Anticipatory nausea and vomiting in cancer patients undergoing chemotherapy treatment: Prevalence, etiology, and behavioral interventions

Clinical Psychology Review, Vol. 8, pp. 517-556, Printed in the USA. All rights reserved.

1988 Copyright

0272.7358/88 $3.00 + .OO 0 1988 Pergamon Press plc

ANTICIPATORY NAUSEA AND VOMITING IN CANCER PATIENTS UNDERGOING CHEMOTHERAPY TREATMENT: PREVALENCE, ETIOLOGY, AND BEHAVIORAL INTERVENTIONS Gary R. Morrow Patricia L. Dobkin University

of Rochester

School of Medicine

and Dentistry

ABSTRACT. Nausea and vomiting are comrrwn side effects thut occur after chemotherapeutic drugs are given to treat cancer. Recent attention has been focused on the occurrence of nausea and vomiting in anticipation of a treatment. Prevalence data from a variety of studies show that approximately 25 70 of patients receiving chemotherapy treatments for calzcer develop anticipatory nausea and vomiting (ANY by the time of their fourth treatment. A review of studies examining models and variables associated with potential etiology indicates that no single demographic, clinical, or psychologic characteristic appears to be as related to the development of anticipatory side effects as are several characteristics in concert. Spectfically, patients under 50 years of age who are treated with potentially emetic chemotherapy and who experience distressing nausea and vomiting after treatment appear to be at risk for the development of anticipatory nausea and vomiting. The development of anticipatory side effects a@ ears to follow a classically conditioned model; no compelling data contradict a view that anticipatory side effects are learned. A review of both case and controlled studies on the biobehauioral treatment of ANV indicates that: (a) while studies on Hypnosis have been methodologically weak, they provide support for continued controlled inuestigations in children; (b) P rog ressive Relaxation Training appears effective in controlling posttreatment nausea and vomiting; and (c) Systematic Desensitization appears effectiue in controlling both anticipatory and posttreatment nausea and vomiting. The review concludes with a brief consideration of several methodologic and practical issues thut point to promising directions for future research in the area.

This research was supported by a Research Career Development Award K04-CA-01038, by research grants CA11198 and CA26832 from the National Cancer Institute, and by research grants PDT 2 17 and CH 35 1A from the American Cancer Society. Request for reprints should be sent to Gary R. Morrow Ph.D., Associate Professor of Oncology in Psychiatry (Psychology), Department of Psychiatry, University of Rochester School of Medicine and Dentistry, 300 Crittenden Blvd., Rochester, NY 14642. 517

G. R. Morrow and P L. Dobkin

518

The nausea

most common and vomiting.

side effects of chemotherapy If not adequately controlled,

drug treatment for cancer are nausea and vomiting can lead

to further complications, such as anorexia and metabolite imbalance, along contributing to a general deterioration of the cancer patient’s psychological physical condition. Cancer patients may develop as much apprehension dread about treatment cancer patients treated

and its side effects with chemotherapy

with and and

as they do about their disease. Many drugs request dosage reductions or

even terminate potentially curative treatment regimens prematurely due to poorly controlled nausea and vomiting (Holland, 1977; Hoagland, Morrow, Bennett, & Carnrike, 1983). In addition to becoming nauseous and/or vomiting following chemotherapy treatment, effects prior to a treatment

a number of patients session (Morrow, 1981,

experience 1982,

these

aversive

side

1986a).

Increasing experimental attention is being paid to this phenomenon referred to as anticipatory nausea and vomiting (ANV), for it is a side effect that appears to link psychological, neurological, and physiological systems.’ In addition, anticipatory side effects may provide an unusual opportunity to study the natural occurrence of what appears to be a form of aversive learning in humans (Burish & Carey, 1986). Here we examine and critique studies on the prevalence, and treatment of anticipatory nausea and vomiting, with a view toward ing past research to identify further directions for exploration.

PREVALENCE Since

1979

at least

OF ANTICIPATORY

28 studies

NAUSEA

(summarized

etiology, organiz-

AND VOMITING

in Table

1) have reported

the preva-

lence of anticipatory nausea and vomiting in adult and pediatric cancer chemotherapy patients. As shown at the bottom of the table, a range of prevalence rates of anticipatory nausea and vomiting have been reported in the different studies. On the lower end of estimates, rates of anticipatory side effects of 18% of 71

‘To appreciate response these

side

muscles tion

how psychological

of nausea effects.

vomiting structure.

is thought

Hawken, unique

of being

area

poisoning,

toxicity,

input

tracts

from

associated substrate

and nausea

are cortical areas with

which

nausea areas

in the

memory makes

sion of nausea/vomiting.

possible

is an area

near

from

due

spinal

system

associated

to the vomiting

the involvement

ventricle

takes

the

center.

of psychologic

expression This

(Borison,

that is in the It is thought such

as food

place.

The

final

There

are

fiber

of emotion

provides

phenomenon

consystem.

and vomiting

postrema

challenges,

drugs

chemo-

the vestibular

and the cerebrum. with

during

fluid and blood.

to chemical

the

to this

is an emerging

of nausea

the fourth

called

inputs

occur

is the area

to chemotherapy

the limbic

system

connect

cerebral

vomiting

also the sensa-

major

there

and expression input

with both and

three

of

respiratory

of the medulla

first is from

A second

attributable

limbic that

This

and probably

1983),

The

physiology

of the

and vomiting

(Lazlo,

the physiologic the

coordination

at least

nausea

influence review

formation

are

are involved.

1975).

in contact

in which

There

in the development 1975).

the

reticular

by which

& Brand,

& Sirett,

for

characterized

inputs

to be involved (Reason

to be that large

major

may

to briefly

in the act of vomiting 1953).

the mechanisms

effects

be helpful

pathway

involved

are not totally

Hubbard, position

common

& Wang,

that at least three

due to motion

it may

of the dorsolateral

(Borison

treatment

sensus This

structures

Although

therapy

final

is an area

center

and biobehavioral

vomiting,

The

and other

of nausea

and

and

a neurologic in the expres-

519

Anticipatory Nausea and Vomiting

TABLE 1. Prevalence of Anticipatory Side Effects in Cancer Chemotherapy Patients Anticipatory Number studied

Authors

Anticipatory nausea

Andrykowski

(1987)

78

33%

Andrykowski

et al. (1985)

71

37%

Andrykowski

et al. (1987)

77

57%

Cella

et al. (1984)

Cohen

(1982)

Anticipatory vomiting

60

63%

-

149

42%

27%

Dobkin

et al. (1985)

125

32%

12%

Dolgan

et al. (1986)

80

20%

Fdez-Arguelles Fetting

et al. (1985)

et al. (1983)

Jacobsen

et al. (1985)

72

29% -

123

14%

27

Ingle

et al. (1984)

60

Love

et al. (1983)

126

38%

406

9%

225

24% -

736*

26%

18

44%

8% -

(in press)

Morrow

et al. (1982)

Morrow

& Dobkin

Nesse

(1985)

et al. (1980) (1982)

71

24% -

Nicholas

(1983)

50

42%

50

40%

14%

22% -

9% -

Olafsdottir

61

et al. (1986)

Palmer

et al. (1980)

24

Schultz

(1980)

68

Scogna

& Smalley

van Komen

& Redd

Weddington Weddington

(1979) (1985)

33%

11%

(1982)

17

et al. (1984)

50

53% -

12% 33%

Wilcox

et al. (1982)

52

-

Wilson

(1986)

66

20%

2,452

Median Range *These

80 17-736

patients

were part of a consecutive

31% 38% 21% -

18% -

31%

-

41 100

Total

31%

-

Nicholas

et al. (1986)

-

25%

-

Nerenz

series;

18

10 12%

therefore,

37% 38% -

-

8%

33% 14-63%

and/or

vomiting

-

57% -

Morrow

nausea

9 18-38%

9-27% the patients

reported

in Morrow

(1982) and Morrow (in press) are part of the n = 736 in Morrow & Dobkin (1985).

patients were reported by Nicholas (1982). A rate of over 50% was reported by Cella, Pratt, and Holland (1984) for 60 patients treated for Hodgkin’s Disease. Several factors have been proposed to account for variation in prevalence rates (see discussions by Andrykowski, 1986; Burish & Carey, 1986; Duigon, 1986; Morrow, 1982, 1984b; Nicholas & Hollandsworth, 1986). These reviews generally point to one or more of the following explanations of variation in prevalence rates of anticipatory nausea: (a) Some researchers have studied anticipatory nausea and vomiting symptoms independently of each other, whereas other researchers combined them and viewed the symptoms as one phenomenon. Often it is not

G. R. Morrow and P L. Dobkin

evident which phenomenon was being reported. effects may occur during chemotherapy treatment.

(b) Nausea and vomiting side While this may represent antici-

patory phenomenon, it is more likely a physiological response since some chemotherapy drugs may be reacted to during the time that they are actively given. (c) Prevalence rates may be influenced by the type of chemotherapy drugs administered to cancer patients, since side effects vary across different treatment regimens.2 (d) The time frame in which ANV symptoms were studied has differed across (1982)

studies. assessed

For example, patients prior

Morrow, Arseneau, Asbury, Bennett, and Boros to their fourth chemotherapy cycle whereas Wilcox,

Fetting, Nettescheim, and Abeloff (1982) did so prior to the tenth chemotherapy cycle. The data currently available support the view that there is a positive correlation between ANV and the number of chemotherapy treatments (see Figure 2). (e) A portion of the variation in prevalence rates may be due to measurement artifacts. A variety of self-report measures have been used to assess ANV across

studies

(Morrow, 1984b). F or example, some studies involved interviewing by asking retrospective questions (e.g., Dobkin, Zeichner, & Dickson1985), whereas others used patient-completed logs during and following

patients Parnell,

treatment (Burish & Carey, 1984). As a means of approximating how much due

to the

factors

mentioned

above,

variation

data

from

in prevalence

a series

rates

of 1,480

may be

consecutive

chemotherapy patients reported by Morrow (1986b) can be compared with the rates cited in Table 1 summarizing data collected on 2,452 patients. By aggregating the data from Table 1, it appears that the overall prevalence rate of anticipatory nausea is 33 %. From Morrow’s data, collected on 1,480 patients from a single cancer center, assessed Emesis), at a standard that the overall factors

mentioned

with the same scale (Morrow Assessment of Nausea and time (prior to the fourth chemotherapy cycle), it appears

prevalence

rate is 23 %

Thus,

10 % of the variance

may be due to

above.

Summary and Suggestions for Future Research Directions Comparing

the results

for anticipatory

nausea

shows

that

approximately

one in

four patients was found to have experienced anticipatory nausea based on a standard criteria in a series of consecutive patients. A 10% higher figure of approximately one in three patients found to experience anticipatory nausea was found for a sample of chemotherapy patients aggregated from different studies, using potentially disparate definitions. from a variety of geographic locations, These

rates are comparable

A different assessment advantage in this area.

from a clinical

as well as practical

standpoint.

framework for measures of nausea would be of potential For example, if selected physiologic parameters were

Chemotherapy treatment is typically given repeatedly over a number of weeks in episodes that are called treatment cycles. For example, it is common to be treated with a chemotherapy drug once every three weeks for a total of perhaps a dozen cycles. During each of these treatments, patients spend several hours in the treatment clinic. Chemotherapy drugs are given through a small needle typically placed in a vein in the back of the hand. Some drug regimens (especially those that are more toxic) are administered slowly over a period of more than one hour. Thus, an individual treatment may occupy a significant portion of the day.

521

Anticipatory Nausea and Wmiting

shown to be replicable concomitants of patient reported used to systematically examine variation in patient report

nausea, they could be by providing an objec-

tive frame of reference (Nicholas & Hollandsworth, 1986; Burish & Carey, 1986; Andrykowski, 1986; Morrow, in press). Some preliminary work has been reported in the motion sickness literature, where physiological measures such as skin pallor have been shown related to motion-induced nausea (Oman, Lichtenberg, & Mooney, 1984). If motion-induced nausea is analogous to chemotherapy druginduced nausea, then this methodology may be useful. Morrow has modified and relined assessment instruments for the measurement of several physiologic parameters associated with nausea (pallor, skin temperature, blood volume pulse, heart rate, physiologic

and body motion). Preliminary data (Morrow, in press) have shown pattern that matches patient report of drug-induced nausea. Over

a a

series of repeated chemotherapy cycles, eight patients have shown an increase in pallor and decrease in skin temperature, blood volume pulse, and body motion that corresponds to self-reported nausea. During vomiting episodes, pallor decreases and skin there is a gradual If these findings

temperature transiently decrease of up to 4.6”C are supported

increases during over ten vomiting

by larger

scale studies,

expulsion, episodes.

although

this framework

of physio-

logic assessment may be applicable to studies not only on the prevalence of anticipatory side effects but also on determinants and the efficacy of treatment. Such an assessment technology could prove useful in studies examining aspects of patients’

self-report

of both anticipatory

ETIOLOGY

and posttreatment

OF ANTICIPATORY

NAUSEA

nausea.

AND VOMITING

A wide variety of correlates of anticipatory nausea and vomiting have been explored. In this section, studies exploring associations among demographic, clinical and/or psychological variables and ANV are presented and discussed. This is followed

by a critique

ANV develops. The data support a model Univariate

Correlates

A number anticipatory

of the four basic

models

that have been

proposed

of Anticipatory

Nausea and Vomiting

of characteristics have been examined for their possible side effects. They may be conveniently grouped under

of demographic,

for how

section ends with a detailed examination of how available of ANV development based on learning principles.

clinical,

and psychological

characteristics

association to the categories

studied.

