Japanese Journal of Clinical Oncology

¹Department of Cancer Detection, Tochigi Cancer Center, ²Department of Surgery, Tochigi Cancer Center Hospital, Utsunomiya, Tochigi, ³Department of Public Health, Saitama University, Saitama, Japan

Many psychosocial factors have been reported to influence the duration of survival of breast cancer patients. We have studied how family members, hobbies and habits of the patients may alter their psychosocial status. Female patients with surgically treated breast cancer diagnosed between 1986 and 1995 at the Tochigi Cancer Center Hospital, who provided information on the above-mentioned factors, were used. Their subsequent physical status was followed up in the outpatients clinic. The Cox regression model was used to evaluate the relationship between the results of the factors examined and the duration of the patients' survival, adjusting for the patients' age, stage of disease at diagnosis and curability, as judged by the physician in charge after the treatment. The following factors were revealed to be significant with regard to the survival of surgically treated breast cancer patients: being a widow (hazard ratio 3.29; 95% confidence interval 1.32-8.20), having a hobby (hazard ratio 0.43; 95% confidence interval 0.23-0.82), number of hobbies (hazard ratio 0.64; 95% confidence interval 0.41-1.00), number of female children (hazard ratio 0.64; 95% confidence interval 0.42-0.98), smoker (hazard ratio 2.08; 95% confidence interval 1.02-4.26) and alcohol consumption (hazard ratio 0.10; 95% confidence interval 0.01-0.72). These results suggest that psychosocial factors, including the family environment, where patients receive emotional support from their spouse and children, hobbies and the patients' habits, may influence the duration of survival in surgically treated breast cancer patients.

Key words: breast cancer patients - duration of survival - family environment - hobbies - habits

Introduction

In the past few decades, numerous studies have revealed a relationship between psychosocial factors and the survival of cancer patients. Previous clinical studies have also revealed a prognostic association between severe life stressors and recurrence of breast cancer. Ramirezet al. (1) reported that severe life-threatening events and difficulties were significantly associated with the first recurrence of breast cancer. In these severe situations, emotional support seems to be linked to the behavioral component of adjustment to cancer. In addition, it has been reported that family support is an important factor in the adjustment of patients with chronic disease (2). Supportive familial relationships are particularly important to cancer patients, because of the fear and uncertainty associated with cancer (3-6).

Previous studies have also demonstrated that being married is associated with increased survival of patients with any disease (7,8) and in cancer patients (8,9). Other evidence points to the role social support plays in alleviating the impact of stress on the individual and in facilitating adjustment when illness occurs (5,10-12). The effect of psychosocial intervention on the survival time was studied by Spiegelet al. (13) in women with metastatic breast cancer. At the 10 year follow-up, patients who had had psychosocial intervention lived on average twice as long as the control group. Involvement in satisfying work (14,15) and being religious (16) have also been found to be related to adjustment to illness.

Hobbies and habits such as alcohol consumption and smoking are considered by some people to alleviate their emotional stress. However, these two factors have rarely been studied in relation to prognostic factors for cancer patients (14,17).

The purpose of the present study was to explore the existence of relationships between selected psychosocial factors which are considered to buffer emotional stress, such as the fear and uncertainty caused by cancer and/or life-threatening events and survival of the surgically treated patients with breast cancer.

Subjects and Methods

The study population for this analysis included female patients who had undergone surgery for primary breast cancer between September 1, 1986 and January 31, 1995 at the Tochigi Cancer Center Hospital, Utsunomiya. Patients who were diagnosed with in situ carcinoma or with multiple primary cancers were excluded. This left a total study population of 398 patients with histologically confirmed breast cancer. Some of these patients were treated with radiation and/or chemotherapy in addition to the surgical treatment. Almost all patients were followed up as outpatients after treatment, so we were easily able to obtain follow-up data. From each medical record, we took the patient's age and stage of the disease at diagnosis, as well as marital status, occupation, constitution of family, religion, habits and hobbies, which were obtained at interview with the patients themselves at first admission. In addition, we obtained the records of surgical method, subsequent final curability judgment made by the doctor who had treated the patient after all treatment was completed during the first admission, outcome and causes of the patient's death if she had died during the following-up period. Some of the patients enrolled in this study formed a social circle from 1990 and held parties every 2 months. They also had occasional trips or meals with other people in the group. The purpose of the party was to facilitate mutual communication and psychological support. Interactions within the group often involved self-disclosure and sharing of mutual fears and concerns and also often fostered telephone calls among members between parties, visits to their houses and visits to members when they were hospitalized. We checked whether or not each patient enrolled in the present study joined the circle.

