Winning the military utilization analytics battles
Originally published November 2020 on https://broadcastedge.blogspot.com/
Utilization of healthcare services has become a central story point when examining healthcare delivery. Utilization is important for cost containment, has potential to benefit quality of care through best practice alignment and provides insight into substandard or deficient current operations.
Military healthcare is largely held to cost containment pressures, and these pressures seek utilization innovation to improve and assure delivery. Military healthcare is largely a government funded operation in most countries, and utilization strategies are great opportunities everywhere.
There are three rules that could be applied to the large and complex utilization arena. These three rules should be applied with the tandem concept of utilization improvement for the military alongside healthcare access delivery improvements for all countries.
1. Everyone is a genius when playing in a united center.
Large quantities of utilization data, with variance to tools, methodologies and interpretations, are difficult to consume with focus. It makes more sense to standardize some basics, streamline expectations and sharpen the process.
How can we work toward utilization focus as it remains abstract, with everyone contributing?
How can we find opportunity and put forth our unique value?
There are many ways to contribute to the united arena of utilization, placing military healthcare at center:
* Emphasize definition and training. Utilization management and utilization metrics should be understood by those in healthcare operations, even as its application evolves.
* Standardize metrics. Utilization metrics, usually tracked through ED visits, inpatient visits, outpatient and primary care visits and ED encounters, should follow analytical consistency. Rather than steer calculation and methodology based on data available, require the data necessary for formulaic processes (1).
* Get serious about overcoming the limitations to what utilization metrics tell us. Limitations to utilization metrics can be addressed with consensus: the construction of strong evaluation models, obtaining complete data, estimation of times, accounting for external care such as housing and criminal justice, accounting for geographical and regional variance, minimizing selection bias, accounting for natural progression of life course estimates, inclusion of qualitative and narrative data, and adopting a framework that integrates cost with functional status outcomes would align the utilization analytics. Additionally, clarification of when we want to see increases and decreases of utilization is necessary. This should be achieved with structure.
* Appreciate the military center's contributions. The US VA system (2, 3) and MHS (4) continue to provide utilization data in annual reporting, for transparency and to better understand service delivery. VA studies have noted external determinants of utilization, such as employment, alternative insurance, economics, hosing non-VA physician access and other components (5). Additionally, drug utilization data has significantly improved MHS spending. The utilization of compounded drugs, .5% of all 2015 TRICARE prescriptions, taking up 20% of all 2015 TRICARE costs, was examined emphatically. In 2018, compounded drug use was reduced to account for just .14% of all 2018 costs (6). Drug utilization data continues to present opportunity for evidence based practice analyses, such as a recent clozapine study (7) of which there was no benchmark for guideline adherence. This work also holds potential to identify gaps and shift improved pharmaceutical contracts. Military work also examines use of provider by specialty. Dutch military planning (8) can be improved with noted differences in genera practioner versus specialist care based on mental health symptom severity. Barriers to care have also been examined using utilization metric concepts. Mental health service utilization and barrier data for specific populations, such as the OEF/OIF/OND deployed groups, have potential to enhance interventions (9). Improvements to budgeting and planning can follow diagnoses utilization data. With data reporting that VHA outpatient care utilization is twice as high for OEF/OIF/OND veterans with CMI versus veterans without CMI, operations can be tailored (10). Additionally, data that showcases when military population utilization is more optimal than the general population is critical. This provides general public health and medicine with positive pressure to improve, with opportunities to provide innovation to military counterparts. Studies that show Canadian Armed Forces utilization of mental health services as more optimal than general population utilization could contribute to improved overall delivery (11).
* Acknowledge the military center's efforts to sharpen Value Based Care. Assessments of Low Value Heathcare (12) continue to take shape for military units, just as VBC continues to take shape all over the world. VBC and the identification of low value care also offers great opportunity for quality improvement to care delivery, such as seen in MHS pediatric patient studies (13).
