Moving death certificates to the digital age: Unintended consequences

Moving from paper death certificates to an online process called the Washington State Electronic Death Registry System, or EDRS. Sounds like a simple 21st century process that should make the process of completing a death certificate faster and easier right?  Oh, the government designed the system.  Oops.

I’ve been so frustrated by the unintended consequences of a new state governmental process for completing death certificates.

Beginning in early 2012 Washington state started an online procedure for completion of death certificates called the electronic death registration system or EDRS.  This was presented as an easy and efficient way to complete a task often assigned to the family physician when a patient dies.  This is usually not unreasonable, and is certainly expedient, but has become a source of great drama and frustration.  In the first few months of using this system I have yet to successfully complete an EDRS form online in time to allow the family to bury their dead without stress that the death certificate would not be done in time and the funeral could be delayed.  You can imagine the frustration of family members that their doctor can’t accomplish such a simple task.  I must seem like I just don’t care enough to get it done.  Actually I do care, and simply have been unable to complete the form to the satisfaction of the web-based form’s programmed rules.

First it will help to understand what happens when a person dies and death certificate needs to be completed.  When there is no concern about foul play or liability concerns the usual procedure is to ask the family who the patient’s primary care physician is and to request that physician to complete the death certificate.  This has historically been the best and most expedient way to accomplish this relatively mundane task.  The primary physician usually has a good idea of the patient’s health concerns and can usually complete the task in a timely fashion and even be reasonably accurate.  Sometimes we have to make assumptions and give or best estimate of the cause of death, but it has seemed to work pretty smoothly most of the time.

In the old paper system someone from the funeral home would bring me the death certificate, I’d complete it  as accurately as I could, and the grieving family could bury their beloved. That has all changed now.  In an attempt to speed the process up and make it more efficient, the state has created a website where I log on and complete the form.  I thought this would be an improvement, and it was presented to me as a way to simplify, expedite and improve the process.  I’m reasonably computer savvy, and so when the first EDRS (electronic death registry service) request came to me I logged on and tried to complete the form.

Here lies the problem. In an attempt to be sure that physicians are accurate and complete the form has underlying rules.  Maybe someone understands these rules, but I certainly do not.  I’ve read the instructions and it all sounds simple. The first task is to assign a cause of death, along with naming any contributing factors, and to answer several questions.  The questions include whether the patient was pregnant, whether tobacco contributed to the death, and whether trauma played a role.

These questions seem innocuous, except that the answers seem to trigger red flags to the reviewers that prevent the acceptance of the death certificate.  In addition there are only a finite number of acceptable causes of death it seems.  If I assign a cause of death that is not in the list of acceptable causes then the death certificate is not accepted.   The computer program “spell checker” does not have a human’s ability to recognize synonyms or various ways of saying the same thing.  If I say a patient died of COPD, a commonly accepted acronym on paper, the program rejects the cause and prompts me to spell out chronic obstructive pulmonary disease.  That’s OK because it leads me to an acceptable answer.  Other answers though trigger hidden problems that don’t reject the answer and lead to delays.

When an EDRS form has any type of answer that the program will not accept it goes to a real human somewhere who looks at it and they contact my office and ask me to fix the issue.  This tends to take about a day before I am notified that I need to fix the answers.  I then go back online and try to figure out what is wrong, and fix it.  So far my first attempt at completion of the EDRS form has never worked, nor has my first attempt to fix the issue.  When the  attempt to fix this does not satisfy the program, it again goes to a human, who again contacts me and I again try to fix it.  By this time drama is creeping into the scenario.  The patient’s family is anxious and often angry, because the funeral director tells them that they cannot bury or cremate their beloved until they have a death certificate.  I now start to get urgent calls from the funeral home, the EDRS office, and often the family that I need to get it right.  Just fill out the form correctly and do it right now so they can get on with the funeral.

This happened again this week.  An elderly blind woman with tobacco related chronic lung disease had fallen at home, clearly in my mind a type of trauma.  This had led to a hospitalization, respiratory failure and her death.  I completed the form, describing the trauma, etc.  The problem apparently was that because I said trauma was involved the program required review of the case by the medical examiner.  No one told me about this until mid-morning today, in the middle of a busy office schedule.  I was then told that the medical examiner had reviewed the case and that the trauma played no role in the death.  All I needed to do was remove the discussion of the trauma from the EDRS form and the computer program could allow it to be accepted.

This is clearly silly. Everyone knew that trauma played a role.  In a paper world the medical examiner could have just signed off on the death certificate and all would have been fine.  Now as an unintended consequence of the inflexible nature of this EDRS program, it seems we just have to be selective in our choices of contributing factors on the death certificate so that families can bury their dead and the computer programmers can have the answers they want.  I had been told that the new electronic form was to insure more accurate and complete death certificates.  It seems that the result is that only answers the program likes are acceptable causes and contributing factors to death.

I hope I never need to complete a complicated death certificate in a complicated death scenario.  So far even simple ones have confounded me.

Edward Pullen is a family physician who blogs at DrPullen.com.

email

Comments are moderated before they are published. Please read the comment policy.

  • http://twitter.com/GinsbergMD Daniel Ginsberg, MD

    I too found it frustrating at first, though most of the time now I find it pretty quick. There are a couple of things that can seem confusing. After you enter the data, it makes you run a spell check before you can move to the next step. If you make any changes you have to spell check again. I’ve never tried to see if you can force it to accept words it doesn’t recognized. I’ve also learned it doesn’t like abbreviations.

    If you enter any malignancy, it prods you to make sure you entered as much information about the cancer as you can.

    I’ve never had to enter a trauma case but I would argue that your patient did not die of trauma unless she fell purely accidentally and everything else that followed was a result of the trauma. If instead she fell because she was lightheaded because she had pneumonia, then the trauma was not the cause of the death, though it might have contributed. If it truly won’t accept a trauma diagnosis that you feel is appropriate, then you can always insist that the medical examiner file the certificate.

    One thing I never liked about the paper or electronic form was the appearance that everything was a stepwise progression, for example, a myocardial infarction, as a result of hypertension, as a result of hyperaldosternism. Often it’s multifactorial. The heart attack was caused by hypertension, high cholesterol and diabetes. I was much more satisfied when I discovered that you can enter more than one diagnosis on a line.

  • http://www.facebook.com/people/Joan-M-Stephens/735781946 Joan M. Stephens

    Have a funny feeling that every smoker will have a “smoking related” death.

  • http://www.texmed.org/ stlevine

    Texas physicians have “enjoyed” similar problems with our new electronic death registry — complicated by the problem that you must sign up for the system 14 days before you use it. Read more in this excellent testimony Fort Worth physician Gary Floyd, MD, provided to the Texas House Committee on Public Health: http://www.texmed.org/WorkArea/DownloadAsset.aspx?id=24121