Outcome measures in Clinical Practice

							Outcome measures in Clinical
Practice
Hugh Sanderson
Outcome measures
 Physical, social and mental health
 Health care equation
 Standardisation
 Specific measures
 Generic measures
 Routine data
 Specific data
 Issues around measurement
WHO
 Physical health… deviation from
                physiologic norms


 Mental health…. anxiety, distress


 Social health…… ability to perform
          normal social roles
Healthcare Equation

Condition + Intervention = Outcome

( outcome is the difference between the
  actual and expected state of health)
                Resource



Need   Demand     Use


                Outcome
Standardising the input
  Randomised controlled trial
  Risk adjustment through severity scoring
     Possum, Cancer Staging
     Casemix adjustment
        HRGs/DRGs
        Disease Staging
  Standardisation ( age/ sex/ social class etc)
Specific measures
  Single dimension/ single condition
     Raised liver enzymes
     Raised blood pressure
     Cancer staging
     Disease staging
     Arthritis score
  Multiple dimension/ single condition
     Oxford Hip score
     Visual disability score
Generic Measures
  Rosser and Kind Disability / distress
  Karnovsky scale
  Short form (SF) 36
  Patient experience surveys
Using Existing Routine Data
  Complications
  Adverse Events
  Deaths within 30 days
  Readmissions
  Delayed discharge
  Delays in operation
                        Oesophagectomy death rates and annual number of cases.
                                          England 1997/98

             1.2

               1

             0.8
death rate




             0.6
                                                                            Series2
                                                                            Linear (Series2)
             0.4

             0.2

               0
                    0       20      40         60          80   100   120
             -0.2
                                         number of cases
Specific data
  Cancer Data sets
  Myocardial Infarction National Audit Project
  (MINAP)
  Central Cardiac Audit Database(CCAD)
  Confidential Enquiry into Peri-operative
  Deaths (CEPOD)
  Breast Cancer Screening Audit Database
  Cancer registration
  NHS CR flagged records
Cancer Minimum dataset

 Demographics       Pathology
 Referral           Chemotherapy
 Imaging            Radiotherapy
 Diagnosis          Palliative care
 Cancer care plan   Clinical trials
 Staging            Clinical Status
 Surgery            Death
Staging

 Final pre-treatment T category
 Certainty factor for T category
 Final pre-treatment N category
 Certainty factor for N category
 Final pre-treatment M category
 Certainty factor for M category
 Overall pre-treatment stage group
 Certainty factor for TNM stage
 Site Specific Classification
 Overall Pathological TNM stage grouping –
 integrated stage
Site specific classification
 e.g.
  Dukes for colorectal,
  NPI for breast,
  “Limited’ or ‘Extensive’ stage disease for
  Small Cell Lung Cancer
Issues around measurement
  Validity
  Repeatability
     between observers
     within observers
  Measurement biases
     Terminal digit bias
     Observer bias
  Practicality
  Ethics
  Process control analysis
Outcome measures
 Physical, social and mental health
 Health care equation
 Standardisation
 Specific measures
 Generic measures
 Routine data
 Specific data
 Issues around measurement