Enabling Healthy Decisions

Healthcare is such an easy target for BPM opportunities.  I am living one right now.

In December, we had to refill a prescription for my son.  It happened to be 2 days before Christmas.  Since I had switched insurance carriers, Aetna now required a PA (Prior Authorization) for the drug (although my son had been on it for a year).  By the time, I got the message and tried to call the physician, he (a specialist) was gone until the new year.

I had to fill the prescription and pay cash since I needed my son to stay on the drug.  So, in January, I downloaded the appeal form; completed it; had the pharmacist complete it; and mailed it in.  In March, I heard back from Aetna that it was rejected.  Fortunately, after working in the industry, I know that you can appeal it multiple times, and I understand the coding and reject process.  After talking with a call center agent, I had them change some of the notes on the claim to reflect the situation with my physician and appealed it a second time.

It is now May, and I got a letter telling me they haven’t finalized the review process but have received my appeal.  I should know something in 60 days.

This has to be easier.  All parties have to be frustrated.  The Aetna PBM has to spend expensive pharmacist time reviewing these multiple appeals; answering multiple calls by me; and sending me multiple letters.  My employer and Aetna will eventually have to pay a claim from 2006.  I am out the cash and spending valuable time trying to work the process.  And, none of us have much if any visibility into the number of appeals, the average cycle time, the status, and how to change this process.

BPM could help with this in many ways including simply streamlining the process.

From a pharmacy perspective in healthcare, one of the more elusive solutions has been e-prescribing (or eRx).  There are numerous companies (e.g., Allscripts, Prematics, Purkinje, RxNT, Zixcorp, iScribe) that have solutions and many more that have died over the years.  When I was at Express Scripts, we funded RxHub which was an industry solution by the PBMs (Pharmacy Benefit Management companies) to facilitate eligibility checks and other electronic transactions.

The goal of e-prescribing is to eliminate safety errors associated with wrong prescriptions, drug-drug interactions, non-compliance, and many other issues.  Anyone who has ever seen a doctor’s hand written prescription can understand.  Here is an example from the Prematics site:

Some of the managed care companies have spent millions rolling out these solutions.  In general, most of them are adopted as interesting tools and then disgarded by MDs at the first sign of trouble.  Honestly, there is very little in it for them unless they have a quality bonus.

Physicians are one step removed from the filling of the Rx.  They care about patient safety, but it is hard to change their working habits.

So, I wondered the other day whether BPM could solve the problem.  Doctors all use some type of system for managing their office.  This ranges from simple systems that just capture data to much more involved PPMS (Physician Practice Management Systems) or EMR (Electronic Medical Record) systems.

Could a BPMS use a process to:

1. Push data from the physician’s desktop using messaging queues;
2. Run the data through some eligibility and adjudication logic;
3. Identify issues real-time; and
4. Then either route it back to the MD for questions or send it to the patient’s pharmacy of choice.

An interesting spin on this would be to do an auction at the end where pharmacies could bid on the prescription (e.g., BidForRx).  Patients could create customized rules in the BPMS solution that processes the claims and managed the auction based on their interests.

Just a thought.  This seems to make adoption easier (no change in process); deployment cheaper (no ongoing support); integration cheaper and easier; and add additional value (patient involvement).

Anyone who has ever been sick knows that our healthcare process is inefficient.  There is lots of paper.  There are numerous handoffs.  Things get dropped.  Patients spend lots of time resolving things that should be automated.  Well, I am not going to give you the answer since there are millions out there trying with limited success.

But, I do think a BPM framework (strategy, process, and technology) makes sense here.  First, I think you need to abstract the healthcare industry.  Think of each component (MinuteClinic, doctor, hospital, outpatient clinic, home health nurse, pharmacist, insurance company, PBM, case manager, therapist, acupuncturist, personal trainer) as a system within an overall company.  Second, think of the patient experience as a process filled with logical steps and driven by rules.

Conceptually, you have a systemic view of healthcare that parallels a typical process view of a inter-company process.  You have constituents.  You have a process.  You have rules.  You have systems (i.e., companies).  Ideally, each of these systems (e.g., the physician’s office computer) could be accessed by the process using a common language.  (I am speaking to a Services Oriented Architecture (SOA) without going there technically.)

Each time a patient engages the system through a doctor’s appointment or a visit to the ER, a new process instance would be initiated.  The patient could be tracked through the process and know where they were at any time.  All the constituents would have shared data (i.e., a personal medical record) about the history of this patient.  Rules would be codified so hopefully there were less errors or miscommunications.

Now, who pays for this is the question which I am not going to debate.  But, the system is broken and finding some fix is important.  I think this is a lens through which to view the problem.

Here is another example from a previous life.  As part of my direct marketing program, we used to use automated outbound calls from a company called Silverlink.  They do a great job with the calls and providing management tools to monitor the calls.  But, a BPM solution could have helped us to automate the process around Silverlink.  The following scenario is how it could be:

• When a new client is enrolled, the set-up form could include a checkbox for participation in the direct marketing program.  The set-up form and process would also capture client approval for co-branding along with a jpg of their logo and other information to personalize the communications.
• That checkbox could initiate a process which pulled in client data from the set-up form, ran a query of their member data (name, address, phone number), and automatically enrolled them in the direct marketing program.
• A rule could exist that would check against any stipulations in their contract to determine whether they were active participants in message approval or whether they had already signed off on the general messaging within the program.
• The existing CRM system could be used to manage the campaigns or a BPM rules-based system could be used.  In either case, the system is looking at each case or process instance and determining which direct marketing path to follow, which letter to use, when to trigger a call, and providing status to the call center agents.
• Any change to the messaging (client, patient, regulatory, research based) could be made and all notifications and approval requests would be triggered by the system and managed by the system.
• Coordination with outside vendors such as Silverlink would be automatic and rules could be used to monitor their call volume versus inbound call capacity to determine automatically how to throttle up or down the number of events.
• Stoplight type triggers could be used by the system to monitor different events and rules written to trigger process changes or initiate problem resolution automatically.

To some people this may seem very basic, but I know I am not alone in managing many of these processes with lots of paper, implicit rules, offline databases, and other tools.  The discipline of BPM as an approach and the technology to automate and manage it can add a lot of value here.

In a previous job, I managed a direct mail program where we mailed several million pieces per year.  It was a complex process which included getting client sign offs, interrogation of data against a customer segmentation model, managing several hundred letter variations, and coordination with a inbound call center for responses.  Although we had a decent Visio diagram for our process and well defined SLAs (Service Level Agreements) and rules, a BPM system would have been ideal.

Let me just focus on the letter approval process.  Several things drove letter changes – regulatory changes, corporate branding changes, lessons learned from our campaign results, client requests, physician requests, and consumer requests.  Each letter change went through my product manager who had to sign off.  It then went to marketing for sign off which often took at least one revision back to product management.  It then went to account management for sign off.  It then went to legal to sign off.  And, many of them then went to clients for sign off with many edits in between.

If someone was out of the office or busy, this could take weeks if not months to complete while the total time of the task was probably about one day from start to finish.  The challenges were the need for discussion and supporting documentation around each change (e.g., why is it better to call someone a patient versus a member); access to the previous versions to understand other changes; and a way to know who had signed off to date.  With a BPM system, this could have been managed very easily taking advantage of the workflow, the business rules for routing and escalating tasks, reporting for understanding the status, document management, and collaboration for a discussion tagged to the process instance.

Continue reading →…