TIP OF THE WEEK
Risk Adjustment, Odds Ratios, and Acute Care Hospitalization
In the next two weeks, the Tip of the Week will focus on the acute care hospitalization utilization outcome measure. This week, let's take a closer look at the acute care hospitalization outcome measure and how it is risk-adjusted, using a comprehensive approach.
A comprehensive approach addresses not only the potential clinical factors but also the potential system factors that can lead to success or failure. The following diagram depicts the process to be followed to ensure that all potential factors are considered for problem or strength identification.
RISK ADJUSTMENT
The intent of risk adjustment is to provide a level playing field so that comparisons take into account important factors that can influence outcomes. For example in home healthcare, improvement and stabilization in functional activities may be affected by numerous factors including pre-existing conditions, age, the presence of a caregiver, cognitive functioning and others. Risk adjustment models, then, look at the outcome taking into account the factors (OASIS items in this case) that were expected to affect the outcome.
The risk adjustment model for the acute care hospitalization outcome measure varies from the other quality measures in that improvement in this category is identified with a decrease in outcome percentage (equaling a decrease in utilization or less acute care hospitalizations), as opposed to, for instance, the increase in outcome percentage desired when trying to improve stabilization in bathing.
ODDS RATIOS
Within the different risk adjustment models for each quality measure, odds ratios are reported for each OASIS item. Odds ratios are values that indicate the likelihood that the outcome will occur, taking the OASIS predictor and other predictors into account.
For example, one of the OASIS indicators for acute care hospitalization is “patient lives alone” with an odds ratio of 1.27. Thus, identifying that the patient lives alone means that a patient is 1.27 times as likely be hospitalized than if he did not live alone, taking into account all the other OASIS predictors. Another way to say this is that a patient that lives alone is 27 percent more likely to be hospitalized. Look also at the presence of a urinary catheter prior to the last two weeks. With an odds ratio of 1.54, the patient who has had a urinary catheter prior to the last two weeks is 1.54 times more likely to be hospitalized, or 54 percent more likely to be hospitalized than if she did not have a urinary catheter within the last two weeks.
Odds ratios of 1.0 are even odds--odds ratios close to 1.0 are close to even odds. For odds ratios less than 1.0, the interpretation is based on 1.0 minus the odds ratio. Using acute care hospitalization again and examining the effect of having surgical wounds present, an odds ratio of 0.58 means that a patient with a surgical wound is less likely to be rehospitalized than if he did have a surgical wound present.
Remember that multiple odds ratios are not compounding. For example with this outcome, a patient with blood disease (odds ratio 1.36) and skin/subq tissue disease (odds ratio 1.31) does not have an odds ratio that is 2.6.
DRILLING DOWN
Using the odds ratios, agencies can identify which OASIS items are most associated with the acute care hospitalization outcome. One common question is where to set the thresholds--one approach is to use a 25 percent rule of thumb where odds ratios for OASIS indicators less than 0.75 and more than 1.25 are used to identify the most influential factors.
Using the same example for acute care hospitalization and the 25 percent threshold, the following items would be identified as influential in this outcome:
| Odds Ratios | |
| 1.28 | Inpatient discharge from hospital |
| 1.54 | Urinary catheter prior to past 2 weeks |
| 1.53 | Stage 3-4 pressure ulcer(s) present (0-1) |
| 1.27 | Status of most problematic statis ulcer (0-3) |
| 0.58 | Surgical wounds present (0-1) |
| 1.47 | Urinary catheter (0-1) |
| 1.44 | Bowel ostomy (0-1) |
| 1.84 | Acute condition: mental/emotional (0-1) |
| 1.32 | Acute condition: oxygen therapy (0-1) |
| 1.71 | Acute condition: IV/Infusion therapy (0-1) |
| 1.29 | Acute condition: enteral/parenteral disease (0-1) |
| 1.29 | Diagnosis: infectious/parasitic disease (0-1) |
| 1.50 | Diagnosis: neoplasms (0-1) |
| 1.36 | Diagnosis: endocrine/nutritional/metabolic (0-1) |
| 1.36 | Diagnosis: blood diseases (0-1) |
| 1.31 | Diagnosis: skin/subcutaneous diseases (0-1) |
| 1.29 | Diagnosis: other injury (0-1) |
| 0.46 | Length of stay: more than 31 days (0-1) |
These items give agencies some indication of the kinds of patients who are more likely to be hospitalized. Identifying high-risk patients (upon admission) for hospitalization is a first step. Patients with a history of prior hospital admissions have a greater likelihood of being rehospitalized. Consider the possibility at start of care of asking patients about the frequency of their hospital admissions. Upon intake, identify these influencing factors and, if appropriate, initiate a care plan with higher visit frequencies at the start of care. Agencies may also consider a care path/clinical pathway that is disease/condition-specific.
