Analysis of factors associated with quality of life in breast cancer patients after surgery

Breast Cancer Vol.1 No.2 1994

Original Article Analysis of Factors Associated with Quality of Life in Breast Cancer Patients after Surgery

K O J I R O SHIMOZUMA .1, H I R O S H I S O N O 0 .1, K I Y O S H I I C H I H A R A .2, JUNICHI K U R E B A Y A S H I *~, K A Z U M A S A M I Y A K E .1, K E I I C H I Y O S H I K A WA .1, A N D K I K U K O OTA .1

The objective of this study was to investigate the factors associated with the quality of life (QOL)

in breast cancer patients after surgery. The QOL in 83 primary breast cancer patients after surgery

was prospectively assessed using a newly developed Japanese QOL questionnaire: The QOL Questionnaire for Cancer Patients Treated with Anticancer Drugs (QOL-ACD). The demographic

and medical factors relating to the QOL score in the subjects were investigated by multiple

regression analyses. Factors related to the four categories of the QOL-ACD (activity, physical

condition, psychological condition, and social relationship) were also analyzed.

The results revealed that only hospitalization had a strong negative relation to the overall QOL score. With regard to the four categories of the QOL-ACD, hospitalization had strong negative

relations to the three categories (activity, physical condition and social relationship) and had weak

negative relation to the psychological condition. Other demographic and medical factors such as types of surgery or adjuvant therapy had no significant association with the QOL score, except for

the performance status which had positive, but not significant, association with the activity.

In conclusion, this study yielded useful information concerning the management of the breast

cancer patients after surgery. We had better be more concerned with the hospitalized period than

other demographic or medical factors such as types of therapy to improve the QOL.

Breast Cancer 1:123-129, 1994

Key words: Quality of life (QOL), Breast cancer, Breast conservation, Adjuvant therapy, Multivariate analysis

The response rate (RR), disease-free survival (DFS), and overall survival (OS) can be objective-

ly assessed with relative ease, and these factors

Division of *1Endocrine Surgery and Department of *2Clinical Pathology, Kawasaki Medical School Reprint requests to Kojiro Shimozuma, Department of Endocrine Surgery, Kawasaki Medical School, 577 Matsushima, Kurashiki 701 -01, Japan

Abbreviations: ADM:Doxorubicin CPA:Cyclophosphamide DFS:Dis- ease-free survival DFUR:Doxifluridine FLIC:Functional Living Index for Cancer KPS:Karnofsky Performance Status MPA:Medroxyprogesterone acetate OS:Overall survival QOL:Quality of life QOL-ACD:QOL Questionnaire for Cancer Patients Treated with Anticancer Drugs RR:Response rate TAM:Tamoxifen UFT:Tegafur" uracil

Received March 30, 1994; accepted October 12, 1994

have been used as endpoints in determining the

efficacy of cancer treatments. However, the quality of life (QOL) in cancer patients, which should

be another important endpoint in cancer treat-

ments, is not usually assessed scientifically, since

QOL is generally thought to be related to per-

sonal or subjective attributes which are difficult

to quantitate.

Since the Karnofsky Performance Status

(KPS) ~) score was developed, various attempts 2 6~

have been made in Western countries to establish

measures representing the overall aspects of

QOL. In Japan clinicians have generally been indifferent to QOL assessment measures and only

some have used KPS and ECOG Performance

123

S h i m o z u m a K, et al Analys i s of fac tors assoc ia ted wi th the qual i ty of life in breas t cancer pa t i en t s a f te r su rge ry

Status (ECOG-PS) 7) scores, which only represent the physical aspects of the QOL. In recent years, some Japanese researchers 8-m began to evaluate QOL in cancer patients using Japanese transla- tions of measures developed in Western countries such as the Functional Living Index for Cancer (FLIC) 2) and the EORTC-QOL questionnaire%

Since those Western scales were not totally compatible with the customs and life style of Japanese patients s'~, Kurihara et al 1~ developed the first QOL assessment measure for Japanese cancer patients, "The QOL Question- naire for Cancer Patients Treated with Antican- cer Drugs (QOL-ACD, as abbreviated by us by permission of Kuriharal~ '', in Japan in 1993. Its

validity and reliability were verified during its development. The QOL-ACD was developed as a core questionniare for evaluating the QOL of cancer patinets applicable to any organs involved or any types of therapy, although it was mainly designed to reflect changes in QOL following treatment with anticancer drugs.

