previous   next   contents

Description and quality of baseline data: Awareness and treatment of high blood pressure and cholesterol Olli Saarela1 and Sangita Kulathinal1,2 for the MORGAM Project3

¹ Department of Health Promotion and Chronic Disease Prevention, National Public Health Institute, Helsinki, Finland
² Since January 2007 at Indic Society for Education and Development (INSEED), Nashik, India
³ See Annex for the sites and key personnel of contributing MORGAM Centres

© National Institute for Health and Welfare and the MORGAM Project investigators Last updated: 11 May 2011 For more information, please contact Olli Saarela (firstname.lastname@thl.fi) or Kari Kuulasmaa (firstname.lastname@thl.fi)

Contents

• Data items considered
• Approach to the description and quality assessment
  • Quality scores
• Assessment of the individual data items
  
  • Availability and distributions of the data items
• Quality assessment scores
• Assessment of the data items on high blood pressure
  • HIBP
  • DRUGS
  • BPRECD
  • Discussion and conclusions on high blood pressure items
• Assessment of the data items on high cholesterol
  • HICH
  • CHDT
  • CHRX
  • CHRECD
  • Discussion and conclusions on high cholesterol items
• References
• Updates to this document
• Specific tables
  • Table 1 Scores for questionnaire items
  • Table 2 Distributions for high blood pressure items
  • Table 3 Distributions for high cholesterol items

Data items considered

We consider the following items related to awareness and treatment of high blood pressure and cholesterol in Data transfer format - MONICA Survey Data (Form 20):

• HIBP: Have you ever been told by a doctor or other health worker that you have high blood pressure?
• DRUGS: Are you taking (in the last two weeks) drugs for high blood pressure?
• BPRECD: Have you had your blood pressure measured in the last year?
• HICH: Have you ever been told by a doctor or other health worker that you have high blood cholesterol?
• CHDT: Are you on a special diet prescribed by a doctor or other health worker to lower your blood cholesterol level?
• CHRX: Are you taking (in the last two weeks) pills or other medicine prescribed by a doctor to lower your blood cholesterol level?
• CHRECD: Have you had your blood cholesterol measured in the last year?

Derived items in Form 31 utilising the above items are

• HIGHBP1: High blood pressure

Approach to the description and quality assessment

The individual data items are assessed by availability and distributions of the data items and comparison of the local questions with the MORGAM data items. This approach is based on the principles adopted from the comprehensive approach described in Quality Assessment of Data on Hypertension Control in the WHO MONICA Project [1] and Quality Assessment of Data on Awareness and Treatment of High Cholesterol in the WHO MONICA Project [2].

Quality scores

Questionnaire Comparability Score (QCS), Insufficient Data Score (IDS) and Summary Score (SS) defined in the Introduction are used to assess the quality of above questionnaire items.

Assessment of the individual data items

Availability and distributions of the data items

Hyperlinks to the distributions of the data items are under the respective data item names:

• HIBP: Have you ever been told by a doctor or other health worker that you have high blood pressure?
• DRUGS: Are you taking (in the last two weeks) drugs for high blood pressure?
• BPRECD: Have you had your blood pressure measured in the last year?
• HICH: Have you ever been told by a doctor or other health worker that you have high blood cholesterol?
• CHDT: Are you on a special diet prescribed by a doctor or other health worker to lower your blood cholesterol level?
• CHRX: Are you taking (in the last two weeks) pills or other medicine prescribed by a doctor to lower your blood cholesterol level?
• CHRECD: Have you had your blood cholesterol measured in the last year?
• Derived variables:
  • HIGHBP1: High blood pressure

Distributions of the data items for the high blood pressure data are summarized in Table 2 and for the high cholesterol data in Table 3.

Data on high blood pressure items HIPB and DRUGS are available for all Cohorts except FIN-ATB and data on BPRECD are available for some or all cohorts in AUS-NEW, DEN-GLO, FIN-ATB, FIN-EAS, FIN-WES, ITA-BRI, LTU-KAU, POL-TAR, POL-WAR, RUS-NOV and SWE-NSW.