The relative importance of multiple prognostic factors on survival was estimated using the Cox proportional hazards analysis method (18). This data analysis was performed using the SAS procedure PHREG. A p-value of <0.05 was considered statistically significant.

Results

The study involved 398 patients. TNM staging was performed by correlating the post-operative histological findings: 116 (29.1%) patients had TNM stage I disease; 208 (52.3%) patients had stage II disease; 61 (15.3%) patients had stage III disease; and 13 (3.2%) patients had stage IV disease (Table 1). The age at diagnosis of most of the patients was from 40 to 49 years (35.7%) (Table 1). The demographic and clinical characteristics of the patients studied are shown in Table 2. Partial mastectomy was performed on 53 (13%) patients and simple mastectomy, modified radical mastectomy and radical mastectomy on 4 (1%), 226 (57%) and 115 (29%) patients, respectively. In addition to the surgical treatment, 258 patients (65%) had adjuvant chemotherapy. Hormonal therapy and radiotherapy were provided to 177 patients (44%) and 50 patients (13%), respectively. Of the patients who received adjuvant chemo-hormonal therapy, 42% were prescribed oral UFT (tegafur-uracil, 300 or 400 mg/day, 2 years) alone, 7% were administered oral tamoxifen (20 mg/day daily, 2 years) only, 15% received combination chemotherapy with or without tamoxifen, which consisted of oral cyclophosphamide (100 mg/day, from the first to the fourteenth day), 5-fluorouracil (5-FU) (500 mg/m² i.v. on the first and eighth days) and adriamycin (20 mg/m² i.v. on the first and eighth days) every 4 weeks, 8% had chemotherapy with or without tamoxifen, which consisted of oral cyclophosphamide (100 mg/day, from the first to the fourteenth day), methotrexate (40 mg/m² i.v. on the first and eighth days) and 5-FU (500 mg/m² i.v. on the first and eighth days) and 20% received chemotherapy with 5-FU or associated agents with or without tamoxifen. The other patients who received chemo-hormonal therapy were administered a combination of some of the aforementioned anticancer agents with or without adriamycin-related compounds. Combination chemotherapy was repeated every 4 weeks with a maximum of six courses. Radiotherapy was usually 50 Gy delivered for local control. During the follow-up period, contact was lost with two patients. All the remaining patients were followed up to the end of October 1995 or death. At that time, 48 patients had died of breast cancer and 348 patients were alive.

Since it is well known that there are many clinical factors that influence the prognosis of breast cancer patients, the crude effect of each factor was first evaluated. First, age, TNM stage, surgical methodology and curability according to treatment were incorporated into the model as ordinal categorical variables. Accordingly, TNM stage I, II, III and IV disease was scored as 1, 2, 3 and 4, respectively. With regard to surgical methodology, partial mastectomy, simple mastectomy, modified mastectomy and radical mastectomy were scored as 1, 2, 3 and 4, respectively. For curability according to treatment, temporary cure, non-curable and permanently curable were categorized as 1, 2 and 3, respectively. All four clinical variables, i.e. age, TNM stage, surgical method and curability according to treatment, were found to be significant predictors of survival as follows: age (hazard ratio 0.69; 95% confidence interval 0.51-0.94); TNM stage (hazard ratio 2.98; 95% confidence interval 2.07-4.29); surgical methodology (hazard ratio 2.07; 95% confidence interval 1.29-3.32); and curability according to treatment (hazard ratio 0.35; 95% confidence interval 0.21-0.56).