Demographic Variables. Table 2 summarizes demographic variables examined for a correlation with anticipatory side effects. Several investigators (Cohen, 1982; Fetting et al., 1983; Ingle, Burish, & Wallston, 1984; Morrow, 1982; van Komen & Redd, 50 years

1985) h ave found an association between age (that is, being younger than old) and experiencing ANV. Only two of the eleven studies investigating

gender (Fetting et al., 1983; Wilson, Rahdert, Black, Taylor, & Holloway, 1986) found it to be associated with ANV symptomatology. None of six studies including data on ethnicity (Fetting et al., 1983; Ingle et al., 1984; Morrow, 1982; Weddington, Miller, & Sweet, 1982; van Komen & Redd, 1985) have reported a relationship between race and anticipatory side effects. Education level and socioeconomic and marital status do not appear to be critical factors for the development of ANV symptoms (Fetting et al., 1983; Morrow, 1982; Weddington, Mill-

522

G. K. Morrow and P L. Dobkin

TABLE

2. Demographic

Variables Associated

with Anticipatory

Nausea and Vomiting Authors

n

Andrykowski et al. (1985) Cohen (1982) Dolgan et al. (1985) Fetting et al. (1983) Ingle et al. (1984) Love et al. (1982) Morrow (1982) Nesse et al. (1980) Schultz (1980) van Komen & Redd (1985) Weddington et al. (1982) Weddington et al. (1984) Wilson (1986) Association

Gender

Age

Race

x

-

x

younger

-

71 149 80 123 60 126 225 18 68 100 17 50 66

found:

X

X

female x -

X

younger younger younger younger x

x X

X

x x -

X

X

-

female

younger x x younger

2/11 (18%)

7/13 (54%)

O/6 (0%)

X X X

X X

N&:x indicates that the factor was investigated and was not found to correlate with ANV. -indicates

er, & Sweet,

that the factor was not investigated.

1982).

Overall,

with the marked

little association between patient ment of anticipatory side effects.

exception

demographic It is possible,

of age, there

characteristics however, that

tients are more likely to receive highly and of itself, is not a crucial factor.

emetic

chemotherapy

Clinical

results

of clinical

Variables.

Table

3 summarizes

appears

to be

and the developsince younger pa-

regimens

variables

that age, in

examined

studies for their association to ANV. Nesse, Carli, Curtis, & Kleinman reported that cancer patients with anticipatory nausea, when compared

in 14 (1980) to those

cancer patients without anticipatory nausea, had been treated for a significantly longer period of time (M =9.3 months versus M =4.2 months, respectively). Morrow (1982; 1984~) noted that, compared to patients who did not report ANV symptoms, ANV palients experienced more post-chemotherapy nausea and vomiting, of longer duration and greater severity. All three studies that examined a potential relationship between how much nausea and vomiting a chemotherapy drug typically caused after treatment (its emetic potential) and ANV reported a significant relationship. Seven out of seven studies, and seven out of nine studies reported a relationship between post-chemotherapy nausea and vomiting, respectively, and ANV symptoms. The neural pathway between the vomiting center (an area in the dorsolateral reticular formation of the medulla) and the vestibular system has been implicated in motion induced nausea and vomiting as well as vomiting from poisons given to animals (Money & Cheung, 1983). Th us, a possible relationship between motion sickness susceptibility and chemotherapy nausea/vomiting has been studied. Using a case control methodology, Morrow has shown that patients who report a susceptibility to motion sickness (compared to patients without a susceptibility) had (a) signifi-

Anticipatory Nausea and Emiting

523

cantly more side effects from chemotherapy drugs (Morrow, 1985); (b) significantly more posttreatment nausea and vomiting (Morrow, 1984f); (c) significantly more

anticipatory

nausea

and vomiting

(Morrow,

1984e).

~yc~ological kriabks. Consistent with a view that emotions and cognitions may contribute to the development of ANV (Ahles et al., 1984; Schultz, 1980; van Komen & Redd, 1985), anxiety, depression, hostility, and coping styles in cancer chemotherapy patients have been studied. As shown in Table 4, 10 out of ii investigations found that state anxiety levels were significantly elevated in patients with ANV compared to patients without ANV. Two out of three investigations found trait anxiety related to ANV. Love, Nerenz, & Leventhal(1982), using a prospective research design, studied 126 cancer patients being treated with chemotherapy for breast cancer (72 =94) and malignant lymphoma (n = 32). Patients were interviewed repeatedly during their initial six months of chemotherapy. Thirty-eight percent of the patients

TABLE

3. Clinical

Variables Associated Nausea and Vomiting

with Anticipatory

Post Rx Nausea Authors

n

Andrykowski et al. (1985) Andrykowski et al. (1987) Cohen (1982) Dobkin et al. (1985) Dolgan et al. (1985) Fetting et al. (1983) Ingle et al. (1984) Morrow (1982) Morrow (1984~) Nesse et al. (1980) Nerenz et al. (1982) Nicholas (1983) van Komen & Redd (1985) Weddin~ton et al. (1982) Weddington et al. (1984) Wilcox et al. (1982) Association

No. of Rx

Emetic Potential

a *

-

X

-

-

-

71 77 149 125 80 123 58 225 176 18 61 71

D

S

* *

-

* * * * 8

*

-

b* * * *

-

-

X

*

-

*

-

*

-

*

-

-

X

17 50 52

S

*

X

-

*

-

-

100

found:

-

Post Rx Vomiting

* *

-

7110 (70%)

313 ~100%)

* indicates

that the factor was found to correlate regimen;

“indicates

length of infusion

bindicates

frequency

(rather

S=severity; than duration).

significantly

D=duration

* * * X *

* -

*

bx -

5/7 (71%)

7/7 (100%)

719 (78%)

that the factor was not studied

Rx=chemotherapy

X

bx -

Note:x indicates that the factor was studied and was not found to correlate -indicates

b* b*

* -

* 9J9 (100%)

-

D

with ANV

with ANV

524

G. R. Morrow andP L. Dobkin

TABLE 4. Psychological Variables Associated with Anticipatory Nausea and Vomiting Authors

n

Anxiety

Ahles et al. (1984) Altmaier et al. (1982) Andrykowski et al. (1985) Andrykowski et al. (1987) Cohen (1982)

9 9 71 77 31

* (state) * (state) * (state) + +(state) * (state) x (trait) * (state) * (state) * (state) * (state/trait) * (trait) +

Houts et al. (1982) Ingle et al. (1984) Nerenz et al. (1986) Shultz (1980) van Komen & Redd (1985) Wilson (1986)

90 58 18 68 100 66

Summary Total

509

Depression

indicates

-

-

X

0

-

-

-

-

-

-

* -

* ** -

X

lo/11 (state) 2/3 (trait)

II3

214

l/3

patients

that the factor was not investigated

* indicates

that the measure

was significantly

o indicates

that the measure

was depressed

** indicates

that “more attempts

+ indicates

that conflicting

+ + indicates

Coping

*

Note:x indicates that there was no difference between ANV and non-ANV -

Hostility

that anxiety

elevated in ANV patients in ANV patients

to cope” were made

results were found between

was significantly

measures

elevated only in late-onset

of ANV patients

developed ANV, with patients who experienced anxiety during injections significantly more likely to develop anticipatory nausea than non-anxious patients. Interestingly, the association between anxiety and anticipatory nausea was not statistically significant during the first two chemotherapy cycles but became significant by the sixth cycle. Andrykowski et al. (1987) also reported that the relationship between anxiety and ANV should be qualified according to the treatment time frame. To determine if there were different patterns of infusionrelated state anxiety and posttreatment nausea prior to AN onset, AN patients were divided into early (i.e., prior to chemotherapy cycle number 7) and late (i.e., following chemotherapy cycle number 7) onset groups. According to this distinction, anxiety appeared to contribute to the development of AN only for patients who were in the late onset group. A simple, direct relationship between anxiety and ANV development has not been found. Depression was found to be significantly elevated in ANV patients in two out of four studies, whereas hostility and coping styles were not found to be consistently different in ANV patients. Altmaier, Ross, & Moore (1982) reported that patients with ANV exhibited a coping style which was “inhibitive rather than facilitative in of “attempts to nature.” In contrast, Ingle et al. (1984) f ound a greater number cope with compared

chemotherapy and to patients without

a higher level of hostility in patients with ANV ANV. Hostility and coping styles are, however,

Anticipatory Nausea and limiting

difficult

hypothetical

these

divergent

ences

in responses.

constructs

findings

reflect

to operationalize; measurement

525

it is, therefore,

variance

rather

than

possible actual

that differ-

Summary. Methodological differences in research designs used and assessment techniques employed render firm comparisons among these investigations difficult. In addition, the retrospective nature of studies examining the possible association between single variables and the development of anticipatory limit the conclusions that can be reached from these data. Multivariate Several

Correlates

recent

of Anticipatory

studies

have

built

side effects

Side Effects on some

of the

earlier

promising

univariate

findings and used multivariate procedures to examine potential joint or interactive relationships among variables that might be associated with the development of anticipatory side effects. Morrow (1982) reported results from a two-group discriminant analysis on 225 cancer patients, Overall, an 80% accurate classification was achieved with 42 % of patients experiencing ANV and 9 1% of patients not experiencing ANV, correctly classified based on a combination of age, severity, and duration of postchemotherapy nausea and vomiting. Cohen (1982) reported that the frequency of vomiting during/after chemotherapy sessions and age (younger than 50 years-old) accounted for 32% of the variance in the occurrence of anticipatory nausea. Similar patterns were observed in analyses involving anticipatory vomiting. In a follow-up study (Cohen, 1982) involving 14 patients with ANV, five variablesnausea, anxiety, noxious sensations, frequency of post-chemotherapy and age-accounted for 88 % of the variance for anticipatory nausea. iables - nausea,

frequency

of

vomiting

during/after

chemotherapy,

vomiting, Four varanxiety

and age-accounted for 65 % of the variance in anticipatory vomiting. Since complete data were obtained from only 14 patients, these results require cautious interpretation. In a study of 58 adult cancer patients, Ingle et al. (1984) used a three-group (“conditioned, may be conditioned, and not conditioned”) discriminant analysis involving seven variables in order to determine the best set of “predictors” for patients with ANV. These authors reported an overall assignment rate of 71% the three groups based on: age (younger than 50 years old), postchemotherapy nausea and vomiting, anxiety, and “coping effect I

to

Dolgan, Katz, McGinty, & Seigel (1985), in a study of 80 pediatric (mean age of nine years-old) cancer patients, identified five variables which accounted for 23 % of the variance in group membership; these were: postchemotherapy nausea and vomiting, emetic potential of chemotherapy regimen, administration of the drug cyclophosphamide, time since diagnosis, and parental anxiety. van Komen and Redd (1985), in an investigation of 59 cancer patients, administered the Millon Behavioral Health Inventory (Millon, Green, & Meagher, 1979) and the Spielberger State-Trait Anxiety Inventory (Spielberger, Gorsuch, and Lushene, 1968) in order to examine potential personality factors hypothesized to be associated with ANV symptomatology. The most important discriminator variables and gastrointestinal

found were: susceptibility.

social Trait

alienation, future despair (depression), anxiety was also shown to be higher in

526

G. R. Morrow and P L. Dobkin

patients with ANV compared to patients without ANV. Interestingly, patients who received chemotherapy in a group setting were more likely to experience ANV compared to patients who were treated on an individual basis. Andrykowski, Redd, and Hatfield (1985) studied 26 patients who displayed anticipatory side effects and 45 who did not. While seven variables entered into a hierarchical regression accounted for about half (47%) of the variance in group membership, posttreatment nausea alone accounted for about one quarter (24%). State anxiety and length of time it took to give the chemotherapy drug were the only two other variables giving a significant increment in explained variance. A recent study (Morrow, 1984c) provided support for a view that clinically relevant characteristics are associated with anticipatory side effects. One hundred seventy-six consecutive ambulatory patients with histologically confirmed cancer who were being treated at three geographically separate hospitals at the University of Rochester Cancer Center were studied at the time of their fourth chemotherapy treatment. Patients found to experience anticipatory nausea and vomiting were significantly more likely to have four or more of the following characteristics compared to patients who did not report anticipatory side effects: (a) age (less than 50 years); (b) the experience of nausea and/or vomiting after their last chemotherapy treatment; (c) describe nausea after the last treatment as “moderate, severe, or intolerable”; (d) describe vomiting after the last treatment as “moderate, severe, or intolerable”; (e) report the side effect “warm or hot all over” after their last treatment; (f) susceptibility to motion sickness; (g) experience “sweating after their last treatment”; and (h) experience “generalized weakness after their last chemotherapy treatment.” These results were recently replicated in a prospective study (Morrow, submitted for publication). Five hundred thirty consecutive cancer patients were asked the previously reported eight questions following their first chemotherapy treatment. The outcome measure was whether or not they had developed anticipatory nausea/vomiting by the time of their fourth chemotherapy treatment. Those patients who had four or more of the eight characteristics were predicted to develop anticipatory side effects. Those with three or less were predicted not to. A significant association was found between these characteristics and subsequent development of anticipatory side effects for 345 patients entered into the study and followed through their fourth chemotherapy treatment. This finding provides a reasonably robust test of the importance of the eight characteristics in the development of the anticipatory side effects. Several of these factors were implicated in univariate investigations. The fact that there are consistent findings regarding patient and treatment characteristics that contribute to the development of anticipatory side effects, irrespective of patient sample and definition used, point to the stability of these factors. and Suggestioffs for Future Reseafc~ ~i~ctio~s. De~nitional differences may account for some of the variation in correlates of ANV found. For example, most studies have set up criterion groups of those who have developed the symptoms and those who have not, and then examined potential correlates based on those criterion. At least one study has used a three-group criterion of those patients who were “conditioned,” “may be conditioned,’ and “not conditioned” based on the MANE (Tngle, Burish, & Wallston, 1984). A multiple regression model yielded no clear Sumnary

