Lecture 15 - Tobacco As A Cause Of Disease Flashcards
Which cancers does tobacco have sufficient evidence of causing?
- lung cancer
- nasal cavity and nasal sinus cancers
- oropharyngeal canceer
- laryngeal cancer
- oesophageal cancer
- stomach cancer
- cervical cancer
- liver cancer
- pancreatic cancer
- urinary tract cancer
- bladder cancer
- myeloid leukemia
Diseases that have sufficient evidence to be caused by smoking: CV disease and resp disease
1) CV disease
- abdominal aortic aneurysm
- atherosclerosis
- cerebrovascular disease
- coronary heart disease
2) Respiratory disease
- COPD
- pneumonia
- asthma
- impaired lung function
Sufficient evidence that smoking causes Repriductive diseases + other diseases
- reduced fertility
- foetal death
- pregnancy complication
- pre-term delivery
- low birthweight
- SIDS
- cataract
- low bone density
- hip fracture
- peptic ulcer disease
4 steps for establishing causality
- the exposure precedes the outcome
- the association is unlikely to be due to chance
- the association is unlikely to be due to bias
- the association is unlikely to be due to confounding
Other criteria for causality
- the association is strong
- there is a dose-response relationship
- the association is consistent
- the association is coherent with other epidemiological data
- the association is specific
- the association is biologically plausible
Evidence that smoking precedes outcome
- cigarette smoking is generally in early adult life
- lung cancer is rare under 40 years of age and peaks in frequency very late in life
- association demonstrated in a number of large, long lasting cohort studies
- lung cancer rare before 40-50
Studies: smoking and life years
- men who smoked only cigarettes and continued smoking died on average 10 years younger than lifelong smokers
- cessation at age 60, 50, 40 or 30 years gained about 3, 6, 9, 10 years of life expectancy
Why is the smoking and lung cancer association unlikely to be due to bias
- very strong association
- case-control studies and cohort studies have given similar results
What are possible confounding factors that could alter the assocation
- demographics
- occupation
- age
- pollution
- alcohol
- those are controlled by statistical models
Smoking and lung cancer : dose-response
- the longer you’ve been smoking, the higher your chances of cancer
- if you stop, kind of goes back to baseline
The associatin is coherent with other information
- prevalence of smoking in men began to rise around the turn of the century, lung cancer mortality began to rise in the 1940s
- women started smoking much later than men, lung cancer mortality rise began later
- 20 years delay between smoking and cancer
The association is biological;y plausible
- cigarette smoke contains over 200 different chemical compounds that can increase the incidence of cancer in experimental studies
- condensates of tobacco smoke painted on mouse skin cause skin cancer
- respiratory tract tumours caused in dogs, rats and hamsters forced to inhale tobacco smoke
The association is reversible
- cessation at 30 avoids almost all of adverse effects on survival compared with continuing smokers
- if you stop you can gain some of your years of life expectancy back
Attributable risk
= Iexposed - Iunexposed
Attributable fraction
= (Iexposed - Iunexposed)/ I exposed x 100
- the proportion of the disease in an exposed group that is due to exposur
- x% of lung cancer among smokers is attributable to smoking
