CBS Video

See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/22630083

Behavioral treatment of caffeinism: Reducing excessive coffee drinking

Article  in  Journal of Applied Behavior Analysis · February 1979

DOI: 10.1901/jaba.1979.12-335 · Source: PubMed

CITATIONS

54 READS

4,283

2 authors, including:

Some of the authors of this publication are also working on these related projects:

Social skills training: The current status of the field View project

Richard M Foxx

Penn State University Harrisburg

129 PUBLICATIONS   5,196 CITATIONS   

SEE PROFILE

All content following this page was uploaded by Richard M Foxx on 24 October 2016.

The user has requested enhancement of the downloaded file.

JOURNAL OF APPLIED BEHAVIOR ANALYSIS

BEHAVIORAL TREATMENT OF CAFFEINISM: REDUCING EXCESSIVE COFFEE DRINKING

R. M. FoxX AND ANDREA RUBINOFF

UNIVERSITY OF MARYLAND, BALTIMORE COUNTY

Excessive coffee drinking can have deleterious effects because of the large amounts of caffeine that are ingested. Caffeine is thought to be addicting, and prolonged and exces- sive use can lead to caffeinism, a condition that has serious behavioral and physiological side effects. The present study developed and evaluated a treatment program to reduce excessive daily coffee drinking to moderate and presumably safer levels. Three habitual coffee drinkers received individualized changing criterion programs that systematically and gradually reduced their daily caffeine intake. The coffee drinkers were required to self-monitor and plot their daily intake of caffeine. They received monetary prizes for not exceeding the treatment phase criteria and forfeited a portion of their pre- treatment deposit when they did. Their coffee drinking decreased from almost nine cups per day (over 1100 mg of caffeine) during baseline to less than three cups per day (less than 343 mg) at the end of treatment or a reduction of 69%. The treatment effect was maintained during a 10-month follow-up, averaging a 67% reduction from baseline. The program appears to be a reasonable method of reducing and then main- taining daily caffeine intake at less harmful levels. DESCRIPTORS: self-recording, coffee drinking, caffeinism, changing criterion de-

sign, adults

Caffeinism describes a set of behavioral and physiological symptoms caused by the exces- sive consumption of caffeine-containing sub- stances (Greden, 1974; McManamy and Schube, 1936; Powers, 1925; Reimann, 1967). The symptoms include nervous irritability, tremu- lousness, occasional muscle twitchings, insom- nia, sensory disturbances, tachypnea (an ab- normally rapid rate of breathing), palpitations, flushing, arrhythmias (an alteration or abnor- mality of normal cardiac rhythm), diuresis, and gastrointestinal disturbances (Greden, 1974; Powers, 1925; Reimann, 1967; Truitt, 1971). Individuals suffering from caffeinism are some- times misdiagnosed as anxiety neurotics because of the similarity of the symptoms (Greden, 1974). The deleterious effects of caffeine on hu-

The authors wish to thank Mark Moody and Rich- ard Wolff for helpful comments on an earlier version of the paper. Reprints may be obtained from R. M. Foxx, Psychology Department, UMBC, 5401 Wilkens Avenue, Baltimore, Maryland 21228.

mans also may include increasing the possibility of coronary heart disease in susceptible persons, promoting the progress of atherosclerosis and affecting chromosomal structure or action (Punke, 1974).

Caffeine is America's most widely used cen- tral nervous system stimulant; it is contained in various popular beverages such as coffee, tea, and colas; prescription medications such as APCs; over-the-counter analgesics such as An- acin; over-the-counter stimulants such as No- Doz; and cold preparations such as Dristan (Greden, 1974). Of all the sources of caffeine, coffee contains the greatest amount of caffeine in a single serving and is ingested more often by more adults than any other caffeine source or beverage. In 1972, Americans spent over 1.4 billion dollars for over 150 billion cups of cof- fee or a per capita consumption of 36 gallons, and more persons over age 10 drank coffee daily (64%) than milk (5 1 %), soft drinks (47 %), or tea (27%) (Ray, 1974).

