There are several measures used to quantify the public health impact of exposures and diseases. Relative measures include relative risk (RR) and odds ratio (OR), which indicate strength of association but not absolute impact. Absolute measures include risk difference (RD), attributable risk (AR), and attributable risk percent (AR%), which represent the actual burden of disease preventable by reducing exposure. Population measures include population attributable risk (PAR) and population attributable risk percent (PAR%), which estimate the impact of exposure on the entire population. These measures can be calculated from cohort and population-based case-control studies but usually not regular case-control studies.
In this presentation we will make an attempt to answer the question of how far AI from revolutionising healthcare and what is the current progress in this area. I have looked into the latest groundbreaking medical innovations driven by deep learning and evaluate their potential impact on medical practice. I have discussed the main challenges that deep learning engineers face and recent advances that have been proposed in deep learning in order to address these challenges. Most of the presentation is based on the recently accepted review paper – Computational biology – deep learning by William Jones, Kaur Alasoo, Dmytro Fishman et al.
The document discusses natural and abandoned settings that are suitable for horror films. Forests, abandoned locations like bunkers and hospitals, cemeteries, carnivals, and deserts are all mentioned as atmospherically conducive to horror due to being dark, isolated, or subverting expectations of safety. The settings provide ready-made feelings of darkness, isolation, and danger that can heighten the scares in horror movies.
The document outlines 13 steps taken to design a movie poster using software. It describes selecting an image, editing it by rotating, changing colors and adding effects. Backgrounds and textures were added. The movie title, slogan, release date and production details were incorporated using different fonts and colors. Age ratings and logos were also included to complete the poster.
Scott King was born on January 3, 1953 and passed away on October 6, 2010. The document provides basic biographical information about Scott King including his birth and death dates but does not include any other details about his life or accomplishments.
This document provides a storyboard for a movie trailer consisting of 9 shots. Shot 1 shows a victim peeking through curtains as the killer appears behind her. Shot 2 shows the killer holding his weapon, implying he is chasing the victims. Shot 3 shows victims talking, with fake blood suggesting the killer already attacked one. Future shots include the victims running down stairs, being held captive by the killer in a dark room, working together to escape, peering through a door, and hiding from the killer under something. The final shot provides a close-up of the killer's face. The order of shots may change during editing.
This poster creates a sense of mystery by only showing the film's title, producers, main actors, and setting, without including other expected information like the director or age rating. It aims to intrigue viewers by featuring the antagonist alone and leaving the victims unknown. Showing the dark setting and the killer's chainsaw is meant to generate a feeling of danger and suspense regarding who will be targeted in the film.
The document outlines 13 steps taken to design a movie poster using various software tools. Key steps included selecting an image from still shots, editing it in Pixlr by rotating, changing colors, and smudging, then adding scratch textures as backgrounds. Text elements like the title, slogan, and production details were added using different fonts and colors. Rating icons and logos were also included to complete the poster.
In this presentation we will make an attempt to answer the question of how far AI from revolutionising healthcare and what is the current progress in this area. I have looked into the latest groundbreaking medical innovations driven by deep learning and evaluate their potential impact on medical practice. I have discussed the main challenges that deep learning engineers face and recent advances that have been proposed in deep learning in order to address these challenges. Most of the presentation is based on the recently accepted review paper – Computational biology – deep learning by William Jones, Kaur Alasoo, Dmytro Fishman et al.
The document discusses natural and abandoned settings that are suitable for horror films. Forests, abandoned locations like bunkers and hospitals, cemeteries, carnivals, and deserts are all mentioned as atmospherically conducive to horror due to being dark, isolated, or subverting expectations of safety. The settings provide ready-made feelings of darkness, isolation, and danger that can heighten the scares in horror movies.
The document outlines 13 steps taken to design a movie poster using software. It describes selecting an image, editing it by rotating, changing colors and adding effects. Backgrounds and textures were added. The movie title, slogan, release date and production details were incorporated using different fonts and colors. Age ratings and logos were also included to complete the poster.
Scott King was born on January 3, 1953 and passed away on October 6, 2010. The document provides basic biographical information about Scott King including his birth and death dates but does not include any other details about his life or accomplishments.
