Lecture 8 Cancer Epidemiology Writing assignment due on Friday 10 2 Chimney Sweeps and Cancer Chimney sweeps coal tar adenocarcinoma What is epidemiology defined populations Epidemiology is the study of patterns causes and effects of health and disease in Looks for causes and risk factors of disease in order to propose factors of disease in order to proper effective strategies for disease prevention and control Using the scientific method Epidemiology Research Populations physician what treatment is appropriate Or politics How healthy is the population How can health in the population be maintained How effective are preventative measures Or Individual Which risks can I avoid How do I stay healthy Or healthcare Which treatment is reasonable How good is the quality of treatment The Epidemiological Approach This approach attempts to Determine if an observed difference is a real difference If it is then it tries to determine why there is a difference Gather epidemiological evidence Form hypothesis Test hypothesis Statistics support hypothesis make recommendations for changes that will protect the public Stats do not support either revise hypothesis or do not Multiple analysis many generate conflicting reports distrust and skepticism An Example of an Epidemiological Study Observations Pollution from chemical factory makes chemtown stink Hypothesis Chemical pollutants in the air cause people to develop bladder cancer 100 cases of bladder cancer in cleantown 200 cases of bladder cancer chemtown at a rate in chemtown How do we Evaluate Evidence P value statistical test Estimates probability that an observed difference would appear by chance even if no difference really exists Is the cancer rate really higher in the chemtown vs cleantown Calculating a p value will tell us if a real difference in the rate of cancer development exists or if it just looks like it does really just random Statistics significance Likelyhood that observed difference is real Ex P value 0 05 means 5 chance is false and 95 is real Cancer is Chemtown vs Cleantown People of chemtown want to know if they are at a higher risk for developing cancer than people that live in cleantown just because they live in chemtown Epidemiologists decide to evaluate chemtown and gather epidemiology evidence They are trying to determine if the number of cancer cases that have occurred in chemtown is significantly greater than the expected of cases given the age gender and racial distribution of the group of people at risk of developin the disease They then calculate the expanded number of cases and compare that number with the observed number of cases If they calculated numbers are different then they compare the numbers to determine if the difference is statistically significantly This generates a p value Bias can cause Conflicting Results Possible reasons why difference exist between investigated group results Experimenter bias Detection bias Selection bias Recall bias Publication bias The Effect of Bias Experimenter preconceptions influence results Detection methods are unequal Selection nonrandom volunteering Recall self reporting errors Publication relationship publish and no relationship no publish So stories 19 shows no relationship while 1 shows a relationship misleading is the one published Epidemiologic data is Age Adjusted Age is a confounding factor and must be accounted for in comparisons of groups Types of Epidemiology Observational Observational is like the cleantown vs chemtown example Experimental Human equivalent of animal testing providing or withholding a substance or determine its toxic or beneficial effects Greatly limited by ethical and legal considerations as well as difficulties securing the cooperation of a large number of people Descriptive Analyze data on the distribution and extent of health problems in Attempts to find correlation among characteristics such as diet air quality and Comparisons are frequently made between countries and smaller geographic various populations occupation regions Attempts to uncover and portray the occurrence of the condition or problem Limitations of Epidemiology Epidemiological studies can never prove causation it cannot prove that specific risk factor actually causes the disease being studied Epidemiology evidence can only show that a risk factor is associated correlated with a higher incidence of disease in the population The strength of an epidemiology study depends on the number of cases and controls included in the study p value gets smaller relationship less likely due to chance The Epidemiological Triad Host Organism In cancer it is the person or human Agent Can be genetic composition of the host infectious agent or environmental factor Environment Place in which the agents and hosts interact Main Causes of Human Cancer Age At 60 your risk is 1000x greater than at 40 Environment Radiation viral exposure chemicals Lifestyle Tobacco use diet alcohol consumption Hereditary what do you get from your parents and grandparents The Ames Test for Potential Carcinogens Liver homogenate substance being tested incubate Bacteria that require histidine to grow add bacteria medium lacking histidine let bacteria grow bacteria with necalate grew more Alcohol A Promutagen with Complex Effects Activates smoke food environmental chemicals Metabolized to acetaldehyde mutagen that binds DNA Cirrhosis step to liver cancer Affects estrogen levels Epidemiological Studies The most common types Case control studies and cohort studies These are both Observational studies Experimental studies Randomized control trial done to test a preventive or therapeutic regimen or diagnostic procedure Aims to remove the effects of unknown sources of bias Case Control Study Case control studies use patients who already have a disease or other condition and look back to see if there are characteristics of these patients that differ from those who don t have diseases Start with two similar groups of people One group cases has a disease and other controls is healthy What happened in the past to give any different results Case Study Study Question What is Risk Epidemiological risk refers to a probability the chance that something may occur but not a guarantee that it will Risks are determined by studying large groups of people to discover the probability that any given person or category of people will develop the disease over a certain period of time Looks at what characteristics or behaviors are associated with
View Full Document
Unlocking...