Making Good Decisions in Medical Coding and DRG Assignment Joel - PowerPoint PPT Presentation

Making Good Decisions in Medical Coding and DRG Assignment Joel Moorhead MD, PhD, CPC

Goals • Review principles of good decision‐making • Identify and eliminate sources of bias • Explore differences of opinion – Hospitals – Insurance companies / audit contractors • View differences as learning opportunities • Consider ways to resolve differences 2

Are we more likely to be killed by … • Falling Airplane Parts or • A Shark 3

Decision‐Making ‐ 1 • Are we more likely to be eaten by a shark or killed by falling airplane parts? – 30 times more likely to be killed by falling airplane parts – Plous, Scott. The Psychology of Judgment and Decision. McGraw‐ Hill Higher Education, 1993. 4

AVAILABILITY • An Event is Judged to be More Likely if it … – Is Easier to Imagine – Is Familiar – Took place recently – Is Highly Emotional 5

ANCHORING AND ADJUSTMENT • ANCHOR = intuitive first impression • ADJUSTMENT = change d/t new info • Effect on decisions – Initial impression often too extreme – Insufficient adjustment to new information • Reluctance to change – Overconfidence in intuitive decisions • “Assumptions are your windows on the world. Scrub them off once in a while, or the light won’t come in.” Isaac Asimov (http://www.goodreads.com/quotes/tag/opinions) 6

Decision‐Making ‐ 2 • Mental shortcuts (heuristics) – We may be more likely to use a code if we: • Use that code frequently – But we may overlook important differences in current situation • Thought of that code first – But further analysis might support different coding approach – Mental shortcuts appeal to our “gut” instincts • Uncritical use of shortcuts → overconfidence – Factors important to good decision‐making • Intuition • Analysis ‐ unbiased examination of each alternative » Kahneman, Slovic, and Tversky, 1982 7

MENTAL SHORTCUTS • RISKS • BENEFITS – Unconscious – Speed of decision‐making – Oversimplify complex – Make information situations manageable – Accuracy of decision – Often reliable and useful depends on accuracy of cues – Often biased – Intuitive appeal leads to overconfidence 8

MENTAL SHORTCUTS CAN INTRODUCE BIAS • BIAS – Systematic error • collecting and interpreting data – Often unconscious – Often consistently in one direction 9

DECISION STRATEGIES • INTUITIVE • ANALYTICAL – Unconscious – Conscious – May not be based on logic – Based on logic – Often not systematic – Systematic – Subjective – Objective – Not easily measured – Measurable – Uses shortcuts uncritically – Reduces bias – Not based on probabilities – Based on probabilities – Very important to making – Very important to making good decisions good decisions 10

Screening • Evaluating a large number of subjects to identify those with a particular set of attributes or characteristics. • http://www.businessdictionary.com/definition/screening.ht • Criteria for “clinical validation” would reasonably be considered “screening” 11

Screening Criteria • High sensitivity – Sensitivity • The ability of a test to identify patients with the disease – The probability of a positive test given that the person has the disease – Use data that apply to groups • Identify cases that require closer attention – Not intended to establish final diagnoses for individual patients 12

Confirming Criteria • High specificity – The ability of a test to identify persons who do not have the disease • Probability of a negative test in persons who are disease‐free – Adds information specific to individual patients • Deductive inference – general to specific – Sherlock Holmes – Designed to make final decisions affecting individuals 13

Predictive Value • Measures ACCURACY of a diagnostic or screening test – Accurate measure of usefulness of a test in diagnosing disease in an individual patient. • Predictive Value ‐ Positive – Percentage of patients with a positive test who actually have the disease • Predictive Value ‐ Negative – Percentage of patients with a negative test who are disease‐free 14