Data on high cholesterol items are not available at all in FIN-ATB, GER-AUG, ITA-PAM, ITA-ROM, RUS-NOVc and UNK-SHH.

Quality assessment scores

Table 1 gives the quality assessment scores QCS, IDS and SS for each Cohort. The table also shows the availability of the data items by Cohort. The Questionnaire Comparability Score (QCS) is derived by comparing the compatibility of the local questionnaire with the MORGAM data transfer format. Most of the local questionnaires are available in pdf format in Appendix 2. The item and Cohort specific differences between local questionnaires and MORGAM format are discussed in more detail below. The Insufficient Data Score (IDS) is derived by calculating the percentage of insufficient data (coded as 9). The denominator used in calculating the proportions of missing data was all subjects in the cohort, except for those data items which were preceded by a filter question, namely HIBP for DRUGS and HICH for CHDT and CHRX. In these cases, subjects for whom the filter item was coded irrelevant (8) were excluded.

Assessment of the data items on high blood pressure

The percentage distributions by Cohort for these data items are given in Table 2. The data availability and quality scores are given in Table 1. The explanation for a lower QCS score is given in the cohort specific comments below.

HIBP: Have you ever been told by a doctor or other health worker that you have high blood pressure?

This data item indicates the awareness of (past or present) hypertension. The most common deviation from the MORGAM format was that the local question did not specify that the diagnosis was done by a doctor or other health worker.

Known issues by Cohort:

• AUS-NEW: In the first two Cohorts, the local question did not specify that the one who told about high blood pressure was a health professional (QCS = 1). In the third Cohort the question was changed to follow the MORGAM format.
• DEN-GLO: The local question was "Have you ever been told that you had high blood pressure?" (QCS = 1). However, the questionnaire included another item where the question was "Has a doctor ever told you that you suffered from the following:", and high blood pressure was listed as one of the alternatives. Apparently this item has not been used when extracting the MORGAM data.
• FIN-EAS/WES: The local question did not specify that the diagnosis was done by a health professional (QCS = 1).
• ITA-ROM: The local question was "Are you aware to have, or have had, or have been treated for hypertension?" (QCS = 1). This question was not asked for Cohorts 21 and 22 but for these Cohorts the individuals with DRUGS = 1 have been coded as HIBP = 1 and the rest are coded as insufficient data (QCS = 0 and IDS = 0). For Cohort 01 the proportion of insufficient data is 17% (IDS = 0).
• UNK-CAE: The local question was "Have you had any of the following illnesses in the last five years?" and high blood pressure was listed as one of the alternatives (QCS = 1).

DRUGS: Are you taking (in the last two weeks) drugs for high blood pressure?

This item indicates the current use of antihypertensive drugs. It has been used in defining the derived item HIGHBP1. It may also be used in the analysis of measured blood pressure levels if it is needed to restrict the analysis to healthy (or pre-intervention) individuals. This item is irrelevant for those with HIBP = 2. The most common deviation from the MORGAM format was that the time specification was something else than in the last two weeks. Also, many local questionnaire items did not include the option "uncertain". This would be used for example if the respondent was taking medication but was unsure if they were for lowering blood pressure. In those cohorts that did specify this option, the proportion of DRUGS = 3 is usually low, so this is unlikely to be a serious problem.

Known issues by Cohort:

• AUS-NEW: In the first Cohort the local question was "Are you on tablets for blood pressure?" (QCS = 1). In the second Cohort the question was "Are you having treatment for high blood pressure at present?", with a later question specifying the kind of treatment (QCS = 1). In the third Cohort the question was "Are you having treatment with medications for high blood pressure at present?" (QCS = 1). In the third Cohort the numbers for HIBP = 2 and DRUGS = 8 do not match.
• DEN-GLO: The local question in the first and second Cohorts was "Are you treated for hypertension now?", with an option of "yes, tablets or pills" (QCS = 1). The question in the third Cohort and in Cohort 21 was "Are you taking (in the last week) drugs for high blood pressure?" (QCS = 1). The proportion of insufficient data is 20%, 21% and 54% for Cohorts 01, 03 and 21 respectively (IDS = 0).
• FIN-EAS/WES: The local question in all four Cohorts was "Have you ever used antihypertensive drugs?", with a later question "When did you last take antihypertensive drugs?", with predefined options. The exact time frame of two weeks can not be extracted from these (QCS = 1).
• FRA-LIL/STR/TOU: The local question was "Do you presently take tablets for high blood pressure?" (QCS = 1).
• ITA-PAM: The local question was "Do you regularly take antihypertensive drugs?" (QCS = 1).
• ITA-ROM: The local question was "Are you regularly taking drugs for lowering your blood pressure now?" (QCS = 1). In Cohorts 21 and 22, no-one was coded as HIBP = 2 and consequently here no-one has been coded as DRUGS = 8. The proportion of insufficient data for Cohort 01 was 50% (IDS = 0).
• LTU-KAU: The local question was "Have you ever taken any medicine for high blood pressure?", followed by a question "How many weeks ago have you taken medicine for high blood pressure last time?" (QCS = 1). The proportion of DRUGS = 3 is unusually high for the first Cohort. The MONICA quality assessment [1] states that "For LTU-KAU the MCC informed that there are two possible explanations for the decrease in the proportion of uncertain data on DRUGS: 1) the question was interpreted by physician in the middle and final survey and 2) the awareness of the patients about the drugs has increased during the last years."
• POL-TAR: The proportion of insufficient data is 16% and 17% for Cohorts 02 and 03 respectively (IDS = 0).
• UNK-BEL: The local question was "Do you presently take tablets for high blood pressure?" (QCS = 1).
• UNK-CAE: The conversion rule provided by the centre states that the item DRUGS is derived from the local question:  "Are you taking drugs for cardiovascular disease?" and anyone taking drugs containing antihypertensives are coded as uncertain (QCS = 0). The proportion of insufficient data is 18% (IDS = 0).
• UNK-EDI/GLA/SHH: The local question was "Are you now taking any medication for high blood pressure?" (QCS = 1).

BPRECD: Have you had your blood pressure measured in the last year?

This item is not available for as many cohorts as HIBP and DRUGS. However, the item itself is more straightforward than the other two and therefore there was less variation in the local questions. One type of variation was "When was your blood pressure last measured?", with predefined alternatives. This matches with the MORGAM item. Some local questionnaires included an option for uncertain answer ("do not remember"). In some cases these may have been coded as something else than BPRECD = 9. However, for most cohorts the exact conversion rule is unknown.

Known issues by Cohort:

• DEN-GLO: The proportion of insufficient data is 24% and 19% for Cohorts 03 and 21 respectively (IDS = 0).
• FIN-ATB: Such item was not actually included in the questionnaire, but has been coded as BPRECD = 1 for all cohort members, because the blood pressure had been measured for everyone in previous examination visits (QCS = 0).

Discussion and conclusions on high blood pressure items

The conclusions made in the discussion section of MONICA Quality Assessment for Hypertension Control [1] apply also here, even though in MORGAM these data are likely used for other purposes than trend analyses. The most common variable to be used in the analyses is likely to be the derived variable for hypertension, HIGHBP1, which classifies as hypertensive those with DRUGS = 1 or 3. Because of the differences in the local questions used for deriving DRUGS, the derived variable HIGHBP1 may not be exactly comparable between different cohorts.

Assessment of the data items on high cholesterol

The percentage distributions by cohort for these data items are given in Table 3. The data availability and quality scores are given in Table 1. The explanation for a lower QCS score is given in the cohort specific comments below.

HICH: Have you ever been told by a doctor or other health worker that you have high blood cholesterol?

This data item indicates the awareness of (past or present) high cholesterol. The most common deviation from the MORGAM format was that the local question did not specify that the diagnosis was done by a doctor or other health worker. This item is not available for as many cohorts as the comparable item on high blood pressure.

