Angry Doctor

Thursday, May 18, 2006

Smoke gets in your eyes

Encouraging news from the Health Promotion Board (HPB):


Smokers heed graphic warnings on cigarette packs: HPB
By S Ramesh, Channel NewsAsia

SINGAPORE : Graphic warnings introduced on cigarette packaging since August 2004 seem to have had an effect on smokers, according to a survey by the Health Promotion Board.

Nearly half (47 percent) said they smoked less frequently after seeing the health warnings, while 57 percent said they became more concerned about the health effects of smoking.

The survey covered some 1,300 smokers and non-smokers from the ages of 18 and 69.

It was conducted between November and December 2004, a few months after the health warnings were introduced on cigarette packaging.

The survey also showed that the labels were effective in reinforcing health messages among smokers. Seventy-one percent said they knew more about the health effects of smoking as a result of the warnings, while a quarter said they were motivated by the warnings to quit smoking.

The warnings also helped kick-start the quitting process among smokers -- 28 percent said they smoked fewer cigarettes; 14 percent said they made it a point to avoid smoking in front of children; 12 percent said they avoided smoking in front of pregnant women; and 8 percent said they smoked less at home.

Among non-smokers, 46 percent said they had advised smokers to quit smoking after seeing the health warning labels; 17 percent of them were wives who had advised their husbands to quit smoking.

The Health Promotion Board said it was looking at introducing a new set of graphic health warnings by the end of the year.

(You can also see the text of the Press Release from HPB here.)


angry doc is all for public health measures to reduce smoking, but this report has brought a few questions to my mind.

I wonder why, if the survey was done almost 18 months ago, that they waited so long to release the findings.

The sample size of 1300 is not really very big, considering that Singapore has a population of 4 million, and 24.9% of men and 4.1% of women are smokers. The report does not state percentage of smokers in the 1300 persons interviewed, so it's hard to see how representative the survey is.

Also, being a survey, the study relied on responses from those interviewed rather than objective parameters - and in angry doc's experience every smoker he has had the privilege to know smokes 'very little', or has been 'cutting down', or 'has stopped smoking' (usually just prior to admission to the hospital or arrival at the clinic).

The report and the press release do not actually tell you how many smokers had managed to successfully quit smoking as a result of the graphic health warnings. Granted the survey may have been a once-off questionnaire, but as we all know quitting is a (long) process, and an indication of a desire to quit smoking may sometimes be just that - a desire. A follow-up survey would have provided more information.

More importantly, when conducting a health study, one should really keep the objective in mind. In this case, the real question should not be how people feel about the graphic warnings, but whether they do achieve their purpose of cutting down smoking-related illnesses through cutting down smoking.

Now since the ill-effects of smoking may not manifest themselves for many years, a good surrogate and a more accurate way to assess the effects would be simply to look at whether cigarette sales per capita has dropped since the introduction of the graphic warnings. Certainly there may be other factors influencing smoking patterns (such as the rise in price of cigarettes and the banning of smoking in certain public areas), but at the end of the day it's the end results we are interested in, isn't it?

( I am willing to bet, however, that the sales of cigarette cases have increased since the introduction of graphic warnings on cigarette packets.)

Still, I suppose the public likes to see a survey with lots of numbers and percentages.

More than pictures of diseased organs anyway.

Labels:

7 Comments:

  • we can't really trust the media can we?

    By Anonymous Anonymous, At May 19, 2006 12:28 pm  

  • When it comes to health news, I tend to want to try to look beyond the figures presented, which is not to say whether or not the media can be trusted.

    Specifically in this case, they probably received the 'news' as a press release from the HPB (see link in the post), and edited it to make their piece. Basically they rearranged the paragraphs a little - there was no real analysis.

    The purpose of the press release is probably to demonstrate that the use of the graphic warnings was effective, and that the practice should be continued (note the last paragraph which tells us that a new set of pictures will be introduced this year). But I think as people working for the ultimate aim of health promotion, it would be more important to know if the responses from those interviewed actually translated to action - and the best way to do this would be to look at actual cigarette sales.

    In fact, I would be more interested to know if per capita cigarette consumption actually went down since the introduction of the graphic warnings, rather than what 1300 people felt about them.

    By Blogger angry doc, At May 19, 2006 4:45 pm  

  • They also did not include baseline figures for intention of change prior to the survey, because with every sample there is a natural number of people who are attempting to quit, or cut down, etc. Whether it's an improvement or decrement we don't know.

    And yes per capita cigarette consumption would be a good figure to use as comparison and perhaps easier.

    Anon: every research paper is skewed just like every single human is, just how skewed the report is that's all.

    By Blogger Unknown, At May 21, 2006 12:48 am  

  • haha blindcat a well designed study should minimise random and systematic errors. Skewed studies are poorly designed or used to promote an agenda. A good study should not deviate too much from the truth.
    If all studies are skewed,it wouldn't make much sense to use them to make decisions would it???
    :)
    To me this piece of work sounds more like propaganda than truth. With out basing on stats, i feel that the society seems to moving towards a smoking culture. I could be wrong but first reaction to the report is not one of trust but doubt. It doesn't fit my instinct.
    I guess people just ask for the happy family pack. I don't think it serves the purpose. Too much graphic exposure might just serve to desensitise people. Just my hypothesis.

    By Anonymous Anonymous, At May 21, 2006 1:59 am  

  • Of course, there are other decisions to be made besides how a study is designed - such as why the topic was chosen to be studied, how to interprete the findings, and finally how much if any publicity is put on the interpretation.

    Smoking prevelance in Singapore may actually have been decreasing over the years, at least according to the MOH National Health Survey:

    http://www.moh.gov.sg/corp/hottopics/NationalHealthSurvey/index.do

    But the methodology used here is that of a questionnaire too, so instead of concluding that fewer Singaporeans smoke now, one may also conclude that fewer Singaporeans admit that they smoke now.

    Wouldn't a simple figure on the number of cigarettes imported/consumed per capita be more informative?

    I'll try to see if I can find the figures...

    By Blogger angry doc, At May 21, 2006 12:38 pm  

  • Anon: true that, but i'm a skeptic so i believe that every resarcher has his own agenda, so everything's skewed. Plus, is objectivity an ideal or reality? I think it's still only an ideal that we can work towards to. Actually got a really long explanation to the 1st pt but it's not my blog, don't wanna take up too much space. It's my own little skewed theory. Heh.

    By Blogger Unknown, At May 21, 2006 5:23 pm  

  • Thank you for your article, really effective piece of writing.
    glance | 2 | 3 6 | my | 3 3 | here

    By Anonymous Anonymous, At February 17, 2013 12:27 am  

Post a Comment

Subscribe to Post Comments [Atom]



<< Home