Documents
Resources
Learning Center
Upload
Plans & pricing Sign in
Sign Out

Advantages - DOC

VIEWS: 172 PAGES: 2

									                                                    APPENDIX E
                                              Data Sources Pros and Cons

        Data Source                               Advantages                                        Limitations

       Mortality Data       Timely                                                Relatively rare event
                            Readily available and free                            75% occur in elderly where diagnosis is least
          (Outcomes)        Population-based                                      clear
                            Long-term historical data available                   Usual death certificate limitations
                            Can compare to other states and nationally
   Behavioral Risk Factor   Population-based                                      Based on self-report
Surveillance System (BRFSS) Ten years of lifetime prevalence data                 Confined to adults (children asked about one
                            Now includes data on current asthma                   year) with telephones
         (Prevalence)       Performed annually                                    Misses frequent movers, cell phone only users
                            Includes risk behavior information                    Expensive to add questions
                            Could use as basis for call backs for more detailed   Relatively small sample size makes it difficult to
                            information                                           draw inferences about sub-groups
   Hospital discharge data  Potentially preventable outcome                       No information on race/ethnicity
                            Covers 85% of hospitals and 95% of discharges         No individual identifiers so # of events only
          (Outcomes)        Collected on a standardized form (UB 92)              Validity concerns
                            Zip code data available                               No information on out-of-state hospitalizations
                            Charge data available (not reliable)                  Gender codes may be unreliable
                            Predictive value reasonably good                      No data on federal hospitals (VA, IHS)
                                                                                  Misses snowbirds and residents who receive
                                                                                  care out of state
                                                                                  Bias towards DRG codes with highest
                                                                                  reimbursement
   Emergency room data         Potentially preventable outcome                    Statewide data only available recently
                               Covers 85% of hospitals and 95% of discharges      No information on race/ethnicity
        (Outcomes)             Zip code data available                            No individual identifiers so # of events only
                               Charge data available                              Validity concerns
                                                                                  Mixes outpatient, UC, ED visits
                                                                                  Misses snowbirds and residents who receive
                                                                                  care out of state




                                                              1
                                                  APPENDIX E
                                            Data Sources Pros and Cons

       Encounter data        Underlying population can be described so rates    Owned by health plans (MDH data from
                             can be calculated                                  Medica, Blue Cross Blue Shield, Minnesota
                                                                                Department of Human Services)
         (Incidence)         Pharmacy and procedure data available (but can't   has no identifiers
        (Prevalence)         distinguish no Rx from no Rx benefits)             Datasets not uniform across health plans
         (Outcomes)          Zip code of subscriber                             Primary purpose of data is for billing, validity
            (CQI)                                                               not fully evaluated
                                                                                Excludes those without insurance
                                                                                Data may be incomplete or biased
                                                                                Lone dx code (493) used for asthma- may be
                                                                                both over- and under-used
         School data

   Minnesota Student Survey MSS- comprehensive survey of high school-aged      Question asks about asthma and other chronic
                             children.                                          conditions
   School District specific
                                                                                Data on asthma in school children comes from a
                                                                                variety of different sources in different districts,
                                                                                not compiled systematically. Not all schools
                                                                                have school nurses.




                                                             2

								
To top