Descriptivelongitudinal Study

1
Descriptive-longitudinal Study

• Colleen So
• Aaron Bryks
• Maile Capistrano
• Crystal Fong

2
Physical Activity and Lung Cancer Risk in Male
Smokers
3
Authors Affiliations

• Center for Cancer Research, National Cancer
  Institute, Bethesda, MD, USA
• Lisa H. Colbert, Joseph A. Tangrea, Philip R.
  Taylor
• Department of Nutrition, Pennsylvania State
  University, University Park, PA, USA
• Terryl J. Hartman
• Department of Epidemiology and Health Promotion,
  National Public Health Institute, Helsinki,
  Finland
• Pirjo Pietinen, Jarmo Virtamo
• Division of Cancer Epidemiology and Genetics,
  National Cancer Institute, Bethesda, MD, USA
• Demetrius Albanes

4
Grant Sponsor

• National Cancer Institute, USA

5
Purpose

• Was to examine the association between physical
  activity (occupational and leisure-time) and lung
  cancer risk in a large group (27,087) of Finnish
  white male smokers, ages 50-69 over a period of
  10 years.

6
Type of Research

• Prospective study
• extrapolates to unknown outcomes
• slow process
• Longitudinal
• assesses changes in behavior in one group of
  subjects at more than one point in time
• e.g. over 10 year period
• subjects act as their own controls
• intra-individual variability is minimized

7
Type of Research

• Descriptive
• reveals current patterns associated with a
  phenomenon (e.g. lung cancer) and how they relate
  to other factors
• gives researchers something to build on for
  future research
• does not include treatment or control group
• does not attempt to determine causal relationships

8
Research Methods

• Study population white male smokers enrolled in
  the Alpha-Tocopherol Beta-Carotene Cancer
  Prevention (ATBC) study
• living in the southwestern Finland area
• all subjects volunteered
• smoked 5 or more cigarettes per day
• between ages 50-69
• subjects recruited from 1985-1988

9
Research Methods

• Cases of lung cancer were identified through the
  Finnish Cancer registry
• Examined the associations between occupational
  and leisure time physical activities and lung
  cancer
• age
• smoking characteristics years of smoking, number
  of cigarettes smoked daily

10
Research Methods

• Collected baseline data
• general medical history questionnaire
• blood sample
• Other question included in the questionnaire
• working environment
• -risk factor to lung cancer

11
Research Methods

• Crudely classified them as occupationally
  exposed
• asbestos
• other causes of lung cancer (e.g. mining and
  quarrying)
• Completed a self-administered food-use
  questionnaire
• report usual frequency of consumption and portion
  size of more than 276 food items, mixed dishes
  and beverages during previous 12 months

12
Research Methods

• Two questions to assess usual occupational and
  leisure-time physical activity
• 1st Question Describe your activity during work
  in the past year as

13
Research Methods

• 2nd Question
• Describe your usual
• leisure-time activity in
• the past year as

14
Research Method

• Sedentary men used as the reference group
• Non-workers kept as a separate occupational
  activity category
• includes retirees
• Regular and heavy leisure-time physical activity
  combined to create an active leisure group

15
Research Methods

• Cox proportional hazards models were used to
  estimate the relative risk (RR) and 95
  confidence intervals (CI) of lung cancer
  associated with level of physical activity
• All models were run excluding the first 2 years
  of follow-up
• no significant changes in any of the estimates

16
Research Methods

• Validity of the proportional hazards assumption
  was checked by examining the cross-product terms
  of follow-up time and the covariates of interest
• no departures from this assumption for any
  covariate in the final models

17
Main Findings

• No inverse association between occupational and/
  or leisure-time physical activity and risk of
  lung cancer in long-term cigarette smokers
• may be due to length of time smoking
• any beneficial effect of occupational activity
  may be outweighed by the related increase in risk
  due to exposure to asbestos and other high-risk
  exposures

18
Main Findings

• Only age appeared to modify the relationship
  between leisure-time activity and lung cancer
• among younger men, approximately 20-25 of active
  men at lower risk
• may be a protective relationship for leisure
  physical activity among younger smokers
• no risk reduction among older men
• may be due to the accumulation of more
  carcinogenic insults
• slight increase of physical activity may go
  unnoticed

19
Main Findings

• Neither age nor amount of smoking modified the
  occupational physical activity association

20
Main Findings

• Non-workers appear to be at increased risk of
  lung cancer
• Non-workers tend to be
• older
• smoked longer
• less educated
• eat fewer vegetables
• exposed to higher levels of carcinogens
  (asbestos)

21
Main Findings

• Men in moderate and heavy occupational activity
  categories, like non-working men, at increased
  risk to lung cancer
• less education, higher exposure to carcinogens
  (e.g. asbestos)
• perhaps non-working category may largely consist
  of men retired from occupations requiring
  moderate to heavy physical activity

22
Main Findings

• Men who participated in physical activity during
  their leisure time
• smoke fewer cigarettes/day
• inhale smoke less often
• consume foods of greater dietary energy sources
  and vegetables

23
Weaknesses of Research Study

• Sample is not generalizable to the entire
  population
• reliability
• if we were to conduct a similar study in the
  future, would we end up with the same results?
• Homogeneous population (target sample)
• not representative of entire population
• population differences are controlled
• amount of tobacco exposure
• white males between ages of 50-69 from
  southwestern Finland

24
Weaknesses of Research Study

• Subjects derived from a previous study
• no link between study objectives
• subjects volunteered (convenience sampling)
• selection bias

25
Weaknesses of Research Study

• Threats to internal validity
• possible history threat
• an uncontrolled event that influences the
  participants testing outcomes over the extended
  period of time
• selection bias
• volunteers (convenience sampling)
• whoever was available from the ATBC study
• mortality
• potential for high drop-out rates
• changing the characteristics and qualities of the
  original sample

26
Weaknesses of Research Study

• Threats to internal validity (cont.)
• instrumentation
• studied white, long-term male smokers
• used only one question each to measure usual
  occupational and leisure-time physical activity
  in the past year
• questions are insufficient
• many things can happen over a year
• limited assessment (not specific enough)

27
Weaknesses of Research Study

• Threats to external validity
• reactive arrangement
• social desirability
• questionnaire participants answer how they think
  you want them to answer
• Does not evaluate long-term effects of asbestos
  exposure
• non-working category (I.e. retirees)
• Does not specify how often testing occurs over 10
  year period

28
Strengths of Research Study

• Attempted to control for the influences of
  education, food consumption (in particular
  vegetables), and asbestos and other high-risk
  carcinogens
• may still be residual confounding effects from
  these and/or other factors in our risk estimates
• Controlled the type of people who were at the
  extremes of the population

29
Strengths of Research Study

• Prospective nature

30
Considerations for Future Research

• Use a more diverse population
• careful considerations of smoking status
• smokers and/or non-smokers
• wider range of ages
• different ethnicities and gender
• Question more extensively to determine the level
  of physical activity at different ages associated
  with lung cancer
• frequency, duration, type, and intensity of
  exercise

31
Research Questions

• Prospective research addresses changes that occur
  over time
• e.g. "violence in children and television"
• Descriptive research (e.g. studying a different
  culture) does not have any treatments
• observational
• no manipulation
• observes natural progression that occurs within
  an environment

32
Reference

• Colbert, L.H., Hartman, T.J., Tangrea, J.A.,
• Pietinen, P., Virtamo, J., Taylor P.R.,
• Albanes, D., (2002). Physical Activity and
• Lung Cancer Risk in Male Smokers.
• Int. J. Cancer, 98, 770-773.

33
Questions?