Table 1. Age at diagnosis and TNM classification of the 398 patients studied

Age at diagnosis (years)	TNM classification
	I	II	III	VI
<40	14	36	15	5
40-49	49	80	9	4
50-59	21	43	21	3
>= 60	32	49	16	1
Total	116	208	61	13

Table 2. Demographic and clinical characteristics of 398 patients with breast cancer

	No. of patients	Percentage
Surgical method
Partial mastectomy	53	13
Simple mastectomy	4	1
Modified radical mastectomy	226	57
Radical mastectomy	115	29
Methods of adjuvant therapy
Chemotherapy
Yes	258	65
No	140	35
Hormone therapy
Yes	177	44
No	221	56
Radiation therapy
Yes	50	13
No	347	87
Unknown	1	0
Employment
Employed	207	52
Unemployed/retired	191	48
Marital status
Married	328	82
Widowed	31	8
Divorced	17	4
Single	22	6
Participation in the meeting of patients
Yes	98	25
No	300	75
Hobby
Yes	281	71
No	107	27
Unknown	10	2
Religion
Yes	60	15
No	325	82
Unknown	13	3
Child/children
Yes	350	88
No	45	11
Unknown	3	1
Male child
Yes	267	67
No	128	32
Unknown	3	1
Female child
Yes	260	65
No	135	34
Unknown	3	1
Habit of smoking
Yes	68	17
No	325	82
Unknown	5	1
Habit of drinking alcohol
Yes	74	19
No	320	80
Unknown	4	1

Second, the above four variables were incorporated into the model as sets of dummy categorical variables. The results are shown in Table 3. For age, the hazard ratio between 40 and 49 years decreased in comparison with that for patients under 40 years. However, the hazard ratio between 50 and 59 years increased in comparison with that between 40 and 49 years and again decreased for patients aged over 60 years. With regard to TNM stage, the hazard ratio increased as the stage of the disease at diagnosis increased. For curability according to treatment, the permanently curable category had the lowest hazard ratio and appeared to show a decreasing tendency. With regard to methodology, there appeared to be no definite hazard ratio tendency. However, as the choice of surgical method seemed to be determined by TNM stage, survival analysis was performed adjusting for age, TNM stage and curability according to treatment and excluding the surgical method. Moreover, on the basis of the above analytical results, the analysis was conducted incorporating all three variables into the model as sets of dummy categorical variables. The Cox proportional hazards method was used to calculate each estimate of the instantaneous relative death rate for each psychosocial factor separately, after adjusting for the three significant clinical factors (age, TNM stage and curability according to treatment). The results of analysis are shown in Table 4. The number of hobbies and children were studied as continuous variables. Six variables, namely being a widow, having a hobby, number of hobbies, number of female children, smoking and alcohol consumption, were revealed to be significant predictors of survival.

Table 3. Clinical prognostic factors for surgically treated breast cancer patients

Variable	Hazard ratio	95% confidence interval	p-value
Age at diagnosis (years)
<40
40-49	0.30	0.14-0.65	0.003
50-59	0.63	0.30-1.30	0.208
>= 60	0.22	0.08-0.61	0.003
Stage of disease
I
II	2.45	0.83-7.20	0.104
III	7.57	2.51-22.81	<0.001
IV	23.96	6.73-85.22	<0.001
Surgical method
Partial mastectomy
Simple mastectomy	2.63	0.85-8.18	0.094
Modified mastectomy	0.94	0.50-1.76	0.838
Radical mastectomy	1.90	1.05-3.45	0.034
Curability according to treatment
Temporary cure
Non-curable	0.54	0.14-2.12	0.377
Permanently curable	0.15	0.04-0.48	0.001

Table 4. Psychosocial prognostic factors for surgically treated breast cancer patients, multivariate survival analysis: results of the Cox proportional hazards regression model

Variable	Hazard ratio*	95% confidence interval	p-value
Employment
Unemployed/retired	1
Employed	1.34	0.71-2.51	0.370
Marital status
Widowed/divorced/single	1
Married	0.64	0.33-1.25	0.190
Married/divorced/single	1
Widowed	3.29	1.32-8.20	0.011
Married/single/divorced	1
Single	1.46	0.55-3.86	0.446
Participation in the meeting of patients
No	1
Yes	1.05	0.51-2.15	0.891
Hobby
No	1
Yes	0.43	0.23-0.82	0.010
Religion
No	1
Yes	0.60	0.23-1.55	0.290
Child/children
No	1
Yes	0.54	0.24-1.19	0.127
Habit of smoking
No	1
Yes	2.08	1.02-4.26	0.044
Habit of drinking alcohol
No	1
Yes	0.10	0.01-0.72	0.023
Number of hobbies	0.64	0.41-1.00	0.048
Number of children	0.79	0.58-1.07	0.127
Number of male children	1.05	0.72-1.53	0.819
Number of female children	0.64	0.42-0.98	0.039

*Adjusted for age at diagnosis, TNM stage and curability according to treatment.

