MORGAM Manual

Conversion from MONICA Survey core data to MORGAM Form 20

© National Institute for Health and Welfare and the MORGAM Project investigators Last updated: 27 March 2001 For more information, please contact Zygimantas Cepaitis (firstname.lastname@thl.fi)

• MORGAM Data transfer format: MONICA survey data (Form 20)
• Conversion of variables related to smoking behaviour
• Conversion of other variables from the older data collection forms (versions 3,6)
• Conversion of dates
• Calculation of variable EAGE - age of the subject on the date of examination

The MORGAM Data transfer format: MONICA survey data (Form 20) is very similar to version 7 of the MONICA Core data transfer format - survey data (Form 04). For versions 3 and 6 of the MONICA data transfer formats, which were used in the initial and middle MONICA surveys, the conversion to the MORGAM format is not straight forward. The MDC has drafted a conversion rule, which you can find at the end of this page. If you have any comments on the conversion rule, please let me know.

For testing the data checking system, a conversion from the MONICA data to the MORGAM data has been done in the MDC for the cohorts where the MONICA data are available in the MDC. This concerns MPCs:

These converted data have not been entered to the MORGAM data base. If you ask for a copy of your data, I will send them to you. The data can also be entered to the MORGAM data base, if so agreed with the MPC. However, it is often preferred to extract the MONICA survey data locally, because the MPC may have data for some of the data items, which were not collected to the MDC in the initial and/or the middle MONICA survey.

Confirmation of constraint violations of the MONICA survey data: The MDC will not ask for reconfirmation of the constraint violations which the MPC already confirmed during the MONICA data transfer.

Conversion rule from versions 3 and 6 of MONICA Core data transfer format: Survey data (Form 04) to MORGAM Data transfer format: MONICA survey data (Form 20)

The rule for the non-trivial conversions is given in SAS language. If you have difficulty in understanding the rule, please contact Zygimantas Cepaitis (firstname.lastname@thl.fi).

Conversion of variables related to smoking behaviour

OTHERSM

• if VERSN = 3 then do;
  • if CIGS = 1 then OTHERSM = 88;
  • else OTHERSM = 99;
• end;
• if VERSN = 6 and CIGS = 1 then OTHERSM = 88;

CIGARSM, PIPESM

• if VERSN = 3 then do;
  • CIGARSM = 9;
  • PIPESM = 9;
• end;

CIGAGE

• if VERSN = 3 then do;
  • if EVERCIG in(2,3) then CIGAGE = 88;
  • else CIGAGE = 99;
• end;
• if VERSN = 6 then do;
  • if CIGAGE = 88 then CIGAGE = 99;
  • if EVERCIG = 3 then CIGAGE = 88;
• END;

MAXCIGS

• if VERSN = 3 then do;
  • if EVERCIG in(2,3) then MAXCIGS = 888;
  • else MAXCIGS = 999;
• end;
• if VERSN = 6 and EVERCIG = 3 then MAXCIGS = 888;

IFLYEAR

• if VERSN in(3,6) and IFLYEAR = 8 and CIGS = 3 then IFLYEAR = 9;

STOP

• if VERSN in(3,6) and STOP = 8888 and CIGS = 3 then STOP = 9999;

EVERCIG

• if VERSN in(3,6) then do;
  • if EVERCIG = 3 then EVERCIG = 2;
  • if EVERCIG = 8 and CIGS = 3 then EVERCIG = 9;
• end;

DAYCIGS

• if VERSN = 3 then do;
  • if CIGS in(1,2) then DAYCIGS = 8;
  • else DAYCIGS = 9;
• end;
• if VERSN = 6 and CIGS = 1 then DAYCIGS = 8;

Conversion of other variables from the older data collection forms (versions 3,6)

• if VERSN = 3 then do;
  • BPRECD = 9;
  • HICH = 9;
  • CHDT = 9;
  • CHRX = 9;
  • CHRECD = 9;
  • ASP = 9;
  • if SEX = 1 then do;
    • MENOP = 8;
    • AGEM = 88;
    • HORM = 8;
    • PILL = 8;
  • end;
  • if SEX = 2 then do;
    • MENOP = 9;
    • AGEM = 99;
    • HORM = 9;
    • PILL = 9;
  • end;
  • CUFF = 9;
  • ARM = 99;
  • TIMEBP = 9999;
  • RTEMP = 99;
  • DHDL8 = 99999999;
  • COTIN = 9999;
  • CARBMON = 99;
  • WAIST = 9999;
  • HIP = 9999;
  • WHCODER = 99;
• end;
• if VERSN = 6 then do;
  • CUFF = 9;
  • ARM = 99;
• end;

Conversion of dates

In MORGAM Form 20 all years are 4-digit numbers. The rule for conversion of date fields from MONICA Form 04 (Survey data) to MORGAM Form 20 (MONICA Survey data) is as follows:

• if date is valid, then '1900' was added to the year, so DDMMYY in old format becomes DDMM19YY in new format
• if date is missing, then '9900' was added to the year, so 999999 in old format becomes 99999999 in new format
• if date is irrelevant, then '8800' was added to the year, so 888888 in old format becomes 88888888 in new format

Calculation of variable EAGE - age of the subject on the date of examination

Note that in calculations was used variable DBIRTH (full date of birth) and variable DEXAM from MONICA Core data transfer format - survey data (Form 04). Calculation of variable EAGE is given in SAS language. If you have difficulty in understanding this calculation, please contact Zygimantas Cepaitis (firstname.lastname@thl.fi).

Evaluate missing fields in date of birth (DBIRTH)

• select(verify(DBIRTH,"9"));
  • when(3) db = input(translate("15" || substr(DBIRTH,3,4),"0"," "),??ddmmyy6.);
  • when(5,6) db = input(translate("3006" || substr(DBIRTH,5,2),"0"," "),??ddmmyy6.);
  • when(0) db = .;
  • otherwise db = input(translate(DBIRTH,"0"," "),??ddmmyy6.);
• end;

Evaluate missing fields in date of examination (DEXAM)

• select(verify(DEXAM,"9"));
  • when(3) dex = input(translate("15" || substr(DEXAM,3,4),"0"," "),??ddmmyy6.);
  • when(5,6) dex = input(translate("3006" || substr(DEXAM,5,2),"0"," "),??ddmmyy6.);
  • when(0) dex = .;
  • otherwise dex = input(translate(DEXAM,"0"," "),??ddmmyy6.);
• end;

Calculation of age on the date of examination

• EAGE = floor((dex-db+(day(dex)=day(db)))/365.25);
• if EAGE = . then EAGE = 99;

Note: The floor() function returns the largest integer that is less than or equal to the argument.