* Acknowledge the inefficiencies of piecemeal approach and advocate for real-time, ongoing data. It is likely that advanced militaries analyze similar utilization metrics throughout the world, and equally as likely that other advanced militaries also take piecemeal research approaches. There is no ongoing, real-time utilization data of service personnel, anywhere. Published work reliant on 1999-2006 data, such as a recent study of Medicaid expansion from secondary 1999-2006 (14), only furthers the case for real-time data improvements. Real-time ongoing surveillance for utilization purposes is feasible, with ethics and privacy compliance.
* Build expansion of service utilization research around current laws. These may be different by country, and countries can lead. The US can lead by creating structure to the studies that examine impact of the Affordable Care Act, as well as expansion of Medicaid and Medicare services. Strategic structure to these insights make advancements more efficient. It is important to know that veterans with spinal cord injuries and disorders (SCI/D) sought more care and VA admissions were higher after ACA (15). It would also be of benefit, however, to coordinate utilization research on ACA with pertinent laws already in place (16). Build around the requirements already in place and mandate funding for the entirety of utilization.
* Outline needed improvements to expertise insight. Notably, the Dartmouth Atlas has opportunities to expand upon external determinants, consumer satisfaction and patient health (17).
*Outline the necessary definitions and formulas to determine accuracy and validity of utilization data, with military healthcare as the initial focus. Reproducability would be a start.
2. We can and should avoid drowning in opportunities.
Utilization in healthcare remains abstract, the analysts and labor supply do not work as one system, and healthcare remains fragmented.
At the end of the day, what matters most to the average individual is that the best possible care was provided to them, in a timely manner, from a trusted entity.
Utilization efforts can help to assure this by encouraging cost effective care, appropriating funding efficiently, securing visible quality and offering tangible process improvement. Healthcare is at the beachfront of the impending Big Data wave, and healthcare must formalize a structured approached to utilization science. We can and should avoid drowning in opportunities blurred by abstract, evolving concepts. We should strive for meaningful utilization metrics.
Here are some efforts to focus on:
* Clarify utilization increases and decreases, and the significance of the numbers. Ideals in utilization increases and decreases are not case by case, they can and should be standardized. In example, utilization increases for primary healthcare and mental healthcare are a goal when examining barriers that affect Australian military personnel (18). A defined increase goal should be applicable and scaled across borders. In another example, prescription use and lack of prescription utilization should be clearer. What is the ideal utilization of prescriptions for elderly veterans, and what is the alternative care provided for those with fewer prescriptions (19)?
* Define Value Based Care (VBC) and the utilization components to it in a globally cohesive way. Many countries have begun VBP considerations (20, 21), technological considerations (22) and other advances. This is a battle we need to tackle before taking on others, and one we must structure ourselves around before the Big Data wave grows.
* Standardize well adult criterion, with documentation of benchmarks. Benchmarks not only help utilization improvements for patients; benchmarks helps with appropriate labor supply (23). Military units and veteran care should benefit from proactive planning with improved well adult and physical benchmarks.
* Identify opportunities for evidence-based guidance embedded in the utilization research, without delay. The healthcare of military personnel with mTBI (24) is a story that utilization metrics provide descriptors for, and better long-term planning, early interventions and funding for this population should be an expected follow up.
* Improve prior authorization evidence. What evidence is there for prior authorization and outcomes? For prior authorization and consumer satisfaction? For prior authorization and patient safety?
* Detail consistent utilization practices across countries. Successful drug utilization strategies such as step therapy, quantity limits, prior authorization, negoatied pricing, gener drug conracs, channel management, constant conversation about pipeline drugs and novel drug mechanisms, new indication insight and conducting cost-effectivenss evaluations are not unique to one population or one military. Align and improve.
* Detail electronic record and data capabilities for utilization in healthcare, with MHS Genesis compatibility evaluation as a start (6).