This document provides a storyboard for a movie trailer consisting of 9 shots. Shot 1 shows a victim peeking through curtains as the killer appears behind her. Shot 2 shows the killer holding his weapon, implying he is chasing the victims. Shot 3 shows victims talking, with fake blood suggesting the killer already attacked one. Future shots include the victims running down stairs, being held captive by the killer in a dark room, working together to escape, peering through a door, and hiding from the killer under something. The final shot provides a close-up of the killer's face. The order of shots may change during editing.
This poster creates a sense of mystery by only showing the film's title, producers, main actors, and setting, without including other expected information like the director or age rating. It aims to intrigue viewers by featuring the antagonist alone and leaving the victims unknown. Showing the dark setting and the killer's chainsaw is meant to generate a feeling of danger and suspense regarding who will be targeted in the film.
The document outlines 13 steps taken to design a movie poster using various software tools. Key steps included selecting an image from still shots, editing it in Pixlr by rotating, changing colors, and smudging, then adding scratch textures as backgrounds. Text elements like the title, slogan, and production details were added using different fonts and colors. Rating icons and logos were also included to complete the poster.
This document provides definitions and examples of key concepts for estimating risk from epidemiological studies, including probability, odds, relative risk, and absolute risk. It discusses how relative risk is calculated from cohort and case-control study designs. Relative risk compares the risk of an outcome between exposed and unexposed groups to determine if exposure is associated with increased risk. The odds ratio, which estimates relative risk, is presented as the measure used to assess association in case-control studies. Examples are provided to demonstrate calculating and interpreting these risk measures.
The document discusses various absolute measures used to quantify the association between an exposure and disease occurrence in a population, including attributable risk (AR), attributable risk percent (AR%), population attributable risk (PAR), and population attributable risk percent (PAR%). It provides formulas for calculating each measure and examples to illustrate their interpretation and how they depend on both the strength of the exposure-disease association and the prevalence of exposure in the population. Alternative formulas are presented that allow estimating these measures from case-control study odds ratios when only relative measures can be directly estimated.
relative risk, odds ratio, attributable fractions and lots of statistical measurements which can improve epidemiological problems have been discussed. Besides, some reliability instruments' description have been added.
R software codes for ICC, Kappa, Lin's agreement and Bland agreement is exist.
Statistics are important to understand data, summarize results, and make informed decisions based on research. Key aspects of understanding statistics include:
- Comparing results to peers to determine if an outcome is unusual or expected
- Calculating measures like mean, median, and range to summarize variables
- Understanding concepts like p-values, confidence intervals, relative risk, and number needed to treat to properly analyze and interpret data.
1) Statistics are important to understand data collected in research projects, summarize large datasets, and make evidence-based decisions.
2) Key statistical concepts include measures like relative risk, odds ratios, and confidence intervals which are used to evaluate research results and size of treatment effects.
3) Proper statistical analysis and reporting of values like p-values, absolute risk reduction, and number needed to treat are needed to accurately interpret results and clinical significance.
Cohort study design.ppt Epidemiology medicalSoravSorout
This document describes prospective and retrospective cohort studies. A prospective cohort study follows groups of exposed and unexposed individuals over time to see who develops a disease in the future. A retrospective cohort study identifies exposed and unexposed groups based on past exposure and disease occurrence data collected from medical records. Cohort studies aim to calculate the relative risk of disease between exposed and unexposed groups to determine the strength of association between an exposure and disease.
The document discusses risk assessment and various measures used to quantify risk such as relative risk, attributable risk, odds ratio, prevalence rate, and incidence rate. It provides examples of how to calculate these measures from cohort and case-control study data and interpret the results. Key points are that relative risk is used for cohort and experimental studies, odds ratio for case-control studies, and prevalence and incidence rates help measure disease burden. Attributable risk helps identify excess risk from an exposure. The examples help illustrate how to apply these concepts to public health practice.
This document describes different types of study designs used in research. It discusses observational studies, which involve analyzing exposures without intervention. Within observational studies, it distinguishes between descriptive studies, which generate hypotheses but no comparisons, and analytical studies, which attempt to establish associations. It provides details on specific observational study designs including cross-sectional studies, case-control studies, and cohort studies. Key aspects like study populations, exposures, outcomes, and advantages/disadvantages are compared across these study designs.
