Reducing the Cost of Medicaid: A Multistate Simulation

813311 HIS0010.1177/1178632918813311Health Services InsightsLinder et al research-article2018 Reducing the Cost of Medicaid: A Multistate Simulation Stephen H Linder1,2 , Kimberly Aguillard1,2, Kelsey French3 and Arthur Garson1 Health Services Insights Volume 11: 1–9 © The Author(s) 2018 Article reuse guidelines: sagepub.com/journals-permissions DOI: 10.1177/1178632918813311 https://doi.org/10.1177/1178632918813311 1Health Policy Institute, Texas Medical Center, Houston, TX, USA. 2Institute for Health Policy, School of Public Health, The University of Texas Health Science Center at Houston, Houston, TX, USA. 3Formerly Jones Graduate School of Business, Rice University, Houston, TX, USA. ABSTRACT: According to some estimates, the United States wastes as much as 30% of health care dollars. Some of that waste can be mitigated by reducing certain costs associated with Medicaid. We chose 5 areas of savings applicable to Medicaid: (1) modification of physician payment models to reduce unnecessary care, (2) development of a medication adherence program for patients dually eligible for Medicaid and Medicare support (“dual eligibles”), (3) improvement in unnecessary admissions and readmissions for dual eligibles, (4) reduction in emergency department visits among children in Medicaid and dual-eligible beneficiaries, and (5) improvement in adoption of endof-life advance directives. We chose the states from both ends of the spending spectrum: the 5 with the lowest annual Medicaid expenditures: Wyoming, South Dakota, Montana, Vermont, and Alaska, and those with the highest: California, New York, Texas, Pennsylvania, and Florida. This spectrum demonstrates the range of potential cost-saving measures, from US $23.6 million in Wyoming to US $3.4 billion in California. We conclude that there are a number of ways to reduce Medicaid spending and improve quality. To the extent that states have already adopted programs addressing the same problems, our approach may be supplementary but the total savings may be achieved with a combination of current initiative and those described here. As Medicaid creates savings, physician payment could be increased to attract more physicians into caring for Medicaid patients. KEYWORDS: Medicaid, dual-eligible beneficiaries, hospital readmissions RECEIVED: October 15, 2018. ACCEPTED: October 20, 2018. TYPE: Perspective DECLARATION OF CONFLICTING INTERESTS: The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article. FUNDING: The author(s) received no financial support for the research, authorship, and/or publication of this article. CORRESPONDING AUTHOR: Stephen H Linder, Health Policy Institute, Texas Medical Center, 6550 Bertner Avenue, Houston, TX 77030, USA. Email: [email protected] Introduction care spending, a framework Berwick and Hackbarth2 refined to examine a variety of cost-saving initiatives to target wasteful spending in key areas such as failures in care coordination/delivery and overtreatment. The key areas identified in the Wedges Model led us to identify 5 approaches to savings, which could contribute to stabilizing health care spending through streamlining and strengthening care coordination and minimizing unnecessary treatment. The innovative initiatives highlighted are adaptable, relatively low-cost investments, yielding meaningful savings to Medicaid. The programs also aid vulnerable and costly health care populations. The proposed initiatives, applied to the 10-state sample to represent the full range of potential cost savings, include reduction in unnecessary care, improved medication adherence in dual-eligible beneficiaries, improved care for dual-eligible beneficiaries to reduce hospital readmissions, reduction in emergency department (ED) visits among children in Medicaid and dual-eligible beneficiaries, and improved coordination for end-of-life care. According to the Institute of Medicine, 30% of dollars spent on health care is waste in the American medical system.1 Berwick and Hackbarth2 conducted additional analysis and placed a midpoint estimate of waste at 34% of national health spending. Although Medicaid programs operate with extremely low margins, and prospective savings will not approach 30%, savings are possible. Medicaid in the state of Texas, for example, represents just over 30% of the state budget and covers 4.4 million people. In 2016, Texas spent US $18 billion on Medicaid; with the Federal Medical Assistance Percentage (FMAP) at 57.13%, the federal government