Background for Elixhauser Comorbidity Index:
In December of 2002, the US Department of Health and Human Services (HHS) and the Centers for Medicare and Medicaid Services (CMS) in a collaboration with the American Hospital Association, the Association of American Medical Colleges, and the Federation of American Hospitals, formed the Hospital Quality Alliance (HQA). The HQA set out with the mission to provide consumers with information regarding provider and/or hospital performance, allowing the consumer to make an informed choice as to which provider and where to receive their care. Over time, legislation formed surrounding the HQA data, forming the policies found in the quality and pay-for-performance provisions embedded in the Affordable Care Act (ACA) in 2010.
CMS developed a website, Care Compare, that publicly reports hospital performance and rates the hospital’s performance by assigning them to one of three performance categories:
- “Better than the National Rate”
- “No Different than the National Rate”
- “Worse than the National Rate”
When CMS set out to score a hospital’s performance, they had to consider differences in the patient populations, so for each of the measures, CMS included an adjustment for what are called risk-adjusted factors, such as age, comorbid disease, and indicators of patient frailty.
For each patient, risk-adjustment variables (or risk-adjusted factors) are obtained from inpatient, outpatient, and physician Medicare administrative claims data extending 12 months prior to the index admission and all claims for the index admission itself.
Then, using statistical analysis, the hospital’s data is risk-adjusted and an estimated death rate called a risk-standardized mortality rate (RSMR) is calculated for each performance category (AMI, COPD, CHF, PNA, CVA, post-CABG). To categorize hospital performance, CMS compares each hospital’s RSMR to the national observed mortality rate.
Elixhauser Comorbidity Index identifies pre-existing medical conditions (risk factors) that would result in an impact on certain outcomes, such as mortality. Coding these pre-existing medical conditions allows the data to be risk-adjusted and therefore impacts the expected mortality ratio for the patient.
If these diagnoses are not captured in the documentation, then there may be a poor representation of the expected mortality.
Clinical Example:
H&P: 55 y/o male presented to ER with vomiting for 3+ days. Patient states he often drinks 6-7 beers and a couple shots of hard liquor in a day and has done so for 10+ years. Patient states he feels weak and irritable.
Admitting Principal Diagnosis: AKI
PMH: CAD, CKD
Vitals: Initial: Temp 98.9, HR 112, RR 13, BP 104/75, 97% on room air. EKG shows sinus tachycardia.
PE: Patient appears thin and in moderate distress, lungs clear, abdomen soft, normal bowel sounds, no cardiac abnormalities noted with auscultation.
Labs: Admission labs as follows:WBC 8.2,hgb 8.4, hct 23%, AST/ALT ratio> 2, Potassium 3.0. Vitamin B6 Profile shows PLP of 6; folate 2ng/mL. Crt 3.2 with GFR 22 on admission, f/u Crt on day 1 of admission 2.2, GFR 34. Lactic acid 2.4 on admission; repeat LA after IV bolus: 1.5. Baseline labs from 4 months: Crt 2.1, GFR 37
Radiology: liver US shows hepatomegaly and increased echogenicity; EGD with normal mucosa, no evidence of overt gastritis.
Treatment: CIWA protocol using prn Ativan, IV fluid bolus of 500mL banana bag in the ER then 150mL/hr once admitted. IV Zofran q 4hr prn, IV Reglan q 12hr, telemetry, and nutritional consult. Potassium supplementation, Boost with all meals. Vitamin B complex and Fe supplementation daily po. Liver panel and liver US. EGD.
Documentation: Dietary consult notes state that the patient has had weight loss, but does not meet the criteria for malnutrition and recommends dietary supplements once the patient can tolerate PO. H&P states, “AKI likely related to poor po intake and dehydration—elevated crt from baseline of 2.1 per chart review. Patient with low hgb and hypokalemia. Poor dietary—likely nutritional etiology—RD consult. Liver panel elevated, will f/u with US. EGD to assess for etiology of N/V as no evidence of viral gastroenteritis. Pt on home BP meds but with low BP in ER—will hold meds for now.” Per DCS, “Patient with AKI, prerenal related to dehydration from vomiting-returned to baseline and stable. Acidosis resolved with fluids. No evidence of pathological abnormality for etiology of nausea—likely related to chronic alcohol consumption. No w/d symptoms during this admission. Counseled patient on ETOH cessation. Low Hgb likely also due to poor nutrition. Patient to continue with Vit B and Fe supplementation and follow up with PCP. Resume home BP meds.”
Discharge Dx: AKI, acidosis, dehydration, hypokalemia and CAD.
Question: Are there query opportunities based on the scenario stated above?
Discussion: While this is a relatively simple case that appears to be maximized for relative weight, SOI, and ROM, and there are several query opportunities for diagnoses that are often overlooked, but have impact on the AHRQ Elixhauser Comorbidity Index. The patient exhibited the following risk factors and clinical indicators:
- Heavy drinking history and placed on CIWA
- Low Hbg/Hct
- Radiology evidence of fatty liver disease
- RD notes weight loss
- Elevated baseline creatinine
- On home BP meds
UASI Recommends:
- Query to clarify alcohol use/abuse
- Query for nutritional anemia
- Query for alcohol fatty liver
- Query for abnormal weight loss
- Query for CKD and stage
- Query for HTN/dx associated with ‘home BP meds’
Documentation without Clarification:
- Principal Diagnosis: AKI N179
- Secondary Diagnoses: acidosis E872, dehydration E860, hypokalemia E876, CAD I12510
- Procedure: Inspection of Upper Intestinal Tract, Via natural or Artificial Opening Endoscopic 0DJ08ZZ
- Working DRG:683 RENAL FAILURE WITH CC. Relative weight 0.8793. SOI/ROM 2/2.
- AHRQ Elixhauser Score: 11
- Van Walraven Elixhauser Score: 5
Documentation with Clarification:
- Principal Diagnosis: AKI N179
- Secondary Diagnoses: acidosis E872, dehydration E860, hypokalemia E876, CAD without angina I12510, alcohol abuse uncomplicated F1010, nutritional anemia D539, CKD stage 3b N1832, HTN I129, alcohol fatty liver K700, and abnormal weight loss R634
- Procedure: Inspection of Upper Intestinal Tract, Via natural or Artificial Opening Endoscopic 0DJ08ZZ
- Working DRG:683 RENAL FAILURE WITH CC. Relative weight 0.8793. SOI/ROM 2/2.
- AHRQ Elixhauser Score: 26
- Van Walraven Elixhauser Score: 25
| Code | Description | Elixhauser Comorbidity Category | Uses POA indicators for assignment? |
| F1010 | Alcohol Abuse, uncomplicated | Alcohol Abuse | No |
| D539 | Nutritional Anemia, unspecified | Deficiency Anemia | Yes |
| I129 | Hypertensive CKD with Stage 1-4 or Unspecified CKD | Renal Failure, moderate | No |
| K700 | Alcoholic Fatty Liver | Liver Disease, mild | Yes |
| R634 | Abnormal Weight Loss | Weight Loss | Yes |
CDI Educational Tips:
Creatine levels must be stable to appropriately diagnosis CKD. Be cautious as creatinine levels often vary during the admission. When stable, the lowest creatinine level, highest GFR, should be used as the baseline for staging.
This is a short synopsis of a possible patient record and is not intended to be all inclusive. This is for educational purposes only and not intended to replace your institutional guidelines.