SEARCH

SEARCH BY CITATION

Keywords:

  • case-control study;
  • metabolic equivalent tasks;
  • ovarian cancer;
  • physical activity;
  • risk factor

Abstract

  1. Top of page
  2. Abstract
  3. MATERIAL AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES
  8. APPENDIX

A case-control study was conducted in China during 1999–2000 to investigate the effects of intensity and duration of physical activity on the risk of epithelial ovarian cancer. Cases were 254 patients with histologically confirmed epithelial ovarian cancer. The 652 controls comprised 340 hospital visitors, 261 non-neoplasm hospital outpatients and 51 women recruited from the community. Physical activity was measured by a validated questionnaire. The risks of ovarian cancer were assessed using multivariate logistic regression analysis accounting for age, demographic, lifestyle and familial factors, hormonal status, family ovarian cancer history and total energy intake. The study found that increasing total physical activity was associated with a lower ovarian cancer risk among Chinese women. The odds ratio was 0.54 (95% CI 0.34–0.87) for high vs. low levels of total weekly metabolic equivalent tasks. Ovarian cancer risk tended to decline with increasing duration of strenuous sports and frequency of activity-induced sweating among pre-menopausal women, with adjusted OR 0.13 (95% CI 0.03–0.64) and 0.45 (95% CI 0.24–0.85), respectively. Increasing duration of moderate activity in post-menopausal women also appeared to be protective against ovarian cancer, with adjusted OR 0.36 (95% CI 0.18–0.73). © 2003 Wiley-Liss, Inc.

Ovarian cancer is the 7th most common cancer in women and the leading cause of mortality among gynaecological malignancies.1 There are considerable variations in the incidence of ovarian cancer. For example, the incidence in Europe is 13.92 per 100,000 females, whereas China has a relatively low incidence (5 per 100,000), which is only ¼ of the incidence in northern European countries.2, 3

The etiology of ovarian cancer is relatively unknown. There is some evidence that physical activity may be associated with a reduced risk of estrogen-dependent cancers, including ovarian cancer,4 breast cancer5 and endometrial cancer.6 However, the relationship between physical activity and ovarian cancer has been inconsistent. Three case-control studies recently reported that physical activity has a protective effect on the disease,4, 7, 8 whereas 2 other studies found that occupations requiring low energy expenditure can increase the risk of ovarian cancer, although the evidence obtained did not reach levels of statistical significance.9, 10 A Finnish retrospective cohort study that compared physical education teachers with language teachers found no significant difference in ovarian cancer risk between the 2 groups.11 Furthermore, 2 cohort studies observed that women reporting high levels of recreational activity were at greater risk of ovarian cancer compared to inactive women.12, 13 Despite these inconsistencies in the literature, it is still suggested that the type, intensity, duration and frequency of physical activity can affect the risk of ovarian cancer.4 Because of the complication in physical activity exposure, its assessment has posed a challenge in investigating the association between physical activity and chronic diseases.6

The physical activity exposures of women living in developed and developing countries are quite different. The geographical variations in incidence provide an opportunity to study ovarian cancer etiology. To examine the effects of intensity and duration of physical activity on epithelial ovarian cancer, which accounts for more than 90% of all ovarian malignancies,14 a case-control study was conducted in Hangzhou, China.

MATERIAL AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIAL AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES
  8. APPENDIX

Research setting

A hospital based case-control study was conducted in Hangzhou, the capital city of Zhejiang province, between July 1999 and June 2000. Most cases and controls were recruited at the Women's Hospital, School of Medicine, Zhejiang University and Zhejiang Cancer Hospital. The remaining cases were identified in 9 other general hospitals, but their pathological diagnoses had been reviewed and confirmed by pathologists of the former 2 hospitals. There was no significant difference in pathological diagnosis (serous cell vs. other types) between the 2 teaching hospitals and the 9 community hospitals. Both hospitals are public teaching hospitals with over 500 beds and receive patients from all over the province. In 1999, there were about 12,000 inpatients with 8,000 operations performed and 450,000 outpatients at the Women's Hospital.15