paid US $20.5 billion, for the total state + federal shares at US $38.5 billion.3 Even with only a 5% reduction, annual savings could equal well over US $1 billion, which could be used to improve care and potentially redistribute funding to chronically underfunded areas, such as payments to physicians. In Alaska, where the FMAP is 50%, the total spending on Medicaid (including state and federal contributions) in 2016 was US $1.42 billion3; Alaska’s portion of this payment is $710 million. Applying a 5% savings would yield US $35.5 million for other state programs and priorities. This article adds to current work on Medicaid reform by exploring specific pathways to find and estimate savings that, at the same time, maintain or improve quality of care. These results should be applicable to Medicaid programs across the states, as well as to other parts of the US health care system. We use the “Wedges Model” for examining proposed reductions in health Pathways to Savings Health workforce initiatives To reduce overutilization of EDs. The overuse of EDs is a large drain of health care dollars. Routine care provided in an ED setting can be 2 to 5 times more expensive than the same care provided in an alternate setting such as an urgent care clinic.4 A Health Partners study discovered charges for treating strep Creative Commons Non Commercial CC BY-NC: This article is distributed under the terms of the Creative Commons Attribution-NonCommercial 4.0 License (http://www.creativecommons.org/licenses/by-nc/4.0/) which permits non-commercial use, reproduction and distribution of the work without further permission provided the original work is attributed as specified on the SAGE and Open Access pages (https://us.sagepub.com/en-us/nam/open-access-at-sage). Health Services Insights 2 throat in the ED to be US $328, US $130 at an urgent care center, and US $122 in a primary care office.5 Clearly, based on these figures, it is critical to find ways to treat as many patients as possible outside of an ED due to the 261% price premium that ED care costs. Massachusetts conducted an in-depth analysis of their ED usage and the emergent status of patient’s health when visiting the ED. Of all ED visits, 42% were classified as avoidable.6 If we extrapolated this percentage to California, this would translate to 5 749 000 avoidable ED visits per year, and in South Dakota, this percentage would translate to 114 000 annual avoidable ED visits.7 The Massachusetts report also highlighted disparities between incomes: the residents in the lowest income quartile, after adjusting for age and sex, had greater than 3 times the avoidable ED rate than residents in the highest income quartile. The lowest income quartile would represent uninsured and Medicaid populations. The most beneficial program to reduce ED usage is one that prevents unnecessary trips to the ED. This can involve enhancing programs such as “Grand-Aides” to assist patients in health care management and reduce their perceived need for ED treatment. Grand-Aides are nurse aides who are closely supervised by nurses and foster relationships with patients and family with the goal of appropriate use of the ED. Calculations indicate Grand-Aides could potentially reduce Medicaid ED visits by 74% in Medicaid children and patients dually eligible for Medicaid and Medicare support (“dual eligibles”).8 payment methods. For physicians who are already part of hospital systems or Accountable Care Organizations (ACOs) it would be reasonable to convert to a salaried system (at their current yearly income), with a relatively modest bonus (ie, 5%-10%) for quality. Physician income does not need to decrease with these changes. As Medicaid generates savings in this and other areas, physician payment should increase to attract more physicians into caring for Medicaid patients. Salary + bonus would be the dominant method of payment. For physicians not in systems, Medicaid could test the resultant effect on patient care of paying a certain amount per patient with a bonus for quality. It could also change to FFS payment with not only incentives for quality but also disincentives for doing what physicians’ own specialty societies determine in their guidelines to be unnecessary or harmful. The Centers for Medicare and Medicaid Services (CMS) has announced that by the end of 2018, more than half of Medicare dollars will be paid via alternative payment models that focus on reducing the negative incentives associated with paying physicians based on FFS. Of note, the