335

1979, 129 335-344 NUMBER 3 (FALL 1979)

R. M. FOXX and ANDREA RUBINOFF

Thus, it is not surprising that the primary cause of caffeinism is excessive coffe drinking. A daily intake of over 1000 mg of caffeine (ap- proximately eight cups of brewed coffee at 125 mg per cup) is considered to be potentially quite harmful (Greden, 1974). Currently, caffeinism is a little recognized yet widespread problem be- cause of the number of individuals whose daily ingestion of brewed coffee exceeds eight cups. A major problem in treating caffeinism by

reducing an individual's daily coffee intake is that caffeine may have addicting properties (Ray, 1974), as shown by documented withdrawal and tolerance effects. Individuals who ingested more than six cups of coffee daily have described physiological withdrawal effects, such as head- ache, irritability, and nervousness, when their daily intake was reduced substantially (Goldstein and Kaizer, 1969), and these symptoms disap- pears when they were given 300 mg dosages of caffeine (Goldstein, Kaizer, and Whitby, 1969). Caffeine withdrawal effects, especially headaches, have also been produced by giving large doses of caffeine (in capsule form) over time and then substituting placebo capsules (Dreisbach and Pfeiffer, 1943). Excessive coffee drinkers also appear to acquire a tolerance to two actions of caffeine, diuresis (Eddy and Downs, 1928) and salivation (Winsor and Strongin, 1933), since regular coffee drinkers require large caffeine dosages to produce these physiological actions, whereas occasional coffee drinkers or abstainers require small dosages. Furthermore, coffee drinkers can be given caffeine at bedtime with- out experiencing ill effects, while nondrinkers on the same dosage report difficulty falling asleep (Colton, Gosselin, and Smith, 1968). Fi- nally, coffee drinkers tend to increase their daily consumption over a period of years in order to produce the same desirable physiological ef- fects of caffeine that had been experienced from smaller amounts in earlier years (Goldstein and Kaizer, 1969). To date, there are no behavioral studies of

caffeinism. However, recent smoking research (Foxx and Brown, 1979) may provide a partial

model to apply to caffeinism, because coffee drinking and cigarette smoking are somewhat similar. Both involve the ingestion of sub- stances, caffeine and nicotine, that are consid- ered to produce physiological dependence, and both have a multitude of behavioral aspects associated with them that could be described as psychological dependence. The Foxx and Brown (1979) program considered both the physio- logical and behavioral factors involved in the smoking habit by systematically weaning smok- ers from nicotine (they changed brands until they were smoking the lowest nicotine cigarette commercially available, at which time they were to quit smoking), while also providing them with positive feedback regarding their success (they plotted their daily intake of nicotine throughout treatment).

The present study developed a program, mod- eled in part on the Foxx and Brown smoking study, that would decrease the caffeine ingested by excessive coffee drinkers to a moderate level. Moderation was the goal rather than absti- nence since, unlike nicotine which is harmful in any amount, the ingestion of reasonable amounts of caffeine (five cups of brewed coffee or less) has not been shown to be harmful.

METHOD

Subjects

Individuals interested in reducing their cof- fee intake were asked to complete a question- naire that asked them the following: (1) demo- graphic and biographical information; (2) the number of cups of coffee consumed daily; (3) the number of years they had drunk coffee daily; (4) the brands and types of coffee consumed, i.e., Maxim instant; (5) their reasons for wanting to reduce their coffee consumption; (6) their longest period of abstinence since they had be- come regular coffee drinkers; (7) to list and rank order all other beverages that they con- sumed regularly; and (8) to list any prescrip- tion medications and over-the-counter stimulants

336

TREATING CAFFEINISM

taken daily and their dosages.1 Three of seven respondents were chosen. The three met the following criteria: (1) They consumed eight or more cups of brewed coffee per day (over 1000 mg caffeine); (2) They reported being bothered by "coffee nerves" and other physiological and behavioral symptoms associated with excessive caffeine intake; and (3) They expressed a desire to reduce their daily coffee consumption. Sub- ject 1 was a female school teacher married to Subject 2, a graduate student who also was em- ployed full time. Subject 3 was a psychologist. They had been regular coffee drinkers an av- erage of 12 years.

