Friday, April 29, 2016

Back to Paper After U.S. Coast Guard EHR Debacle: Proof of Hegel's Adage "We Learn From History That We Do Not Learn From History"?

I have become blue in the face writing about healthcare information technology mismanagement over the years.  In fact, the original focus of my 1998 website on health IT (its descendant now at http://cci.drexel.edu/faculty/ssilverstein/cases) was on HIT project mismanagement.

If this industry actually had learned anything from history, I would not be reading nor writing about brutally mismanaged HIT endeavors in 2016.  Sadly, that is not the case.

The Coast Guard, founded by Alexander Hamilton, has this as its motto and mission:

http://www.gocoastguard.com/about-the-coast-guard
Semper Paratus - Always Ready.

The Coast Guard is one of our nation's five military services. We exist to defend and preserve the United States. We protect the personal safety and security of our people; the marine transportation system and infrastructure; our natural and economic resources; and the territorial integrity of our nation–from both internal and external threats, natural and man-made. We protect these interests in U.S. ports and inland waterways, along the coasts, on international waters.

We are a military, multi-mission, maritime force offering a unique blend of military, law enforcement, humanitarian, regulatory, and diplomatic capabilities. These capabilities underpin our three broad roles: maritime safety, maritime security, and maritime stewardship. There are 11 missions that are interwoven within these roles.

It seems the Coast Guard personnel need personal protection from the HIT industry, for the motto of that industry, sadly appears to be something like "Stupra Acetabulus" (Screw the Suckers).

From Politico, one of only a few publications that in recent years has taken a critical approach to this industry and pulls no punches:

http://www.politico.com/story/2016/04/ehr-debacle-leads-to-paper-based-care-for-coast-guard-servicemembers-222412
EHR debacle leads to paper-based care for Coast Guard servicemembers
By Darius Tahir
04/25/16

The botched implementation of an electronic health records system sent Coast Guard doctors scurrying to copy digital records onto paper last fall and has disrupted health care for 50,000 active troops and civilian members and their families.

Five years after signing a $14 million contract with industry leader Epic Systems, the Coast Guard ended its relationship with the Wisconsin vendor, while recovering just more than $2.2 million from the company. But it couldn’t revert back to its old system, leaving its doctors reliant on paper.

This state of affairs is simple inexcusable.  It represents gross negligence and severe multi-axial incompetence at best - but likely primarily not by the Coast Guard, whose core competency does not include HIT.

There’s no clear evidence the EHR disaster has harmed patients, and a Coast Guard spokesman said the use of paper records hasn’t affected “the quality of health care provided to our people.”

Proof by lack of evidence is not reassuring in a debacle of this kind.  However, the Coast Guard admits that paper records aren't the clear and present danger the IT pundits make them out to be.

Politico is skeptical of the claim:

That seems unlikely. Without digital records, if a patient goes outside a Coast Guard clinic, it can take weeks for the paper record to follow him or her back to the Coast Guard, says Michael Little of the Association of the United States Navy. And since the Coast Guard primarily provides outpatient, rather than hospital, services, many of its patients seek outside care.

“It’s one thing if you’re doing paper-based [care] in Ohio, but what about if you’re on paper records in [an] icebreaker or cutter in Alaska, and you need your gall bladder removed?” said Little, the organization’s director of legislative affairs.

In this case, I disagree that the lack of records is so dangerous.  There's the telephone, FAX machines, the patient himself or herself, and the hand-carried note.  Used with care, those serve care reasonably well. 

With the Department of Veterans Affairs weighing whether to buy a top-of-the-line commercial electronic health record and the Pentagon beginning a multibillion-dollar EHR implementation, the Coast Guard case displays how poorly the process can go for the government, even when the biggest names in health IT are involved.

Not just the government.  I'd also argue that this shows that the "biggest names" are, at best, overextended, and at worst, badly needing external investigation as to their software development, customization, implementation and support practices, as well as hiring practices (e.g., see my August 15, 2010 post "EPIC's outrageous recommendations on healthcare IT project staffing"
at http://hcrenewal.blogspot.com/2010/08/epics-outrageous-recommendations-on.html) and contracting.

Reversion to a purely paper-based system is a rare event in the recent annals of electronic records, said Thomas Payne, a health IT expert at the University of Washington. “I can think of examples where that has happened, but in the last decade that is much less common.”