On the other hand, there are still many problems concerning treatment options for breast cancer, such as (1) the choice between conven- tional mastectomy and breast conserving surgery, which has been gaining world-wide accep- tance in recent years, or (2) the choice among various postsurgical adjuvant or neoadjuvant

(presurgical) therapies. When discussing those options, we should be concerned with QOL as one of the major goals. However, there have been few reports that have prospectively assessed the QOL in a scientific manner under those settings.

In this study, we applied the newly developed QOL-ACD to breast cancer patients after surgery and carried out multiple regression analyses to reveal the factors associated with the QOL of the patients among demographic and medical factors, such as the recency of surgery, the types of surgery as well as the modes of adjuvant therapy.

Patients and Methods

Recruitment and Characteristics of Patients

From February 1993 to June 1994, 83 non-recur-

rent postsurgical breast cancer patients treated by a surgery within the past 5 years at our clinic were asked to participate in this study after being

informed of its purpose by the authors. The questionnaires were filled out by the patients and returned to us. None of the patients refused to take part in this study. All the patients had already been notified of the cancer, and had been well informed of the treatment they received and of their prognoses. The religious backgrounds of the subjects were obscure and not take into account, because the majority of Japanese do not consider themselves to be religious.

The demographic and medical factors possibly associated with the QOL in the 83 patients are listed in Table 1. The clinical stage of each case was determined by UICC TNM classifications. All the subjects underwent resection of their primary lesion and all the patients who received breast conserving surgery had adjuvant radiotherapy to the residual glands.

T a b l e 1. T h e D e m o g r a p h i c a n d M e d i c a l F a c t o r s P o s s i b l y

A f f e c t i n g t h e Q O L in B r e a s t C a n c e r P a t i e n t s a f t e r Sur-

g e r y

Factor No. of cases Median (Range)

Recency of surgery (M) 83 6 (0-52)

Factor No. of cases Mean_+SD

Age 83 52.6 + 10.9 Body weight (kg) 7P > 55.6+8.7

Factor No. of cases Scores b~

Stage Tis 5 0 I 33 1 II 36 2 IIl 9 3

Performance status (PS) 0 77 1 1 6 0

Hospitalization Not hospitalized 67 0 Hospitalized 16 1

Types of surgery None 0 0 Lumpectomy 3 1 Quadrantectomy 16 2 Auchincloss 19 3 Patey 45 4

Types of adjuvant therapy r

Chemotherapy No/Yes dl 34 / 49 0 / 1

Endocrine therapy No/Yes dl 48 / 35 0 / 1

Radiotherapy No/Yes el 78 / 5 0 / 1

a~The body weight was not measured in the remaining 12 cases. U~The score set for each independent variable for the use of multiple regression analyses. ~ of adjuvant therapy given within one month prior to the questionnaire. d~Twenty three chemoendocrine therapy cases were included. e~Two chemotherapy cases and one endocrine therapy case were included.

124


Types of adjuvant therapy adopted in this analysis were those given within one month prior to the questionnaire. Forty-nine cases received chemotherapy and 35 received endocrine therapy, among which 23 received chemoendocrine therapy. Five cases received radiotherapy, which included two chemotherapy cases and one endocrine therapy case. As for orally administered drugs, tamoxifen (TAM) and medroxyprogesterone acetate (MPA) (only in one case) were used for endocrine therapy, and cyclophosphamide (CPA), doxifluridine (DFUR) and tegafur'uracil (UFT) for chemotherapy. Three cases were exceptionally given intravenous chemotherapy in the form of sequential CAF therapy (CPA + doxorubicin (ADM)+5FU) (two of them in com- bination with TAM, all t h ree cases w e r e in stage II l).