Experimental Design

A changing criterion design (Hartmann and Hall, 1976) was employed to reduce systemati- cally and gradually the subjects' average base- line caffeine intake over a 4-week treatment period.

Procedure Baseline. During baseline, the subjects re-

corded the number and types of all caffeine- containing beverages they drank daily. The base- line was 16 days for Subjects 1 and 2, and 14 days for Subject 3. At the end of baseline, they met individually with the second author who calculated their baseline daily intake of caf- feine in milligrams. The treatment goal for all subjects was to decrease their daily caffeine in- take from their baseline level to 600 mg (fewer than five cups of brewed coffee). This 600 mg figure was subtracted from each baseline mean and then divided by four (the number of treat- ment phases) to yield the number of mg of caffeine reduction for each phase. For example, during baseline, Subject 2 consumed an average of 1148 mg of caffeine daily. His caffeine reduc- tion for each phase was 137 mg (1148 mg minus 600 mg divided by four).

1A copy of the Coffee Drinker's Questionnaire, a copy of the treatment contract, and copies of the various recording sheets can be obtained by writing R. M. Foxx.

Treatment. At the post-baseline meeting, the subjects signed a contract that explained the study's procedure and rules (see Footnote 1). They were required to deposit $20, half of which was returned in four equal steps during the treatment, and the other half at the 3-month follow-up. When the subjects did not exceed the criterion level on any day during a treat- ment phase, they received one-quarter of their $10 deposit ($2.50), a $1 bonus, and remained eligible to receive a $10 bonus at the end of treatment. If the subjects exceeded the treatment phase criterion on any day, however, they would forfeit the $2.50 portion of their deposit (to be donated to American Heart Fund), the $1 bonus for that phase, and the $10 end of treatment bonus. When a subject exceeded the criterion, the treatment phase was recycled from that day and kept in force until the required number of consecutive days at criterion for that phase were met. The subject remained eligible to receive the remaining portion of the deposit and the $1 bonus for not exceeding criterion during each of the remaining phases. However, if a subject exceeded the criterion more than once during any phase, the remaining deposit would be do- nated to charity.

At the beginning of treatment, the subjects were given a list of all caffeine-containing bev- erages and their caffeine content. Caffeine-con- taining medications were not listed, because none of the subjects had indicated in their ques- tionnaires that they took any. This beverage list was modified from one compiled by Greden (1974). Whereas Greden listed the range of caf- feine (mg) contained in a cup of a particular beverage, e.g., brewed coffee contains 100 to 150 mg of caffeine, in the present study the mean of each range was used, e.g., brewed coffee was listed as containing 125 mg of caffeine per cup. Subjects were encouraged to switch from brewed coffee (125 mg caffeine) to instant cof- fee (93 mg caffeine) or to a beverage containing less caffeine, e.g., tea (68 mg), cola (50 mg), or, better yet, to decaffeinated coffee (only 3 mg caffeine). The meeting ended with the second

337

R. M. FOXX and ANDREA RUBINOFF

author describing the hazardous physical and behavioral effects of excessive caffeine intake.

Recording. Each day during treatment, the subjects recorded the number of cups of coffee and other caffeine-containing beverages they drank. Their recording sheets contained a list of the caffeine value (mg) for each beverage (see Footnote 1). After consuming a beverage, the subject recorded its caffeine content, the activity associated with drinking the beverage, and cal- culated the total intake of caffeine to that mo- ment. (The requirement of keeping a daily running account of total caffeine intake had been specified in the contract.) At the end of each day, they plotted their total caffeine intake, caffeine intake of coffee, number of cups of all caffeine-containing beverages, and the number of cups of coffee on graphs that had been pro- vided at the beginning of treatment. On the total caffeine intake graph, a criterion line was drawn across each treatment phase that indi- cated the maximum mg of caffeine that could be consumed per day. This criterion level of mg of caffeine was written also on the daily re- cording sheets. At the end of each treatment phase, the recording sheets were collected and examined for correctness. At this time, the sub- jects were instructed about the next phase and were given the money that they had earned.

