Item Details

title

STATISTICAL CONSIDERATIONS FOR EVALUATING OPIOID-RELATED POLICIES IN NORTH CAROLINA

author

Qiu, Chenhui

abstract

The opioid crisis has intensified nationwide and in North Carolina in recent years, prompting both federal and state governments to implement policies to address this crisis. Therefore, evaluating the effectiveness of opioid policies is very important. The lack of high-quality outcome data makes it difficult to evaluate policy effectiveness. Also, the evaluation of policy effectiveness inherently depends on the outcome being used. In this thesis, I construct a Poisson autoregressive model to examine the effect of the North Carolina STOP Act on five different county-level outcome rates. By estimating the average treatment effect, we find that the estimate of the causal effect of the policy depends on the outcome variable that is used. The STOP Act has the most significant negative impact on the rate of people who misuse opioids (PWMO) per population, which is the main goal of the STOP Act. It is important to note that, when conducting spatial causal inference, one of the four necessary assumptions is that there is no interference between different geographical units, otherwise the estimation of the treatment effect will be biased. Considering that in reality, population movement between adjacent counties to evade local opioid restrictive policies is possible, we need to take into account the potential spatial process where the policy implementation in neighboring counties also affects the outcome variables of the county. In the second part of this thesis, I consider models that account for spatial spillover to assess the impact that Syringe Service Programs have on rates of Hepatitis C. I use Moran's I to assess the amount of spatial autocorrelation in the residuals of models with and without spillover.

subject

Bayesian

causal

opioid

policy

spatial

contributor

Hepler, Staci (advisor)

Beavers, Daniel (committee member)

Evans, Ciaran (committee member)

Kline, David (committee member)

Hepler, Staci (committee member)

date

2024-05-23T08:36:09Z (accessioned)

2025-05-22T08:30:08Z (available)

2024 (issued)

degree

Statistics (discipline)

embargo

2025-05-22 (terms)

identifier

http://hdl.handle.net/10339/109407 (uri)

language

en (iso)

publisher

Wake Forest University

type

Thesis