Case definition and identification

Cases were identified from the medical records in the hospitals. Inclusion criteria for cases were defined to be women under 75 years of age, who had been residents of Zhejiang province for at least 10 years and who had been histopathologically diagnosed with epithelial ovarian cancer within the past 3 years. To ensure complete ascertainment of cases, all medical records and laboratory pathology reports were reviewed during the study period. Pathological diagnoses were based on the International Histological Classification of Ovarian Tumors recommended by the IFGO.14 A total of 255 patients with epithelial ovarian cancer were identified and only 1 patient did not participate in the study (nonresponse rate 0.04%). Most of the cases (75.2%) were recent patients recruited within 12 months from diagnosis. In terms of interview type, 241 cases (95%) were interviewed in person, whereas 13 patients (5%) were interviewed via their next of kin. For these patients, 226 women (89%) were either inpatients or at hospitals for follow-up treatment, 25 women (10%) were interviewed at their homes, while 3 women (1%) completed their interviews at their work places. Details of cases recruited are provided in Table I.

Table I. Distribution of Diagnosis and Recruitment of Cases
 Frequency (%)
Pathological diagnosis of epithelial ovarian cancer 
 Serous cystadnocarcinoma106 (41.7)
 Mucinous cystadnocarcinoma35 (13.8)
 Endometrioid cystadnocarcinoma21 (8.3)
 Mixed epithelial cystadnocarcinoma7 (2.8)
 Undifferentiated carcinoma36 (14.2)
 Borderline malignancy40 (15.7)
 Clear cell carcinoma5 (2.0)
 Transitional cell carcinoma2 (0.8)
 Malignant Brenner's tumor2 (0.8)
Months between diagnosis and interview 
 1–12191 (75.2)
 13–2434 (13.4)
 25–3629 (11.4)
Type of interview 
 In person241 (95)
 Next of kin13 (5)
Place of interview 
 Hospital226 (89)
 Home25 (10)
 Work place3 (1)

Control definition and recruitment

During the same period 652 controls were recruited and interviewed. The criteria for controls were women not to have a neoplasm, bilateral oophorectomy, nor to have been on long-term physical activity limitation. Women recruited for controls were matched with cases by age and geographical area. A total of 601 women were recruited in the Women's Hospital. This hospital-based control sample consisted of 340 hospital visitors (15 women declined to be interviewed, nonresponse rate 4.4%) and 261 outpatients (nonresponse rate 1.2%). The hospital visitors were recruited while they visited their relatives or friends having deliveries, when they came for their own family planning or when they came for a routine medical examination. All outpatients were recruited after they had consulted their doctors and their diagnoses were confirmed as minor gynecological diseases (84.7% had vulvitis, vaginitis or cervicitis, 6.5% diagnosed with urethritis and 5.7% had menopausal symptoms). The 51 community women (nonresponse rate 7.8%) were interviewed at community council buildings, their homes or work places.

To control for bias in selecting the hospital controls, consulting rooms for outpatients and ward numbers for healthy women were chosen using random numbers. If no suitable subjects were found in the chosen room/ward, the adjacent room/ward was used instead. This systematic selection process was adopted throughout the entire recruitment period. The sample of 51 community women was recruited by curbside sampling from 9 different districts of Hangzhou with assistance from local community councils. The project was approved by the Hangzhou hospital authorities and the Human Research Ethics Committee of the researchers' institution.

Assessment of physical activity

The intensity and duration of physical activity were assessed by an interviewer-administered questionnaire. The questionnaire was designed to measure household, occupational, and recreational activities. Participants were asked to report their weekly average time spent in 3 intensity activity, namely, strenuous sports (such as jogging, cycling up hill, tennis, racquetball, lap swimming or aerobics), vigorous work (such as moving heavy furniture, shoveling, lifting objects, loading/unloading trucks or equivalent manual labor) and moderate activity (such as housework, brisk walking, bowling, cycling on level ground, gardening or Taichi). The time spent in 3 intensity activity was categorized into the following categories: never, 1/2 to 1 hr, 2–3 hr, 4–6 hr, 7–10 hr, 11–20 hr, 21–30 hr and ≥ 31 hr per week. In addition, individuals reported the frequency of activity-induced sweating, ranging from 0 to 7 times or more per week. A “reference” recall period was set at the year that was 5 years prior to diagnosis for cases or 5 years prior to interview for controls. Although calendars (e.g., 1994 or 1995) were used to help the participants to recall their physical activity exposure 5 years ago, they were blinded to the purpose of the study. Specific questions on physical activity are provided in the Appendix.

