In this issue, Rosen and colleagues report on an exquisitely well-designed and well-implemented research effort to examine a strategy to improve outcomes for veterans with posttraumatic stress disorder (PTSD) after discharge from residential treatment. However, the aim of the study, although well intentioned, seems naïve given current realities, and it misses the mark in regard to lessons learned over the past 30 years about treatment outcomes for veterans with PTSD and the challenges that treatment presents.

If active residential treatment is itself ineffective, as past and current research suggests, what is the value in examining continuity-of-care strategies after residential treatment? Perhaps part of this study’s value is as a cautionary note to others. However, the history of the Department of Veterans Affairs (VA) treatment efforts for veterans with PTSD suggests that future generations are unlikely to heed the available lessons.

In the late 1990s, residential and inpatient programs operated by the VA were shut down all over the United States for lack of demonstrated efficacy. One wonders why such expensive programs have been restarted when there is no evidence to support them. It seems that the VA continually attempts to reinvent the wheel as each new generation of politicians and administrators fails to learn what others before them have already learned.

It also is time for VA leaders and researchers to reflect on why veterans with PTSD are not benefiting from treatment, especially when civilians with PTSD do. Although politically inconvenient to say, the VA's current disability policies, which reinforce illness roles and provide disincentives to work and to recover, likely have much to do with this disparity. About 45% of Iraq and Afghanistan veterans have already applied for VA disability benefits, a remarkably high rate compared with 11% and 16%, respectively, for veterans of World War II and Vietnam. This represents a historically unprecedented and dramatic rise in disability applications—and for a war with a different pattern of combat exposure and comparatively lower typical battle casualties.

Rosen and colleagues indicate that almost 70% of the patients in their study were already receiving VA cash disability payments, and the other 30% may have already applied or will apply in the future. Thus virtually all patients have a powerful secondary-gain incentive to remain symptomatic.

For several years the VA has required completion every 90 days of a PTSD symptom checklist by all veterans with a diagnosis of PTSD in order to monitor symptom trajectories. Have these national data on treatment effectiveness ever been published or disseminated? What do they indicate?

The VA now faces a dilemma of logic: Either VA evidence-based treatments and programs work as expected and most veterans will recover, in which case they no longer require disability payments, or these expensive treatments do not work, in which case it is difficult to justify the expanding VA budgets for inert mental health programs. Logically, it cannot be both.