Known issues by Cohort:

• AUS-NEW: In the second Cohort, the local question did not specify that the one who told about high blood cholesterol was a health professional (QCS = 1). In the third Cohort the question was changed to follow the MORGAM format.
• FIN-EAS/WES: The local question did not specify that the diagnosis was done by a health professional (QCS = 1). The proportion of insufficient data is 26% and 28% for FIN-EASa Cohorts 03 and 24 and 39% and 33% for FIN-WESb Cohorts 03 and 24 (IDS = 0).
• ITA-ROM: This item was not asked for Cohorts 21 and 22 but for these Cohorts the individuals with CHRX = 1 have been coded as HICH = 1 and the rest are coded as insufficient data (QCS = 0 and IDS = 0).
• POL-WAR: The proportion of missing data for Cohort 03 is 36% (IDS = 0).
• RUS-NOV: The proportion of missing data for Cohort 03 is 18% (IDS = 0).
• UNK-CAE: The local questionnaire included an item "Have you taken any medicines prescribed by your doctor or a specialist within the last 7 days? If yes, specify". It appears that cholesterol medicine listed here has been used to derive HICH = 1 for five individuals (QCS = 0 and IDS = 0).

CHDT: Are you on a special diet prescribed by a doctor or other health worker to lower your blood cholesterol level?

This item indicates if the respondent is currently complying a diet prescribed by a health professional. It may be used in the analysis of measured cholesterol levels if it is needed to restrict the analysis to healthy (or pre-intervention) individuals. This item is irrelevant for those with HICH = 2. The most common deviation from the MORGAM format was that the local question did not specify that the diet was prescribed by a doctor or other health worker.  Also, many local questionnaire items did not include the option "uncertain".

Known issues by Cohort:

• AUS-NEW: The local question was "If you are on a treatment for high cholesterol, what sort of method(s) are you using?", with diet as one of the alternatives. However, for the second Cohort it was not specified that the diet had to be prescribed by a health professional (QCS = 1).
• FIN-EAS/WES: The local question was "If your blood cholesterol level has been examined, did you in that examination receive diet instructions to lower the blood cholesterol level?" (QCS = 1). The proportion of insufficient data is 47% and 48% for FIN-EASa Cohorts 03 and 24 and 65% and 57% for FIN-WESb Cohorts 03 and 24 (IDS = 0).
• FRA-LIL/STR/TOU: The local question was "Do you presently follow a diet for cholesterol and/or blood fats?" (QCS = 1).
• LTU-KAU: The local question was "Did you ever try to decrease your blood cholesterol level?", with options "No", "Yes, taking medicine only", "Using diet only", "Medicine and diet" and "I don't know" (QCS = 1). The third and fourth options were coded as CHDT = 1.  The proportion of missing data for Cohort 03 is 71% (IDS = 0).
• POL-TAR: The proportion of missing data for Cohort 03 is 50% (IDS = 0).
• POL-WAR: The proportion of missing data for Cohort 03 is 71% (IDS = 0).
• RUS-NOV: The proportion of missing data for Cohort 03 is 87% (IDS = 0).
• UNK-BEL: The local question was "Do you presently follow a diet for cholesterol and/or blood fats?" (QCS = 1).

CHRX: Are you taking (in the last two weeks) pills or other medicine prescribed by a doctor to lower your blood cholesterol level?

This item indicates the current use of drugs for high cholesterol. It may be used in the analysis of measured cholesterol levels if it is needed to restrict the analysis to healthy (or pre-intervention) individuals. This item is irrelevant for those with HICH = 2. The most common deviations from the MORGAM format were that the time specification was something else than in the last two weeks or that it was not specified that the medicine had to be prescribed by a doctor. Also, many local questionnaire items did not include the option "uncertain".

