The Hospice Journal
ISSN: 0742-969X (Print) (Online) Journal homepage: http://www.tandfonline.com/loi/ippc19
Denial in Terminal Illness: To Intervene or Not to Intervene Stephen R. Connor To cite this article: Stephen R. Connor (1992) Denial in Terminal Illness: To Intervene or Not to Intervene, The Hospice Journal, 8:4, 1-15, DOI: 10.1080/0742-969X.1992.11882739 To link to this article: http://dx.doi.org/10.1080/0742-969X.1992.11882739
Published online: 21 Sep 2017.
Submit your article to this journal
Article views: 2
View related articles
Citing articles: 5 View citing articles
Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=ippc19 Download by: [University of New England]
Date: 10 November 2017, At: 05:30
Denial in Terminal Illness: To Intervene or Not to Intervene
Downloaded by [University of New England] at 05:30 10 November 2017
Stephen R. Connor
ABSTRACT. This research examined effects of psychosocial intervention on denial-related coping ability in terminally ill can cer patients. Most patients considered deniers were seen using interpersonal denial in the service of preserving relationships important to them. Denial use as a result of intrapsychic impover ishment was seen infrequently. Twenty-four patients with various terminal cancers were assigned to either an experimental (n = 13)' or a control (n = 11) group. A pretest versus posttest multifacto rial analysis of variance model was used. Experimental subjects received a structured psychosocial intervention just before posttest. Results showed that, as predicted, interpersonal deniers in the experimental group had a statistically significant (p < .005) decrease in denial scores at posttest, whereas denial scores in creased in the control group. Data indicate that, in the absence of psychological dysfunction, patients using interpersonal denial may respond favorably to sensitive psychosocial intervention. Still others may benefit only after they have had sufficiept time to Stephen R. Connor, PhD, was Hospice Director, Kaiser Permanente Medical Center, Walnut Creek, CA, when he conducted this research. This work was presented in part at the 10th annual meeting of the National Hospice Organization, Los Angeles, November 15-18, 1987. The research was supported in part by the Community Service Program of Kaiser Foundation Hos pitals and in part by the Professional School of Psychology in San Francisco. The author wishes especially to thank the following people for their valuable advice and critical review: Charles Dithrich, PhD; Philip Fischbacher, MD; An drew Kneier, PhD; Dale Larson, PhD; and Elliott Rosen, EdD. The Medical Editing department of Kaiser Foundation Hospitals provided editorial assistance. Address correspondence to Stephen Connor, PhD, Executive Director, Hos pice of Central Kentucky, 2919 Ring Road, P.O. Box 368, Elizabethtown, KY 42702-0368. The Hospice Journal, Vol. 8(4) 1992 � 1992 by The Haworth Press, Inc. All rights reserved.
J
2
THE HOSPICE JOURNAL
muster adaptive coping strategies. Areas for further research are suggested.
Downloaded by [University of New England] at 05:30 10 November 2017
INTRODUCTION Like Hamlet, we who care for the terminally ill often find our selves in an equivocation of hesitations and uncertainties when confronted with patients who appear to use denial. Should we in tervene psychosocially to help them develop more adaptive coping strategies, or is their use of denial essential for their emotional survival and thus in need of our support? The answer varies with the patient. This research was undertaken to provide guidance to caregivers about an important clinical question: What effect does psychosocial intervention have on denial-related coping behaviors in terminally ill patients, and can those who benefit from intervention be distin guished from those for whom intervention will not be helpful? Denial is commonly thought of as a repudiation of a painful shared reality and the replacement of the painful reality with a more agreeable version. For purposes of this inquiry, the definition of denial was expanded to include other denial-related coping strat egies, such as suppression, repression, negation, and reaction forma tion, used as defenses aimed at excluding the threat or its signifi cance from awareness (Verwoerdt, 1972). On the basis of my clinical experience with hundreds of hospice patients, I have come to believe that most who use denial do so to preserve important interpersonal relationships. I call this interper sonal denial. I have also observed that a smaller number of patients use denial to help preserve a weak ego. I call this intrapsychic denial. Interpersonal deniers seem to use denial primarily to cope with guilt about the effect of their condition on others, to protect others from the emotional distress they might feel if the patient were to openly acknowledge his or her condition and feelings, or out of fear of abandonment. I maintain that these patients can be helped by sensitive intervention. Some intrapsychic deniers appear to deny because of serious
Stephen R. Connor
