Estimating the Prevalence of Alcoholism: Modified Values in the Jellinek Formula and an Alternative Approach

Abstract

The critique by Brenner² of PD in the numerator of the “Jellinek formula” for the estimation of the number of alcoholics is perfectly valid. His substitution of (D–Dn) 100 for PD is a definite improvement and his use of 4L is–for the time being–the best estimate of the unknown quantity Dn, but it tends toward a maximum estimate of the quantity D–Dn. That P is not a constant is evident but it had seemed, when the formula was first developed, that within a range of 10 years the fluctuations would not be large enough seriously to interfere with the estimates. More important is the fact that P cannot be assumed to be the same in all the states of the U.S.A. and that PD does not make allowance for the possibility of deaths from alcoholic cirrhosis of the liver dropping to zero. My interest in the estimation of the number of alcoholics had dwindled when it became obvious that alcoholism was a public health problem of such magnitude that preventive, therapeutic and research efforts on a large scale were imperative. Furthermore, other aspects of alcoholism filled my interest and time to such an extent that I abandoned the much needed further development of the Jellinek formula or the search for another. In discussions following lectures which had nothing to do with the estimation of the number of alcoholics the matter of the Jellinek formula was nevertheless frequently touched upon and on those occasions I expressed the opinion that not only the value of P but also the value of K and R require at least periodic revision.