Upload
edan-pitts
View
28
Download
0
Embed Size (px)
DESCRIPTION
Objectives. To discuss threats to validity for the two articles To review for the midterm examination. Threats to Validity. Smoking prevention paper (Cameron, et al) Growth retardation paper (McDermott, et al). Figure 2 . Stroke mortality rates by birth cohort and age group in females. - PowerPoint PPT Presentation
Citation preview
Objectives
• To discuss threats to validity for the two articles
• To review for the midterm examination
Threats to Validity
• Smoking prevention paper (Cameron, et al)
• Growth retardation paper (McDermott, et al)
Figure 2. Stroke mortality rates by birth cohort and age group in females.
From: Medrano: Stroke, Volume 28(1).January 1997.40-44
Discussion• Age effect: Mortality increased markedly with age in all
cohorts for men and women alike, with rates being somewhat higher in men, especially in the younger generations. For any given age group, mortality was observed to vary with cohort, increasing in the older cohorts and decreasing in the more recent ones. The negative slope was less evident for men, the younger age groups, and the more recent cohorts.
• A period effect: was also to be observed, since mortality peaked for all age groups in consecutive cohorts, meaning that the downward shift occurred in the same period for all birth cohorts and age groups. From age and year of birth, it can be deduced that this period must have been circa 1974 (1972-1976).
Table 1. Age-specific rates (per 100000) for schizophrenia.
From: Takei: Psychol Med, Volume 26(5).September 1996.963-974
Age--period--cohort Analysis of the Incidence of Schizophrenia in Scotland Takei,
Noriyoshi; Lewis, Glyn; Sham, Pak C.; Murray, Robin M. Psychological Medicine Volume 26(5) September 1996 pp 963-974
Discussion
• Cohort effect is present
• Age effect is present
Rates
• a. incidence density
• b. cumulative incidence
• c. point prevalence
• d. period prevalence
• e. odds for the disease of interest
Study Design
• a. case-cohort• b. case-control• c. case-control study within a defined cohort• d. nested case control• e. cohort• f. cross-sectional• g. ecologic• h. experimental
Sun exposure and melanoma• A random-digit telephone survey of 580 person
was conducted to examine the association of sun exposure and skin melanoma
• Sun exposure was classified as low, medium and high using a standardized algorithm
• Persons who had skin melanoma were compared to those who did not for sun exposure?
• What type of study is this?• What other confounders should be considered?
The association between sun exposure and skin melanoma
Low exposure
Medium
Exposure
High exposure
Skin melanoma
20 20 40
No skin melanoma
200 100 100
What is the prevalence of skin melanoma?What is the prevalence of high sun exposure?
What is the odds of having med/high exposure given that you Have skin melanoma compared to low exposure?
Collapsing Categories
Skin Melanoma
No Skin Melanoma
Low exposure 20 200
Med/High exposure
60 200
ACL tear and gender
Women
(n=1000)
Men
(n=1000)
ACL tear 40 80
No ACL tear 960 920
What is the odds of being a women given you haveAn ACL tear compared to the odds of being a man?
Crude and stratified Odds Ratios
Crude ACL No ACLwoman 70 960 1000
man 80 920 1000
Plays SportsACL No ACL
woman 50 950 1000man 20 980 1000
Doesn't play sportsACL No ACL
woman 20 980 1000man 60 940 1000