In addition to the section on physical activity, the structured questionnaire contained questions on demographic characteristics, lifestyle, usual diet, reproductive history, factors relevant to hormonal status and family history of cancer. Anthropometric information on height and weight (5 years ago) were also collected.16 The interviews usually took between 40 and 50 min. They were undertaken by the first author after obtaining written consent from the participants.

Validity and reliability of physical activity questionnaire

The specific questions used to assess physical activity was taken from the Hawaii Cancer Research Survey,17 and such questions were also used in the Australian Health Survey 1995.18 Both questionnaires have been validated in studies of large multi-ethnic populations including Chinese immigrants whom are comparable to our study population of Chinese women.

The questionnaire was pre-tested on 51 Chinese women recently migrated to Perth, Australia, to assess its feasibility, face and content validity. The participants in the preliminary study were requested to report their physical activity back in China. The participants could classify their activity into 3 intensity levels and estimate their time spent at each type of activity without difficulty. Their positive feedback confirmed that the instrument was appropriate to measure physical activity among Chinese women.

A separate test-retest study was then conducted to assess the reproducibility of the questionnaire. Forty-one women in Hangzhou were interviewed twice within 11 weeks (SD = 6 weeks). Results of the test-retest are given in Table II. No significant difference was found in mean time spent at the 3 intensity activities and activity-induced sweating between the 2 interviews. The high intraclass correlations shown in Table II further supported the reproducibility of the questionnaire.

Table II. Results of the 41 Chinese Women in the Test-Retest
Physical activityFirst interview Mean (SD)Second interview Mean (SD)Intraclass correlationp-value
Strenuous sports (hrs/week)0.7 (3.9)0.5 (2.4)0.93<0.001
Vigorous work (hrs/week)2.5 (7.2)3.4 (9.5)0.90<0.001
Moderate activity (hrs/week)21.2 (9.2)18.5 (9.2)0.66<0.001
Activity-induced sweating (times/week)0.6 (1.6)0.8 (1.9)0.600.002

Data analysis

All data were checked for completeness at the end of each interview. The data were coded and analyzed using the SPSS package.19 To assess the differences between the control groups, physical activity, demographic and lifestyle variables were initially compared between the 2 hospital control groups and then compared between hospital and community controls. The consistency of exposure levels among the 3 groups would suggest that hospital bias was minimal. To assess potential survival bias, data for the 127 patients (interviewed within 3 months from diagnosis) and data for all cases (interviewed within 3 years from diagnosis) were analyzed separately.

Demographic characteristics and potential risk factors between cases and controls were compared using t test for continuous variables and χ2 test for categorical variables. The overall activity level was expressed in terms of metabolic equivalent tasks (MET). Firstly, MET scores 7.5, 6, and 4.5 were assigned respectively for strenuous sports, vigorous work and moderate activity based on a compendium of physical activities.20 The categorical variables of activity duration were converted into continuous measures, e.g., 45 min for category “1/2 to 1 hr,“ and 2.5 hr for category “2–3 hr.” MET-hour per week for each intensity activity was then calculated by multiplying MET score by activity duration. Finally, the total physical activity for each individual as measured by weekly MET-hr was obtained by summing over the 3 intensity activities.

Following previous studies, duration in strenuous sports21 and activity-induced sweating22 were categorized into low, medium and high levels to facilitate statistical analysis. Vigorous work was also classified into low (0 hr/week), medium (≤ 20 hr/week) and high (> 20 hr/week) levels. Moderate activity was categorized into low (≤ 10 hr/week), medium (11–30 hr/week) and high (≥ 31 hr/week). The total physical activity measure (weekly MET-hr) was divided into tertiles according to the distribution of controls, with the lowest tertile being the reference category.

Univariate analysis was undertaken to screen potentially significant variables for subsequent multivariate analysis. Adjusted odds ratios (OR) and associated 95% confidence intervals (95% CI) of ovarian cancer risk for physical activity variables were computed using unconditional multivariate logistic regression models. Each fitted regression equation included terms for adjusting age at diagnosis, education, living area, BMI, smoking, alcohol consumption, family income, menopause status, parity, tubal ligation, oral contraceptive use, hormone replacement therapy, family history of ovarian cancer, total energy intake and interview season. These variables were included in the multivariate models because they were either established risk factors of ovarian cancer3, 4, 7 or significant confounders with physical activity according to the univariate analysis. Each quantitative physical activity variable was subjected to a linear trend test in ovarian cancer risk. The association between moderate activity and ovarian cancer was assessed after excluding women who performed strenuous sports and vigorous work. Finally, model adequacy was assessed using the Hosmer and Lemeshow goodness-of-fit statistic.

