First Advisor

Junghee Lee

Term of Graduation

Summer 2022

Date of Publication


Document Type


Degree Name

Doctor of Philosophy (Ph.D.) in Social Work and Social Research


Social Work





Physical Description

1 online resource (xiii, 104 pages)


Cancer survivors' perceptions on how cancer has impacted their lives has been identified as a "critical predictor" of psychosocial well-being and quality of life outcomes. Given the dramatic increase in survival rates and the long-term health and psychosocial challenges, as well as survivorship care barriers, this study focuses special attention on childhood and adolescent cancer survivors and is guided by Tedeschi and Calhoun's theory of posttraumatic growth and Hammond's theory of distinctiveness. This study aims to examine the influence of demographic (i.e., age at diagnosis, biological sex, race/ethnicity, type of cancer) and environmental factors (i.e., geographical location and insurance type), as well as the interaction effects (i.e., age at diagnosis x biological sex, and geographical location x insurance type), in predicting negative impacts of cancer (NIOC) and positive impacts of cancer (PIOC). Study findings provide insights to better understand the differential perceptions of female adolescent cancer survivors, as well as those residing in rural locations with public health insurance.

Study respondents were cancer survivors attending Oregon Health and Science University's Doernbecher Cancer Survivorship Clinic (DCSC) in Portland, Oregon, who were diagnosed with and treated for cancer between the ages of 0-19 years old and were in remission from cancer for a minimum of two years at the time of the study. During their first visit to DSCS, the Impact of Cancer for Childhood Cancer Survivors (IOC-CS) questionnaire was completed by participants 13 years and older (otherwise, after they turned 13 years old) to measure NIOC and PIOC scores. Demographic and environmental variables were derived from the Survivorship Repository and a retrospective chart review through OHSU's electronic medical record (EPIC). Of the 470 respondents, years since diagnosis ranged from 6-50 years, including 79% who were childhood cancer survivors (i.e., diagnosed between the ages of 0-14 years old, vs. 21% were adolescent cancer survivors diagnosed between 15-19 years old); 51% identified as male (vs. 49% were female); and 73% identified as non-Hispanic White (vs. 21% were non-White). Participants' current age ranged between 14-55 years old, with 67% residing in urban geographical locations (vs. 33% in rural) and 65% with private insurance (vs. 35% with public). Type of cancer featured three main groups, hematological cancers (60%), CNS/Brain tumors (11%), and solid tumors/soft tissue tumors/other cancers (29%).

To examine how well a new set of study variables predicted the outcome variables over and above the previously entered set of variables, hierarchical ordinary least squares (OLS) regression was utilized. Controlling for years since diagnosis, three sets of data analysis, including only demographic variables were entered in the first model, environmental variables were then added to the second model, and the interaction effects added in the final model. Each model was developed and tested separately for each dependent variable (NIOC vs. PIOC). Prior to OLS regression analysis, factor analysis was conducted to examine the factor structure of NIOC and PIOC.

Interestingly, PIOC scores were consistently higher than NIOC scores for all groups. Preliminary data analysis using t-tests found that NIOC was higher among female, adolescents, and participants with public health insurance compared to their counterparts; PIOC was higher among females, and carriers of private insurance; but no significant differences were found between non-White vs. non-Hispanic White, rural vs. urban, and types of cancer. All three OLS models were significant in predicting outcome variables. The final model including the interaction effects significantly increased model description above the second model for both NIOC (additional 1.6% of the variance, F(10, 433) = 3.88, p < 0.001) and PIOC (additional 1.8% of the variance, F (10, 433) = 3.31, p< 0.001). Being a female adolescent cancer survivor was identified as a risk factor for increased NIOC and decreased PIOC. Having public health insurance significantly increased NIOC, while the main effect disappeared after adding the interaction term (i.e., rural x public insurance). In addition, being a childhood and adolescent cancer survivor living in a rural location with public insurance was a significant risk factor for increased NIOC. Furthermore, having public insurance as a main effect was significant in decreasing PIOC.

Study respondents (i.e., childhood and adolescent cancer survivors) consistently endorsed higher PIOC than NIOC. This finding is not surprising and may be indicative of being in remission from cancer for a minimum of two years, and as years since diagnosis elapses, these individuals may experience an improvement in their physical health, personal growth, as well as greater health literacy. Furthermore, their social networks may have improved alongside family support. These individuals' stories may motivate survivors newly transitioning into post-treatment survivorship in navigating and overcoming the challenges involved in adjusting to life after cancer.

Consistent with previous studies, while it appeared that having public health insurance was a risk factor for increased NIOC, the contributing factors were complicated -- study participants residing in rural areas with public health insurance had significantly increased NIOC. The accessibility and quality of health care services may explain these differential experiences.

Study findings also highlight that the experience of being a female survivor diagnosed with cancer during adolescence may feature distinct experiences. These survivors reported not only reduced PIOC, but increased NIOC. Mixed-method study designs may help in deepening an understanding of their lived experiences to better inform the development and implementation of gender sensitive and developmentally specific cancer survivorship support programs.

Interestingly and contrary to prior research, being diagnosed with a CNS/brain tumor cancer type did not yield significant differences in NIOC as previously reported. Study findings provide insights on the need for survivorship cancer research to move towards frameworks of social determinants of health in further examining the accumulated effects of medical and non-medical indicators including socioeconomic factors and exposure to early childhood trauma that may influence NIOC and PIOC.


In Copyright. URI: This Item is protected by copyright and/or related rights. You are free to use this Item in any way that is permitted by the copyright and related rights legislation that applies to your use. For other uses you need to obtain permission from the rights-holder(s).

Persistent Identifier