Tobacco Documents Online

---

At this stage in our thinking we are questioning the indices that have been used as measures of smoke exposure. We have documented well the unreliability of the number of cigarettes smoked as such an index and we have recently come to suspect large variations in the amount of the smoke actually retained from the puff of smoke drawn into the mouth. Our current ~,.~. .... .~. effort addressed to the question is focused upon the behaviors /."~~i." ~,~ ... involved in the inhalation of the smoke. • ,,~.~.,;:,. .... .~ .... ~". ,' .... -. "~'~-"--: Underlying all of our work in this area is the conviction that ,:"< ..... ' " .. what the smoker gets in the way of smoke is independent of • '"~;~'~, ....... ~ ..... s, moke concentration levels as delivered within the range of ... commercially available cigarettes. He has a variety of regul " .tory maneuvers at his disposal for accommodating supply to a ~fairly constant need. To monitor all of these maneuvers " ~~ simultaneously is a major objective of our Behavioral Research

- 6 - But there are two considerations that would constrain us from inferring that these few smokers have been or are being alienated and thus are responsible for the decline in sales. First, there has been a five-year interval between these two Marlboros. It would be ludicrous to speculate that a smoker would be able to meaningfully compare today's Marlboro with his recall of the 1970 Marlboro. Nor would the relative sales performance of Marlboro and Winston (probably the closest cur- rently commercially available brand to the 1970 Marlboro in terms of delivery levels) suggest a Marlboro to Winston movement Secondly, during the past five years there has been a downward drift in the delivery levels of those brands which have been increasing their market shares. That this drift has not been _,v~merely a function of the increasing determinancy of the health- scare in brand choice is to be seen in trends in acceptability ratings A series of studies* conducted by the Behavioral .... Research Laboratory from 1972 to 1974 permits the generalization that regular filter smokers find 14 mg tar delivering cigarettes as acceptable as cigarettes comparable to their 16-18 mg pre- , .ferred brands. This observation begs the acceptability question because there was only marginal evidence that differences wer~ Another recent study (report in preparation) was a well designed and cleanly executed methods study in which cigarettes at 12 mg and 22 mg of tar were s~oked by a large POL panel of 85 mm non- menthol filter smokers in comparison with a 17 mg Marlboro-like control. Nicotine/tar ratios were constant• In this study the differences in delivery levels were detected. The 22 mg ciga- -rette was clearly rejected in favor of the control, but a low- draw 12 mg cigarette was marginally preferred over the control ~.:~.~ Such preference results could not have been had in 1970. -.~.,in sum, ~hen, our R&D information on smoker sensitivity and smoker preferences marshals rather convincingly in support of . the statement that smokers are,~with but few exceptio " ns, ~nsen- . sitive to the delivery changes that have occurred in the and in those cases where difference may be detected, preference trends are such that-the probabilities favor preference for the lesser delivering product. .~ LesC this conclusion be construed as inconsistent with the postu late that the seasoned smoker smokes for his nicotine quota, some words of explanation are in order. Both the literatur~ and the internal P,M.-R&D findingsto date ........ ,,~, are not fully .supportive of the postulate. Suggestive, yes, but there has been no study which revealed the smoker to be titrating~C:~ to t~e degree necessary to maintain constancy in his nicotine s~pply when he is faced with variations in delivery levels of the~:~ .c~garettes smoked~ or when he is pre-loaded or pre-deprived~.