health care systems in the United States that are routinely ranked the highest for quality (eg, Mayo Clinic, Cleveland Clinic, and Kaiser Permanente) have salaried physicians, some with and some without a bonus. Such systems have demonstrated savings between 20% and 46% due to a decrease in tests ordered and procedures performed.13,14 For the purposes of this analysis, we assume 15% savings. To improve medication adherence among dual eligibles. About 50% of patients with chronic diseases take their medications appropriately.9 Medication nonadherence among the other 50% generates a significant cost burden. Dual eligibles represent 15% of the national Medicaid population but require 33% of Medicaid spending.10 This high level of spending can be partially attributed to the dual-eligible population’s vulnerability and complicated chronic health conditions. Significantly, the Grand-Aides program has achieved a 91% medication adherence in patients with heart failure 1 month after discharge.11 Advance directives To reduce avoidable hospitalization among dual eligibles. Dualeligible beneficiaries are at a higher risk for potentially avoidable hospitalizations—admissions and readmissions. Among hospital visits in this population, just over one quarter (26%) of hospitalizations have been determined to be unnecessary, many due to readmissions.12 The Grand-Aides program is one initiative achieving the aim of reducing readmissions with a demonstrated ability to reduce hospital readmissions by 58%.11 Payment Initiatives To reduce unnecessary care Most physicians are still predominately paid on a fee-forservice (FFS) basis. Medicaid programs could propose new To improve end-of-life care. Approximately US $205 billion is spent in the United States on patients in the last year of life or 13% of the annual total spending on health care.15 A number of strategies are incorporated to improve the quality of a person’s last days. These approaches must be exquisitely sensitive to improving the quality of life of the patient and loved ones address the mislabeled “death panels” from the past. The most successful approach involves recording the wishes of the individual patient and family, broadly called “advance directives,” which fall into 3 categories: living wills, power of attorney and health care proxy. One calculation places the savings through advance directives at US $5585 per patient.16 This figure was the most recent study reported from a 2016 systematic review of advanced care planning cost savings. Estimates in the review varied widely from US $1041 to US $64 830 per patient, based on the length of the study and the method for measuring cost.17 These savings are realized by reduced usage of EDs and reductions of extraordinary life-saving measures while honoring the patient’s and their family’s wishes. Only 65% of nursing home patients have an advance directive.18 There is a great opportunity, as up to 90% of nursing home patients and families will complete advanced directives if a physician initiates the Linder et al 3 419 828 000 1 869 608 000 Abbreviations: AK, Alaska; CA, California; FL, Florida; MT, Montana; nY, new York; PA, Pennsylvania; Sd, South dakota; TX, Texas; VT, Vermont; wY, wyoming. 1 143 292 000 915 635 000 1 024 674 000 59 909 000 68 265 000 54 505 000 23 680 000 Total cost savings to Medicaid 43 333 000 84 075 000 45 760 000 49 099 000 23 154 000 35 378 000 1 270 000 1 587 000 1 380 000 818 000 Emergency department visit reduction 1 273 000 684 338 000 373 275 000 221 689 000 190 094 000 198 485 000 7 406 000 14 313 000 8 397 000 3 453 000 reduction in avoidable dual-eligible hospital readmissions 6 914 000 406 393 000 221 669 000 131 652 000 112 887 000 117 871 000 4 398 000 8 500 000 4 985 000 2 052 000 End-of-life coordination of care 4 105 000 2 078 809 000 1 133 896 000 673 438 000 577 447 000 602 945 000 22 498 000 43 492 000 25 498 000 20 998 000 95 008 000 67 414 000 12 053 000 69 995 000 24 337 000 373 000 PA, US $ AK, US $ VT, US $ MT, US $ 14 245 000 10 043 000 10 499 000 Improved adherence was calculated from 50% to 75% for 4 chronic conditions: hypertension, diabetes, heart failure, and dyslipidemia.23 State populations for each chronic condition were estimated using data from the CMS Chronic Conditions dual-eligible medication adherence program Improved Medication Adherence Table 1. Overall proposed cost savings. Reduction in Overutilization of EDs Sd, US $ This research explored the states from both ends of the spending spectrum: the 5 with the lowest annual Medicaid expenditures—Wyoming, South Dakota, Montana, Vermont, and Alaska—and those with the highest—California, New York, Texas, Pennsylvania, and Florida. This spectrum demonstrates the range of potential cost-saving measures, from US $23.6 million in Wyoming to US $3.4 billion in California. 