Lessons from the Rise—and Fall?—of VA Healthcare

Around 2005, freelance journalist Phillip Longman was asked by Fortune magazine to write a feature article about a healthcare organization that was doing everything right. After extensive investigation, he wrote a story about the remarkable transformation of the Veterans Health Administration (VHA), but Fortune found this answer unsatisfactory. So he wrote a book¹ that cited extensive evidence claiming that VHA healthcare was second to none. In this issue of JGIM, O'Hanlon et al. present an updated review of the evidence on VHA’s quality of care.² Given the recent controversies surrounding the VHA, some might find these authors’ review difficult to believe. It is certainly true that some of the articles cited could be criticized for potential favorable selection (such as over-sampling academic-affiliated VHA facilities) and that the measurement approaches inside and outside the VHA differed. However, most of evidence the authors cite comes from national or representative VHA samples and present a very strong scientific case to support their main conclusion—that at least for common performance metrics, the available evidence suggests that the quality and safety measures of the VHA were truly as good as or better than those in the private sector, even top-rated managed care organizations.²

More remarkably, the VHA had been widely viewed in the 1970s and 1980s as an inefficient safety-net provider at best. Its turnaround in the 1990s under VHA Undersecretary Ken Kizer was one of the most dramatic in healthcare history. How did this happen? Is the controversy over wait times evidence that the VHA has since reverted to old ways? If so, how can the VHA find its way again?

Loose–Tight Leadership/Management

I believe that three organizational changes were particularly important in the VHA’s transformation: 1) VHA facilities were divided into manageable (just more than 20) Veterans Integrated Service Networks (VISNs) that had reasonably similar patient populations and healthcare delivery challenges, making comparison of their performance much easier; 2) both resources and decision-making authority were principally distributed to these VISNs, thereby delegating day-to-day decision-making power to those with the most information about their local resources and environments; and 3) each VISN leader was held strictly accountable for success, with success monitored by carefully chosen measures of quality and efficiency. In other words, the VHA Central Office meddled little in day-to-day operations (loose = minimal micromanagement), but held VISNs strictly accountable for central values such as quality and efficiency (tight = strict accountability for performance). Moreover, the VHA Central Office evaluation system was created such that a given VISN’s success was not predicated on another’s failure. This form of managed competition between VISNs created a cooperative, positive culture of collaborative continuous improvement. Further, when leaders in the field brought strong arguments to the national leaders, the Central Office generally listened and tried to expediently refine the system to eliminate perverse incentives. For example, research is a congressionally mandated VA mission, but since there were initially no VHA performance measures for research productivity, some VISNs ignored, or even stole resources from, their research mission. As soon as this became known, VHA leadership reportedly demoted an extreme violator and instituted new performance measures and incentives to hold VISN leadership more accountable for fostering high-quality, efficient research operations. I believe that parts of the VHA's success were exaggerated (much of what is most important in healthcare is hard to measure accurately), but its transition and accomplishments during this period were inarguably impressive.²

There Are No Perfect Systems

Managed competition has great merit, but it is impossible to build a faultless organizational structure out of the crooked timber of humanity.³ It did not take long before both internal and external political and social forces began to push for increased central planning and micromanagement (such as centralizing IT management and centrally mandated report cards on individual providers).⁴ The push for adopting more, often poorly conceived, performance measures was perhaps the worst mistake, but this was a problem that occurred both inside and outside the VHA.^{5 , 6}

The VA Access Scandal of 2014

So how does the recent wait-time scandal fit into the VHA performance story? What are the root causes of the wait times, and how can we improve veterans’ access to needed services? It was common knowledge that VHA was struggling to meet the increasing demand for services as more and more Operation Enduring Freedom (OEF) and Operation Iraqi Freedom (OIF) veterans were returning to civilian life. Wait lists began to balloon, and miracle-seeking committees were formed. One such committee recommended a new intensified performance measure for wait times—minimize the proportion of patients who had to wait more than 2 weeks for a visit. Reportedly, the committee members fully realized that this was not necessarily a good performance goal. Optimal wait list management is complex and nuanced. After all, a veteran with PTSD and suicidal ideation should be seen immediately and followed closely, while a veteran who feels well and simply wants to establish care can generally wait for months. Encouraging a focus on average wait time can focus energy and resources away from the highest priority—the urgent medical needs of the severely ill with time-sensitive care needs. Still, this committee felt that something had to be done, and that this wait-time measure would just be a “stretch goal” that should not be taken too seriously.

