Using BPM to Assure Information Quality

I know, there is no such thing as 100% quality, but we need to challenge ourselves to get there. To catch potential data quality issues, you’ll need to create a set of validations, added to a process that will identify and fix them.  This set of validations can be automated within a workflow, identifying rules and actions to perform given certain conditions. Finding and fixing issues is an ongoing task: it requires a balance of vigilance and curiosity, as well as caring. Usually issues come about because there is spotty accountability somewhere in the flow of information from the source to the downstream systems.

Goal

Close any data integrity gaps by applying validation checks and fixes.

How does this happen?

It can happen very gradually and surreptitiously. Within a company, unless there’s a strong information quality department, there will inevitably be data inconsistencies because each department has different priorities and validation requirements. All it takes is one form in one application with lax data input requirements. It could also be a lack of validation check during data input. Lastly, when software is upgraded or data is merged from one system to another, we assume wrongly that one source data is fully vetted.

Examples of how this happens

Downtime

Let’s say your system has an account lookup feature, but that feature is down, so you have to enter in the account information manually. The feature is fixed in a few hours, but by then you’ve entered 100 accounts. Does this data get validated later? If it doesn’t, a downstream application could have quality issues.

Patient information merges and updates

Lets’ say you work at a hospital. There’s a patient referral with the same medical record name (MRN) as an existing patient with the same birth date. The referral is entered in at the existing patient. This error is caught later and the patient information is fixed, but did the patient already get treated?

Towards Quality with Process Automation

By inserting workflows into the process, specific types of data inconsistencies can be identified, investigated and resolved. Below are some general design components for building a quality validation workflow:

·         Figure out how to funnel all data/content through the validation workflow. Using by doc type or input sources, the information can be collected and filtered as appropriate.

·         Create the rules to route issues into buckets. Here are some typical queues:

o   Routing: this queue has validation checks which compare metadata values against source of record values.

o   Issues queues: these correspond to the common issues that get identified.

o   Routing issues: this queue holds any doc that doesn’t match the issues queues.

·         These buckets can be evaluated during their initial manual fixes for potential automated solution, and identifying upstream, root cause, data issues.

The Power of Visualizing Checklists

“…the volume and complexity of what we know has exceeded our individual ability to deliver its benefits correctly, safely, or reliably.” ― Atul Gawande, The Checklist Manifesto: How to Get Things Right

We all make lists of activities for today and in the future. As we write them down and/or type them up, we form categories if the list is long. We rearrange and rename the categories. When everything looks good, we can relax, right? Not really; I’ll explain why.

After a month of working on a checklist for a large parallel upgrade project, I decided use Microsoft Visio to represent the checklist’s categories visually as a diagram. Each box in the diagram would correspond to set of tasks. As I was thinking about these boxes, a certain order became clear; an order of what was being done and which team was responsible.

The major sections of an upgrade are Preparation, Pre-Upgrade, Upgrade, Post-Upgrade, and Support. These sections are natural top level boxes. Underneath each of these top level boxes are their corresponding task categories.

As I created and placed the boxes under the sections, I renamed some of the categories and consolidated others. The visual representation allowed me to “see” that some were the same, and some tasks were missing. I could reorder the task categories. I could also see that we needed more support tasks after the upgrade.

I know I could have used MS Project to do this, but it’s not as flexible. Project also presupposes activities that may not add any value to this type of visualization. If you look at the end visual result above, hopefully you’ll see what I mean; by breaking out categories as boxes, it’s easier to see gaps and similarities than if you just had a long list.

ECM Organization Structure Evolution

A few years ago, Greg Clark posted on AIIM’s Community blog called, “Creating an ECM Organization Structure…” He divided ECM structures into small, medium, and large organizations. At the end of the two part blog, he talks about the need to have a “Change Management” role in the mix for all company sizes. Although he provided structures were are static in terms of growth, most cases small companies grow and evolve.

Understanding how ECM structures go from small to medium and medium to large, as well as the deformed nature of most ECM structures is the goal here. Let’s work through change management at the management level, not just the application level of a typical healthcare IT organization.

“You may ask yourself, well, how did I get here?” – Talking Heads

In general, healthcare ECM is either in the EMR or the Finance IT camp. As Mr. Clark outlines, a large ECM should stand on its own, outside of IT along with EMR. This depends on which IT group and business area developed the ECM solution first. Both finance and health information management (HIM) departments are inundated with paper.

