May 1, 2012  

Curing military health care

From 2000 to 2010, the Defense Department’s health care costs rose from $17.8 billion to $43.5 billion — growing more than twice as fast as economywide medical inflation. As a share of total DoD spending, health care costs rose from 4.5 cents of every dollar spent by DoD in 2000 to 6.1 cents of every dollar spent in 2010, leading former Defense Secretary Robert Gates to warn that these obligations were “eating the Defense Department alive.”

Though knowledge of the magnitude of these costs has spread, the mechanisms at work behind them have remained somewhat obscured. To maneuver within the new constrained fiscal environment, policymakers must understand the dynamics behind this extraordinary cost growth, going beyond the convenient but often factually bankrupt conventional wisdom.

One such piece of conventional wisdom blames this monumental growth on the wars in Iraq and Afghanistan, forecasting automatic savings as the conflicts draw down. The data say differently. For example, in 2000 — before the wars — 23 percent of the total inpatient, outpatient and prescription drug costs incurred by DoD were for care received by active-duty service members. A decade and two wars later in 2010, end strength had risen 1.4 percent, but active-duty service members’ share of total health care costs had dropped to the low double digits. Approaching the issue from another angle, we can look at overseas contingency operations (OCO) funding, which — at least in theory — represents money appropriated for war-related medical needs. At its highest point in the decade — 2009 — OCO comprised a mere 4 percent of total defense health spending.

If the exigencies of war have not been the chief driver behind cost growth in military health care, then what has? To answer this question, I collected and analyzed over a decade’s worth of DoD health care data relating to budgets, beneficiary populations, benefit use and other topics. Based chiefly upon primary sources, my analysis extends from 2000 through 2010, the latest year for which reliable data were available, with earlier data included where relevant and possible. (In this piece, all years are fiscal years unless otherwise noted; dollar values are fiscal 2012 constant dollars; budget data represent DoD Total Obligational Authority [TOA]; and growth calculations are cumulative real growth.)

Through my research, I identified six key trends behind the extraordinary cost growth in defense health care. They suggest several broad conclusions and inform a series of recommendations for reform.


• Expansion of the Tricare-eligible population. In 2000, 6.8 million people were eligible for Tricare; in 2010, 9.7 million people were — and nearly 85 percent of them were not active-duty service members. This represents 43 percent real cumulative growth in the eligible population.

This growth can be attributed to four sub-trends. First, Congress granted Tricare access to new beneficiary groups, including retired Guard and reserve personnel, Medicare-eligible retirees, and the families of both. Second, Congress and DoD lengthened and expanded eligibility terms for existing programs. Third, the percentage of the active-duty force that is married rose. Lastly, overall end strength increased.

• Enhancement of the Tricare benefit. DoD and Congress added 17 new programs to Tricare between 2000 and 2010 and expanded existing coverage for all eligible individuals multiple times throughout the decade. Some expansions affected all beneficiaries; others targeted specific Tricare options. The broadest expansions chiefly came in two guises: covering new procedures and loosening plan restrictions.

• Decrease in the percentage of costs paid by beneficiaries. When Tricare was established, a cost-sharing relationship between DoD and beneficiaries was instituted to control costs and bound beneficiary behavior. However, Congress never pegged beneficiary cost shares to inflation, as originally intended. Consequently, cost shares that had not changed in face value between 1997 and 2010 — the majority of them — lost 41 percent of their real value to inflation, a cost burden picked up by DoD.

In other cases, Congress went further than inflation, decreasing cost shares or eliminating them outright.

• Increased use of the benefit. This trend has three separate dimensions. First, there was growth in overall inpatient, outpatient and prescription drug utilization rates. This means that over time, the average beneficiary was hospitalized for longer, visited his or her doctor more often, and/or filled more prescriptions each year.

Second, though military beneficiaries have historically had higher utilization rates than comparable beneficiaries in civilian plans, that gap widened during the 2000s.

Lastly, the percentage of eligible beneficiaries who elected Tricare coverage in lieu of civilian plans increased. Between 2001 and 2010, 22 percent of retirees under 65 dropped their civilian health care coverage, while the number enrolled in Tricare Prime increased by 20 percent. Though the data cannot say why this is the case — or even whether the two groups represent the same beneficiaries — a comparison of the Prime enrollment fee and public-sector premiums since 2001 suggests that relative cost may be key. Average private-sector premiums went from $2,214 in 2001 to $4,022 in 2010, while the Prime enrollment fee went from $585 to $474. Additionally, DoD has suggested that the recent economic downturn may have diminished the percentage of working-age military retirees eligible for employer benefits.

