Using the data from the New York University Women\'s Health study (1985), descri
ID: 122006 • Letter: U
Question
Using the data from the New York University Women's Health study (1985), describe the public health problem in terms of magnitude, person, time and place. Discuss the basic epidemiology measures used in the study. Specifically what measure of risk was used? How was risk calculated? (Hint: Consider a 2 by 2 table to display the data). Do you agree with the authors' conclusions? What would you infer from the data? Why? see data below: The New York University Women’s Health Study enrolled 14,274 women ages 35 to 65 who attended a mammography breast-screening clinic in New York City between 1985 and 1991. A total of 7054 women (49.4% of the cohort) were postmenopausal at enrollment. The cohort has been described in detail previously (5, 6). Women who were pregnant or using exogenous hormones (oral contraceptives or HRT) within the 6 months prior to enrollment were not eligible for inclusion. At enrollment, participants answered a self-administered questionnaire with questions about their medical and reproductive history, use of medications, and demographic characteristics. Participants donated 30mL of peripheral venous blood at enrollment, and aliquots of serum were stored at 80°C. Cases of incident malignant disease are identified through self-administered questionnaires mailed to participants every 2 to 4 years, with telephone follow-up of non-respondents. Medical records are retrieved for reported malignancies and reviewed to confirm colorectal cancer diagnoses. Active follow-up is supplemented by record linkages to state cancer registries in New York, New Jersey, and Florida. Ascertainment of vital status and cause of death is accomplished through record linkage with the National Death Index. A capture-recapture analysis estimated the cancer ascertainment rate to be 95% in the NYUWHS cohort (7).
Explanation / Answer
Basic epidemiology measures used in the study
1. Women were postmenopausal at the time of enrollment.
2. Pregnant women or those using exogenous hormones (oral contraceptives or HRT) within the 6 months prior to enrollment were not included.
3. Incidents of malignant disease were identified through self-administered questionnaires mailed to participants every 2 to 4 years, with telephone follow-up of non-respondents.
4. Medical records were retrieved pertaining to reported malignancies and reviewed for confirmation of colorectal cancer diagnoses and then active follow-up is supplemented by record linkages to state cancer registries in New York, New Jersey, and Florida.
5. Exclusion of Women who had a previous cancer but non-melanoma skin cancer.
Calculation of risk
Smoking at enrollment
Non-smokers (N, %)
Current smokers (N, %)
Former smokers (N, %)
59 (44.4)
21 (15.8)
53 (39.8)
118 (45.6)
46 (17.8)
95 (36.7)
Alcohol intake
Non-drinker (N, %)
Less than 7 drinks/week (N, %)
7 or more drinks/week (N, %)
80 (60.2)
32 (24.1)
21 (15.8)
183 (69.3)
59 (22.3)
22 (8.3)
No vigorous exercise (N, %)
11 met hours/week (N, %)
>11 met hours/week (N, %)
< 2.5 met hours/week (N, %)
< 7.5 met hours/week (N, %)
7.5 met hours/week (N, %)
91 (61.5)
20 (13.5)
37 (25.0)
54 (41.5)
35 (26.9)
41 (31.5)
183 (62.5)
54 (18.4)
56 (19.1)
92 (34.6)
91 (34.2)
83 (31.2)
MET= Metabolic equivalents, it has been concluded that P-value for differences between cases and controls <0.05 for ever use of aspirin (p=0.002), family history of colorectal cancer (p=0.003), alcohol intake (p=0.02), and estrone (p=0.02).
There was a positive corelation betweeen levels of estrone and risk of colorectal cancer. BMI acts as an antecedent to estrone on the causal pathway to colorectal cancer. Biases are common in retrospective studies and this study is no exception in this regard.So,these biases might have affected the measurement of these covariates in participants who had already been diagnosed with disease when they completed follow-up questionnaires. However, risk estimates remain unaffected by the inclusion of the potential confounders as covariates in the regression models except for BMI, which was calculated at enrollment for all participants.
Reference
Clendenen,T,V.,Koenig,K,L.,Shore,R,E.,Levitz,M.,Arslan,A,A.,& Jacquotte,A,Z.(2009). Postmenopausal Levels of Endogenous Sex Hormones and Risk of Colorectal Cancer. Cancer Epidemiol Biomarkers, 18(1), 275–281.
Characteristics Cases (n=148) Controls (n=293) Age (years) at enrollment, median (10th–90th percentiles) 60.4 (53.7–64.3) 60.4 (53.8–64.2) Age (years) at diagnosis, median (10th–90th percentiles) 69.9 (61.2–77.7)Body mass index (kg/m2), median (10th–90th percentiles) 25.6 (21.1–33.3) 24.7 (20.3–31.6) Age (years) at menopause, median (10th–90th percentiles) 50 (42–54) 51 (41–55) Ever use of aspirin 3 or more times/week for 6 months or longer (N, %) 22 (16.9) 76 (28.6) First-degree family history of colorectal cancer (N, %) 33 (30.3) 41 (16.5) Ever use of hormone replacement therapy (N, %) 30 (22.7) 77 (28.1)
Smoking at enrollment
Non-smokers (N, %)
Current smokers (N, %)
Former smokers (N, %)
59 (44.4)
21 (15.8)
53 (39.8)
118 (45.6)
46 (17.8)
95 (36.7)
Alcohol intake
Non-drinker (N, %)
Less than 7 drinks/week (N, %)
7 or more drinks/week (N, %)
80 (60.2)
32 (24.1)
21 (15.8)
183 (69.3)
59 (22.3)
22 (8.3)
Estrone (pg/mL), median (10th–90th percentiles) 21 (11–41) 18 (10–33) Estradiol (pg/mL), median, (10th–90th percentiles) 7 (4–16) 7 (4–16) Physical activity Vigorous exercise at baseline (MET5.5)No vigorous exercise (N, %)
11 met hours/week (N, %)
>11 met hours/week (N, %)
Light activity at baseline (MET<5.5)< 2.5 met hours/week (N, %)
< 7.5 met hours/week (N, %)
7.5 met hours/week (N, %)
91 (61.5)
20 (13.5)
37 (25.0)
54 (41.5)
35 (26.9)
41 (31.5)
183 (62.5)
54 (18.4)
56 (19.1)
92 (34.6)
91 (34.2)
83 (31.2)
Related Questions
Navigate
Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.