Questionnaire The English edition of the QOL-ACD used in

this study as a measure of QOL is shown in Fig 1. It is a patient-rated QOL evaluation system

developed for Japanese cancer patients by a scientific group called "The Research Group For

Scientific Assessment Measures of Treatments with Anticancer Drugs" established by the Minis- try of Health and Welfare in Japan (April 1989 to March 1991, director: M. Kurihara), and by "The Research Group for Scientific Assessment of Drug Treatment" (April 1991 to March 1993, director: M. Kurihara).

The following is a brief description of the QOL-ACD. It consists of 22 questions, which are divided into five categories: (1) six questions to evaluate activity, (2) five for physical condition, (3) five for mental or psychological condition, (4) five for social relationship, and (5) one for overall aspects of QOL as represented by a face scale, which includes five different face figures selected from the 20 original faces by Lorish and Maisia TM. Patients are instructed to answer all questions by circling the number that best represents their state. The score for each question is summed to give an overall score, the minimum being 22 and the maximum being 110. A higher score represents a higher QOL. The mean scores were also calculated for the four categories with the exception of that evaluating the global QOL by the face scale. Each ranged from one to five.

N A M E : DATE : / /

AGE : _ _ SEX : 1. MALE 2. FEMALE B. W, : Kg

This questionnaire is being done to help us understand your present state,

Please read the questions and circle the number above the scale which best describes your state dur in 9 the past several days.

( Y o u r pr ivacy wi l l be str ict ly protected and there wi l l be no negat ive inf luence on your medical t reatment, so please answer these questions freely w i t hou t concern.)

(during the past several days) 1. Have you been able to perform

normal daily l ife activities ?

2. Have you been able to go ou t by yoursel f ?

3. Have you been able to go for a 30 m inu te wa lk ?

4. Has it been hard for you to walk even a short distance?

5, Have you been able to go up and d o w n the stairs ?

6. Have you been able to take a bath by yoursel f ?

7. H o w wel l have you fel t ?

8. Have you had a good appetite?

9. Have you enjoyed your mea ls?

10. Have you experienced vom i t i ng?

I 1. Have you lost we igh t ?

12. Have you slept wel l ?

2

Completely Unable

2

Completely Unable

2

Completely Unable

4

Not At All

2

Completely Unable

2

Completely Unable

I 2

Extremely Poor

I 2

Extremely Poor

2

Not At All

5 4

Not At All

5 4

Not At All

I 2 Not At All

Completely Able

Completely Able

Completely Able

Completely Able

Completely Able

5

Extremely Well

5

Extremely Good

5

Great Deal

1

Very Often

2 1

A Great Deal

4 5

Extremely Well

(during the past several days) 13. Have you been absorbed in anything ?

14. Have you coped wel l w i t h everyday stress ?

15. Do you feel unable to concentrate ?

16.

17.

18.

19.

20.

21.

22.

Are you encouraged by anyone or anything ( fami ly. friends, re g on, hobb es. e tc , )?

2 4

Not At All A Great Deal

2 4

Not At All Extremely Well

4 2 1


2 4


Have you fel t ill at ease about the condition of your disease?

Have you felt unwi l l ing to meet anyone other than your f am i l y?

H o w much trouble do you feel you give your fami ly o v e r your medical t rea tment?

Are you anxious about your social life in the f u tu re?

Are you worr ied about the medical costs of your t reatment ?

4


4


4

None A Great Deal

4


4


Please circle the number of the face expressing your state dur ing the past several days,

5 4 3 2 1

Yr A t last. please confirm that you have answered all of the quest ions.