Common measures of association in medical research (UPDATED) 2013Pat Barlow
This is an updated version of my Common Measures of Association presentation. I've updated it to include (1) more detail on rates, risks, and proportions, (2) Absolute Risk Reduction (ARR), Attributable Risk (AR), Number Needed to Treat (NNT) and Number Needed to Harm (NNH). Feel free to email me for a full version of the slideshow.
This document provides definitions and explanations of key statistical and epidemiological concepts:
- A 95% reference interval contains the central 95% of a population distribution, calculated as the mean +/- 2 standard deviations for a normal distribution.
- Sensitivity measures the proportion of true positives detected, specificity measures the proportion of true negatives detected. Sensitivity and specificity do not change with prevalence.
- Prevalence refers to the proportion of a population with a disease. Higher prevalence increases the positive predictive value of a test.
This document discusses various measures used to quantify the effect of exposures or treatments on health outcomes. It describes relative measures like relative risk (RR) and odds ratio (OR), which compare outcomes between exposed and unexposed groups. It also describes absolute measures like risk difference and absolute risk reduction, which provide the actual difference in risk. The number needed to treat (NNT) metric is also introduced, which indicates the number of patients needed to treat to prevent one adverse event based on absolute risk reduction. Examples are provided to illustrate how these measures are calculated and interpreted.
The document discusses concepts related to measuring associations between exposures and diseases in epidemiology. It defines different types of associations and measures of association, including relative risk, odds ratio, and attributable risk. It explains that an association between two variables does not necessarily imply causation and discusses several approaches used in epidemiology to help establish whether an observed association may be causal.
Stratification and multivariate analysis are methods used in epidemiologic studies to reduce confounding. Stratification involves dividing the data into subgroups based on potential confounding variables, and examining the relationship between exposure and outcome separately within each subgroup. The Mantel-Haenszel method provides a pooled estimate that averages the stratum-specific estimates to provide an overall effect estimate adjusted for confounding. Evaluating differences between adjusted and non-adjusted estimates helps determine if confounding is present.
This document discusses measures of association used to compare disease rates between exposed and unexposed populations. It defines key relative measures like risk ratio and odds ratio that describe the direction and strength of association. Absolute measures like attributable risk and population attributable risk are also introduced, which quantify the excess disease rate due to exposure in numeric terms. Examples are provided to illustrate the interpretation and proper communication of these different effect measures.
This document discusses measures of association used in epidemiology to quantify the strength of relationships between categorical variables. It defines relative risk, odds ratio, and attributable risk, and provides formulas for calculating each. Examples are given for how to calculate and interpret these measures using data from cohort and case-control studies. Relative risk reflects the likelihood of disease in the exposed group compared to unexposed. Attributable risk quantifies the excess risk among the exposed that can be attributed to the exposure.
Attributable risk and population attributable riskAbino David
This document defines risk factors and describes methods for identifying and quantifying risk. It defines a risk factor as an attribute or exposure associated with disease development. Epidemiological studies help identify risk factors and estimate degree of risk. Relative risk compares incidence between exposed and unexposed groups, while attributable risk indicates how much disease can be attributed to exposure by comparing incidence rates. Two examples are given to illustrate these concepts and how attributable risk informs potential public health interventions.
Risk factors relate to the effect that an exposure may have on a person's health. Exposures can include drugs, environmental factors, or a person's physical characteristics. The outcome of an exposure can be any effect on a person's health, either positive or negative. Risk is reported as the likelihood of an outcome occurring due to an exposure, but this does not necessarily imply causation. Measures like relative risk, hazard ratio, and odds ratio are used to compare the likelihood of an outcome between an exposed group and unexposed group.
This document discusses measures of association used in epidemiology to quantify the relationship between an exposure and disease. It defines key terms like relative risk, odds ratio, and attributable proportion. Relative risk compares the risk of disease between an exposed and unexposed group. Odds ratio makes a similar comparison but uses odds instead of probabilities. Attributable proportion estimates the percentage of disease risk in the exposed group that can be attributed to the exposure. Examples are provided to demonstrate calculating and interpreting each measure. Overall, the document outlines the main epidemiological measures used to determine the strength of association between an exposure and health outcome in a population.
