Module 4 Overview Context

1. Module 4 Overview Context Content Area: Interpretation of Epidemiological Evidence Essential Question (Generic): Is the association causal? Essential Question (Drug Abuse Specific): Is an association with drug abuse causal? Enduring Epidemiological Understanding: Causation is only one explanation for finding an association between an exposure and a disease. Because observational studies are complicated by factors not controlled by the observer, other explanations must also be considered. Synopsis: In Module 4, students explore the rationale and methods of interpreting epidemiological studies. Students develop skills to assess possible explanations for an association found in a study, with consideration of explanations of causality, chance, confounding, reversed time order, and bias. Module 4 concludes with consideration of methods for weighing the overall evidence for an association. Lessons: Lesson 4-1: Introduction to Interpreting Associations Lesson 4-2: Causality Lesson 4-3: Chance Lesson 4-4: Bias Lesson 4-5: Confounding Lesson 4-6: Reverse Time Order Lesson 4-7: Weighing the Evidence

2. Module 4 - Interpretation of Epidemiological Evidence • Lesson 4-5 Confounding • Content • Definition and explanation of confounding • Examples of potential confounding in associations between watching R-rated movies and smoking and alcohol use • Controversies about the role of confounding in the study of gateway drugs • Big Ideas • Association does not necessarily mean causation • When an association is found, several possible explanations must be considered, including the possibility that the association is due to confounding • Confounding that is not measured and controlled will result in study findings that are not accurate. This project is supported by a Science Education Drug Abuse Partnership Award, Grant Number 1R24DA016357-01, from the National Institute on Drug Abuse, National Institutes of Health.

3. Where Are We? Essential Questions Enduring Understandings

4. Confounding Is the association due to confounding? 1. Cause 2. Chance Chance 3. 4. Confounding

5. Scenario 1 - Guilt by Association Hugo Lennie

6. Association Cause X Lennie Empty Beer Bottles Cause Guilt By Association Not Observed Hugo Observed

7. Definition Confounding A situation in which an association between a given exposure and an outcome is observed as a result of the influence of a third unobserved factor, called a confounder.

8. Review Confounding A situation in which an association between a given exposure (match carrying) and an outcome (lung cancer) is observed as a result of the influence of a third unobserved factor, called a confounder (smoking) Obviously we should have measured smoking

9. Association Cause X Cause Lung Cancer Match-Carrying Review - Diagram of Confounding Confounder Not Observed Smoking Observed

10. Scenario 2

11. R-Rated Movies No R-Rated Movies Scenario 2 Try Alcohol Not Try Alcohol Risks Relative Risk % or a b 15 c d 90 or %

12. Cause Scenario 2 “If we can keep our teenagers from watching R-rated movies, then we can go a long way towards preventing teenagers from trying alcohol.” “The study establishes a clear progression that begins with watching R-rated movies and leads to teenagers trying alcohol.” Watching R-Rated Movies Trying Alcohol

13. Association Cause X Watching R-Rated Movies Trying Alcohol Cause Scenario 2 Not Observed ? Observed

14. Not Observed Not Observed Hugo ? Association Association Cause Cause Observed Observed X X Watching R-Rated Movies Lennie Broken Beer Bottles Teenagers Trying Alcohol Cause Cause Scenarios Scenario 1 Scenario 2

15. Association Cause X Watching R-Rated Movies Teenagers Trying Alcohol Cause Scenario 2 Not Observed ? Observed

16. Association Cause X Watching R-Rated Movies Teenagers Trying Alcohol Cause Scenario 2 Confounder Not Observed ? Observed

17. Review - Definition of Confounding Confound To confuse Confounding A situation in which an association between a given exposure and an outcome is observed as a result of the influence of a third unobserved factor, called a confounder.

18. Association Cause X Watching R-Rated Movies Teenagers Trying alcohol Cause Scenario 2 Confounder Not Observed ? Lack of Adequate Supervision Observed

19. Epi Teams

20. Association Cause X Cause Matches Lung Cancer Confounding Confounder Not Observed Observed

21. Association Cause X Cause Playing Volleyball Skin Cancer 1 Confounding Confounder Not Observed Observed

22. Association Cause X Cause Low Birth Weight Babies Drinking Alcohol during Pregnancy 2 Confounding Confounder Not Observed Observed