6 858 000 FL, US $ wY, US $ Results Using a statistic of 24.8% of Medicaid children21 and 44% of dual-eligible beneficiaries,22 ED visits are calculated for each state for these populations. Next, 50% of the maximum possible reduction from the Grand-Aides program is applied, which is a 37% reduction. The savings applied include the cost of the Grand-Aides program. If the Grand-Aides program were implemented to assist these key populations (assuming 50% of the possible benefit = 37% reduction), it could result in Medicaid savings of US $243 million in this 10-state sample, with state savings ranging from US $409 000 in Wyoming to US $42 million in California (Table 2). TX, US $ 5 STATES wITH HIGHEST MEdICAId EXPEndITUrES 5 STATES wITH LOwEST MEdICAId EXPEndITUrES Using Urban Institute calculations as of January 2017, based on CMS Form 64, the 5 lowest total Medicaid expenditure states (Wyoming, South Dakota, Montana, Vermont, and Alaska) and the 5 highest Medicaid expenditure states (Florida, Pennsylvania, Texas, New York, and California), were selected for evaluation. Note that North Dakota and Idaho were likely among the lowest Medicaid expenditure states, but they did not have complete reporting to generate adequate data for equal comparisons; therefore, these states were not used in this analysis. Enrollment figures for Medicaid, full dual-eligible beneficiaries, and children enrolled in Medicaid were obtained from the December 2016 MACPAC Databook for various calculations.20 The FY17 FMAP was applied to reflect the state share of Medicaid savings. Each approach to savings was applied to the 10-state sample to evaluate potential cost savings. If all programs are implemented, the total financial benefit to states ranges from US $11.8 million in Wyoming to US $1.7 billion in California (Table 1). nY, US $ Methods Salaried physicians for reduction in unnecessary care CA, US $ discussion.19 The percentage of patients aged 65 and older with recorded advance care plans or surrogate decision makers designated in their medical records is a quality communication and care coordination process measure in the Merit-Based Incentive Payment System for many disciplines. 166 213 000 3 4 Table 2. Savings calculations for reduction in emergency department use for children and dual eligibles in Medicaid. 5 STATES wITH LOwEST MEdICAId EXPEndITUrES wY Total Medicaid Pop (FY13, average monthly enrollment), n Sd MT 5 STATES wITH HIGHEST MEdICAId EXPEndITUrES VT AK FL PA TX nY CA 2 159 000 4 081 000 5 115 000 9 307 000 107 000 114 000 170 000 111 000 3 386 000 65 59 57 34 55 51 42 63 35 36 Children on Medicaid, n 44 000 63 000 65 000 58 000 61 000 1 727 000 914 000 2 590 000 1 815 000 3 340 000 Average Ed use for Medicaid children 24.8%, n 10 912 15 624 16 120 14 384 15 128 428 296 226 672 642 320 450 120 828 320 Grand-Aides reduces Ed visits by 37% (assume 50% of opportunity of 74%), n 4037 5780 5964 5322 5597 158 469 83 868 237 658 166 544 306 478 Grand-Aides US $158 savings per Ed visit, including cost of program, US $ 637 846 913 240 942 312 840 876 884 326 25 038 102 13 251 144 37 549 964 26 313 952 48 423 524 dual-eligible population, n 7000 14 000 17 000 29 000 15 000 402 000 385 000 449 000 756 000 1 386 000 Average Ed usage for dual-eligible population 44%, n 3080 6160 7480 12 760 6600 176 880 169 400 197 560 332 640 609 840 Grand-Aides reduces Ed visits by 37% (assume 50% of opportunity of 74%), n 1139 2279 2767 4721 2442 65 445 62 678 73 097 123 076 225 640 Grand-Aides US $158 savings per Ed visit, US $ 180 057 360 113 437 280 745 918 385 836 10 340 310 9 903 124 11 549 357 19 446 134 35 651 246 Total Medicaid savings through reduction in Ed visits, US $ 817 903 1 273 353 1 379 592 1 586 794 1 270 162 35 378 412 23 154 268 49 099 321 45 760 086 84 074 770 50 45 34 45.54 50 38.9 43.82 50 50 408 951 573 772 475 131 722 625 635 081 13 762 202 21 515 322 22 880 043 42 037 385 Children as % of Medicaid population State share, % State savings, US $ 48 11 164 988 Abbreviations: AK, Alaska; CA, California; FL, Florida; MT, Montana; nY, new York; PA, Pennsylvania; Sd, South dakota; TX, Texas; VT, Vermont; wY, wyoming. data obtained from Garson et al8; MACPAC20; Cubanski et al22; and Gindi