Tobacco Documents Online

p. o ~. - ! IXCO+Xi| 11~ T|I 510111 D.F~. Yerbur~on, Ph. E. Department of Psycholo£y, ReadlnC Un~versi ~y, ~eadin~, RG6 ~AL, U~ITED KINGDOF. J Received 1 5 OCT 1984 l ~__~_~ .i ; me.mdb BAT Co LTD - MINNESOTA TOBACCO LITIGATION

°. ABSTRACT Xn s~i~e of the considerable ~ublici~y about the heal~ hazards of .~/~f/ smok~J~j, people continue to smoke and so smokers must =onmider tha~ LA~ ! the risks are outveig~ed by snok~J~'s benefit. ~hin" hAch.ly selected Ct i-~ j reviev of nicotine and the enok4~g habit reveals that nicotine does ¢~t~ ~ ~ave positive effects. Niootine ~elesees hormones vhioh reduce fsticue and acts O~ the centre2 nes'vous system 4:0 ]produce nora e£ficlent processing of Infornetlon. The increased eff£ciencp produced by nicotine enables bo~.h smokers and nonsmokers to perfor= better in v©rk sltuatAona. In addition, nicotine has a sedative ectioc reducing sz~ziet7 and anger. Smokers tltrate their nicotine intake o© "that t.he7 o~.ain the appz'opria~e dome of ~cotine for these kinds of effects. The pharnacok.lnetlc propertAes elv nicotine make smok-~nd~ doses rez~ax'kab27 eLSe £or noz~nL1 bea2tby adults ~ com~riso: vi~ other available stimulan~ and sedative Sube~4Luces and no 'there is a ~ benefAt-risk x'8~£o for n.ico~Lxte versus other coe, p~rsble agents. ~£ the other components of cig8~tte smoke could be ,,ads less active, the unique pharmacological properties of nicotine make it an ideal substance ~or self-aedAca~:ion by Az~halatlon. w 0 0 BAT Co LTD - MINNESOTA TOBACCO LITIGATION

b INTRODUCTION Smoking has been indicted by health authorities arotumd the world as a habit which impalra health and shortens life. According to epidemlologists, smoking causes ill health and premature death through cancers of the upper respiratory tract and Zung, chronic bronchitis, emphysema, and coronary heart disease (USPHS, 1979)- It is felt by many that, even if it does not always cause death or disablement, there is hypermorbidity for these diseases among smokers. Joseph Califano, former Secretary of the US Department of Health, Education and ~'elfsre described smoking as "Preventable Public Realth Enemy Number One" which has been convicted, "beyond reasonable doubt, of crimea against the public health" (Califano, 1980). In this climate of political and medical opinion, it is not surprising that researchers are reluctant to state publicly that smoking cam have any positive effects for the smoker, add yet common sense argues that it must have. Smokers are exposed to considerable publicity about the health risks amd so every smoker must have made some ~udgement about ~C,.;.~ the additional risk to health of continuing to smoke. Since cigarettes continue to be purchased, we can only cohclude that smokers ~..~ consider that the risks are outweighed by smokin~'s positive effects. This belief of smokers about the beneficlal effects of smoking is substantiated by this selective but balanced review of the literature on nicotine and the smo~ng habit. II NICOTINE PHARRACOKINETICS A 1.2 mg cigarette will give a mouth level of 150-250 ug of nicotine (Armitage, 1973)- The pH of the mainstream smoke ranges between 5-5 and 6.2 for flue cured cigarettes and between 6.5 and 8.8 for cigar and pipe smoke. The level of alkalinity is crucial in determining the sits and amount of nicotine absorption from the smoke aerosol. A. Absorption Nicotine from tobacco smoke is absorbed from the mouth, nose, and lungs and digestive tract with the alkalinity determining the amount absorbed. I. Oral Absorptiom Nicotine base is readily absorbed by the buccal membrane. However, the amount of free base depends on pH, so that where the ~F. is 5-3~, about O.4 per cent of the nicotine is present as the free base, while 6~ p~ 8.5 (alkaline), e~ per cent of the nicotine is ]~reaent as the free base. Beckett and Trigge (1967) found that people take about 6% from a 1.2 mg nicotine solution at pP." 5-5 and 25% at pH 8.5. Umpublished work by Dr F. #. R Russell and Dr ¥ Washes showed that ors! absor~tio~ fro= tabletE conteir4~£ 1.5 =g nicotime gave venous levels of 6.0 fig/ml at pF" 6 and 10.5 ng/m] at pP 9. Thus, as little ms ~C~ of the nlcotine is absorbed orally free cigarette smoke (pE 5.=~ Ic 6.2), although much more is taken uI from cigar smoke (Arritsge, 1973). Ae s consequence of the poor ora~ absorption, a crucial ~er~ cf cigere~t~ smoking is further smoke manipulation by i~haling i~ an.~ ~hez empe!l!ng it through the mose and mouth, or eve~ Just thrc, u~,_ ~h~ 3 BAT Co LTD - MINNESOTA TOBACCO LITIGATION O O C) O O~

nose • 2. Rassl Absorl~tion The habit of snuff taking is found in many cultures and it has been suggested that nicotine is absorbed by the nasal mucosa. Recently, Russell described the time course of plasma nicotine after snuff taking (Russell, Jarvis and Feyerabend, 1980).. Uptake of nicotine was extremely rapid and concentrations of over 20 ng/ml were found in blood samples from a forearm vein. Thus some nicotine must be absorbed from the Dose but it is a small amount in comparison with the uptake from inhalation. Inhalation The ma~or site of nicotine absorption for the majority of smokers is the lungs. During inhalation, the smoke aerosol passes down the bronchii into the alveoli and absorption occurs through the thin alveolar membrane into the pulmonary capillaries. Nicotine diffuses so rapidly across the alveolar membrane and the velocity of blood flow through the capillaries is so slow that equilibrium is probably reached between alveolar nicotine and capillary nicotine. On the basis of the previous estimates of a TO to 250 ug mouth level of nicotine from each puff, over 100 ug would be taken up during each inhalation from a medium delivery cigarette (Armitage, 1973) giving over 1.0 mg of n/cotine per cigarette. The time course of nicotine in human plasma has been studied most extensively by Russell (1976). Smokers puffed ten times on a cigarette and there was a rapid increase in p3asma nicotine in the forearm vein with each puff with irregularities in the ascent profile from the puff by puff boli of nicotine. Peak venous nicotine levels of 15.5 to 38.4 ng/=l were reached at the end of the cigarette, about one fifth or one sixth of the carotid artery levels. The estimated overall half-life in humans is round 20 mine after finishing the cigarette falling to baselim~ levels of about 7 rig/el in 40 mine. Gastric Absorption Gastric absorption only plays a small part in nicotine uptake from cigare~tz smoking in normal circumstances. Trsvell (1967) showed that nicotine was rapidly absorbed from a cat's stomach when the solution pP. was alkaline but not when the solution was acidic, the normal gastric pH. 5- Summary The complicated practice of puffing on burning tobacco leaves, inhalin£ the smoke and blowing it out through the nose an~ mouth enables the efficient trane£er of nicotine from the tobacco to the sv.oke r's bloodstream. Distribution After absorption i~to the pulmonary capillaries, the nicotine-loaded bleed leaves the lungs via the pulmonar~j veins into the h~art. Prom •hert, the z/~'otlne ia pumped out into the aorta from ~hich tx:e larg~ arteries branch off. The significant branch for the smchi--c ha%it is ~h~ carcti~ artery which lead2 directly tc the brain. Abou~, a fifth cf ~ze b~.mcd fro= the heart ascends im th~ csrcti~ ar~ry e: ~hat a f--ftL cf ~he sbsorbe~ nicotine passe.- to the braim with!~_ ~C ~ecs. BAT Co LTD - MINNESOTA TOBACCO LITIGATION O O O O"- Co

(Ollendorf, 1977) i..e. • dome of x'ou~l~'~fron m n~ll±'uum delive~Qr cldarette on the baals of the earlier u~ptions. In order to act on the bra£n, a substance must penetrate the lipid blood-bra±n barrier to the brain eztracellular fluid. Hicotine t8 soluble in Ztpido and so passes posses throush t)sJ.o I~rrAer. )tat studios hove compared the percentage of nieot£ne rena£n~n8 in the brain I~ secosdo after I z~]~Ld intracoro~id ~ectlo= v4th tz~Lt~1~d voter u a standard. B~ne~ ]per ¢e~t of the tritiats~ voter 4a take~ up on the fAreS pass through the brain, and nicotine uptake is 1~1~ lil ~-~ l~hat o£ tz-Ltistod voter. (Oldondorf, HTnan, B~wn and Oldendor~, 1972) ~X ~,~7,-~ i~e. %:~ually all the D£col~Lne that ~s delivered to the brLl~ leaves "- the blood i.e. about 250 u8 nicotine per c£larette. AO • result o~ this efficient uptake of ntcoT~ne, doses affecting the brain can be ob'lut:Lned v~th a~elativol7 ~ov blood levels vhtr.b m~Ln~oes the r£sd~ Of ~iz~ tO~L~C:~r tO other orKonn. Hicl'oentored*od~Tons o-~ttr 14 C-~cot:'u0e ~-~l~ ~;.~::~ H-n£co*Lue show rsdioaol~y ~n sorties1 ce~la, h£dh levels 1: the -~ t~/~ ~ ~ t~- h~ppocupue, the cerebellum, a~d :mclei o~ the ~y~othsiamus sJ~ brown soot (S~hmi~erlov, Kansas, appleKw~n and Boffnan, 195~). Th£s patter= of nicotine dist~bution throughout the brg£n 811ov| vide scope for ph~rnOCOdJq~aL~C intor~ction. The broS.= does not motsbo~ieo n£cot~se but l* washes out qudckly From the b~8~u v~b ~ve8 a 8ho1~t duration of sct~. Thus a£cott~e 4s 8 drug which is rapidly o~orbe~ Lu~o the bra~u, widely distributed and then quickly renewed; the idoa~ o~ecificotion8 for • suba~ce that is ~equAred for 8 short durat~c~ oF sc~oa. ~.'. ,~C L"r¢"! ~:; ~-." Sico:Ane's metobo~ion add ezcretion have been discussed ~n numerous publmcations (RusselZ, 1~76). Rovever there are two aspects of ms,or ~mport&nce to the onokinK habit. ?ireS, i~ is ¢~ear from the ~i~e course that nicotine in metabolined vet7 efficiently by the liter, vbtcb 34mite ~icot4~e'o d:tion of lCt£o~ ~n ~he body. Second, the metobol£tss e.I. oot~n4ns, non-nicotine, etc. appear to be ~£rtua~y ~oct£ve and so the phlrmlcodT=lnic effects m de~e~4~ed s~somt completely by the so~Lo-~ o£ ~oot~ne alone. C ODC~USiO~ ~tudAes of.nicotine pharmscukinetics have revealed that it is a substance v~4~.~ is a~orbed very efficiently, readily e~tors and is quickly e14m~.uated from the bra~Ln, and is rapidly notabol~oed to relat~vely ~.wsetive netabolites. Th~n ~hs~nocok~netic ~ntter= L~OVS a brief duration of action and the poosib£~£t7 of centra~ nervous action with minieus side effects from actions on the rest of the body. The z~e&li=otion oF this possibili~y depends on the smoker's ab~ll~7 to c~ntrol nlcot~no 1=take. III R~COTZEE CONTROL £ ~a~or argument fi~r the relevance oT nicotine to the smoking habit co~es ~ro~ evidence tha~ people smoke ~c ob~&in nicotine 8=~ even regulate nicotine in~ake to obtain speoi~io leve~s o~ nicotine in their ~oodo~re£:. q~.e 1=cp~.Icaticr, cf ~itro~ion is ~hs~ smokers have a need for nicotine and possess a mechanisn An ~be body which i~ sensitive to nAcotAn¶ so e=abllng then tc c~ntr~l ~be dose. Evidence • or nicotine need arid control has co:e fro: su~eys of clgare~te 0 mmmd~ 0 0 Q m BAT Co LTD - MINNESOTA TOBACCO L1TICATION

preference, studies of L:~L~ation, s~u~ies of daily 8moL-L~ ]~te~s, the titration of IDdlvldual cigarettes, Gicotine preloading and s nicotine Lu~aEonist e%udiea- A. Preferre~ Cigarette Brands I~ 1957, only one per cent of ~he cigarettes i~ the United States were falter tip and the average cigarette delivered 2o~ mg of nicotine and around 26 mg of tar. After the adverse publicity from the Surgeon Genera1°s report (USPHS, 1964), and the report of the Royal College of Physicians (RCP, 1964) in ~ritaln, the average levels cf nicotine in the USA, as determined by smoking machine analyses had dropped to 1.5 mg in 196A. It might have been expected that the comhinatlc= of governJent publicity and co,niL:tony promction would have pr~uced an even more remarkable switch i~ brands to flower nicotine and l~ar yields in the followi31g ten ysa~s, and ~he sales-weighted 8verlge nicotine per cigarette would have decreased a further 1.0 rag. In fact, cigarettes with a nicotine co:tent of 0.~ mg and correspondingly icy levels of particulate matter are not popular, and the sales-weighted average level of nicotine in ~[ cigarettes has been 1.5 - ~.~ mg per cigarette for the past 8 years. Nicotine-free cigarettes have been a tota~ disaster, even though many people have tried them. 22 could be argued that cigarette preferences do not provide convincing evidence for the importance of nicotine because nicotine usualZy cavaries with tar. However, one controlle~ study has tested cigarettes wlth independently varied nicotine and tar levels. (Goldfarb, Ja~vik a~d Click, 1970). Smokers were a!Io~'e~ tc smoke as many as t~ex wished of these cigarettes an~ this number corre~ste~ • -Ith nicotine content but not tar content. ~evertheless, i~ is interest~ that non-nicotine cigarettes were smoked tc some exten~ over the three weeks of the teat when there ~ere nc o~hsr alternatives. JL1though low and sere nicotine cigarettes allow the smoker to go through the rituals of lighting, manipulating, and puffing the cigarette as yell as inhaling the smoke, the lack o~" nicotine results in" lower consumption. Ho~ever, it cannot be concluded that flavour from the ~ar of to~acco smoke plays no part i~ cigarette acceptability.. Nevertheless, the inference from this work is that nicotine is an essential ingredient of the cigarette for the smoker and this conclusion becomes eve= clearer where e more complete measure of smoking behaviour, than number of cigarettes smoked, is used • 5. Inhalation Smoke inhalation results in very efficient abso:-pticn of nicotine (S+s Section I~) and the large ~e~entags of smokers who do inha~e provider evidence '~at a.~ aim of' smok~.ng is to o1:ta:L= nicotine. Doll e.~8 ~-i!2 (1964) found that 80- 90% of cigarette smokers re,offing inhaling, and Hknmon~ (1966) found ~hat 96.4 per cent thought the•. they vere moderate to dee]~ iJ~haler~. C. ~Cnohin~ ps t', e .'~: BAT Co LTD - MINNESOTA TOBACCO LITIGATION