527

Anticipatory Nausea and Vomiting

correlates those

of the intermediate,

subjects

“may be conditioned,”

who fell in a “conditioned”

group

group. were

The

study found

significantly

younger

that and

experienced greater severity of posttreatment nausea and vomiting and were generally more anxious than patients who did not. These findings are consistent with previous investigations outlined in Table 2 through 4. Thus, while a variety of research methodologies involving a variety of samples from populations that may have varied on a number of clinical characteristics, a reasonable consistency of results has been shown. While they have differed in sample, assessment, and analysis, the more recent multivariate studies support an emerging consensus on which factors contribute to the development of anticipatory side effects. At least three summary contentions are supported by the data. First, ANV is multiply determined in that no single demographic, clinical, or psychological characteristic appears to be as related to ANV development as are several characteristics in concert. Second, a subset (usually no more than three or four) of any larger set of characteristics accounts for as much variance in group membership (has ANV versus does not have ANV) as the larger set. And third, patients under 50 years of age who are treated with potentially emetic chemotherapy and who experience distressing pear to be at risk for the development There has been a general increase

postchemotherapy of an anticipatory in the methodologic

nausea and vomiting apresponse. sophistication of studies

investigating ANV correlates; however, improvements can still be made. Early studies were basically correlational in nature. More recent studies have focused on multivariate methods; regression models have been used primarily. While an improvement over simple univariate studies, multivariate methodology introduces difficulties and potential pitfalls of its own. There are potential problems of validity in analyzing large numbers of variables with a small data set. The stability of multivariate statistical procedures such as regression-based statistical models depends on the number of observations available for each of the variables being examined (or, rather, parameters being estimated) (Cohen & Cohen, 1975). A generally accepted lower bound is on the order of ten subjects for each variable entered in a regression equation (Chatterjee & Price, 1977; Lewis-Beck, 1980). Few previous studies have met even this minimal standard. Several suggestions for improvements in studies designed to explore etiology may be made. The first of these is that little is likely to be learned from one-shot assessment studies. Longitudinal designs assessing the development of the side effects over time are preferred. While methodologically more sophisticated, such longitudinal studies unfortunately involve practical difficulties in attrition and compliance. For example, attrition rates on the order of 25-50% may be expected for studies that follow patients even through four chemotherapy cycles (Morrow, 1984d). Part of the attrition is due to medical complications. A sizable percentage of patients treated on chemotherapy also have their drug treatment changed or modified prior to their fourth or fifth chemotherapy cycle. Such changes in pharmacologic agents may have a pronounced etiology of side effect development.

effect

on studies

examining

potential

Models of Etiology Theories of how anticipatory nausea/vomiting develops take either a physiologic or psychologic view. The physiologic view is that anticipatory symptoms may be

528

G. R. Morrow and P L. Dobkin

produced by brain metastasis or local cancer involvement of the gastrointestinal tract (Chang, 1981). This potential explanation is not consistent with at least two findings. The first is that metastatic spread of cancer to the brain or the gastrointestinal tract is significantly less evident than the greater than 25% prevalence of anticipatory collaborating tatic spread row, 1981; At least

side effects. Second, in a series of randomized clinical trials, we (and oncologists) carefully screened for any clinical evidence of metasand found no association between metastatic disease and ANV (Mor1982; 1984c; four different

1984d). psychologic

how psychological processes range from a psychodynamic classical

viewpoints

have

been

proposed

to explain

may cause anticipatory side effects. These hypotheses conceptualization to a learning paradigm involving

conditioning.

Psychodynamic Model. Chang (1981) has proposed a psychodynamic anticipatory side effects. According to this view, nausea and vomiting always direct side effects of chemotherapy, but rather may be surfacing tions of underlying psychological readjustment problems associated threatening

illness” (p. 707).

Chang

further

suggests

origin of “are not manifestawith life-

that these side effects

may be

caused by “psychological mechanisms, including anger, anxiety, and frustration” (p. 707). This theory has received no direct empirical support (Redd, Burish, & Andrykowski, 1985; Morrow & Dobkin, in press), although the data regarding elevated levels of anxiety theory, albeit indirectly. Coping Model.

If having

in patients

cancer

with ANV

is conceptualized

may be interpreted

as being

in a health

to support

crisis,

the

then it

follows that cancer patients are subjected to multiple stressors that call for effective coping strategies. Altmaier et al. (1982) suggested that patients who develop ANV may be deficient in their ability to adjust to, or cope with, the physical and emotional challenges associated with their chemotherapy treatments. The fact that patients

with ANV

tend to be anxious

about

treatment

supports

this hypothe-

sis. High levels of anxiety may be seen as a failure to cope effectively. Altmaier et al. (1982) reported that patients with ANV demonstrated significantly less desire and ability to cope than patients without ANV. Although this their method of assessing coping has never appears to support their hypothesis, been validated or replicated. Ingle et al. (1984) a 1so investigated the relationship between coping styles and ANV symptomology. They found a “high correlation” between the ANV score and “coping effort,” indicating that the “more conditioned” a patient was, the more ways s/he had tried to cope with having cancer. These apparently that the patients

contradictory results may be reconciled if one accepts the notion in Ingle et al.‘s study made more attempts to cope because their been inadequate/deficient). Yet, since Ingle et al.‘s had failed (i.e., these findings are preliminary, at best. was also problematic,3

initial trials methodology These two studies

approached

the area

of coping

in a very broad

and general

31ngle et al. (1984) administered the Horowitz Coping Inventory (Horowitz & Wilner, 1980) as a means of assessing coping styles. Reportedly, the nurse investigator observed that many subjects experienced difficulty in understanding statements on the coping inventory. Thus, the validity of the instrument has yet to be established.

Anticipatory Nausea and Vomiting

529

manner. Perhaps by narrowing the focus and organizing one’s thoughts around a model of coping we could gain insight into its contribution to the development of ANV. Gil (1984), for example, proposed a model of coping which recognizes a three-way interaction between physical and psychological demands of the medical situation and the coping resources of the individual. These, in turn, are broken down into three channels of responses: overt behavioral, physiological, and cognitive. Thus, patients may differ on any of these factors and thereby respond quite differently. In the context of the cancer chemotherapy setting, the physical and psychological demands might include the administration of potentially toxic chemicals, symptoms of the disease itself, concern about prognosis, fear of injections, etc. As for coping resources, this constitutes an area replete with individual variation. Examples of coping resources are: overt behavioral control (e.g., seeking information), cognitive control (e.g., using a cognitive structuring), and physiological control (e.g., deep breathing

strategy such as retechniques). These

control responses may be used adaptively to reduce the individual’s arousal and thereby reduce the individual’s awareness or interpretation of the demands of the situation. The main point to be retained from Gil’s model is that coping is multidimensional

and

that

individuals

may

vary

on any one or combination

of the factors

involved. It is likely that little is known about how cancer patients cope with their disease and its treatment because the methodology used to investigate this phenomenon thus far has failed to address the complexity adequately. Anxiety Model. Several investigators have speculated that anxiety may be involved in the development of anticipatory side effects. Houts, Morrow, Lipton, Harvey, and Simmons (1984) h ave proposed four potential ways that anxiety might relate to anticipatory (b) pretreatment

side effects: (a) anticipatory anxiety causes anticipatory

nausea causes pretreatment anxiety; nausea; (c) anticipatory nausea and

pretreatment anxiety are both caused by posttreatment nausea; and (d) pretreatment anxiety facilitates the conditioning process of anticipatory side effects. There are some data that lend partial support to each of the four potential explanations. However, hypotheses,

in order experiments

to examine

separately

each

would have to be designed

potentially

competing

that would isolate

of the

a particular

portion of the chemotherapy treatment. This, unfortunately, is not possible in a clinical setting since chemotherapy is given in repeated cycles. Andrykowski et al. (1985) recently provided evidence that elevated levels of state anxiety may precede the initial occurrence of anticipatory symptoms. These authors caution, however, that the relationship between anxiety and anticipatory side effects may not be a strictly causal one. Their data also support a view that the anxiety may be heightened following a particular chemotherapy treatment and that, in turn, the increased anxiety may increase posttreatment nausea/vomiting which, in turn, may increase susceptibility to conditioning on the next chemotherapy cycle. This circular process may facilitate and promote a conditioning process rather than serve as a direct cause itself. Thus, the explanation that anxiety is involved in some form in the conditioning of side effects is the most plausible of the four. It is likely that some degree of anxiety facilitates the conditioning process by alerting or sensitizing the patient in much the same way that somebody who is mildly anxious may be quite prone to suddenly notice and become concerned over a

G. R. Morrow and El L. Dobkin

530

physical

sensation

a period

of time.

learned

such as irregular

heart

beats

that had probably

been present

for

Etiology Model

The development

of anticipatory

side effects

closely

follows the principles

of learn-

ing. While potential differences between them are often complex and controversial, there are basically two principal models of learning: operant and classical. An operant model suggests that behavior develops and continues because it is reinforced.

Such

a view of the development

of anticipatory

side effects

would state

that the patient received some reward for nausea and vomiting behavior. While it is certainly the case that nausea and vomiting side effects are attended to by treating staff and others, it stretches credibility that someone would continue these behaviors for the rather meager reinforcement being provided. The high prevalence explanation.

of ANV

casts

further

doubt

on

the

validity

of this

potential

A classical conditioning explanation of how ANV might develop is outlined below. The conditioning process proposed involves the elements in the top part of the model of Figure 1 which was modified from Burish and Carey (1984). An unconditioned unconditioned

response stimulus

(posttreatment nausea and vomiting) (the chemotherapy drugs administered)

which follows in the context

an of

potentially conditionable stimuli (sensations, thoughts, images of the clinic and/or nurse) will, after a number of repeated trials (chemotherapy treatments), give rise to a conditioned stimulus (such as the chemotherapy nurse) eliciting a conditioned response of anticipatory nausea and vomiting. A number of indirect lines of evidence

support

such an etiology.

The First Few Chemotherapy r-

-~~~~~~~

Condltloned Stimulus

-b

Treatmenta No Rerponae

(nuree) +

1 Uncondltloned Stimulus II, ~ (chemotherapy

drugs)

Uncondltloned Fbrponoe (nausea,voml ting)

After Several Chemotherapy Treatments

FIGURE

1. Conditioning Model of how Anticipatory Side Effects Develop.

531

Anticipatory Nausea and limiting

loo! 80

t 60 i

0

5

10 Treatment

FIGURE

Parameters

Cycle

2. Anticipatory Nausea and Vomiting (ANV) Number of Chemotherapy Treatments.

of Classical Conditioning

Which Support

15

Prevalence

the View that ANV

versus

is Learned:

Anticipatory nausea and vomiting develops only after a number of administrations of chemotherapeutic agents have been given (Morrow et al., 1982; Love et al., 1982). Clinical reports, along with clinical studies (Wilcox et al., 1982), support the notion that anticipatory nausea and vomiting is virtually never seen before a number of administrations of chemotherapeutic agents have been given. The percent of patients who develop anticipatory nausea and vomiting increases with the number of courses of chemotherapy given (Morrow, 1982; Morrow & Dobkin, in press; Olafsdottir, Sjoden, & Westling, 1986). Figure 2 combines data from several studies and suggests that the development of anticipatory symptoms is a function of the number of treatment cycles administered. This observation is consistent with the development of a learned response where the strength of that response increases with the number of conditioning trials given. In a clinical situation, each administration of chemotherapy agents would correspond to a conditioning trial. Course of Development.

Generalization. The learning phenomenon of stimulus generalization, where a response may be elicited by a stimulus similar to the original conditioned stimulus, fits the available clinical data. Patients’ anticipatory side effects may be elicited by an increasing constellation for stimuli and situations as treatment continues. It is not uncommon, for example, for patients to report nausea when they see the clinic nurse who administers their drugs, and then report, after a few that the sight of any clinic muse elicits nausea more chemotherapy treatments, (Redd & Andrykowski, 1982).

Stimulus

G. R. Morrow and II L. Dobkin

532

Correspondence tigations have

Between Unconditioned and Conditioned Responses. Several invesshown a relationship between the occurrence of posttreatment

nausea/vomiting and the likelihood of an individual patient developing anticipatory nausea and vomiting. There has not been a study that reported the presence of anticipatory nausea/vomiting in the absence of posttreatment nausea/vomiting. The fact that the anticipatory nausea and vomiting so closely resemble the posttreatment nausea and vomiting fits a traditional Pavlovian or classical conditioning model, in which the posttreatment response and the anticipatory nausea

nausea and vomiting is an unconditioned and vomiting is a conditioned response.

Intensity of Unconditioned Response. The intensity of an unconditioned response has been shown to affect the development of a conditioned response (Dragoin, 1971; Nachman & Ashe, 1973). When applied to ANV, this finding from classical conditioning would hypothesize that the severity of posttreatment nausea/vomiting (intensity of unconditioned response) would be related to the development of anticipatory nausea/vomiting (the conditioned response). All eight studies (summarized in Table 3) that have examined severity of posttreatment nausea/vomiting have found it significantly associated with the development of ANV, the greater the severity of posttreatment anticipatory nausea/vomiting. Higher-Order process er-order

Conditioning.

Aspects

nausea/vomiting,

of ANV

the higher

development

the incidence

are consistent

of

with the

of higher-order conditioning,

(also termed second-order) conditioning. Through highneutral stimuli can come to elicit a conditioned response by would be being paired with a conditioned stimulus (CS,). A more exact definition that a neutral stimulus, when paired with CS,, can come to elicit a CR even though it has never been paired directly with the UCS (Konorski, 1948; Rescorla, 1973, 1978, 1979). Typically, a higher-order conditioned stimulus (CS,) does not elicit as great a magnitude or frequency of CR’s as does the CS, (Rizley & Rescorla, 1972). This finding fits well with the data concerning cancer chemotherapy patients, as a patient approaches the treatment center, the intensity of ANV tends to increase (Nicholas, 1982; Olafsdottir et al., 1986). According to this process, the increased intensity of ANV would be the result of the patient coming CS’s as the distance into contact with successively strong (i.e., more conditioned) from the CS, decreases. Higher-order conditioning

would

proceed

through

this

association

of stimuli

with the CS, which, in some manner, predict the occurrence of this stimulus. Specifically, these stimuli could be an oncology nurse or doctor, clinic odors, of other stimuli that are sights or sounds of the treatment room, or a myriad predictive of chemotherapy. As these stimuli come to be associated with chemotherapy, other stimuli such as the hospital parking lot which predict their occurrence would begin to be conditioned. Thus, over time, an association among these stimuli is formed for which ANV is the overt behavioral manifestation. Dobkin et al. (1985) asked patients to indicate the location where anticipatory symptoms occurred; sixteen patients identified their home, three reported becoming ill while traveling to the clinic, five identified the clinic, and three reported other locations. In the interview, the cancer patients described the cues occurring just prior to the anticipatory response. Sixteen patients reported thinking about the chemotherapy and four reported visual cues (e.g., visualizing needle inser-

533

Anticipatory Nausea and Vomiting

tions), while four others reported olfactory cues (e.g., smell of clinic). None patients reported gustatory cues. Similar cues were reported by Olafsdottir (1986). Andryk owski (1985) also reported patients’ initial two chemotherapy infusions ing their initial report of ANV, were ANV. Thus, the cognitive conditioning Summary

and Suggestions

for Future

that tastes and during

of the et al.

and odors, both during the two infusions preced-

unrelated to subsequent development model appears to be a viable one.