This document provides definitions and examples of key concepts for estimating risk from epidemiological studies, including probability, odds, relative risk, and absolute risk. It discusses how relative risk is calculated from cohort and case-control study designs. Relative risk compares the risk of an outcome between exposed and unexposed groups to determine if exposure is associated with increased risk. The odds ratio, which estimates relative risk, is presented as the measure used to assess association in case-control studies. Examples are provided to demonstrate calculating and interpreting these risk measures.
The document discusses various absolute measures used to quantify the association between an exposure and disease occurrence in a population, including attributable risk (AR), attributable risk percent (AR%), population attributable risk (PAR), and population attributable risk percent (PAR%). It provides formulas for calculating each measure and examples to illustrate their interpretation and how they depend on both the strength of the exposure-disease association and the prevalence of exposure in the population. Alternative formulas are presented that allow estimating these measures from case-control study odds ratios when only relative measures can be directly estimated.
relative risk, odds ratio, attributable fractions and lots of statistical measurements which can improve epidemiological problems have been discussed. Besides, some reliability instruments' description have been added.
R software codes for ICC, Kappa, Lin's agreement and Bland agreement is exist.
Statistics are important to understand data, summarize results, and make informed decisions based on research. Key aspects of understanding statistics include:
- Comparing results to peers to determine if an outcome is unusual or expected
- Calculating measures like mean, median, and range to summarize variables
- Understanding concepts like p-values, confidence intervals, relative risk, and number needed to treat to properly analyze and interpret data.
1) Statistics are important to understand data collected in research projects, summarize large datasets, and make evidence-based decisions.
2) Key statistical concepts include measures like relative risk, odds ratios, and confidence intervals which are used to evaluate research results and size of treatment effects.
3) Proper statistical analysis and reporting of values like p-values, absolute risk reduction, and number needed to treat are needed to accurately interpret results and clinical significance.
Cohort study design.ppt Epidemiology medicalSoravSorout
This document describes prospective and retrospective cohort studies. A prospective cohort study follows groups of exposed and unexposed individuals over time to see who develops a disease in the future. A retrospective cohort study identifies exposed and unexposed groups based on past exposure and disease occurrence data collected from medical records. Cohort studies aim to calculate the relative risk of disease between exposed and unexposed groups to determine the strength of association between an exposure and disease.
The document discusses risk assessment and various measures used to quantify risk such as relative risk, attributable risk, odds ratio, prevalence rate, and incidence rate. It provides examples of how to calculate these measures from cohort and case-control study data and interpret the results. Key points are that relative risk is used for cohort and experimental studies, odds ratio for case-control studies, and prevalence and incidence rates help measure disease burden. Attributable risk helps identify excess risk from an exposure. The examples help illustrate how to apply these concepts to public health practice.
This document describes different types of study designs used in research. It discusses observational studies, which involve analyzing exposures without intervention. Within observational studies, it distinguishes between descriptive studies, which generate hypotheses but no comparisons, and analytical studies, which attempt to establish associations. It provides details on specific observational study designs including cross-sectional studies, case-control studies, and cohort studies. Key aspects like study populations, exposures, outcomes, and advantages/disadvantages are compared across these study designs.
Common measures of association in medical research (UPDATED) 2013Pat Barlow
This is an updated version of my Common Measures of Association presentation. I've updated it to include (1) more detail on rates, risks, and proportions, (2) Absolute Risk Reduction (ARR), Attributable Risk (AR), Number Needed to Treat (NNT) and Number Needed to Harm (NNH). Feel free to email me for a full version of the slideshow.
This document provides definitions and explanations of key statistical and epidemiological concepts:
- A 95% reference interval contains the central 95% of a population distribution, calculated as the mean +/- 2 standard deviations for a normal distribution.
- Sensitivity measures the proportion of true positives detected, specificity measures the proportion of true negatives detected. Sensitivity and specificity do not change with prevalence.
- Prevalence refers to the proportion of a population with a disease. Higher prevalence increases the positive predictive value of a test.
This document discusses various measures used to quantify the effect of exposures or treatments on health outcomes. It describes relative measures like relative risk (RR) and odds ratio (OR), which compare outcomes between exposed and unexposed groups. It also describes absolute measures like risk difference and absolute risk reduction, which provide the actual difference in risk. The number needed to treat (NNT) metric is also introduced, which indicates the number of patients needed to treat to prevent one adverse event based on absolute risk reduction. Examples are provided to illustrate how these measures are calculated and interpreted.