3
Downloaded by [University of New England] at 05:30 10 November 2017
psychological dysfunction and may thus benefit from their use of denial. Others may use denial temporarily to gain some emotional distance so that they can more gradually face and finally come to terms with their condition. In these cases intervention may not be helpful. LITERATURE REVIEW The concept of denial as a defense first appeared in the early writings of Sigmund Freud in the early 1920s and in those of Anna Freud in the 1940s. Sigmund Freud referred to denial as disavowal of reality. Later authors have examined denial as an adaptive cop ing process (Beilin, 1981; Beisser, 1979; Dansak & Cordes, 197879; Hackett & Cassem, 1970) or as a useful strategy during the initial stages of response to cancer (Detwiler, 1981; Falek & Britton, 1974; Kubler-Ross, 1969; Weisman, 1972). More recently, investigators have begun to look at both the positive and the negative functions of denial in seriously ill and aged patients. Lazarus and Golden (1981) proposed that a person's circumstances, the timing of denial, a situation's ambiguity, and the pervasiveness of denial should be examined. Weisman (1972) considered the degree of denial crucial. First order denial is a patient's obvious denial of the primary facts of an illness. Second-order denial, which may develop after a patient accepts the diagnosis, is denial of the clinical significance or impli cations of an illness. Third-order denial is a patient's inability, after fully accepting the diagnosis and its import, to believe that the disease will progress to death. Such patients believe they will remain in their incapacitated state indefinitely. Breznitz (1983, p. 258) ex panded on Weisman's concepts and proposed seven distinct types of denial "each related to a different stage in the processing of the threatening information." Weisman proposed that the primary purpose of denial is not sim ply to avoid danger but to prevent the loss of important interper sonal relationships. Weisman and Worden (1975) as a result of their work on social support with terminally ill cancer patients suggested that patients who remain in effective touch with others
Downloaded by [University of New England] at 05:30 10 November 2017
4
THE HOSPICE JOURNAL
and who can accept the help of others without alienating them live longer than patients who withdraw and become depressed. The research reported here was an outgrowth of Weisman and Worden's (1975) observations about the interpersonal nature of denial as well as the more recent research by Lazarus and Golden (1981) suggesting a need to understand the conditional variables on which denial depends so that people can be grouped on the basis of similarities and differences for the purpose of devising interven tions. This research is also an attempt to develop greater under standing of psychosocial intervention with dying patients. It chal lenges the prevailing belief that intervention with patients using denial is usually risky.
METHOD Because we studied two groups during two time periods, we used a multifactorial analysis of variance for data analysis. We did not expect to find differences between terminally ill cancer patients using intrapsychic denial-related coping behaviors who received intervention and those who did not receive it. We did, on the other hand, anticipate differences between those receiving intervention in the interpersonal denial-related behavior group and those not re ceiving it. Instruments The Defense Mechanism Inventory (DMI) Reversal Scale (Gieser & Ihilevich, 1969) because of its established reliability and validity (Gieser & Sacks, 1973) was used as the primary measure of denial. The Templer/McMordie Death Anxiety Scale (DAS) (McMordie, 1979) was used to help distinguish intrapsychic from interpersonal denial. We presumed that patients who used intrapsy chic denial would not deny their anxiety on a self-report measure like the DAS. Patients with statistically significantly low DAS scores were considered intrapsychic deniers, whereas those with high or normal scores were considered interpersonal deniers. The DAS has been found to be acceptably reliable, valid, and useful as
Downloaded by [University of New England] at 05:30 10 November 2017
Stephen R. Connor
5
an auxiliary measure of denial (Kastenbaum & Costa, 1977). Users of this score are cautioned, however, by Kastenbaum and Costa to measure and control for social desirability and general anxiety influ ences. We, therefore, added the Social Desirability Scale (SOS) (Crowne & Marlowe, 1964) and the Manifest Anxiety Scale (MAS) (Taylor, 1953) to control for the influences of desire for approval and of trait anxiety. Reliability and validity for both of these scales has been well established (Byrne, 1974; Crowne & Marlowe, 1964; Gocka & Holloway, 1963; Taylor, 1953; Wiggins, 1968). The pretest consisted of the DMI, DAS, SOS, and MAS. The posttest included only the DMI and DAS. A recent review of denial measurement in terminally ill patients has been published elsewhere (Connor, 1986).
