Effect of Ban on Sale of Flavored E-cigarettes on Adolescent Vaping

Operationalizing outcome as a continuous and nominal variable

The US issued a ban on sale of flavored e-cigarettes in February 2020 to decrease the number of people who were vaping.  Dr. X is interested in examining whether this ban lead to decreases in adolescent vaping.  In March 2020 Dr. X surveyed 2750 adolescents in England, 2380 adolescents from the US and 2775 from France.  He found the following:

Number vaping

England        276

France          655

US                432

1.Describe one way to operationalize the outcome as a continuous variable. Describe one way to operationalize the outcome as a nominal variable.

2.Based on the results above, does US adolescent vape more than British or French adolescents? For full credit you must calculate appropriate measure of frequency and relative measures of association. You also must also state your null hypothesis.

3.Because you have taken health system research methods, you realize Dr. X’s research study has many problems. In a short paragraph, describe TWO major problems with Dr. X’s current study research design based on what he conceptually wants to research and what he actually did.  Please also describe how you would have done this study differently to address the two major problems that you identified.

4.In 2008, Oregon initiated a limited expansion of a Medicaid program for uninsured, low-income adults, drawing names from a waiting list by lottery. This lottery created a rare opportunity to study the effects of Medicaid coverage by using a randomized controlled design. By using the randomization provided by the lottery and emergency-department records from Portland-area hospitals, we studied the emergency department use of about 25,000 lottery participants over about 18 months after the lottery.

Table 1:  Comparison of ED outcomes by insurance status

Please write the research question.

5.Which of the numbers given in Table 1 are a rate, proportion or count (excluding the p-values)?

6.Interpret the results shown for the % of any ED visits in Table 1 above.  For full credit, you must calculate the absolute measure of association.  For full credit, you also must also state your null hypothesis and use the p-value to interpret the results

7.The researchers were worried that hospitalizations in the year before the study began is a potential confounder.  They conducted additional analysis shown in Table 2.  Based on the information given in Tables 1 and 2 does hospitalization in the year before the study began confound the association between insurance status and any ED visits?  For full credit, you must clearly explain the numbers you used and comparisons you made to come to your conclusion.  Please also explain whether your result and interpretation of confounding indicate there are any problems with the researcher’s study design.

Table 2:  Percent with any ED visits stratified by number of hospitalizations in the year before the study began

8.Describe one strategy the researchers could have used to control for confounding in the study design phase.

9.A study looked at the effectiveness of a needle exchange program in number of overdose deaths in Connecticut in 2009.  Researchers were worried that age, gender, education and income are potential confounders. The researchers calculated a relative risk ratio of 0.75 (95% CI=0.66, 0.82) in the crude analysis.  Participants in the study were matched on gender.  Stratified analysis showed a relative risk ratio of 0.74 for those younger than 21 and a relative risk ration of 0.74 for those 21 and over.  The relative risk ratio adjusted for all potential confounders was 1.2 (95% CI=0.89, 1.3) Please explain whether this needle exchange program was effective in reducing overdose deaths using the information given. Please explain which variables – age, gender, education, income – were confounders in this example

10.You can only calculate incidence rates from the following research study design for what type of research study designs?

11.Health systems consists of environmental, political, cultural, demographic and socioeconomic factors. Complex interplay of these factors influence health of populations and health services.  In 2-3 sentences, describe how one of these factors, besides the health system, that you think is contributing to the difference in mortality from COVID-19 in different countries.

12.You are interested in examining the prevalence of food insecurity of elementary students in New Jersey.  Provide two reasons why you would need to take a sample to complete your study. 

13.You find the following results.  Write one sentence interpreting these results.  For full credit please be sure to include whether this result is statistically significant.