spacer
Dalhousie University PHRU Dalhousie University Faculty of Medicine spacer
spacer
   
 
spacer
spacer
spacer
spacer
spacer
spacer
spacer
spacer
spacer

Concept Dictionary


The following is a brief explanation of the variables that are found in the PHRU data sets. For more information on these and other variables, send an email with your query to PHRU@dal.ca

  • Patient Age

  • Age should be calculated in all databases.
    • In Medicare and Hospital databases, AGE=int((Date-DOB)/365.25)
      • Due to missing century information in Medicare 1989/90, DOB is unreliable. Analyses by age should not be performed on this dataset.
    • In Pharmacare, there is only Year of Birth (YOB), so age must be calculated at a specific date (ie. start or end of FY): AGE=FY-YOB*

      * YOB includes century information


    • Calculating age in only one line:
      age = floor((intck('month',birth,somedate) - (day(somedate) < day(birth))) / 12); 
      See details. More SAS notes at pauldickman.com
  • Approved = 0 in MSI Data

  • APPROVED=0 in the MSI data is a legitimate value:

    Some services are legitimately paid at zero, but must be submitted for the next services to be paid appropriately. Example: The first office visit after most major surgeries is paid at zero. If it is not submitted to be paid at zero, it will affect the payment of any subsequent office visits submitted. (Sandra Dares, ABCC)
  • Cape Breton Amalgamation

  • The creation of the Cape Breton Regional Municipality amalgamated 10 CSDs into 1 RGM. The amalgamation of Halifax had no effect on the CSDs.

    1991 Census
    1. Sydney, C
    2. Sydney Mines, T
    3. Cape Breton Subd. A, SCM
    4. Cape Breton Subd. B, SCM
    5. Cape Breton Subd. C, SCM
    6. Dominion, T
    7. Glace Bay, T
    8. Louisbourg, T
    9. New Waterford, T
    10. North Sydney, T
    1996 Census
    1. Cape Breton, RGM
  • Cape Breton Counties:


  • DOH Counties
    10    Inverness County
    15    Richmond County
    03    Cape Breton County
    17    Victoria County
    Statistics Canada Counties
    15    Inverness County
    16    Richmond County
    17    Cape Breton County
    18    Victoria County
  • Consultations in FEEGRPC:
    Consultations would be claimed for patients referred from one physician to another. Normally the claiming physician would be a specialist. Consults are paid at specialty rates only if there is a referral. If the patient went on his/her own account, MSI would not pay the specialty rate. Contacts initiated by the patient would normally be paid as visits and the fee groups would be different than those that apply to consults.
  • Deaths in Hospital


  • In ASD 1989-94:

    if SEPTO in ('05','06','07','08','09','10');
    or
    if CDEATH in ('1','2');

    In CIHI 1995-97:

    if EXITALIV='N';

    * EXITALIV='N' if (OPDEATH ne ' ' or SUPDEATH ne ' ' or
        SCUDEATH ne ' ' or RIWTYPE = '4')


  • Emergency Room Visits in Medicare Data:

    Prior to 1996 it was the fee codes that identified emergency room/outpatient services. The codes were:

    Emergency Care Centres
    A072 - 8am - 8pm
    A077 - 8pm - 8am
    A075 - Sundays and statutory holidays

    Emergency Care Centres had to be approved by the Department of Health. The last list of Emergency Care Centres dated September 22, 1995 were:

    Dartmouth General Hospital
    Victoria General Hospital
    Yarmouth Regional Hospital
    Camp Hill Medical Centre
    I.W.K. Hospital
    Cape Breton Regional Hospital
    Colchester Regional Hospital
    St. Martha's Regional Hospital (effective April 1, 1995)
    Aberdeen Hospital (effective April 1, 1995)
    Valley Regional Hospital (effective July 25, 1995)
    Highland View Hospital (effective August 14, 1995)

    Special Visit to Emergency or Outpatient Department (first patient seen)

    0601 - 8am to 5pm, Monday to Friday
    0602 - 5pm to midnight, Monday to Friday
    0603 - Midnight to 8am, all days
    0604 - 8am to midnight, Saturday, Sunday and holidays

    Extra Patients seen at any Emergency or Outpatient Department (not Emergency Care Centres)(second and subsequent patients seen while the physician was still on the premises)

    0605 - 8am to 5pm, Monday to Friday
    0607 - 5pm to midnight, Monday to Friday
    0608 - Midnight to 8am, all days
    0609 - 8am to midnight, Saturday, Sunday and holidays

    Doctor on Duty in Hospital (Emergency or Outpatient Department (not Emergency Care Centres)

