History of incarceration and heart disease

Heart disease affects over 30.3 million adults in the United States and is a leading cause of mortality, morbidity, and disability.Data suggests that incarcerated individuals have a higher burden of heart disease compared to individuals in the general population. Therefore, the aim of this study was to assess the relationship between history of incarceration and heart disease. The study included 8,699 adults who enrolled in the National Longitudinal Survey of Youth (NLSY) in 1979 when they were 14 - 22 years of age. Following the baseline data collection in 1979, data was collected annually through 1996 and biennially since then. Starting in 1998, the "health 40" and the "health 50" modules were administered to the participants as they turned 40 and 50, respectively, to assess health status. History of incarceration was defined as any episode of incarceration in a correctional institution. Heart disease, was defined as self-reported diagnosis of heart disease. Regression models were used to model the relationship between history of incarceration and heart disease.

(Olukotun O, et al. Health Place 75 (2022) 102808).

1. What is the primary exposure of interest? (5 points)

1. What is the primary outcome of interest? (5 points)

1. In Table 1 (below), study investigators report participants' demographics stratified by incarceration status. For each demographic, indicate whether or not it appears to be a potential confounder.Briefly justify your response. (30 points)

Variables	Ever Incarcerated
	No (n = 8,111)	Yes (n = 588)	Confounder? (yes or no)	Justification
Age in 1979 mean(SD)	18 (2.23)	18 (2.19)
Body Mass Index >25 kg/m2	21.4%	18.6%
Race - White	65.2%	37.2%
Sex - Male	46.7%	87.8%
Unemployed	46.7%	56.3%
Household poverty	22.5%	40.1%
Years of education, mean (SD)	13 (2.56)	12 (1.69)
Ever drug use	14.5%	29.1%
Current/former smoker	45%	75.1%
Family History of Type-2 diabetes	7.3%	11.2%

The study investigators report the following results:

The crude odds ratio (OR): 2.29 (95% CI: 1.40-3.75)

An adjusted OR (AOR - adjusting for all characteristics listed in Table 1): 3.46 (95% CI: 2.04-5.88)

1. which method to control for confounding is being used by the investigators? (5 points)

1. For the association between history of incarceration (No v. Yes) and heart disease, calculate the magnitude of confounding by this collection of risk factors (covariates). Show your work. (5 points)

1. Is there joint confounding by this collection of risk factors? (5 points)

Motor vehicle traffic accidents are the leading cause of death among teenagers in the United States. Despite decades of declines in adolescent deaths, the death rate among teens due to traffic accidents has been increasing since 2013¹. The Youth Risk Behavior Surveillance System² (YRBSS) is a national survey of youth that is designed to measure the prevalence of health behaviors that contribute to death and disability in adolescents. The survey is administered every two years to youth in grades 9-12 in the United States.As of 2017, 46 states are participating in the YRBSS and data has been collected from more than 4.4 million high school students in 2017 YRBSS survey from three US cities: Chicago, IL, Fort Worth, TX, and San Diego, CA.

¹Curtin SC, et al. National Vital Statistics Reports. 2018 Jun;67(4):1-16.

This research question examines the association between texting while driving (yes/no) and alcohol use versus no alcohol use among youth aged 10-19. Below is the 2x2 table of the data:

	Texting While Driving
	Yes	No
Alcohol Use	553	562	1115
No Alcohol Use	325	886	1211
	878	1448	2326

1. Compute and interpret the prevalence ratio comparing the prevalence of texting while driving among those who use alcohol to those who do not use alcohol (10 points).

The analysis was stratified by sex.

Association between Texting While Driving and Alcohol Use Among Female Youth

	Texting While Driving
	Yes	No
Alcohol Use	151	250	401
No Alcohol Use	150	47	197
	301	297	598

Association between Texting While Driving and Alcohol Use Among Male Youth

	Texting While Driving
	Yes	No
Alcohol Use	402	312	714
No Alcohol Use	175	839	1014
	577	1151	1728

1. Compute and interpret the prevalence ratio comparing the prevalence of texting while driving among those who use alcohol to those who do not use alcohol among females. (10 points)

1. Compute and interpret the prevalence ratio comparing the prevalence of texting while driving among those who use alcohol to those who do not use alcohol among males. (10 points)

1. Is sex an effect modifier on the association between the prevalence of texting while driving and alcohol use? Justify your response. (5 points)

1. When the authors describe the results of the study, which of the following would be the most appropriate for them to present? Justify your answer. (5 points)

1. A weighted average of the stratum-specific estimates (a pooled MH sex

adjusted estimate)

1. Report stratum-specific estimates