and Jones.21 Health Services Insights 68 000 Linder et al Chartbook.24 Assuming that 50% of patients take their medication appropriately, an improvement to 75% adherence would produce savings displayed in Table 3. These significant savings also incorporate increased drug cost as a result of drug adherence. Importantly, those expenses are offset due to overall reductions in health care expenditures for costly services such as ED visits, hospital admissions, additional diagnostic testing, and increased pharmacy expenses related to treatment. For patients with hypertension, potential savings in the 10-state sample equal US $2 billion; for heart failure, total savings equal US $1.55 billion; for diabetes, US $1.17 billion; and for dyslipidemia, US $260 million. The potential cost savings for state Medicaid range from US $5.2 million in Wyoming to US $1 billion in California. Reduction in Avoidable Hospitalizations Assuming the Grand-Aides program would achieve 50% reduction in hospital admissions, and calculating the cost of a readmission based on US $15,435,25 the net savings to the Medicaid program could range from US $3.4 million in Wyoming to US $684 million in California, including the expense of operating the Grand-Aides program (Table 4).26] Reduction in Unnecessary Procedures Assuming that paying physicians a salary plus bonus could result in a 15% reduction in tests and procedures; in the 10-state sample (based on the 2016 Medicaid expenditure data),27 these measures result in state savings ranging from US $3.4 million in Wyoming to US $83.1 million in California (Table 5). Vermont classifies most of the Medicaid expenditures as “other services” so the state savings for this innovation are small, US $170 000. Improved End-of-Life Care About 21% of dual-eligible beneficiaries are in long-term services and supportive living according to the 2017 MedPAC Databook.10 End-of-life care cost savings were estimated for each state by applying a 25% increase in advance directives among this population with the estimated saving of US $5585 per directive. This results in state savings ranging from US $1 million in Wyoming to US $203 million in California (Table 6). Discussion and Conclusions Several caveats are important related to the estimated improved adherence calculations: first, the savings are in the short term (ie, hospital admissions over several years) and do not take into account the cost of future disease if the current disease is wellcontrolled (eg, hypertension is well-controlled and a stroke is avoided, only to have the patient get cancer). Second, there is clear overlap in the patient diagnoses (ie, many patients with diabetes have heart disease). Improvement (and medication adherence) in one of these diagnoses will likely have a positive effect on the other diseases in the patient and therefore these are, again, maximal numbers. Personal reinforcement and teaching are among the most promising approaches to 5 improving medication adherence, as the American Diabetes Association recognizes and recommends.28 Programs such as Grand-Aides with a 91% medication adherence could be extremely beneficial. Physician Payments Although combining alternative physician payment models are the basis for part of Medicare Access and CHIP Reauthorization Act of 2015 (MACRA), it is unwieldy. In addition, there will be added expenses as managed care companies and physicians switch to a value-based payment system because the data infrastructure to track these metrics must be in place. These expenses must be netted against potential savings (ie, paid to systems and physicians) as the requirements for financial outlays are real. This issue should be addressed through more sophisticated and interoperable Electronic Health Record (EHR) systems, which are, unfortunately, likely a decade away. An alternative method to reduce unnecessary care would be to examine the 20 most expensive procedures and tests (because of high volume, high price, or both) and compare the indications for the tests or procedures reported by the ordering physician to national guidelines produced by the physician’s specialty society. A recent such analysis revealed that 34% of echocardiograms performed on preoperative patients were unnecessary and were outside the recommended practice guidelines.29 We are recommending the physician payment change because it could be achieved more simply (eg, the MACRA regulations could provide an incentive for programs in which at least 50% of their physicians are salaried). These