t The hypothesis ~hat smokers attempt to maintain minimum, or above, levels of nicotine is supported by studies of plasma levels of nicotine throughout the day (Russell, Wilson, Patel. Cole and Yeyerabend, 197~; Russell, Wilson, Petal, Yeyerlbend and Cole, 197~). The half-life of nicotine i~ plasma (see Section II) is about 20 mine and habitual smokers consume 15 to ~0 cigarettes per day, i.e. a cigarette eve~ ~0 ~o 50 minutes, ezcluding, meals and sleep. Russell's studies demonstrated ~at the mid-mcr~nK levels of plsJtma nicotine were remarkable con•intent within 8ub~ects •cross days, although individual levels varied widely. Five hours later in the ~ternoon, these levels were either virtually the same or higher. ~icotine Titration In this part, Z wtl~ consider the direct evidence for s:ckers controlling their nicotine intake lu terms of cigarette consumption, smoke generation ~nd smoke manipulation. I. Cigarette Consumption The number smoked is ~he most obvious way for smokers tc control %heir nicotine intake and it is the eaaies~ to study. The first studies examined changes in cigarette brand a~d consumption. a. Cigarette Switching: The findings of RusselZ st al (1975), which strongly hinted that smokers control their nicotine intake were erplored further by switching the subjects fro= their usual brand (1.5 mE average) to either • ~.2 mg or a O.D m£ nicotine cigarettes on different days. The number smoked was recorded for five hours. For • he high nicotine cigarette, consumption dropped aicnifica~tly by ~7 per cent and increased by 17 per cent when they changed to • lo~ nicotine cigarette, which was not si~nlfican%. ~owever, su~ects would have needed to smoke five times as many ~o compensate if the.~ did not chan~e ~-y other upect• of ~heir smok4n~ behaviour. In • two hour experi~ont, smokers were switched to cigarettes which were either r- • 0.2 mE n~cotine delivery or a 2.0 m~ nicotine delivery (Jar~ik, ~opek, Schneider, Baer-¥eiss and Grit-, 1978), with identical tar levels as that the taste difference was small. Sub~ects smoked significantly more o£ the lover delivery than the higher delivery cigarette which gives ~ersuasive support for nicotine regulation. ~n a study of smoking in real life, smoke~s were given a I.£ rag. nicotine cigarette for the fir•t week, a 0.8 mE ciE•rette in the second week, amd a 0.~ rag. cigarette in the third week (Turner, Sillett and Ball, 1974)- The subjects compensate8 b v smoking sigm.lfica~tly more ciKarettes when they switched fro= the ~-4 rag. to the 0.8 mg cigarettes, but there was little change after the secon4 switch from C~8 mg to the 0,~ mg of nicotine product. ~ ~onger study (Yr~eedmmm ~ Fletcher, 1976) examined the oh•ragas im consumption over a 20 month period when smokers were switche4 from a conventions/ 1.~9 m£ nicotine cigaretSe t0 • 1.01 mg cigarette which contained ~O per cent tobacco substitute. The average number of cigarettes smoke8 stayed conJ~ant.~r the first 15 months, but increased slightl.~ in the las~ 5 months. A reanalysis od thls data by Bavbone (1976) shoved a 8iE:nificamtlT h/gher ratio of obae:-wed to expected n~cotine ~'hem the emoke:~ were s=okim£ %he 1o~er delivery product (grou~ I P<O.O01; ~TOU; 2 P<0.OO~)which i~icetes compensation im reletio= t~ the h/Eher O 0 0 0 BAT Co LTD - MINNESOTA TOBACCO LITIGATION

delivery product. sophisticated study of cigarette switching (Fageretrom, 19B2) studied the titration of smokers gives either their own brand, average yield of 1.1 Bg of nicotine, or one of two experimental cigarettes, one was a 0.52 mg. nicotine cigarette while the other had identical composition but was sprayed vith niaotine to deliver 1.1 mE. The subjects could differentiate them by taste but not by any other characteristics. Some evidence of i~creeaed consumption of the lower delivery product was seen but the change was not large. In an earlier similar s~udy by Yinnegan, Larsom and Haag (1945) used tobacco leaf with a naturally low nicotAne content made into a 0.34 mg nicotine cigarette or the same leaves sprayed with nicotine to give a 1.96 mg cigarette vith the same pressure drop and tar level, a~tbough ~be nicotine would give a slightly different taste. Subjects were given the high nicotine product followed by a month on the low nicotine brand and then switched back to the high nicotine cigarette. There was no correlation between the number of cigarettes smoked and the nicotine level, but an Interesting picture emerged when Russell (1976) examined the data on wlthdrawal sylnptoma in the Finnegan, et al study; subjects who did not increase their consumption am the Io~ nicotine cigarette experienced lack of satisfaction, irritability and poorer concentration. Further evidence for ebmtimence symptoms in the absence of compenaatiom ~as found in a'compariaon of an heavy smoker group (st least 20 cigarettes e day) and a light 5maker group. Scbachter (1977) found that, as a group, the heavy smokers smoked 24 per cent more cf e O.3 mg nicotine cigarette than a 1.3 rag. nicotine brand, ~hile the light smokers used 16 per cen~ more of the low nicotine compared vith the high nicotine cigarette. Schachter noted that three heavy smokers, ~hc experienced severe abstinence symptoms, had on~y Increased their consumption ~y 14.3 per cent compared with 33-6 per cent for the rester the group. Thus the last ~wo studies suggest that smokers who do not regulate their nicotine ~take, suffer deprivation syap t ores. b. Ventilated Nolder: Another method of varying smoke delivery to the smoker is to use a norma~ cigarette but smoked through a ventilated cigarette holder. Two holders producing nicotine dilutions of 2 per cent and 56 per cent were used to study titration (Sutton, Russell, Feyerabend amd Saloo~ee, ~978). Consumption remained constant throughout the week of study. ~o~ever, as ~e will see later, compensation was achieved by ~ucreasi~ag the amoumt of smoke inhaled. c. Partial Cigarettes: The number smoked i~ a ~eek has been examined when subjects ~ere on~y allowed to smoke part of a cigarette. Goidfarb sad Jarvik (|972) gave smokers cigarettes ~hich ~ere either cut Am half Or ~ad a line drawn around them halfway do~n, and the number of smoked cighrettes ~'as cou~ted. Smoking increased ~'ith the l~med cigarettesamd the cut cigsrztte, but during the fourth week oz their o~-m brand smoking ~as alac higher. The data give limited support for titration because ~2 subjects did i~crease by s~ average of five a 6av o~ the lined cigarettes and by a~ average of eevem e day am the cut cigarette~. A more exte~si~-~ study (P, usse~, Sut~cr. O O C~D Cr~ BAT Co LTD - MINNESOTA TOBACCO LITIGATION

~eyerabend and Cole, 1978) tested a full length 1.08 mg nicotine ci6arette, a threequarter length 0.83 mg D/cotine cigarette and a half length 0.67 mg cigarette in the laboratory. There was a clear and significant increase in consumption with decreasing delivery. d. Nicotine Preload: A fourth approach has been to preload smokers with nicotine. Johnston (1942) injected 20 mg of nicotine intravenously and reported that smokers did not want to smoke for some time afterwards. No details were given of number consumed. In a follow-up study (Lucchaai, Schuster and Emley, 1967), subjects were given intravenous infusions of nicotine and their cigarette consumption examined. Low doses of n/cotine had no effect, but 4 mg par hour very significantly reduced consumption cigarettes of unspecified delivery by 27 per cent. Nevertheless, this decrease in numbers is small in comparison vith the amount of nicotine given. Similarly daily doses of five tablets of 1.O mg of nicotine reduced the number of cigarettes used by only eight per cent although the decrease was significant (Jarvik, Click and Rakamura, 1970). Nicotine chewing gum is used as a stopping-smoking aid, and in a test of it (Russell, Wilson, Yeyerabend and Cole, 1976), subjects were given either alkaline gum with nicotine (a high pH increases buccal and gastric absorption), or a placebo, both highly spiced to hide the taste. Subjects reduced their smoking on both gums but significantly more with ~he nicotine gum. Plasma nicotine determinations showed that nicotine was absorbed from the nicotine gum although blood levels were lower t~an after smoking. Clearly, im spite of a placebo effect, there was reduction of consumption by the absorbed nicotine. e. Eodified Excretion: A fifth type of study manipulated nicotine body levels by changing urine acidity. If the urine is alkaline them less them one per cent of the nicotine is excreted unchanged end if the urine is acid then 35 per cent is excreted unchanged (Beckett and Triggs, 1967). Thus, Schachter predicted that increasing acidity should increase smoking (Schachter, Kozlowski and Silverstein, 1977). Ascorbic acid or glutamic acid hydrochloride were given to smokers to increase excretion and there were increases in consumption of 15.6 per cemt and 21.6 per cent respectively. Once again there ~as some evidence for control of nicotine intake by changing consumption. f. Nicotine Antagonists: A complementary study to those on nicotine loading" used mecamylamine, ~'hich blocks nicotinic synapses in the brain (Stolerman, Goldfarb, 7ink and Jarvik, 1973). Smokers were giveD this mecamylamine or pentolinium, s nicotinic ~locker which does not enter the brain, and asked to record the number that they smoked of their usual cigarettes. Y. ecamylamine, increased cigarette consumption by as much as 30 per cent, as smokers smoked more to overcome the partial nicotine blockade in the brain. Pentc!iniu= produced no change ~'hich ruled out any peripheral effects of nicotine on consumption. Clearly smokers were using cigarettes tc ottain plasma levels of =ifotine suff~ciemt tc affect th~ brain. g. SummarT,: ~-n~se studies sho~- that some subjects titrate nicctlne ty changing the number of cigarettes tha~ ~hEy smoke. Some smokers, ~-hc dc not cor~,er.sate, suffered abstinence symptoms from the effects of micctine de~rivaticr,. ~o~ever, negative studie~ may be er~Isi~ed in C BAT Co LTD - MINNESOTA TOBACCO LITIGATION O O o",

terms of changes in either puffing or inhalation. Data, that ebo~" these factors play a part, come from an experiment by Critz, Bser- Yeiss and Jarvik (1976) i~ which subjects were given an equal number of ful~ length cigarettes and half length cigarettes to smoke in a week. Urinary nicotine measures shoved that subjects were able to obtain considerably more nicotine than expected from the half length cigarettes, and indeed almost as much am from the ful~ cigarettes. 2. Smoke Generation Im the last section, cigarette consumption was related to the nicotine content of the cigarettes. These nicotine leve~e were calculated on the basis of nicotine deliveries that were obtained from standard smoking machlnea. These machines enable comparisons of cigarettes, but only produce an approximation of human smoking. In am innovative aeries of studies, Creighton and Lewis (1978 a and b) recorded the pattern of smoking in terms of number of puffs, puff interval, puff volume and puff shape. They found that there were marked interindividual differences and clear, but smaller, intraindividual variations, am that it was inevitable that am individual's pattern of generation rarely matched the smoking machine's standard parameters. This same equipment was used to study smokers smoking their own brand twice iz the laboratory (~attig, Buzsi and ~il, 1982) and they found that smokers had a consistent individual patterm of puffing cigarettes and that vQmen took larger smoke volumes than men which could have been due to them smoking lower delivery brands. The practical consequence of these variations in smo~ing patterm, in terms of nicotine deliveries, was examined using a puff duplicator (Creighton and Lewis, 1978 a). The divergence of the smoking machine yields from the actual values delivered cam be seen by comparing the average amount of nicotine generated after duplicating the smoking patterns for a cigarette having a machine estimated yield of I M mg of nicotine. Smokers derived 2.25 mg of nicotine for males and 2.0 mg of nicotine for females. Clearly the machine estimated delivery is a limited index of the nicotine dose entering a smokers mouth. Some of the previous studies of nicotine regulation recorded puff variables as well as numbers of cigarettes consumed. Lucchesi et al (1967) found that intravenous nicotine reduced the num~er of puffs and the subjects discarded their cigarette earlier. However, a similar study by Kumar (Eumar, Cooke, Lader and Russell, 1977) found no changes in puff number, imterpuff interval, puff duration or puff volume on a 1.3 mg cigarette after 10 injections of nicotine at one minute intervals in order to simulate 10 puffs on a 0.55 mg or a 1.17 Eg cigarette. A companion study, with controlled smoking of either a herba~ cigarette, a 1.3 mg nicotine cigarette or t~o 1-3 mg nico~in~ cigarettes did reduce the number of puffs taken from the cigarette in a dose dependent fashion. Thus smokers can contro~ their nicotine intake by changing their smoke generatiom. Direct studies of this behaviour have employed Cigarette s~-itching and partial cigarettes. a. Cigarette Switching: Estimates of nicotine entering the smoker's mouth have been made from the nicotine deposited in the cigaratt~ fi~ter and the tip's filtration efficiency. One of the firsl studies on butt nicotine al~owed smokers to smoke either a 1.0 mg cr 2.1 mg I0 CZD O Cy~ BAT Co LTD - MINNESOTA TOBACCO LITIGATION

nicotine cigarette (Ashton Imd Watson, 1970). Depth of inhalation wan unchanged hut puff number was larger on the medium delivery cigarette and butt nicotine data showed that about the same amount o£ nicotine was delivered to the mouth from both cigarettes. Titration had occurred by smoking the lower delivery cigarette more intensely and puffing the higher delivery product leas intensely. Zn our own studies (¥arburton and Weanee, 1978) we found similar changes. Smokers smoked both a (M3 mg nicotine cigarette ind a 0.7 mg nicotine cigarette more intensely but s=oked a 1.64 nicotine cigarette, slightly lens intensely. In s longer study by Freedman and Fletcher (1976), in which there were only smal~ changes in the numbers smoked when subjects svltche~ from a I.~9 mg nicotine cigarette to • 1.01 mg product, butt nicotine levels also showed more intense smoking of the lower brand in comparisiom with the higher delivery cigarette. As part of their studies using the puff recorder and puff duplicator, Creighton and Lewis (1978 a) studied cigarette switching. Smokers smoked a 1.4 w,g cigarette for one month, a 1.8 mg nicotine cigarette for the second month and the 1.4 mg cigarette again for the third month. The estimated accoumt cf nicotine that was delivered to the mouth of the smoker was assessed. The level of nicotine delivered to the mouth stayed constant because subjects reduced their puffing intensity when they s~-itched from medium delivery to high delivery cigarettes and increased their necking intensity for the opposite switch. Clearly, smokers cha~ged their smoking intensity in the direction of equalini~g nicotine deliveries. There was no evidence, in this study, that smokers modified the number smoked each day. In another series of recordings of smoke generation, Adams (1978) measured puff number, puff volume and puff duration, as well a butt nicotine. He also found that smokers behaved differently with e 0~ mg nicotine c~garette in comparison with a 1.4 mg delivery cigarette; they puffed harder on the lower delivery product and left a shorter butt. Butt nicotine Lnalysis confirmed that more intense smoking resulted in proportionally more nicotine being taken into the mouths of smokers from the low nicotine brand. A similar switching experiment (Ravbone, Rurphy, Tare and Kane, 1978) studied smokers before and after they switched from their own brands (average 1.22 mE of nicotsne) to am undefined "low" delivery product. Subjects did not increase consumption, but smoked harder on the lower delivery cigarette, obtaining 0.83 mg for low against 0.96 mg for their own medium delivery brands as e result of increased puff volume. A similar finding was obtained in a study that tested smokers with either a 2.5 rag, 1.2 rag, or a 0.4 mg nicotine %-ith similar tar levels (Hern~n~, Jones, Bachmsm and Eines, 1981). Puff volumes were measured and smokers took larger puffs with cigarettes delivering less nicotine than their usual product. In another precisely controlled study, Stepney (1981) switched smokers from their own brands (1-5~ mg nicotine and 19 mg tqr) to either a 1.1 mg or 0.7 mE nicotire product with similar tar levels for three %-asks. At ~eekly imter'.'als, ~he cigarettes were smoked in the laboratory and the puffing recorded using a puff recorder and again puff volumes increased. Puff duplication of the teat brands showed that mouth ]eve! of n,_'c:tin.~ wa~ greater th&z would have bee~ predicted by standard s~c)-ing ~achln~ zm~ BAT Co LTD - MINNESOTA TOBACCO LITIGATION cZ~ CZ) O O',, ¢..j.n