* Support the decision-making on weighting, with ongoing, continuous third party review of evidence. For MHS, there are metrics where Relative Weighted Product (RWP) replace the Relative Value Units (RVU) (25, 26). The whole of resource-based relative value scale (RBRVS) should be clear for an average consumer, with a standard entity accountable to continuous evaluation. Ideally, this entity would cross borders, examining weighted metrics across the globe. When questions arise, such as the MHS use of IBM Watson and weighting for DoD (4), where it is unclear to us how the pharmaceutical utilization benchmark is determined and if outcomes and external experiences are weighted, ongoing evaluation could be ready response.
* Standardize the use of diagnostic codes. The use of ICD-9 code and HCC designation as indicator of disease burden, without analysis for clinician assignment or appropriateness, could be challenged (19) and should be subject to stronger standardization in utilization science.
* Address gaps in utilization research. Utilization data is often centered on specific populations, even though care structures are remarkably similar in many delivery systems (reliance on VA studies and VA registries as opposed to all veteran and military reach). Additionally, utilization data may be drawn from a single item metric on a tool (9). Gaps that are addressed, such as studies seeking to clarify utilization between VA and non-VA users (27), should be supported with stronger ongoing data sets.
* Account for external determinants such as transportation (28), housing stability (29), and caregiver support (30, 31, 32). Too, external determinant studies need outcome data to accompany the metrics. Peer support should also account for healthcare literacy (33). Without comprehensive approach, best practice is lost. In example, identification of the pharmaceutical mail delivery system for veteran and military service persons, without addressing homelessness, is misleading.
* Clarify utilization management science, including the use of utilization management nursing, and how it fits into greater healthcare utilization analytics (34).
* Provide pragmatic approach to utilization data. Utilization of community services, policies and reimbursement should be examined pragmatically. In example, environmental regulations on base and utilization of respiratory healthcare, military nutritional purchases and blood pressure utilization, or increased stress reducers at the worksite and veteran mental health could be examined. This is applicable for militaries throughout the world and the data should be shared.
* Act on the data around clinician supply, military health and utilization. This is another area that should require ongoing analytics, particularly when US providers will not see persons with military coverage (35) or when UK clinicians report a lack of knowledge regarding veteran needs (36). It is clear that despite differences in mandates, including with Covenant policies, ill-equipped providers contribute to utilization variance. Lack of provider awareness on service person needs, as well as insufficient payment structures, are solvable. Make use of the data provided, globally.
* Structure the endgame for health equity and utilization (37). What are goals and timelines for differences to gender, race, disability and other populations subject to utilization variance? What are the additional support structures for health equity with utilization differences? How are we accounting for military populations and variance within the service person populations? In one example, UK service persons do not have equitable amputation trajectories and outcomes (38). Where is the utilization guidance, measurement and formal accountability?
* Refine utilization research and publications so that systematic reviews are required and acknowledged. What is already known, and what risk adjustments have been made as we examine future projections and policies for military healthcare utilization (39).
* Request that unwarranted care expertise up their game for military inclusion. Dartmouth, Wennberg (40) and other unwarranted care expertise (41) are capable of improvements to patient health status incorporation (42) as well.
* Request that country-specific healthcare quality agencies incorporate military utilization work into large data sets. For the US, AHRQ would be a great start (43).
* Consider how we can incorporate consumer satisfaction and healthcare outcomes into visual data that tells the story of utilization (44).
* Fix the lag time in national healthcare data availability. For the US, Medicaid and Medicare lag times need to improve, billing systems have become quite advanced, and there is no excuse (45).
* Identify areas where utilization research continues to be an excellent opportunity for aligned, EBP military healthcare intervention worldwide: pain management is one area, complimentary and alternative care another.
* Reiterate identified areas for global military utilization metric coordination: provider access, delivery times, Dartmouth and expert expansion, consumer satisfaction and health outcome incorporation, evidence based guidance development and adherence, definition and clarity to utilization management, and telehealth opportunities. In example, reiterate how NATO can initiate TMED utilization metrics (46) and COMED can work with unwarranted care alongside uninterrupted care metrics (47).
