Look at the System, Not the Individuals
I’ve been wanting to blog about the “VA Scandal,” but have avoided it because
- it is somewhat politicized (as tends to happen),
- the allegations and story are moving quickly, and
- the whole thing just makes me sad, so I stop writing.
Building on Monday’s post… when the news first broke about scandals at the VA (the Veterans Health Administration), a lot of the focus was on “secret waiting lists” and individuals in Phoenix “gaming the system” (or fudging the numbers):
Right away, commentators and politicians were calling for people’s heads. I thought, “Wait, this seems like it’s a really systemic problem… let’s not rush to blame people in Phoenix or elsewhere.”
The allegation, from internal whistleblowers, was that appointment waiting times were made to look shorter than they were through the use of off-the-books “secret waiting lists.”
“The VA requires its hospitals to provide care to patients in a timely manner, typically within 14 to 30 days, [Dr.] Foote said.”
One problem with goals and targets (targets set by upper management, especially from headquarters) is that they are arbitrary. Why is the target 14 days? Why not 10 or why not 20?
Update: The target was 30 days until 2011, when it was reduced to 14 days.
The goal can be skewed because it doesn’t mean “14 days of waiting from the call requesting an appointment,” it means providing an appointment “within 14 days of the date requested by the patient,” which opens things up to shenanigans.
Another problem occurs when the target isn’t achievable without cheating… this inevitably leads to cheating, “fudging the numbers” or “gaming the system.” It’s rational behavior and I don’t blame and doctors or managers who did this, given their circumstances. A VA undersecretary called the 14 day goal “unrealistic” a few years back… so the trouble that resulted isn’t the least bit surprising to me.
Look Good or Actually Be Good?
This WSJ video explains the four “scheduling schemes” or ways of cheating the system:
If it’s easier to “look good” than it is to “actually be good,” local leaders will make things look good, especially when faced with the loss of bonuses or other consequences from NOT hitting the targets. $8.8 million in bonuses were paid out to leaders at seven VA facilities accused of cheating because they hit their numbers (even though the numbers were false.
“According to Foote, the elaborate scheme in Phoenix involved shredding evidence to hide the long list of veterans waiting for appointments and care. Officials at the VA, Foote says, instructed their staff to not actually make doctor’s appointments for veterans within the computer system.
Instead, Foote says, when a veteran comes in seeking an appointment, “they enter information into the computer and do a screen capture hard copy printout. They then do not save what was put into the computer so there’s no record that you were ever here,” he said.”
Bad Systems, Not Bad Apples
Many people react by wanting to punish those who cheated the system.
I reacted differently — thinking about the leaders who created the targets, the bonuses, and the fear.
This, from the beginning, smelled like a systemic problem and it’s not solved by punishing individuals who acted quite rationally… their bosses wanted things to look good (I’m guessing), so the local leaders made things look good. The VA in Phoenix reports “fictitious” numbers to DC and everybody is happy… except for the veterans who are waiting for care.
More evidence of this “gaming” being a systemic problem are the allegations of different types of cheating at different VA sites and cities across the country. There isn’t just a “bad apple” or two in Phoenix.
Other cases and allegations keep piling up (there might be more, it’s hard to keep up – at least 26 sites are under investigation):
- Doctor says Temple (Texas) VA manipulated waiting lists
- Exclusive: VA Scandal Hits New Hospital (Albuquerque)
- VA clinic employee on leave after e-mail about manipulating appointments (Wyoming)
- Exclusive: Texas VA Run Like a ‘Crime Syndicate,’ Whistleblower Says
The fact that the same “gaming” behavior is taking place in multiple cities. We have secret off-the-books waiting lists. We have patients being intentionally guided toward appointment dates that are available instead of what was truly requested (artificially pushing back the patient’s “requested” date to what’s actually available) making the numbers look better – closing the gap between “requested” and “scheduled” to within 14 days.
The gaming taking place in MULTIPLE locations only emphasizes how systemic the problem is. We don’t have uniquely ruthless or horrible managers in these different cities. Again, they are acting rationally by cheating in the face of pressure to hit unachievable targets. I’ve seen people “game the numbers” in different businesses (including GM)… the people doing this usually realize it’s wrong, but they rationalize it in some way (including saying, “well, that’s what the boss wants”).