closer to that obtained from their usual brand. As e conaequemce, their exposure to tar was greater than predicted and so smokers were not getting the health benefit expected from the reduced delivery. b. Partial Cigarette: In a partial cigarette study (Ashram, Stepney a~d Thompson, 1978) subjects vere tested with two versions of their usual cigarette:- a full length and a two-thirds length, which was calculated for each individual. The amount of the cigarette normally smoked wan measured and then a mark was made on the paper st two thirds of this length. The aub~ects were issued with the same number of marked cigarettes as they smoked noFmally, so that they were deprived ~o the same degree but not by the ease absolute amount. Puff duration increased and puff interval decreased with the two thirds cigarette shoving more intensive smoking. However, the butt nicotine estimates shoved that subjects did not obtain their usual amount of nicotine, as smokers had done in the study of Gritz et sl (1976), because about 61 ~er cent of nicotine in the smoke aerosol comes from the half of ~he cigarette nearer the filter, and only 39 per cent from the other half (Crits et al, 1976). c. Summary: Estimations of the nicotine delivered to the mouth of smokers shays that there are many different ways of smoking. Subjects tend to smoke low delivery cigarettes more intensively and smoke high delivery brands less imtenslvely, even allowing for uncertainties of the filter tip analysis (Creighton and Lewis, 1978 a). However, estimations of mouth nicotine do no~ represent the effective dose because nicotine is not readily a~sorbed by the mouth and cigarette smoke must be manipulated for maximum absorption of nlco~ine. 3- Smoke Manipulation One index of manipulation is the amount of carbon monoxide exhaled after a cigarette compared with before smoking. Carbon monoxide is absorbed into the bloodstream froz the lungs and not the mouth. When the residual smoke has been expelled from the lungs after smoking, carbon monoxide exchange from the blood to the lungs will occur ec tha~ the level of exhaled, end-tidal carbon monoxide provides an index of lung uptake. a. Cigarette ~witching: In a study referred to twice before, switchin~ from I.~ mg nicotine cigarettes to either ~.2 mg nicotine or low 0.3 mg nicotine cigarettes was compared by measuring exhaled carbo~ monoxide (Russell et el, 1973). Exhaled carbon monoxide for both switches declined and Russell argued post hoc that the decrease with the high cigarettes represented less inhalation while the decrease with low cigarettes was attributable to the lower carbon momoxi6e yield of that product. Plasma nicotine measures showed clear titration in half of the sub~ects when switched to the high n/coSine cigarette while there was some evidence of titration for the group as a whole when switched to the low nicotine cigarette. Strong evidence was obtained from subjects who switched from a 1.7 =g nicotine cigarette with 17.~ m~ of carbon monoxide to a 0.7 mg nicotine cigarette wi~h only 11.4 m£ of carbon monoxide for five reeks (Guillerm, Easursl, Brouaaole, Hyacinthe, Simon, and Ree, 1974). Comsum~tion increased slightly but carbon monoxide levels increased froz 5.7~ per cent to 7.43 per cent even though the machine estimated O G O Cr~ C~D 12 BAT Co LTD - MINNESOTA TOBACCO LITIGATION

delivery of carbon monoxide from the low-medium nicotine cigarette was lower. Two of the studies which recorded smoke generation with varying nicotine, similar tar also measured smoke manipulation (Stapney, 1981; Yagerstrom, 1982). Carbon monoxide levels were higher for the lower nicotine product indicating more inhalation and plasma nicotine and cotinine levels ~u the Fagerstrom study revealed'that similar amounts of nicotine were absorbed from the two products. Additional evidence comes from an eleven week study (Ashton, Stepney and Thompson, 1978) in which smokers switched from their usua~ 1.4 mg nicotine brand to either a 1.84 mg nicotine cigarettes i.e. an increase of 31 per cent in delivery or O.6 mg cigarettes i.e. a decrease of 57 per cent in delivery. The exhaled carbon monoxide levels showed an increase of only IO per cent on the 1.84 mg nicotine cigarette, and were only 15 per cent lower on the 0.6 mg cigarette. Plasma nicotine levels aho~ed a similar patter~ indicating that considerable poet puffing titration had occurred. Urinary nicotine excretion on the high nicotiDe cigarette gave no support for titration when the machine delivery of 3~ per cent was higher than the usual brand. However with the Io~" nicotine cigarette, the 24 hour excretion was 94 per cent. Altogether these studies give strong support for nicotine titration by smoke mani pul ati on • b. Partial Cigarettes: The experiment with half and three-quarter length cigarettes of Russell-et al (1978) also included determinations of exhaled carbon monoxide and plasma nicotine. Although smokers used a greater number of the partial cigarettes, the carbon monoxide and plasma nicotine levels ware as expected, emggestinE no titration. ~e will retur1~ to this contradictor7 result im the nex~ sub-section. c. Ventilated Rolder: The ventilated holder study (Sutto~ el al, 1978) gave no support for titratio~ in terms of the number smoked, but a comparison of the o~served reduction of carbon monoxide by the more ventilated holder at two days and severn days with the expected reduction showed partial but significant compensation which was maintained throughout the test week. There seemed to be no compensation with the less ventilated holder. The outcome, i~ terms of plasma nicotine, was a reduction of only 40 per cent instead of the expected 58 per cent which confirms the effectiveness of smoke manipulation as a titration mechanism. This clear fi~din~ contrasted ~ith the partial cigarette study of Russell et al (1978; Russell, 1980) in which smokers consummed more cigarettes, but their exhaled carbon monoxide and plasma nicotine levels were the same as those predicted if no titration had occurred. This study implies that smoke comcentration is the cue for smoke manipulation because total smoke inhaled increased for smoke diluted by ventilation but no~ wheL the smoke concentration was the same as in the partial cigarettes (Russell, 1980). d. Nicotine Preloa4in~: As one of a series of titration studies ~'e administered an oral 1.0 mg dose of nicotine to smokers ~rior to amokin£ a 0.6 ~g nicotine cigarette (~'esnes, Pitkethly and ~'arburton, i~ preparstio~). The subjects thought they were participating in a study of smoking an~ the pure nicotine o~ head tremor. .~uffin6 behaviour, tut~ nicotine and exhaled carbo~ monoxide ~'ere measured. O O Cr~ ~3 BAT Co LTD - MINNESOTA TOBACCO LITIGATION

No differences were seen in puffing or butt nicotine levels for nicotine and placebo conditions. ~owever, there was a |ignificant reduction of exhaled carbon monoxide a~ter the subjects had received a nicotine tablet indicating reduced inhalation. Clearly the smokers were titrating their nicotine dose by smoke manipulation according to some internal mechanism aenmitlve to plasma nicotine levels. e. Summary: Nicotine intake can be controlled b~ the amount of smoke inhaled as well as by cigarette consumption and by smoke generation. Rawbone et al (1978) found no correlation of carbon monoxide levels with smoke generation measures. Thus, the mechanisms are not necessarily interdependent. As far as nicotine absorption is concerned, inhalation is the final control on intake. 4- Conclusion The experiments in this section show that nicotine intake is titrated to obtain more nicotine from lower delivery cigarettes and less nicotine from higher delivery brands. Unfortunately, most studies have only studied one titration method and found that some subjects do titrate, some titrate partially and some may not titrate at all by that method. Indeed it may be impossible to titrate completely with some low delivery brenda. Evidence for contro~ over the nicotine done is important; it argues not only for nicotine being a necessary condition for smoking but also that smokers are trying to obtain a dose which will produced desired or needed effects. IV. NICOTINE PHAEMACODYNAMICS In this section, ~ will discuss nicotine's action on the nervous system and on l~ehaviour which may be used to account for the smoking habit. ~'ith such a vast literature, only illustrative experiments cam be ci ted. A. Neurochsmical Action of Nicotine Nicotine could modify neural transmission by interfering with one or more of the processes that are responsible for the regulation of transmitters i.e. (I) transmitter synthesis; (2) Presynaptic storage; (3) Transmitter inactivation (Warburton, 1975). There is evidence from in vivo animal studies ~hat nicotine produces changes in the brain levels of catecholaminee, indoleamines and acstylcholine. The crucial questions are how cam these changes in levels bs related to the dynamics of transmitter systems and whether these changes can be extrapolated meaningfully %o explain the human smoking habit. A major problem, with the majority of theme animal studies, is the unrealistically high doses of nicotine that have been tested. A 75 kg person, who takes between O-75 and 3.0 mg of micotine from a cigarette into his mouth, ~'ill receive a dose of between 10 and 40 ug per kg. __ There will obviously be differences because of the route of administration and the metabolic rates of different species, but it is safe to conclude that in mice, rats or cats, over I0 times this does (O~. mg/kg) is well outside the "smokimg" dose range. I. Catecholamimes Studie~ cm the cstecho]ami~es, have been carried out by Lichtensteiger CZD O Cy~ CO BAT Co LTD - MINNESOTA TOBACCO LITIGATION

and his co-workers; (Lichtensteiger, 1979), by Fuxe and hie colleagues (Fuxe, Agnsti, Enero~h, Custafson, Bokfelt, Lofstrom, Sk•tt and Skett, 1977) and Vestfall's laboratory (Westfall, 1974). These studies have given the same answer in spite of wid• difference• in nicotine dose level•, which ranged fred 0.~ mg/kg in one of Lichteneteiger's studies to four doses of 3 hE/kS at ~O sin intervals in one of the experiments of Fuxe's group. Lichtensteiger rap.erred that increased fluorescence intensity of c•techolamine neuroses in the suhetantia nigr8 nuclei, zone compacts and .one iDcerta with doses of O.~ mS/kS of nicotine, suggesting increased activity in these neuroses. This conclusion van supported by recording from the neuroses and biochemical analysis which showed some depletiom of both transmitters. Nicotine produced • 2~ per cent depletion of catecholami~e levels in the median palisade zone of the medis~ eminence of male rats (~xe et el, 1977) hut hexamethonium, • cholinergie blocker, prevented depletion. No changes were found in dopamine or noradrenalim receptor activity in in vitro studies so the ahmn~s are secondary to effects on choliner~i-~ neuro----'--nes. ~q~ ?~ 6~4 ; ~L:/ CO - 2. Indoleamimee Changes in the concentration and turnover of serotonin have been found after doses of nicotine. A dose of 1 mg/kE of nicotine, imtraperi%oneally in,acted into mice, markedly increased the levels of serotonin in the mesencepha~on and diencephalon within 1~ mine but not in the cortex (Essman,1971). In the same study serotoniz's major metabolite, 5-hydror3indoleacetic acid, was also increased but there was a decreased serotonin tux'nover rate of 20 per cent add increased serotonin turnover time (÷500 per cent). It is not clear whether the changes in serotonin in rats after a large dose of nicotine ere important for human smoking. Acetylcholine Nicotine depletes whole brass acetylcholine in ~he mouse (Essman, 1971) in doses of I ms~ks. Depletion of transmitter c0uld be s consequence of decreases synthesis, increased release from storage, increased release into the synaptic cleft or more effective enzymatic activation. There is no evidence that nicotine modifies acetylcholime synthesis (Hrdina, 1974) and the enzymatic inactivaTiom cf acetylcholine ie extremely effective, ~hich argues for a change in either storage or release. There is also for increased free acetylcholi~e at the cortex after a "smoking" dose in the cat (Armitage, Hall, Sellers, 1969) which is consistent ~'i~h both ideas. The question of nicotine-induced changes in ecetylcho~ine storage pools was tackled by Eesmen (1971). He found evidence o~ ~ ~e=rease of scetylcholine im synsptio vesicles and in bound scary]choline a~ the nsoccrtex which suggests that scetylcholine was bet=c re,eased from storage by nicotine. However there was no imcreess ir ±h~ free ecetylcholine pool concentratiom ~hich argues for increase~ :'elease cf th~ unbound transmitter em~ autsequent inacti%~tion ~y ec~%'r~cholinest~rase. T~e phencmen~m of incre-~ze~ release a% the cortex would be explained if n/cotine enhanced presyl~a~tic release but there is nc in vitro evidence Of entrancement (P.rdina, 1974). Thus we ere lef~ with the hypothesis ~hat it is increased ectivi~F in th~ cholinergic neuroses tc the cortex which produces the i= wive ~5 <D <D O -O BAT Co LTD - MINNESOTA TOBACCO LITIGATION

depletion. 4. Conclusion There is evidence that moderate to enormous doses of nicotine change levels of catecholamlmes and indoleamimes although the cstecholamine changes were mediated by choliner£ic pathways. The major effect of "smoking" doses has been increased cortical release of acetylcholine but there is no evidence that it is due to a direct effect on cortical neurones, but must be due to indirect activation of the cortical neurones and this is discussed next. B. NeurgphTsiological Action of Nicotine The action of many pharmacological agents has been explored by using meuronea in the peripheral nervous system which are more accessible than those In the central nervous system. Although cautioz must be exercised when using this data to explain central nervous system phenomena, in feneral, the principles that have been derived from such s~udies have proved useful in understanding nicotine's action on the brain. Thus the first studies will describe nicotine's action o= these neural Junctions. I. Peripheral Rer~ous System The action of nicotine on the nervous system has been know~ since 1914. Dale (1914) established that n/coSine mimicked acet-Ilcholine al autonomic gamglia and neuromuscular Junctions. The effects of doses of nicotine on cell membrane depolarization and subsequent action potentials was examined by Paton and Perry (1953) using the cervical ganglion preparation of the cat. Yhen 0.2 ml cf the solution containing 50 ug of nicotine %artrate was injected into the external carotid artery there was depolarization of the membrane and some reduction of the subsequent action potentials, a change which was similar to, but more transient than, the effects of a small dose of aceTylcholine. Six times the above dose, 0-3 mg of n/coSine, produced prolonged depolarization and the action potentials %-ere abolished i.e. ganglionic blocP~n£. ~ challen~e with • second dose of nicotine after the original depolarization, but before recovery of the actico potentials, produced less depolarization which demonstrated tha~ nicotine was producing s competitive block of cholinergic recepYor~. Exactly the same phenomenon was observed with high doses cf acetylcholine st the motor end-plate so that once again nicotine is mimicking acetylcholine. However it seems unlikely that bloo~ concentrations some IO0,OOO times that found in the human forearm vein ever occur in the smoker's brain and hence unlikely that e depolarization block occurs in brain neuromes after nicotine(the biphasic effect). In summary, "smoking" doses of nicotine precisely mimic acetyloholine and produce the same neural changes that would occur after natural activation of that synapse. The reason for the exact mimicking of acetylcholine by nicotine at some cbolinergic synapses .is .the .remarkable similarity .~etweem the structures of ihe two molecules. 2 Adrenal ~Sed u!la Cos pregang]~oni: nerve goe.- tc the adrenal medulla an~ iniTiaTes the release cf ca~echclanimes in~c the bloodstream. Smc~'ing a sinf]~ cigarette ;ro~uces hiFher ~eveie c.f ncra'S.-enaliz than r~sti~£ su~ir~e, BAT Co LTD - MINNESOTA TOBACCO LITIGATION 0 0 0