Once treatment ended, the subjects ceased self-monitoring and the follow-up began. Dur- ing follow-up, the subjects were contacted by telephone or in person every 2 weeks and told to record the number and types of all caffeine- containing beverages they drank for the next 2 days. They were contacted the day after the follow-up and their reports were recorded. In addition, they mailed their records to the ex- perimenters or presented them in person.

Reliability Treatment reliability. The reliability prob-

lems associated with the recording of coffee drinking behavior were characteristic of those found in many other self-control studies, since the subjects self-recorded their behavior. As a

result, the reports of other individuals, "signifi- cant others," are necessary to corroborate the subjects' reports (Lichtenstein and Danaher, 1976). In the present study, each subject sub- mitted the names and addresses of two indi- viduals (one of whom could not be a relative) who were familiar with the subject's daily cof- fee drinking. Letters were mailed to the signifi- cant others, requesting that they monitor the subject's coffee drinking, together with a state- ment that was to be signed and returned, if they agreed to do so. Enclosed in the letter was the subject's signed release form indicating that he or she agreed to be monitored. Furthermore, all subjects made a public announcement at work that they intended to reduce their coffee drinking.

The significant others were telephoned at the end of treatment and asked, "Have you seen any differences in the subject's coffee drinking?" This end of treatment check was conducted to ensure that the subjects had noticeably reduced their coffee drinking during treatment. Because it was impossible to determine exactly whether the subjects met criteria every day, the signifi- cant others' reports were used to provide a gross reliability measure of the subjects' records. All significant others stated that they had observed either a dramatic reduction in the number of cups of regular coffee consumed across treat- ment phases and/or a noticeable increase in the number of cups of decaffeinated coffee consumed. An additional reliability check was to deter-

mine if the subjects were falsifying their records. At each subject meeting, all records were scru- tinized to determine if they all concurred. No errors were ever discovered, which suggested that the subjects had honestly reported their caffeine intake.

Follow-up reliability. In self-control studies, the most meaningful data are the posttreatment (follow-up) results because of the common re- lapse problem. Thus, the ultimate determinant of the caffeine reduction program's success would be whether it resulted in a stable and last-

338

TREATING CAFFEINISM

ing modification of coffee drinking after treat- ment. Accordingly, special attention was paid to the veracity of the subjects' follow-up reports by contacting the significant others at each fol- low-up check.

One, and often two, of the subjects' significant others was telephoned immediately following each 2-week follow-up check and asked the following questions: (1) The brand(s) of coffee the subject drank most often; and (2) Whether the subject was consuming approximately more, less, or the same number of cups of coffee daily during that follow-up period as he/she had been consuming at the end of treatment and during any previous follow-up periods. In all cases, the significant others' reports agreed with the subjects' reports regarding brands of coffee and were within one cup regarding the number of cups consumed. Inasmuch as Subjects 1 and 2 were married and served as each other's sig- nificant other, their second significant others were always contacted to corroborate both their self-reports and their reports on each other.

BASELINE

w

4

I- w

z

-J

4

M-

0 IW6

1300'

1200a

1100'

1000'

900

800

700

600

500

400

300 a

200

TREATMENT PHASES

1 2 3

RESULTS

Figure 1 shows that Subject 1 never exceeded criterion and that her mean daily intake of caffeine decreased from her baseline of 1008 mg to 357 mg (a decrease of 651 mg) during the fourth and final treatment phase. Her mean daily intake of caffeine during the 10-month follow-up was 298.4 mg.

Figure 2 shows that during treatment Subject 2's daily intake of caffeine decreased from his baseline mean of 1147 mg to a mean of 357 mg (a decrease of 790 mg or over six cups of brewed coffee) and that he never exceeded cri- terion. His mean daily intake of caffeine during the 10-month follow-up was 250 mg. Subject 2 reported that his low, stable posttreatment con- sumption of caffeine through the 3-month follow-up was because he had stopped brewing coffee at work which prior to treatment had been his major source of caffeine. Between the 3- and 10-month follow-up, Subject 2 reported

FOLLOW-UP 4

SUBJECT 1

v

1554 ,

7 .105 10 is 20 25 30 3S 40 44 SS 72 86 loo 114 128 310 324

DAYS Fig. 1. Subject's daily caffeine intake (mg) during baseline, treatment, and follow-up. The criterion level

for each treatment phase was 102 mg of caffeine less than the previous treatment phase. Solid horizontal lines indicate the criterion level for each phase. Broken horizontal lines indicate the mean for each condition.