• Decrease in use of the most cost-efficient care options. A greater proportion of care shifted to the options most expensive for DoD to provide. Several factors contributed to this trend, including a decline in beneficiaries’ ability to access military facilities and the loosening of regulations governing beneficiary behavior. Inpatient, outpatient and prescription drug workloads all more than doubled over the 2000s, and in each case, the bulk of total workload shifted to the most expensive private-sector option(s).

• Increase in the cost to DoD of providing Tricare benefits. Three factors contributed to this increase. The first, medical inflation, accounts for about one-third of the growth in military medical costs over this period. Its effects could not have been avoided entirely; however, its impact was exacerbated by congressional rule changes and other factors that shifted a greater proportion of beneficiary care to the private sector.

Second, advancements in medical technology to serve both wartime and domestic needs increased DoD’s cost of doing business. DoD incurs additional costs for new technologies beyond those reflected in economywide medical inflation. DoD both buys private-sector technology off the shelf and develops new products in house. In the latter case, these costs often hit DoD several times over, as the department first pays for the research, development, test and evaluation (RDT&E) that results in the technological innovation, only to then pay again to procure the resultant product, often also expending further capital to upgrade facilities to accommodate it. In this way, the cost of technological advancement bleeds through the RDT&E, military construction and procurement accounts.

Lastly, the shift to accrual accounting for the future health care costs of current service members greatly increased DoD’s apparent costs, adding $9.7 billion to the books in its first year alone. That cost is directly attributable to Tricare for Life. Before Tricare for Life made Tricare the second payer after Medicare for dual-eligible military retirees and their dependents (originally, at no cost to the beneficiary), DoD paid for all medical care in the year it was provided. The switch to accrual accounting — in which an amount actuarially determined to cover the future health care costs of current service members is placed in a Treasury Department trust fund each year — was undertaken in an attempt to make the fiscal consequences of benefit expansions such as Tricare for Life more visible to law- and policymakers in the present.


I draw four chief conclusions from my examination of the factors behind this cost growth.

• It’s not the wars. Despite the rise in spending on battlefield medicine, military medical RDT&E and wounded warrior care, it is clear that the Iraq and Afghanistan wars and those who have been on the front lines waging them are not behind the majority of recent military medical cost growth.

The Pentagon’s medical costs had been rising long before 2000. Further, growth that is clearly attributable to the wars — like the spike in RDT&E funding — is so small relative to overall defense health spending as to contribute very little to systemwide cost expansion. Certainly, logic dictates that the wars’ impact on defense health care cost growth exceeded the few discrete areas identified here; unfortunately, the degree to which that is true cannot be determined from the publicly available data.

Though a cursory look at the facts may imply direct links between periods of steep cost growth and key military events in the global war on terror — such as the beginning of the war in Afghanistan or the first Iraq surge — the apparent synchronicity is due to a combination of coincidence and misreading of the data. For example, though the war in Afghanistan began in calendar 2001, costs related to the war do not appear in DoD budgets until fiscal 2002, as the war began Oct. 7, 2001, a week into that fiscal year. Likewise, spikes in DoD health care costs in fiscal 2001 are directly attributable to Tricare for Life and Tricare Senior Pharmacy. Additionally, though the war in Iraq began in calendar and fiscal year 2003, the spike in DoD health care costs that fiscal year is overwhelmingly due to the $9.7 billion introduction of accrual accounting, not any wartime requirements.

• It’s not the war fighter … yet. Active-duty service members are not responsible for the majority of the cost increases in defense health care over the 2000s. Even considering two long wars and an increase in overall end strength, the data show that active-duty service members were the least costly beneficiary group overall between 2000 and 2010. Service members were second-to-last in terms of per-capita costs and last in terms of per-capita utilization rates.

Tellingly, the cost of private-sector care and its share of total workload in all three care areas dwarfs DoD-provided care — and those deployed to Iraq and Afghanistan and wounded in battle are not the chief users of domestic civilian health networks.

However, future disability payments and health care costs related to veterans of the recent wars will be a substantial cost. Estimates differ, but analysts agree that these costs will increase substantially over the coming decades, as the more than 2.2 million veterans of these conflicts — a higher proportion of whom survived battlefield injuries and are seeking care from the Department of Veterans Affairs (VA) — age and decline in health. The vast majority of these costs, however, will be borne not by DoD but by VA.

