Population Health Needs Patterns of Service Usage

1
Session 5
2

National Requirements
Learning from Others
Resources/ Costs
Population Health Needs Patterns of Service
Usage
Where want to be (Strategy)
Where are now
Commissioning Cycle
How get there (Transition Plan)
Review Impact and quality
Putting into operation
3
Program/Methods ForMonitoring/Influencing
Behavior of Physicians

• QUALITY IMPROVEMENT
• PRACTICE GUIDELINES
• DISEASE/CASE MANAGEMENT
• FEEDBACK/PROFILING

4
A Working Definition of Quality of Care

• Quality of care is the degree to which health
  services for individuals and populations increase
  the likelihood of desired health outcomes and are
  consistent with current professional knowledge.

5
Quality Improvement

• HMOs have an advantage in improving quality of
  care
• gtA defined population permits tracking
  population-based outcome measures.
• gtRelatively comprehensive benefit coverage.
• gtCommon medical record (group, staff models).
• gtCommon billing data

6

• Obstacles to Improving Quality of Care
• gtSkepticism of physicians and employers regarding
  organizational commitment to quality improvement.
• gtDifficulty in moving quality improvement efforts
  from the philosophical to the practical level.
• gtLack of MIS capacity needed to support quality
  improvement efforts.
• gtLack of expertise in evaluation of quality
  initiatives and their outcomes.

7
Physician Education

• Little evidence of effectiveness in changing
  behavior when used alone.
• Best employed in conjunction with more
  comprehensive efforts to alter practice patterns,
  such as clinical practice guidelines, profiling,
  and financial incentives.

8
Best Practices Aggregate Goals

• Define aggregate performance targets regarding
  clinical practice or patient outcomes
• Recruit a panel of physicians to develop strategy
  for achieving targets
• gtShare plan data
• gtCollect best practices of other managed care
  organizations
• Implementation
• gtFocus on what is best for patients rather than
  what is convenient for providers
• gtBe aware that change will generate fear of staff
  reductions and therefore resistance
• gtMake goals achievable under most circumstances

9
Quality ImprovementMajor Concerns

• Different explanations for poor plan performance
  can exist cookie-cutter responses are not likely
  to be appropriate.
• The problem is not always a clinical one health
  plan factors and community characteristics may be
  critical.
• Careful fact finding is necessary before data are
  released or corrective actions taken.

10
Clinical Practice Guidelines

• Premise if one can determine best practice for
  a clinical condition, and more physicians can
  learn to practice in this way, outcomes can be
  improved and costs reduced.
• At their best, guidelines represent the medical
  professions knowledge about how to best address
  a clinical problem
• Early clinical practice guidelines were not
  effective
• gtNot written for practicing physicians
• gtDistrust of guidelines written by experts - Not
  coordinated with physician financial incentives

11
Clinical Practice Guidelines(cont.)

• Recent generation of clinical guidelines hold
  greater promise
• gtStructured as tools to assist the physician
  clear that they are not intended to replace
  clinical judgment
• gtDevelopmental process involves local physicians
  and other health providers
• gtClear explanation of scientific basis
• gtEmphasis on implementation as well as
  development
• gtRecognition of need to continually revise and
  update guidelines
• gtCoordination with financial incentives remains
  an issue

12
Impact of Cystitis Clinical Practice Guideline on
Costs and Outcomes of Care

• Guideline developed to reduce variation and
  improve quality of care for women age 18 to 64
  with uncomplicated cystitis.
• Study involves sample of women with cystitis
  attending five primary care practices
• About half of all urinary tract infections seen
  in these practices were eligible for care under
  the guideline.
• Analysis compared 201 eligible cases seen before
  the guideline was implemented to 145 cases seen
  after the guideline was implemented.

13

Impact of Cystitis Clinical Practice Guideline
on Costs and Outcomes of Care

• Results
• gtUse of an antibiotic recommended by the
  guideline rose from 88 of cases to 95 of cases
• gtUse of 3 day course of the antibiotic rose from
  28 of cases to 52 of cases
• gtUse of a urine culture decreased from 69 of
  cases to 38 of cases
• gtVisits managed solely by physicians fell from
  65 of cases to 32 of cases.

14
Impact of Cystitis Clinical Practice Guideline
on Costs and Outcomes of Care

• Results
• gtProportion of cases coordinated primarily by the
  nurse rose from 19 to 57.
• gtSavings per case of 25.69 were 35 of total
  pre-guideline direct costs.
• Statistically significant

15
Feedback/Profiling Techniques

• Internal profiling
• External profiling

16
Feedback/Profiling Techniques

• General Issues
• Peer Comparison
• gtReport performance back to an individual
  physician or physician group along with
  comparative data from others
• gtMeaningful only if physicians agree that the
  comparison group is relevant
• Individual Feedback
• gtProvide an individual physician or physician
  group with data about own performance without
  comparative information. Meaningful only if there
  is an accepted norm or standard otherwise there
  is no anchor for the data

17
Feedback/Profiling Techniques

• Group or Aggregate Feedback
• gtProvide individual physicians with information
  on the performance of the larger group, but not
  on individual performance
• gtMeaningful only if individuals in the group can
  relate their performance to the groups
  performance

18
Retinal ScreeningTarget 90 - 5 points
19
COPDMax 45 points
20
Oxygen PrescribingItems per 1,000 Registered
21
J40-J44 - COPDNon-Elective Last FCEs XYZ PCT
discharge dates between January and December
2003 (Number of Discharges per 1000 List Size)
22
Experience with Feedback

• Individual feedback and peer comparison has been
  more effective than aggregate feedback because
  aggregate feedback has often been used with no
  norm for comparison
• Feedback is most useful in conjunction with other
  efforts
• Feedback is most effective when the objective is
  to increase utilization
• The effect of feedback is at its greatest when
• gtThere is a strong scientific base for
  comparative norms or goals
• gtThe data being fed back to physicians are
  relatively stable over time
• gtThere is case mix adjustment
• gtThere are clear actions physicians can take in
  response to the data (linkage to quality
  management)

23
Characteristics of a Potentially Effective
Profiling Approach

• Severity adjustments are essential.
• Outlier removal is desirable.
• Use more than one benchmark if possible (for
  example, for inpatient comparisons, length of
  stay and average charges).
• For each patient, calculate variance from
  comparison group average.
• gtVariance defined as (actual LOS-average LOS)
• Aggregate severity adjusted categories to
  meaningful service lines that approximate
  medical staff structures.
• Calculate average variance for each physician for
  each service line in which physician exceeds
  threshold volume of patients.
• Determine comparison groups
• gtPeer group at the hospital
• gtPeer group in local competitive area
• gtPeer group in region

24
Groupwork

• Consider COPD
• Which levers/incentives would be appropriate to
  your strategy from yesterday?
• What information/data do you need to access to
  support you in using these?