Since the biological mechanisms for ovarian cancer may differ between pre- and post-menopausal women, a separate analysis stratified by menopausal status was also conducted. Women were defined as post-menopausal if they indicated they had ceased having periods for at least 1 year at the time of interview and (for cases) had stopped them prior to the cancer surgery. Women who had had a surgical menopause at age over 50 were also assigned the post-menopausal group.

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIAL AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES
  8. APPENDIX

There were no significant differences between the 2 hospital control groups with respect to mean time spent on strenuous sports, vigorous work and moderate activity, total energy intake, demographic characteristics and other variables related to the study outcomes. Although the community controls spent less time in vigorous work, their time devoted to other activities and activity-induced sweating was comparable with the hospital controls. Data from the 3 control groups were thus combined to form a single control group in subsequent analyses. Results from the 191 recent patients and all cases were also similar, suggesting that survival bias was minimal. Therefore, we report the results of all cases below.

Table III contrasts the sample characteristics of women with and without ovarian cancer. There were no significant differences between cases and controls in age, locality (urban or rural areas), education, BMI (5 years ago), smoking, alcohol consumption, menopausal status and tubal ligation. Compared to controls, patients with epithelial ovarian cancer tended to have less oral contraceptive use and lower parities but higher family income. More of cases were unmarried and had professional occupations and apparent family susceptibility.

Table III. Selected Characteristics of Participants With and Without Ovarian Cancer1
 Cases (n = 254)Controls (n = 652)p′–value2
Number(%)Number%
  • 1

    Data missing for 2 controls on alcohol consumption, menopausal status, parity, hormone replacement therapy, oral contraceptive use, tubal ligation and ovarian cancer in first degree relatives.

  • 2

    Differences between cases and controls assessed using chi-square test for categorical variables and t-test for continuous variable.

Age at diagnosis (years): mean (SD)46.8 (12.5)48.0 (10.2)0.14
 < 4512841.421533.0 
 45–5411035.629044.5 
 55–644815.510816.6 
 ≥ 65237.4396.0 
Area of resident    0.28
 Urban14858.335454.3 
 Rural10641.729845.7 
Income in 1998 (yuan/month)    0.001
 ≤ 1,00019376.057187.6 
 1,001–2,0005019.76910.6 
 ≥ 2,001114.3121.8 
Education    0.04
 None4417.313320.4 
 Primary7328.721633.1 
 Secondary9537.523035.3 
 Tertiary4216.57311.2 
Occupation    0.003
 Laborer/farm worker6023.619429.8 
 Factory worker/machine operator9537.423836.5 
 Clerical/officer114.3213.2 
 Sales31.2172.6 
 Manager/administrator135.1436.6 
 Craftsperson20.830.5 
 Small business owner176.7568.6 
 Professionals/technical3814.9578.7 
 Housewife/others156.0233.5 
Body mass index (5 years ago, kg/m2)    0.20
 ≤ 18.5166.3588.9 
 18.5–24.519074.849676.1 
 ≥ 254818.99815.0 
Tobacco smoking    0.50
 Never24998.063497.2 
 Ever52.0182.8 
Alcohol consumption    0.23
 Never19777.645269.5 
 Ever5722.419830.5 
Marital status    0.001
 Never married72.881.2 
 Married23190.961093.6 
 Widowed, divorced, separated166.3345.2 
Menopausal status    0.67
 No15059.139460.4 
 Yes10440.925639.6 
No. of delivery of full-term pregnancy    0.03
 03513.8426.5 
 18633.922033.8 
 27730.323335.8 
 ≥ 35622.015523.8 
Hormone replacement therapy    0.15
 Never24897.664398.9 
 Ever62.471.1 
Oral contraceptive use    < 0.001
 Never19878.041664.0 
 Ever5622.023436.0 
Tubal ligation    0.28
 No18472.444768.8 
 Yes7027.620331.2 
Ovarian cancer in first degree relatives    0.001
 No24897.664999.8 
 Yes62.410.2 
Interview season    < 0.001
 Summer (June–August)6827.810716.4 
 Winter (December–February)7228.212018.4 
 Spring and Autumn (the other months)11444.042565.2 

Table IV summarizes the proportion of subjects and mean time spent at each intensity activity and activity-induced sweating by the cases and controls. Compared to the controls, fewer cases participated in the 3 intensity activities, with less time spent at each activity and less frequent activity-induced sweating.