Research Directions.

The

case

of

for a condi-

tioned response developed above provides circumstantial support for a learned etiology of anticipatory side effects. There are no carefully controlled studies that demonstrate conclusively that the anticipatory nausea and vomiting seen prior to chemotherapy treatment are unequivocally would, of necessity, involve the experimental,

a result of learning. Such a finding deliberate development of anticipa-

tory nausea and vomiting within a conditioning paradigm. While of scientific interests, such a study would not be ethically feasible. Clinical studies carried out “in the field” (e.g., oncology wards) may not permit the experimental precision afforded in carefully controlled laboratory settings. Nonetheless, while there are no objective scientific data from which to state conclusively that ANV are the result of learning, a number of converging lines of indirect evidence strongly support such a view. None of them singly may be viewed as either necessary or sufficient; however, viewing anticipatory

taken together, they provide reasonably strong support for nausea and vomiting as a learned phenomenon. To date,

there are no compelling data which contradict this viewpoint. While prevailing data fit the learning theory-based model presented above, at least two characteristics of ANV development do not fit as comfortably within a classical conditioning framework as they might with a different model of learning: the relatively long interval between the US (p resentation of chemotherapy and UR (posttreatment nausea/vomiting), and the intermittent nature

drugs) of the

CR. In classical

of sec-

conditioning,

the US-UR

interval

is typically

on the order

onds or possibly minutes. A common finding is that the shorter the interval, the more pronounced the conditioning. The US-UR interval in the chemotherapy treatment context is typically on the order of hours as opposed to seconds or minutes. There is some evidence (Garcia, Ervin, & Koellinger, 1966; Rusiniak, Palmerino, Rice, Forthman, & Garcia, 1982) that conditioning involving the gastrointestinal system (such as would be present in nausea and vomiting) can occur with CS-US intervals of up to 75 minutes and has been shown to occur in special circumstances with a delay up to eight or nine hours. The fact that learning can take place with this longer CS-US interval suggests that anticipatory side effect development may also be consistent with a learned taste-aversion model (Robertson & Garcia, 1985, 1986) as well as the cognitive model of classical conditioning examined by Rescorla (1978). The conditioned response of anticipatory nausea and vomiting present prior to each chemotherapy treatment. While no hard data clinical observation that sometimes the patient will show anticipatory not show the CR again for prior to one chemotherapy treatment, treatments, and then show the anticipatory side effect again. This is variance response

with (CR)

a classical conditioning is observed each time

model in that after CS-UCS the CS occurs while the UCS

may not be exist, it is a side effects one or more somewhat at pairings a is still being

534

G. R. Morrow and P L. Dobkin

paired with CS. One potential explanation for the intermittent nature of this CR may be that the CS might not be present prior to each chemotherapy treatment. For example, if the nurse has become part of the CS, the patient may not be treated patient

by the same chemotherapy may also travel a different

tion of sensations this possibility.

present.

There

nurse at each administration of the drug. The route to the clinic or have a different constellaare no firm data at present

to support

or refute

The development of ANV may provide a common set of observable phenomena in humans with which to discuss and contrast various learning theories. While detailed examination of these is beyond both the scope and intent of this review, an example might illustrate how ANV might be useful. The relatively long US-UR interval has been mentioned as a possible contrast between

a classical

conditioning

model

and learned

taste-aversion

explanation.

The learned taste-aversion model would further predict that the organism (patient) does not have to be aware of the UR (nausea) for conditioning to occur. The theory would thus predict that chemotherapy patients who were medicated below the threshold of conscious awareness could nonetheless develop ANV without awareness of having experienced postchemotherapy nausea (Garcia, personal communication, 5 Dee 1986). Given that a state of altered awareness seems often to be clinically (Laszlo, 1985) It would be research in the

present in several antiemetic drugs such as high dose Lorazepam a relevant experiment might be feasible. important to keep in mind, however, that the vast majority of area of both classical conditioning and food aversion learning has

involved laboratory animals and/or blinks. Among the more problematic findings causing

reasonably trivial responses such as eye “leaps of faith” needed to extend and apply

from this body of research are the fact that a rat cannot vomit (thereby problems in indirect measurement) and the fact that the original work

commonly tionability”

mentioned as showing a relationship between “anxiety” and “condiinvolved eye blinks and an effect size so small that dozens of trials were

needed to show any effect. Nonetheless, it may be possible for ANV to fit together clinical treatment and studies on learning theory. The challenge is to devise experiments that can coexist within a medical delivery system. Causal modeling approaches based on sophisticated statistical modeling procedures such as LISREL (Joreskog & Sorbom, 1981) are being used more often in theory building and confirmation. They might, at first glance, seem applicable to a study of models of ANV development. A major problem, however, is that LISREL-type analyses typically require that each hypothesized construct (like anxiety) be measured by two or more different assessment instruments or approaches (or both). A longitudinal assessment is also important in sorting out directionality (or causality) of effects (for example, changes in anxiety must precede changes in anticipatory nausea if anxiety causes anticipatory nausea). Sample sizes of several hundred are often recommended. Given the logistics of repeated psychological measurement in an ongoing medical treatment setting with marked attrition and compliance problems, it is unlikely definitive data could be gathered. BEHAVIORAL Both

anticipatory

and

TREATMENT

posttreatment

nausea

OF SIDE EFFECTS and

vomiting

are

not

completely

controlled by antiemetic medicines (Laszlo, 1983; Morrow, 1982, 1984d; Morrow, Loughner, & Bennett, 1984). Three principal interventions (Hypnosis, Progressive Relaxation Training, and Systematic Desensitization) have been investigated for their effectiveness in controlling nausea and vomiting. Table 5 summarizes selected aspects of case studies and controlled investigations of these interventions. Hyponosis Hypnosis has been defined as a state of intensified attention and receptiveness and an increased responsiveness to an idea or set of ideas (Erickson, 1959). Six case histories of hypnotherapy with cancer patients have reported positive therapeutic results (e.g., Dempster, Balson, & Whalen, 1976; Ellenberg, Kellerman, Dash, Higgins, & Zeltzer, 1980; LaBaw, Holton, Newell, & Eccies, 1975; Margolis, 1983; Olness, 1981; Zeltzer, Kellerman, Ellenberg, & Dash, 1983). The majority of these investigations have been carried out with a pediatric population probably since children are more readily hypnotized than adults (Olness, 1981). This may be also because children often experience undersirable side effects from antiemetic drugs and thus some antiemetics are not used with children (Cotanch, Hockenberry, & Herman, 1985). LaBaw et al. (1975) studied 27 children and adolescents, aged 4-20 years, who were treated with self-hypnosis over a two-year period (the exact number of treatment sessions per patient was not specified). “Progressive Body Relaxation” was employed as an inductive technique which was followed by “psychic imagery” of fantasied idyllic scenes common in the patient’s experience. Dependent measures included patients’ self-report and therapist observations. Results indicated that “varying degrees of success” were obtained where “success” was defined as improved sleep, increased caloric intake and retention, increased fluid intake, and greater tolerance for therapeutic procedures. In general, few objective data were collected from these case studies. The “hypnotic” intervention was not standardized and sometimes not described. These methodological Iimitations preclude definitive conclusions. Three controlled studies have examined hypnotherapy. Redd, Andresen, and Minagawa (1982) treated six patients experiencing ANV with hypnosis in a multiple baseline design. During treatment, each patient was individually instructed in focusing attention, achieving deep muscle relaxation, and imagining pleasant scenes. Training sessions were audiotaped and patients were instructed to practice the treatment daily. Due to unforeseen events, the hypnotic intervention was temporarily interrupted for several of the patients. At this time, the symptoms that had been apparently controlled by hypnosis reappeared, suggesting that the intervention was involved in the initial positive effects. Given the small sample size, these findings require replication before conclusions can be drawn. Cotanch et al. (1985) also combined relaxation training with hypnosis in a study of 12 pediatric inpatients (aged lo-18 years), who were referred by oncologists for treatment since they were experiencing troublesome chemotherapyrelated nausea and vomiting. The children were randomly assigned to an experimental group (n=6) or a control group (n=6) who received standard care. On the day of chemotherapy, the children in the experimental group were trained by a Both groups were followed through two consecutive therapist in self-hypnosis. chemotherapy cycles. Child self-report and nurse observations were obtained on

Olness

(1981)

25

1

Ellenberg et al. (1980)

with Cancer

2 F; 2 M 7-13(%=11) Not reported Leukemia Lymphoma Hodgkins disease Ewing’s sarcoma F 12 Inpt Leukemia

Gender: Age: Status: Dx:

Gender: Age: Status: Dx:

Gender: Age: Status:

Not reported 3-1w Inpt

F 21 Stage IV Hodgkins disease

Gender: Age: Status: Dx :

Age: Status: Dx:

Mixed (about 50-50) 4-20 Not reported Not reported

Gcndcr:

Demographic/Clinical data

Intervention

Consecutive referrals Serial case studies

Pt referred Case study

Pts referred

4

(1980)

Dash

case studies

Case study

Serial

Design

1

27

N

5. Biobehavioral

Dempster et al. (1976)

Case studies La Baw et al. (1975)

TABLE

KcSults

Investigations

No RI, 4-7 TX ( + opti ma1 qoup

1 BL (baseline) 1 TX

of improvtmcnt

Self-report NV

of pain,

Reduction

in pain,

associated

(but RX

NV for 10 pts

Self-report of anxiety, Reduction ot P(:N\’ frquency pain, anorexia, also rrducrd) nansca (N), Reduction of pain and anxiety vomiting (V) bvith rnedic,al proccdurcts Nurse r&scrsations

rrllorts

Anecdotal

dara

No ohjccti\,c

1-4 TX

AN\’ cxtinquishcd Krduction in duration of PCK lncrrasrs in “quality of lift”

PC self-report

of TX

Unspccificd number

Varying degrees of“success” reported (“Success” defined as: improved sleep. increase caloric intake and retention, increase fluid intake, and greater tolerance for therapeutic procedures.)

-

Case and Controlled

Pt self-rcporr Therapist ohscrvations

Mvasurcs

Patients:

Unspecified number of’& (Treatment)

HYPNOSIS

Protocol

Chemotherapy

Single subject Multiple baseline (ABAB design) Pts referred for ANV

Pt referred Serial case studies

Pt referred Serial case studies

Time series design Stratified (age & severity) Random assignment (1) Hypnosis (n=9) (2) Supportive Counseling (n=lO)

6

6

12

19

Redd et al. (1982)

Margolis (1983)

Zeltzer et al. (1983)

Zeltzer et al. (1984)

4 2 1 3 2

1 1 3 1

BL TX

l-3

1 BL TX 1 FU (6 m)

Pt s

No BL TX varied across

7-14 5-7

work)

Not reported 2 BL 6-17 (x=11.3) 2 TX Not reported FU (n=9) Hematologic 11 Lymphoma 3 Bone 5

5F; 7M x=14.2 Not reported Hodgkins Hematologic GYN Brain Other

Gender: Age : Status: Dx:

Gender: Age: Status: Dx:

3 F; 3M 27-54 (x =39) In- and Outpt Lymphoma Lung GYN Pancreas

Gender: Age: Status: Dx:

1 1 4

15 4 6

All F 24-54 Not reported Lung Hematologic Breast

Leukemia Lymphoma Other

Gender: Age: Status: Dx:

DX:

Pt self-report “bother” Parent report

Pt self-report of V Anxiety Health Locus of Control Illness Impact Self-esteem

Self-report of N, V, insomnia, pain

Pt self-report of N Nurse observations OfV

Bothpou,tn: Reduction in NV, and “bother”

(25% of pts rejected ‘lx) 73% reduced - PCV frcyuency -PCV intensity Reduction in trait anxiety (No changes in health locus of control, impact of illness, self-esteem)

Reduction in suffering (No objective data)

Decreased NV during all Rxsessions all pts Elimination of ANV

for

“Convenience” sample Serial case studies

Pts referred Serial case studies

Pt referred Single case

Design

2 Pts referred

10

Scott et al (1983)

Weddington et al. (1983)

12

1

Cotanch (1983)

Case studies Burish & Lyles (1979)

N

Status: Dx:

Gender: Age:

Dx:

Gender: Age: Status:

Lung

M Pt l-58 Pt Z-46 Outpt

All F 42-67 Stage III or IViInpts Ovarian No BL Pt 1: 1OTx Pt2: 4Tx

No BL 2 TX

1 BL 5F;?M 1 Xx (audiotapr) 17-49 (x =34) Outpt 4 6 FU Inpt 8 Melanoma 3 Soft cell 5 Adenocystic 1 Breast I Testicular 2

Gender: Age: Status: Dx:

F 30 Outpatient Lymphoma

Measures

of the

of

Pi 1: Pt self-report ui‘ ANV and PCNV Anxiety - srlfreport Pt 2: Pt self-repot? of ANV and PCNV

Observations cm&s Diarrhea Pt “pcrccption experience”

Anxiety N/V Caloric intake PR, UP

Anxiety, dcprcssion Puisc rate (PR) Blood Pressure (BP) NIV

TRAINING

1 BL, 11 TX (PRT)

RELAXATION

Protocol

Gender: Age: Status: Dx:

Dernographiclclillical data

TABLE 5. Continued

PCNV caloric intake, in pulse rate suggested)

‘1%

Pt 1: ANV red~ecd by 50% (intcnsit~) ‘Anxiety reduced F’t 2: ANV rcduccd hy 60% (scvcrity)