The Core Scandal is the Long Waiting Times
To me, the real scandal is that it takes so long for vets to get the care they need… and these delays can be deadly (for one man and at least 23, in total). 1,700 patients in Phoenix were waiting up to 115 days instead of getting treatment within 14 days (the usual target / goal). There are real human costs to this and that shouldn’t be forgotten.
This whole thing shouldn’t be a partisan issue. I was complaining about systematic VA delays back in 2007, under the George W. Bush administration. Then-candidate Barack Obama complained about it too.
Therefore, it’s not a new issue. The current leadership, including VA Secretary Eric Shinseki has known about this problem (of long wait times) for years. What has been done? Not enough, apparently. Do you blame Secretary Eric Shinseki? Do you blame President Obama? Do you blame the Republican Congress?
Again, long VA waiting times of various sorts are NOT a new problem, as I’ve blogged about before, including these posts:
- We STILL Need to Reduce Waiting Times for Veterans
- These Wounded Soldiers Need Lean To Get Their Benefits?
- Honoring Our Troops with Better Service through Lean
The Obama-Biden transition team was specifically warned NOT to trust the VA waiting time data:
Veterans Affairs officials warned the Obama-Biden transition team in the weeks after the 2008 presidential election that the department shouldn’t trust the wait times that its facilities were reporting.
“This is not only a data integrity issue in which [Veterans Health Administration] reports unreliable performance data; it affects quality of care by delaying — and potentially denying — deserving veterans timely care,” the officials wrote…
In particular, the 2008 transition report referred to a VA inspector general recommendation to test the accuracy of reported waiting times…
“Audits of outpatient scheduling and patient waiting times completed since 2005 have identified noncompliance with the policies and procedures for scheduling, inaccurate reporting of patient waiting times and errors in [electronic waiting lists],” the briefing papers state…
The briefing materials do not reveal any concerns about outright fraud in manipulating waiting times, but they make repeated references in summarizing past audits and reviews about data accuracy.
Well, I guess it has turned out to be fraud.
Choices to Make When Demand Exceeds Capacity
As a hospital leader, you’re given a target of “14 days” for waiting times. You’re told that your annual bonus and future promotions depend on hitting those targets. What do you do?
It’s important to keep in mind that demand for VA treatment has increased dramatically in the past decade.
From 2002 to 2012, the VA handled a steady increase in outpatient visits, from 46.5 million annually to 83.6 million. The number of veterans receiving mental health treatment grew from less than 900,000 in 2006 to more than 1.2 million in 2012. The 1 million new disability claims filed in 2009 were an all-time high, until 2010 and 2011, when new records were set.
Spending up “using 2011 dollars, America spent $88.8 billion on the VA in 2007, and $125.3 billion on the VA in 2012.”
But has it been enough?
If spending would have to increase proportionally to volume/demand (probably not a good assumption), we had a 44% increase in outpatient visits and a 29% increase in spending. Was that enough to keep up? Was the VA already under capacity before these wars started?
Update (from the previous VA Secretary):
The VA’s budget has more than tripled, to $154 billion in 2014 from $49 billion in 2001, the year I became secretary. In that time, the veteran population has declined to 21.9 million from 25.5 million.
So maybe “lack of spending” isn’t the problem and wouldn’t be the solution?
If you’re a manager of a VA facility and you see that the number of appointment requests is greater than your capacity, you have some choices and you can:
- Figure out how to increase capacity and throughput (perhaps, using Lean) without spending more: Possible, but moderately difficult?
- Ask for additional spending: Unlikely and pretty impossible?
- Cheat the system and make things LOOK better: Possible, and fairly easy?
Too many local leaders chose to fudge the numbers by creating secret waiting lists and the like.
As Brian Joiner wrote in his outstanding book Fourth Generation Management: The New Business Consciousness, there are three things that can happen when you have a quota or a target:
- Distort the system
- Distort the numbers
- Improve the system
It seems like there was too much focus on #1 and #2 at VA locations.