Table IV. Proportion of Participation and Mean Time Spent on Physical Activity and Activity-Induced Sweating by Cases and Controls
Physical activityCases (n = 254)Controls (n = 652)p-value1
PercentMean time (SD)PercentMean time (SD)
  • 1

    t-test for differences in mean time between cases and controls.

All women     
 Strenuous sports (hr/week)5.90.43 (2.75)9.50.56 (2.33)0.48
 Vigorous work (hr/week)20.53.54 (8.88)30.15.63 (10.59)0.005
 Moderate activity (hr/week)98.420.81 (9.65)99.822.78 (8.38)0.002
 Activity-induced sweating (times/week)26.41.37 (2.60)40.52.17 (2.99)<0.001
Pre-menopausal women     
 Strenuous sports (hr/week)2.90.33 (2.58)12.90.78 (2.56)0.13
 Vigorous work (hr/week)19.22.79 (7.85)27.75.24 (10.40)0.03
 Moderate activity (hr/week)93.324.03 (8.02)94.124.02 (7.67)0.99
 Activity-induced sweating (times/week)27.91.38 (2.61)45.32.52 (3.14)0.001
Post-menopausal women     
 Strenuous sports (hr/week)8.00.42 (2.16)7.40.50 (2.87)0.72
 Vigorous work (hr/week)21.34.06 (9.52)31.55.90 (10.74)0.07
 Moderate activity (hr/week)76.718.57 (10.06)89.321.97 (8.72)<0.001
 Activity-induced sweating (times/week)25.31.37 (2.61)37.61.95 (2.88)0.03

Table V reports the adjusted odds ratios and associated 95% confidence intervals from separate multivariate logistic regression fits for each physical activity measure. The risk of ovarian cancer tended to decline with increasing total physical activity. The adjusted OR was 0.54 (95% CI 0.34–0.87) for high vs. low levels of total weekly MET, as well as a significant dose-response relationship. In pre-menopausal women, strenuous sports and activity-induced sweating were inversely associated with ovarian cancer risk, the respective adjusted OR being 0.13 (95% CI 0.03–0.64) and 0.45 (95% CI 0.24–0.85) for high vs. low levels. In post-menopausal women, the adjusted OR was 0.36 (95% CI 0.18-0.73) for moderate activity and the corresponding linear trend was also significant. However, no significant association was found for vigorous work. Finally, the Hosmer and Lemeshow goodness-of-fit statistic ranged between 3.91 (p-value = 0.87) and 10.15 (p-value = 0.26), indicating no lack of fit for the logistic regression models.

Table V. Adjusted Odds Ratios and Associated 95% Confidence Intervals of Epithelial Ovarian Cancer Risk for Physical Activity Among Chinese Women
Physical activityOdds ratio (95% CI)1
All womenPre-menopausal womenPost-menopausal women
  • 1

    Estimates from separate logistic regression models include terms for age (in years; continuous), living area (urban, rural), education (none, primary, secondary, tertiary), family income (in yuan; <200, ≥201–500, ≥501–1,000, ≥1,001–2,000, ≥2,001–3,000, ≥ 3,001–5,000, ≥5,001 or more), BMI (5 years ago; continuous), tobacco smoking (never, ever), alcohol drinking (never, ever), parity (full-term pregnancy; continuous), hormone replacement therapy (never, ever), oral contraceptive use (never, ever), tubal ligation (no, yes), family history of ovarian cancer (no, yes), interview season (summer, winter, spring and autumn), total energy intake (continuous) and menopausal status (no, yes; only included the models for all women).

  • 2

    Reference category.

  • 3

    Women reported having strenuous sports and vigorous work were excluded.