- 5 - A more incisive question would be whether the downward step-wise nicotine change from 1971 to the presen~t (plateaued from '73 to '74) has resulted in a net nicotine decline that has made for a discernible difference. The 1970 Marlboro delivered 19.4 mg of tar and l 33 mg of nicotine. The 1975 Marlboro (April C.I.) delivers i6.9 mg of tar and 1.06 mg of nicotine. Are these two cigarettes d iscernibly different? . . ~ ~ L . ..~,.~., _..,.~..,. .. There are three relevant studies which would support the judgment that smokers at large would not detect this difference even when smoked comparatively at a point in time. -<. ~... ,:~, .... -" ~. • 8N0-I (Sus~ Noticeable Difference), ~epo~ed in a ~e~o &ugus~ 8, ..... '~I~7Z~ S~ho~i-~o Dunn, ~s a ~a~o~ ¢ield ~es~ ~i~h 550 ~egula~ ., '~ filter smokers smoking 2 packs each of five cigarettes ranging -.,~.':~; " incrementally in tar(and nicotine) deliveries from 23.9 mg (1.30 rag)to 18.1 mg (I.05 mg). There were no differences among the acceptability nor the strength ratings ~-..,..., .,,~,...:... .... . ~ND-Z, ~epor~ed in ~e~o qo~ Ma~ch 30, 1973,"Scho~ ~O"Dunn, • . a ~ollo~-up ~o ~ND-I in ~hich ZOO smoke~s o~ 85 ~ non~en~ho] ti1~e~ cigarettes s~oked $ packs e~ch o~ ~o ~a~Ibo~o-like c~g~-.~,_~ rettes deliverin~, in tar (and nicotine) 20.7 mg (1.43 mg) and .... ,o ,or ratings were found. ~s qui~e si~ila~ ~o ~he interval bounded bd ~he I~70 and I~75 Marlboros. ' .~.,,~ :"~~,~":Th~ ~ep~t~of the third study is in preparation. Frank Ryan ...... . ~ .randomly mixed cigarettes delivering 7 mg of tar with cigarettes " .. ,~: ~ delivering 14 mg of tar. Nicotine varied concomitantly with tar. ...... ::~. Five packs of these mixed packs were smoked against 5 packs of . ~ "I0 mg tar delivering cigarettes Responding smokers (250) gave - . no evidence of awareness of the 2-to-I delivery level differences in the mixed pack, even though they were provided the opportunity These three studies bear out the remarkable degree of insensitivity ~ of consumer panel,s to major delivery level differences in ciga- rettes, and the findings are wholly con,sistent with our general experience with smoking panels. On the other hand our 15 .years of smoker testing would lead us to believe that there is a small percentage (5%?) o~ smokers who would be able to perceive the 1970-1975 Marl.,boro difference in a side-by-side comparative test~;~.~,;~. . . ~.~ .... :

O~ ~U M/W • I x 100 ! I

BILLIONS OF UNITS ~T~:~O000¢ ~0 0 ~ r-

- 2 - If one then plots the corresponding*-nicotine delivery levels 18 16 I I I 69 70 71 72 73 74 75 YEAR It would be approprCate before addressing the question to chrono- logically fix the onset of the sales Cncrement decline. This is not so simple a task as it might at first appear; but for this discussion we shall~ assume that the time of onset is mid-'74. The annual incremental percentage declines subsequent to 1970 up to 1973 are not interpreted to be so portentous as the more recel~t developments. In fact there is some question as to whether they constitute real declines. Myron Johnston is preparing a statement that treats this issue in some depth, ..:~ .... ..~;~..~.. -- . ~., .,'. ....~,~ - " *Taken from A r~l C,I. Re orts~~~. ;~~ ........ ~.~,~ ~-,- ~.~ .... Fig. 2 - Marlboro Annual Sales'Increment in Percent. plotted with Nicotine Delivery Levels (from April C.I.) " .:-.... for those years, one observes a close parallel in the two curves ~.~ .... -"~~.~..~throughou~ the period. . . .51~. ~. " [Note needs to be made of the fact that the same pattern would -..~-~.- be evident were one to plot tar levels; nicotine has been singled out for attention because many investigators of smoking behavior, " including some P.M. R&D people, have contended that the seasoned .... cigarette smoker smokes for the nicotine in the smoke. In view of that contention, t~.e question has been raised as to whether " the declining nicotine delivery values reported for Marlboro Red could be responsible for the declining sales increment.]