Smoking index is generally defined as the number of cigarette-years (or pack-years) smoked. This ignores the parameters such as age at initiation, time elapsed since cessation in case of ex-smokers, passive smoking and other forms of smoking such as cigars and pipes. The burden of smoking in a person depends on all these and much more.
While trying to locate a comprehensive index, the only one I could find is the one given by Indrayan on pages 228-230 of his book Medical Biostatistics, Second Edition, (Chapman & Hall/CRCPress). He defines:
Smoking index, S = (3 - a/15)[1/2 sq.root(sum of psubk*nsubk*xsubk) - y]
(S=0 for S<0),
where a= age at initiation in years; a=30 for a>30;
psubk=adjustment for the type of smoking for xsubk years; (Suggested values are psubk= 1 for regular cigarette, psubk = 0.67 for filter cigarette, psubk= 0.15 for passive smoking, psubk= 5.0 for cigar and psubk= 2.5 for pipe)
nsubk= number of cigarettes (or others) smoked per day;
xsubk=number of years of specified smoking;
y=number of years elapsed since quitting.
Indrayan¢s smoking index is more comprehensive than any other I could locate but is based on several assumptions and lack validation.
Does anyone know or can provide a reference for more comprehensive index of smoking which is validated for a variety of setups?
Nezhat Shakeri
|