[ A co lumn fil led out by doctors or nurses]

a. 1. inpat ient 2. outpat ient e. Remarks :

b. PS:

c. B .W. : K9

d. Date : / / Recorder :

Fig 1. "The QOL Questionnaire for Cancer Patients Treated with Anticancer Drugs (QOL-ACD)" (English edition).

125

Shimozuma K, et al Analysis of factors associated with the quality of life in breast cancer patients after surgery

Data Management Questionnaires with no more than two un-

answered question among the 22 questions (just one case in this study) were judged valid. The

unanswered questions were substituted by the

mean score for the rest of the questions. By this criterion, all the 83 questionnaires were valid

(100%).

Statistical Methods The factors that were associated with the QOL

were evaluated by multiple regression analysis. The overall QOL score was set as a dependent variable. The following factors were set as independent variables: shortness of the period of

time since surgery, age, body weight, clinical stage of the disease, PS, hospitalization, types of surgery, and types of adjuvant therapy given

within one month prior to the questionnaire (Table 1). For the use of multiple regression analyses, we set the score for each independent variable as indicated in Table 1. In the analyses of the factors that were related to each of the four categories of the QOL-ACD, the mean score of each category was set as a dependent variable.

Lotus 1-2-3 (Lotus Development Co) was used for accumulation of the data and calculation of the

score. The statistical analyses, including the

multivariate analyses, were carried out by general purpose statistical software, Stat Flex (View Flex Co, Tokyo).

Resul ts

Factors Associated with the Overall QOL Score The results of the overall QOL scores in 83

patients in relation to the period after surgery

are shown in Fig 2. Significant improvement was

observed in the mean scores from 0 to 3-6 months period after surgery (P = 0.016 by Kruskal-Wall is

test). The results of the multiple regression analyses

after a stepwise selection of the variables (Table 2) revealed that only hospitalization had a strong

negative relation to the overall QOL score (P <0. 001). The rest of the demographic and medical

factors were not significantly related to the overall QOL score. The multiple correlation coeffi-

cient (R) between the estimated overall QOL score and the observed overall QOL score was 0. 4775 (P--0.0001).

Factors Associated with the Mean Score of Each of the Four Categories of the QOL-ACD

The results of the multiple regression analyses for the four categories after stepwise selection (Table 2) were as follows: hospitalization had strongly negative relations to the three categories (activity, physical condition and social relationship) (P<0.005, 0.005, 0.001, respectively) and had a weakly negative relation to the psychological condition (P < 0.05). The rest of the demographic and medical factors had no significant

association with the mean score of each of the

four categories, although PS had a positive, but not significant, marginally significant associa-

tion with the activity (P--0.0576). The final multiple correlation coefficients (R) after stepwise selection were 0.5638 (P -- 0.0000), 0.3933 (P=0.0039), 0.3317 (P=0.0260), and 0.4570 (P=0 .

0003), respectively.

Discuss ion

Currently we have a host of strategies for

cancer treatment, among them, surgery, medical

o 100

O 75

ID > 0

50

! I 8 88 8 �9

�9 �9

9 �9 �9

8

1"2 3"6 7-'12 13"24 25"36 37"60 Months after Surgery

Fig 2. The overall QOL scores in 83 patients in relation to the period after surgery. The bar represents the mean of the score. Significant improvement was observed in the mean scores from 0 to 3-6 months period after surgery (P=0.016 by Kruskal-Wallis test). The mean score in 0 month period after surgery was significantly lower than those in 3 6, 7-12, 13-24, and 25-36 months period (P<0. 01, 0.01, 0.005, and 0.05, respectively, by Mann-Whitney test).

1 2 6 - -


Table 2. Multiple Regression Analysis of Factors Associated with the QOL Score in Breast Cancer Patients after Surgery

Overall QOL

Factor Regression Standard t value P value coefficient error

1. Hospitalization - 15.490 3.48710 4.442 0.0000 d) 2. Chemotherapy 2.1824 2.72151 0.802 0.4250 3. Age 88.199 0.12685 0.551 0.5829

Re1=0.4775 (P 0.0001).