I

339

R. M. FOXX and ANDREA RUBINOFF

that he had given up coffee and had substituted herbal teas.

Figure 3 shows that Subject 3's mean daily intake of caffeine decreased during treatment

from his baseline of 1175 mg to 314 mg (a de- crease of 861 mg or almost seven cups of brewed coffee). His mean daily intake of caf- feine during the 10-month follow-up was 537.3 mg. He exceeded criterion twice, on days 27 and 42. Although Subject 3's caffeine intake during follow-up was quite high in comparison to the other subjects' follow-ups, he did average below the 600 mg treatment goal of less than 5 cups

of brewed coffee per day. Table 1 lists the mean number of cups of caf-

eine-containing beverages consumed per day during each treatment phase and follow-up. The table reveals the strategy that the subjects em-

ployed in order to avoid exceeding the criterion levels, since it shows the amount and types of reductions that were made. Subjects 1 and 2 simply reduced the number of cups of brewed

BASELINE

A... 1500

_ 1400

1300

1200

1100.

'I 1000

Z 900a

ui 800z - 700aLA.

4 600 a

mi 500 a

49 400. 0

300 a

0.

1

DAYS

Fig. 2. Subject 2's daily caffeine intake (mg) during baseline, treatment, and follow-up. The criterion level for each treatment phase was 137 mg of caffeine less than the previous treatment phase. Solid horizontal lines indicate the criterion level for each phase. Broken horizontal lines indicate the mean for each condition.

TREATMENT PHASES

2 3

coffee they drank. They did not substitute bev- erages containing less caffeine except during the first treatment phase. Subject 3, however, achieved and maintained his reductions by de- creasing his intake of brewed coffee, switching to instant coffee and substantially increasing his consumption of decaffeinated coffee.

Amount of Money Earned

Subjects 1 and 2 earned $14, or $4 for meet-

ing each treatment phase criterion ($1 per

phase) and the $10 bonus for never exceeding criterion. They were refunded their entire $20 deposit ($2.50 per phase for not exceeding cri- terion and $10 for participating in the follow- up). Subject 3 exceeded criterion once in two

different treatment phases. As a result, he earned only $2 and forfeited the $10 bonus. He also lost $5 of his deposit for exceeding two treat-

ment phase criteria. He' was refunded $5 of his deposit at the end of treatment and the remain- ing $10 after follow-up.

FOLLOW - UP

4

SUBJECT 2

.......................

ssfOI I - * I 5 1 1 2 w 3 35 44 5 7 - 310 3

5 10 is 20 25 30 35 40 44 58 72 86 100 114 128 310 324

-W 5 0 m IVAT- - v . . ff a .mw-

D

1 0

4 p

4

340

30MO-I

TREATING CAFFEINISM

BASELINE

0X

w

4 "I z

'a. -J

z

49 I~- 0

TREATMENT PHASES FOLLOW-UP

1 5 10 15 20 25 30 35 40 45 50 64 78 92 106 120 134 316 330

DAYS Fig. 3. Subject 3's daily caffeine intake (mg) during baseline, treatment, and follow-up. The criterion

level for each treatment phase was 144 mg of caffeine less than the previous treatment phase. Solid hori- zontal lines indicate the criterion level for each phase. Broken horizontal lines indicate the mean for each condition. Arrows indicate days on which the treatment phase criterion was exceeded.

Anecdotal Results All subjects reported positive physiological

and behavioral side effects at the end of treat-

ment. Subject 1 said that she felt "healthier." Subject 2 reported that he noticed he was less tense and "hyper" at work and in his relations with others, and that he felt better physically. Subject 3 reported that he felt less irritable.

DISCUSSION

The results showed that the program was

effective in reducing excessive coffee drinking to moderate and presumably safer levels. The subjects' coffee drinking decreased from almost nine cups per day (over 1 100 mg caffeine) dur- ing baseline to less than three cups per day (less than 343 mg) at the end of treatment. By the final treatment phase, Subject 1 had decreased her mean daily intake of caffeine by 64%, Sub- ject 2 by 68.8%, and Subject 3 by 73.2%. This

effect was maintained during a 10-month follow- up; the subjects averaged a 67% reduction from baseline.