Total physical activity (weekly MET-hr)   
 Low (<110)21.001.001.00
 Medium (110–140)0.75 (0.52–1.09)1.10 (0.58–2.01)0.54 (0.32–0.91)
 High (>140)0.54 (0.34–0.87)0.63 (0.30–1.32)0.55 (0.29–1.05)
 χ2 for trend6.10 (p = 0.014)2.32 (p = 0.13)1.98 (p = 0.16)
Strenuous sports (hr/week)   
 Low (0)21.001.001.00
 Medium (<2)1.02 (0.29–3.60)0.49 (0.05–4.76)1.31 (0.22–7.60)
 High (≥2)0.33 (0.14–0.75)0.13 (0.03–0.64)0.42 (0.14–1.28)
 χ2 for trend0.25 (p = 0.62)2.05 (p = 0.15)0.13 (p = 0.72)
Vigorous work (hr/week)   
 Low (0)21.001.001.00
 Medium (≤20)0.88 (0.53–1.46)1.03 (0.45–2.35)0.94 (0.47–1.89)
 High (>20)0.73 (0.41–1.31)0.56 (0.20–1.56)1.02 (0.47–2.19)
 χ2 for trend1.34 (p = 0.25)1.78 (p = 0.18)0.04 (p = 0.85)
Moderate activity (hr/week)3   
 Low (≤10)21.001.001.00
 Medium (11–30)0.65 (0.35–1.20)1.81 (0.46–7.09)0.60 (0.28–1.29)
 High (≥31)0.42 (0.22–0.81)1.60 (0.39–6.53)0.26 (0.11–0.60)
 χ2 for trend8.17 (p = 0.004)0.01 (p = 0.91)10.69 (p = 0.001)
Activity-induced sweating (times/week)   
 Low (0)21.001.001.00
 Medium (1–2)0.59 (0.31–1.12)0.78 (0.31–1.99)0.49 (0.19–1.29)
 High (≥3)0.56 (0.37–0.84)0.45 (0.24–0.85)0.67 (0.38–1.18)
 χ2 for trend7.79 (p = 0.005)7.39 (p = 0.007)1.39 (p = 0.24)

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIAL AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES
  8. APPENDIX

This hospital-based case-control study investigated the associations between epithelial ovarian cancer and the intensity and duration of physical activity exposure, including exercise at home, at work and for recreation, using a validated and reliable instrument.17 Since dietary factors had little confounding influence on physical activity,23, 24 they were not included in the multivariate logistic regression models for covariates. We found that the cancer risk tended to decline with increasing total physical activity among Chinese women, supporting findings from previous case-control studies that physical activity can reduce the ovarian cancer risk.4, 7, 8 It is worth noting the different effects between pre- and post-menopausal women. Increasing duration of strenuous sports and frequency of activity-induced sweating by pre-menopausal women was inversely associated with epithelial ovarian cancer, whereas increasing duration of moderate activity by post-menopausal women appeared to be protective against the disease.

Ovarian cancer is an estrogen-dependent cancer.25, 26 There are several plausible mechanisms to explain the protective effect of physical activity against ovarian cancer. Physical activity may influence energy balance and maintain low body fat and moderate estrogen. Central obesity is associated with decreased levels of sex hormone-binding globulins, and increased levels of serum estrogen and androgens.27 Secondly, physical activity may prevent ovarian cancer by reducing the number of ovulatory cycles and subsequent diminution of lifetime exposure to endogenous estrogen. This hypothesis is supported by observational studies that physical activity can delay the onset of menses28 and amenorrhea,29 and prolong menstrual cycle and anovulatory cycles.30 The third plausible biologic mechanism is that physical activity may enhance the immune system by improving the capacity and number of natural killer cells, which can recognize and eliminate neoplastic cell.31 Finally, experimental studies have found that exercise can improve the innate immune mechanisms.32

Several issues should be considered when interpreting the findings. In our study, questions on regular physical activity related to a single reference time, since assessment of lifetime physical activity was impractical. Moreover, it has been reported that physical activity patterns later in life are quite similar to those earlier in life,33 so that measuring physical activity level at a single reference time can reflect long-term physical activity.34 Although physical activity levels generally decrease with age, as the study population was quite young (the majority of women at reference time of 5 years ago were premenopausal), their regular physical activity would be expected to maintain the same level during the premenopausal period. As shown in Table IV, the women have similar physical activity patterns pre- and post-menopausal. Therefore, the findings are expected to hold for older women.