23. Association Cause X Cause Eating Pretzels Auto Accidents 3 Confounding Confounder Not Observed Observed

24. Association Cause X Cause Watching TV Acne 4 Confounding Confounder Not Observed Observed

25. Association Cause X Cause Coffee Cancer 5 Confounding Confounder Not Observed Observed

26. Association Cause X Cause Driving Motorcycles Hepatitis C 6 Confounding Confounder Not Observed Observed

27. Association Cause X Cause Playing Baseball Oral Cancer 7 Confounding Confounder Not Observed Observed

28. Association Cause X Cause Skin Cancer Sand 8 Confounding Confounder Not Observed Observed

29. Association Cause X Cause Eating Ice Cream Drowning 9 Confounding Confounder Not Observed Observed

30. Confounding Is the association due to confounding? 1. Cause 2. Chance Chance 3. 4. Confounding

31. In the News - Gateway Drugs Read 1994 Article About Gateway Drugs

32. In the News - Gateway Drugs National Study Shows "Gateway" Drugs Lead to Cocaine Use Columbia University Record -- November 18, 1994 -- Vol. 20, No. 10 http://www.columbia.edu/cu/record/archives/vol20/vol20_iss10/record2010.24.html The Center on Addiction and Substance Abuse at Columbia (CASA) study shows that children (12 to 17 years old) who use gateway drugs - tobacco, alcohol and marijuana - are up to 266 times - and adults who use such drugs are up to 323 times - more likely to use cocaine than those who don't use any gateway drugs. Compared with people who used only one gateway drug, children who used all three are 77 times - and adults are 104 times - more likely to use cocaine. "This study, the most comprehensive national assessment ever undertaken, reveals a consistent and powerful connection between the use of cigarettes and alcohol and the subsequent use of marijuana, and between the use of cigarettes, alcohol and marijuana and the subsequent use of cocaine and other illicit drugs," said Joseph A. Califano, Jr., CASA's president and former HEW secretary. "An increasing number of American children and teens believe there is little risk in chugging a beer or smoking a tobacco or marijuana cigarette. With the recently reported rise in drinking and using marijuana by children and teenagers, this report is a wake-up call for parents to discourage their children from smoking and drinking and for governors and mayors to enforce the laws prohibiting the sale of cigarettes, beer, wine coolers and other alcoholic beverages to minors," he said.

33. In the News - Gateway Drugs (continued) The CASA study establishes a clear progression that begins with gateway drugs and leads to cocaine use: nearly 90 percent of people who have ever tried cocaine used all three gateway substances first. More than half followed a progression from cigarettes to alcohol to marijuana and then on to cocaine. The CASA study also concludes that the earlier a child starts to use these gateway drugs, and the more frequently, the greater the likelihood of using hard drugs. For example, children who smoke daily are 13 times more likely to use heroin than children who smoke less often. The study is the most comprehensive to date using national data that looks at both children and adults and all gateway substances. The research is based on the National Household Survey on Drug Abuse conducted by the National Institute on Drug Abuse. "No matter how we looked at the numbers, whether the user was white, black, male or female, the statistical connection between smoking, drinking or using marijuana and subsequent illicit drug use is clear," said Califano. CASA's analysis reveals: • Children who use marijuana are 85 times more likely to use cocaine than non- marijuana users. (Ninety percent of children who used marijuana smoked or drank first.) - Children who drink are 50 times more likely to use cocaine than non-drinkers. - Children who smoke are 19 times more likely to use cocaine than nonsmokers.

34. In the News - Gateway Drugs (continued) - The CASA study also links the use of gateway drugs by children with subsequent regular use of illicit substances as adults: • Adults who used marijuana as children are 17 times more likely to be regular cocaine users. (Ninety-one percent of adults who used marijuana as children smoked or drank first.) • Adults who drank as children are six times more likely to be regular cocaine users. Califano noted that the study finds a far more compelling relationship between the use of gateway drugs and subsequent use of cocaine and other illicit drugs than the 1964 Surgeon General's report found between smoking and lung cancer, the 1968 Framingham study found between cholesterol and heart disease, and the 1981 Selikoff study found between asbestos and lung cancer. "Each of those studies led not only to major investments in biomedical research, but to major changes in personal conduct among millions of Americans," said Califano. "Ultimately, prevention is our only hope for stemming the tide of new addicts. If we can keep our children and teens from smoking, drinking and using marijuana, then we can go a long way towards preventing the use of all dangerous drugs," said Califano. The study was conducted under the direction of Jeffrey C. Merrill, vice president for policy and research. Columbia University Record -- November 18, 1994 -- Vol. 20, No. 10 http://www.columbia.edu/cu/record/archives/vol20/vol20_iss10/record2010.24.html

35. Tobacco AlcoholMarijuana Cause Cocaine Association In the News “The study establishes a clear progression that begins with gateway drugs and leads to cocaine use.” “If we can keep our children and teens from smoking, drinking and using marijuana, then we can go a long way towards preventing the use of all dangerous drugs.”