Reduction in duration, frequency. and intensity ofemesis; in cliarrhca Improved pt pcrccption of cxpcricncc

All patients-rcduccd All patients-increased - dccrcasc (Anxiety-reductions

Nausea rcducod during TX Nryatise affect reduc-ed during Vomiting eliminated Physiological arousal rcduccd

Kcsults

50

24

45

Burish et al. (1985)

Carey & Burish (1985)

studies 16

Lyles et al. (1982)

Controlled Burish & Lyles (1981)

on chemoemetogenic

(4) Nn=

(1) Professional (2) Volunteer (3) Audiotape

to:

Stratified according to Dx and Rx Random assignment

(*based therapy values)

(2) NRC

(1) PRT

to:

“Select” sample* Stratified according to Dx and Rx Random assignment

(3) Nn=

Pt referred Stratified according to Dx and Rx Random assignment to: (1) PRT (2) Attention control

Pt referred Stratified according to Dx and Rx Random assignment to: (1) PRT (2) Control

Outpt Breast Lung Ovarian Hodgkins

Status: Dx:

Not reported Breast GYN Hcmatologic Lung Other

19 F; 5 M 23-69

Gender: Age:

Status: Dx:

Not reported Outpt Breast Lung Ovarian Testicular Lymphoma Hodgkins Other

Age: Status: Dx:

25 F; 20 M 25-73

31 F; 19 M

Gender:

Gender: Age:

14 F; 2 M Not reported Outpt Mixed

Gender: Age: Status: Dx:

9 10 10 9 7

9 4 10 1

14 10 10 5 5 3 3

1 BL 3 TX 1 FU

(* pts seen at Rxl)

No BL* 6 TX

3 TX 1 FU

only) 2 FU

1 BL 2 TX (written instructions

anxiety, depression PR, BP Nurse observations

Anxiety, depression PR, BP, NIV Nurse observations

Anxiety, dcprcssion PR, BP, NIV Nurse observations

Anxiety, depression PR, BP, N/V Nurse observations distress

and

in “ma,jority”

(conhed)

Pts treated by professional cxpcricnced leas distress than did pts treated by volunteers or with audiotape

NTC group: ANV developed

PR7‘~wup: During Rx-reduction in anxiety and N Post-Rx - reduction in anxeity, N, PR, BP (ANV did not develop in PRT pts)

PRY‘ p@ : During Rx-reduction in anxiety and N Post-Rx-reduction in PR and BP nausea-less severe, shorter duration

nausea Post-Rxreduction of PR and BP (Nurse and pt reports agreed) (No changes in vomiting)

PR7‘puj~: During and post-Rx-less

a ‘-7n -r.cJ--__e-

2

92*

Morrow press)

Burish et al. (1981)

(1) (2) (3) (4)

SD PRT Counseling NTC

1

Single subject Multiple baseline Dx:

Status:

Gcndrr: Age:

from

Status: Dx:

Gender: Age:

Consecutive pts (with ANV) Group assignment to:

(1) SD (2) N-E

Gender: Age: Status: Dx:

Gender: Age: Status: Dx:

Consecutive pts Random assignment to:

(1) SD (2) Counseling (3) NTC

Consecutive pts (with ANV) Random assignment to:

*Some overlap with patients Morrow and Morrell (1982)

40

Dobkin & Morrow (1985)

(in

60

Morrow & Morrell (1982)

15 4 5 7 4

29 13 8 10

F 44 Not reported Adenocarcinoma

yrs Inpt Mixed maj. = breast

66% F; 33% M Median > 50

26 F; 8 M x =52.2 Out- and Inpt Breast Lung Lymphoma Ovarian Other

42 F; 18 M 19-76 Outpt Breast Hematologic Lung Other

3 BL 4 TX (EMG biofeedback) 3 FU

OTHER

Anxiety NV, BP, PR,

Pt self-report Anxiety

EMG

NV

Pt self-report NV Physical symptoms checklist Self-monitoring NV, tension Anxiety

1 BL 2 TX 3 FU

2 BL 2 TX 2 FU

Pt self-report NV, ANV Anxiety Health Locus of Control

2 BL 2 TX 2 FU

in terms

in severity

(conlinued)

and duration

of severity and

BP, PR, RMG I~vcls anxiety and N

rcduccd

Reduced Reduced

PCN

SD 63 PRT ,pou~ :

ANV reduced duration

Downward trend for SD pts PCNV Upward trend for NTC pts’ PCNV Anxiety higher in NTC than SD at FU3

Pts in the SD group rcportcd lrss scvcrc ANV of shorter duration than the pts in the other 2 groups No group differences in anxiety or locus of control

5

b

Kolko & RickardFigueroa (1985)

LeBaron & Zeltzer (1984)

Moore & Altmaier (1981)

3

8

9

N

case studies

Serial single case Multiple baseline (ABAB)

Single subject Multiple baseline

Serial

Design 6 F; 3 M 19-66 (x=47) Outpt Breast Sarcomas Other Not reported 10-17(x=12) Not reported Hematologic Bone

M 11-17 Outpt Hematologic

Gender: Age: status: Dx:

Gender: Age: Status: Dx:

Gendrr: Age: Status: Dx:

Demographic/Clinical data Protocol

6 2

Pt self-report N, V “hother”, “disruption” Parent rrport N, V “bother”, “disruption” Pt self-report of anxirty, distress, and chcmo-rclatcd symptoms Observer rating of distress

2-3 BL 2-3 Tx(distraction)

3-5 RI, 3 TX (vidro game distraction) 3 Withdrawal 3 TX

interview

Measures Structured Anxiety

5. Continued

6. TX (stress inoculation; including PRT) 4 3 2

TABLE

in anxiety

Reduction

in ANV

01

of PCNV

and disruption

and severity

Reduced NIV, bother, activities

Reduction

Results

543

Anticipatory Nausea and Vomiting

the parameters of nausea and vomiting (intensity, amount of oral intake 24 hours postchemotherapy.

severity, frequency) and on the Data were collected by staff

nurses and research assistants who were unaware which group the child was assigned to. In the experimental group, there was a significant reduction in nausea and vomiting both in terms of intensity and severity, and a significant increase in oral intake. These changes were not evident in the control group. Although this study is suggestive regarding the effectiveness of hypnotherapy, the results could experimental

of an attention effect. The children in the attention from a therapist whereas the chilare suggestive, they dren in the control group did not. While these findings require replication with an attention-placebo group to rule out this alternative hypothesis. Zeltzer, counseling acteristics

be explained in terms group received “extra”

LeBaron, and Zeltzer (in press) compared hypnotherapy to supportive and suggested that nonspecific therapy effects, such as demand charand/or attention, may contribute to treatment changes found. In their

study, 19 children, aged 16-17 years, were randomly assigned to a hypnotherapy or a supportive counseling group. Children in both groups reported reductions in nausea and vomiting and rated chemotherapy as “less noxious” following intervention. There were, however, no statistically or clinically significant differences in outcome found between the two approaches. Overall, it appears that studies using hypnosis for ANV control that this intervention may be beneficial for children. Less evidence demonstrating hypnotherapy

its usefulness for adults. for cancer chemotherapy

in which patients groups are needed.

are

Progressive Relaxation Progressive

Relaxation

randomly

have shown is available

In order to establish the effectiveness of patients, studies with larger sample sizes

assigned

to treatment

or appropriate

control

Training Training

(PRT)

is a behavioral

technique

which

involves

learning how to relax by actively tensing and relaxing muscle groups in a progressive manner. Typically, an individual is taught deep muscle relaxation by a therapist, a training audiotape is made, and the individual is requested to practice PRT at home in order to acquire the skill. Four case studies and five controlled investigations of PRT have been reported thus far. Results from case studies (Burish & Lyles, 1979; Cotanch, 1983; Weddington, Blindt, & McCracken, 1983) have suggested that PRT can benefit cancer patients by reducing side effects, such as postchemotherapy nausea and vomiting, and negative states such as depression and anxiety. Lyles, Burish, Krozely, and Oldham (1982) studied cancer patients experiencing anxiety, depression, nausea, and vomiting; 50 patients were randomly assigned to one of three treatment conditions: (a) relaxation training with guided imagery; (b) therapist control, in which a therapist spent an equal amount of time with the patients as in condition (a); and, (c) no-treatment control. Patients who received PRT were found to be less anxious and nauseated both during and following their chemotherapy treatment sessions compared to the patients in the two other groups. Additionally, patients in the PRT group evidenced less physiological arousal (pulse rates) and were less depressed following chemotherapy. The positive treatment effects found in the PRT condition, but not in the therapist

544

G. R. Morrow and E L. Dobkin

attention-control condition, suggest that result of “nonspecific” treatment factors.

the improvements

were

not simply

the

These results were supported by Cotanch (1983) in a study in which 43 cancer patients were randomly assigned to one of three conditions: (1) PRT, provided by audiotape; (2) tape control, in which patients listened to soothing music and were requested took part Cotanch

to focus on positive thoughts; and (3) no-treatment control. in one baseline session and “varying numbers” of training reported

that

67%

of the

patients

in the

PRT

conditions

Patients sessions.

showed

in-

creases in postchemotherapy nausea and vomiting (PCNV), whereas 85% of the patients in the two control groups showed increased in PCNV. This 18% difference between the groups suggests that PRT can be helpful even with a minimal amount of therapist contact. In the first reported PRT (1987) investigated whether

prevention study, Burish, Carey, Krozely, PRT could be used to prevent or at least

and Greco ameliorate

chemotherapy side effects through early intervention. Thirty-two cancer patients about to start their first course of emetogenic chemotherapy were randomized to either a PRT group or a no-treatment prior to the initiation of chemotherapy treatments. Patients in the PRT group following chemotherapy. Additionally,

control group. PRT and during the first

sessions were held five chemotherapy

reported feeling less nauseated these patients reported fewer

during and occurrences

of vomiting and lower physiological arousal (e.g., heart rate and blood pressure) compared to the patients in the control group. PRT also resulted in less dysphoria and a progressive reduction in PCNV symptoms as treatment with chemotherapy continued over time. Reportedly, experienced PCN whereas 54%

by the fifth session only 10 % of the PRT patients of the control patients experienced PCN. These

findings are both impressive and encouraging. Overall, the data regarding the use of PRT with cancer chemotherapy patients show that this type of intervention can be effective in reducing side effects which are present during and after cancer toms occurring before a treatment

chemotherapy session were

sessions. Since nausea sympnot assessed in the early work

carried out by Burish and his associates (Burish, personal communication), any potential impact PRT has on ANV symptoms defined strictly as pretreatment remains uninvestigated. Given the relationship between postchemotherapy side effects, and the subsequent development of ANV, using this technique before the initial chemotherapy treatment could potentially block the conditioning process and thereby

prevent

its occurrence.

System tic Desensifiza tion Systematic Desensitization (SD) is a well-developed, standardized behavioral technique which has been shown to be useful in altering maladaptive learned 1983). In SD, patients are first taught a responses such as phobias (Wolpe, response incompatible with the maladaptive response they presently have to particular stimuli. This alternative response is then paired in imagination with the original stimuli so as to countercondition the maladaptive response. At the University of Rochester Cancer Center, ANV patients have been taught a modified version of Progressive Muscle Relaxation as a competing response to the maladaptive response of anticipatory nausea/vomiting. During the Systematic

545

Anticipatory Nausea and Emiting

Desensitization related

treatment,

to chemotherapy

patients treatment

imagine (such

scenes

as driving

from

a hierarchy

to the cancer

of events

center)

while

remaining deeply relaxed. In this way, treatment stimuli become associated with relaxation so that when the patient encounters stimuli (such as the clinic nurse), they respond with relaxation rather than nausea and vomiting. Two case studies and four controlled investigations of SD have been reported (Dobkin, 1987; Hailey & White, 1983; Hoffman, 1983; Meyer, 1982; Morrow, 1986; Morrow & Morrell, 1982). Morrow and Morrell(1982), in a study designed to examine the antiemetic efficacy of SD for the control of ANV, randomly assigned 60 patients based on a Rogerian,

with ANV to one of three client-centered approach;

groups: (a) SD; (b) counseling, and (c) no-treatment control. It

was found that only patients in the SD group showed a significant reduction in the frequency, severity, and duration of ANV. The efficacy of SD appeared unrelated to antiemetic medications used by the patients. Additionally, patients in the SD group reported the counseling tion)

no greater expectation for improvement than did the patients in group. Thus, apparently, nonspecific therapy effects (e.g., atten-

were not responsible

for the reported

positive

effects.

In a follow-up study, Morrow (198613) compared the effectiveness of SD to (a) relaxation only, (b) counseling, and (c) no-treatment control in order to examine the essential treatment components in the SD procedure. Relative to the other three groups, patients treated with SD reported a significant decrease in the severity

and duration

of anticipatory

SD and relaxation patients and severity of posttreatment

nausea

from

had a significantly nausea compared

baseline

to follow-up

sessions.

greater decrease in the duration to patients who were in the other

two groups. These results were independent of patients’ ratings of their expectations for success or the credibility of the experimenter. Results support a view that

both the cognitive stimulus hierarchy ponents for the successful treatment In another investigation, retarding, or preventing

and relaxation of ANV.

Dobkin (1985) the development

examined of both

response

are necessary

com-

the possibility of reducing, post- and anticipatory side

effects in new-to-chemotherapy cancer patients. Forty consecutive patients were randomly assigned to either a SD or Waiting-List Control group. SD was administered in two separate one-hour sessions prior to the second chemotherapy cycle. A repeated measures design was employed with one baseline and three follow-up periods; dependent measures included: postchemotherapy nausea and vomiting, anticipatory nausea and vomiting, anxiety (trait at baseline, state at all periods), and tension

level postchemotherapy.