* Identify data extraction tools for real-time utilization analysis. Extraction by military keyword has been proven effective, yet time consuming, for the UK (48). The development and validation of a Canadian Armed Forces tool to find neck and back pain cases provided an accepted algorithm for diagnosis identification (49). Structure data tools to reflect best research practice, now. The development of data aggregation algorithms should be accompanied by acceptable sensitivity and specificity criteria. Structure these tools to advance beyond case identification, in aspiration of real-time data for case aggregation. Structure these tools for utilization standardization adherence, and for identification of performance improvement opportunities in EHR and documentation.
* Build a portfolio of cross-walks, tools, surveys, methodologies and other components for utilization analyses.
* Detail ways in which utilization goals should seek to deliver care to the patient, as opposed to measures barriers of the patient seeking care delivery where medicine dictates.
* Make utilization metrics meaningful. Healthcare administrators, policymakers, adminstrators and the public need to understand and value utilization. Ongoing efforts for real time and predicted use of military coverage must account for reactionary changes to civilian and private coverage. The ongoing, interdependent relationships between civilian and military healthcare delivery is true everywhere. The degrees to cause, effect and consequences may be a consumed by big data conundrums, yet meaningful data aggregation should be implemented with consensus.
* Acknowledge that at the end of the day, what matter is that the best care was delivered, in an acceptable time, from a trusted entity. Data analytics is a partner for this aspiration, not the time-consuming only focus, so let's get it efficient and right.
3. We should win the analytics battles we are in before taking on new battles.
Utilization improvements for military healthcare require improvements to the global healthcare delivery system. Advanced militaries support host country health needs and support structure to host country militaries. It is logical that utilization improvements to healthcare be accomplished in tandem with global healthcare improvement. This is how we win the analytics battles we are already in.
We should:
* Identify healthcare reimbursement systems, relevant to military inclusion, by country. European work has been a great start (50).
* Support universal healthcare infrastructure, acknowledging that 50% of the world cannot access essential health services (51).
* Support universal healthcare infrastructure by requesting sincerity to financial protection (52, 53) as well as labor pool essential service structure.
* Support universal healthcare infrastructure through accountability to those that would politicize and those who choose to insert "free care" propaganda, stimulate discord and decry healthcare access as priority.
* Overcome the data gaps and request financial accounting for what it will take to get countries to submit solid data for OECD and other expert analysis (54 a-f).
* Support Latin America with drug utilization work, ethics structure efforts (55) and value based management coordination (56, 57) to advance utilization metrics structure.
* Support Asia with pricing and negotiation efforts where utilization metrics could assist (58, 59), support Asia with improvement to claims data validity (60), support home visit utilization considerations (61, 62), and support stronger data availability for policy analyses (63). Additionally, we should support health and economic efforts to address serious differences in health costs between Asian countries (64).
* Support cross-regional work, such as Asia and Eastern European (Albania, Armenia, Georgia, Kazakhstan, Kirov Province in Russia, and Tajikistan) efforts in clinical care quality. Assessments could incorporate utilization components and this should be funded. Reimbursements for funding could be accompanied by metric structure around diagnostics, allied health referrals, testing, treatment and utilization analyses of these components (65).
* Support African work around utilization insight with primary care (66), for MURIA medication utilization collaboration across borders (67) and for Sub-Saharn work with a paucity of health data(68). Support should address reliance on maternal-child health data, exclusion of transient populations and exclusion of field work, all areas where stronger data, expansion of definition and surveillance support could accompany delivery funding.
* Support Gulf Cooperation Council countries in health economics (69), utilization management opportunities (70), migrant population complexities (71) and technological advancements in healthcare (72). Work with utilization alongside philosophical and and delivery priorities for GCC (73).