Activity


1. Hospitalization -0.6880 0.21817 3.154 0.0023 c) 2. Performance status (PS) 0.6456 0.33501 1.927 0.0576 a) 3. Stage 0.1776 0.11046 1.608 0.1120 4. Types of surgery 0.1476 0.09604 1.537 0.1284

R 0.5638 (P=0.0000).

Physical Condition


1. Hospitalization -0.4209 0.13951 3.017 0.0034 c) 2. Stage 0.1027 0.07737 1.328 0.1881 3. Chemotherapy 0.1533 0.11975 1.280 0.2042

R 0.3933 (P-0.0039).

Psycological Condition


1. Hospitalization -0.6808 0.26139 2.605 0.011@ ) 2. Performance status (PS) -0.6355 0.39975 1.590 0.1159 3. Types of surgery 0.1129 0.09517 1.186 0.2391

R 0.3317 (P-0.0260).

Social Relationship


1. Hospitalization - 1.2146 0.31853 3.813 0.0003 e) 2. Age 0.0116 9.549E-3 1.216 0.2274 3. Performance status (PS) -0.5729 0.47380 1.209 0.2302

R = 0.4570 (P - 0.0003).

a>Marginally significant; b~p<0.05; c~p<0.005; d)p<0.001; e~Multiple correlation coefficient. All the results of the multiple regression analyses were those after stepwise selection of the variables.

t r e a t m e n t wi th an t i cance r drugs, and i r rad ia t ion .

H o w e v e r , these s t r a t eg ies do not necessa r i ly

i m p r o v e the s t a tus of cance r pa t i en t s as a whole,

even if DFS or OS m a y be prolonged. Po ten t i a l

benef i t s ob ta ined by their p ro longa t i on mus t be

b a l a n c e d aga ins t the d i s a d v a n t a g e s or tox ic i ty of

the t r ea tmen t .

W e bel ieve the QOL in cance r pa t i en t s should

be assessed in a scientif ic m a n n e r for the follow-

ing four r ea sons in addi t ion to the a b o v e reasons:

First , the QOL score can p rov ide objec t ive infor-

m a t i o n for cl inicians and pa t i en t s as to the supe-

r io r i ty of va r i ous t r e a t m e n t options, espec ia l ly

when they w a v e r in choos ing a m o n g opt ions

whose RR, DFS, or OS are known to be s imi lar .

Second, the QOL score can be used for judging

the overa l l e f f i cacy of a d rug to be licensed, by

p rov id ing a useful m e a s u r e as one of the end-

points of a phase study. Third , the QOL score can

be used to p red ic t the fu ture course of the QOL in

the da i ly ca re of pat ients , when a p rospec t ive

s tudy revea l s po ten t ia l f ac to r s a s soc ia t ed wi th

127

Shimozuma K, et al Analysis of factors associated with the quality of life in breast cancer patients after surgery

the QOL. Many measurements have already been devel-

oped for the evaluation of the QOL, in Western countries 1-6), some of which are specifically for cancer patients. Those QOL measures were developed and codified after questionnaire items were gathered from cancer patients, doctors, nurses, and psychologists. The validity and/or reliability were verified by factor analysis. How- ever, there has been no "gold standard" for the measurement of the QOL 14), and there are a number of problems with the QOL measurements currently in use, including: (1) the notification of cancer, (2) the patient's age, (3) the validity of applying a single QOL measure to people in different languages and cultures, (4) the choice between a patient-rated and a professional-rated questionnaire. First of all, most measures from Western countries were devised on the premise of notification of cancer. In Japan, however, most cancer patients are usually not notified, although 88% of experimentally notified patients with breast cancer approved of the notificationl~L Secondly, a positive relationship between age and the QOL has been reported la-~9/, and some age adjustment may be required in the use of QOL assessment scores. Thirdly, in a field study using the EORTC core questionnaire, Aaronson et al 2~ revealed that there was difference in the psychometric properties of the questionnaire between European and Japanese people. In our experience using a Japanese version of the FLIC score, the linear analogue scale for each question was occasionally misunderstood by Japanese patients; ie the percentage of patients who could correctly answer the questionnaire was 83.1% 9). Ikeda TM also pointed out that a questionnaire with a categorical scale was much easier to answer for Japanese patients. Therefore, the original linear analog scale of FLIC was modified in the QOL-ACD. As for the fourth problem, we believe a majority of investigators currently agree that a patient-rated questionnaire is better than a doctor-, nurse- or other professional-rated questionnaire 22).

