3M™ Potentially Preventable Admissions (PPA)

3M PPA can be useful to payers, employers, government agencies, researchers, integrated health delivery systems (e.g., managed care organizations and accountable care organizations), and residential nursing care facilities (e.g., nursing facilities, intermediate care facilities, and residential treatment centers).

For example, employers or government insurance programs can use 3M PPA to measure the performance of contracted, integrated health delivery systems or residential nursing care facilities. Those integrated health delivery systems and residential nursing care facilities themselves can also use the 3M PPA methodology to enhance their own performance by improving outcomes from year to year.

While 3M PPA classification logic is the same for every client, an organization can use different versions of the methodology and apply their own reimbursement calculation tables. 3M always recommends that clients use the latest version of the software, but each organization can decide how to apply it (for research, public reporting, reimbursement, or a combination). Note: At present, 3M does not offer software that replicates a specific organization’s PPA analysis.

3M experts are available to advise health plans, government agencies and other interested parties on how to obtain maximum value from the 3M PPA. For example, 3M consultants can help you measure the incidence of potentially preventable admissions, compare those findings with appropriate benchmarks, and help design improvement programs. 3M consultants can also help payers and other organizations measure PPAs across health plans and other patient populations, design pay-for-outcomes incentive methods and facilitate shared learning collaboratives.

The 3M PPA methodology can identify PPAs using standard, inpatient claims data derived from institutional and professional claims (i.e., the UB-04 and CMS-1500 paper forms and their corresponding X12N 837 electronic formats). Consistent, unique patient identifiers are essential. Comparing 3M PPA rates across different populations requires the 3M™ Clinical Risk Grouping (CRG) Software to perform risk adjustment by individual health status. Thus, 3M PPA analysis typically involves creation of a static data set comprising at least one full year of data. If available, pharmacy data in NCPDP format is optional but recommended. Note: 3M CRG and 3M PPA methodologies do not need to be built into the claim processing systems.

Every preventable hospital admission represents both a significant cost to the health care system and evidence of an ambulatory care system failure. The challenge has been to define “preventable” with an understanding that the preventability of these admissions depends on the overall burden of illness of the individual patient. In 2012, 3M met this challenge by releasing the 3M™ Potentially Preventable Admission methodology as one of the three Population-focused Preventables. Learn more. (The others are Potentially Preventable Emergency Department Visits and Potentially Preventable Services.)

As with all 3M PPE methodologies, three core concepts are essential. First, we recognize that not all admissions are potentially preventable. Second, what matters is not the individual admission, but rather the overall rate of potentially preventable admissions. Instead of approaching quality with the mindset of “This should never happen,” we use a more realistic and meaningful approach of “This has happened too often.” Third, any comparisons across populations of patients must be risk-adjusted. In practice, that means that the actual 3M PPA experience of a population is compared with the experience that would be expected for a population with the same case mix.

The 3M PPA logic is divided into two phases.

1. Identify patients with potentially preventable initial admissions

All inpatient stays are assigned to a 3M™ All Patient Refined Diagnosis Related Group (APR DRG). For the general population, approximately 41 base 3M APR DRGs are considered potentially preventable. For some 3M APR DRGs, the 3M PPA logic also takes into account individual diagnoses and patient age. In a Minnesota all-payer analysis, for example, the most common 3M PPAs were pneumonia, heart failure and COPD. [www.health.state.mn.us/healthreform/allpayer/potentially_preventable_events_072115.pdf] When an integrated health delivery system (such as a managed care organization or accountable care organization) accepts responsibility for a specific population, then additional 3M APR DRGs are considered potentially preventable, for a total of approximately 100. Examples of additional 3M APR DRGs include bipolar disorder, sickle cell anemia crisis and inflammatory bowel disease. When a population is under the care of a residential nursing care facility (such as a nursing facility, intermediate care facility or residential treatment center), additional 3M APR DRGs are considered potentially preventable, for a total of approximately 126. For example, 3M analysis showed that 3M APR DRG 720, Septicemia and Disseminated Infections, was the most common 3M PPA in a population of more than 400,000 Medicare nursing facility patients. [https://www.liebertpub.com/doi/10.1089/pop.2018.0065]

2. Determine patient risk adjustment

In any rate-based comparison of outcomes, risk adjustment is essential to a fair comparison across populations. Although 3M PPAs are generally preventable, they will never be totally eliminated, even with optimal care. As a result, there will be a residual rate of 3M PPAs in even the best-performing systems. More importantly, the rate at which 3M PPAs occur depends on the burden of illness of the population which is measured using 3M™ Clinical Risk Grouping (CRG) Software. For example, 3M CRG 70602 is a person with congestive heart failure, diabetes and chronic obstructive pulmonary disease, severity 2. Hospital admissions for this person would be more likely to be preventable than for a person in severity 5 (i.e., 3M CRG 70605).

Further information on the 3M PPA logic is available in the Population-focused Preventables Methodology Overview. Detailed information is available to licensees in the online PFP definitions manual.

The 3M PPA clinical logic is maintained by a team of 3M clinicians, data analysts, nosologists, programmers and economists. The methodology is updated annually to reflect changes in the standard diagnosis and procedure code sets as well as 3M enhancements to the clinical logic.