Dissertations & Theses (Open Access)

Date of Award

12-2020

Degree Name

Master of Public Health (MPH)

Advisor(s)

Ruosha Li

Second Advisor

Vahed Maroufy

Abstract

An Adolescent Wellness Care (AWC) visit is an annual preventive doctor visit predestined to assess the overall health of adolescents aged between 12 and 21 years old. Compared to their younger counterparts, adolescents who are using public insurance are at risk of not completing AWC visits due to several factors related to age or providers disparities in their living environment. This research was a quasiexperimental study aiming to compare two methods of automated Reminders/Recalls (RR) interventions, ‘phone-only’ versus ‘phone-and-text’, in order to evaluate which method is more effective in helping parents/guardians or young adults to schedule an AWC visit. A control group, characterized by the delivery of in-person telephone calls, was also designed to compare the marginal effect of each automated RR intervention. Recipients of those RR messages were followed-up with over a six-week period in each intervention arm, after entering the study. An overall and subgroup analysis based on Kaplan Meier survival estimates and Cox Proportional Hazard models were respectively performed to compare the hazard of scheduling an AWC visit across the different intervention arms and between the two automated RR interventions. The vii models were adjusted for demographical variables such as age, gender and race of the patients. A propensity scores matching analysis was implemented in order to account for selection bias and confounding effect in the study design due to non-randomization. A total of 516 patients were included in the analysis. By the end of the study period, 28.3% of patients in the control group, 31% of patients in the ‘phone-only’ group and 40.7% of patients in the ‘phone-and-text’ group scheduled an AWC appointment (P=0.0146). The crude hazard of scheduling an AWC visit was 73% (HR: 1.73, 95 % CI: 1.161 – 2.577) greater in the ‘phone-and-text’ group as compared to the control group (P=0.0070). The crude hazard of scheduling an AWC visit was 70.5% (HR: 1.705, 95 % CI: 1.157 – 2.513) greater in the ‘phone-and-text’ group as compared to the ‘phone-only’ group (P=0.0070). The adjusted hazard of scheduling was 51% (HR:1.510, 95 % CI: 1.014 – 2.249) greater in the ‘phone-and-text’ group as compared to the ‘phone-only’ group (P=0.0427). A one-year unit increase in age was associated with a 10.6% (HR: 0.894, 95 % CI: 0.837 – 0.955) decrease in the adjusted hazard of scheduling (P=0.0008) when the overall analysis was performed; and with a 11.3 % (HR: 0.887, 95 % CI: 0.818 – 0.963) decrease in the adjusted hazard of scheduling (P=0.0040) when the subgroup analysis was performed. Results from the propensity scores matching showed that the age and demographical area (based on patient’s zip codes) variables prompted some selection bias in the study design. Results from the propensity scores matching also showed that the race variable was a confounding factor in the quasi-experimental study, and that when the overall Cox PH model was adjusted for demographic variables, there was no statistically significant difference in the hazard viii of scheduling an AWC visit among the three intervention arms. The combination of automated telephone calls and automated text messages or ‘phone-and-text’ could be used as an effective tool to help healthcare professionals in the management of the scheduling flow of their patients, especially among their cohorts of children and adolescents. Future research should focus on randomized control trial (RCT) studies aiming to assess the rate of completion for AWC visits while using this combination of automated RR intervention.

Share

COinS