Whether or not breast cancer patients were employed was found not to be a significant prognostic factor. The risk for survival among widowed patients was significantly higher than among patients with other marital status (including all three other categories, married, divorced and single) (hazard ratio 3.70; 95% confidence interval 1.52-9.09). Only one deceased patient was divorced, so survival analysis was not performed for that category.

Participation in the social circle of breast cancer patients was not a significant prognostic factor. Patients who had a hobby proved to have longer survival times than those who had no hobbies, adjusting for age, stage of disease and curability according to treatment (hazard ratio 0.46; 95% confidence interval 0.25-0.85). Additionally, a positive relationship was found between the number of hobbies the patients had and survival time (hazard ratio 0.64; 95% confidence interval 0.42-0.98).

Religion did not have a significant impact on the survival of surgically treated breast cancer patients. The presence of a child did not influence survival. However, female, not male, children of surgically treated cancer patients had an impact on survival; as the number of the female children increased, the survival time of the patient increased (hazard ratio 0.65; 95% confidence interval 0.43-0.98).

Smoking was found to be a significant negative predictor of survival (hazard ratio 2.04; 95% confidence interval 1.01-4.17). However, drinking alcohol had a positive impact on the survival of breast cancer patients (hazard ratio 0.09; 95% confidence interval 0.01-0.68).

Discussion

Previous studies have revealed a significant association between severe life events and recurrence of breast cancer (1). Recent work has shown that social relationships and contacts can promote health and protect against disease (12). In addition, it has been shown that improving the quality of life should be a major objective in the treatment of advanced breast cancer, because there may be a causal relationship between quality of life scores and survival duration in cancer patients (19). However, hobbies, which are considered to be closely associated with quality of life, are seldom taken into account as prognostic factors (14).

In this study, we tried to elucidate the association between psychosocial factors, including hobbies and habits, which are assumed to alleviate the impact of stress on individuals and facilitate adjustment when the illness occurred and the survival of surgically treated breast cancer patients.

After adjusting for clinical factors, six variables, namely marital status, having a hobby, number of hobbies, number of female children, smoking and alcohol consumption, were revealed to be significant predictors of survival.

In a previous study, involvement in satisfying work has been found to be related to adjustment (14,15). In this study, 207 patients (52%) had occupations. However, we did not find an association between patients having an occupation and their survival. In this regard, the degree of satisfaction in their occupation might have to be taken into consideration.

In the present study, the survival time of widowed patients was shorter than for patients of other marital status. Goodwinet al. (9) demonstrated that unmarried persons with cancer have decreased overall survival after examining the survival of patients with cancer in various sites on the basis of population-based data. Others have also demonstrated the survival advantage of married persons (7,8,15,20). The widely accepted explanation as to why married people have lower mortality from cancer is that marriage provides social support, which has been postulated to buffer the effects of stressful events (21). Widowed women might have taken care of their spouse before they suffered from breast cancer and have experienced many sad events, especially the death of their husband. In addition to these sorrowful experiences, they have lost their main provider of social support. We think that a series of such stressful events and the loss of familial support might shorten the survival of the widowed women.

Social affiliation refers to the mutual dependence between people. Social contact can provide opportunities to exchange information, obtain reassurance and reduce feelings of loneliness and isolation (7,11,22). We expected that the patients who belonged to the social circle, which provided social support for isolated individuals, would have longer survival times than non-participants. Unexpectedly, our study did not demonstrate a beneficial effect of participation in the circle on survival among the breast cancer patients. We do not have a definite explanation for this finding. The presence of those patients with a short duration of participation (median duration 3 years, maximum 6 years) and those with low attendance at the parties might cause such a result. Further investigations are needed to determine whether the social contact associated with group activities can prolong the survival of breast cancer patients.