* Request utilization metric standardization with formal funding. In example, when the World Bank supports survey data, payer data could accompany dependent variables. There should be consensus and best practice assistance when supporting utilization research, such as support for Bosnia and Herzegovina utilization studies (74). The selection of five variables in this study, FP, hospitalization, urgent care, dentistry and private may or may not be what other countries have examined. Funders like the World Bank are more than capable of structuring themselves to assist in utilization metric structure for supported ministries of health, with comparative pictures in mind.
* Figure better ways to address utilization in spending from a global perspective (75) . This includes health coverage expansion in wealthy nations (76, 77, 78). Figure better ways to account for normal aging, and external retirement determinants (79).
* Shape utilization efforts with SHARE data from Europe (80, 81), and shape utilization efforts when SHARE is used as a template in other regions (82).
Military healthcare can and is improved with utilization science. Unfortunately, the science continues to evolve abstractly.
Utilization measures should be supported, meaningful and advanced. Standards can be aligned across militaries and across countries. Addressing global healthcare improvements will also improve the military center in our united court.
At the end of the day, what matters most to the average individual is that the best possible care was provided to them, in a timely manner, from a trusted entity.
Utilization science can sharpen us, and we can sharpen utilization science.
Refs for traveling the court:
1. https://www.chcs.org/media/CCIL-cost-and-utilization-paper_032317.pdf
2. https://www.va.gov/vetdata/Utilization.asp
5. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6232405/
6. https://www.jmcp.org/doi/pdf/10.18553/jmcp.2019.25.11.1195
8. https://www.sciencedirect.com/science/article/abs/pii/S0022395616300875
9. https://pubmed.ncbi.nlm.nih.gov/26237497/
10. https://link.springer.com/article/10.1007/s11606-018-4479-6
11. https://link.springer.com/article/10.1007/s11606-019-05529-y
12. https://pubmed.ncbi.nlm.nih.gov/31381388/
13. https://bmchealthservres.biomedcentral.com/articles/10.1186/s12913-020-05640-5
14. https://pubmed.ncbi.nlm.nih.gov/32205790/
15. https://pubmed.ncbi.nlm.nih.gov/33085584/
16. https://www.everycrsreport.com/files/20130422_R41198_0dbca027e05850c5be305d7adadc1561e579417f.pdf
17. https://www.dartmouthatlas.org/faq/#research-methods-faq
18. https://journals.sagepub.com/doi/abs/10.1177/0095327X19852652
19. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6235816/
20. https://catalyst.nejm.org/doi/full/10.1056/CAT.17.0558
21. https://bmchealthservres.biomedcentral.com/articles/10.1186/s12913-017-2104-8
22. https://link.springer.com/chapter/10.1007/978-3-030-24318-0_64
24. https://www.liebertpub.com/doi/abs/10.1089/neu.2016.4910?journalCode=neu
25. https://apps.dtic.mil/dtic/tr/fulltext/u2/1004034.pdff
26. https://scholar.afit.edu/cgi/viewcontent.cgi?article=2851&context=etd
27. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6352911/
28. https://pubmed.ncbi.nlm.nih.gov/27115074/
29. https://www.ncbi.nlm.nih.gov/pmc /articles/PMC5359170/
30. https://academic.oup.com/hsw/article-abstract/42/2/e111/3089953?redirectedFrom=fulltext
31. https://journals.sagepub.com/doi/pdf/10.1177/1077558717697015
32. https://pediatrics.aappublications.org/content/early/2014/03/11/peds.2013-1630