Finance

In the past, paper coming in from 3^rd party invoices and employee expenses was enormous. The ROI on implementing a scan/store solution was an easy sell for Accounts Payable. This is why many ECM solutions were first introduced into the finance IT area.

HIM

Health information went from purely paper to EMR relatively quickly. This meant that some areas were left alone because of the need to implement “80%” of the solution. Part of every EMR solution is dealing with scanning and storing paper forms, especially ones that need to be signed, or ones that are brought in from other healthcare providers. ECM can easily be introduced with in the EMR solution. If your hospital has EPIC as its EMR, chances are good that Hyland OnBase is there as well.

Who wins Finance or HIM?

Many issues from project priorities to information governance are usually after thoughts when it comes to thinking about change management of ECM organization. Moving the ECM group from either Finance or HIM into its own program can be a  battle. How do you split up the project intake process when there’s no governance or program management specifically for ECM?

What happened to ECM governance?

When ECM was initially rolled out there was most likely some type of group or consultant that wrote out the goals, strategy and deployment plans. The documents and diagrams have dust on them. The regular meetings went from weekly to yearly. The intake process is on fire. And so on… Chris Riley in his http://searchcontentmanagement.techtarget.com/ article, “Taking charge: Structuring and managing an ECM governance committee”, explains the need for sustaining a governance committee:

“A strong ECM governance committee reduces the burden on IT and establishes a more robust and broadly supported governance framework than IT could build on its own. Not only that, committee members who are properly involved in creating governance processes will stand by a system and be the first defense against ECM naysayers.”

How the Program Management going?

With ECM governance intact, Program Management can exist beyond the project management software and “project managers” pushing milestones. At the program level, according to Mr. Clark, we are talking about the following responsibilities:

·         Drive the creation and execution of the ECM strategy

·         Liaison to the steering committee

·         Manage vendor relationships

·         Hire and manage the ECM team

·         Contribute to or lead the development of an appropriate information architecture

·         Identify, track and act  upon key metrics for the ECM program

Ok, so here’s where the split between Finance and HIM could have the potential to merge. It would take a program level management structure to efficiently handle ECM projects. Otherwise, the project priorities are dependent on the old school influences which only worked in smaller organization.

Here’s what Mr. Clark says:

“The most important aspect of this particular structure is that the ECM group is not part of the IT department, Legal group or even an administrative or business services group. This organization has chosen to align ECM with overall business operations in a group called Technical Operations and Competence. This division is responsible for helping the organization achieve operational excellence across their core business, and includes such functions as Engineering Standards, Maintenance and Reliability, Environment, Health and Safety (EHS), Training and Development, and, finally, the ECM function.”

Pulling a mature ECM program out of IT may seem impossible, however the reasons for doing so should outweigh the reasons for status quo. A liberated ECM program strategy is free to pursue the higher level goals of the organization with being held back by IT. 

99% Accuracy is Good, Really?

Ok, I know, it’s obvious that, when there is one mistake in a hundred, quality is pretty good. But 1% failures add up when there are 10,000 records: that’s 100 failures. Even small or medium systems can process a million transactions a day. I guess it’s a matter of how important the accuracy of the data matters to the business. You’d be surprised though how many failures are dealt with manually without fixing the root causes of the failures. Let’s consider what types of failures this 1% might impact:

Pharma: good clinical trial practice

Clinical trial documentation keeps track of test results by patient demographics. If there are any data quality issues, the effects ripple throughout the trial results. The safety and efficacy of the trials can easily be influenced by any data quality issues, let alone 1%. Audits for CFR part11 compliance will be disastrous otherwise.

Healthcare: patient safety

If registration systems are not 100% accurate, you could have a patient safety issue. For example, an allergic reaction to a specific medication is not information that should be 99% accurate. I have experienced this first hand and I can tell you it is not fun. Try requesting information on your visit to a hospital and see if it is all there. Even .001% failure is not acceptable.

Finance: entities of entities

At a hedge fund, following the account numbers between entities and fund of funds, accuracy has to be 100%. Documentation metadata has to be validated against source data at every possible opportunity. SOX fines would add up otherwise.

If anyone at your company considers 99% information accuracy a viable goal, ask them if they would care if they got 1% less pay in the paycheck.