Several factors may have contributed to the success of the program: the changing criterion treatment, the individualized programs for each subject, the self-monitoring procedures, positive feedback, the deposit and contract, and the posi- tive and negative short- and long-term mone- tary consequences. However, at this time, we do not know which factors, either singularly or in combination, produced the treatment effect because no component analysis was conducted. Although doubtful, it is possible that none of the program components were necessary and that other factors were responsible for the treat- ment effect. Other factors that could have in- fluenced the subjects include: (1) the informa- tion they received on the deleterious effects of caffeine and on the relative amounts of caffeine in various beverages; (2) the minimal encour- agement they received from the experimenters;

341

R. M. FOXX and ANDREA RUBINOFF

Table 1 Mean number of cups of caffeine-containing beverages consumed per day by each subject per condition

Caffeine-Containing Beverage Brewed Instant Decaffeinated

Length of Coffee Coffee Coffee Colas Tea Condition Subjects Phase (Days) (125 mg) (93 mg) (3 mg) (50 mg) (68 mg)

Baseline Si 16 7.43 0 0 1.06 .43 0 .5

0 0

1.0

5.6 0 5.6 0 3.4 .8

4.7 0 4.5 0 2.4 1.0

2.85 2.85 1.84

3.0 0 3.0 0 3.5 1.5

0 0

.68

.07 .68 0

0 1.28 0 1.57 .57 0

0 0 0

0 .6 0 0 1.2 0 oa 0 0

0 0

4.7

0 0 0 0 .69 4.69

0 0

4.75

0 0

0 0

0 2.75

0

0

.5

Si 4 2.0 0 S2 4 0 0 S3 4 4.5 .25

3 forgot to record decaffeinated coffee during this phase.

0

0

3.5

0

0

5.0

.77 0

.44 0

.20 0

0 0 0 0 0 0

0 0 0 0

1.0 0

0 0 0 0 0 0

0 0 0 0 0 .25

0 0 0 0 0 0

(3) their motivation to reduce their coffee con- condition in which the subjects were merely sumption as evidenced by volunteering for the asked to reduce their caffeine intake for a speci- project; and (4) the social pressure associated fied period of time. with having made a public announcement that This study depended on the reports of signifi- they were reducing their coffee consumption. In cant others to corroborate the subjects' self- summary, the present study failed to control for reports. The follow-up reports seemed accurate

self-induced change. This variable could have because: (1) each report was verified by one and been controlled for by including a pretreatment in many cases two significant others; (2) the re-

S2 S3

16 14

Si S2 S3

8.5 9.0

5.85 6.57 6.1

Treatment Phase 1

Treatment Phase 2

Treatment Phase 3

Treatment Phase 4

Follow-up Month 1

Follow-up Month 2

Follow-up Month 3

Follow-up Month 10

'Subject

7 7 7

S S S

9 9 10

7 7

13

4 4 4

4 4 4

4 4 4

Si S2 S3

Si S2 S3

Si S2 S3

Si S2 S3

Si S2 S3

Si S2 S3

2.5 2.75 3.6

2.25 2.25 3.0

342

TREATING CAFFEINISM 343

turn of the remaining half of the deposit was dependent on reporting the number and types of beverages consumed rather than caffeine con- sumption; and (3) the effects of feedback were minimized somewhat because the subjects were not required to total or plot their daily caffeine intake. Several factors suggested that the sub- jects honestly reported their treatment data: (1) all significant others reported that the subjects had noticeably reduced their coffee consump- tion; (2) careful scrutinization of the subjects' graphs and records revealed no errors or falsifi- cations; (3) the married couple subjects' outside significant others always verified both the couples' self-reports and their reports on each other; and (4) Subject 3's self-reported infrac- tions added credibility to the assumption that at least one subject was honestly and reliably re- porting his treatment results.