but leas than quiet standing and m~ld exercise and certainly less than moderate and vigorous exercise. Adrenalin levels after one cigarette are about the name aa •fret moderate exerclae and greater than resting supine, quiet standing and mild exercise, but about • quarter of the levels seen after heavy exercise (Cryer, 1980). These small increases in pl•sma catecholaminee ~ust exceed the threshold for cardiovascular and some metabolic effects, much as increased heart rate, systolic presser effects and llpolyaia but not for hyperglycemia, ketogeneaia, glycolywia and suppression of insulin secretion. It should be noted that "not only are noradrenalin and adrenalin rapidly cleared but they also accelerate their own metabolic cle•rance" (Cryer, 1980, p438) so that only very heavy smoking could produce accumulations of adrenalin and measurable metabolic effects. Thus, in order to demonstrate th•t cigarette smoking could produce a~ increase in plasma free fatty acids and so might contribute to coronary heart disease, atherogenesis, and atheroscleroais, Kershbaum and Belier (1968) had to give three cigarettes in 20 mim. The lack of re•liam of this dose can be seen from the fact that a 40 a day smoker smokes a cigarette every 20 miss. For average smokers the effect of nicotine on the adrenal medulla is no more than taking moderate exercise. The fatigue alleviation that vould result from the metabolic changes will he discussed in the next sub-section. 3. Adrenal Cortex Unlike the adrenal medulla, T~lease from the adrenal cortex into the circulation is controlled from the brain, vis the anterior pituitary. The glucocorticoids like the catecholanines have marked metabolic effects but small or no haemodynamic action. Ae yell as conversion of amino acids to glucose, they also speed up extensive mobilization of depot lipid reserves, inhibit lipid synthesis and reduce glucose catabolism. Together these effects elevate plasma concentrations of glucose. Thus,the too eats of hormones mobilize energy sources and prepare the person for action and reduce fatigue. Adrenocortical secretion does increase after smoking although it does not seem is occur after one cigarette: Habitual smokers had to chain smoke four unspecified cigarAttes in thirty minutes to elevate the plasma corticosteroid levels (Kershbaum, Pappsjohn, Ballet, Hirabayashi and Shafiiha, 1968. In s second mtudy (Winternitz and Quillen, 1977), habitual smokers were given eight 2.5 mg nicotine cig•rette to smoke in two hours (equivalent to over 100 cigarettes per day). Their graph shows that oortisol levels are the same or lower after the second cigarette amd it is after the third cigarette in half an hour that ccrtisol had increased. The best conclusion that can be drawn from these studies is that high nicotine doses from extreme smoking rates can else-ate plasma glucocorticoids. An alternative interpretation is that the rapid smoking was aversive and stressed the subjects. If we accept the moat charitable ~nterpretation that smoking Froducem some release of glucocorticoids then combine~ wit~_ catecholami~e release, they will make evaila~]~ emergv sources. These would be available for use by the ~rsin and the res~ of the body (Astrand and Rodahl, 1970) and, in %his way, smoZi~g would alleviate =enta~ and muscle fatigue. !7 C_-> c=.~ BAT Co LTD - MINNESOTA TOBACCO LITIGATION

Central ~ervous System Studies with iontophoretically applied acetylcholine have revealed that acetylcholine excites brain neurones.(Phillis, 1970) including the medullary and mesencephallc reticular formationt lateral and medial geniculste, caudate, ventr©basal complex of the thalamus, hippocampua, cerebellum, infex-ior colllculus and the Betz cells of the cerebral cortex. Cortical cells and caudate cells clearly have muscarinic receptors which were relatively insensitive to nicotine while acetylcboline receptors in the geniculate, ventrobasal thalamus, hippocampus and reticular formation nuclei were sensitive to both nicotinic and muscarinlc compounds. in spite of the clear evidence that the cholinergic neuroses at the cortex are ~redomina~tly muscarlnic, "smoking" doses of nicotine (eg 20 ug/kg in the cat) produce excitation of cortical cells (eg Kavamura and ~omino, 1969; Armltage, ~all and Sellers, 1969) and release of acetylchcllne at the cortex (Armitage et al, 1969)- Cortical scetylcholine release and cortical excitation can be produced by stimulation of the mesencephalic reticular formation and this phemomenou cam be reduced in one hemisphere by unilateral destruction of this region ipsilaterally (Celeste and Jasper, 1966). In an analysis o~ the effects of "smoking" doses of nicotine after destruction of the midbrain (Ka~amurs and Domimo, 1969), nicotine ~roduced cortical desynchronization and hi~pocampal synchronization im ca~s with s caudal midbraln transection st the junction of the pans in exactly the same way as intact animals ~iven nicotine. After bilateral lesions in the tegmental region of the nidbrain, n/coti~e in doses up ~c five times %he "smokin£'dose did not activate the cortex. Clearly, nicotine's action on the cortex depends on an intact tegmental region. The ventral tegmental region of the mesencephalic reticular formation is the origin of s cholinergic pathway which projects to the cortex (Shute and Lewis, 1967)- It terminates on the Betz cells of the sensory cortex and produces electrocortical arousal (see revie~ by ~-Yarburton, 1981). The most parsimonious conclusion is that "smoking" doses of nicotine ascend in the carotid artery and excite nicotinic receptors on the tegmental-neocortical cholinergic pathway in the midbrain. Nicotine does not act directly oD the cortex but the outcome of activation of the pathway at the midbrain is release of sce~Icholine at the cortex and cortical desynchronization. C. Nicotine and Euman Psychophysiology In this section we will consider the action of nicotine on the human cortex in the context of psychophysiology, the correlatio~ of neurophysiological evemts with human behaviour. (See extensive revie~ ~u Edwards a~d Yar~urton, 1983). Cortical Desyuchroniza tics Kany human s~udies hs~ shown that smoking increases tb~ amount o~ cortical da~v~chronimatiom in th.~ f~rr o." a~ uF~-ard shift i~ dominant a!uha freouemcy ~g ~.auser, Sch~-artz, Eoth end ~ick~ord, 195~}, lees tcta) alpha activity, and more bets activity (eg Yr.ott, 1979; Kurphree, ~979). Thus these hu-~ studies shc~ them smokUnf ~roducss cortzcal desy~chroni~etior jus'. as nicotine dces ir a~-ira.~ si-udie~. In • study BAT Co LTD - MINNESOTA TOBACCO LITIGATION O

correlating performance with electrocortical activity, Werburton and %'eases (1979) found that both cigarettes and nicotine tablets increased the dominant alpha frequency and beta activity, the alert pattern similar to that found when a person is concentrating. In the same study, concentration performance was improved by these treatments. It is significant that Hurphree (1979) was unable to distinguish the cortical arousal changes produced in smokers and non- smokers by 1.86 ag of nicotine intravenously" i.e. there was no evidence of tolerance. This lack of tolerance explalnJ why smokers do not increase their smoking over the yearn. These results make sense because of the molecular similarity of nicotine a-d acetylcholine because the bralm cannot become tolerant to its own transmitters. In summary, the shifts in cortical activity are within normal limits and are indistimguishable from thai seem when s person is concentrating hard. There ie certainly no evidence of the EEC abnormalities. Smokers claim that they smoke to help them think and concentrate and nicotine's action on cortical desynchronization produces the neural state for satisfying this need. Evidence that smokers can control their nicotine intake to produce a specific brain state has come from ccmtingent negative variation studies. 2. Contingent Negative Variation The contingent negative variation (CI~V) appears as a slob negative shift of cortical potential during the period l~et~een a'warming signal and an imperative stimulus which requires a motor response or a decision i.e. planned action. Significant correlations have beam reported between C~ amplitude and measures said to represent mamy psychological processes, but selective attention or concentratiom is the most commonly cited concept. Experiments on the CNV and smoking (Ashtom, ~arsh, M illmam, Rawlims, Telford and Thompeon,1978) have correlated ~ with subjective reports of feeling either "relaxed" or "stimulated following smoking. Ashto- collected personality information and estimated the amoumt of nicotine taken into each subject's mouth from the cigarette butts. After smoking C~ amplitude increased in seven subjects, decreased in elevem and four showed a hi-directional response (Ashton e~ al, 1974). Puffing am umlit cigarette by three other subjects resulted Im no CNV amplitude changes. Repetition of the 8moki~ sessionn for I I of the 22 subjects produced the same directional changes in the CN~ i~ these individuals. Ashto~ assumed that there is a positive relationship between C/';V magnitude and the individual's level of activity in the neural systems. Accordingly, they related individual differences in CNV personality and smoking behaviour by dividing their group at the extreversion mean. They found that the eight more extraverted subjects took a smaller oral dose cf nicotlne-and showed a mean increase in C~V magnitude after smoking, while the eight more i~trover~ed su~ject~ showed the reverse. There was some evidence from subjects thai self-report of e "sedative" effect of smoking was aesociate~ with a fall in CNV whereas "stimulation" was associated wi%b a rise in CEV. BAT Co LTD - MINNESOTA TOBACCO LITIGATION O O O O

Ashton and her co-vorkers (1978) also examined the effects of a dose of nicetAne similar ¢o that obtained by an inhaling smoker from a 1 - 2 m6 nicotine cigarette. The marne direction of change in CNV occurred in individuals who took part in ~oth the smoking and nicotine sessions. Than Alhton et ILl (1978) examined whether the CNV changes are dependent on the dose of nicotine or, alternatively, on both the characteristics of the smoker and dose. Over a range of doses there was an inverted U relation between C~Y magnitude and dose for all seven subjects, althouKb the precise dose-response relation was not the same for each of them. In summary, nicotine increased or decreased the CNV amplitude depending on the dose and subjects seemed to control their smoking behaviour in order ¢o obtain a particular dose. These findings were combined by Ashton to give the hypothesis that individuals adjust ~heir smoking behaviour ~o the amount of nicotine for a particular brain state and so control their psychological state. Extending this concept the desired state will depend on the outcome of interaction between the individual and the situation and, by adjusting the nicotine dose, the ease individual may use a cigarette to provide s "stimulLut" effect on one occasion and a "sedative" effect on another. Event-Related Potentials (ERP) Event-Related Potentials, sometimes known as ~versged Evoked Potentials, are the complex electrical changes recorded at the same time as • physical or mental event add represent the activity o~ groups of neuroses in the brain. The obvious advantage of the ERP technique is that it provides s continuous record of events occurring in the brain during psychological processes. With repetitive stimulation, the waverers ~etween 0 to 250 msec post-stimulus consists of components which are essential!y constant for s given stimulus. These exogenous components occur irrespective of the person's stats. Of particular interest are the later so-called endogenous componeutE which are elicited or emitted in the absence of stimulation, and whose characteristics are partially independent of the stimulus characteristics. The major endogenous component which has been identified is the PD'or P3OO whose latency ranges from 275 - 600 meet. The P~00 is particularly sensitive to the subject's prior ezperlence, intentions and decisions and varies according to the task requirements and instructions. In an analysis of the effects of smoking on the EP, Ps to correct tareet detections in a rapid informatiom processing task (Edwards, ~'esnes, ~'arburton and Gale, in preparation), subjects were instructed io detect and respond to triplets of three odd or even digits in • sequence of single numbers that were presented on a TV screen e~ the rate of 100 per minute. Good performance depended on subjects maintaining their concentration throu6hout the 20 minute session. Smoking produced l~rge decreases in the latency of P300 (20-30 msec decrease) and smal~ decreases im the amplitude of P~O0. A re~uctiom in P~OO latency ie interpreted as s decrease in the.time "~aken to evaluate or categor~ee a stimulus while a decrease in P300 an~litud~_ is thought to mee.u that the subject is using les~ neural "resources' in ~roceesiL£. Smoking else increased the ~argets detected azd 20 BAT Co LTD - MINNESOTA TOBACCO LITIGATION

decreased reaction time (nee sub-section 1~ D). This iuprovement of performance fits neatly with the ]r~OO changes which are indicative of more efficient neural processing of the input. ps~cho~harmacology, of Nicotine In this section evidence will be presented on the implications for human behaviour of the neural states that are induced by nicotine. As we have seen, nicotine stimulates and produces increased neural changes that have been interpreted as L sedation effect. ~uestionnaire surveys (eg McXennell, 1970; Thomas, 1973; Russell, Pet, and Patel, 1974) have shown that desires for stimulation and for sedation are major smoking motives. Thus, this section will concentrate on the evidence for more efficient performance after nicotine and the effects of nicotine on mood states. I. Performance This section discusses (1) sensation (the input of information); (2) attention (selection of information); (3) processing of information; (4) learning and memory (storage of information); a.~d (5) motor Output (the response). An extensive review of the effects of smoking and nicotine on performance has been published (Wesnes and Warburton, 1983a) and so this section will be very selective. a. Sensation: There is l~ttle evidence from subjective reports of smokers ~het smoking has amy effect On sensation, except perhaps for a possible blunting of taste and smell. Some research has suggested that smoking raises sensory threshold~ slightly i.e. the smokers are less sensitive to stimuli (Late.n, Haag and Silvette, 1961) but there are also reports that smoking doses of carbon monoxide increase "~hresholde (Ralperin, Ec~arland, Niven and Roughton, 1959) which would explain these changes. Even for taste and smell the evidence for changes after smoking are not clear cut. Pangborn and Trabue (1973) reviewed 16 experimental studies between 1937 and 1970 of which 11 reported impaired taste and, or, smell with heavy smoking and five which found no difference. The effects of nicotine alone on sensory thresholds have not been studied. A second test, which has been used to assess the effects of smoking on mensation, is the critica~_ flicker fusion test. ~riefly, this task requires the subject to discriminate between.intermittent photic stimulation which either appears to flicker or appears to be s steady light. The dependent variable is the frequency at which the change in appearance occurs so that a rise in threshold could indicate better sensory processing. Smokin~ increased the threshold slightly (Let's.n, Finnegan and X~aaE, 1950) and a later study using an oral dose of 0.1 mg of nicoline (Warwick and Eyaemck, 1963) found an elevation of the fusion threshold by about seven per cent. It is believed that the mechanism for critical flicker fusion is at the cortex and as Yarwick and Eyaenck claim that nicotine is improving cortical reactivity rather than sensory processing. Im the m~xt section we will be considering the process of attention which, is a/so related to cortical arousal. b. Attention: Smoking motive questionnaires (e.g. Russell, e~ el, I~74) imdicste clearly that smokers believe tha~ emc~ng hel~s then tc thin~- end to cenceztrste. The firzt experimeztal evidence t~ su££eE~ 2~ m ~D CD CD Cr~ BAT Co LTD - MINNESOTA TOBACCO LITIGATION