The QOL-ACD was developed to eliminate the various above problems. As expected, all the eligible patients accepted to participate in our study and all of the 83 questionnaires were valid.

The results demonstrated that the QOL-ACD was easier for Japanese breast cancer patients to fill out than the Japanese version of FLIC.

As for the relults of the multiple regression analyses in this study, it is noteworthy that hospitalization had stronger association with the QOL score than any other demograhic or medical factors, although PS had a marginally significant association with the activity. It is still a matter of controversy TM among investigators as to which types of surgery; ie, breast conservation or mastectomy, or which types of adjuvant therapy is superior in terms of the patient's QOL. The results of this study failed to reveal significant relations between the patient's QOL and types of therapy, probably because of the insufficient number of cases studied. Furthermore, we are currently trying to investigate the superiority of the treatment modes by applying the QOL-ACD in a variety of clinical settings and patient status.

The current result may have been as expected since the QOL-ACD are known to be sensitive to the status of hospitalization and PS as proved during the development and verification of the measurel~ In any case, we believe the results of this study will help clinicians to assess the QOL of breast cancer patients after surgery.

The prediction of the QOL in breast cancer patients after surgery may not be possible from various clinical factors, since the multiple correlation coefficient between the estimated overall QOL score and the observed overall QOL score was only 0.4775. This may indicate insufficient performance of QOL-ACD as a QOL measure, but we considered that the low predictability was mainly due to that the general condition of the subjects examined were not so severe. Actually, the QOL of most patients had not been com- promised very much at the start of treatment (data not shown). Therefore, the potential degree of improvement in the QOL score was minimal. In other words, if we had examined the QOL in severer cases, we probably would have observed prominent changes in the score and obtained a higher predictable equation of the QOL score. Accordingly, we are planning to apply the QOL score to more advanced or recurrent cases.

In conclusion, this study yielded useful information concerning the management of the breast

' 128


cance r pa t i en t s a f t e r surgery . W e had be t t e r be

m o r e concerned with the hosp i ta l ized per iod than

o ther d e m o g r a p h i c or medica l f ac to r s such as

types of t he r a py to i m p r o v e the QOL.

Acknowledgment

We wish to express our gratitude to Prof M. Kurihara of Showa University, Prof S. Arita of Kansai Medical University, and Emeritus Prof K. Sarai of Hiroshima University for their guidance.

References

1) Karnofsky DA, Burchenal JH: The clinical evaluation of chemotherapeutic agents in cancer. In: MacCleod CM ed, Evaluation of Chemotherapeutic Agents, Columbia University Press, New York, pp191-205, 1949.

2) Schipper H, Clinch J, McMurray A, et al: Measuring the quality of life of cancer patients; The functional living index-cancer; Development and validation. J Clin Oncol 2:472-483, 1984.

3) Spitzer WO, Dobson AK, Hall J, et al: Measuring the quality of life of cancer patients. J Chron Dis 34:585- 597, 1981.

4) Selby PJ, Chapman JA, Etazadi-Amoli J, et al: The development of a method for assessing the quality of life of cancer patients. Br J Cancer 50:13-22, 1984.

