Information Technology Reference
In-Depth Information
PatIent rIsk measures
Because some patients are at higher risk compared to others, differences in patients can mask differ-
ences across healthcare providers. In many cases, higher volume is correlated with better outcomes.
However, higher volume can also mean a higher proportion of higher risk patients. If this is the case,
then comparing just volume to outcome may not end up statistically significant. (Enzinger et al., 2007)
Sometimes, the attempt at risk adjustment is fairly simple, defined by just using a count of co-morbid
conditions.(Shugarman, Bird, Schuster, & Lynn, 2007) However, it is not usually given just how the
co-morbid conditions were decided upon. Other choices for co-morbid conditions will result in differ-
ent risk adjustment values.(Cerrito, 2006) Still other studies just give a very vague idea of how patient
co-morbidities are accounted for in the study without providing specific codes that are used to define
them.(Fang et al., 2006) Occasionally, the paper will give a reference to a defined severity index without
going into much detail.(Donald H Taylor & Hoenig, 2006; Ellis et al., 1996)
In a study by Schlenker, et. al., a list of 37 different risk factors was defined; most by using a 0-1
indicator function, but others on a scale of 0-5, including urinary incontinence severity and disability in
ambulation.(Schlenker, Powell, & Goodrich, 2005) Treatments such as the use of oxygen and the presence
of a urinary catheter are also included. This list is derived from the OASIS, or Outcome and Assessment
Information Set that is routinely collected by home health agencies. The measures, as defined through
a consensus of experts, include the following: (Anonymous-OASIS, 2007)
•
Three measures related to improvement in getting around:
◦
Percentage of patients who get better at walking or moving around
◦
Percentage of patients who get better at getting in and out of bed
◦
Four measures related to meeting the patient's activities of daily living:
Percentage of patients who have less pain when moving around
•
◦
Percentage of patients whose bladder control improves
◦
Percentage of patients who get better at bathing
◦
Percentage of patients who get better at taking their medicines correctly (by mouth)
◦
Two measures about how home health care ends:
Percentage of patients who are short of breath less often
•
◦
Percentage of patients who stay at home after an episode of home health care ends
◦
Three measures related to patient medical emergencies:
Percentage of patients whose wounds improved or healed after an operation
•
◦
Percentage of patients who had to be admitted to the hospital
◦
Percentage of patients who need urgent, unplanned medical care
◦
Percentage of patients who need unplanned medical care related to a wound that is new, is
worse, or has become infected
Unfortunately, it is difficult to determine the magnitude of “better” without more specifics.
Daley, Iezzoni, & Shwartz (2003) suggests that we can use the following strategies to find potential
risk factors to include in a patient severity model:
•
Published reports from randomized trials and clinical studies
•
Ask clinical experts or panels of practicing clinicians
Search WWH ::
Custom Search