• It’s Congress. The vast majority of the increase in defense health care costs during the 2000s resulted from congressional initiatives or the effects of congressional failures to act. Congress’ willingness to expand benefits began before the wars — see, for example, Tricare for Life and Tricare Senior Pharmacy, which began as congressionally mandated pilot programs in 1997. Over the decade, Congress enhanced Tricare benefits and expanded beneficiary populations; pushed an increasingly greater proportion of beneficiary care into the private sector; decreased many beneficiary out-of-pocket costs outright; and allowed cost shares that remained to lose over 41 percent of their value due to inflation, all the while blocking DoD’s attempts to stop the bleeding. Only Congress had the power to stop the erosion of Tricare’s original cost-sharing relationships and return rationality to the system. It did not.

• It’s the beneficiaries farthest from the fighting. Those who gained the most from changes to Tricare in 2000-10 — and who represent the greatest cost to DoD — are not active-duty service members or their families, though those groups were often invoked by Congress to justify its benefit expansion initiatives.

Retirees over 65, who in 2000 were eligible only for space-available care in DoD facilities, saw their benefits greatly expand in 2001 with the introduction of Tricare for Life. Though some former service members claimed before Tricare for Life that they had been promised “free health care for life” in the past by the military, no evidence to support this claim was found over the course of several investigations by DoD and the Congressional Research Service, among others. That did not stop Congress from citing that claim in its expansion efforts, however. Tricare for Life beneficiaries are the most expensive per capita to DoD and have utilization rates far exceeding the average of all other beneficiaries in all three components of care.

The beneficiary group that poses the greatest total cost to DoD is retirees under 65 and their families. This group already had Tricare access in 2000, so did not gain as much as Tricare for Life beneficiaries over the decade in question. However, their large numbers and strong cost growth have also put significant strain on the system.


Reform of the military health system is long overdue and increasingly urgent in DoD’s new fiscal environment. While Congress rejected outright the multiple cost-reduction proposals DoD presented in the later 2000s, questions of affordability have now seriously re-entered the military compensation debate for the first time perhaps since the 1970 all-volunteer force commission report.

Limited progress has been made, but much remains to be done. My recommendations are:

• Approach health care reform as part of a comprehensive overhaul of military compensation — and be willing to take risks on promising new ideas. Today’s military compensation system is a relic from a prior era. Built up piecemeal over time with no one governing rationale, the system is inflexible and lacks many features modern employees have come to expect. Service members and their families have few choices and value their benefits far less than it costs DoD to provide them. This is especially true for noncash benefits, including health care. As the military has become increasingly diverse, composed of a wider range of service members with varied family situations, preferences and needs, compensation has expanded — but it has not become more flexible. Service members are denied the opportunity to design a benefits package that makes the most sense for them and their families, a practice increasingly common in the private sector.

Choice theory literature argues that especially in less-differentiated cultures — of which the military is a prime example — the very act of choosing one’s compensation package has value in itself. DoD need not slash benefits to reduce its costs and increase morale. All that is necessary is to give service members the option to make limited rational trades in line with their families’ unique preferences and requirements.

Considering health care as part of the broad compensation package opens up a wide range of new possibilities. DoD could offer, for example, a period of stabilization while a service member’s children are in high school in exchange for that family paying higher deductibles and cost shares in Tricare. Instead of paying more out of pocket for health care, service members could agree to put their dependents on outside insurance available through their spouse’s employer, or limit themselves to only filling prescriptions at the most cost-efficient venue for DoD, etc. The realm of possibilities is vast. Several alternatives to the current compensation system have already been tested successfully in service-level pilot programs. For example, options including billet, platform or home port choice have been offered on a small scale in the Navy and valued very highly by those service members who selected them.

• Reduce defense health program exposure to economywide medical inflation. Increased reliance on the private sector for care of Tricare beneficiaries has exposed DoD to unsustainably high rates of medical inflation and limited its ability to ensure care is being managed efficiently. DoD must find a way to lessen its exposure to these out-of-control costs. Whether the answer is renegotiating the terms of contracts with civilian providers to cap allowable cost increases, opening additional military treatment facilities to handle greater workload in-house, or repealing rule changes allowing beneficiaries to see private-sector providers at the same cost they would incur at a DoD facility — or a combination of some or all of the above — the status quo cannot stand.

• Reinstate the original cost-sharing percentages in Tricare for all beneficiaries and tie them to the most accurate measure of inflation possible.