The document discusses concepts related to measuring associations between exposures and diseases in epidemiology. It defines different types of associations and measures of association, including relative risk, odds ratio, and attributable risk. It explains that an association between two variables does not necessarily imply causation and discusses several approaches used in epidemiology to help establish whether an observed association may be causal.
Stratification and multivariate analysis are methods used in epidemiologic studies to reduce confounding. Stratification involves dividing the data into subgroups based on potential confounding variables, and examining the relationship between exposure and outcome separately within each subgroup. The Mantel-Haenszel method provides a pooled estimate that averages the stratum-specific estimates to provide an overall effect estimate adjusted for confounding. Evaluating differences between adjusted and non-adjusted estimates helps determine if confounding is present.
This document discusses measures of association used to compare disease rates between exposed and unexposed populations. It defines key relative measures like risk ratio and odds ratio that describe the direction and strength of association. Absolute measures like attributable risk and population attributable risk are also introduced, which quantify the excess disease rate due to exposure in numeric terms. Examples are provided to illustrate the interpretation and proper communication of these different effect measures.
This document discusses measures of association used in epidemiology to quantify the strength of relationships between categorical variables. It defines relative risk, odds ratio, and attributable risk, and provides formulas for calculating each. Examples are given for how to calculate and interpret these measures using data from cohort and case-control studies. Relative risk reflects the likelihood of disease in the exposed group compared to unexposed. Attributable risk quantifies the excess risk among the exposed that can be attributed to the exposure.
Attributable risk and population attributable riskAbino David
This document defines risk factors and describes methods for identifying and quantifying risk. It defines a risk factor as an attribute or exposure associated with disease development. Epidemiological studies help identify risk factors and estimate degree of risk. Relative risk compares incidence between exposed and unexposed groups, while attributable risk indicates how much disease can be attributed to exposure by comparing incidence rates. Two examples are given to illustrate these concepts and how attributable risk informs potential public health interventions.
Risk factors relate to the effect that an exposure may have on a person's health. Exposures can include drugs, environmental factors, or a person's physical characteristics. The outcome of an exposure can be any effect on a person's health, either positive or negative. Risk is reported as the likelihood of an outcome occurring due to an exposure, but this does not necessarily imply causation. Measures like relative risk, hazard ratio, and odds ratio are used to compare the likelihood of an outcome between an exposed group and unexposed group.
This document discusses measures of association used in epidemiology to quantify the relationship between an exposure and disease. It defines key terms like relative risk, odds ratio, and attributable proportion. Relative risk compares the risk of disease between an exposed and unexposed group. Odds ratio makes a similar comparison but uses odds instead of probabilities. Attributable proportion estimates the percentage of disease risk in the exposed group that can be attributed to the exposure. Examples are provided to demonstrate calculating and interpreting each measure. Overall, the document outlines the main epidemiological measures used to determine the strength of association between an exposure and health outcome in a population.
4. Measures of Public Health Impact • Attributable Risk (AR) Number • Attributable Risk Percent (AR%) Percentage • Population Attributable Risk (PAR) Number • Population Attributable Risk Percent (PAR%) Percentage
5. Measures of Public Health Impact IMPORTANT! They all assume (require) that a cause-effect relationship exists between the exposure and the outcome.
6. Relative Risk vs. Attributable Risk Relative Risk: Measure of the strength of association , and indicator used to assess the possibility of a causal relationship. Attributable Risk: Measure of the potential for prevention of disease if the exposure could be eliminated (assuming a causal relationship).
10. Attributable Risk (AR) Among the EXPOSED: How much of the disease that occurs can be attributed to a certain exposure? AR AR% This is of primary interest to the practicing clinician.