Predictive Value Diagnosis, and Treatment • Predictive value depends on prevalence – Pre‐test probability of disease • C diff colitis – If pre‐test suspicion high for C diff, consider empiric therapy regardless of test results • Negative predictive values for C diff colitis tests are not sufficiently high to exclude disease in patients with high pre‐test suspicion of disease • Surawicz CM et. al. Guidelines for Diagnosis, Treatment, and Prevention of C diff Infections. Am J Gastroenterol 2013 (108):478‐ 498http://gi.org/guideline/diagnosis‐and‐management‐of‐c‐difficile‐ associated‐diarrhea‐and‐colitis 15

Confirmation • Citing ways that individual conforms to screening criteria may still be screening if no evidence of deductive reasoning • Confirmation requires analysis of – Mitigating factors – Ways that clinical indicators specific to the individual affect interpretation of criteria • Decision based on “Weight of Evidence” 16

Weight of Evidence • Respected methodology; basis for – Meta‐analysis – “More likely than not” legal determinations • All evidence as a whole may justify a conclusion ... – ... that none of the individual pieces of evidence alone can justify. • Melnick, M., The weight of the evidence ‐ or ‐ More likely than not. Journal of Craniofacial Genetics, 1986. 6 : p. 203‐206. • Edwards, A., et al., Judging the 'weight of evidence' in systematic reviews: Introducing rigour into the qualitative overview stage by assessing signal and noise. Journal of Evaluation in Clinical Practice, 2000. 6 (2): p. 177‐184 17

Clinical Validation • "Clinical validation is performed by a clinician (RN, CMD or therapist).” – “Clinical validation is beyond the scope of DRG (coding) validation, and the skills of a certified coder.” • “This type of review can only be performed by a clinician or maybe performed by a clinician with approved coding credentials.” – 2013 CMS Statement of Work for the RAC ‐ DRG Validation vs. Clinical Validation 18

Validity • Concurrent / Criterion validity – Correlation between one measure and another that is assumed to be superior. • “Gold standard” • Coyne KD et. al.; Heart Lung 1998;27:263‐73 19

Whose opinion is “superior?” Hospital Auditor • • Physician makes clinical Auditor disagrees with diagnosis physician’s diagnosis – History, physical exam, – Non‐physician who never diagnostic testing examined or treated the patient • Multiple physicians agree with • Auditor makes diagnosis diagnosis – based on criteria chosen by – supported by clinical the insurance company indicators in the EMR – Without confirming – consistent with published methodology medical literature 20

GOLD STANDARD • Error‐free identification of true status • Most errors in measuring test discrimination can be traced to problem of learning the true state of the patient 21

How do we decide? • Unreasonable to assume that auditor’s opinion is “truth.” • Unbiased way to resolve conflict is needed. 22

Valid Conflict Resolution • Impartial third party review – ALJ Hearing is credibly impartial and valid – Vendor under contract to insurance company is not credibly impartial • Obvious potential for bias • Agreement between hospital and audit contractor – Resolution of conflict by mutual agreement 23

The Goal of Coding • The most ACCURATE and SPECIFIC codes that are SUPPORTED by – Medical record documentation and – Coding guidelines • What this goal accomplishes – Accurate numerical representation of … • Severity of illness • Resources required to care for the patient 24

Isn’t Accurate and Specific Coding Everyone’s Goal ? • Good data is good for everyone – Physicians – Hospitals – Coders and coding consultants – Auditors • Quality Improvement Organizations • Insurers and Audit Contractors – The Feds 25

Accurate Coding • Not arbitrary – Arbitrary • Not bound by rules • Unreasonable and unsupported • Not capricious – Capricious • Erratic; inconsistent • Subject to change without reason • Not biased – Bias • Systematic error • Not over‐coded • Not under‐coded 26

Decisions by Auditors • Bias ‐ Systematic errors – Making final decisions based on screening criteria ... • Without credible analysis of clinical indicators ... – specific to the individual and ... – outside of internal (screening) “criteria” ... » affecting probability of disease in the individual patient • Arbitrary – Final decisions based on internal “criteria” or published criteria different from hospital‐cited criteria but not more “authoritative” • Without opportunity for discussion or fair hearing • Capricious – Criteria inconsistent between auditors and review organizations – Subject to change without explanation or discussion 27

Back to Basics 28