The impact of slashing cost shares goes beyond the obvious; lower out-of-pocket costs have ripple effects, especially on utilization rates. By 2010, the original Tricare cost-sharing infrastructure had been gutted by a decade and a half of inflation erosion combined with congressional slashing of certain beneficiary costs — often to zero. The majority of waived cost shares must be reinstated, and all shares whose value has been allowed to erode — by nearly half — must be adjusted to the amounts they would be worth today if they had been tied to a measure of inflation in the mid-1990s as originally designed.

Ideally, cost shares should be adjusted by a measure of inflation that accurately reflects increases in costs to some extent — the Consumer Price Index for Medical Care, for example. Any less accurate adjustment measure, like the CPI for Urban Consumers, which measures overall inflation, will only ensure that DoD again takes on an increasing share of beneficiaries’ costs in the decades to come.

Congress did allow some limited but real changes to Tricare cost shares and co-pays in the fiscal 2012 and 2013 DoD budget rounds. For example, DoD’s 2013 budget request put forth new fee increases for working-age retirees in Tricare Prime, which would bring their share of out-of-pocket costs up to about 14 percent of the cost of their care — closer to, though still short of, the 27 percent of costs beneficiaries paid when Tricare was established in the mid-1990s.

• Alter beneficiary incentive structures to change usage patterns and reduce Tricare beneficiaries’ overall utilization rates. Restoring the original cost-sharing relationships between DoD and beneficiaries will go part way toward reining in the runaway utilization rates of the 2000s. However, further reform is necessary to reduce Tricare utilization rates — far higher than those of comparable beneficiaries in the private sector — and encourage greater use of the Tricare options that DoD provides most cost-efficiently. DoD should continue to provide preventative services free to all beneficiaries — which has been proven to reduce long-term costs — but for other procedures and services, DoD must reinstall a financial disincentive to deter beneficiary overuse of the most costly care options.

This problem is most striking in prescription drugs. Conveniently, DoD has the authority to make changes to the prescription drug formulary without congressional approval, one of its few prerogatives in military health care. Beneficiaries have three options to choose from when filling their prescriptions: DoD pharmacies at military facilities, participating retail pharmacies and the Tricare mail-order pharmacy. The difference in cost to DoD between prescriptions filled via the most cost-efficient options (DoD pharmacies and mail order) and the least cost-efficient option (retail pharmacies) is significant. However, the difference between beneficiary co-pays in the three options is not. DoD has had an extremely difficult time shifting workload to mail order, despite high rates of satisfaction among those who use it. Changing the beneficiary co-pay structures so that the difference in DoD costs is proportionally reflected in the difference in beneficiary costs would bring rationality to the system and redistribute workload in a way that makes sense for all parties without decreasing beneficiary coverage.

• Aim initial reforms at the areas of greatest cost and fastest cost growth. In terms of beneficiaries, this means retirees (both those under 65 and Medicare-eligible) and beneficiaries of any age or plan receiving large amounts of care in the private sector. In terms of budget areas, this means private-sector care, the pharmacy benefit and medical research. Getting retiree costs under control is crucial to slowing overall cost growth, as is lessening DoD exposure to private-sector care and private-sector pharmacy costs. Medical RDT&E, though fast-growing, remains so small relative to other costs that it is significantly less concerning; however, DoD should consider public-private partnerships, X prizes, and/or collaboration with universities as ways to decrease these costs and inject fresh thinking into the process at the same time.

• Consider a major reorganization of the military medical system. The piecemeal approach that DoD and Congress have taken in building Tricare out from the Civilian Health and Medical Program of the Uniformed Services and expanding the system year by year since the late 1980s, combined with convoluted budgeting practices, has resulted in an inefficient and focusless system extremely difficult to both oversee and navigate. Splitting the defense health program’s two missions — battlefield medicine and patient care — into separate organizations, or setting them up as two independent arms of the same organization (various proposals exist), would create cost-saving efficiencies and enhance transparency of costs related to war fighting and costs related to non-active-duty service member beneficiary care.

Redesigning the military health system from the top down would allow for a much more rational bureaucracy that could better serve all its constituencies. If DoD is determined to provide health care to such a large population of beneficiaries increasingly distant from military service — a goal that was unthinkable to military planners 50 years ago — an independent or quasi-independent agency that can better serve these beneficiaries and institute best practices for care and cost control may be a better and more affordable fit.

Brittany Gregerson is a research associate at the Institute for Defense Analyses. She holds an M.A. in security studies and a B.S. in foreign service from Georgetown University.