The Role of MISC in ECM

I used to loathe working on categorization projects because I knew eventually someone would say, “I don’t know, put it in a miscellaneous folder”. This meant that the overall design of the categories was flawed or that we didn’t have enough time and energy to work out every minute detail, only to have it change in a few months anyways.

Flexible categorization makes sense, but the tools are still designed to tag content with fixed values. Big Data solutions might eventually help with this if you have millions of dollars to spend on them. For small to medium sized systems, we are stuck with good old fashioned indexing and search. However, this might prove to be a better long-term solution to content mining. I still believe the better the metadata, the better search results.

The Misc folder suits a number of different purposes:

In general, this folder can be used to analyze new trends in metadata values, that is, some patterns of values will become apparent as more content goes there. Over time, the patterns will become folders/categories and there metadata values will become part of the indexing process. Likewise, categories that are almost empty will be merged with others because their index values are too restrictive.

In taxonomy, a miscellaneous folder is a black box, something that gets all that is outside the scope of the people working on it at the time. Emphasis on “at the time” here because as ways of organizing information changes, so goes the taxonomy.

In workflow, miscellaneous really means a place/bucket where all the routing mistakes are sent, or more likely where any new unanticipated content types go. This works well as it is obvious as the content builds up which queue they should be routed too. The alternative would be to ignore the outliers which would leave them for discovery projects in the future.

Drivers and Followers and ECM

Futurama (New York World's Fair)

In The Option of Urbanism, Chris Leinberger, describes how there are “drivers” and “followers” in the product types of real estate. For example, a driver would be residential home development and a follower would be a mall constructed close to where the new homes are built.

The same is true for enterprise content management. For example, a driver in healthcare applications would be an EMR, and the follower would be all the other applications, except registration and billing. EMRs were late to the game, but are now front and center. ECM systems, like Hyland’s Onbase, fill in the gaps as a follower in the long line of other applications.

Gaps occur over time as applications are moved in and out of favor, CIO’s change things up to reduce costs, requirements change, and priorities are mandated. Gaps are filled with invoice scanning solutions for paper processing. When regulations change, ECM has the flexibility to meet the requirements until the larger “driver” systems and adjust.

Hospital merger and acquisitions are “drivers” which spur multiple content and information migrations. The “followers” help clean up the information messes caused by disparate systems. Content types and metadata have to be mapped and reclassified. Who does this? ECM applications and professionals are well versed in these types of situations.

Understanding ECM’s flexibility and its major role as an IT “follower” is essential when making strategic decisions on where to deploy it. It innovates when “drivers” need it to. It fills in the gaps quickly to provide the needed bridge for unforeseen obstacles of business change. It’s easy to get excited about innovation “drivers” and much harder to involve the “follower” applications in this excitement. 

I'm Not Just An ECM Flounder!

Hi, I’m not just an ECM flounder at the bottom of your information ocean. I’m affected by any changes in how or what information is dumped into the sea. Let’s look at some of the ways that my environment gets polluted.

According to CIO’s Howard Baldwin, here are some IT complexity issues that should be looked at in order to decrease pollution levels. Baldwin calls it “complexity”, I call it pollution. As a bottom feeder, I need a lot of coordination and cooperation from above so I can live and prosper. I’m only as fit as the polluters will allow me to be. If I’m not in shape, I can’t innovate. Sorry about that last rhyme, unintended…

Documented Knowledge

How many application integrations are undocumented? Or, documented years ago and not revised? Every undocumented process turns into an oil leak into the ocean, and I have to deal with it during upgrades and new development. I know I have to make sure everyone is aware of changes I make to the ECM system, but same goes for all you applications at the surface.

Consolidate to a Cloud

As an ocean dweller, I would have no problem if all information was filtered in the cloud. This would necessitate more accurate requirements, functional specs, and testing documentation. The rain would be less polluted. The oxygen levels would be excellent. The bottom would be less murky. The ongoing maintenance of the infrastructure and integration points would drop considerably.

Complexity ebbs and flows

Like ocean tides, information inputs and outputs ebb and flow. Complexity can accumulate quickly with director churn, as they need to prove themselves and usually build new applications beside existing ones. Talk to anyone dealing with SharePoint beside an existing ECM solution and you’ll get an ear full.