and $374, respectively. The adjusted mean expected cost was $488, $408, $395, $569, $429, $610, $698, $530, $802 in East North Central, East South Central, Mid-Atlantic, Mountain, New England, Pacific, South Atlantic, West North Central, and West South Central regions, respectively. The adjusted mean expected payment was $281, $527, $187, $319, $245, $574, $434, $567, $476 East North Central, East South Central, Mid-Atlantic, Mountain, New England, Pacific, South Atlantic, West North Central, and West South Central regions, respectively. The cost and payment of treating an opioid related visit were highest in West South Central region and West North Central, respectively. The costs associated with the visit were typically higher than the average payment. Opioid-related visits are a highly prevalent and costly epidemic placing a large burden on US hospitals.

PHS30 - Geographic Variation in Opioid Related Inpatient Healthcare Utilization in the United States: A Retrospective Study of a Large Health System’s Claims Data

Despite the rise in opioid abuse and dependence, there has been little research on the geographic variation of healthcare utilization. The objective was to estimate the geographic variation in costs, payments, and length of stay (LOS) associated with inpatient opioid related hospital visits in a large hospital claims database. The Vizient Health System Database was used to identify opioid related patient visits between January 2014 and June 2017. Eligible visits had a primary diagnosis of opioid use or dependence defined by International Classification of Diseases, 9th and 10th Revision (ICD-9/10). Geography was defined using the 9 US Census sub-regions. Multivariable regression models were performed to estimate the mean expected cost and payment per visit for each region controlling for patient demographics, comorbid conditions, hospital characteristics, and year. A total of 32,713 hospital inpatient visits were identified in 359 hospitals. The overall adjusted mean cost, payment, and LOS for an opioid visit was $4,383, $6,689, and 4.35 days, respectively. The adjusted mean expected cost was $3,336, $2,894, $5,835, $4,472, $4,885, $6,671, $3,979, $5,192, $3,520 in the East North Central, East South Central, Mid-Atlantic, Mountain, New England, Pacific, South Atlantic, West North Central, and West South Central regions. The adjusted mean expected payment was $7,984, $4,038, $7,070, $9,001, $7,897, $8,183, $5,551, $6,490, $5,617 in the East North Central, East South Central, Mid-Atlantic, Mountain, New England, Pacific, South Atlantic, West North Central, and West South Central regions. The Pacific and East North Central Region had the highest (4.95 days) and lowest (3.77 days) LOS, respectively. The cost and payment of treating an opioid related visit was highest in Mid-Atlantic and Pacific regions, respectively. Visit-associated costs associated were lower than the payment. Opioid related visits are a highly prevalent and costly epidemic placing a large burden on US hospitals.

Geographic Variance of Cost Associated With Hysterectomy

(Abstracted from Obstet Gynecol 2017;129:844–853) With recent changes in access to health care services and the aging populations in the United States, increasing attention is being given to understanding factors contributing to the cost of health care. A recent study conducted by the National Academy of Medicine assessed Medicare costs across the country and found significant geographic variation in cost in the United States.

Geographic variations in the cost of spine surgery.

Retrospective review. To define the geographic variation in costs of anterior cervical discectomy and fusion (ACDF) and posterolateral fusion (PLF). ACDF and lumbar PLF are common procedures that are used in the treatment of spinal pathologies. To optimize value, both the benefits and costs of an intervention must be quantified. Data on costs are scarce in comparison with data on total charges. This study aims at defining the costs of ACDF and PLF and describing the geographic variation within the United States. Medicare Provider Utilization and Payment data were used to investigate the costs associated with ACDF, PLF, and total knee arthroplasty (TKA). Average total costs of the procedures were compared by state and geographic region. Combined professional and facility costs for a single-level ACDF had a national mean of $13,899. Total costs for a single-level PLF had a mean of $25,858. Total costs for a primary TKA had a national mean of $13,039. The cost increased to an average of $22,138 for TKA with major comorbidities. Analysis of geographic trends showed statistically significant differences in total costs of PLF, TKA, and TKA, with major complications or comorbidities between geographic regions (P < 0.01 for all). Three of the 4 procedures (PLF, TKA, and TKA with major complications or comorbidities) showed statistically significant variation in cost between geographic regions. The Midwest provided the lowest cost for all procedures. Similar geographic trends in the cost of spinal fusions and TKAs suggest that these trends may not be limited to spine-related procedures. Surgical costs were found to correlate with cost of living but were not associated with the population of the state. These data shed light on the actual cost of common surgical procedures throughout the United States and will allow further progress toward the development of cost-effective, value-driven care. 3.