Procedures Patients being treated in the oncology clinic at the Kaiser Permanente Medical Center in Walnut Creek, California, who were identified as potential subjects by a medical oncologist, were con tacted and asked if they would be willing to participate in a study of how people cope with serious illness. To be eligible, patients had to have been diagnosed with metastatic melanoma or with non small-cell cancer of the lung, breast, pancreas, stomach, colon, or rectum. In addition, they had to be terminally ill, with a limited prog nosis; to not be cognitively impaired; and they had to be ambulatory, with a rating of 50 or higher on the Kamofsky Activity Scale (Kamofsky & Burchenal, 1949). Those patients willing to participate signed an informed consent statement and were given the pretest. Patients whose pretest score was one or more standard deviations above the mean DMI Reversal Scale score were included in the sample. Patients whose scores were less than one standard devia tion below the mean were not considered deniers and were exclud ed from the study. The subjects were randomly assigned to either the experimental or the control group. Experimental group subjects received psychosocial intervention immediately before the posttest. Control group subjects were given the posttest with no intervention. A two- to three-week period between pretest and posttest was established to correspond with DMI reliability studies.
THE HOSPICE JOURNAL
6
The Intervention Definition
Downloaded by [University of New England] at 05:30 10 November 2017
I define sensitive intervention as the practitioner's ability to talk openly about an illness, the impact it has on the patient and family, and the patient's hopes and fears about it.
The Questionnaire For this study I developed a structured psychosocial interven tion based on questions I have asked for many years in making psychosocial assessments of patients newly referred to hospice. I have found these encounters remarkably effective in allowing patients to become more open about experiences shared with important people in their lives. My emphasis has been on how a person's interpersonal world has been influenced by an incur able illness. The purpose of the intervention in this study was to create an opportunity for the patient to explore and to gain in sight into his or her own coping processes. Questions were direct but sufficiently open-ended for subjects to be able to choose their own level of self-disclosure. The interview was paced so that those who gave guarded responses were not required to elabo rate; subjects who were less guarded were encouraged to talk more about their feelings, perceptions, and memories. The inter view included the following questions: 1. Different people experience different kinds of difficulties when they are ill. What, for you, have been some of the most difficult aspects of having cancer? 2. Are there any things you do, or that other people do, that make these difficulties easier to deal with? 3. Is there anything you or other people do, that make these difficulties harder to deal with? 4. Do you believe you will or will not recover from this illness? 5. Have you had any close encounters with death? 6. What effect has your illness had on your family and close friends, and how have they reacted to it?
Stephen R. Connor
7
Downloaded by [University of New England] at 05:30 10 November 2017
7. How do you feel about the way in which your family and friends have been effected by or have reacted to your illness? 8. Is there anything good that has come out of your having cancer? 9. Are there any other thoughts or feelings about your illness or the questions I've asked that you'd like to talk more about? Sample A sample size of 24 was determined sufficient to test interper sonal denial. Because only two subjects met the criteria for intra psychic denial, it was estimated that enrollment by about 300 sub jects would have been necessary to find an adequate number to test for it. Given the limited time allotted for the study, this test was deemed infeasible. Of the 60 subjects referred to the study during the three years of data collection, 10 declined to participate, 18 died before comple tion of the study, and 6 did not use denial sufficiently to qualify. A total of 26 patients randomly assigned to either the experimental or the control group completed the pretest and the posttest. Of the 26 only 2 met the criteria for intrapsychic denial and since we had deemed it infeasible to test for that, we excluded them from the study. We thus had a final sample of 24 who met the criteria for interpersonal denial-13 experimental and 11 control. Nineteen were women and 5 were men; their mean age was 61; range 35-80. All were from the middle socioeconomic class and had at least a high school education. Disease sites varied: 9 (38%) had breast cancer, 8 (33%) had gastrointestinal malignancies, 6 (25%) had lung can cer, and 1 (4%) had metastatic melanoma. The following results are based on the data on these subjects. RESULTS Standard t-tests were used to determine whether social desirabili ty or trait anxiety influenced results on denial scores. A chi-squared tabulation was done to determine sex bias in the two groups. Be-
Downloaded by [University of New England] at 05:30 10 November 2017
8
THE HOSPICE JOURNAL