In the present study, patients who had a hobby lived longer than those who had no hobbies and, as the number of hobbies increased, the risk of death decreased, after adjusting for age, stage of disease and curability according to treatment. Hobbies such as painting, composing haiku poems, dancing, keeping a garden beautiful, etc., may afford patients strength to live and mitigate the disease-induced stress in accordance with the improvement of quality of life. In addition, the affiliative networks which were made through hobbies might provide opportunities for social contact, reducing feelings of loneliness and isolation. These emotional factors might have significant effects on adjustment and result in longer survival. In a previous study, spending more time on leisure activities appeared to be related to breast cancer patients experiencing less distress, but this result did not reach statistical significance (14). In a Tecumseh community health study, the predictive effect of leisure activities on mortalities, which involved socializing with other people, either friends and family (e.g. pleasure drives and picnics) or others (e.g. cultural and sports events, classes and lectures) proved to be consequential (10); however, passive/solitary leisure activities such as reading, listening to the radio and watching television were associated with an increase, rather than a decrease, in mortality. In this regard, Houseet al. (10) suggested that to have beneficial effects a leisure activity or relationship must involve greater active effort by the individual and some contact with other people. Therefore, the importance of hobbies as a prognostic factor in cancer patients might have to be further analyzed considering whether the hobbies are active or passive.

Religions such as Buddhism and related religions and Christianity did not have a significant effect on survival in our study. Being religious has been reported to be related to adjustment (16). However, adjustment to the illness may depend on the deepness of the faith. In this context, we agree with Carey's opinion that the quality of religious orientation, rather than mere religious affiliation or verbal acceptance of religious beliefs, is the most important religious variable. Further investigation is needed to elucidate the association between religion and survival of cancer patients.

Having a confidant may mitigate the effects of a traumatic event or difficulty and prevent the onset of psychiatric disturbance. Among other things, emotional support is often best provided by the family. Many previous studies have suggested that adjustment to cancer is better in a close family environment where feelings are openly expressed and there is an absence of family conflict (3,4,6,23). We demonstrated that in spite of a lack of effect on survival by male children, female children have an impact on the survival of breast cancer patients; as the number of female children increased, the survival time of the patients increased. The causal aspects of the relationship between female children and survival of the patients are not clearly defined. Recent findings offer support for the hypothesis that it is important to the physical health of cancer patients that they express their emotions and receive support from others (24). The results of the present study suggest that a female child being of the same gender as the patient might be an important factor as a receiver of her mother's emotional expression and as a supporter of her mind. In this context, we would like to stress the family environment as a significant predictor of breast cancer outcomes.

Smoking was found to be a significant negative predictor of survival. This is consistent with a previous report. Scanlonet al. (17) examined the association between cigarette smoking and the development of lung metastases in a group of 835 breast cancer patients. They found a significant association between a history of cigarette smoking and risk of development of lung metastases. In addition, the risk of lung metastases increased as the number of cigarettes smoked in a lifetime increased. In their opinion, the subtle changes in the lung tissue that might result from exposure to cigarette smoke could be the explanation for the predisposition to lung metastases. A history of cigarette smoking may be used as an important prognostic factor for breast cancer patients.

Unexpectedly, alcohol consumption had a positive effect on the survival of breast cancer patients. We have no definite explanation for this positive impact. However, patients who had a habit of drinking in the present study generally drank a small to moderate amount of alcohol and usually drank three or four times a week. A small to moderate amount of alcohol might settle the uneasy mind and might improve the patient's adjustment to cancer.

In this study, we have explored the significance of psychosocial factors, focusing merely on the family environment, hobbies, habits and several other factors on the survival of breast cancer patients. However, the potential contribution of psychosocial factors to cancer survival has attracted great attention and an increasing number of studies have been reported (25). On the basis of the results obtained, at least four broad categories of psychosocial factors have been proposed as being linked to the survival of cancer patients: (1) adjustment to illness; (2) emotional expression (fighters vs compliers); (3) will to live; and (4) emotional stress (25). The prognostic factors we have proposed seem to have some direct or indirect relationship to the above-mentioned psychological factors.