I believe that is because of the general invisibility of, and immunity from, the risks and harms that occur from "making do" with bad health IT due to financial pressures.  Hence one sees hair-raising examples like I wrote of at my Nov. 17, 2013 post "Another 'Survey' on EHRs - Affinity Medical Center (Ohio) Nurses Warn That Serious Patient Complications 'Only a Matter of Time' in Open Letter"at http://hcrenewal.blogspot.com/2013/11/another-survey-on-ehrs-affinity-medical.html where going back to paper to allow a complete rethinking of the EHR implementation would likely have been the safe response.

See also, for example, my July 2013 post "RNs Say Sutter’s New Electronic System Causing Serious Disruptions to Safe Patient Care at East Bay Hospitals" at http://hcrenewal.blogspot.com/2013/07/rns-say-sutters-new-electronic-system.html (there are links there to still more examples).

The Coast Guard is tight-lipped about the causes, timeline and responsibility for the debacle. “Various irregularities were uncovered, which are currently being reviewed,” a spokesman said.

The causes are all covered at http://cci.drexel.edu/faculty/ssilverstein/cases/, and have been since the late 1990s.  In the alternative, the book "Managing Technological Change: Organizational Aspects of Health Informatics" (http://www.amazon.com/Managing-Technological-Change-Organizational-Informatics/dp/0387985484) by Lorenzi & Riley does likewise for an even longer period, since the mid 1990s - for those willing or able to learn from history and from the pioneers

There’s no shortage of candidates: the service relied on five separate vendors to build the new system, and its own planning seems to have been at fault.

Lawmakers are looking into the matter, said a spokesman for the Senate Appropriations Committee, which is “monitoring the situation."

This is symptomatic, in my view, of the fact that there are a lot of "Beltway Bandit" IT consultant companies doing business, few of them very good.

Bungled implementation, followed by chaos

In September 2010, the Coast Guard bid out the contract to Epic Systems, then added an array of other contracts to software vendors and consultants to help implement it. Since 2010, the agency spent, on net, just more than $34 million on health IT.

In a January 2011 speech, Coast Guard Chief Medical Officer Mark Tedesco cited the success of Epic installations at Kaiser Permanente and Cleveland Clinic. He predicted that the Epic implementation would improve the health of its population and save money.

Overall it’s a cheaper system for us to run than to upgrade to [the next generation military EHR], because of what that would’ve meant to us infrastructure-wise and support-personnel wise,” he said.

It's stunning to think what this says about the next-generation military EHR.  The previous one was not very good, either (see my June 4, 2009 post "If The Military Can't Get Electronic Health Records Right, Why Would We Think Conflicted EHR Companies And IT-Backwater Hospitals Can?" at http://hcrenewal.blogspot.com/2009/06/if-military-cant-get-electronic-health.html). 

Trouble, apparently, struck quickly. The solicitation for the EHR contract envisioned rolling out the software within six months at two to three pilot sites, before deploying it to a total of 43 clinics and the sickbays aboard the Coast Guard’s fleet.

That didn’t occur; the system never deployed to any clinic or cutter, said Eric Helsher, an executive with Epic. The next missed deadline was March 2012, which Trent Janda — the Coast Guard doctor serving as project leader — announced in a summer 2011 newsletter of the Uniformed Services Academy of Family Physicians.

One can only wonder what penalties the contract called for if the goals and timelines were not met.  That software was not deployed even to any pilot sites is nearly unimaginable to me.

As Janda set the new goal, he acknowledged there had been “multiple hurdles and delays,” and explained that the service had expanded its ambitions.

“Immediately upon award of the contract, we began a comprehensive analysis of the clinical workflows and existing information systems,” Janda wrote. “Many of the weaknesses became apparent as we compared ourselves to industry standards and best practices. Frequently, a weakness would lead to others, ultimately leading to the need for an additional system. The work-flow analysis quickly grew into a system wide re-engineering project like a snowball rolling down the mountainside.”

This sounds like a groundbreaking level of project mayhem and chaos, even for HIT.

The comment reveals that the agency failed to do necessary advance planning, says Theresa Cullen, an informatics executive with the Regenstrief Institute who formerly worked with Veterans Health Affairs and the Indian Health Service.

“They should have done a full needs assessment,” she said. “One would have normally done the workflow evaluation prior to the release of the RFP.”

If true, I believe it was an obligation of EPIC and the multiple contractors to have pointed that out to their future customer, and adjusted their bids accordingly, taking into account the time and resources needed for this type of work - or not placed a bid at all.  Such deficiencies and what they mean towards project progress and failure are obvious - to anyone who's learned from history.