that this experience of mnhanced.eencentretlon van due to nicotine cane fro• a complex vigilance etudy vith • main guiding task and s subsidiary visual monitoring (Terriers smd Rartemann 1964). They counted correct detections of perApherL1 missile, false detections and gvAdance errors and measured plasma nicotine, heart anate and /ZG. be test group• vere non-seekers, smokers deprived for 20 hour• and enoke~ alloyed to smoke unspecified c~garetto• oMen•oar ~bey y£shed. The7 do not give l~e plasma n4cotine Aevel• or ~G data but an elevation of heart rate, after smoking, indicate• none nicotine abBot-pales. T~gilaace perfornance (per cent onlasioas) de~llned in non-smoker• and deprived smokers during the ~oat bat nicotine naAnl~tined detact~en pez-foz~na~c• ~hroughouT the see•ion. IRe other behav:Loural data yore given on detection errors oz guidance errors, presumably becauJe they d~d not ~hange. Fz~u~enJ~aeuser ~d her oollesgt~a (]~mJxMemJ~teuser, l~yz~en, Post a~d ~ohsn••on, 1975) also toyed that smoking during an 80 mAn. reaction time ~a•k enabled gabfest• ~.,o malntaln their perfo~ance. All sub~ecta yore depraved o£ cigarette• for 10 hour• prior to te•~ing. BaAf of them remained deprived ~hx.Dugho~t ~he ~eat vh~le ~he o~her half ve~e a11ove~ a 1~ ~ nAcot4~e e~ja~et~e e~t•r 20 m~, 4~ mi=s and ~0 mane of testing. Heart rate increased vh£ch suggested ~he~ nicotine bed been absorbed. The reaction tames for the deprive~ smokers Increas•~ during the n•maio= vh£1s •noklng maintained ~be reaction ~ino. The amok•rn else reported that they felt more ~ler~ and more relaxe~ omen •noM~J~. ~sentially, the sane f~ndings vere obtained vhen the •~udy ~es rejected uaA~g either • similar prolonged ~•t or this te•~ vi~h ~e added stress of iv= ether tasks ~o be perfoz~ed at ~be eame ~i=e - a MASh a~ousa~ test ()ty'A-81:em, Poa~, ~enbaeumer and JoManssc:, 15"7~b; • yrs~en. And•resort, ~rankenJ~aeueer and Eardh, 1972a). The smokers vere sued-divided into smokers who smoked nora often An leo arousal ait~atiens and those who smoked more often 1= hig~ arousal situations. Re,~ot-kon ~imes vere betluer for lay 8,r~un~ smoker1 vhc smoked dul-Ang l~he lov a~-ouna~ test than vhem t3aey vere deprived but smoking van not beneficial dur~=g the high •roueal tes~. ~he opposite flnd£ng ~an obtained for the high arousal smoker•. These ~o studies vere the fix'•~ tO reveal that performance van the outcome of a smoking z i~l~ua,1 • •i~ut~io= ~.ute15acIMLon. Veeries and Ya~bu~'~on (1978) l~nste,5 smoker8 dcpz'Avod ~or 12 hours a.~ non-smokers on a visual ~A~Ala~e l~tsk vhAeh ~naisted of de~ecti=g brief pauses A: the minute hand o~ a clock. Detection per~c.-msnce dec].i:ed over ~he 80 ni=ute test in L3~ 8ub~ects vhe did no* •:~Me bu~ deprived snckere vho masked a 1.,4 mg cigarette et 20 :ins, 40 :!=-s a=~ 60 miss (as i~ ~he Yranke~baeueer studies) shoved no decre=e:~ in perfcrnL~ce. Ciga~e~tee had heaped smoker• to maintain better detection perforzanoe than either deprived smokers or no~-amoke~s. The too n=n-•mo~=~g groups did not differ mIL~;i~cant~F e¢ there ~a~ no suy~or~ for t~-e hypothesis tha~ cigarette dep.-~vatAcr, ~'as e~ ~or~J~=t factor. E~-Adence ~-~•~ +~me pf:~ect van due ~o r,~cc~ine ca~e fr~ e.~c~'-'.~r z"-..'.c~ (¥esmes, b'arbur~o~ and gang, 19S~) vitM this taa~ uei~g C, ~ a~ 2 ~g ( • , . '. l, .t =. It ". . BAT Co LTD - MINNESOTA TOBACCO LITIGATION

nicotine on alk~lioe ~ablets held i~ ~he mouth at the sa=e 20 minute intervals as before. During the first 20 minutes, performance in the three conditions was matched to give a base,ins and a comparison of the three conditions showe~ a clear dose related improvement after the first and second tablets at 20 miss and 40 mins. However, after the third tablet st 60 mine, performance was worse in the 2 mg group than the Img group, probably due to the adverse effects of s cumulative dose of 6 mg of nicotine in an hour. In a third study, cigarettes yielding O.31, 0.71 and 1.65 mg of nicotine were compared using a version of the vigilance ask, with well defined detection intervals aud the same experimental design as before. (¥eanes and ¥arburton, 1978). The ciEarette butts were analy~ed %0 estimate the amount of n/cotine which was generated by the smoker. The bes~ group performance during the first 20 minutes after a cigarette was obtained after smoking the 0.71 mg nicotine cigarette rather than after smoking the O.D~ mE nicotine cigarette. ~owever, the butt analyses showad that smokers obtained estimated mouth levels of 0.63 mg of nicotine from the 0.31 mg nicotine cigarette, I.D mg nicotine from the 0.71 mg nicotine cigarette and 1.5 mg nicotine from the 1.65 mg nicotine cigarette giving evidence of titration. Analyses of individual differences gave clear evidence that subjects who scored high on neuroticism smoked harder st the beginning of the session than io~" neuroticism subjects while there was some evidence that subjects who scored high on extrsversio~ smoked more intensely st the end of the session than the more imtroverted subjects. These studies suggest that subjects had a need for nicotine to help them perform ~'hich determined the smoking behaviour no that performance ~as the outcome of the interaction of the situation, the individual and smoking. AB we~l as examining the effects of nicotine and smoking in a long boring task we have also examined sustained attentiom i~ a shorter 2C mim rapid information processing test, that was described in Section IV C3. In a set of eight atudi-es (see Veeries and Yarburtcn, 197~: 198~; 1984a ~ud h) ~e have tested cigarettes with nicotine deliveries ranging from O mg (herbal cigarette) to 2 mg as well as nicotine %a~le~a. Each subject was tested for 10 minutes before a further 20 miautes after a cigarette or tablet. Performance was compared with both the pre-treatment baseline and the no smoking, herbal cigarette or placebo tsble~ conditions. The overall pattern of findings showed that per$ormance declioe~ over the session in exactly the same way for herbal cigarettes and no smoking , showing that the smoking act itself had no influence on performance. Smoking not only improved performance relative to not smoking but produced an absolute enhancement of 11 - 15 per cent relative to the pre-n/cotine baseline in the firs~ I0 minutes in 68 - 76 per cent of the subjects for each cigarette in each study. In general, greater imprcvement was seen afte- high delivery cigarettes than lob" delivery products but there was overlap between cigarettes of similar delivery probably due to nicotine titration~by the subjects. C~mclusive evid~ce that the better detection performance was Eue to nicotine came from a study ~ith non smoker~ using alkaline tablets containing n~ccTine (Wesnes and Yarburton, 1984a). Doses of 0.~, ~.C and 1.5 mg of nicotine significantly reduced the decrememt ~het BAT Co LTD - MINNESOTA TOBACCO LITIGATION Gr,,

occurred in the placebo condition and the 1.5 mg dose produced a significant absolute improvement relative to the pre-treatment baseline. In this task the stimuli were well above threshold, the memory component was small and the motor response was a simple bu%ton press so it can be concluded that nicotine was enhancing either the attent~onal process, information processing, or both processes. Support for enhancement of the attentional pro~essing came from a study of nicotine tablets and the Stroop effect (%'esnes, and Warburton, 19~a). In the Stroop test, subjects marne the colours of either a set of colour spots or a as% of colour wor~s printed im smother colour. There is a difference between the time taken to read the two sets, because of the distractio~ produced ~y the incor.gruity between print colour and the word. Nicotine doses of I mg and 2 mg reduced the magnitude of the Stroop effect in both smokers and mot- smokers and there was no difference in the size of the effect iu smokers and nOD-smokers. This evidence argues strongly for the hypothesis that nicotine helps a subject to i6nore distracting informatioz and concentrate on relevant information. In summary, these data support the subjective experience of smokers ~hat smoking helps them to concentrate. Nicotine increases etectrocortice~ arousal and so this work fits neatly w~th research demonstrating that compound.- which modify electro.cortical arousal change attentional performance (See review Warburton, 1981). f c. Information Processing: ~e studies of smoking and micotine tablets and rapid information processing (Wesnes and ~arburton, 19B3b; 1984a and b) provide ~equivoca~ evidence for more efficient processing in terms of improved detection of triplets of three odd or three even digits in a string of single numbers flashed oz • com]~uter screen. Nowever, nicotine from cigarettes and tablets also reduced reaction time. This improvement yam around seven ]~er cent and between 65 and 77 ])er cent of the subjects were improved depending on the ciFarette'e nicotine delivery. This finding is very important because it show~ that there is no speed and accuracy trade-off and so there ~'as an overall imyrovement in processing efficiency. This behavioura~ data fi~a with the study of Edwards et al (in press) which showed that smoking decreased the latency of the P300 wave in the rapid imformatiom processing task (Sub-sac%ion YV C3). Thus both be~avio~ral and electrocortical measures indicate that nicotine is enabling more efficient prooesslx~g of i~formatiom i~ the braim. d. Lesraing and Memory: The effects of nicotine on ~earmimg have been investigated extensively by Karts Andereson. The first stu~y, (Anderssom and Post, 1974) compared the effect of two nicotime-free cigarettes and two 2.~ mg nic~tlme cigarettes on light smokers (~ees than five per day) who ~ere 6eprived for ~2 hours prior %0 the testin£, when learning s nonsense syllable list. Significant increases in hear~ ,rate in~iceted thai nicotine was a~scrbe~ from im.~ nicotime cigarettes. The first c~garette wee givez after the first ten trials ~f~'earning the list add a second cigarette, ~f the ~az~ kin~ wa£ Given after 20 trials gives every, two m~_ouies. The ]e'-rninf curves were i~emTical for the two conditions ~rior tc ezohizE, tu~ after nicotine the num~er correct 6ecreased an~ remained "t,~]ow ~h~ 2L BAT Co LTD - MINNESOTA TOBACCO LITIGATION Co

scores in the nicotine-free condition suggesting impaired short term memo~7 although the learnin~ curves were parallel. After the second nicotine cigarette, the number of correct syllables increased significantly to the same level of acquisition performamce as in the nicotine-free cigarette condition. Thus, relative to the previous perform~ce, nicotlne had improved recall of the syllables. The difficulty of interpretinC the effects of n~cotine im this study is that learning and fecal2 are occurring over a 20 minute period while plasma and brain levels of nicotine were falling to half of their pea~ levels. These date give no evidence of nicotine impairing acquisition because the learning curves are parallel after the nicotine cigarette. However, it appeared that after the first nicotine cigarette, the information stored in the non-n/cotine state was less available in the nicotine state, a phenomenon ~-novn as state- dependent learning, perhaps due %o the different selection of information in the two states. P Another study by Amdsrssmn (Amderssoz and ~ockey, 1977) used a memory task of relevant and irrelevant information which enables s test of this selectional interpretation. A list of eight adjectives were presented, one at a time, in one of the four corners of s screen and the subjects were asked to remember the words in preaemta%ion order and by implication location was irrelevsmt. Two groups of deprived smokers received either a 2.3 mg nicotine cigarette or remained deprived. There were no differences between the nicotine end no nicotine groups for percentage of words that were recalled in the correct order, or for the percentage of words that were recalled correctly regardless of yard order which coniradic%~ Andersson's previous evidence on recall. The most interesting finding was that position on the screen was recalled significantly less well with nicotine. In a second test in which subjects were asked to remember words, word order and screen poaltion, the groups did not differ sigzLificamtly in their recall although %bare was a trend for locatic= %0 be recalled better after nicotine tha~ when de,rived. These data on 1-ecall also contradict the previous findings. This study suggests that nicotine can enable more selective processing of information bul on/y when the information is thought to be irrelevant by the subjects i-e. nicotine is not acting via a passive processing system. The possibility of nicotine produoimg state dependent learning in human subjects was mentioned earlier and numerous anise] stnudiee have given evidence of state dependency with cholinergic agents (see review in %'arburton, ~977)- I= a state dependent design, one ~roup of subjects leax-ns after s dose of compound while a second group learru~ after a placebo or nothing. For the recall lest beth groups are divided, one half of each group are tented with the a~ent of learning and half are switched %0 the other condition. If the recall score~ are better for those groups which learned i~ the same chemical state, then state depem~en% leaz-ning is sai~ to have occurred. This state dependency hypothesis with nicotine ha~ been investigated in several studies. ." I~ the first publishes study (Boustom, Schneider amd Jarvik, Io7E) 2~ heEv=r smokers, deprived of ci~are%tes for three hour~, read a list of O O O O Cr~ BAT Co LTD - MINNESOTA TOBACCO LITIC;ATION

words and were matched on a free recall test prior to smoking. One group smoked a ~-5 mg nicotine cigarette and the other group amoke~ a non-nicotine cigarette and ~hen the subjects were given three lists with free recoil tests after each one. The immediate recall scores showed that the m/cotlne group had significantly poorer recall than the placebo group, matching the findings of Anderaaon and Post (1974). Yorty-eight hours later, only eight of the subjects from each conditiom returned for re%eating and these were s~5-divided into four groups of four subjects. The recall scores of the subjects, who smoked a nicotine cigarette at the time of learning, were significantly worse than %home who smoked a non-nicotime cigarette which was not consisteD% with the findings of Andersson add Post (1974). ]~o evidemce of state dependent learning was obtaimed which is Do% surprising with such small groups. In contras~ %0 the last study, three studies have found evidence for state dependent learning with nicotine (Peters and ~cGee, 1982; Warburton, Wesmes Lud ~hergold, 1982). The first experiment used the state dependent design to test nicotine's effect on recall and recognition memory. After smoking s 1.4 mg cigarette, the subjects were shown a list of norms and immediately aske~ to %'rite down as ma~y as they could. There was no evidence of Lny difference on immediate recall in agreement with Andersson and Hockey (1977) and Scum%on et al (1978). Bowever the recognition teat on the next day showed a clear state dependent effect but no difference between the same state groups. In smother recognition study (~arburton et al, 1982), smokers who were deprived of cigarettes for 10 hours were give~ s 1.4 mE D/cotine cigarette or nothing immediate3y before seriai presentation of a set of Chinese radicals. They were divided into four groups - no smoking prior to learning m~d recall; no smol.-ing ~rior to learning but a cigarette prior to recal2; a cigarette prior to ]earning and recall; and a cigarette prior to learning but some prior to recm~]. Subjects who recelved nicotine prior to learning had significamtly better recogniTiom scores than the subjects who did not smoke in the firs~ part of the experiment. There was no effect of nicotine on recall performance. A significant imterac%ion term indicated that changing the chemical state ~uterfered with recognition. A more elaborate study examined the effect of s dose of nicotine on both short-term and long-term recall, using a state dependent design. Subjects, who were smokers deprived of cigarettes for ten hours, herd a 1-5 mg alkalime nicotine tablet, or a placebo teb]et i~ their mouths for five minutes to allow nicotine absorption, before swallowing. After the tablet the subjects listened to a list of words and the~ did successive subtractions for one minute to prevemt rehearsal. A free recall test was given and then the subjects were told that they would be called im one hour for a second unspecified test. ]'-'ring the hour they worked and were then recalled, given either nicotine or placebo tablets, depemding on their group ~nd asked to recall as mamy of the %'o~a a the.~ could im another 10 minutes free recall %est. The shor~ tern recall data revealed e very significant superiority of the micotime ~rou~.~ver the placebo group. Long tern recall was also significantly halter when subjectE had taken nicotine prior ~o Icarming but not when taken prior to recall. The very significamt "izieraction term again gave evidence ~or s state dependemt effect of 26 BAT Co LTD - MINNESOTA TOBACCO LITIGATION m O O O O O'-, CZD