When top leadership sets up measures and just accepts what’s reported at face value… inviting and REWARDING gaming of the system, that’s the fault of top leadership for being naive and for not helping fix the real problems (the lack of capacity and long waiting times).
It would have been a better countermeasure to speak up and ask for help, since the VA is allowed by law to pay for private healthcare for veterans (and spends 9 to 10% of its budget on this). That’s been one “short-term countermeasure” recently – putting more focus on outside resources – but it’s a shame that it took a national embarrassment for this to occur.
Who Do We “Hold Accountable?”
When there’s anger and top leaders are “mad as hell,” people want “accountability.” It’s a good start (but not enough) for leaders to be mad about bad performance… this anger can lead to improvement. But, if the anger just leads to scapegoating and punishment (such as throwing individuals “under the bus”) then, that’s not as good. I put the term “accountability” in quotes because it’s a very loaded to term. In this day and age (and this is true in many private hospitals), accountability means blame and punishment. Punishing individuals doesn’t address the underlying systemic issue.
As Dr. Lucian Leape, one of the fathers of the modern patient safety movement says:
We need to quit blaming and punishing people when they make mistakes and recognize that errors are symptoms of a system that’s not working right, and go figure that out and change the system so no one will make that error again, hopefully. We have to change the culture, so everyone feels safety is his or her responsibility, and identifies hazards before someone gets hurt.
I think that applies to more than safety – it applies to nearly every other performance management issue.
Earlier in the investigation, President Obama said:
“I know that people are angry and want swift reckoning. I sympathize with that. But we have to let the investigators do their job and get to the bottom of what happened. Our veterans deserve to know the facts. Their families deserve to know the facts. Once we know the facts, I assure you if there is misconduct it will be punished,” the president said.
So who do you punish? People at individual VA locations? I’d argue that if you’re going to punish anybody, it’s those responsible for the design of the system. Who set the targets and who set up the bonus incentives? Who didn’t create an environment where local leaders could speak up? This probably goes up pretty high to the top, close to if not to Secretary Shinseki. That’s why there are bi-partisan calls for Shinseki to resign. I’d call for that too, if anybody cared.
Today, President Obama doubled down on accountability, saying:
Anybody found to have manipulated or falsified Veterans Affairs records “will be held accountable,” President Obama said Wednesday…
Speaking about reports of long wait times — and efforts to cover up the delays — Obama said that if they’re proven true, the behavior is “dishonorable” and “disgraceful.”
“I will not stand for it,” Obama said. “None of us should.”
This means punishment of lower level people:
“…he also noted that some employees had already been put on administrative leave.”
Again, I think that’s not the right approach. Gaming the numbers might not have been a high-integrity move or was it the bravest thing to do… but when people’s bonuses and paychecks are at stake, I understand why most of the local leaders and doctors went along with the ruse.
It’s tempting for Shinseki to deflect responsibility by blaming those below him in the chain of command:
“Earlier this month, Shinseki told NPR’s Robert Siegel that he would “take swift and appropriate action” if reports that the Phoenix Veterans Affairs Health Care system kept two lists of veterans waiting for care – one for sharing with Washington and another showing wait times that sometimes reached beyond a year — were substantiated.”
He’s pointing to a problem in PHOENIX, when it’s more broader and more systemic than that.
So What Now?
What do you think of the government’s problem solving and corrective action plans so far? Are you optimistic that things will be truly fixed or will we just see more “naming, blaming, and shaming” that doesn’t get to any root causes?
Will a scapegoat be named? Will it be Shinseki? Will it matter?
I think the more important questions with longer-lasting impact are:
- How will we increase capacity and throughput to get REAL waiting times down, providing better care?
- What sort of systemic changes will be made regarding targets, performance evaluation, bonuses, and the such?
How do we REALLY fix this?
Note: As new information comes in, I’ll post comments instead of editing the post…
About LeanBlog.org: Mark Graban is a consultant, author, and speaker in the “lean healthcare” methodology. Mark is author of the Shingo Award-winning books Lean Hospitals and Healthcare Kaizen, as well as the new Executive Guide to Healthcare Kaizen. Mark is also the VP of Innovation and Improvement Services for KaiNexus.