cause of the small number of men in the study (5), the Fisher exact test was also used. Results of both the SDS and the MAS showed no statistically significant differences between experimental and control subjects. Neither social desirability nor trait anxiety seemed to influence test results in these groups; nor were statistically significant differences found between the groups according to sex. The DAS was analyzed using a two-factor repeated-measures analysis of variance. The two factors were group (experimental vs. control) and time (pretest vs. posttest). Neither the interaction effect nor the main effect for group difference was statistically significant. Although a statistically significant overall decrease was found in DAS scores between pretest and posttest periods for all subjects (p < .05), the death anxiety decrease in the experimental group was not statistically significant. To control for pretest scale differences, an analysis of covariance was also conducted. Posttest DAS group means did not differ from the pretest. All group means were within one standard deviation of the means reported by McMordie. Table I shows the F-test results of the DAS scores (see also Figure 1). For the DMI measure, the main effects of group and time did not yield statistical significance. However, the interaction test showed a statistically significant tendency for greater group differ ences at posttest than at pretest [F( l, 22) = 9.782; p < .005]. Table 2 shows the F-test results of the DMI pretest and posttest (see also Figure 2). Denial means for both groups were nearly identical at pretest. At posttest, the experimental group mean de creased, whereas the control group mean had increased. Test results indicated that, as predicted, interpersonal deniers who received the intervention scored statistically significantly lower at posttest on the DMI Reversal Scale and within the normal range on the DAS. DISCUSSION Results of this study would seem to indicate that, in the absence of psychological dysfunction, terminally ill patients using denial
Stephen R. Connor
9
Table 1 Pretest and Posttest ANOVA for Death Anxiety
Downloaded by [University of New England] at 05:30 10 November 2017
Scale Scores in 24 Terminally Ill Patients
Source
DF
SS(H)
Mes
.r
Between subjects Group Error 1
23 1 22
3932.81 27.57 3905.24
27.57 17.7.51
.16
Within subjects Time Group X time Error 2
24 1 1 22
823.50 130.02 45.20 648.28
130.02 45.20 29.47
4.41* 1.53
\
* 12
ISSN: 0742-969X (Print) (Online) Journal homepage: http://www.tandfonline.com/loi/ippc19
Denial in Terminal Illness: To Intervene or Not to Intervene Stephen R. Connor To cite this article: Stephen R. Connor (1992) Denial in Terminal Illness: To Intervene or Not to Intervene, The Hospice Journal, 8:4, 1-15, DOI: 10.1080/0742-969X.1992.11882739 To link to this article: http://dx.doi.org/10.1080/0742-969X.1992.11882739
Published online: 21 Sep 2017.
Submit your article to this journal
Article views: 2
View related articles
Citing articles: 5 View citing articles
Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=ippc19 Download by: [University of New England]
Date: 10 November 2017, At: 05:30
Denial in Terminal Illness: To Intervene or Not to Intervene
Downloaded by [University of New England] at 05:30 10 November 2017
Stephen R. Connor
ABSTRACT. This research examined effects of psychosocial intervention on denial-related coping ability in terminally ill can cer patients. Most patients considered deniers were seen using interpersonal denial in the service of preserving relationships important to them. Denial use as a result of intrapsychic impover ishment was seen infrequently. Twenty-four patients with various terminal cancers were assigned to either an experimental (n = 13)' or a control (n = 11) group. A pretest versus posttest multifacto rial analysis of variance model was used. Experimental subjects received a structured psychosocial intervention just before posttest. Results showed that, as predicted, interpersonal deniers in the experimental group had a statistically significant (p < .005) decrease in denial scores at posttest, whereas denial scores in creased in the control group. Data indicate that, in the absence of psychological dysfunction, patients using interpersonal denial may respond favorably to sensitive psychosocial intervention. Still others may benefit only after they have had sufficiept time to Stephen R. Connor, PhD, was Hospice Director, Kaiser Permanente Medical Center, Walnut Creek, CA, when he conducted this research. This work was presented in part at the 10th annual meeting of the National Hospice Organization, Los Angeles, November 15-18, 1987. The research was supported in part by the Community Service Program of Kaiser Foundation Hos pitals and in part by the Professional School of Psychology in San Francisco. The author wishes especially to thank the following people for their valuable advice and critical review: Charles Dithrich, PhD; Philip Fischbacher, MD; An drew Kneier, PhD; Dale Larson, PhD; and Elliott Rosen, EdD. The Medical Editing department of Kaiser Foundation Hospitals provided editorial assistance. Address correspondence to Stephen Connor, PhD, Executive Director, Hos pice of Central Kentucky, 2919 Ring Road, P.O. Box 368, Elizabethtown, KY 42702-0368. The Hospice Journal, Vol. 8(4) 1992 � 1992 by The Haworth Press, Inc. All rights reserved.
