Project SMART

Go to content


Meetings and Conference

SMART Meeting (Florence, 28-29 September 2009)


Main Points from SMART Meeting- Florence 28-29 September 2009

Delegates present:
Airi-Alina Allaste, (AAA) (Estonia), Kim Bloomfield (KB) (project's expert), Ladislaw Csemy (LC) and Hana Savinova (HS)(Czech Republic), Zsuzsanna Elekes (ZE) (Hungary), John Foster (JF) (UK), ( Ann Hope (AH) (Ireland), Esa Ísterberg (EO) (Finland), Daniela Piontek (DP) (Germany), Jacek Moskalewicz (JM) and Janusz Sieroslawski (JS) (Poland), Lidia Segura,(LS) (Spain), Allaman Allamani (AA) and Karin Panzer (KP) (Italy)

Apologies: Peter Anderson (project's expert), Joan Colom and Antoni Gual (Spain), Ludwig Kraus (Germany), Betsy Thom (UK),

Main agenda items
included discussion of a literature review on survey methodology prepared by KB, EU surveys review report by JS and then extensive discussion on consecutive parts of the pilot questionnaire, implementation plan, focus group themes to assess utility of the survey instrument, first proposal of statistical analyses and finally administrative and financial issues.

Literature Review
Literature review was discussed and felt to be very helpful to the SMART project. The main concerns were around table 3 on recommended drinking limits as there appeared to be a number of inaccuracies. Delegates were requested to email updated information to Kim.
It was also felt that the review should include some discussion of the RAPS. Finally KB stated that the SMART team needed to consider a unique angle for this review with a view to publication.

Survey Review
The latest draft of this was presented by JS. This will be distributed to the delegates for further comments. It was felt that attempts should be made to check that the information/interpretation is correct. If checking is not possible a rider with this regard should be added to the report.

There is still some missing data from Sweden and Denmark amongst others (EO will attempt to obtain this).

KB suggested that perhaps some of the GENACIS data in Lausanne could be used. Finally before the report was approved there needed to be a number of collective recommendations made.

On the final day KB suggested that there was a need for more tables and less text- e.g. in relation to what country had used which approach and measures of dependence and abuse. A good approach would be to focus upon areas where confusion exist- e.g. P13 Table 4- Spain, Holland, Estonia etc.
Each country was asked to read the report and feedback if there are any obvious errors.

Pilot survey instrument

Questions on Alcohol Consumption and Drinking Context:

JS led this discussion. There was a feeling that there was a need to avoid interviewer fatigue and confusion and it was agreed to drop questions relating to the last 30 days . However it was also decided to introduce a filter question - along the lines of "Have you drunk any of the following during the past 12 months." The word "alcohol" will not be used as this is culturally specific in some countries e.g. wine is not considered alcohol. JS will distribute the finalised version shortly. Following the sending of the final version, when training the interviewers please ensure that the consumptions methods (i.e. graduated frequency or last occasion) are presented in random order.

Unrecorded Consumption
Following discussion a number of suggestions were made to shorten this part of the questionnaire and to ensure consistency between purchase and consumption. JS will distribute final version shortly

Heavy Consumers and Problems
The state of play following the Copenhagen meeting was fed back. After some discussion it was decided to keep CIDI dependence scale but to amalgamate the CIDI abuse items into a social problems section of the questionnaire. LS questioned the veracity of using DSM-IV related items for a general population survey. Thus the RAPS and AUDIT (full version) have been added to this part of the measure to test their psychometric effects. When the survey is administered these should be presented in random order.
The drunkenness items were accepted though the question should be worded to avoid a double negative. "Binge drinking" was also accepted in principle, though for RSOD_2 it was suggested that time intervals were used.
"Alcohol Harm to Others"- AH raised concerns about altering the items from Robin Room's questionnaire. Items H1-11 will be included in the final version, and a number of "Attitudes to Alcohol Policy" questions will be included in the final version.
AH will circulate the final version shortly.

Implementation of the Pilot Study
All interviews will be face to face and interviewers will have to be trained and a protocol will be sent to ensure consistency.
It will be necessary to keep some record of the numbers of refusals (e.g. record gender and make an estimate of age). The pilot survey will apply to individuals who have drunk within the past 12 months only and a filter question will be employed. The aim is to perform a household study (i.e. knock on doors) and for the main study recruit 180 individuals- (90 from urban areas, 90 from non- urban areas). The aim is a 50% gender split in the following age bands 18-29-30-49, 50-74 (these categories may be changed in the final protocol.) Again if viable these should be related to census data so that it reflects the age and gender spilt of the general population. If possible there should be a class/education spread.
There will also be a survey of a further 20 heavy drinkers (10 urban/10 non-urban)- these can either be snowballed from the original sample or taken from alcohol treatment centres or similar. The final protocol will define how "heavy drinker" is to be assessed.
Re-Qualitative data: A protocol will be developed to add this and a number of semi-structured questions will be asked as part of the interview. The interviewer will take notes and then be asked to write a short reflection on each interview. At the end of this process each country will write a global report concerning the experience of completing these questionnaires etc.
The final English version of the questionnaire and protocol should be distributed before the end of October. Where relevant the survey tools will have to be translated into the native tongue, this is likely to take 1-2 weeks. The aim is to complete collection of the pilot data before Xmas.

Focus Groups
The aim is conduct three focus groups that reflect the age ranges that will be set out in the final protocol and a group of heavy drinkers. It may be possible to select the heavy drinkers with reference to their replies to the consumption questions. All participants will be asked in the pilot study (above) if they are willing to participate in further focus groups. The interview will also provide the interviewer with a chance to assess the suitability of the participant for the focus groups (i.e. do they understand the concepts sufficiently and are they able to articulate their thoughts to provide valuable data). It will be up to each country to find the best way of conducting these in their chosen environments and it is unlikely there will be need for full taping and transcription. It may be best to consider them as group discussions. The key will be taking full notes of the main points raised in the discussion and that the interviewers are trained and understand the concepts being discussed. It may be helpful to tape the groups as this will aid subsequent note taking. The final protocol and the questions will be distributed at the end of October by the Polish team.

Cost- Benefit Analysis
Peter Anderson's draft report was distributed at the meeting and feedback was requested. However as PA was not able to be present at the meeting there was no further discussion.

Statistical Analysis
JS introduced his tentative ideas for statistical analysis based upon the principles of General Linear Modelling. One of his aims is to look at the predictors of problems/troubles and to do this he will aim is to minimise the interaction between countries and control for individual differences. It is likely he will collapse the countries into three or four groups and use any of the following either alone or in combination- cluster analysis, multi-dimensional scaling or factor analysis. Other aims will be to get an estimation of consumption levels and test the sensitivity/specificity of the troubles measures- i.e DSM-IV/CIDI, RAPS and AUDIT).
One of the main concerns of the delegates was to ensure uniformity of data collection and coding. JS will design a data set with coding instructions. Each country will then enter this and send it on to JS.

Financial Reports
JM stated that were still some financial reports outstanding and requested they be sent to him by Wednesday 7th October.

JS presented the web-site, links and the number of hits etc. A list was made of those individuals who should be added to the list of participants. Each institution was asked to send a web link and to provide a web link to the SMART web-site.

Next Meeting
A data has not been fixed for the next meeting. However it will be held at some point March/April 2010 following analysis of the pilot data.

Rapporteur: John Foster

Back to content | Back to main menu