11. Attributable Risk (AR) AR = I exposed – I nonexposed = “Risk Difference” Develop CHD I SM = 84 / 3000 = 0.028 = 28.0 / 1000 I NS = 87 / 5000 = 0.0174 = 17.4 / 1000 (background risk) AR = (28.0 – 17.4) / 1000 = 10.6 / 1000 Smoke Yes No Yes 84 2916 3000 No 87 4913 5000
12. Attributable Risk (AR) AR = (28.0 – 17.4) / 1000 = 10.6 / 1000 Among SMOKERS , 10.6 of the 28/1000 incident cases of CHD are attributed to the fact that these people smoke … Among SMOKERS , 10.6 of the 28/1000 incident cases of CHD that occur could be prevented if smoking were eliminated.
20. AR: Drunk driving Dead Not dead Risk RD Drunk 45 255 300 0.150 Not d. 135 9565 10000 0.014 0.136
21.
22. Attributable Risk Percent (AR%) AR% = (I exposed – I nonexposed ) / I exposed = “Etiologic fraction” Develop CHD AR% = (28.0 – 17.4) / 28.0 = 37.9% I SM = 84 / 3000 = 0.028 = 28.0 / 1000 I NS = 87 / 5000 = 0.0174 = 17.4 / 1000 (background risk) Smoke Yes No Yes 84 2916 3000 No 87 4913 5000
23. Attributable Risk Percent (AR%) AR% = (28.0 – 17.4) / 28.0 = 37.9% Among SMOKERS , 38% of the morbidity from CHD may be attributed to smoking… Among SMOKERS , 38% of the morbidity from CHD could be prevented if smoking were eliminated.
24. Attributable Risk Percent I exposed – I unexposed RR - 1 ------------------------------- = ------------ x 100% I exposed RR
25. AR%: Fast driving Dead Not dead Risk AR% Fast 100 1900 2000 0.05 Slow 80 7920 8000 0.01 0.05 – 0.01 0.05 = 80%
26. AR%: Drunk driving Dead Not dead Risk AR% Drunk 45 255 300 0.150 Not d. 135 9565 9700 0.014 0.150 – 0.014 150 = 91%
27.
28. Population Attributable Risk (PAR) Among the EXPOSED and NONEXPOSED (e.g. total population): How much of the disease that occurs can be attributed to a certain exposure? PAR PAR% This of interest to policy makers and those responsible for funding prevention programs.
29. PAR and PAR% Example: We want to estimate how much of the burden of diabetes among Tampanians is attributed to obesity.
30. PAR and PAR% CAUTION! In order to calculate PAR and PAR%, we have to be reasonably sure that the results of the study can be generalized to the population of Tampa. (e.g the incidence rates drawn from the sample approximate the incidence rates in the entire population).
32. Population Attributable Risk (PAR) PAR = I total – I nonexposed Diabetes I T = 1100 / 10000 = 0.11 = 110 / 1000 I NE = 250 / 5500 = 0.0455 = 45.5 / 1000 (background risk) PAR = (110 – 45.5) / 1000 = 64.5 / 1000 Weight Yes No Obese 850 3650 4500 Slim 250 5250 5500 1100 8900 10000
33. Population Attributable Risk (PAR) PAR = (110 – 45.5) / 1000 = 64.5 / 1000 In Tampa , 64.5 of the 110/1000 incident cases of diabetes are attributed to obesity … In Tampa , 64.5 of the 110/1000 incident cases of diabetes that occur could be prevented with sufficient weight loss.
34.
35. Population Attributable Risk Percent PAR% = (I total – I nonexposed ) / I total Diabetes PAR% = (110 – 45.5) / 110 = 58.6% I T = 1100 / 10000 = 0.11 = 110 / 1000 I NE = 250 / 5500 = 0.0455 = 45.5 / 1000 (background risk) Weight Yes No Obese 850 3650 4500 Slim 250 5250 5500 1100 8900 10000
36. Population Attributable Risk Percent PAR% = (110 – 45.5) / 110 = 58.6% In Tampa , 59% of the cases of diabetes may be attributed to obesity in the population… In Tampa , 59% of the cases of diabetes could be prevented if Tampa residents lost sufficient weight.