Preliminary results showed a downward trend for the SD patients’ PCNV side effects, in that nausea and vomiting were less frequent, severe, and of shorter duration as chemotherapy progressed. In contrast, there was an upward trend for the control group patients’ PCNV side effects in that nausea and vomiting were more frequent, severe, and of longer duration as chemotherapy progressed. In addition, the control patients reported higher levels of tension following chemotherapy compared to SD patients. These preliminary findings suggest that introduction of SD early in treatment can reduce and retard the development of conditioned side effects resulting from chemotherapy in cancer patients. In summary, consistent findings regarding the effectiveness of SD have been reported in both case and controlled studies. Although the majority of the SD

G. R. Morrow and P L. Dobkin

546

investigations one controlled replicated

have been carried out in one research investigation (Meyer, 1982) in other

these promising

Other Behavioral Five

studies

above.

results.

Interventions

have

Moore

center, two case studies and geographical locations have

used

behavioral

and Altmaier

interventions

(1981)

reported

other

than

(6 female, 3 males; mean Training (the six-session

age=47 years) who were treated treatment “package” consisted

Modification

with

patients

combined

were interviewed

PRT

and

and completed

the three

a pilot study of nine

Education).

patients

with Stress Inoculation of Cognitive Behavior

Prior

the Multiple

discussed

cancer

Affect

to the

intervention,

Adjective

Checklist

(MAACL: Zuckerman, Lubin, Vogel, & Valerius, 1964) which is designed to measure anxiety, depression, and hostility. Five of the nine patients exhibited ANV prior to treatment. Since the authors discontinued monitoring the patients at the last training session (i.e., no follow-up data), it is impossible to determine the effectiveness of this approach. Three patients did report, however, that they felt less anxious prior to treatment having learned effective coping skills. Burish, Shartner, and Lyles (1981) treated a 44-year-old female cancer patient with EMG biofeedback combined with relaxation in order to help control anticipatory

and posttreatment

symptoms.

Baseline

measures

of affect (anxiety,

depres-

sion, and hostility, MAACL: Zuckerman et al., 1964), muscle tension, pulse rate, and blood pressure were taken. Following ten training sessions, the patient was able to reduce her physiological arousal levels (as measured and blood pressure) and reported feeling less nauseated.

by EMG, pulse rate, These changes were

maintained during three follow-up periods. There were no reported improvements in affect. These results, which are similar to those found for PRT, suggest that EMG Biofeedback may be another means of teaching relaxation to cancer patients. Three studies, LeBaron and Zeltzer (1984) Kolko and Rickard-Figueroa (1985),

and Redd,

have investigated mer study involved

Jacobsen,

Die-Trill,

interventions

based

directing

Dermatis,

McEvoy,

on cognitive

diversion

adolescents’

attention

and Holland techniques.

(1987) The

for-

away from the administration

of chemotherapy by having them play games and be actively engaged with a therapist (modified relaxation training was also included in some cases). The latter two studies involved the use of video games to distract pediatric and adolescent patients. LeBaron and Zeltzer (1984) reported a decrease in postchemotherapy nausea, vomiting, and “bother” during the intervention, along with fewer disruptions of activities following treatment. These improvements were maintained at follow-up (time period not reported). The investigators concluded that “since a repeated measures design found no symptom reductions prior to intervention, it can be assumed that some aspects of the intervention itself were responsible for the reductions found, rather than attention or expectations related to assessment alone” (p. 180). This conclusion seems premature given a sample size of n=8, no control groups, and no theoretical rationale of how distraction alone could reduce chemotherapy-related side effects. Kolko and Rickard-Figueroa (1985) used (Kazdin, 1982) with three pediatric oncology

a multiple-baseline patients (all male,

(ABAB) design aged 11, 16, and

Anticipatory Nausea and Vomiting

547

Anxiety 17 years). Anticipatory “distress,” PCNV and state anxiety (State-Trait Inventory, Spielberger, Gorsuch, & Lushene, 1968) measures were taken. A Modified-Procedure Behavioral Rating Scale (Katz, Kellerman, & Siegel, 1980) was used to gather observational data five minutes prior to chemotherapy. Introduction of video games concurrent with administration of chemotherapy was associated with reduction of self-reported and observer-reported anticipatory distress. These improvements symptoms (not ANV) as well as postchemotherapy were reversed when the return-to-baseline phase was initiated. Symptoms decreased again when the video game procedure was reintroduced in two of the three cases (the third case could not be evaluated due to admission of the patient just before the final condition). Kolko and Rickard-Figueroa’s study addressed the relaxation versus distraction issue regarding the therapeutic mechanisms underlying behavior therapy. Although the results imply that distraction may effectively reduce symptomatology, ANV and PCNV side effects per se were not reduced; only “distress” levels were altered. one of which employed an Redd et al. (1987) conducted two experiments, ABAB design, to evaluate the effect of video game playing in pediatric oncology patients with anticipatory nausea and anxiety. In the first experiment, patients were alternately assigned to either the experimental group or to the control condition. In the second experiment, patients from the first experiment were carried over and exposed to an ABAB design presentation of the video games. Unlike Kolko and Rickard-Figueroa’s (1985) study, the ABAB presentation took place baseline within the same chemotherapy session, that is, after a no-video-game assessment (A), patients played video games for 10 minutes (B), followed immediately by a lo-minute no-video-game period (A), and then a second lo-minute video game period (B). Nausea and anxiety were assessed using a 10 cm Visual Analogue Scale; pulse rate and blood pressure measures were taken to assess physiological indices. Results indicated that anticipatory nausea but not anxiety decreased significantly following video game playing in both studies. Reportedly, pulse rates and systolic/diastolic blood pressure rates were variable, with one measure (systolic blood pressure) showing a significant increase from the previous no-game level following the second exposure to video games within the session. The authors interpreted these findings as support for the hypothesis that cognitive distraction, and not relaxation, is the critical component of the intervention. Redd et al.‘s (1987) findings complement Kolko and Rickard-Figueroa’s investigation quite well; however, a few methodological flaws need to be rectified before conclusions can be drawn. First, consecutive patients should be randomly assigned to groups. Second, an attention-control group should be included. Third, follow-up measures, both subsequent to chemotherapy for at least three days and following several chemotherapy cycles, need to be documented in order to evaluate treatment effects. Moreover, larger sample sizes should be employed.

Mechanisms

of Actions

Several hypotheses attempting to explain how and why behavior therapy is effective for cancer patients have been proposed. These explanations range from simple placebo effects to theories involving the conditioning process discussed in the previous section.

G. R. Morrow and I? L. Dobkin

Intuitively, “nonspecific factors” such as attention received from a therapist, demand characteristics, patient expectation for success, or other “placebo” effects may seem involved in the effectiveness of behavior therapy with cancer patients. While nonspecific treatment effects may contribute to the improvements reported, several studies have demonstrated that other treatment components are necessary for behavior therapy to be effective. For instance, Morrow (1982) found that expectations for success and demand characteristics did not distinguish patients receiving Systematic Desensitization from patients in counseling and no-treatment control groups. In support of these findings, Burish and his colleagues (1982) presented data from a credible attention-control group suggesting that specific behavioral techniques, not the nonspecific factors involved in the intervention, were responsible for treatment efficacy. A second hypothesis proposed to explain the underlying mechanisms of behavior therapy is that these techniques distract the patients sufficiently to divert their attention away from the chemotherapy treatments such that conditioning no longer occurs. In other words, by removing the conditioned stimuli, one would prevent the occurrence of the CR (i.e., ANV). Although appealing, this explanation is unlikely. Most oncology clinics are equipped with televisions and patients are often accompanied by friends or relatives; these stimuli have failed to diminish the side effects experienced by cancer patients. Also, studies including control groups have demonstrated that behavior therapy is more effective than attentiondiverting alternative procedures. Finally, although hypnotherapy and PRT use guided imagery to divert the patients’ attention awayfrom the aversive chemotheraSD focuses attmtion on these stimuli. The cognitive hierarchy empy experience, ployed involves scenes which depict going to the chemotherapy clinic. A third hypothesis proposed to explain the effectiveness of behavior therapy is that these techniques change the patients’ self-perceived sense of control over their disease process. According to this view, the patients’ psychological state improves and their subjective experience of helplessness diminishes. Although appealing, this hypothesis has no empirical data to support it. In fact, Morrow and Morrell (1982) noted that reductions in ANV were not accompanied by change in locus of control as measured by Wallston, Wallston, and DeVellis’ (1978) Health Locus of Control Scale. Burish and Carey (1984) note that one common element of the various behavioral interventions is the induction of the relaxed state. Progressive Relaxation Training has been shown to reduce anxiety and physiological arousal and, therefore, may be responsible for the reduction in chemotherapy aversiveness. Redd and Andrykowski (1982) suggest that deep muscular relaxation may directly inhibit muscular contractions that generally precede vomiting, thereby eliminating this response. Data from the Morrow (1986a) study suggest that relaxation itself is a necessary but not sufficient treatment component for reducing anticipatory nauseaivomiting. The data suggests that counterconditioning is a necessary element in the treatment package. As was previously discussed, ANV symptoms are thought to be learned, maladaptive responses which are associated with chemotherapy treatment effects. Using Systematic Desensitization (which involves both relaxation and cognitive elements) to break the associative bond affords the patient the opportunity to learn a new, adaptive response. Relaxation training does, however, appear to provide symptom relief during and following the administration of chemotherapy.

Anticipatory Nausea and Vomiting

FUTURE

549

RESEARCH DIRECTIONS

Methodology Several points can be made regarding the methodology of future investigations of chemotherapy side effects. First, these symptoms are best described using a prospective, longitudinal research design. A “one-shot” study gives a flat picture of multidimensional experience. The process of being treated for cancer requires a repeated-measures design. Second, adequate sample sizes and appropriate control groups are essential. Treatments should be described in sufficient detail so that comparisons across studies can be made. It would be helpful if findings from these investigations were reported in terms of both clinical and statistical significance (Morrow, 1980, 1982a, 1984b, 1984d, 1986). Finally, a more critical selection of instruments is needed for the assessment of cancer patients. naires developed on a psychiatric or college student population are

Questionfrequently

used with cancer patients. Due to both illness and treatment factors, these instruments may be inappropriate for a medical population (cf. Morrow, 1980; Dobkin & Morrow, 1986).

Treatment

Comparison

Studies

Four

studies have compared different interventions for the treatment of cancer 1982; Morrow, 1986; Morrow & Morrell, 1982; patients’ side effects (Meyer, Zeltzer et al., 1984). Three of these compared Systematic Desensitization to other techniques and one compared Hypnosis to Supportive Counseling. Since different modalities may be beneficial for particular side effects (e.g., Hypnosis for pain management, or Systematic Desensitization yet to be explored. Results from comparative ing mechanisms

Maintenance Behavior

of the intervention

and Generalization

therapy

has been

strategies.

of Treatment

shown

for ANV), this is an important area studies may also clarify the underly-

Effects

to be the “treatment

of choice”

for numerous

disorders (e.g., phobias, enuresis, sexual dysfunction, obesity), yet a recurrent problem is the maintenance and generalization of treatment effects. Burish and his colleagues have consistently reported a reduction of positive treatment effects when the therapist is no longer present to direct the patient in PRT. Redd et al. (1982) showed, through the use of a multiple baseline treatment design, that when Hypnosis was not used, symptoms recurred. These findings were replicated by Kolko and Rickard-Figueroa (1985) and Redd et al. (1987). A lack of generalization of treatment effects comes as no surprise when one considers the poor track record of long-term maintenance of behavioral change for many different problem areas (e.g., alcoholism, smoking, weight management). Future research needs to be directed at developing “built-in” relapse prevention strategies so that the benefits gained can be maintained.

Treatment There could

Exportation

are too few psychologists working with oncology benefit from their interventions. For this reason,

patients to assist all who it is important to explore

550

G. R. Morrow and P L. Dobkin

the “exportability” of biobehavioral treatments. That is, can behavior therapy be administered without the presence of a psychologist? Morrow (1984a) examined the possibility of reducing the necessary professional time required to administer SD by comparing a live presentation to an audiotaped presentation to the relaxation component of the SD procedure. In the live presentation condition, individual instruction in PRT was provided to the patient; the session was tape-recorded and given to the patient for subsequent home practice. In the audiotaped presentation condition, the patient was provided with a prerecorded audiotape which instructed the patient in PRT methods. All patients were requested to practice PRT daily. Surprisingly, 4 of 5 patients in the audiotaped presentation condition spontaneously reported that listening to the audiotape triggered a nausea response. None of the five patients in the live presentation condition reported this phenomenon. When interpreted within a conditioning paradigm, the voice on the tape, a novel stimulus, may have been incorporated in a stimulus configuration which was able to elicit the conditioned nausea response. In contrast, during the live presentation there was a wider range of novel, potentially conditionable stimuli present (e.g., the objects in the room, smells, etc.) such that no single stimulus was easily incorporated into the hierarchy of stimuli which triggered a conditioned response. Redd, Rosenberger, and Hendler (1983) a 1so found that patients listening to self-hypnosis audiotapes complained that listening to the tapes made them feel nauseous. Redd et al. (1983) also speculated that the therapist’s voice became a conditioned stimulus, eliciting nausea. Considering the fact that using audiotapes alone does not seem to work, an alternative approach may be to instruct other health professionals in the administration of behavior therapy. Thus, an important research direction involves answering the questions: Can biobehavioral techniques be taught to health professionals and still be as effective? and Will patients perceive physicians and nurses as being as credible results suggest that

as a psychologist in the use of these procedures? the answers to these questions are yes (Morrow

Preliminary & Dobkin,

1987). Carey (1985), however, found that patients trained in progressive relaxation by psychologists experienced less distress than did patients who were treated either by a trained volunteer (women from the community) or an audiotaped presentation of the treatment procedure. These results support Morrow’s (1984a) and Redd et al.‘s (1983) conclusions concerning audiotape interventions and also suggest that a health professional may be required to administer the treatment. Patients seem to be more likely to respond to a health professional’s rather than a volunteer’s efforts to intervene in their cancer treatment. Clearly, more work is warranted in this area in order to maximize the number of cancer chemotherapy patients who may benefit from biobehavioral interventions. Past and current studies show clearly the promise of fruitful and productive further research in understanding the development and treatment of anticipatory side effects from chemotherapy treatment. ANV continues to be as challenging an area to study as it is relevant and worthwhile; potentially useful a model for addressing important theoretical issues as it is a model for studying aspects of behavioral interventions. It is as useful an area for interdisciplinary research collaboration as it is for treatment and patient care collaboration; as good an area for teaching students the frustrations and fruits of research with medically ill

Anticipatory Nausea and Vomiting

patients as it is for continually reminding it is as personally rewarding an area challenging.