Next week, the Tip of the Week will address case mix and system factors that can impact agency outcome measures.
RESOURCE SPOTLIGHT
Overview of Risk Adjustment and Outcome Measures for Home Health Agency OBQI Reports: Highlights of Current Approaches and Outline of Planned Enhancements
This document provides additional details about outcome measurement, risk adjustment, and the rationale for risk adjusted outcome reporting in the context of outcome-based quality improvement (OBQI) for home health care. It also summarizes the current risk adjustment and outcome measurement approaches as well as plans for their continued refinement.
IN THE NEWS
FDA CLEARS GLOVE MADE FROM NEW TYPE OF LATEX
April 23, 2008 -- FDA -- The U.S. Food and Drug Administration has cleared for marketing the first device made from a new form of natural rubber latex, guayule latex. The product is derived from the guayule bush, a desert plant native to the Southwestern United States. Read more >>
$300 FEDERAL REBATE IS OPEN TO NURSING-HOME RESIDENTS
April 22, 2008 -- Cleveland Plain Dealer -- With a pen and a 1040A form from the Internal Revenue Service, you can help nursing home residents on Medicaid to get $300 of spending money with the federal tax rebate. Read more >>
SENIORS NO STRANGERS TO HAPPINESS
April 22, 2008 -- HealthDay -- Older Americans are quite satisfied, while Baby Boomers are the least content, survey shows. Surveys of Americans taken since 1972 suggest that plenty of older people are quite happy, with more than half of black men and women over the age of 80 saying they're "very happy," with older white men and women following suit. Read more >>
CMS PROPOSES MORE ACCURATE PAYMENTS FOR INPATIENT REHABILITATION SERVICES IN FY 2009
April 21, 2008 -- CMS -- Due to statutory requirements in the Medicare, Medicaid, and SCHIP Extension Act of 2007 (MMSEA), the Centers for Medicare & Medicaid Services (CMS) will apply a zero percent increase factor for Inpatient Rehabilitation Services (IRF) payment rates for FY 2009. As a result, CMS expects to spend roughly $5.6 billion in FY 2009 and approximately $30 billion in the next five years for IRF services. Read more >>
CMS RELEASES NEW QUALITY REPORTING OPTIONS FOR PHYSICIANS
April 17, 2008 -- CMS -- CMS released new reporting options, including registry-based reporting, to make it easier for physicians to participate in the Physician Quality Reporting Initiative (PQRI). Read more >>
OHIO KEPRO EVENTS
OBQI TRAINING
The Outcomes-Based Quality Improvement (OBQI) Process is a systematic, data-driven method of investigating patient health outcomes and developing improvement strategies that are tailored to suit your agency’s unique population.
Join us for a refresher course in your area. We are offering two courses:
- 9:00 to 12:00 -- OBQI Basics
- 1:00 to 4:00 -- Beyond the Basics
| May 15 | Summa Home Health at St. Thomas Hospital, Akron |
| May 20 | Mount Carmel Home Health, Columbus |
| May 22 | St. Rita’s Hospital Home Care Aux. Conference Center, Lima |
| May 28 | Kettering Network Home Health, Miamisburg, Ohio |
| June 4 | Union Hospital Auditorium, Dover |
For more information or to register, go to www.ohiokepro.com/providers/homehealth/events.asp.
COMING ATTRACTIONS
Ohio Home Care Organization – Upcoming Education
- May 12 – Palmetto GBA Updates for Hospice and Home Care
- May 13-14 – Home Health Spring Conference
Ohio Council for Home Care – Upcoming Education
- May 13-15 – Breaking the Code – Competency in Diagnostic Coding
Health Policy Institute of Ohio: Regional Meetings on the Ohio Family Violence Prevention Project
The Tip of the Week Archive is available on the Ohio KePRO Web Site.
Clinically driven. Client focused. Value based.