There are a couple of ways that additional partial reliability checks of the subjects' treat- ment reports could be made. The significant others could be contacted intermittently and asked to record on a given day the exact number and types of beverages consumed by the subject in their presence. However, unless the subject and significant other spent all their waking hours together, only a portion of the day's record could be compared. Another way would be to distribute single serving packages of in- stant coffee, labeled with a D (decaffeinated) or C (caffeinated) and to require the subjects to return their dated used packages after each treat- ment phase.

The present study should be viewed as a first effort in the development of an effective be- havioral program for treating caffeinism. We hope that it calls attention to the serious and widely experienced problems caused by the ex- cessive consumption of coffee. It appears that many individuals unknowingly suffer from caf- feinism and that, in general, professionals in health-related fields are unaware of the existence of a "caffeine syndrome." As mentioned earlier, individuals suffering from caffeinism are some- times misdiagnosed as anxiety neurotics because

of the similarity of the symptoms (Greden, 1974). Accordingly, useful diagnostic informa- tion could be obtained if therapists would rou- tinely question their neurotic clients concerning the extent of their coffee drinking and caffeine consumption.

REFERENCES

Colton, T., Gosselin, R. E., and Smith, R. P. The tolerance of coffee drinkers to caffeine. Clinical Pharmacology and Therapeutics, 1968, 9 (1), 31-39.

Dreisbach, R. H. and Pfeiffer, C. Caffeine-with- drawal headache. The Journal of Laboratory and Clinical Medicine, 1943, 28, 1212-1219.

Eddy, N. B. and Downs, A. W. Tolerance and cross- tolerance in the human subject to the diuretic effect of caffeine, theobromine and theophylline. Journal of Pharmacology and Experimental Ther- apy, 1928, 33, 167-174.

Foxx, R. M. and Brown, R. A. A nicotine fading and nicotine monitoring program to reduce ciga- rette smoking. Journal of Applied Behavior An- alysis, 1979, 12, 111-125.

Goldstein, A. and Kaizer, S. Psychotropic effects of caffeine in man. III. A questionnaire survey of coffee drinking and its effects in a group of housewives. Clinical Pharmacology and Thera- peutics, 1969, 10 (4), 477-488.

Goldstein, A., Kaizer, S., and Whitby, 0. Psycho- tropic effects of caffeine in man. IV. Quantitative and qualitative differences associated with habitu- ation to coffee. Clinical Pharmacology and Thera- peutics, 1969, 10 (4), 489-497.

Greden, J. F. Anxiety or caffeinism: A diagnostic dilemma. American Journal of Psychiatry, 1974, 131 (10), 1089-1092.

Hartmann, D. P. and Hall, R. V. The changing criterion design. Journal of Applied Behavior Analysis, 1976, 9, 527-532.

Lichtenstein, E. and Danaher, B. G. Modification of smoking behavior: A critical analysis of theory, research and practice. In M. Hersen, R. M. Eisler, and P. M. Miller (Eds), Progress in behavior modification (Vol. III). New York: Academic Press, 1976.

McManamy, M. C. and Schube, P. G. Caffeine in- toxication: Report of a case the symptoms of which amounted to a psychosis. New England Journal of Medicine, 1936, 215 (14), 616-620.

Powers, H. The syndrome of coffee. Medical Jour- nal and Record, 1925, 124, 745-747.

Punke, H. H. Caffeine in America's food and drug habits. The Journal of School Health, 1974, 44 (10), 551-562.

344 R. M. FOXX and ANDREA RUBINOFF

Reimann, H. A. Caffeinism: A cause of long-con- tinued, low-grade fever. Journal of the American Medical Association, 1967, 202 (12), 1105-1106.

Ray, 0. S. Drugs, society, and human behavior (2nd ed.). St. Louis: C. V. Mosby Company, 1974.

Truitt, E. B., Jr. The xanthines. In J. R. Dipalma (Ed), Drill's Pharmacology in Medicine (4th ed.). New York: McGraw-Hill, 1971.

Winsor, A. L. and Strongin, E. I. A study of the development of tolerance for caffeinated bev- erages. Journal of Experimental Psychology, 1933, 16, 725-744.

Received 18 September 1978. (Final Acceptance 16 March 1979.)

View publication stats