Factors associated with geographic variation in cost per episode of care for three medical conditions.

ObjectiveTo identify associations between market factors, especially relative reimbursement rates, and the probability of surgery and cost per episode for three medical conditions (cataract, benign prostatic neoplasm, and knee degeneration) with multiple treatment options.MethodsWe use 2004–2006 Medicare claims data for elderly beneficiaries from sixty nationally representative communities to estimate multivariate models for the probability of surgery and cost per episode of care as a function local market factors, including Medicare physician reimbursement for surgical versus non-surgical treatment and the availability of primary care and specialty physicians. We used Symmetry’s Episode Treatment Groups (ETG) software to group claims into episodes for the three conditions (n = 540,874 episodes).ResultsHigher Medicare reimbursement for surgical episodes and greater availability of the relevant specialists are significantly associated with more surgery and higher cost per episode for all three conditions, while greater availability of primary care physicians is significantly associated with less frequent surgery and lower cost per episode.ConclusionRelative Medicare reimbursement rates for surgical vs. non-surgical treatments and the availability of both primary care physicians and relevant specialists are associated with the likelihood of surgery and cost per episode.

Standardizing Medicare payment information to support examining geographic variation in costs.

Examination of efficiency in health care requires that cost information be normalized. Medicare payments include both geographic and policy-based facility type differentials (e.g., wage index and disproportionate share hospital), which can bias cost comparisons of hospitals and averages across geographic areas. Standardizing payment information to remove the area- and policy-based payment differentials should normalize much of the observed geographic variability in payments, allowing for a more accurate comparison of resource use between providers and across geographic regions. Use of standardized payments will ensure that observed payment variation is due to differences in practice patterns and service use, rather than Medicare payment differences over which the providers have no control. This paper describes a method for standardizing claim payments, and demonstrates the difference in actual versus standardized payments by geographic region. We used a nationwide cohort of Medicare patients hospitalized with an acute myocardial infarction (AMI) in 2007, then limited our study to those with Medicare Part A and Part B fee-for-service (FFS), and Part D coverage (n = 143,123). Standardized payment amounts were calculated for each Part A and Part B claim; standardized and actual payments were summed for all services for each patient beginning with the index hospitalization through 12 months post discharge. Without standardization of payments, certain areas of the country are mischaracterized as either high or low healthcare resource-consuming areas. The difference between actual and standardized payments varies by care setting. Standardized payment amounts should be calculated when comparing Medicare resource use across geographic areas.

Geographic Variations in Cost of Living: Associations With Family and Child Well‐Being

The effects of geographic variations in cost of living and family income on children's academic achievement and social competence in first grade (mean age = 86.9 months) were examined, mediated through material hardship, parental investments, family stress, and school resources. Using data from the Early Childhood Longitudinal Study-Kindergarten Cohort (N = 17,565), higher cost of living was associated with lower academic achievement. For poor children only, higher cost of living was also detrimental to parental investments and school resources. Parental investments and school resources were more strongly associated with achievement for lower income than higher income children. Results suggest that cost of living intersects with income in meaningful ways for family and child well-being and should be accounted for in the poverty measure.