    0606 - per patient

  • General Practitioners

  • In Medicare Data In Hospital Data
    1989/90-1996/97:   DSPECIAL='00'
    1989/90-1996/97:   DSPECIAL='GENP'
    1989/90-1994/95:   PHYSERV='10'
    1995/96-1998/99:   PHYSERVn='10'
  • Global Physician Cap
  • MUV: Master Unit Value - dollar value of a unit to a physician Unit: The value of a procedure is defined in terms of units Payment Rate: Payment Rate on Master Unit Value - ie. Global Cap

    APPROVED = MUV x Units
    (This is always true -- approved is a very reliable variable)

    PAID = MUV x Units x Payment Rate x Individual Cap (if applicable)
    (Usually. There are a number of conditions where this is not true)

    In general, if PAID < [APPROVED x Payment Rate], the difference is due to the individual physician cap

    Health Regions are to be replaced by District Health Authorities 1 October 2000

    Nova Scotia Health Regions (with Region & County Codes)
    Western (1) Central (3) Northern (2) Eastern (4)
    Shelburne (16) West Hants* (9) East Hants* (9) Guysborough (7)
    Yarmouth (18) Halifax (8) Colchester (4) Antigonish (2)
    Digby (6)   Cumberland (5) Inverness (10)
    Queens (14)   Pictou (13) Richmond (15)
    Annapolis (1)     Cape Breton (3)
    Lunenburg (12)     Victoria (17)
    Kings (11)      
    * Hants is divided by municipality between the Central and Northern Regions.  When assigning Health Region by County, assign Hants to the Central Region.

    Click to view larger map...
    NS Health Regions Map


  • Hospital Data
  • Fiscal Year Diagnosis Codes Procedure Codes
    1989/90 primdiag, secdiag1-secdiag4 p1code-p8code
    1990/91 primdiag, secdiag1-secdiag4 p1code-p9code
    1991/92 primdiag, secdiag1-secdiag6 p1code-p9code
    1992/93 primdiag, secdiag1-secdiag6 p1code-p9code
    1993/94 primdiag, secdiag1-secdiag6 p1code-p9code
    1994/95 primdiag, secdiag1-secdiag6 p1code-p9code
    1995/96 dxcode1-dxcode16 pcode1-pcode10
    1996/97 dxcode1-dxcode16 pcode1-pcode10
    1997/98 dxcode1-dxcode16 pcode1-pcode10
    1998/99 dxcode1-dxcode16 pcode1-pcode10
    1999/00 dxcode1-dxcode16 pcode1-pcode10
    2000/01 dxcode1-dxcode16 pcode1-pcode10


  • ICD-9 Dx Code Formats
  • Hospital Data  1989/90 - 2000/01
    Fiscal Year ICD Version Format
    1989/90 ICD-9 No Decimal
    1990/91
    1991/92* ICD-9-CM Decimal
    1992/93
    1993/94
    1994/95
    1995/96 ICD-9-CM No Decimal
    1996/97
    1997/98
    1998/99
    1999/00
    2000/01
    * ICD-9-CM was introduced 1 October 1991 (middle of FY 1991/92)
  • Income Quintile Cutoffs

    1991 $ 2nd Quintile 3rd Quintile 4th Quintile 5th Quintile
    Median Income $26,046 $31,132 $36,276 $43,745
    Average Income $30,915 $35,874 $40,268 $47,947
    Source: 1991 Census
  • Identifying Inpatients:

  •   ASD (1989-1994) CIHI (1995- )
    Inpatients/Acute Care PATCAT = '1' RECTYPE ne '1'
    Outpatients/Day Surgery PATCAT = '2' RECTYPE = '1'
  • Kosovo Refugees: Valerie Shaffner at the IWK Grace recommended that patients from Kosovo be identified by residence code 95.
  • Laspeyres Index

  • The Laspeyres Index calculates the individual contributions of price, volume, price-volume interaction, new drugs and exiting drugs to changes in drug expenditure, holding all other factors constant. It is a forward looking index that expresses prices and quantities in terms of the previous (base) period.
  • Length-of-Stay

  • Should be calculated as LOS=(SepDate-AdmitDate). If (SepDate-AdmitDate)=0, then set LOS=1.
    The calculated value and the datafile value will differ in 1991 ASD and 1995 & 1996 CIHI data due to inconsistencies in how Day Surgery LOS is treated.
  • Newborns
    Newborns can be identified in ASD data be selecting SEX=3,4 (3=Newborn Male 4=Newborn Female 5=Stillborn?)
    In CIHI 1995, newborns are identified with BIRTHS='NB'
    In CIHI 1996, newborns are identified with ENTRYCOD='N'
  • Nova Scotia Residents

  • To select Nova Scotia residents in ASD data, exclude missing MSI numbers. In CIHI data, select '11' le RESIDENC le '77'.
  • Opt In/Opt Out Dates

  • For the purposes of determining a physician enrollment in the NS Medicare plan, select physicans where OPTIN le End of FY and OPTOUT ge Beginning of FY or Missing (Physician has not opted out of program to date).
  • Payment Q & A:
    During our recent trip to Victoria, a number of questions arose about data collection in NS that I was not able to answer... these questions were forwarded to Twyla Taylor at ABCC. The following is a list of her responses, which I thought would be useful to share with everyone.