Faced with a performance measure that many felt was unreasonable and unattainable, some of those in the field began cheating. Multiple reports suggested that cheating might be widespread. Why these reports were not acted upon is unclear, but I can say that the VHA had a clear short-term incentive to not crack down on the cheating, since it was, at least in my opinion, impossible for many, if not most, facilities to meet the 2-week metric in the face of a shrinking per capita VHA budget and greatly increasing demand for VHA services (see below). (Note: I am in no way excusing those who cheated, especially those who put veterans at risk; I am merely noting the incentives and circumstances surrounding this scandal.)

What happened next is well documented. A retiring physician at the Phoenix VA turned whistleblower, making both true and false accusations. Although the initial accusations of how many people died at the Phoenix VA due to long wait times was discredited, many heart-wrenching stories of truly terrible mismanagement of severely ill veterans have since been well documented nationwide. However, an important lesson is often missed. Many of these appalling cases of delayed care and poor care management, rather than indicating too little accountability for average wait times, may actually support the opposite. Putting too much pressure on reducing average wait times can make matters worse, if it decreases nuanced management decisions, such as allowing the many with non-urgent needs to wait longer, so that the neediest patients receive the timely care and intensive follow-up that they require. Making tough and nuanced local decisions is especially important in a system that is dealing with excess demand for services.

Why the Long Wait Times?

My research focuses on quality and safety, not scheduling and efficiency, so I had not been rigorously following the VHA’s increasing wait list problem. Even after the scandal, I relied mainly on journal and medical media reports and commentaries. However, in my background reading for this commentary, I was truly shocked by a series of particulars that I had never heard mentioned in these academic and media reports. The most common theme in published reports was that VHA’s visit wait times continued to rise despite them getting much larger increases in their budget than most other federal agencies. Although this summation is not untrue, it lacks quantification. Publicly available statistics suggest that between 2008 and 2014, the VHA increased outpatient visit availability by over 25 % per capita, despite a substantial reduction in their inflation-adjusted budget per capita.^{7 , 8} The increase in VHA funding seems impressive until you consider the increase in high-need returning veterans, and that private sector per capita costs increased by more than 30 % during this period.⁹ In short, if the reported government statistics are correct, the real question is why we would think that the access problem is due to the VA system rather than due to a dramatic increase in demand for VHA services, during a time when the VHA had an inflation-adjusted decrease in its per capita funding.

What to Do Next

What the future of the VHA should be is a matter of intense interest and debate. Regardless of what people think about how well it was functioning in 2008, the VHA clearly faces major problems and difficult challenges today. The recent Commission on Care report¹⁰ offers an excellent starting point for thinking about the VHA’s future, but hard choices will likely be needed, and like any government program, what one feels is optimal will depend on one’s worldview and preferences.

Policy leaders and the public will face difficult choices in trying to find the best way to provide veterans with high-quality care at a price for which there is adequate political will. The high rate of disability associated with the OEF/OIF campaigns and the aging Vietnam veteran population will continue to put stress on a federal budget that is already struggling with increasing Medicare, Medicaid, and Social Security costs.

Policy leaders should not underestimate how hard it will be for the VHA to increase access if managing demand is off the table. Competing for good primary care providers will also prove difficult in the current sellers’ market. Good primary care providers are in great demand, salaries are soaring, and scarcity and salaries vary greatly by region. The areas in which primary care wait times are the highest in the VHA are often the same areas in which primary care wait times are highest outside the VHA. Some flexibility, whether in hiring or in contracting with private practices, may be needed. Although checks and transparency in federal contracting are certainly needed, the resource-intensive process and glacial pace of federal contracting makes a quick response to the current access crisis all but impossible.

Finally, the VHA’s difficulty in implementing the portion of the Veterans Choice Act of 2014 allowing many veterans to seek care outside the VHA surprised few in the healthcare sector. Providing health insurance coverage not only requires sophisticated systems for verifying and reimbursing for care, but needs to protect against fraud and limit inappropriate and wasteful care. Having patients seen inside and outside of a provider network can compromise both quality and efficiency, unless communication between providers in each system is facilitated. Providing better access by contracting with private sector providers has long been a VHA policy strategy for improving access, and I personally see no path to improving access to veterans that would not increase such contracting to some degree. However, how such arrangements are made will require careful thought, competent implementation, local flexibility in making decisions, and periodic refinements. A return to the VHA’s earlier lessons of the value of decentralized decision-making, tight accountability for true quality and efficiency, and respect for two-way communication between the field and central management, might result in a systematic review of the VHA 5–10 years from now reaching the same conclusions as those reached by O'Hanlon et al., but including success in both quality and access. Let us hope, as we approach the VHA’s current problems, that everyone will at least try to set partisanship aside and concentrate on our national duty to those who served our country, body and soul.

Compliance with Ethical Standards

Conflict of Interest

The author has no conflicts of interest to report, other than he is a part-time VA researcher. All opinions are his alone, and do not represent those of the University of Michigan or the US Department of Veterans Affairs.