Recent studies have examined the possible effects that immunological and central nervous system-mediated mechanisms might have on the tumor-host relationship (26). Several investigators have shown that marital disruption is associated with measurable abnormality of immune function (27,28). Levyet al. (26) demonstrated that natural killer (NK) cell activity was a strong predictor of disease outcome. Higher NK activity at follow-up predicted non-recurrence of early-stage breast cancer during the follow-up period (26). In this context, the prognostic factors we have proposed could be associated with an alteration of immune function (29), and might contribute to the outcome variance of breast cancer patients.

Socioeconomic status has also been shown to have a relationship with cancer survival (30). Cellaet al. (30) reported that the survival of patients with lower annual incomes and educational level was significantly shorter than those with higher income or education, respectively. Since we have no data about the socioeconomic status of our patients, we are unable to address the relationship between these factors and the length of survival of breast cancer patients and cannot rule out the possible influence of socioeconomic status on our results, which may have some relationship with the variables we propose.

Jamisonet al. (25) followed patients with metastatic breast cancer and assessed the significance of various psychogenic variables on their longevity. Their results revealed no significant relationship between psychosocial factors and length of survival and suggested that, for breast cancer patients with metastatic disease, disease-related variables probably outweigh the influence of psychosocial factors. In our study, patients with stage II disease comprised about half of all the patients and the proportion of stage IV patients was small (3.2%). It should be noted that in advanced stage, disease-related factors might outweigh the influence of psychosocial factors in determining the length of survival.

In addition, we obtained the necessary data from patients' medical records only at the time of their admission. Subsequently, we neglected changes in the constitution of each patient's family, hobbies and habits during the follow-up period.

Overall, we demonstrated that having female children, hobbies and a habit of drinking small to moderate amounts of alcohol were positive predictors whereas being widowed and smoking were negative predictors for survival of surgically treated breast cancer patients after adjusting for age, TNM stage and curability according to treatment. Each of these factors could be used as a psychosocial predictor of survival. However, further investigations enrolling a larger number of patients are needed to determine the impact, if any, of these factors on survival and to elucidate the mechanism underlying their influence.

References

2. Lithman TJ. The family as a basic unit on health and medical care: a social-behavioral overview. Soc Sci Med 1974;8:495-519.

3. Friedman LC, Baer PE, Nelson DV, Lane M, Smith FE, Dworkin, RJ. Women with breast cancer: perception of family functioning and adjustment to illness. Psychosom Med 1988;50:529-40. MEDLINE Abstract

4. Spiegel D, Bloom J, Gotteil E. Family environment as a predictor of adjustment to metastatic breast carcinoma. J Psychosoc Oncol 1983;1:33-4.

5. Peters-Golden H. Breast cancer: varied perception of social support in the illness experience. Soc Sci Med 1982;16:483-91.

6. Singer BA. Psychosocial trauma, defense strategies and treatment consideration in cancer patients and their families. Am J Fam Ther 1983;11:15-21.

7. Berkman LF, Syme SL. Social networks, host resistance and mortality: a nine year follow-up study of Alameda County residence. Am J Epidemiol 1979;109:186-204. MEDLINE Abstract

8. Klerman GL, Izen JE. The effects of bereavement and grief on physical health and general well-being. Adv Psychosom Med 1977;9:63-104.

9. Goodwin J, Hunt WC, Key CR, Samet JM. The effect of marital status on stage, treatment and survival of cancer patients. J Am Med Assoc 1987;258:3120-5.

10. House JS, Robbins C, Metzner HL. The association of social relationship and activities with mortality: prospective evidence from the TECUMSEH community health study. Am J Epidemiol 1982;116:123-40. MEDLINE Abstract

11. Feflic M, Goldman A, Kennnedy BJ. Group counseling in adult patients with advanced cancer. Cancer 1979;43:760-6.

12. House JS, Landis KR, Umberson D. Social relationships and health. Science 1988;241:540-5. MEDLINE Abstract

13. Spiegel D, Kraemer HC, Bloom JR, Gottheil E. Effects of psychosocial treatment on survival of patients with metastatic breast cancer. Lancet 1989;2:888-91. MEDLINE Abstract

14. Bloom JR. Social support, accommodation to stress and adjustment to breast cancer. Soc Sci Med 1982;16:1329-38.

15. Waxler-Morrison N, Hislop TG, Mears BKL. Effect of social relationships on survival for women with breast cancer: a prospective study. Soc Sci Med 1991;33:177-83.