As far as potential sources of biases are concerned, selection bias appeared to be minimal in view of the low refusal rate. The majority of cases were recently diagnosed, while the recruitment and identification procedures ensured that ascertainment of cases was maximized and complete. Despite some differences in interview season between the cases and controls, there were no significant differences in physical activity by seasons among the study participants. Recall bias was minimized by reporting regular physical activity and by using a “reference” recall period. Given the lack of knowledge by the participants on the association between physical activity and ovarian cancer and that they were blinded to the purpose of the study, information bias in their responses is unlikely. Finally, although the small number of community controls spent less time in vigorous work than the hospital controls, their exclusion did not influence the findings of our study.

Acknowledgements

  1. Top of page
  2. Abstract
  3. MATERIAL AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES
  8. APPENDIX

The authors acknowledge with gratitude the participation of patients and controls in Hangzhou. We are grateful for the collaboration received from the participating hospitals and their staff. In particular, we would like to thank Professor Xie Xing and Chief Pathologist Zhao Cheng Luo of Women's Hospital, School of Medicine, Zhejiang University, for their kind assistance. We are indebted to 2 anonymous reviewers for their constructive comments and suggestions.

REFERENCES

  1. Top of page
  2. Abstract
  3. MATERIAL AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES
  8. APPENDIX
  • 1
    Greenlee RT, Murray T, Bolden S, Wingo PA. Cancer statistics: 2000. CA Cancer J Clin 2000; 50: 733.
  • 2
    WHO. Cancer Epidemiology Databases. France: International Agency for Research on Cancer, 2001. (web site: http://www.iarc.fr).
  • 3
    World Cancer Research Fund. Food, nutrition and the prevention of cancer: a global perspective. Washington: World Cancer Research Fund/American Institute for Cancer Research, 1997. 28893p
  • 4
    Bertone ER, Newcomb PA, Willett WC, Stampfer MJ, Egan KM. Recreational physical activity and ovarian cancer in a population-based case-control study. Int J Cancer 2002; 99: 4316.
  • 5
    Levi F, Pasche C, Lucchini F, La Vecchia C. Occupational and leisure time physical activity and the risk of breast cancer. Eur J Cancer 1999; 35: 7758.
  • 6
    McTiernan A, Ulrich C, Slate S, Potter J. Physical activity and cancer etiology: associations and mechanisms. Cancer Causes Control 1998; 9: 487509.
  • 7
    Tavani A, Gallus S, La Vecchia C, Dal Maso L, Negri E, Pelucchi C, Montella M, Conti E, Carbone A, Francesschi S. Physical activity and risk of ovarian cancer: an Italian case-control study. Int J Cancer 2001; 91: 40711.
  • 8
    Cottreau CM, Ness RB, Kriska AM. Physical activity and reduced risk of ovarian cancer. Obstet Gynecol 2000; 96: 60914.
  • 9
    Dosemici M, Hayes RB, Vetter R, Hoover RN, Tucker M, Engin K, Unsal M, Blair A. Occupational physical activity, socioeconomic status, and risks of 15 cancer sites in Turkey. Cancer Causes Control 1993; 4: 31321.
  • 10
    Zheng W, Shu XO, McLaughlin JK, Chow WH, Gao YT, Blot WJ. Occupational physical activity and the incidence of cancer of the breast, corpus uteri, and ovary in Shanghai. Cancer 1993; 71: 36204.
  • 11
    Pukkala E, Poskiparta M, Apter D, Vihko V. Life-long physical activity and cancer risk among Finnish female teachers. Eur J Cancer Prev 1993; 2: 36976.
  • 12
    Mink PJ, Folsom AR, Sellers TA, Kushi LH. Physical activity, waist-to-hip ratio, and other risk factors for ovarian cancer: a follow-up study of older women. Epidemiology 1996; 7: 3845.
  • 13
