An explanation of AP's analysis of VA wait time data

AP News
|
Posted: Apr 09, 2015 4:02 AM

Everyone agrees on the problem: Many patients at Veterans Affairs hospitals and clinics are waiting too long to see the doctor. But what's the best way to measure that problem?

In setting out to evaluate the performance of 940 hospitals and clinics in the VA system, The Associated Press chose to focus on the number of appointments that failed to meet the VA's own timeliness goal, which calls for patients to receive non-emergency care within 30 days.

In some ways, that number — 30 days — is arbitrary. A one-month wait for a routine annual physical is hardly a burden. For someone in physical or mental pain, 30 days could be an eternity.

The standard, though, is a significant one in the VA bureaucracy. Under a law passed in August, VA patients who have to wait longer than 30 days for an appointment are supposed to be offered a chance to switch to a private-sector doctor, at the VA's expense.

And, despite its clinical insignificance, the 30-day standard is a useful barometer for identifying VA sites that have a problem providing timely care.

Ideally, any analysis of patient waits at the VA would include a look back of at least a year, if not much longer, but here that was not possible.

The VA has made some adjustments in the way it calculates its statistics on delays, including doing away with a system of measuring delays from the time an appointment is entered into the VA's scheduling software, rather than from the date when the patient actually wanted or needed to receive care.

Those changes, made in late summer, had the effect of roughly halving the number of appointments that failed to meet the VA's 30-day timeliness standard and radically changing the average wait times reported by facilities. For that reason, the data the VA releases now cannot be compared directly to numbers it generated in the spring.

The AP analysis includes a look at system-wide VA data from Aug. 1 to Feb. 28, and statistics on individual VA facilities from Sept. 1 to Feb. 28. A detailed breakdown of facility performance in August was not available.