P nicotine and sboved that nicotine was facilitating the input o~ information to storage but had no direct effect on storage but had no direct effect on storage or retrieval. These last findings are consistent with ~revious results which showed that ~cotine facilitated i~formation processing. However, the state dependency effect in these studies and those Of Anderseon and Post (1974) and Peters and McGee(19S2) resulted'in a poorer recall when subjects either learned with nicotine but recalled without nicotine or vice versa. A state dependent effect is not surprising because of the differences in the processing of information by nicotine's action on electrocortical arousal. e. Motor Output: There is no evidence of gross changes i~ motor output after smoking doses of nicotine or micotine tablets and the only change is a reliable increase i~ hazed %remor after emo~_i~ (Larsom et el, 1950). This fine tremor has no practical significance. 2. Eood States The illuEtrative studies i~ this section have provided ms evidence of nicotine absorption from cigarettes, and so it is not usually possible to separate the pharmacological effects of nicotine from those of the smoking ritual. a. A~xiety: Evidence for a ~ra~quillizi~ actio~ of smoF~ng comes from questionnaire studies, stress studies, personality tests o~ smoker~ a.ud non-smokers, smoking abstinence studies a~d situations affecting smoking behaviour. ~n a factor a~alysis of smoking motives, McF, e~4~ell (1970) emtracted a factor, "nervous irritation" smoking, which included "smokes when anxious or worried" and "smokes when nervous" and also ~amoke~ when a~gry'. ~'hem cigarette consl-m~tion was equated, female smokers ha~. higher scores for "nervous irritatioz" smoking. A factor am=_~v~is of smoking situations (Frith, 1971) found a similar factor which he called high arousal smoking, s atromg desire for a cigarette in stressful situations, and women were more likely tha~ men to smoke i= these situations. ~sing a Smoking Motives questionaire, ~ussell et al (19~4) found that ~3 per cent of smokers who attem~ed a smoking clinic, 7~ per cen~ of non-climic smokers said that they smoked when worried while and ~'arburton Lud ~esnes (I~) found tha~ 88 per cent of students answered im the came way. A~together, this evidence shows that smokers believe that smoking is beneficial where they are amxious. ~eebitt (I~73) tested heavy smokers (2C or more cigarette~ s day) amd men-smokers, uzing ~illingueas to accep~ electric shocks as a msa~urm of amxiety. 2oth smokers and nor-smoker~ were aske~ to ~uff an~ impale fro~ either s 1.6 rag, or 0.3 mg nicotine cigarette, or ~uff om a~ u~lighted c~gar~te. Smokers tc!sra~e~ much stronger shocks after smoking and this e#fect ~'as ~ose ~e~,endent. 0~, the other hand, the ncz-smo~erz +.c:*~erate~ zcr~ a~och ~m ~he no smoking com~ition ar, d £!i£ht~y less after th~ high nicotine cigarette. ~If i~ is ecc~le~ the'. e~o=l: tolers.oce is • measure c~ at_misty ~s_~d i~ 2"~ BAT Co LTD - MINNESOTA TOBACCO LITIGATION O (=z:) m

P is a big IF) amd thai non-smokers can tolerate smoking (another biG IF), the eimple expl•n•tiom th•t smok±n£ reduces anxiety does not fit this data because the •bsence of nicotine in smokers could have resulted in more anxiety than non-amokera. Smokers who puffed an unlighted cigarette tolerated leas shock than men-smokers im the same condition which suggests that nicotine from the smoke is more important than the smoking ritual for an effect on anxiety of this sort. The•e findings could mea~ either that • nic6tine deficit in the smokers resulted in abstinence symptoms due to physiological dependence on nicotine or that smokers are constitutionally more anxious. The latter possibility can be examined from studies of ~be neuroticism level of smokers, which correlates highly with anxiety. Eysentk's studies of 6,000 men gave no evidence for a significant correlation (Eysenck, 1963) and Rae's (1975) survey of 253 female students revealed me differences in" the degree of neuroticism between non- smokers, ex-smokera, light smokers or medium smokers (less than 15 per dey). However, st least seven studies have suggested • positive relationship of smoking with neuroticis= and this association is particularly strong for women smokers (Guilford, 1966; Mesres, el el, 1971; Waters, 1971; Dunnell and Cartwrigh%, 1972; Shiffman, 1979b; ~'arburton, Wesmes and Revell, 1983). Although this evidence suggests that smokers are constitutionally more anxious, it might still be argued that repeated exposure to nicotine had caused anxiety and a higher meuroticiam score. Evidence against this argument comes from a crucial study dome by Cherry and Kiernan (1976; 1978) which have followed a cohort of 2,853 young people for 25 years. ~'hen they were 16 years old they completed the Eauds]ey Personality Imventory before most of them had begum to smoke. At 20 years and 25 years of age they completed • smoking habits questionnaire eu~d It was found that the cigarette smokers, as a group, scored more highly on extraversiom and neurot~cism. Statistical amalyaes revealed that the two personality dimensions were independent and additive in their effect on the likelihood of becoming a habitual smoker and suggests that constitutional factors underlying anxiety results in some people smoking and not the other way round. Given smokers can titrate for nicotine (Section IIID), it would be predicted that the situational demands would modify titration if nicotine reduces anxiety. Schachter (1978) gave smokers electric shocks ostensibly to determine their tactile sensitivity. In the 3c~" stress condition, they on~y received • tingling shock whi~e the high stree~ group were occesional]y given s painful shock. Schschter ccunted ~he number of cigarettes smoked between tests in the laboratory and found more cigarettes were smoked im the interval_ after the high stress condition &'hem they were seriously anticipating the next test series. Unfortunately, he did not collect amy information abou~. smoke gemers~icn and smoke manipulation. In a recent study (¥arbur%on et al, 19S3), undergradtmtes filled in a detailed ~iary o= their smo~-ing habits during a first year examination ~eriod and e~aim 6 weeks later, after the examination results had come out. Prom th~ cigarette totalE for the aftermoon-, sJ~d eveniLg~ i~ ~a~- 2~ BAT Co LTD - MINNESOTA TOBACCO LITIGATION C~ 0 0 0 C~

p obvious that subjects smoked significantly more ci~aretteB at all times during the examination period than six weeks later. Subjects also 6ave a aub~ective evaluation of their degree of inhalation and there were highly significant increases in degree of inhalation during the examinations with respect to the post-examination period. Thus the students smoked more when faced with examinations and the conoomitaat stress. The situation x individual x smoking interaction was examined by studying the amount of nicotine taken by subjects in the vigilance test mentioned earlier (~arburton and ~'esnes, 1978). A positive correlation was found between neuroticism and the difference A~ smoke generation between the first and last cigarette ~or both the low nicotine cigarette and the high nicotine cigarette, indicating that the more neurotic subjects smoke Bore iJ~tenaely st ~e begix~uing of the session th~ at the add of the test. This result provides further evidence that one aspect of smoking behaviour ia the outcome of the interaction of the situatlon and the individual and increases in stressful situations; smokers use more cigarettes, smoke them more intensely and inhale the smoke more deeply with those who are more anxious and susceptible %o stress sho~-ing the most marked effects. An interesting question is whether smoking does be]T to cope with stres~ and improve behavioura~ efficiency? ~e- have analysed performance increments as a functio~ of neuroticiem and ~oun~ that there was a direct association betwee~ the degree of neurcticism and the amount of performance improvement after smokirg (~'erburto- a~d Yesnes, 197e). The latter asaocistio~ is consistent ~'ith ~a~a obtained by Kucek (197~) in a~ experiment where subjects were tested under conditions of information overload. £ comparison of Deurotic smokers, vho were allowed to smoke, and neurotic smokers ~ho ~ere deprived of smoking showed that smok~.ug had a beneficia~ e~ect on the ]~erformance of neurotic subjects allowed to smoke. Nicotine is the most likely constituent of cigarettes ~or ~roducin£ this improvement in ]~erfor~ance, and we found that for both females and males the performance imTrovemen~s after nicotine tablet ~ere directly correlated with neuroticism scores (¥arburton and Yesnes, ~97~). This significant association with neuroticism gives some evidence for the hypothesis that nicotine is the tranquillizing ~ugredisnt of cigarette smoke. inferences about th~ properties of a pharmacological agen~ can often be ma~e from the symptoms that occur when medication is terminated. Studies of withdraw'el symptoms that follo~- cessation o~ smok-Ang sh~ that one of the major symptoms is anxiety (Resbitt, 197~; Shiffm~ aL~ Jarvi~, I~76) and it is more likely to occur i~ women among ~-hon there is a greater proportion of neurotic smoker~ (Guilford, 1966; Shif[ms.u, ~979 b). ~ ~naly~is of The ~i~uati~ns "~hicb resulte~ im • retur~ tc s~ok_i~g by Earlatt~(1~79) and Shiffman (I~7o a) in~icate~ that 8= ;e: cent of these ~ituations fell into three categories: co~ir,£ ~ith anxiety an~ other negative emotional states (43%), social Treasure (25%) and co~ing ~'ith social stress (12%). Fe~" of Marlst~'s a=~ Shi~fmsn's relapse~ smokers reported that ~hysica~ symptoms triggere~ the relapse ~hich argues against the physical dependence h~-pothe~is. 29 BAT Co LTD - MINNESOTA TOBACCO LITIGATION O O O"-

In the ~.Iff=a~ (".q79 s) stud/, t~¢ ~hlrd[ of the Futje=t~ ,'ere unS(- £~ree.s s% the time of relapse aL~ sr:z:ety was Far%iculsrY/ com.=c: ~oLE ¢~-smokers who relapsed a~ work, ~hich £ug£~ats iba~ v-.r):- red,ted anxiety was a conlributcrr factor. Neuro',ics are =cr~ aY.xie%F-~:rcme r~nd so i~ ~S mot eurprlai-.g %hs~ the~ fired ~ diff'icu~ %c. elo;, enokir~£ or relapse if that ab~t[im (Cherry sad Kierr..z:., ICY=_). Theme data or, smokimc ~o cope %'~th s~re~s fit m~ail~ v'itl, rtud~e.~ ehowZm£ Ths~ de]:raved smokers are more ]ike].~'to feel ~=%~¢u[ "ix ~ree[fu3 situations (Prar, ke=haeuser, e'{ an., IC71; ):.rrelen. e~ a], 1972a; Schacht~r, 1_o75). A large e;idemiclcg~ce] stud.v has cci~.'e~ scKe ev~deDce ~,ha=]'..r..~i ~ndic~:~ ~here are adverse comse~.uenre~ of exokimg cess_~iicn ¢.ee, 1979). Bri:i.-h doctors have reduced ibelr cigarette F"--OILIG~" over 2C Tears, and there has bean some of ~be expected .~" ;rev.~r.~r', i~ ~cr~ali:r rate from "smoke-related" diseases, but tb~ over'-!l dez~i rate ~i~ nc~ fall as much aS vo~]d have bee~, ~redicled from c~Ltr ~rou;~ cf similar socioeconomic status. Kortal~TT frcz olh~r cau~.~,. ha~ imcrzased and, im particular, aTreE=~-:,elste~ cea~.h.~ Lad =r, cre_.~!=~ re!a~ive to c~her ;.rcfe[sioms im the e~.u valemt ~ocia! cla.~ ; more doctors had died from accidemts, poison!mr_, suicide, and clrrLc_~_~ ¢f the liver. Lee ccmcludes that "the r~ative worsezir.£ ir zcr~,_~ <.~ from s~resa-re]aled diseases may h~sve heed due parT]/ to £ ;c~r~i'~ adverse effec~ of giving u~ e=ol:i=£ if sue}finE bed 5cted Tc reduce stress" (; I~3~). This comclusio~ must be treated ~i~h cau=~_cz because ii is a~%'aTe very hazardous ~o arEu~ fro= corr~a'~-n ~: csus-lift but, ~everibe]ess, it is consistem~ with the yrev~cu~ dais. It, summary, s large body of evidence imdicates that smeker~ a=.ok~. "_z order %c reduce anxiety am~ smoke E. ore amd ;reeura'~l.r i~_l:e rcrr rLicctine wheL %u~der stress. Certaizzly, ~am_xiety-]~ron~ su~:c¢~ sr~_) ~ more imt.e=se!.r amd experiments sho~ thai mico¢ine ke~r~ =~.~ c:" e bailer. F.x-smokers report the% emmie~y is one of the z. $~r -.r~'~'.,,~z [ of cessa~ioz amd %hal stre~e' can ~esd tc relapse. ~.:~r~. zr ~cr,. su~gesiio~. %hal those s~okers who do ah~%a~_D ~a.r be mcr~ .;.; c_-~ -,: stress-relate~. ~i~crders. It mus~ be noted that r~cclir,~_ h~s "_~: ~t-~': esta~2ished as the crucial aEe:~ for szzle%y reduct~c._: in, r.. at( a nu=be: cf other possible chemicals i~ c~gsre~te e=ol:e a~_~ z£t~::l~:-- of ~moke i=grediez%~ that could be i~voZved it %ramcui!/ir~t-~¢-_. Nevertheless, nicotime is the mos~ pe~-n¢ peFchcacti~'e agezt :-_ ~ obacco ~zo)-e. b. J.mger. Amcther item that made u~ the "ncr~ou.~ irr~ta%icz" faclcr c: s,'oking ia the EcEen,~ell (1970) %~s~io=,~aire was a:.ger. .-h:zL~ (1973) re~or',e~ that cigareite s=ck~r~ sm~ ~er:zer szck~r~ "La~ h_g;'~r ~ez'el~ cf irritability and anger in ccm;.arieor ~:tL ~--r,_ -c-- smokers. Vhez the smokers were div~ced izto l~gh~ (]~s~ ~L~:. ~" oar) an~ heavy groups, the heavy smokers showed grealer e~ger u~d~r sSress *.ha= ".he ligh~ smokers. Szckizc ~:heD an~r2 ~e,:" . ode :cejcr rea.-_-'m given for -#mcizinc by ax:,):er~ etten'~'_ng s ztc; -~:u!:~=;£ (!:--': (c3%),- ncz-:li_~ic e--okerz (z7~', amd e:'.'de:." szc. kers (6~%). ~u~'.., ~'. e= el, =?-z; ;,accur~c._ &Z;. %'e~,_e-, ~-" -" 0 C~ BAT Co LTD - MINNESOTA TOBACCO LITIGATION