J
2
THE HOSPICE JOURNAL
muster adaptive coping strategies. Areas for further research are suggested.
Downloaded by [University of New England] at 05:30 10 November 2017
INTRODUCTION Like Hamlet, we who care for the terminally ill often find our selves in an equivocation of hesitations and uncertainties when confronted with patients who appear to use denial. Should we in tervene psychosocially to help them develop more adaptive coping strategies, or is their use of denial essential for their emotional survival and thus in need of our support? The answer varies with the patient. This research was undertaken to provide guidance to caregivers about an important clinical question: What effect does psychosocial intervention have on denial-related coping behaviors in terminally ill patients, and can those who benefit from intervention be distin guished from those for whom intervention will not be helpful? Denial is commonly thought of as a repudiation of a painful shared reality and the replacement of the painful reality with a more agreeable version. For purposes of this inquiry, the definition of denial was expanded to include other denial-related coping strat egies, such as suppression, repression, negation, and reaction forma tion, used as defenses aimed at excluding the threat or its signifi cance from awareness (Verwoerdt, 1972). On the basis of my clinical experience with hundreds of hospice patients, I have come to believe that most who use denial do so to preserve important interpersonal relationships. I call this interper sonal denial. I have also observed that a smaller number of patients use denial to help preserve a weak ego. I call this intrapsychic denial. Interpersonal deniers seem to use denial primarily to cope with guilt about the effect of their condition on others, to protect others from the emotional distress they might feel if the patient were to openly acknowledge his or her condition and feelings, or out of fear of abandonment. I maintain that these patients can be helped by sensitive intervention. Some intrapsychic deniers appear to deny because of serious
Stephen R. Connor
3
Downloaded by [University of New England] at 05:30 10 November 2017
psychological dysfunction and may thus benefit from their use of denial. Others may use denial temporarily to gain some emotional distance so that they can more gradually face and finally come to terms with their condition. In these cases intervention may not be helpful. LITERATURE REVIEW The concept of denial as a defense first appeared in the early writings of Sigmund Freud in the early 1920s and in those of Anna Freud in the 1940s. Sigmund Freud referred to denial as disavowal of reality. Later authors have examined denial as an adaptive cop ing process (Beilin, 1981; Beisser, 1979; Dansak & Cordes, 197879; Hackett & Cassem, 1970) or as a useful strategy during the initial stages of response to cancer (Detwiler, 1981; Falek & Britton, 1974; Kubler-Ross, 1969; Weisman, 1972). More recently, investigators have begun to look at both the positive and the negative functions of denial in seriously ill and aged patients. Lazarus and Golden (1981) proposed that a person's circumstances, the timing of denial, a situation's ambiguity, and the pervasiveness of denial should be examined. Weisman (1972) considered the degree of denial crucial. First order denial is a patient's obvious denial of the primary facts of an illness. Second-order denial, which may develop after a patient accepts the diagnosis, is denial of the clinical significance or impli cations of an illness. Third-order denial is a patient's inability, after fully accepting the diagnosis and its import, to believe that the disease will progress to death. Such patients believe they will remain in their incapacitated state indefinitely. Breznitz (1983, p. 258) ex panded on Weisman's concepts and proposed seven distinct types of denial "each related to a different stage in the processing of the threatening information." Weisman proposed that the primary purpose of denial is not sim ply to avoid danger but to prevent the loss of important interper sonal relationships. Weisman and Worden (1975) as a result of their work on social support with terminally ill cancer patients suggested that patients who remain in effective touch with others
Downloaded by [University of New England] at 05:30 10 November 2017
4
THE HOSPICE JOURNAL
and who can accept the help of others without alienating them live longer than patients who withdraw and become depressed. The research reported here was an outgrowth of Weisman and Worden's (1975) observations about the interpersonal nature of denial as well as the more recent research by Lazarus and Golden (1981) suggesting a need to understand the conditional variables on which denial depends so that people can be grouped on the basis of similarities and differences for the purpose of devising interven tions. This research is also an attempt to develop greater under standing of psychosocial intervention with dying patients. It chal lenges the prevailing belief that intervention with patients using denial is usually risky.