Relative Status and Well-Being: Evidence from U.S. Suicide Deaths

This paper empirically assesses the theory of interpersonal income comparison using individual level data on suicide deaths in the United States. We model suicide as a choice variable, conditional on exogenous risk factors, reflecting an individual's assessment of current and expected future utility. Our empirical analysis considers whether suicide risk is systematically related to the income of others, holding own income and other individual factors fixed. We estimate proportional hazards and probit models of the suicide hazard using two separate and independent data sets: (1) the National Longitudinal Mortality Study and (2) the Detailed Mortality Files combined with the 5 percent Public Use Micro Sample of the 1990 decennial census. Results from both data sources show that, controlling for own income and individual characteristics, individual suicide risk rises with reference group income. This result holds for reference groups defined broadly, such as by county, and more narrowly by county and one demographic marker (e.g., age, sex, race). These findings are robust to alternative specifications and cannot be explained by geographic variation in cost of living, access to emergency medical care, mismeasurement of deaths by suicide, or by bias due to endogeneity of own income. Our results confirm findings using self-reported happiness data and are consistent with models of utility featuring external habit or Keeping Up with the Joneses preferences.

Relative Status and Well-Being: Evidence from U.S. Suicide Deaths

This paper empirically assesses the theory of interpersonal income comparison using individual level data on suicide deaths in the United States. We model suicide as a choice variable, conditional on exogenous risk factors, reflecting an individual's assessment of current and expected future utility. Our empirical analysis considers whether suicide risk is systematically related to the income of others, holding own income and other individual factors fixed. We estimate proportional hazards and probit models of the suicide hazard using two separate and independent data sets: (1) the National Longitudinal Mortality Study and (2) the Detailed Mortality Files combined with the 5 percent Public Use Micro Sample of the 1990 decennial census. Results from both data sources show that, controlling for own income and individual characteristics, individual suicide risk rises with reference group income. This result holds for reference groups defined broadly, such as by county, and more narrowly by county and one demographic marker (e.g., age, sex, race). These findings are robust to alternative specifications and cannot be explained by geographic variation in cost of living, access to emergency medical care, mismeasurement of deaths by suicide, or by bias due to endogeneity of own income. Our results confirm findings using self-reported happiness data and are consistent with models of utility featuring external habit or Keeping Up with the Joneses preferences.

The Institute of Medicine committee's clarion call for universal coverage.

The Institute of Medicine (IOM) Committee on the Consequences of Being Uninsured is to be commended for its work, which recently culminated in the release of six volumes on the subject. The concluding volume presents a vision for universal coverage and describes four options for achieving it. The options include an incremental approach, employer and individual mandates, and a single-payer plan. We identify complications involving benefits and geographic variation in costs surrounding attempts to achieve universal coverage. The complications suggest that the committee's cost estimates may be too low and that there may be sizable political barriers to the proposals.

Does rainfall constrain the evolution of tail length in widowbirds?

Widowbirds (Euplectes, Ploceidae) are sexually dimorphic weaverbirds in which the males have elongated tails that show much interspecific variation. One possible cause of such variation is geographic variation in cost due rainfall making it more difficult to fly. This hypothesis was tested by comparing tail length to the modal and maximum rainfall levels in which the populations occur. Overall, there was a significant negative relationship between maximum rainfall and tail length, and a similar but non-significant trend with modal rainfall. Phylogenetic effects were partially controlled for by testing for the same relationships within two species groups and across species groups. In all cases, the trends were as in the overall test, suggesting that phylogenetic effects alone cannot explain the relationship. These results suggest that geographic variation in costs could explain inter-population variation in sexually selected ornaments. Other reasons for the relationship with rainfall are also possible.

A competitive bidding approach to physician payment.

Prologue: For the most part, efforts to reform the way Medicare pays physicians have focused on administrative fee-setting approaches: fee schedules, adjustments of fees to reflect geographic variation in costs of care, rebasing the Medicare Economic Index, and other strategies. The rationale for administered pricing centers on the pervasive view that to continue the current style of paying physicians under Medicare will merely drive prices higher. One voice in the crowd, however, has suggested an alternative approach to meeting the goals of physician payment reform: competitive bidding. In this article, economist Jeff McCombs advocates retaining market-driven physician payment; his proposal draws heavily on the competitive bidding model used in setting payments for durable medical equipment and outpatient clinical laboratories under Medicare. His approach keeps the task of setting prices in the hands of providers, who, he contends, know best what it costs to provide health care services. While similar to...