estimates do not take into account existing state programs that could have already achieved some of the savings. The program overlap poses a significant limitation (ie, the same savings may be attributed to more than one program), as well. The data are likely to be correct within an order of magnitude; rather than focusing on the exact amounts, we suggest that “large (say, 10%), medium (5%) and small (2.5%)” be attached to the possible program savings and be made available as a potential supplement to the cost-saving work already being done by the state Medicaid programs. We have examined a number of possible approaches to reducing the expense of Medicaid. These savings should remain in the Medicaid program and, for example, help to cover more people and increase physician reimbursement. Increased reimbursements will enhance the number of physicians seeing Medicaid patients, thus improving access for the underserved. If all programs are implemented, the total financial benefit to states ranges from US $11.8 million in Wyoming to US $1.7 billion in California, as illustrated in Table 1. These 5 initiatives also could be applied to commercially insured patients or those covered by Medicare, resulting in major savings across the United States. Realizing these savings in achievable ways suggested in this article could make a major dent in the rising cost of health care. 6 Table 3. Savings calculations for improvement in drug adherence by dual eligibles, by disease category. 5 STATES wITH LOwEST MEdICAId EXPEndITUrES wY Total dual-eligible population Sd MT VT 5 STATES wITH HIGHEST MEdICAId EXPEndITUrES AK FL PA TX nY CA 7000 14 000 17 000 29 000 15 000 402 000 385 000 449 000 756 000 1 386 000 60% of pop, n 4200 8400 10 200 17 400 9000 241 200 231 000 269 400 453 600 831 600 50% nonadherence, n 2100 4200 5100 8700 4500 120 600 115 500 134 700 226 800 415 800 4 103 400 8 206 800 9 965 400 16 999 800 8 793 000 235 652 400 225 687 000 263 203 800 443 167 200 812 473 200 1610 3220 3910 6670 3450 92 460 88 550 103 270 173 880 318 780 805 1610 1955 3335 1725 46 230 44 275 51 635 86 940 159 390 3 148 757 6 297 515 7 646 982 13 040 941 6 747 337 180 828 645 173 181 662 201 970 302 340 065 810 623 453 985 36% of pop, n 2520 5040 6120 10 440 5400 144 720 138 600 161 640 272 160 498 960 50% nonadherence, n 1260 2520 3060 5220 2700 72 360 69 300 80 820 136 080 249 480 2 366 280 4 732 560 5 746 680 9 803 160 5 070 600 135 892 080 130 145 400 151 779 960 255 558 240 468 523 440 40% of pop, n 2800 5600 6800 11 600 6000 160 800 154 000 179 600 302 400 554 400 50% nonadherence, n 1400 2800 3400 5800 3000 80 400 77 000 89 800 151 200 277 200 800 600 1 761 200 2 138 600 3 648 200 1 887 000 50 571 600 48 433 000 56 484 200 95 104 800 174 358 800 10 499 037 20 998 075 25 497 662 43 492 101 22 497 937 602 944 725 577 447 062 673 438 262 1 133 896 050 50 45.06 34.44 45.54 50 38.9 48.22 43.82 50 50 5 249 519 9 461 733 8 781 395 19 806 302 11 248 969 234 545 498 278 444 973 295 100 646 566 948 025 1 039 404 712 Hypertension Savings with 75% adherence at US $3908/ patient, US $ Heart failure 23% of pop, n 50% nonadherence, n Savings with 75% adherence at US $7823 per patient, US $ diabetes Savings with 75% adherence at US $3756 per patient, US $ dyslipidemia Total savings State share, % State savings—FMAP 2017 contributions applied, US $ Abbreviations: AK, Alaska; CA, California; FL, Florida; MT, Montana; nY, new York; PA, Pennsylvania; Sd, South dakota; TX, Texas; VT, Vermont; wY, wyoming. data drawn from roebuck et al23 and Centers for Medicare and Medicaid Services.24 2 078 809 425 Health Services Insights Savings with 75% adherence at US $1258 per patient, US $ Linder et al Table 4. Savings calculations for reduction in hospital readmission costs for dual eligibles. 5 STATES wITH LOwEST MEdICAId EXPEndITUrES wY Sd MT VT 5 STATES wITH HIGHEST MEdICAId EXPEndITUrES AK FL PA TX nY CA Full dual-eligible population, n 7000 14 000 17 000 29 000 15 000 402 000 385 000 449 000 756 000 1 386 000 dual-eligible hospitalization (27%), n 1890 3780 4590 7830 4050 108 540 103 950 121 230 204 120 374 220 Avoidable hospitalizations (26%), n 491 983 1193 2035 1053 28 220 27 027 31 519 53 071 97 297 Expense to Medicaid calculated at US $15 667 per readmission, US $ 7 692 497 15 400 661 18 702 661 31 882 345 16 497 351 442 122 740 423 432 009 493 808 173 831 463 357 1 524 352 099 Grand-Aides could reduce readmissions by 50%, US $ 3 846 248 7 700 330 9 351 330 15 941 172 8 248 675 221 061 370 211 716 004 246 904 086 415 