Health Information Management

Know how APR-DRGs can assist with physician profiling

JustCoding News: Inpatient, May 9, 2012

Want to receive articles like this one in your inbox? Subscribe to JustCoding News: Inpatient!

Although MS-DRGs have stolen the spotlight since CMS implemented them in 2007, hospitals are increasingly using All Patient Refined DRGs (APR-DRG) to compile the most accurate assessment of patient severity of illness (SOI) and risk of mortality (ROM). 


APR-DRGs are the most widely-used SOI and ROM-adjusted DRGs in the healthcare industry. They are driven not only by the principal diagnosis (PDX) and any procedures performed, but also take into consideration all secondary diagnoses. MS-DRGs, in comparison, are driven by the PDX, procedures, and the existence of only one CC or MCC secondary condition.

“The important piece [with APR-DRGs] is that all secondary diagnoses are captured if they meet the secondary diagnosis criteria of being monitored, evaluated, treated, extending length of stay, or increasing the nursing resources,” said Cheryl M. Manchenton, RN, BSN, senior inpatient consultant at 3M Health Information Systems in Atlanta. 3M created APR-DRGs in 1990 in conjunction with the National Association of Children’s Hospitals and Related Institutions.

Understand the scores

As with MS-DRGs, APR-DRGs are divided into Major Diagnostic Categories. However, APR-DRGs are adjusted to better reflect a patient’s true clinical picture, which makes them extremely valuable for data analysis, said Manchenton.

Before an APR-DRG is calculated, each diagnosis receives an ROM and SOI score as follows:

  • 1: Minor
  • 2: Moderate
  • 3: Major
  • 4: Extreme

Each diagnosis may not have the same ROM and SOI score. For example, the ROM score for acute cholecystitis is 1 (low risk of mortality), but the SOI score is 3 (significant organ decomposition).

“You can come in with acute cholecystitis and be extremely ill, but your risk of mortality is extremely low, at least in the Western world, because we have antibiotics for that, or worst case scenario, we have surgery,” said Manchenton.

Likewise, a secondary diagnosis of hypotension has an ROM score of 2 and an SOI score of 1.

Sometimes, ROM and SOI scores change commensurate with the progression of a particular disease process, said Manchenton.

Consider the following ways in which SOI scores change as a secondary diagnosis of diabetes progresses through various stages:

  • Uncomplicated diabetes (ICD-9-CM code 250.0x): SOI 1 (minor)
  • Diabetes with renal manifestations (ICD-9-CM code 250.4x): SOI 2 (moderate)
  • Diabetes with ketoacidosis (ICD-9-CM code 250.1x): SOI 3 (major)
  • Diabetes with hyperosmolar coma (ICD-9-CM code 250.2x): SOI 4 (severe)

Consider the following ways in which ROM scores change as a secondary diagnosis of cardiac dysrhythmias progresses through various stages:

  • Premature beats (427.60): ROM 1 (minor)
  • Sinoatrial node dysfunction (427.81): ROM 2 (moderate)
  • Paroxysmal ventricular tachycardia (427.1): ROM 3 (major)
  • Ventricular fibrillation (427.41): ROM 4 (severe)

APR-DRGs and specificity

ROM and SOI scores for each diagnosis drive the specific APR-DRG assigned. APR-DRGs with high ROM and SOI scores generally denote multiple serious diseases as well as the interaction among those diseases, said Manchenton. “You have a much more complex patient who would be more ill and more likely to expire,” she adds.

Unlike MS-DRGs, APR-DRGs reflect the totality of a patient’s clinical picture because ROM and SOI scores change as a patient’s condition becomes more clinically complicated—even when documentation isn’t as specific as it could be, said Manchenton.

Consider the following example: A patient has a PDX of viral pneumonia and no secondary diagnoses. This case maps to MS-DRG 195 and APR-DRG 139 (SOI 1; ROM 1). If the same patient has a secondary diagnosis of congestive heart failure (CHF), unspecified, the case would map to the same DRGs; however, the ROM and SOI scores associated with APR-DRG 139 would increase from 1 to 2.

