Monday, February 22, 2016

'Big Data' Data Quality Issues

Many attempts will be made to incorporate the use of 'Big Data' tools and techniques to obtain insights into the provision of medical care in the United States. However, as the 1st law of computer science is: GIGO (this is also the first law of archeology by the way) - Garbage in, Garbage Out, we need to discuss if their are any systematic ways that clinical practice data are systematically skewed. 

The harsh reality is the Golden Rule now rules the entire clinical enterprise. And the Golden Rule has a potent and biasing effect on data quality. 

I recall when young & idealistic that the pursuit of the best description of the patient was based on a genuine attempt to understand the patient in their enormous and at times beweildering uniqueness. 

Now due to extraordinary time constraints being leveraged against even the most idealistic practitioner & even more perniciously the gaming of the diagnosis to justify the highest DRG (Diagnostic Related Group) payments made by an ever increasing group of hospital employed physicians, I seriously question every diagnosis not made by myself.

The diagnostic related group is a payment method utilized by the payers to compensate hospitals for their care. The process specifies the number of days a typical patient with that diagnosis takes for discharge and the hospital is than compensated for that typical stay. As is obvious, a respiratory failure patient who requires intubation and mechanical ventilation for days, weeks, or sometimes for months, hospital will be compensated at a much higher rate than a patient with mild pneumonia requiring observation for 24 hours to ensure his antibiotics are effective.  

I haven't seen any patient admitted to my current hospital with chest pain who is not admitted as anything other than Acute Coronary Syndrome (ACS) . 

Their are four reasons for this. The first is that its guarantees the highest DRG payment for the hospital, secondly it means not having to justify and defend yourself for NOT giving guideline mandated best practice therapy if you diagnose the chest pain as anything other than ACS, and its turns out that you were wrong.  Increasingly emergency medicine groups are owned directly by the hospital, if the group is not owned outright, than the group is beholden to the good graces of the hospital, as their dissatisfaction with the ED physicians translates into a lost contract and a new ED group in place. While this is not universally normative behavior I have worked in some facilities where the ED physicians are tracked as too how many Medicare age patient’s they admit to the hospital and they are given admission targets (formally or informally) by hospital administration to ensure that the hospital beds are adequately filled.  

The current payment systems reward you for making the numbers come out right and that process is much more important than making the ‘right’ diagnoses. That is, errors in which the diagnosis is wrong and are felt in retrospect to be much more benign are accepted, since even one mischaracterized patient to have a benign chest pain, and who is admitted and than found with further study to have acute coronary syndrome can potentially make the hospital quality statistics fail to meet current standards since the rate of compliance for core measures for ACS has become essentially 100%. That 100% is at the price of at times massive over treatment of benign chest pain syndromes.  

Thirdly it justifies a Cath first strategy, often times directly from the emergency department, as well as lucrative outpatient advanced imaging studies and a think-second approach is adopted as these ED studies are fully reimbursed. 

Fourthly, since 25% of all malpractice claim payouts in Emergency Medicine are related to missed myocardial infarction, most emergency physicians assume the worst and admit all chest pain patients for a formal ‘Rule-out’ admission. As a consequence this force also dilutes the mix of real to non-real coronary syndromes. 

No one ever criticizes my referring emergency medicine MD's when make a diagnosis of acute coronary syndrome, while admitting obvious cases of costochondritis (rib-sternum arthritis), hyperventilation syndrome, or panic attack. Their guideline -evidence based treatment than contributes to the denominator of the appropriate quality metrics, even though in a more deliberative age a more accurate diagnosis, would not be at the economic disadvantage it currently faces. The clinical rule has become 'Not all diagnoses are created equal' and the higher the risk of catastrophic outcome, the more serious the prognostic implications of the diagnosis than the more likely that diagnosis will be utilized for the reasons I have enumerated above. 


If your looking for Nosological purity the only relevant high quality dataset probably belongs to Medicine Sans Frontier or Partners In Health. Their remuneration is not tied to the perverse 'grade inflation' I see at work every day.