METHOD Because we studied two groups during two time periods, we used a multifactorial analysis of variance for data analysis. We did not expect to find differences between terminally ill cancer patients using intrapsychic denial-related coping behaviors who received intervention and those who did not receive it. We did, on the other hand, anticipate differences between those receiving intervention in the interpersonal denial-related behavior group and those not re ceiving it. Instruments The Defense Mechanism Inventory (DMI) Reversal Scale (Gieser & Ihilevich, 1969) because of its established reliability and validity (Gieser & Sacks, 1973) was used as the primary measure of denial. The Templer/McMordie Death Anxiety Scale (DAS) (McMordie, 1979) was used to help distinguish intrapsychic from interpersonal denial. We presumed that patients who used intrapsy chic denial would not deny their anxiety on a self-report measure like the DAS. Patients with statistically significantly low DAS scores were considered intrapsychic deniers, whereas those with high or normal scores were considered interpersonal deniers. The DAS has been found to be acceptably reliable, valid, and useful as
Downloaded by [University of New England] at 05:30 10 November 2017
Stephen R. Connor
5
an auxiliary measure of denial (Kastenbaum & Costa, 1977). Users of this score are cautioned, however, by Kastenbaum and Costa to measure and control for social desirability and general anxiety influ ences. We, therefore, added the Social Desirability Scale (SOS) (Crowne & Marlowe, 1964) and the Manifest Anxiety Scale (MAS) (Taylor, 1953) to control for the influences of desire for approval and of trait anxiety. Reliability and validity for both of these scales has been well established (Byrne, 1974; Crowne & Marlowe, 1964; Gocka & Holloway, 1963; Taylor, 1953; Wiggins, 1968). The pretest consisted of the DMI, DAS, SOS, and MAS. The posttest included only the DMI and DAS. A recent review of denial measurement in terminally ill patients has been published elsewhere (Connor, 1986).
Procedures Patients being treated in the oncology clinic at the Kaiser Permanente Medical Center in Walnut Creek, California, who were identified as potential subjects by a medical oncologist, were con tacted and asked if they would be willing to participate in a study of how people cope with serious illness. To be eligible, patients had to have been diagnosed with metastatic melanoma or with non small-cell cancer of the lung, breast, pancreas, stomach, colon, or rectum. In addition, they had to be terminally ill, with a limited prog nosis; to not be cognitively impaired; and they had to be ambulatory, with a rating of 50 or higher on the Kamofsky Activity Scale (Kamofsky & Burchenal, 1949). Those patients willing to participate signed an informed consent statement and were given the pretest. Patients whose pretest score was one or more standard deviations above the mean DMI Reversal Scale score were included in the sample. Patients whose scores were less than one standard devia tion below the mean were not considered deniers and were exclud ed from the study. The subjects were randomly assigned to either the experimental or the control group. Experimental group subjects received psychosocial intervention immediately before the posttest. Control group subjects were given the posttest with no intervention. A two- to three-week period between pretest and posttest was established to correspond with DMI reliability studies.
THE HOSPICE JOURNAL
6
The Intervention Definition
Downloaded by [University of New England] at 05:30 10 November 2017
I define sensitive intervention as the practitioner's ability to talk openly about an illness, the impact it has on the patient and family, and the patient's hopes and fears about it.
The Questionnaire For this study I developed a structured psychosocial interven tion based on questions I have asked for many years in making psychosocial assessments of patients newly referred to hospice. I have found these encounters remarkably effective in allowing patients to become more open about experiences shared with important people in their lives. My emphasis has been on how a person's interpersonal world has been influenced by an incur able illness. The purpose of the intervention in this study was to create an opportunity for the patient to explore and to gain in sight into his or her own coping processes. Questions were direct but sufficiently open-ended for subjects to be able to choose their own level of self-disclosure. The interview was paced so that those who gave guarded responses were not required to elabo rate; subjects who were less guarded were encouraged to talk more about their feelings, perceptions, and memories. The inter view included the following questions: 1. Different people experience different kinds of difficulties when they are ill. What, for you, have been some of the most difficult aspects of having cancer? 2. Are there any things you do, or that other people do, that make these difficulties easier to deal with? 3. Is there anything you or other people do, that make these difficulties harder to deal with? 4. Do you believe you will or will not recover from this illness? 5. Have you had any close encounters with death? 6. What effect has your illness had on your family and close friends, and how have they reacted to it?
