Introduction: Drug dependence is a chronic medical illness that often requires multiple treatment episodes and the use of health services. However, patterns related to substance use and abuse treatment are not well known. Two critical factors that have not been explored in relation to the number of prior substance use treatment episodes (PSUTEs) are multiple age groups and waiting periods. Hence, the first aim of this study was to examine if the frequency of prior substance use treatment episodes varies by different age categories. The second aim was to assess the extent to which the waiting period prior to receipt of substance use treatment services influences the likelihood of experiencing multiple treatment episodes. Methods: Data used for this research was the 2018 Treatment Episodes Data Set— Admissions (TEDS-A) (N= 1,935,541), which comprised of admissions to alcohol or drug treatment facilities across the United States. Descriptive statistics of participants was conducted. Bivariate and Zero-Inflated Poisson regression (ZIPR) analyses were performed to evaluate the number of PSUTEs associated with age and days waiting to enter a treatment facility while adjusting for other potential confounders. Andersen's healthcare utilization model was used to categorize covariates into predisposing, enabling, needs, and environmental factors. P-value ≤ 0.01 was considered the criteria for rejection of all null hypotheses. Results: Among participants, the average frequency of PSUTE was 1.60. About 34.2% were 25-32 years old, while 19.2% had a waiting period of between 1 to 7 days. Bivariate analysis showed that the number of PSUTEs (0 to ≥ 5) was significantly associated with all age groups and waiting periods, respectively. The results for age showed that 1.4% (12-17 years old), 8.4% (18-24 years old), 14.7% (25-34 years old), 16.7% (35-49 years old), and 18.1% (≥ 50 years old) reported ≥ 5 PUSTEs. For individuals with a waiting period of ≥ 31 days, the number of PSUTEs included 36.5% (no PUSTE), 20.2% (1 PUSTE), 12.3% (2 PUSTEs), 6.8% (3 PUSTEs), 4.2% (4 PUSTEs), and 20.0% (≥ 5 PUSTEs). ZIPR analysis demonstrated that the predicted log count of PSUTE increased significantly for every increase in age category. While for every increase in the number of days waiting to enter treatment, the predicted log count of PSUTEs significantly decreased. All potential confounders including, gender, race, living arrangement, type of treatment or service setting at admission, primary substance used, presence of co-occurring mental & substance use disorder, health insurance, and census region, were significantly associated with the frequency of previous substance use treatment episodes (P-value ≤0.01). Conclusion: This study demonstrated that multiple age categories and wait periods are significant predictors of the number of previous substance use treatment episodes. Notably, younger participants showed fewer prior episodes than older participants. These factors should be considered in order to develop effective strategies to improve treatment use and access to substance use treatment facilities. Therefore, more research is needed to explore these factors as well as other unknown predictors influencing multiple substance use treatment episodes.

In the United States (US), substance use, misuse, and substance use disorders are significant public health problems. Additionally, the difference between needing substance use treatment (SUT) and receiving treatment is becoming more pronounced. Therefore, the aims of this dissertation include, 1) to synthesize evidence of barriers and facilitators to integrating SUT into mainstream health care (MHC) after the Affordable Care Act was nationally implemented in 2014, 2) to examine the rate of treatment completion and dropout, along with its associated factors, among youth who received Medication-Assisted Opioid Therapy (MAT), 3) to determine the relationship between prior treatment episodes and length of stay (LOS) among adults in residential rehabilitation facilities (RRF) and explore other predictors of LOS and, 4) to determine whether experiencing multiple treatment episodes and being in certain age groups was associated with longer wait period to enter SUT. The literature synthesis involved data from five databases and was informed by PRISMA. Quantitative analysis (aims 2-4) included log-binomial, Poisson, and logistic regression models using the Treatment Episode Data Set-Admissions and Discharges. Andersen’s Behavioral Model for Health Services was the conceptual framework used to inform aims 2-4. For aim 1, several patient, provider, and program/system-level barriers and facilitators were identified, which highlights the need to use a comprehensive approach to improve SUT adoption in MHC. For aim 2, among youth who received MAT, 43.9% completed treatment and 56.1% dropped out. Factors positively associated with treatment completion included MAT use, males, self-help group participation, admission to detoxification and residential/rehabilitation settings, and being in the Midwest/Western US; while minority races (excluding Blacks/African Americans) and being in the South resulted in lower likelihood. For aim 3, the average LOS in RRF was 16.4 days, and having ≥ 1 prior SUT episode (PSUTE) was associated with slightly higher LOS. Some predictors associated with lower LOS included age (18-34 years), males, being employed, private insurance, and being in the Northeast. For aim 4, a longer wait period was identified among clients 25-49 years and those with ≥5 PSUTEs. These factors should be considered to improve SUT use, and future studies should corroborate these findings.