    Q: What percent of salaried physicians are not shadow billing?
    A: There are only a few salaried physicians who are not required to shadow bill that I can think of. One example would be institutional psychiatry.

    Q: Do nurse practitioners (ie. primary care projects) shadow bill?
    A: Nurse practitioners are required to shadow bill.

    Q: Are all emergency doctors salaried, or are some FFS?
    A: There is a mixture of payment methods. Most emergency depts are either block funded or the physicians are paid an hourly rate during selected times of the day and are paid FFS for regular hours during the week day.

    Q: Does NS receive claims for patients that visit doctors out-of-province (recipricol billings)? If a patient goes to see a doctor in Ontario, does that show up in our administrative data?
    A: NS does receive reciprocal claims form the other provinces. The claims are in the production environment only. They are not loaded to the data warehouse.

    Q: What optometry procedures are covered, and are 100% of such procedures covered in the MSI data, or are some of these covered outside of MSI?
    A: There is an optometry manual/ fee schedule. The health service codes include exam, continuing care, punctal occlusion. Certain services are covered for selected age groups. Perhaps you should order a manual if you are planning to do any work with optometric services.

  • Pharmacare Co-Pay

  • Period Co-Pay Maximum
    1 June 1990 - 30 June 1991 $3 per Rx $150
    1 July 1991 - 31 December 1992 20% per Rx ($3 minimum) $150
    1 January 1993 - 31 March 1995 20% per Rx ($3 minimum) $150 GIS
    $400 Non-GIS
    1 April 1995 - 31 March 1996 * 20% per Rx ($3 minimum) $200
    1 April 1996 - 10 April 2000 20% per Rx ($3 minimum) $200
    11 April 2000 - Present 33% per Rx ($3 minimum) $350 **
    * Year ends 31 March 1996, not 31 December 1995. Next Annual is based on fiscal year beginning 1 April 1996.
    ** If a senior reached the $200 maximum between 1 April - 11 April 2000, they will be subject to the new $350 maximum. Any co-pay will be applied towards the new maximum.
  • Pharmacare Costs:

    CLAIMED = Unaudited cost claim from pharmacy
    APPROVED* = DRUGCOST + DISPFEE
    PAID = APPROVED - COPAY

    * The definition of APPROVED changes beginning in 1996/97. The PAID field which was new in 1996/97 replaces what was known as APPROVED in previous years. The new APPROVED field represents amounts before adjustments are applied.

  • Report Disclaimer
  • Please include the disclaimer in the publications.

    The data used in this report were made available by the Population Health Research Unit (PHRU) within Dalhousie University's Department of Community Health and Epidemiology.

    PHRU is a university-based research and support group conducting systematic research into population health, health services and their inter-relationships.The Province of Nova Scotia supplies PHRU with complete Medicare, Pharmacare and Hospital files suitable for research purposes. The Unit also has access to a variety of other data sources including clinical databases and large scale population surveys.

    Although this research is based on data obtained from PHRU, the observations and opinions expressed are those of the authors and do not represent those of PHRU.

  1995/96 1996/97 1997/98 1998/99 1999/00
Hospital Budget
($ '000s)
$631,646 $645,026 $707,242 $795,946 $811,375
Total RIWs 183,158 175,466 176,209 160,864 161,702
Cost per RIW $3,449 $3,676 $4,014 $4,285 $5,018
  • SAS Tips:
    • Age Calculation:    See Patient Age


    • Converting Formats from v6 to v8:

      libname v6lib v6 '[whatever.directory]';
      libname v8lib v8 '[whatever.directory]';

      proc copy in=v6lib out=v8lib memtype=catalog;
      select formats;
      run;

    • Many-to-Many Merge using SQL:    Click for SAS Code

    • Parsing Formatted Values:    IF PUT(variable,$format.)=...

    • Summary Array:    Click for SAS Code

  • Vital Statistics Database


    * The large increase in 1995 is due to the inclusion of Registry Data to backfill Vital Stats data.

 
   
  spacer