551

their supervisor of the same points; and of study as it remains professionally

REFERENCES P. M., & Kean, T. M. (1984). Ahles, T. A., Cohen, R. E., Little, D., Balducci, L., Dubbert, Toward a behavioral assessment of anticipatory symptoms associated with cancer chemotherapy. Journal ofBehaviorTherapy and Experimental Psychiatry, 1.5, 141-145. Altmaier, E. M., Ross, W. E., & Moore, K. (1982). A pilot investigation of the psychologic functioning of patients with anticipatory vomiting. Cancer, 49, 201-204. Andrykowski, M. A. (1986). Definitional issues in the study of anticipatory nausea in cancer chemotherapy. Journal ofBehavioralMedicine, 9, 33-41. Andrykowski, M. A. (1987). Do infusion-related tastes and odors facilitate the development of anticipatory nausea? A failure to support hypothesis. Health Psychology, 6, 329-341. Andrykowski, M. A., & Redd, W. H. (1987). Longitudinal analysis of the development of anticipatory na~sea.~o~r~ ~~~ons~~~~~~ and ~li~~al Psyc~lo~, 55, 36-41. nausea: Andrykowski, M. A., Redd, W. H., & Hatfield, A. K. (1985). D evelopment of anticipatory A prospective analysis. ~JournalofConsulting and Clinical Psycholoply,53, 447-454. Andrykowski, M. A., Redd, W. H., Jacobsen, P. B., Marks, E., Gorfinkle, K., Hakes, T. B., Kaufman, R. J., Currie, V. E., & Holland, J. (1987). P revalence,predictors, and course of anticipatory nausea in wornan rec&ing ~~~va~~ c~~t~~apy~ breast cancer. UnpubIished manuscript. Bernstein, I. L. (1978). Learned taste aversions in children receiving chemotherapy. &z&e, 200, 1302-1303. Borison, II., & Wang, S. C. (1953). Physiology and pharmacology of vomiting. Pharmacolo~ I&&W, 5, 193-230. Borison, II., Hawken, M. J., Hubbard, J. L., & Sirett, N. E. (1975). Unit activity from the cat area postrema influenced by drugs. Brain Reseurch, 92, 153-156. Burish, T. G. (1985). Personal communication. Burish, 7: G., & Carey, M. P. (1984). Conditioned responses to cancer chemotherapy: Etiology and treatment. In G. H. Fox, & B. H. Newberry (Eds.), Impact ofpsychaendocrinesystems in cancer immunity (pp. 147-178). Toronto: C. J. Hogrefe. Burish, T. G., & Carey, M. P. (1986). Conditioned aversive responses in cancer chemotherapy patients: Theoretical and developmental analysis. j&r& 4 consulting and clinical Psyc~lo~, 5, 593-600. Burish, ‘I’. G., Carey, M. P., Krozely, M. G., & Greco, A. (1987). Conditioned side effects induced by cancer chemotherapy: Prevention through behavioral treatment. Journal ofConsulting and Clinical P&wlogy, 55, 42-48. Burish, T. G., & Lyles, J. N. (1979). Eff ec t’lveness of relaxation training in reducing the aversiveness of chemotherapy in the treatment of cancer. Journal o~Be~v~ora1 Therapy and Experimental Psychiatry, 10,357-361. Burish, T. G., & Lyles, J. N. (1981). Effectiveness of relaxation training in reducing adverse reactions to cancer chemotherapy. Journal ofBehavioral Medicine, 4, 65-78. Burish, ‘I: G., Shartner, C. D., & Lyles, J. N. (1981). Effectiveness of muhiple-site EMG biofeedback and relaxation in reducing the aversiveness of cancer chemotherapy~ Bj~ed~~k and &lffiegulation, 6, 523-535. Carey, M. P. (1985). ProvLiing relaxation training to cancer chemotherapypa&ts: A comparison of three delivery strategies. Unpublished doctoral dissertation, Vanderbilt University, Nashville. Carey, M. P., & Burish, T. G. (1985). Anxiety as a predictor of behavior therapy outcome for cancer chemotherapy patients. Journal o~consultin~ and ClintGalPsyc~lo~, 53, 860-865. Cella, D. F., Pratt, A., & Holland, J. C. (1984). Long-term conditioned nausea and anxiety persisting in cured Hodgkin’s patients after chemotherapy (Abstract). Proceedings 4 the Annual Meeting oj the American Society oj Clintial Oruology, 3, 73. Chang, J. C. (1981). Nausea and vomiting in cancer patients: An expression of psychological mechanisms? Psychosomatics, 22, 707-709. Chatterjee, S., & Price, B. (1977). Regression adysis by ex~rnp~e. New York: John Wiley & Sons. Cohen, J., & Cohen, P. (1975). Appl’d ze muIt+1 I e Feg ression/correla& amz&ses_f&the behavioralsciences. New York: John Wiley & Sons.

G. R. Morrow and I? L. Dobkin

Cohen, R. E. (1982). Distress associatedwith anti-neoplasticchemotherapy:Prediction, assessment, and treatment. Unpublished doctoral dissertation, State University of New York, Albany. Cotanch, P. H. (1983). Muscle relaxation versus ‘Wention -placebo” in decreasingthe aversivenessof&m&erapy. Unpublished manuscript, Duke University, Durham. Cotanch, P., Hockenberry, M., & Herman, S. (1985). Self-hypnosis as antiemetic therapy in children receiving chemotherapy. Oncology Nursing Forum, 12, 41-46. Davis, C. J., Lake-Bakaar, G. V., & Grahame-Smith, D. G. (1986). Nausea and Vomiting:Mechanisms and Treatment.Berlin: Springer-Verlag. Dempster, C. R., Balson, P., & Whalen, B. T. (1976). Supportive hypnotherapy during the radical treatment of malignancies. The InternationalJournal ofClintia1 and Experimental Hypnosis, 24, l-9. Dobkin, P. (1987). The use of systematic desensitization, a behavioral intervention, in the reduction of aversive chemotherapyside effects in cancer patients. Unpublished doctoral dissertation, University of Georgia, Athens. Dobkin, P. L., & Morrow, G. R. (1985, October). Prevention of chemotherapyinducednauseaand vomiting. Paper presented at the meeting of Biobehavioral Oncology, 6th Eupsyca Symposium, Zaragoza, Spain. Dobkin, P., Zeichner, A., & Dickson-Parnell, B. (1985). C oncomitants of anticipatory nausea and emesis in cancer chemotherapy. Psycholoptial Reports, 56, 671-676. Dobkin, P. L., & Morrow, G. R. (1986). Biopsychosocial assessment of cancer patients: Methods and suggestions. The Hospice Journal, 2, 37-59. Dolgan, M. J., Katz, E. R., McGinty, K., & Seigel, S. E. (1985). Anticipatory nausea and vomiting in pediatric cancer patients. Pediatrtis, 75, 547-552. Dragoin, W. B. (1971). Conditioning and extinction of taste aversions with variations in the intensity of the CS and UCS in two strains of rats. Psychonomic Sctince, 22, 303-305. Duigon, A. (1986). Anticipatory nausea and vomiting associated with cancer chemotherapy. Oncology Nursing Forum, 13, 35-40. Ellenberg, L., Kellerman, J., Dash, J., Higg ins, G., & Zeltzer, L. (1980). Use of hypnosis for multiple symptoms in an adolescent girl with leukemia. Journal of Adolescent Health Cam, 1, 132136. Erickson, M. H. (1959). Hypnosis is painful terminal illness. American Journal of Clinical Hypnosis, 1, 117. Fdez-Arguelles, P., Guerrero, J., Duque, A., Borrego, A., & Marquez, J. (1985, October). Nauseas y vomitos anticipatoriosen pacientes cancerosossometidosa tratamientode quimioterapti. Paper presented at the meeting in Biobehavioral Oncology, 6th Eupsyca Symposium, Zaragoza, Spain. Fetting, J. H., Wilcox, P. M., Iwata, B. A., Criswell, E. L., Bosmajian, L. S., & Sheider, V. R. (1983). Anticipatory nausea and vomiting in an ambulatory medical oncology population. Cancer TreatmentReports, 67, 1093-1098. Frytak, S., & Moertel, C. G. (1981). Management of nausea and vomiting in the cancer patient. Journal of the American Medical Association, 245, 393-396. Garcia, J. (1986). Food for To&an: Cognition and cathexis in concert. Manuscript submitted for publication. Garcia, J. (1986). Personal communication. Garcia, J., Ervin, F. R., & Koellinger, R. A. (1966). Learning with prolonged delay of reinforcement. Psychonomic Science, 5, 121-122. Gil, K. (1984). Coping effectively with invasive medical procedures: A descriptive model. Clinical Psycholo+lReview, 4, 339-362. Gralla, R. J., Itri, L. M., Piski, S. E., Squillante, A. E., Kelsse, D. P., Brown, D. W., Bordon, L. A., Braun, T. J., & Young, C. W. (1981). Antiemetic efficacy of high-dose metoclopramide: Randomized trials with placebo and prochlorperazine in patients with chemotherapy-induced nausea and vomiting. New EnElandJournal ofMedicine, 305, 905-909. Hailey, B. J., & White, J. G. (1983). Systematic desensitization for anticipatory nausea associated with chemotherapy. Psychosomatics, 24, 287-291. Harris, A. L. (1984, September). Mechanisms and treatmentof cytoto.& inducednausea and vomiting. Paper presented at the symposium in the mechanisms and treatment of nausea and vomiting in man. Oxford, England. Hendler, C. S., & Redd, W. H. (in press). Fear of hypnosis: The role of labeling in patients’ acceptance of behavioral interventions. Behavior Therapy. Hoagland, A. C., Morrow, G. R., Bennett, J. M., & Carnrike, C. L. M. (1983). Oncologists’view of cancer patient noncompliance. Amertian Journal of Clinical Oncology, 6, 239-244. Holland, J. (1977). Psychological aspects of oncology. Medical Clinics ofNorth America, 61, 737-748.

Anticipatory Nausea and Vomiting

553

Houts, P., Lipton, A., Harvey, H., Eyster, E., Curry, S., &Thompson, C. (1982, December). The relationship between pretreatment anxiety and pretreatment nausea among chemotherapy patients. Paper presented at the 13th International Cancer Congress, Seattle, WA. Houts, P., Morrow, G. R., Lipton, A., Harvey, H. A., & Simmons, M. A. (1984). The role ofpretreatment anxtity in anticipatory nausea among cancer patients receiving chemotherapy. Unpublished manuscript, University of Pennsylvania. Horowitz, M. J., & Wilner, N. (1980). Life events, stress, and coping. In L. Poon (ed.), Agin.g in the 1980%: Psychological issues. Washington: American Psychological Association. Ingle, R. J., Burish, T. G., & Wallston, K. A. (1984). Conditionability of cancer chemotherapy patients. Oncology Nursing Forum, 11,97-102. National Educational Joreskog, K. G., & Sorbom, D. (1981). LZSREL % U serk Guide. Chicago, Resources. Katz, E. R., Kellerman, J., & Siegel, S. E. (1980). B e h avioral distress in children with cancer undergoing medical procedures: Developmental considerations. Journal of Consulting and Clinical P’sycholo~, 48, 356-365. Kazdin, A. E. (1982). Single-case research designs, New York: Oxford University Press. Kolko, D. J. & Rickard-Figueroa, J. L. (1985). Effects of video games in the adverse corollaries of chemotherapy in pediatric oncology patients: A single-case analysis. Journal of Consulting and Clinical Psychology, 53, 223-225. Cambridge: Cambridge University Press. Konorski, J. (1948). conditionedreflexerand neuronm#ti. Kutz, I., Borysenko, J. Z., Come, S. E., & Benson, H. (1980). Paradoxical emetic response to antiemetic treatment in cancer patients. New England Journal ofMedicine, 303, 1480. LaBaw, W., Holton, C., Tewell, K., & Eccles, D. (1975). Th e use of self-hypnosis by children with cancer. American Journal ofClinical Hypnosis, 17, 233-238. Laszlo, J. (1983). Antiemetics and Gamer chemotherapy. Baltimore: Williams and Wilkins. Laszlo, J. (1985). Lorazepan as an adjunct in cancer chemotherapy. Proceedings of an Investigational Revtiw-June 10, 1983. Oncology Times, 7, 5-10. LeBaron, S., & Zeltzer, L. K. (1984). Behavioral intervention for reducing chemotherapy-related nausea and vomiting in adolescents with cancer. Journal of Adolescent Health Care, 5, 178-182. Lewis-Beck, M. S. (1980). Appl ied res ression: An introduction. Quantitative applications in the social sciences. Sage University. London: Sage Publications. Love, R. R., Nerenz, D. R., & Leventhal, H. (1982). Anticipatory nausea with cancer chemotherapy: Development through two mechanisms (Abstract). Proceedings ofthe Annual Meetiq of the American Society of Clinkal Oncology, 2, 242. Lyles, J. N., Burish, T. G., Krozely, M. G., & Oldham, R. K. (1982). Efficacy of relaxation training and guided imagery in reducing the aversiveness of cancer chemotherapy. Journal of Consulting and Clinical Psychology, 50, 509-524. Macintosh, N. J. (1983). Conditioniq and Associative Learning. Oxford: Clarenden Press. Maher, J, (1981). Intravenous lorazepam to prevent nausea and vomiting associated with cancer chemotherapy. The Lance& January, 91-92. Margolis, C. G. (1983). Hypnotic imagery with cancer patients. American Journal of Clinical Hypnosis, 25, 128-134. Meyer, J. (1982). Systematic desensitization versus relaxation training and no treatment (controls) fw the reduction qf nausea, vomiting; and anxiety resultinlfrom chemotherapy. Unpublished doctoral dissertation, Virginia Commonwealth University, Richmond. Millon, T., Green, C. J., & Meagher, R. B. (1979). The MBHI: A new inventory for the psychodiagnostician in medical settings. Projessional P.&wlogy, 10,529-539. Money, K. E., & Cheung, B. S. (1983). Another function of the inner ear: Facilitation of the emetic response to poisons. Am&m and Space Environmental Medicine, 54, 208-211. Moore, K., & Altmaier, E. M. (1981). Stress inoculation training with cancer patients. Cancer Nursing October, 389-393. Morrow, G. R. (1980). Clinical trials in psychosocial medicine: Methodologic and statistical considerations. Part III. Assessing measurement techniques in psychosocial oncology. Gamer ilieatment Reports, 64, 451-456. Morrow, G. R. (1981). Behavioral treatment of anticipatory nausea and vomiting during chemotherapy. Proceedings of the American Association of Gamer Research and American Society of Clinical Oncology, 22, 396. Morrow, G. R. (1982). Prevalence and correlates of anticipatory nausea and vomiting in chemotherapy patients. Journal of the National Cancer Institute, 68, 585-588. Morrow, G. R. (1984a). Appropriateness of taped versus live relaxation in the systematic desensiti-