Thus, even when the physician doesn’t specify whether the CHF is acute or chronic, the APR-DRG captures the fact that the patient has a secondary condition that complicates his or her care, said Manchenton. If the physician had also documented that the same patient had additional secondary diagnoses of malnutrition, hypotension, and acute respiratory failure, the ROM and SOI scores associated with APR-DRG 139 would increase to the highest level (4), she said.

Obtaining more thorough documentation significantly impacts APR-DRG weights, said Manchenton. Consider the following example: A patient with a PDX of CHF, unspecified, also has secondary diagnoses of chronic obstructive pulmonary disease (COPD) and atrial fibrillation. This case maps to MS-DRG 293 (heart failure and shock without CC/MCC), relative weight 0.7220. It also maps to APR-DRG 194 (heart failure), relative weight 0.7035, ROM score 2, SOI score 2.

If the physician also documents that this same patient has additional secondary diagnoses of acute respiratory failure, acidosis, decubitus ulcer of the heel, severe protein-calorie malnutrition, and cardiogenic shock, the case would map to MS-DRG 291 (heart failure and shock with MCC), relative weight 1.4601. Although the APR-DRG wouldn’t change, the ROM and SOI scores would increase to 4, and the relative weight would more than triple (2.3149).

“Sometimes, the physician is so busy discussing the congestive heart failure and the shortness of breath associated with it that he or she is not giving us the full diagnosis related to the respiratory component,” said Manchenton. “What is the CHF doing to the lungs? If we had gotten that specified—if it was clinically appropriate to be acute respiratory failure—we would be in the highest SOI and ROM subclasses.”

Using ROM and SOI to your advantage

Coding managers and CDI specialists can use these scores as part of a larger initiative to profile physicians and ultimately improve care and ensure accurate reimbursement, said Tamara A. Hicks, RN, BSN, MHA, CCS, CCDS, ACM. Hicks is the manager of care coordination at Wake Forest Baptist Medical Center in Winston-Salem, N.C.

Profiling refers to an analysis of practice patterns using discharge data, such as SOI and ROM scores, to assess physician performance. Hospitals can use other discharge data for this purpose as well, including patient safety indicators, potentially preventable conditions, hospital acquired conditions, and PEPPER, said Hicks.

Many agencies, such as CMS, The Joint Commission, peer review organizations, payers, and others, already use discharge data to profile physician performance, said Hicks. Consumers are also becoming increasingly aware of publicly-available data that ranks performance. They use this data to make more informed healthcare decisions, she adds.

Hicks said Wake Forest Baptist Medical Center has used discharge data in the following ways:

  • Initiate process improvement. Each month, a mortality committee meets to discuss deaths and seeks to understand the balance between quality issues and documentation issues. The physician-led committee includes those from the quality assessment department as well as physicians and representatives from various specialties and disciplines. In particular, the committee helped identify certain physicians who now review all deaths that occur due to acute myocardial infarction, pneumonia, and CHF before final coding to ensure that the code assignment reflects the patient’s total clinical picture.
  • Focus CDI reviews. CDI staff use expected mortality rates to focus chart reviews on patients with an ROM or SOI score of less than 4 who expire. Staff review records after coding but before the bill is dropped to assess for any opportunities to better reflect the patient’s SOI and ROM in the documentation.
  • Provide physician education. CDI staff use actual ROM scores and examples to educate physicians about how better documentation can affect profiling data.

In addition to being mindful of discharge data, Hicks said coders and CDI specialists should:

  • Capture all secondary diagnoses that meet reporting requirements—especially those that aren’t related to the PDX
  • Improve specificity of secondary diagnoses
  • Collaborate with the core measures team and quality department
  • Ensure accurate assignment of the POA indicator

The record is what truly tells the story, and it should include all details about the patient’s complexity. This is helpful not only today, but also in preparing for ICD-10, said Manchenton. “When your record is complete, accurate, and compliant, you get an accurate profile and an appropriate reimbursement,” she said. “When you start focusing on profiling and getting better specificity, you’re already getting a jumpstart on your ICD-10 preparation.”

Editor’s note: Lisa A. Eramo is a freelance writer and editor in Cranston, R.I., who specializes in healthcare regulatory topics, health information management, and medical coding. You may reach her at

Want to receive articles like this one in your inbox? Subscribe to JustCoding News: Inpatient!

Most Popular