done. Schachter (197S) reports a study by Perlick which compared irritability in "unrestrained" smokers and restrained" smokers who ~ere trying to cut dowm their smoking. Each group rated the annoFance caused by aircraft noise, after smoking either a 1.3 =£ ~icctine cigarette, a 0.3 mg n/cotime cigarette or not smoking. Dnrestrained smokers, when depraved or smoking the 0.3 mg nicotine cigarette, were more annoyed than vhen smoking the 1:3 mg cigarette, however, unrestrained smokers, when smoking the 1.3 a~g cigarette, vere mc ~etter tham non-smokers. ~e restrained group were ~ust as irritated in all three conditions as the deprived or low nicotine, unrestrained smokers. There are at least two explanations for these results; either irritability is an abstinence symptom which is the consequence of nicotine dependence, or smokers are const~tutionnlly more irritable and nicotine from smoki~ helps to reduce this feeling. Studies of abstimemce have shown hostility and aggression increase markedly during abstinence (~chechter and Rand, 1974; Shiffman and Jarvik, 1976). Anger commonly triggers relapse amQn6 ez-emckers, and these relapses occurred in the absence of smoking-related cues such as "people around were smoking~. (Shiffman, 1979 a)- Clearl~. the smoker who ie deprived of nicotine is less calm. Evidence Iron a study by Thomas (1973), which shoved similar scores for ameer when stressed for smokers and ex-smckers, suggests thm~ smokers ere constitutionally more angry amd irritable. I~ summary, smoking, and ~y inference nicotine, seems to have anti- anger effects as well as an anti-anxiety action. 3. Conclusion Nicotine seems to be a unique substance because it combines both perfcrmamce enhancement with anti-anxiety and a~ti-amger actioL C~ the basis of the present evidence. This finding alone is interesting, bu~ it is eve~ more significant when it is considered in ±he light of the smoker's precise personal control over nicotine intake. T::e data iz the preceding sections cam be combined to argue tha~ cigarette smoking ca~ b~ regarded ms self-medication for certain groups. ~udies of sel~-med~cstiom suggest that peo~le use those classes of agent wh±cb fulfil their own individual needs (Warburion, I~78) and the research presented here argues persuasively that peo~le titrste nicotine to obtain- a s~ecific dose to fit their needs. Depending om the situation, smokers w~ll adjust their smoking behaviour in terms of the number of cigarette~ smoked, smoke generation, and anoum~ of smoke inhale~ to control th~ nicotine levels reaching their brain amd Am this way their psychological state. Z,e~rivstior of cigarettes results in abs%imsnce s.vm~tom~ which are manifeste~ as a less efficienl function im certain situations, and so relapse is likely to occur. C~nsecuemti.v, the behaviour of s smoker is %he outcome of a cozy~z se'. cf interactions- It is determined, not on-~T by the =haracterist~cs of the situation, but the perscnali:y cf tl~_ imdi~'idual. Persona!itF characteristics w~ll determine ~he ~- ir • ~ick. the i~,~ividua~ selects a~,£ izier~rets infcrmat-~ c n ,'-~.blz th-: O C-D C_"~ ~jn BAT Co LTD - MINNESOTA TOBACCO LITIGATION

situation ie there is an intersctioz going on between individual factors and situatiomsl factors. If it is assumed that i~divi~ua) differences have a constitutional basis (is differences in the neurochemical pathways In the brain) then another complex interaction occurs when a ~harmacolo~ical agent is given to a person. These agents do not affect human performance directly, but they ~roduce chamgee in behaviour as e consequence of their interaction with endogenous chemicals w~thin the body. Changes. in neurochemical activity in the brain alter the way im which a person interacts with his environment and so performance reflects both the nature of the chemical agent and the biochemical state o~" the person. In this Sections, I have discussed the "stimulation" smoker end the "sedlstive" smoker as if they were separate homogeneous groups who smoked for the same reason all the time. However, there must be differences im smo~ng behaviour as s function of the intensity of the individual-situation interaction, its duration and more probably its density (the product of intensity and duration). In other cases Emokimg may not eve= be initiated until the intensi~y, duration or density of the interaction is at a critical level. Clearly, lighting a cigarette a~d the pattern of smoking behaviour controlling nicotime intake depends on the individual-situation interaction. l~terestin~ly, some smokers smoke in anticipation of s future need, before the situation has occurred or the critical stress level has been reached and they are still coping. }{ere the smoking behaviour is very clearly a response ~.o the individual's interpretation of the situation, his expectations a~out when may occur, rather than the objective characteristics of the situation. Smokers have learmed thsl to control their moo~ by smoking ~'il~ enable them to function more efficiently on some tasks and will help them avoid the um~esired consequences of other situations and so anticipatory smoking can te seen as s rational coping strategy on the basis of their past ex pe ri en ce • Advocates of abstinence and the operators of stop smoking clinics have mot considered the smoking habit from the point of view of the smoker's decision making. Smoking and absorbing n/cotine will enable more efficient function now, but future health is at risk. ¥ith abstinence the multiple benefits of xlicotine are lost bu.~ there may be better health in the future. Many smokers choose to continue amokiz~ and abstaining smokers relapse indicating the importance of the perceived ~enefits of nicotine to the smoker. ID the next section we • "il~ consider n~cotine i~ the context of some other substances which are used ~c control mood or to per/or- better. N~COTINE USE IN PERSPECTIVE I. Introduction k'~mL eval~ting m-i~otine use add the smoking habit, it is i~terestin~ ~c oct?ere n~cotime with coz~ouDds ~h~ch produce similar effects. Cc, zparis~ns wi£h reference substances are an essential ~art o~ ~he sssesamen~ o~ any therapeutic agent so that the relative costs and "cer~efits ca~ be es~inated and decisions made e~out the agez~'[ u~efu~nems. !z ±he case of nicotime, the ~elec~ed refsrenc~ ~g~.+,.p =2 BAT Co LTD - MINN~$OYA TOBACCO LITIGATION O u O O O

are two psycbostimulants (amphetamine an~ caffeine), two sedatives ( diazepam, Yalium, and alcohol) and marijuana (active i~£redient is -tetrahydrocannabinol or THC)p which is inhaled for its pleasurable effects. The comparison takes the form of an analysis of bot~ the pharmacokimetic add pharmacody~namlc features which could cons~i+~ute risk to the use i.e. absorption, user control of dose, duration cf action, s~e effects that impsir ~unctioning (acute toxicity), chronic toxicity a=d abuse potential. 2. AbsorF$ion As we have seen, nicotine iz~ha]ed into the lu~Es, is readily absorbed and travels to the brain within ted seconds reaching peak conce=trsticn immediately after the last puff of the cigarette, ~e about seven minutes after lighting up (see Section II). Am oral 6ose of alcohol must travel through the stomach to the major site of rapid absorption in the small intestine. Gastric emptying time and thus the onse~ of absorption results in a variable rate of uptake so tha~ peak plasma and brain concentra±io~ of a]coho2 are reached between 40 and 120 minutes (Lemberger and P, ubin, 1976). Diazepam is also abaorbe~ from the gastrointestinal tract and several hour~ are Deeded to achieve peak blood and brain levels (Eornestsky, I~076). Amphetamine and caffeine are a~aorbed from the stomach and peak comcentrations i~ plasma and brain are reached after ebou~ 30 minutes for caffeine (Marks and Kelly, 1973) and betwee= one an~ three hours for amphetamine (2eckett, 1970). Amphetamime is also takez intravenously and peak doaes in the brain can be ach~e~'ed w~thin ~ive mi=utes (Lemberger and Rubin, 1976). The absorption of TH~ into the bloodstream is rapid when inhaled but uptake into the braim i~ slo~° so that only one per cent appears in rat brain at 10 minutes (Lemberger and Rubim, 1976). Obviously, nicotine reaches potential active s~tes in the brain very quickly in comparisom ~'ith these referemce aubst~ces • Absorption time is significant ~or personal contro~ o~ effect and prevention of over-dosing; a fast absor~tio= time enables the user to adjust his dose to his require=e~ts and so contro~ t~e mag=itude of affect. Am ideal agemt for self-medication would enable the person to fi~ the'dose to his needs. Smokers have puff by puf~ comtrc] ~-ith brai~ effects occurring within 10 second~ whi)e a dri~_ker musl choose a© "appropriate" dose for effects in am hour or so. Thus i= term~ o~ personal comtro~ the experienced smoker has precise control over nicotine's psychoactive effects whereas n.o~t ~ri~k~rs have very haphazard control. Precision of co, trol is also particularly i=portamt whez considering toxicity because the risk of exceeding the optimal dos~ is minimieed. 3. I~rat~oz o~ Aotio= ~ second aspecf of phsrma:ck~r.etics, ~h~ch ie also related to ~rscna! c~m%rol of a person's psychologica~ sta~e, is duration of act~cr. Psychological mead.- vary a~d no~ p~cple wa~t co~ing ~echLi~ueE ~or specific occasions rather tha~ [or chronic states. Thus s~imu!a~i,r C~ O O O Cr~ BAT Co LTD - NIINNESOTA TOBACCO LITIGATION

may be needed for an arduous taak but not some hours later at bedtime. Nicotine has a half-life of only 20 minutes and is cleared from the body in about 40 mlnu~es (see Section If). In comparison, alcoho~ (Beckett, 1970) and caffeine (~arks and Kelly, 1973) have half-lives in the range of 3-4 hours and both may remain in the bo~y for ~-12 hours (Gilbert, 1976)- AAtbough these values are several orders of ma&-~itude more than for nicotine they are imsi~ifican% in comparison with the 20-50 hours for the elimination half-'life of diazepan (Breimer, 1979) and 56 hours for THC (Lemberger and Rubin, 1976). In addition, d~azepam and T~C have active metabolites that further prolong their action whi~e co%inine the major metabolite of nicotine seems to be virtually inactive. Clearly nicotine fits %he specifications of a compound which a person can use to exert fine co©tro~ over his psychological state on an hour by hour basis. 4. Specificity of Action One of the im~rtam% pharmacodynamic criteria for a therapeutically useful agent is that it is relatively specific in its site and mode of action. The acetylcholine, dopamins, moradrenalin and serotonin systems are al~ modified directly by alcohol (¥arburton, 1975), caffeine (Gilbert, ~976), diszepam (Warburton, 1975) mnd THC (Harris, ~978). In contrast, a~phetamine which on/y acts on the dopamine and noradrenalln systems, is relatively specific (Lemberger and Ruble, 1976). From the balance of evidence available, smoking doses of nicotine only act directly on some of the acetylcboline systems im ±he body (the sO called "nicotinic" pathways). This apecificit-/ of actio~ decreases the likelihood %hat a compound will produce unwanted "side effects', which will ~e discussed next. 5- Toxicity Almost all substances have toxic effects if taken in large doses for s sufficiently long time. In this section we will consider the acute, unwanted effects of the reference compounds and the chronic consequences of normally-used doses taken over a long period of time. Although amphetamine csu~ improve performance im laboratory tests, it ales distorts ~udgemen%. Amphetamine also acts as a sleep and appetite suppressant, and chronic use of this agent imvariably leads to loss of weight. Frequent small doses of amphetamines produce psychotic e~isodes in normal subjects (Kormetsky, 1976) amd repeated use of high doses result im severe psychosis. Caffeine is a much less harmful than amphetamia~e but is no% without toxic effects. Am average adult consuming over 600 mg of caffeine a 6ay will be more like]y to experience headaches, insomnia, aLxia~y and depression o~ ~aking. Repeate~ use increases the likelihoo~ of =yocardia] infarction, stomach ulcers, duodenal ulcerE, and carcinoma of the kidneys an~ urinary tract. ~Gilbert, 1976). Thus, although caffeine i6 not harmless, it is e much safer but ]esE potent compound than amph e t aml me. I~ view of alcoho~'F non-specific biochemical action it is nc~ surprising tha~ moderate domes of alcohol hsv~ a v~riety of scule unwauts~ consequences, including inpairment cf thought,, me=sty, <=D O C=D C=D Cr- CXD 3Z BAT Co LTD - ~INNESOTA TOBACCO LITIGATION

co©castratiOn, fans •ovement., •rid ao+.or CODI:Z'O1. ill • co]~aequmDoa, anxiol~£c donee •re incompet:L1Dle v£th most ~7~ou of york. l~.odera~ dz-An)c~ probably cause Yet7 ].~tt~e ~O~l.C T~:B::L~Ty ~t X~]:~ttSd ~Je results in larger and larger donee being requ:Lrod For anti-anziety effects. Long tern use of 1trier amounts 14ado ~o deterioration of ~he bra:L~, liver and other oz'nsJ~ (T(LllCren ~ Ba.z~y, 1971). The nest comma: side effects of diatepam is some drove:imams and inpeirnent of concentrates vh.ich are hazardous 1= • tired person or someone vho has taken s emil2 amount of alcohol. It should alma be remembered that alcohol persists in the body for days and so the alcohol IJ~eract±on could occur ~1o~£ after ¢ak£n( a done. The evidence for chronAc ~oxicA~y vith dlaseya¢ escalation o~ doses Chat ca= occur vi'bb alcohol (It, arks, 1978). Yro,~ the pe:Ln'., of wAev of acute and chronic toxlci~y, d£asepam ~s a much safer anzlolytic than alcohol. Alcohol use is of~ con]~red ~avou~a~ly vi't.h marijuana use, b~ the prob:lem vitb t.h~s tozici~r oompaz~Ason is that ve knov an 1Attle n~ou~ -tetrahydrocannabinol (YEn) and its active me~abolites. Perfoz~nance testing has revealed 1~at impaired oogzi~ive ability (nee reviev in Varburton, 197~). There is no evidence that repea~md casual use has adverse connequencae but an amottvattonal syndrome after frequent mari~ua use has been reported CVarbur~on, lCJ'7~). A typica~ ~ist of ~he ~o=Ic e~ects cT nicotine includes creating, tremor, nausea, vonltlng, abdominal pain, diarrhoea, ]~a~ite~io=, ~ntigue and he•dacha (Cohen and Roe, 1981) but, of these s.vm~to=s, only tremor occurs in experienced smokers vhich suggests ~at either tolerance to these effects has occurred or the smoker nitrates hie done ~ avoid a~vex~e effects. There is no ev:Lda~ce of i~tellect~al £npairaent du~ing use and, on the contrary, nicotine improves psychololAcal performance (nee Section ~T D). Thla benign action could be due to ?.be sinilarit:r betvee: ~.icotine and acetylcholl~,e, 8,0 that nicotine acts i: a naturn~ ray on neuroses and only ~roduces changes i: the brain vhich occur nor~a~_ly in e'tatoa of alertness. As a consequence of ±~s siailarlty to scetylcholine, Tolerance to its ac~o= o= slsc~rocortical arousal ~oes no~ occur (~urphree, ~7~.). This lack o~ tolerance in not surprising because neuroses cannot becol~e toln~t to ~.belr eve che¢ic~ls othe~-vise they vould cease to ~'=.1~c tio¢. ~'ithout tolerzLuce to the desired ef.~ects, smokers do no~ have to kee~ o= ~ncreasing the dose to achieve ~he desired amounts of sedation or s~imula~ion. Evidence for chronic toxicity d~e to nicotine alone, rather than c~garetCe smoke, £e sparse and ~here is little nearS_acing evidence of £ntel~ect~al impairment eve= after a li~etlne of use. Chronic nicotine use has sometimes bee= linked to gaatrointestlna! dist~rbaoce end cardiovascular d-,sorders but no studies ha~e controlled for ca~fei=e and alcohol co=su~p~ion. ~hue, nicc~.ne, i= SmoL..~J:~£ doses, ~s !.t~*,!e ]~o..'~ toXici~"p, i~sel-~ and the heal~ b,~,,ard ~o the smc~nr is A= the other smoke ccnsti~ue=ts. 6. Abuse Potential l o BAT Co LTD - MINNESOTA TOBACCO LITIGATION