... Cullen also found it odd that the Coast Guard didn’t hire consultants to implement the new system until September 2012. The service ended up hiring Leidos, which also maintained its old EHR.

The Coast Guard further complicated the process by deciding to team up with the State Department. Its original request was complicated enough, with installations spanning six time zones. The partnership with State meant implementing across 170 countries. (A spokeswoman for State said the agency was investigating its options, but refused additional comment).

The sheer number of sites led Cullen to question whether Coast Guard and State had devoted enough resources to the project. Between Epic and Leidos, the project was budgeted for roughly $31 million. That was “an inadequate amount of funding for what you’re asking to do,” she said. Consultants receive roughly $100 an hour, and Epic’s work with clinicians is time-consuming.

Again, those hired knew, should have known, or should have made it their business to know that under such conditions, if true, project failure was the predictable outcome.  They are supposed to be the HIT experts, after all, not the Coast Guard.

While a very efficient health care system could implement the EHR, she said, the Coast Guard lacks that reputation. She speculated that Epic intentionally underbid the contract. (Epic’s Helsher said that “the contract was viable and we were fully motivated to lead a successful install.”)

Someone is right, and someone is wrong.  I leave it to the reader to decide who was correct and who wasn't.

Anecdotes of further delays pepper various newsletters and reports from 2012 through 2015. Server failures scuttled a pilot rollout in 2014, then developed into deeper problems, and last July the systems started failing on a more regular basis.

Perhaps the "anecdotes" need to be turned into "teachable moments" through legal discovery by federal law enforcement.

The Coast Guard advised retirees and dependents that month that, due to incompatibility between its EHR and the Department of Defense’s new medication reconciliation system, they couldn’t get their prescriptions filled at Coast Guard clinics.

Around Labor Day, Coast Guard health care personnel were directed to copy information from electronic files onto paper, for fear of losing their data.

That is just about the most pathetic sentence I've ever had to read in my 24 years in Medical Informatics.

... doctors are frustrated. One complained in the Uniformed Services Academy of Family Physicians newsletter of “unique challenges which seemed to revolve around many electronic record keeping changes.” “The question we pose is, how is this affecting shipboard life?” Little said. “This is the most important thing that’s happening right now in the Coast Guard.”

My advice to the Coast Guard is to treat the IT invaders and consultants as it would a invading maritime fleet from a hostile nation.

The vendors who worked with the Coast Guard either don’t know what went wrong, or aren’t telling. Leidos — also the lead company implementing the Pentagon’s EHR project — declined comment, as did Lockheed Martin, which was contracted to implement access to the EHR through mobile devices, and Apprio, which was to provide credentialing services.

I believe they have a very good idea of "what went wrong", and aren't telling (per the Fifth Amendment)?  If they have "no idea" what went wrong, what, I ask, are they doing in the IT consulting business?

... The EHR giant [EPIC] says it’s not entirely clear why the Coast Guard pulled the plug. But the situation wasn’t Epic’s fault, company executive Eric Helsher said.

They pulled the plug out of fear for their members' well-being, hopefully.

It seems everyone seeks to escape culpability, with the blame placed on the customer.

The Coast Guard spokesman said the decision was “driven by concerns about the project's ability to deliver a viable product in a reasonable period of time and at a reasonable cost.”

It seems there's still some who don't continue down the sunk-cost fallacy road (https://www.logicallyfallacious.com/tools/lp/Bo/LogicalFallacies/173/Sunk_Cost_Fallacy) and are willing to walk away from bad HIT.

... In general, software contracts deserve more scrutiny, said Kingston, who served on the House Appropriations Committee. “These things don’t get the scrutiny a weapons system does.”

Considering the reputation of military costs, that's saying quite a lot.  The lesson that should have been learned from history is that HIT is both exploratory, and a relative free-for-all.

Caveat emptor.

One last piece of (free!) advice for the Coast Guard leadership.

Read this paper:

Pessimism, Computer Failure, and Information Systems Development in the Public Sector.  (Public Administration Review 67;5:917-929, Sept/Oct. 2007, Shaun Goldfinch, University of Otago, New Zealand).  Cautionary article on IT that should be read by every healthcare executive documenting the widespread nature of IT difficulties and failure, the lack of attention to the issues responsible, and recommending much more critical attitudes towards IT.  link to pdf

That may be the most valuable learning experience of all for their next attempt to implement EHRs.

-- SS

1 comment:

live IT or live with IT said...

You get what you pay for ... and sometimes not.