Known issues by Cohort:

• AUS-NEW: The local question was "If you are on a treatment for high cholesterol, what sort of method(s) are you using?", with medication as one of the alternatives (QCS = 1).
• DEN-GLO: The local question did not specify that the medicine had to be prescribed by a doctor (QCS = 1).
• FIN-EAS/WES: The local question was "Are you currently using medication prescribed by a doctor to lower blood cholesterol level?" (QCS = 1). The proportion of insufficient data is 47% and 49% for FIN-EASa Cohorts 03 and 24 and 65% and 57% for FIN-WESb Cohorts 03 and 24 (IDS = 0).
• FRA-LIL/STR/TOU: The local question was "Do you presently take tablets for cholesterol and/or blood fats?" (QCS = 1).
• ITA-BRI:  The local question was "Have you taken any medication in the last two weeks for lowering your cholesterol?" (QCS = 1).
• ITA-ROM: The local question was "Are you regularly taking drugs to lower your cholesterol level now?" (QCS = 1). In Cohorts 21 and 22, no-one was coded as HICH = 2 and consequently here no-one has been coded as CHRX = 8.
• LTU-KAU: The local question was "Did you ever try to decrease your blood cholesterol level?", with options "No", "Yes, taking medicine only", "Using diet only", "Medicine and diet" and "I don't know", with a follow-up question "How many weeks ago have you taken medicine for high blood cholesterol level?" (QCS = 1). The proportion of missing data for Cohort 03 is 71% (IDS = 0).
• POL-TAR: The proportion of missing data for Cohort 03 is 51% (IDS = 0).
• POL-WAR: The proportion of missing data for Cohort 03 is 71% (IDS = 0).
• RUS-NOV: The cholesterol items are not available for the second Cohort. However, here 10 individuals have been coded as CHRX = 9 and the rest as CHRX = 2. Apparently some other item has been used for deriving this information (QCS = 0). The proportion of missing data for Cohort 03 is 36% (IDS = 0).
• SWE-NSW: The local question for the second Cohort did not specify that the medicine had to be prescribed by a doctor (QCS = 1).
• UNK-BEL: The local question was "Do you presently take tablets for cholesterol and/or blood fats?" (QCS = 1).
• UNK-CAE: The local questionnaire included an item "Have you taken any medicines prescribed by your doctor or a specialist within the last 7 days? If yes, specify". It appears that cholesterol medicine listed here has been used to derive CHRX = 1 for five individuals (QCS = 0).

CHRECD: Have you had your blood cholesterol measured in the last year?

This is a similar to the corresponding item for blood pressure and the things said about that apply also here.

Known issues by Cohort:

• FIN-EAS/WES: The proportion of insufficient data is 23% and 25% for FIN-EASa Cohorts 01 and 02 and 25% and 27% for FIN-WESb Cohorts 01 and 02 (IDS = 0).
• LTU-KAU: The proportion of missing data for Cohort 03 is 19% (IDS = 0).
• RUS-NOV: The proportion of missing data for Cohort 03 is 20% (IDS = 0).
• UNK-BEL: This item was not included in the local questionnaire but here 45 individuals with CHRX = 1 have been coded as CHRECD = 1 (QCS = 0 and IDS = 0).
• UNK-CAE: The local questionnaire included an item "Have you taken any medicines prescribed by your doctor or a specialist within the last 7 days? If yes, specify". It appears that cholesterol medicine listed here has been used to derive CHRECD = 1 for five individuals (QCS = 0 and IDS = 0).

Discussion and conclusions on high cholesterol items

The conclusions made in the discussion section of MONICA Quality Assessment of Data on Awareness and Treatment of High Cholesterol [2] apply also here. Currently there is no derived variables using the high cholesterol awareness and treatment items.

References

1. Molarius A, Tuomilehto J and Kuulasmaa K for the WHO MONICA Project. Quality Assessment of Data on Hypertension Control in the WHO MONICA Project (October 1998). Available from URL:http://www.ktl.fi/publications/monica/hyperten/hbpdrug.htm, URN:NBN:fi-fe19991084.
2. Tolonen H, Ferrario M and Minoja M for the WHO MONICA Project. Quality Assessment of Data on Awareness and Treatment of High Cholesterol in the WHO MONICA Project (June 1999). Available from URL:http://www.ktl.fi/publications/monica/hich/hchdrug.htm, URN:NBN:fi-fe19991130.

Updates to this document

previous   next   contents