Use of an7 substance can only be evaluated vith respect to the 8i%ua%i~ O~ usa, the msJ~mar of use as we~l a~ the consequences of use (3altar, 5974). Normal use can be defined as oaneumption for a~ innocuous or eonstr~ot£ye purpose in moderate amounts and in the a]rpr~prlste contezt in ~el~ne of place and culture. Abuse focuses on adverse consequences of usa in %er~s of phyalologiea£ or psychological eFFect. The consequences of abus• are arcs: damafe And i=pairmtnt of socia~ and personal funct£oning. Znd£rectly abuse refeTs to the ma~ner oF use by th• person (pattern and amount taken). The problem ©f abuse is 4~ the person's inter•orion v£th the co:pound and is not a fault of the compound £taalf. Hoverer, some substances are nora likely to lead to abuse than others and th£e a=pect is referTed to as able potential. The abase paten•is3 of a substance san be defined i: terms of ~ta £ntrins£c at~ractiweness, the number of users an~ the numkr of users who beast abusers of the 8mbsta~cts. a. Intrinsic £ttract±veneaa: Attractiveness refers to the ezten~ to which moat re•era find the immediate effect of • substance ins~rinstcally pleasing. Zt £8 independent of oultural factors end can be --sensed fro= aelf-adm~n£atrat~on 8~udies ~ animals. ~'h~le enamels will rspAd~y aelf-idmtJ=ister anphetal£nea shov£ng the high intl-l~sic attractiveness of than compound, animals ae~-edm~J~stor alcohol and THC to a much lesser extant. There is no evidence that a~imals rill sel~ administer diasepaze or caffeine, indicating negli~tb~e intrinsic attractiveness. Attempts to train an£nals to self :Lu3ect nAco~ine or "to smoke" h~ve been ra~her unsuccessful ~'hich su~ges~ very low ~u~rin~ic e~tractivene~.-" .......... These laboratory s~die8 fit with ~uns: oo=parisons of the comp~u=ds "- vh~ch 4~dicate ~J;at the product-Ass of euphoria seems t~ be esa~ieI • or high £ntrinsic attractiveness and for strong habLt For=a~on. Intravenous amphetam4ne ~roducea pleasure like s "Fhar~a¢oge=ic ori'um", alcoho3 8rod T'IIC ~u~e mNePlltS eu;d=oria, whale caffeine produces stimulation but no euphoria and diasepan is pleasurably relJu~ but does ~ot produce euphoria. Nicotine cad be either ~ild3y sti~u~at:L~4~ or pl~hly l~laxiz~ dependin~ on the situation, but is not • eu]mhoriant. b. Percentage of Abusers.- An abuser is a paras= vho i• conaumi:g a chemical i: sufficient quantlties to produce damage to heal~h. A co~pa~lson of the number of users and abusers is difficult because nAootkne, ©aFfe~ne and alcohol are readily available vh:i.le a~phs~tmine and 418sepam are controlled and mari.~uana (THC) is i:legal, h'he© amphetamines yore freely available on ~rsacrip~ion iu ~ritain abou~ 20% of ~J~e usex-s ,ere abusi=~ the comp~d (Eiloh an~ ~ra=don, 1952). Alcohol and caffeine are socially scceF~ab~e substances in nos~ Westea-n cultures, and, in these, about 5-7% a~use alccMc~ (}:'xee!n, 1972) and about N% are caffeine abusers (Gi3bert, 1976). It is estimated that ~-7~ oF T~C users are a~user~ (Kiesi~,, 1~T3). Abustr~ of dim•span are rare and current estimates arc less them 1% (K~rks, 1978). Strictly, ~eak~ng, the nun~•~ o,~ ab~er8 o,~ ~cot~--ue is zero ~,-t ~ since ~ev individuals take n.icotine alone I~ x~icct~e ie nc~ E'nc~-= tc ~nTx'Ib~te to the amoke-relate~ ~/aor~ers (Cobs= an~ ~oe, 19~I). (t "~ (':-" / ~..: .: . Abuse potential has been defined am terms o,~ ~ifficui~y cf s~cyylng • "" " ..~ ,,,f.-: :'~ 0 m (_7"I BAT Co LTD - MINNESOTA TOBACCO LITIGATION

use, but comparable data are only available o= slcoho~ •n~ smck~ng. ~u=~ a~ kspal•c (1~4) coapLre~ l~e relJ~se rates for n~cohoZics --~ smokers vl~ ~ero~.D (am a&~mt vt~ vary ~gh ~tl-i~ic •ttrac~iva:eso) and £o~d that the relapse curves yore ve:-y .8~milar. ~ovever, •s • ~•££e awd JarY~lr (1978) ]point out:, 1:h•1: vhi2e one nay wonder vb•t drLves a heroin abuser 1:o relapse ~n 1:he face" o2' the social disapproval, tbe ~hysie•l hasaz~s Lud lsgs~ x~aks of berlin usa, it is not surprising that •z-smoker8 ~elmpse given the mu£t~pla motives for /+ .... smol~.u~,"" ' '.~- "~/ "/ "~£'-t'-:,..,. ,- • I ' ];ico~ne, as used by l~e smoker, ~s s low risk tu~a~•nc• ~n te~ms O~" acute and ch~ozcLc 1:o~ici~y amd 1:he smoke-related dim•nee art dos to the other smoke con~ti~uem1:s. =~g l~tpid uptake ~to ~he brsAn •Stays ~eruonal control o~ their pey~hol~gica~ s~ate at will so that smokers can obtain sA1:~•r stimuls1:ton or mad•tics to betp ~em ~o cope vit~ situations. Bico1:ine mc~s on ~he b~aAm producing changes which •=e within nor~=l limi1:s and so ~roduces -timuls~ion and sedation. 31cotl:a has lov abuse potent:laX bug: sno~ng is a str=ng hab~ because i~ enables s:cksrs to hate ozquisite parson•2 control over their ]~ry cho l~q~ica.l f~c~icn~ng. Y~. CO]NCI~DX]~; COEE~rTS Smoke~n 3~f a rod of bux~iz~ tobacco 8~d ~e ma~orA~y of ~o= inha~le the smoke. As • consequence o~ smoke intake, a number of smoke cons1:ituenta Including nicotine are absorbed. Z1: is absorbed very e£~icient~y, enters the brain very quickly and As me~abolissd quickly giving a ~rAef duration of action. Smokers are sensitive to their ~lasma z~Icott~ne levels and can control 1:heir in1:ake o~ nicotine. Smokex~ cont~:l ,,'~oo1:~.ne by vax-y'A4~ ~slr cigarette oc~m~tic~, their strength o~' ~aaffing and their 4~als1:2~n. ~As contx~] is nora than avoidance 0£ hA~ 1:oxic domes of ~ccl:ine bu~ is ~ at:tempt to tl~rate ~t ~:u~=(. "; for s specific nlco~ine dose. The m=okar's aim I: t~trating is o~.a~ a~ optimal doBe Of ~cot:L~e V~iCM vil2 •c~ on the ~y a~ sstAsfy some need. E~e~-ime~te have ~vttled ~ ~cot~e has ma~.~: tc~ions o~ ~e ~o~ bvt the most cons~s1:snt effect vtt~ s~ok~ng ~ose8 o~ nicotine is stimulation e~ • subset of •the cho!inerg~c neuroses (ie the "nico~ic" ps~hv&ys), including 1:he a~to=omic nervous system a:d pathways to the cortez and linbic ~ysieno The outcome of these actions is an increase In blood 2ovals of ca1:echolsnines and ~lucocorticoids sad great:or amounts Of cortical de~sy~chroz~Ir~tion, vh~c~. are vt~ the normal iAmi~8 of the average persc~. Smokers beZieve tba~ clgarettea alleviate =ental and muscular fatigue an~ these effects OL~ be s~1:ri~ted, i= part, 'to ~e mob£1i:stio~ o~" energy reserves ~.., 1:,b_e. c8~ec~o!a_n.ines and g~ucoccl'ticcida~ S~okel~s also Claim 1~at s~ok:L.1~ hs~s t~e= ~ a:~ concentrate a~ ezper~menta~ studAss have shown 1:hs~ snok.ing can produce absolute enhancements in performance •e vel~ as ~revonting 1:he ~er~or=a:ce dacreu~ts. ~anc~ ~tz~or~ance of ~.he 8a~s magnitude yes fo~ v~ n~co1:ins ta':lets were givan 1:o non-smokerw. ~ ~mlrtl ~jnpcrt~wt e~ect ~7 0 C~ BAT Co LTD - ~IlNNESOTA TOBACCO LITIGATION

of nicotine is its sedative action which has been supported by questionnaire and experimental studies of smoking. Measures of smoking behaviour during performance and while subjects are under stress show that smokers vary their nicotine intake according to the situation. In this way they obtain the required nicotine dose for stimulation or sedation dependin8 on the situation and nicotine's pharmacokinetica make it suitable for hour-by-hour control of a person's psychological state Nicotine taken in smoking doses seems to be relatively safe for healthy adults to use and stands in marked contrast to the deleterious short term and long term consequences of ~.he socially acceptable substance, alcohol. A number of important and controversial implications follow from these facts about nicotine. If smokers derive beneficial effects from nicotine and cigarettes are the most effective method of administering nicotine, then cigarettes should be designed to deliver n/cotine and its beneficial effects with minimum risk from smoke-related diseases. The current trend of continuing ~o reduce n/cotine as well as tar and carbon monoxide cannot have the expected health benefits because smokers are compensating by puffing harder and inhaling more smoke. It has been argued that because compensation can on/y be ~artisl with low delivery brands, intake of tar and carbon monoxide is still less than when high delivery brands were smoked and so these products have less risk (Rawbone, 1979). This argument vould be true if cigarette consumption did not increase. However a recent paper (%'sld et al, 198~) reported that fro= 1949-1974 the sales-weighted average nicotine monotonically decreased in the U.K. by 3-0% per annum. Over these 25 years, cigarette consumption for male smokers (RSL, 1981) increased by 50%, an average annual change of 2.0%, reflecting two-thirds compeusa%ion by consumption alone. The increase i~ consumption for women was eves grea'~er, a.~ it became fashionable for them ~o smoke. A more effective approach for minimising risk and maximising benefit would be a produce with medium nicotine delivery but a re4uction cf some smoke components. For this strategy to be successful, it is obviously crucial that this ci~are%te has sufficient flavour. !1 is therefore essential that the ~roduct would not need to be smoked more intensively, or require more smoke to be inhaled, in order to satisfy taste, or to deliver the dasired amount of nicotine to the smoker. Unfortunately, we have only a limited amount of toxicological information on a few of tha 4000 identified compounds in tobacco and tobacco smoke on which to base a selective reduction strategy. It is said that tobacco smoke is a relatively mild carcinogen (Tsc, 1980) which means either that the hazardous compound or compounds are on/y weak carcinogens or the toxic co~pound is present in very small quantities. In the latter case the task of identification of the constituents to be reduced will be formidable. Toe tobacco industry has methods for =~difying smoke delivery in order tc ~artielly dissociate m~cotine from tar. Some of the ~cre drartic smoke reduction ~rocedures change the chemical nature of the smoke (Guer~n, 1980) an~ the tcrico]ogica! i:.~licatio~ of these shifts i~ umkDcwn. P.usseil (197°) has brief~.v reports5 some tests of re.~ucad %at, re~iur O O O O Cr~ P-D BAT Co LTD - MINNESOTA TOBACCO LITIGATION

nicotine cigarettes But him comments imply that they were less acceptable. Thus the manipulations of smoke delivery have beer unsatiafactory so far, not only because they may leave harmful smoke compounds but because they remove important flavours. ,Clearly, future progress must be in the direction of reduction of specific smoke constitutemta to reduce rlak but maintain flavour by adding flsvourimg in order to preserve the characteristic full £1avour impact for the smoker. The way to a better cigarette is not an easy one, but the development of a medium nicotine cigarette, with reduction of some smoke components, and added flavour, could well represent an important milestone along this path. The end of the road could not only be a safer cigarette but also with important benefits to the smoker. EEFERE~CES ADAMS P I: The influence of cigarette smoke yields on smoking habits. In Smoking Behaviour, ed by R E Thornton, pp 349-360, Churchill- Livingstone, Edinburgh, 1978. ANDERSS0~ K: Effects of cigarette smoking on learning and retention. Psychopharmacologia 41: I-5, 1975s. ANDERSSON K .I HOCKEY G R J: Effects of cigarette smoking on incidental memory. Psychopharmacolegia 52: 223-226, 1977. ANDERSSOE K & POST B: Effects of cigarette smoking on verbal rote learning and physiological arousal. Scand J Psychol ~5: 263-267, 197A. ARHITAGE A K: Some recent observations relating to the absorption of nicotine from tobacco smoke. In Smoking Behaviour, ed by h" 1, Duma, pp ~3-91, ~'inston (J Wiley), ~'ashington DC, 1973. AREITAGE A K, HALL G H and SELLERS C E: Effects of nicotine on electrocortica] activity and acetylcholine release from the cat cerebral cortex. Brit J Pharmacol 35: 152-15C, 1969. ASHTON H, HARSH ¥ R, EILLEA~ J E, P.AWLINE E E, TELPOR~ P,, THCFFSON J %': The ~se of even-related slow potentials of the brain as a means to analyse the effects of cigarette smoking and ZLicotine im humans. In Smoking Behaviour, ed by R E Thornton, p~ 54-68, Churchill- Livingstone, Edinburgh, 1978. ASHTON H, ~TEPh~Y R, and TROKPSON J ~: Smoking behaviour and nicotine intake in smokers presented with s 'two-thirds' cigarette. In Smoking ~ehaviour ed by R E Thornton, pp 315-329, Churchill-Livingstone, Edinburgh, 1978. A~U,-r,, I~, STEPEET'P. and THOMPSON J %': Self titration by cigarette holders. ~rit-Ked J 2: 357-36C, 1°79- ASHTOI; H and ~ATSC.]; I ~: Puffing frequency and nicotine izteke in clgare~%e smokers. 5ri% Eed J ~: 679-6~I, 197C- 0 o o 4 BAT Co LTD - MINNESOTA TOBACCO LITIGATION