33. https://ps.psychiatryonline.org/doi/pdf/10.1176/appi.ps.201600290
34. https://www.ncbi.nlm.nih.gov/books/NBK234996/
35. https://journals.sagepub.com/doi/abs/10.1177/1077558720942700
36. https://journals.sagepub.com/doi/full/10.1177/1757913918785650
37. https://www.ncbi.nlm.nih.gov/books/NBK500097/
38. https://pubmed.ncbi.nlm.nih.gov/26958343/
39. https://pubmed.ncbi.nlm.nih.gov/31381387/
40. https://wennbergcollaborative.org/about/
41. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5650259/
42. https://pubmed.ncbi.nlm.nih.gov/28407769/
43. https://www.ahrq.gov/data/hcup/index.html
44. https://www.pdhealth.mil/utilization_outpatient#slideshow-1
45. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6235813/
46. https://pubmed.ncbi.nlm.nih.gov/19035805/
47. https://www.nato.int/cps/en/natohq/topics_175736.htm
48. https://www.mdpi.com/2227-9032/8/1/1
49. https://jmvfh.utpjournals.press/doi/full/10.3138/jmvfh.2018-0039
50. https://military-medicine.com/media/article/3673/attachment-1548945435.pdf
51. https://www.who.int/news-room/fact-sheets/detail/universal-health-coverage-(uhc)
52. https://www.who.int/health-topics/health-financing#tab=tab_1
53. https://documents1.worldbank.org/curated/en/640121513095868125/pdf/122029-WP-REVISED-PUBLIC.pdf
54. a https://www.oecd.org/els/health-systems/Table-of-Content-Metadata-OECD-Health-Statistics-2020.pdf
54. b https://stats.oecd.org/wbos/fileview2.aspx?IDFile=54b82b5e-5b99-48cc-805c-7b2baf0176b5
54. c https://stats.oecd.org/wbos/fileview2.aspx?IDFile=1aac08d9-99d5-4b52-9aa8-bbc8da33454b
54. d https://stats.oecd.org/wbos/fileview2.aspx?IDFile=d9322291-5553-401b-92e3-34d3732eda6c
54. e. https://stats.oecd.org/wbos/fileview2.aspx?IDFile=587d7574-6ed6-4408-9c32-481d322936e6
https://stats.oecd.org/wbos/fileview2.aspx?IDFile=4cbd825d-6409-4eed-8483-2d5be8524797
55. https://www.sciencedirect.com/science/article/pii/S2212109918302334
56. https://www.jacc.org/doi/full/10.1016/j.jacc.2017.06.050
57. https://www.sciencedirect.com/science/article/pii/S2212109918303625
58. https://www.tandfonline.com/doi/full/10.1080/20016689.2019.1601060
59. https://ueaeprints.uea.ac.uk/id/eprint/74778/1/Accepted_Manuscript.pdf
60. https://onlinelibrary.wiley.com/doi/abs/10.1002/pds.4616
61. https://www.nejm.org/doi/pdf/10.1056/NEJMoa1911965
62. https://www.sciencedirect.com/science/article/pii/S1319016417302037
63. https://onlinelibrary.wiley.com/doi/pdf/10.1002/hpm.2851
64. https://www.tandfonline.com/doi/abs/10.1080/14737167.2019.1650643
65. https://www.ghspjournal.org/content/ghsp/5/3/412.full.pdf
66. https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0013909
67. https://www.tandfonline.com/doi/pdf/10.1586/14737167.2015.1065735 https://muria.mandela.ac.za/
68. https://academic.oup.com/heapol/article/29/7/921/557860
69. https://www.tandfonline.com/doi/full/10.1080/14737167.2018.1479254
70. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5650259/
71. https://spiral.imperial.ac.uk/bitstream/10044/1/74957/1/AlGhafri-A-2019-PhD-Thesis.pdf
73. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5650259/
74. https://journals.sagepub.com/doi/full/10.1177/0020731418762717
77. https://link.springer.com/article/10.1007/s11606-019-05529-y78. https://jamanetwork.com/journals/jama/article-abstract/2674671
79. https://link.springer.com/article/10.1007/s00148-017-0664-x
80. https://infogen.webs.uvigo.es/WPB/WP1801.pdf
81. https://www.share-project.org/data-access.html
82. https://www.sciencedirect.com/science/article/pii/S1319016417302037