36. Is Marijuana a Gateway Drug? • In the book. “Drug Abuse,” (Roman Espijo (Ed.) San Diego, CA: Greenhaven Press, 2002): • “Marijuana is a Gateway Drug” by Joseph Califano. • “Marijuana Use Does Not Lead to Harder Drugs” from • the National Institute of Medicine JE Joy, SJ Watson, Jr, JA Benson, Jr. “Marijuana and medicine: Assessing the science base,” Division of Neuroscience and Behavioral Research, Institute of Medicine. Washington DC: National Academy Press, 1999: “There is no conclusive evidence that the drug effects of marijuana are causally linked to the subsequent abuse of other illicit drugs”

37. Confounding Is the association between marijuana and cocaine use due to confounding? 1. Cause 2. Chance Chance 3. 4. Confounding

38. Unobserved Exposure Possible Explanations for Finding an Association Confounder Tobacco Alcohol Marijuana Cocaine

39. Unobserved Exposure Possible Explanations for Finding an Association Confounder Limited Adult Supervision Tobacco Alcohol Marijuana Cocaine

40. Unobserved Exposure Possible Explanations for Finding an Association Confounder Willingness to Do Something Illegal Tobacco Alcohol Marijuana Cocaine

41. Unobserved Exposure Possible Explanations for Finding an Association Confounder Poor Self-Control Tobacco Alcohol Marijuana Cocaine

42. Unobserved Exposure Possible Explanations for Finding an Association Confounder Poor Social Skills Tobacco Alcohol Marijuana Cocaine

43. Unobserved Exposure Possible Explanations for Finding an Association Confounder Depression Tobacco Alcohol Marijuana Cocaine

44. Depression X … and you avoided or eliminated the hypothesized cause, what would happen to the outcome? Possible Explanations for Finding an Association If the association was found due to confounding, …. found due to confounding, …. ? Tobacco Alcohol Marijuana Cocaine

45. In the News Sequence is not consequence.

46. Postscript - Later CASA Findings Statement of Joseph A. Califano, Jr., Chairman and PresidentThe National Center on Addiction and Substance Abuse at Columbia University on release of 2000 CASA National Survey of American Attitudes on Substance Abuse VI: Teens For the past six years, CASA has been engaged in the unprecedented undertaking of surveying attitudes of teens and those who most influence them-parents, teachers, school principals . . . We seek to identify factors that increase or diminish the likelihood that teens will use cigarettes, alcohol or illegal drugs in an effort to develop the most effective means of helping teens avoid substance abuse. CASA's surveys have consistently found that the family is fundamental to keeping children away from tobacco, alcohol and illegal drugs. The 2000 CASA National Survey of American Attitudes on Substance Abuse VI: Teens is an attempt to assess systematically the impact of parental conduct on the likelihood of teen substance abuse. . . . (continued on next slide)

47. Postscript - Later CASA Findings The loud and clear message of the survey is this: moms and dads should be parents to their children, not pals. Mothers and fathers who are parents rather than pals can greatly reduce the risk of their children smoking, drinking and using drugs. They can counter negative media influences and the availability of marijuana and other drugs in a teen's world. Whatever the family structure, whether the teen lives with both parents, a single mom or a single dad, their risk of smoking, drinking or using illegal drugs in "hands-on" households is dramatically lower than that of the average teen. This year we surveyed 1,000 teens age 12-17 - 526 girls and 474 boys. The most striking conclusion . . . “Hands-On" Parents - parents who establish a household culture where they consistently set down rules and expectations for their teen's behavior and monitor what their teen does - have teens at substantially lower risk of smoking, drinking and using illegal drugs than the average teen.

48. Supplementary Lesson about Confounding Confounding: Working the Numbers

49. No Cocaine Cocaine Marijuana No Marijuana Review Outcome 2x2 Table No Outcome Outcome a b Exposure c d Exposure No Exposure

50. % or % or Hypothetical Study No Cocaine Cocaine Risk Relative Risk Total 775 Marijuana 775 9,225 10,000 7.75 10,000 a b c d 4.08 190 No Marijuana 190 9,810 10,000 1.9 10,000 20,000 Adolescents who smoked marijuana were 4 times as likely to have used cocaine than adolescents who had not smoked marijuana.