731 678 762 176 049 Grand-Aides cost US $800 per individual per year, applied to 26% preventable hospital population, US $ 392 800 786 400 954 400 1 628 000 842 400 22 576 000 21 621 600 25 215 200 42 456 800 77 837 600 Calculated Medicaid savings, US $ 3 453 448 6 913 930 8 396 930 14 313 172 7 406 275 198 485 370 190 094 404 221 688 886 373 274 878 684 338 449 50 45.06 34.44 45.54 50 38.9 48.22 43.82 50 50 1 726 724 3 116 799 2 891 902 6 518 218 3 703 137 77 210 808 91 663 521 97 144 069 186 637 439 342 169 224 State share, % State savings, US $ Abbreviations: AK, Alaska; CA, California; FL, Florida; MT, Montana; nY, new York; PA, Pennsylvania; Sd, South dakota; TX, Texas; VT, Vermont; wY, wyoming. data drawn from Segal12; Garson26; and Fitch et al.25 7 8 Table 5. Savings calculations for reduction in unnecessary care with salaried physicians. 5 STATES wITH LOwEST MEdICAId EXPEndITUrES 5 STATES wITH HIGHEST MEdICAId EXPEndITUrES wY, US $ Sd, US $ MT, US $ VT, US $ AK, US $ FL, US $ PA, US $ TX, US $ nY, US $ CA, US $ 45 722 882 66 956 387 94 969 409 2 488 213 162 246 654 466 630 080 80 351 103 449 423 819 633 387 993 1 108 088 171 6 858 432 10 043 458 14 245 411 373 232 24 336 998 69 994 512 12 052 665 67 413 572 95 008 198 166 213 225 State share, % 50 45.06 34.44 45.54 50 38.9 48.22 43.82 50 50 State savings, US $ 3 429 216 4 525 582 4 906 119 170 011 12 168 499 27 227 865 5 811 795 29 540 627 47 504 099 83 106 612 Medicaid physician, lab, and X-ray, US $ 15% savings Abbreviations: AK, Alaska; CA, California; FL, Florida; MT, Montana; nY, new York; PA, Pennsylvania; Sd, South dakota; TX, Texas; VT, Vermont; wY, wyoming. data obtained from KFF as of January 2017, The structure of Vermont’s Medicaid program formulates most of the state’s Medicaid expenditures in the category of “Other Services.” Table 6. Savings calculations for coordination of end-of-life care. 5 STATES wITH LOwEST MEdICAId EXPEndITUrES wY Sd MT VT 5 STATES wITH HIGHEST MEdICAId EXPEndITUrES AK FL PA TX nY CA 7000 14 000 17 000 29 000 15 000 402 000 385 000 449 000 756 000 1 386 000 21% in long-term services and support, n 1470 2940 3570 6090 3150 84 420 80 850 94 290 158 760 291 060 2 052 487 4 104 975 4 984 612 8 500 370 4 398 187 117 871 425 112 886 812 131 652 412 221 668 650 406 392 525 50 45.06 34.44 45.54 50 38.9 48.22 43.82 50 50 1 026 243 1 849 701 1 714 706 3 871 068 2 199 093 45 851 984 54 434 020 57 690 086 110 834 325 203 196 262 with 25% increase in advance directives at US $5585 per patient, US $ State share, % State savings, US $ Abbreviations: AK, Alaska; CA, California; FL, Florida; Id, Idaho; MT, Montana; nY, new York; PA, Pennsylvania; Sd, South dakota; TX, Texas; VT, Vermont; wY, wyoming. data obtained from nicholas et al16 and The Medicare Payment Advisory Commission and the Medicaid and CHIP Payment and Access Commission.10 Health Services Insights Full dual-eligible population, n Linder et al 9 Author Contributions AG, original concept, design and intervention sections. KF, data sampling and simulation method. KA, expanded data gathering, analysis, results section. SHL, overall organization, integration, discussion and conclusions. 13. 14. 15. ORCID iD Stephen H Linder 16. https://orcid.org/0000-0003-1316-4496 17. REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. McGinnis JM, Stuckhardt L, Saunders R, Smith M. Best Care at Lower Cost: The Path to Continuously Learning Health Care in America. Washington, DC: National Academies Press; 2013. Berwick DM, Hackbarth AD. Eliminating waste in us health care. JAMA. 2012;307:1513–1516. The Henry J. Kaiser Family Foundation. Total Medicaid spending. https://www. kff.org/medicaid/state-indicator/total-medicaid-spending/. Updated December 2016. Accessed March 19, 2017. Montalbano A, Rodean J, Kangas J, Lee B, Hall M. Urgent care and emergency department visits in the pediatric Medicaid population. Pediatrics. 2016;137:e20153100. Weinick RM, Betancourt RM. No Appointment Needed: The Resurgence of Urgent Care Centers in the United States. Oakland, CA: California Healthcare Foundation; 2007. Health Policy Commission. Commonwealth of Massachusetts Health Policy Commission 2015 Cost Trends Report. 2016, https://www.mass.gov/service-details/ annual-cost-trends-report. Accessed November 10, 2018. The Henry J. Kaiser Family Foundation. Hospital emergency room visits per 1,000 population by ownership type. https://www.kff.org/other/state-indicator/ emergency-room-visits-by-ownership/?currentTimeframe=0&sortModel=%7B %22colId%22:%22Location%22,%22sort%22:%22asc%22%7D. Updated 2015. Garson A Jr, Green DM, Rodriguez L, Beech R, Nye C. A new corps of trained Grand-Aides has the potential to extend reach of primary care workforce and save money. Health Aff (Millwood). 