37. PAR: Fast driving Dead Not dead Risk Fast 100 1900 2000 0.05 Slow 80 7920 8000 0.010 180 9820 10000 0.018 PAR = 0.018 – 0.010 = 0.008 PAR% = (0.018 – 0.014) ; 0.018 x 100% = 44%
38. PAR: Drunk driving Dead Not dead Risk Drunk 45 255 300 0.15 Not drunk 135 9565 9700 0.014 PAR = 0.018 – 0.014 = 0.004 PAR% = (0.018 - 0.014) : 0.018 x 100% = 22% 180 9820 10,000 0.018
45. Example: AR% (Case-Control Studies) Case-control study to evaluate the impact of smoking as related to bladder cancer. Bladder Cancer (160 / 90) OR = ------------ (120 / 200) = 2.96 Smoke Yes No Yes 160 120 No 90 200
46. Example: AR% (Case-Control Studies) Question: Among smokers , what proportion (percent) of bladder cancer cases can be attributed to their smoking habit? (OR – 1) AR% = ----------- x 100 OR AR% = ((2.96 – 1) / 2.96) x 100 = 66.2%
47.
48. PAR% (Case-Control Studies) (P E ) (OR – 1) PAR% = -------------------- x 100 [(P E ) (OR-1)] + 1 where P E = proportion of exposed controls (assuming that the proportion of exposed controls is representative of the proportion exposed in the source population)
49. Example: PAR% (Case-Control Studies) Case-control study to evaluate the impact of smoking as related to bladder cancer. Bladder Cancer (160 / 90) OR = ------------ (120 / 200) = 2.96 P E = 120 / 320 = 0.375 Smoke Yes No Yes 160 120 No 90 200
50. Example: PAR% (Case-Control Studies) Question: In this study population , what proportion (percent) of bladder cancer cases can be attributed to smoking? (P E ) (OR – 1) PAR% = ---------------------- x 100 [(P E ) (OR-1)] + 1 PAR% = (0.375) (2.96-1) [(0.375) (2.96-1)] + 1 x 100 = 42.4%
55. Annual Death Rates for Lung Cancer and Coronary Heart Disease by Smoking Status, Males 1000 – 500 = 500 per 100,000 127.2 – 12.8 = 114.4 per 100,000 AR 1000 / 500 = 2 127.2 / 12.8 = 9.9 RR 500 12.8 Non-smoker 1,000 127.2 Smoker Coronary Heart Disease Lung Cancer Exposure Annual Death Rate / 100,000
56.
57. Comparison of RR and RD Gerstman Chapter 8 (partial) Smoking causes more heart disease Smoking has a stronger association with lung cancer An exposure can have a strong relative effect (RR) but make a small difference in absolute terms (RD) Lung Cancer and CHD mortality in smokers and non-smokers (per 100,000 person-years) Smokers Non smokers RR RD LungCA 104 10 10.40 94 CHD 565 413 1.37 152
58. Relative Risk vs. Attributable Risk Relative Risk: Measure of the strength of association , and indicator used to assess the possibility of a causal relationship. Attributable Risk: Measure of the potential for prevention of disease if the exposure could be eliminated (assuming a causal relationship).
60. Summary – Measures of Public Health Impact Measure Cohort study Population-based case-control study Other type of case-control study AR Yes Yes No AR% Yes Yes Yes PAR Yes Yes No PAR% Yes Yes Yes
61. Ringkasan ukuran Tipe Kuantitas Matematis Tanpa denominator Dengan denominator Enumerasi Hitung, angka mutlak Rasio Proporsi Rate
62.
63. Ringkasan ukuran Ukuran dalam epidemiologi Ukuran Frekuensi Penyakit Ukuran asosiasi Ukuran efek /dampak
64. Ukuran frekuensi penyakit Ukuran frekuensi Penyakit Insidens Prevalens Insidens Kumulatif Incidence Density Prevalens titik Prevalens periode Mortalitas
65. Ukuran frekuensi penyakit Ukuran Rasio Risk Ratio Odds Rasio Insidence Density Ratio Prevalence Ratio
66. Ukuran frekuensi penyakit RD = Risk Difference AR = Attributable Risk ER = Excess Risk PAR = Population Attributable Risk PF = Prevented Fraction Ukuran Efek /dampak Perbedaan efek Fraksi Efek RD AR ER PAR AR% PAR% PF
Chapter 8: Association & Impact 10/10/10 Epi Kept Simple Smoking causes more heart disease even though the association between smoking a heart disease is weaker than the association between smoking an lung cancer. This is because heart disease is more common in the population.
Chapter 8: Association & Impact 10/10/10 Epi Kept Simple