5) Aaronson NK, Bullinger M, Ahmedzai S: A modular approach to quality-of-life assessment in cancer clinical trials. Recent Results Cancer Res 111:231-249, 1988.

6) Levine MN, Guyatt GH, Gent M, et al: Quality of life in stage II breast cancer; An instrument for clinical trials. J Clin Oncol 6:1798-1810, 1988.

7) Cella DF, Cherin EA: Measuring quality of life in patients with cancer. In: Proceedings of The Fifth National Conference on Human Values and Cancer, Am Cancer Soc, New York, pp23-31, 1987.

8) Kurihara M, Fukuyama E, Nakano S, et al: Evalua- tion of solid cancer patients' QOL in randomized controlled trials. Karkinos 4:551 - 559, 1991 (in Japanese).

9) Shimozuma K, Tominaga T, Hayashi K, et al: Evalu- ation of quality of life in breast cancer patients. J Jpn Soc Cancer Ther 26:1504-1510, 1991 (in Japanese with English summary).

10) Kurihara M, Shimizu H, Tsuboi K, et al: Assessment

of quality of life in protocols for cancer therapy. CRC

1:174-181, 1992 (in Japanese with English summary). 11) Eguchi K, Fukutani M, Kanazawa M, et al: Feasibil-

ity study on quality-of-life questionnaires for patients with advanced lung cancer. Jpn J Clin Oncol 22:185- 193, 1992.

12) Kurihara M, Eguchi K, Shimozuma K, et al: The QOL questionnaire for cancer patients treated with anticancer drugs. J Jpn Soc Cancer Ther 28:1140 1144, 1993 (in Japanese).

13) Lorish CD, Maisia K: The face scale; A brief, nonver- bal method for assessing patient mood. Arthritis Rheum 29:906-909, 1986.

14) Spitzer WO: State of science 1986; Quality of life and functional status as target variables for research. J Chron Dis 40:465 471, 1987.

15) Shimozuma K, Tominaga T, Hayashi K, et al: Notifi- cation of cancer in breast cancer patients. Jpn J Cancer Chemother 18:2147-2153, 1991 (in Japanese with English summary).

16) Cassileth BR, Lusk EJ, Strouse TB, et al: Psychosocial status in chronic illness; A comparative analysis of six diagnostic groups. N Engl J Med 311: 506 511, 1984.

17) Ganz PA, Schag CC, Heinrich RL: The psychosocial impact of cancer on the elderly; A comparison with younger patients. J Am Geratr Soc 33:429-435, 1985.

18) Maisiak R, Gains R, Lee E, et al: The psychosocial support status of elderly cancer outpatients. In: Eng- strom PF, Anderson PN, Mortenson LF eds, Advances in Cancer Control, Research and Develop- ment, Alan R Liss, New York, pp395-403, 1983.

19) Nerenz DR, Love RR, Leventhal H, et al: Psychosocial consequences of cancer chemotherapy for elderly patients. Health Serv Res 20:961 976, 1986.

20) Aaronson NK, Ahmedzai S, Bullinger M, et al: The EORTC core quality-of-life questionnaire; Interim results of an international field study. In: Osoba D ed, Effect of Cancer on Quality of Life, CRC Press, Boca Raton, pp185-203, 1991.

21) Ikeda T: Quality of life in breast cancer. Jpn J Breast Cancer 5:215-226, 1990 (in Japanese with English summary).

22) Ganz PA, Schag AC, Cheng H, et al: Assessing the quality of life; A study in newly-diagnosed breast cancer patients. J Clin Epidemiol 43:75-86, 1989.

23) Kiebert GM, de Haes JCJM, van de Velde CJH: The impact of breast-conserving treatment and mastectomy on the quality of life of early-stage breast cancer patients; A review. J Clin Oncol 19:1059-1070, 1991.

129

Documents

Analysis of factors associated with quality of life in breast cancer patients after surgery