2012;31:1016–1021. doi.10.1377/hlthaff .2011.0859. Brown MT, Bussell JK. Medication adherence: WHO cares? Mayo Clin Proc. 2011;86:304–314. The Medicare Payment Advisory Commission and the Medicaid and CHIP Payment Access Commission, eds. Data Book: Beneficiaries Dually Eligible for Medicare and Medicaid. Washington, DC: Medicaid and CHIP Payment and Access Commission, January 2017. https://www.macpac.gov/publication/data-book-bene ficiaries-dually-eligible-for-medicare-and-medicaid-3/. Accessed April 25, 2017. Garson A. Grand-Aides and health policy: reducing readmissions cost-effectively. Health Affairs Blog, 2014. http://healthaffairs.org/blog/2014/10/29/grandaides-and-health-policy-reducing-readmissions-cost-effectively. Accessed November 10, 2018. Segal M. Dual eligible beneficiaries and potentially avoidable hospitalizations. CMS Policy Insight Brief, 2011. https://www.cms.gov/Research-Statistics-Data- 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. and-Systems/Statistics-Trends-and-Reports/ Insight-Briefs/Downloads/ PAHInsightBrief.pdf. Accessed April 25, 2017. Gosden T, Forland F, Kristiansen IS, et al. Capitation, salary, fee-for-service and mixed systems of payment: effects on the behaviour of primary care physicians. Cochrane Database Syst Rev. 2000;3:CD002215. Miller TD, Askew JW. The first decade of appropriate use criteria: is the glass half empty or half full? J Am Coll Cardiol. 2015;65:774–776. Aldridge MD, Kelley AS. The myth regarding the high cost of end-of-life care. Am J Public Health. 2015;105:2411–2415. Nicholas LH, Langa KM, Lwashyna TJ, Weir DR. Regional variation in the association between advance directives and end-of-life Medicare expenditures. JAMA. 2011;306:1447–1453. Klingler C, in der Schmitten J, Marckmann G. Does facilitated advance care planning reduce the costs of care near the end of life? systematic review and ethical considerations. Palliat Med. 2016;30:423–433. Jones AL, Moss AJ, Harris-Kojetin LD. Use of advance directives in long-term care populations. NCHS Data Brief. 2011;54:1–8. Wissow LS, Belote A, Kramer W, Compton-Phillips A, Kritzler R, Weiner JP. Promoting advance directives among elderly primary care patients. J Gen Intern Med. 2004;19:944–951. MACPAC. MACStats: Medicaid and CHIP data book, 2016. https://www. macpac.gov/publication/macstats-medicaid-and-chip-data-book-2/. Accessed April 25, 2017. Gindi R, Jones L. Reasons for emergency room use among U.S. children: National Health Interview Survey, 2012. NCHS Data Brief No. 160, 2014. https://www.cdc.gov/nchs/data/databriefs/db160.htm. Accessed April 20, 2017. Cubanski J, Swoope C, Boccuti C, et al. A primer on Medicare. KFF, 2015. ht t p://fi les.k ff.org/at tachment /repor t-a-primer-on-medica re-key-facts -about-the-medicare-program-and-the-people-it-covers. Accessed May, 20, 2017. Roebuck MC, Liberman JN, Gemmill-Toyama M, Brennan TA. Medication adherence leads to lower health care use and costs despite increased drug spending. Health Aff (Millwood). 2011;30:91–99. Centers for Medicare Medicaid Services. Chronic conditions among Medicare beneficiaries (Chartbook), 2012 ed. http://www.cms.gov/Research-Statistics-Data-and-Systems/Statistics-Trends-and-Reports/Chronic-Conditions/ Downloads/2012Chartbook.pdf. Updated 2012. Accessed April 2, 2017. Fitch K, Pelizzari P, Pyenson B. The high cost of heart failure for the Medicare population: an actuarial cost analysis. http://us.milliman.com/uploadedFiles/ insight/2015/heart-failure-cost-medicare-analysis.pdf. Updated 2015. Accessed April 19, 2017. Garson A. Grand-aides: leveraging the workforce for more effective and less expensive care. In: Phillips R, ed. America’s Healthcare Transformation: Strategies and Innovations. New Brunswick, NJ: Rutgers University Press; 2016: 178–198. The Henry J. Kaiser Family Foundation. Distribution of fee-for-service Medicaid spending on acute care. https://www.kff.org/medicaid/state-indicator/ spending-on-acute-care/. Updated 2016. Accessed April 20, 2017. American Diabetes Association. Standards of medical care in diabetes-2017: summary of revisions. Diabetes Care. 2017;40:S4–S5. Adair C, Swart E, Seymour R, Patt J, Karunakar MA. Clinical practice guidelines decrease unnecessary echocardiograms before hip fracture surgery. J Bone Joint Surg Am. 2017;99:676–680.