554

G. R. Morrow

and I? L. Dobkin

zation of anticipatory nausea and vomiting in cancer patients. Joumzl of Consulting and Clinical Psych&~, 52, 1098-1099. Morrow, G. R. (1984b). The assessment of nausea and vomiting: Past problems, current issues and directions for future research. Cancer, 52, 2267-2279. Morrow, G. R. (1984c). Clinical characteristics associated with the development of anticipatory nausea and vomiting in cancer patients undergoing chemotherapy treatment. Journal of Clinical Oncolo~, 10, 1170-1176. Morrow, G. R. (1984d). Prevalence, etiology and treatment of chemotherapy induced anticipatory nausea and vomiting. Proceedings of the American Cancer SocZety 4th National Conference on Human Values and Cancer. Morrow, G. R. (1984e). Susceptibility to motion sickness and the development of anticipatory nausea and vomiting in cancer patients. Cancer Treatment Reports,68, 99-100. Morrow, G. R. (19849. Susceptibility to motion sickness and chemotherapy induced side-effects. Lancet, 52, 1098-1099. Morrow, G. R. (1985). The effect of a susceptibility to motion sickness on the side effects of cancer chemotherapy. Cancer, 55, 2766-2770. Morrow, G. R. (1986a). Effect of the cognitive hierarchy in the systematic desensitization treatment of anticipatory nausea in cancer patients: A component comparison with relaxation only, counseling, and no treatment. Cognitive Therapy and Research, 10, 421-446. Morrow, G. R. (1986b). Behavioral management of chemotherapy-induced nausea and vomiting in the cancer patient. The Clinical Oncologist, 113, 1 l-14. Morrow, G. R. (1987a). Predicting the developnent of anticipatory nausea and vomiting in cancer patients: A prospective validation. Manuscript submitted for publication. Morrow, G. R. (in press). Physiologic measures of chemotherapy-induced nausea and vomiting. Proceedings of the Amertian Association of Cancer Research and American Society of Clinical Oncolo~. Morrow, G. R., & Labrum, A. H. (1978). The relationship between physiological and psychological measures of anxiety. Psychological Medicine, 8, 95-101. Morrow, G. R., Arseneau, J. C., Asbury, R. F., Bennett, J. M., & Boros, L. (1982). Anticipatory nausea and vomiting in chemotherapy patients. New England Journal of Medicine, 306, 431-432. Morrow, G. R., & Dobkin, P. L. (in press). Biobehavioral aspects of cancer treatment side effects. Journal of Experimental and Clinical Cancer Research. Morrow, G. R., & Dobkin, P. L. (1987). A n expl orat ion of nonspectfu therapy effects of systematic desensitization in cancer chemotherapy patients. Unpublished manuscript. Morrow, G. R., 8r Morrell, C. (1982). Behavioral treatment for the anticipatory nausea and vomiting induced by cancer chemotherapy. New England Journal of Medicine, 307, 1476-1480. Morrow, G. R., Loughner, J., & Bennett, J. M. (1984). Prevalence of nausea and vomiting (N&V) and other side effects in patients receiving Cytoxan, Methotrexate, Fluorouracil (CMF) therapy with and without Prednisone. Proceedings of the AmeriGan Society for Clinical Oncolo~, 3, 105. Nachman, M., & Ashe, J. H. (1973). L earned taste aversions in rats and as a function of dosage, concentration, and route of administration of LiCl. Physiology and Behavior, 10, 73-78. Nesse, R. M., Carli, T., Curtis, G. G., & Kleinman, P. D. (1980). Pre-treatment nausea in cancer chemotherapy: A conditioned response? Psychosomatic Medicine, 42, 33-36. Nerenz, D. R., Leventhal, H., Easterling, D. V., & Love, R. R. (1986). Anxiety and drug taste as predictors of anticipatory nausea in cancer chemotherapy. Journal ofClinical Oncology, 4, 224-233. Nicholas, D. R. (1982). Prevalence of anticipatory nausea and emesis in cancer chemotherapy patients. Journal ofBehavioral Medicine, 5, 461-462. Nicholas, D. R. (1983). Anticipatory nausea in cancer chemotherapy: Cognitive, motorzc, and physiological components. Unpublished doctoral dissertation, University of Southern Mississippi. Nicholas, D. R., & Hollandsworth, J. G. (1986). Assessment of anticipatory nausea and vomiting in cancer patients undergoing chemotherapy: Theoretical and methodological considerations. Journal of Psychosocial Oncology, 4, 6 l-75. Olafsdottir, R., Sjoden, P-O., & Westling, B. (1986). P revalence and prediction of chemotherapyrelated anxiety, nausea, and vomiting in cancer patients. Behavior Research and Therapy, 24, 59-66. Olness, K. (1981). Imagery (self-hypnosis) as adjunct therapy in childhood cancer: Clinical experience with 25 patients. AmericanJournal of Pediatric Hematolo~/Oncology, 3, 313-321. Oman, C. M., Lichtenberg, B. K., & Mooney, K. E. (1984). Space motion sickness monitoring experiment: Space Lab One. NATA-AGARD A erasp ace Medical Symposium Proceedings. Palmer, B. V., Walsh, G., McKinna, J. A., & Greening, W. P. (1980). Adjuvant chemotherapy for breast cancer: Side effects and quality of life. British Medical Journal, 281, 1594-l 597.

Anticipate

Nauseu

and Vomiting

555

Pennebaker, J. W. (1982). The Psycholo~ ofP&cal LQmptoms. New York: Springer-Verlag. Penta, J., Poster, D., & Bruno, S. (1983). The pharmacologic treatment of nausea and vomiting caused by cancer chemotherapy: A review. In J, Laszlo (Ed.), An&me&s and cancer c~t~ru~y (pp. 53-92). Baltimore: Williams & Wifkins. Reason, J. T., & Brand, J. J. (1975). Motion Sickness. London: Academic Press. Redd, W. H., Andresen, G. U., Minagawa, R. Y. (1982). Hypnotic control of anticipatory emesis in cancer patients receiving chemotherapy. Journal of Consulting- and C&&al Psychology, 50, 14-19. Redd, W. H., & Andrykowski, M. A. (1982). Behavioral intervention in cancer treatment: Controlling aversion reactions to chemotherap~~~r~l ~Consu~t~ng and ~~in~a~ Psyc~lo~, 50, 1018-1029. Redd, W. H., Burish, T. G., & Andrykowski, M. A. (1985). Aversive conditioning and cancer chemotherapy. In 7: G. Burish, S. M. Levy, & B. E. Meyerowitz (Eds.), Cancer, nutrition, and taste auersisn: A biobehavioral perspective (pp. 117-132). Hillsdale: Lawrence Erlbaum. Redd, W. H., Jacobsen, P. B., Die-Trill, M., Dermatis, H., McEvoy, M., & Holland, J. (1987). ~ognitive/attentional distraction in the control of conditioned nausea in pediatric cancer patients receiving chemotherapy. Journal of Consulting and Clinical Psychology, 55, 391-395. Redd, W. H., Rosenberger, P. H., & Hendler, C. S. (1983). Controlling chemotherapy side effects. American Journal of Clinical Hypnosis, 25, 161-172. Rescorla, R. A. (1973). Effect of US habituation following conditioning. Journal of Comparative and Physio~og~a~ P~c~lo~, 82, 137-143. Rescorla, R. A. (1978). Some implications of a cognitive perspective on Pavlovian conditioning. In S. H. Hulse, H. Fowler, & W. K. Honig (Eds.), Cognitive processes in animal behavior (pp. 15-50). Hillsdale: Lawrence Erlbaum. Rescorla, R. A. (1979). Aspects of the reinforcer learned in second-order Pavlovian conditioning. Journal of Experimental Psychology: Animal Behaoior Processes, 5, 79-95. Rizley, R. C., & Rescorla, R. A. (1972). Association in second-order conditioning and sensory preconditioning. Journaf of Comparative and Physiologicat’ Psychology, 81, l-l 1. Robertson, R. G., & Garcia, J. (1985). X-rays and learned taste aversions: Historical and psychological ramifications. In T. G. Burish, S. M. Levy, &B. E. Meyerowitz (Eds.), Cancer, nutrition, and eating behavior: A btibehavWra1 perspective (pp. 1 l-41). Hillsdale: Erlbaum. Rossi, A., Bonadonna, G., Taruni, G., Bayetta, I%., Marchini, S., Valagussa, P., & Veronesi, U. (1979). Trials ofa&ant c~t~Tapy in breast cancer. The experience ofthe Istituto Nazionale lknori in Milan. Paper presented at the Second Breast Cancer Working Conference, Copenhagen, Denmark. Rozin, P., & Lee, P. (1972). Long extension of effective CS-US interval by anthesthesia between CS and US.Journal of Comparative Physiological Psycholqgy, 80, 43-48. Rusiniak, K. W., Patmerino, C. C., Rice, A. G., Forthman, D. L., & Garcia, J. (1982). Flavorillness aversions: Potentiation of odor by taste with toxin but not shock in rats. Journaf of Comparative Physiology and Psychology, 96, 527-539. Schultz, L. S. (1980). Classical (Pavlovian) conditioning of nausea and vomiting in cancer chemotherapy (Abstract). Proceedings of the American Socz& ofClinical Oncology, 21, 244. Scogna, D. M., & Smalley, R. V. (1979). Chemotherapy-induced nausea and vomiting. Amerkan Journat’ of Nursing, 79, 1562-1564. Silver, B. V., & Blanchard, E. B. (1978). Biofeedback and relaxation training in the treatment of psychophysiological disorders: Or, are the machines really necessary?Jurml of Behavioral Medicine, 1, 217-247. Spence, K. W. (1958). A theory of emotionally based drive (D) and its relation to performance in simple learning situations. Amerkan Psyc~logist, 13, 131-141. Spielberger, C. D., Gorsuch, R. L., & Lushene, R. (1968). Thestate-trait anxiety inuentory &STAI). Palo Alto, Consulting Psychologists Press. Sullan, S. E., Cronin, C., Zelen, M., & Zinberg, N. E. (1980). Antiemetics in patients receiving chemotherapy for cancer: A randomized comparison of delta-9-tetrahydrocannabinol and prochlorperazine. New Eng~a~Jour~~ ofnifediime, 302, 135-l 38. van Komen, R. W., & Redd, W. H. (1985). Personality factors associated with anticipatory nausea/ vomiting in patients receiving cancer chemotherapy. He&i Psychology, 4, 189-202. WaIlston, K. A., Wallston, B. S., & DeVellis, R. (1978). Development of the multidimensional health locus of control (MHLC) scales. Health Edwation Monograph, 6, 160-170. Weddington, W. W. (1982). Psychogenic nausea and vomiting associated with termination of chemotherapy Psyc~~ap~ and Psy~~s~t~s, 37, 129-l 36. Weddington, W. W., Blindt, K. A., & McCracken, S. G. (1983). Relaxation training for anticipatory nausea associated with chemotherapy. Psychosomatics, 24, 281-283.

G. R. Morrow and I? L. Dobkin

Weddington, W. W., Miller, N. J., & Sweet, D. L. (1982). Anticipatory nausea and vomiting associated with cancer chemotherapy. New EnglandJournal ofMedicine, 307, 825-826. Weddington, W. W., Miller, N. J., & Sweet, D. L. (1984). Anticipatory nausea and vomiting associated with cancer chemotherapy. Journal ofPsychosomatic Research, 28, 73-77. Whitehead, V. M. (1975). Cancer treatment needs better antiemetics. New EnglandJournal ofMeditine, 293, 199-200. Wilcox, P. M., Fetting, J. H., Nettescheim, K. M., & Abeloff, M. D. (1982). Anticipatory vomiting in women receiving cyclophosphamide, methotrexate, and 5-FU (CMF) adjuvant chemotherapy for breast carcinoma. Cancer TreatmentReports, 66, 1601-1604. Wilson, J. P., Rahdert, E. R., Black, C. D., Taylor, H. G., & Holloway, H. C. (1986). Identifying anxiety and/or depression in emesis-prone cancer chemotherapy patients (Abstract). Proceedingsof the Annual Meeting 4 the American Society of Clinical Oncology, 5, p 23 9. Wolpe, J. (1983). Tkpractice ofbehavior therapy. (3rd ed.) New York: Pergamon Press. Zeltzer, L., Kellerman, J,, Ellenberg, L., & Dash, J. (1983). Hypnosis for reduction of vomiting associated with chemotherapy and disease in adolescents with cancer. Journal of Adolescent Health Care, 4, 77-84. Zeltzer, L., LeBaron, S., & Zeltzer, P. M. (1984). The effectiveness of behavioral intervention for reduction of nausea and vomiting in children and adolescents receiving chemotherapy. Journal of Clinical Oncology, 2, 683-690. of experimentally Zuckerman, M., Lubin, V., Vogel, L., & Valerius, E. (1964). Measurement induced affects. Journal of Consulting and Clinical Psycholo~, 28, 418-425.