16. Carey R. Living until death. Hosp Prog 1974;55:82-7. MEDLINE Abstract

17. Scanlon EF, Suh O, Murthy SM, Mettlin C, Reid SE, Cummings KM. Influence of smoking on the development of lung metastases from breast cancer. Cancer 1995;75:2693-9. MEDLINE Abstract

18. Cox DR. Regression models and life tables (with discussion). J R Stat Soc B 1972;34:187-220.

19. Coates A, Gebski V, Signorini D, Murray P, McNeil D, Byrne M, et al. Prognostic value of quality-of-life scores during chemotherapy of advanced breast cancer. J Clin Oncol 1992;12:1833-8.

20. Lillienfeld AM, Levin ML, Kessler II. Cancer in the United States. Cambridge, MA: Harvard University Press, 1972.

21. Kennedy S, Kiecolt-Glaser JK, Glaser R. Immunological consequences of acute and chronic stressors: mediating role of interpersonal relationships. Br J Med Psychol 1988;61:77-85. MEDLINE Abstract

22. Wood PE, Milligan M, Christ D, Liff D. Group counseling for cancer patients in a community hospital. Psychosomatics 1978;9:555-61.

23. Grandstaff NW. The impact of breast cancer on the family. Front Radiat Ther Onc 1976;11:146-56.

24. Spiegel D. Psychological distress and disease course for women with breast cancer: one answer, many questions. J Natl Cancer Inst 1996;88:629-31. MEDLINE Abstract

25. Jamison N, Burich TG, Wallston KA. Psychogenic factors in predicting survival of breast cancer patients. J Clin Oncol 1987;5:768-72.

26. Levy SM, Herberman RB, Lippman M, D'Angelo T, Lee J. Immunological and psychosocial predictors of disease recurrence in patients with early-stage breast cancer. Behav Med 1991;17:67-75. MEDLINE Abstract

27. Bartrop RW, Luckhurst E, Lazarus L, Kiloh LG, Penny R. Depressed lymphocyte function after bereavement. Lancet 1977;16:834-6.

28. Kiecolt-Glaser JK, Fisher LD, Ogrocki P, Stout JC, Speicher CE, Glaser R. Marital quality, marital disruption and immune function. Psychosom Med 1987;49:13-33. MEDLINE Abstract

29. Levy S, Herberman R, Lippman M, D'Angelo T. Correlation of stress factors with sustained depression of natural killer cell activity and predicted prognosis in patients with breast cancer. J Clin Oncol 1987;5:348-53.

30. Cella DF, Orav J, Kornblith AB, Holland JC, Silberfarb PM, Lee KW, et al. Socioeconomic status and cancer survival. J Clin Oncol 1991;19:1500-9.

Received March 13, 1997; accepted August 7, 1997
For reprints and all correspondence: Keigo Tominaga, Department of Cancer Detection, Tochigi Cancer Center, 4-9-13 Yohnan, Utsunomiya, Tochigi-ken 320, Japan
Abbreviations: 5-FU, NK

CiteULike

Connotea

Del.icio.us

This article has been cited by other articles:

				S. D. Pressman, K. A. Matthews, S. Cohen, L. M. Martire, M. Scheier, A. Baum, and R. Schulz Association of Enjoyable Leisure Activities With Psychological and Physical Well-Being Psychosom Med, September 1, 2009; 71(7): 725 - 732. [Abstract] [Full Text] [PDF]

				C. M. Tammemagi, C. Neslund-Dudas, M. Simoff, and P. Kvale Smoking and Lung Cancer Survival: The Role of Comorbidity and Treatment Chest, January 1, 2004; 125(1): 27 - 37. [Abstract] [Full Text] [PDF]

This Article Abstract Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to citation manager Search for citing articles in: ISI Web of Science (10) Request Permissions Google Scholar Articles by Tominaga, K Articles by Nagai, M Search for Related Content PubMed PubMed Citation Articles by Tominaga, K Articles by Nagai, M Social Bookmarking          What's this?