Stephen R. Connor
7
Downloaded by [University of New England] at 05:30 10 November 2017
7. How do you feel about the way in which your family and friends have been effected by or have reacted to your illness? 8. Is there anything good that has come out of your having cancer? 9. Are there any other thoughts or feelings about your illness or the questions I've asked that you'd like to talk more about? Sample A sample size of 24 was determined sufficient to test interper sonal denial. Because only two subjects met the criteria for intra psychic denial, it was estimated that enrollment by about 300 sub jects would have been necessary to find an adequate number to test for it. Given the limited time allotted for the study, this test was deemed infeasible. Of the 60 subjects referred to the study during the three years of data collection, 10 declined to participate, 18 died before comple tion of the study, and 6 did not use denial sufficiently to qualify. A total of 26 patients randomly assigned to either the experimental or the control group completed the pretest and the posttest. Of the 26 only 2 met the criteria for intrapsychic denial and since we had deemed it infeasible to test for that, we excluded them from the study. We thus had a final sample of 24 who met the criteria for interpersonal denial-13 experimental and 11 control. Nineteen were women and 5 were men; their mean age was 61; range 35-80. All were from the middle socioeconomic class and had at least a high school education. Disease sites varied: 9 (38%) had breast cancer, 8 (33%) had gastrointestinal malignancies, 6 (25%) had lung can cer, and 1 (4%) had metastatic melanoma. The following results are based on the data on these subjects. RESULTS Standard t-tests were used to determine whether social desirabili ty or trait anxiety influenced results on denial scores. A chi-squared tabulation was done to determine sex bias in the two groups. Be-
Downloaded by [University of New England] at 05:30 10 November 2017
8
THE HOSPICE JOURNAL
cause of the small number of men in the study (5), the Fisher exact test was also used. Results of both the SDS and the MAS showed no statistically significant differences between experimental and control subjects. Neither social desirability nor trait anxiety seemed to influence test results in these groups; nor were statistically significant differences found between the groups according to sex. The DAS was analyzed using a two-factor repeated-measures analysis of variance. The two factors were group (experimental vs. control) and time (pretest vs. posttest). Neither the interaction effect nor the main effect for group difference was statistically significant. Although a statistically significant overall decrease was found in DAS scores between pretest and posttest periods for all subjects (p < .05), the death anxiety decrease in the experimental group was not statistically significant. To control for pretest scale differences, an analysis of covariance was also conducted. Posttest DAS group means did not differ from the pretest. All group means were within one standard deviation of the means reported by McMordie. Table I shows the F-test results of the DAS scores (see also Figure 1). For the DMI measure, the main effects of group and time did not yield statistical significance. However, the interaction test showed a statistically significant tendency for greater group differ ences at posttest than at pretest [F( l, 22) = 9.782; p < .005]. Table 2 shows the F-test results of the DMI pretest and posttest (see also Figure 2). Denial means for both groups were nearly identical at pretest. At posttest, the experimental group mean de creased, whereas the control group mean had increased. Test results indicated that, as predicted, interpersonal deniers who received the intervention scored statistically significantly lower at posttest on the DMI Reversal Scale and within the normal range on the DAS. DISCUSSION Results of this study would seem to indicate that, in the absence of psychological dysfunction, terminally ill patients using denial
Stephen R. Connor
9
Table 1 Pretest and Posttest ANOVA for Death Anxiety
Downloaded by [University of New England] at 05:30 10 November 2017
Scale Scores in 24 Terminally Ill Patients
Source
DF
SS(H)
Mes
.r
Between subjects Group Error 1
23 1 22
3932.81 27.57 3905.24
27.57 17.7.51
.16
Within subjects Time Group X time Error 2
24 1 1 22
823.50 130.02 45.20 648.28
130.02 45.20 29.47
4.41* 1.53
\
* 12