    Bertone ER, Willett WC, Rosner BA, Hunter DJ, Fuchs CS, Speizer FE, Colditz GA, Hankinson SE. Prospective study of recreational physical activity and ovarian cancer. J Natl Cancer Inst 2001; 93: 9428.
  • 14
    CotranRS, KumarV, CollinsT, eds. Robbins pathologic basis of disease, 6th ed. Philadelphia: W. B. Saunders Company, 1999. 1067 p
  • 15
    Xie X. Annual report of Women's Hospital 1999. Hangzhou: Women's Hospital, School of Medicine, Zhejiang University, 2000 (unpublished).
  • 16
    Zhang M, Yang ZY, Binns CW, Lee AH. Diet and the risks of ovarian cancer: a case-control study in China. Br J Cancer 2002; 86: 7127.
  • 17
    Goodman MT, Hankin JH, Wilkens LR, Lyu LC, McDuffie K, Liu LQ, Kolonel LN. Diet, body size, physical activity, and the risk of endometrial cancer. Cancer Res 1997; 57: 507785.
  • 18
    Australian Bureau of Statistics. National health survey 1995: summary of results. Canberra: Australian Bureau of Statistics, 523.
  • 19
    SPSS for windows. SPSS Incorporation. Release 10. 2000.
  • 20
    Ainsworth BE, Haskell WL, Whitt MC, Irwin ML, Swartz AM, Strath SJ, O'Brien WL, Bassett DR Jr, Schmitz KH, Emplaincourt PO, Jacobs DR Jr, Leon AS. Compendium of physical activities: an update of activity codes and MET intensities. Med Sci Sports Exerc 2000; 32: S498S504.
  • 21
    Longnecker MP, De Verdier MG, Frumkin H, Carpenter C. A case-control study of physical activity in relation to risk of cancer of the right colon and rectum in men. Int J Epidemiol 1995; 24: 4250.
  • 22
    Dominguez-Berjon F, Borrell C, Nebot M, Plasencia A. Physical activity assessment in population surveys: can it really be simplified? Int J Epidemiol 1999; 28: 537.
  • 23
    McTiernan A, Stanford JL, Weiss NS, Daling JR Voigt LF. Occurrence of breast cancer in relation to recreational exercise in women age 50–64 year. Epidemiology 1996; 7: 598604.
  • 24
    Thune I, Furberg AS. Physical activity and cancer risk: dose-response and cancer, all sites and site-specific. Mid Sci Sports Exerc 2001; 33: S53050.
  • 25
    Adlercreutz H. Western diet and Western diseases: some hormonal and biochemical mechanisms and associations. Scand J Clin Lab Invest 1990; 50: 323.
  • 26
    Merzenich H, Boeing H, Wahrendorf J. Dietary fat and sports activity as determinants for age at menarche. Am J Epidemiol 1993; 138: 21724.
  • 27
    Park W, Choi JJ, Hwang ES, Lee JH. Identification of a variant estrogen receptor lacking exon 4 and its coexpression with wild-type estrogen receptor in ovarian carcinomas. Clin Cancer Res 1996; 2: 202935.
  • 28
    Kramer MM, Wells CL. Does physical activity reduce risk of estrogen-dependent cancer in women? Med Sci Sports Exerc 1996; 28: 32234.
  • 29
    Frisch RE, Gotz-Welbergen AV, McArthur JW, Albright T, Witschi J, Bullen B, Birnholz J, Reed RB, Hermann H. Delayed menarche and amenorrhea of college athletes in relation to age at onset of training. JAMA 1981; 246: 155963.
  • 30
    Bernstein L, Ross RK, Lobo RA, Hanisch R, Krailo MD, Henderson BE. The effects of moderate physical activity on menstrual cycle patterns in adolescence: implications for breast cancer prevention. Br J Cancer 1987; 55: 6815.
  • 31
    Shephard RJ, Shek PN. Heavy exercise, nutrition and immune function: is there a connection? Int J Sports Med 1995; 16: 4917.
  • 32
    Nieman DC, Pedersen BK. Exercise and immune function: recent developments. Sports Med 1999; 27: 7380.
  • 33
    Lissner L, Bengtsson C, Bjkelund C, Wedel H. Physical activity levels and changes in relation to longevity: a prospective study of Swedish women. Am J Epidemiol 1996; 143: 5462.
  • 34
    Iwai N, Hisamichi S, Hayakawa N, Inaba Y, Nagaoka T, Sugimori H, Seki N, Sakata K, Suzuki K, Tamakoshi A, Nakamura Y, Yamamoto A, et al. Validity and reliability of single-item questions about physical activity. J Epidemiol 2001; 11: 2118.