Date of Degree


Document Type


Degree Name





Kathleen M. Nokes

Committee Members

Arlene Farren

Stacey Plichta

Christine Zarcadoolas

Feleta Louise Wilson

Subject Categories

Health Communication | Nursing


Health Literacy, Electronic health literacy, internet health information, ehealth, Self-care agency, Orem's Theory of Self-Care Agency, HIV, HIV/AIDS, AIDS


Both low health literacy and insufficient electronic health literacy (ehealth) impede access to reliable internet health information for people living with chronic illnesses such as HIV/AIDS. Use of reliable internet health information has been shown to improve self-care through increased understanding of symptoms, disease processes, and improvements in adherence with treatment plans.

This study examined the effectiveness of two interventions that taught people living with HIV/AIDS (PLWH) how to recognize reliable internet based HIV health information. Orem’s Self-Care Theory was the framework for this quasi-experimental study that used a non-equivalent two-group design with two experimental interventions (MEDLINE and E-HELP). Participants (N=100) in both interventions watched the NLM video, Evaluating Internet Health Information, and completed an at-home assignment that consisted of navigating to an HIV health information website and identifying the six criteria of a reliable internet health information website as described on the video. The E-HELP group (n=50) received a teach-back session averaging about 12 minutes with an expert clinician. Participants in both groups returned one week later to demonstrate their ehealth. HIV health literacy, electronic health literacy, and ability to identify reliable internet health information were measured before and after the interventions and one week later; self-care agency was measured before the interventions and one week later.

The mixed model procedure, a Generalized Linear Modeling technique, was used to account for the longitudinal nature of the data. The mixed procedure allows for the modeling of the covariance structure to account for this correlation of the same subject responses. A restricted maximum likelihood approach (REML) was used and the covariance type was unstructured since we were not imposing any constraints on the values. A full model was fitted first for all study variables and included the basic conditioning factors of age, gender, race, ethnicity, education, and AIDS diagnosis. Model fitting was repeated eliminating the non-significant variables with the exception of time and the interaction of time and group. The independent variables (HIV health literacy, ehealth, and ability to identify reliable internet health information) were loaded to determine the effect of the intervention on the dependent variable, self-care agency.

Some differences were found in basic conditioning factors. Non-Hispanics had significantly more self-care agency compared with Hispanics (207.98 vs. 183.36, p=.000); this may be related to English proficiency, acculturation, and cultural factors. Younger age was significant for HIV health literacy (p=.020), ehealth (p=.001), and ability to identify reliable internet health information (p=.024) which is consistent with previous research. Women had significantly better HIV health literacy compared to men (p=.039).

No significant within group differences for self-care agency were found for either the MEDLINE group (p=.780) or the E-HELP group (p=.631). Participants may have needed more than one session of either intervention to produce an increase in self-care agency, or there may not have been adequate time between the two sessions of either intervention to produce increases in self-care agency. Results of this study did not provide sufficient evidence that, within the scope of Orem’s theory of self-care agency, health literacy is a foundational capability or that ehealth and ability to identify reliable internet health information are constituents of power component #7.

There were no significant differences in scores between groups for HIV health literacy (p=.748), ehealth literacy (p=.308), ability to identify reliable internet health information (p=.259), and self-care agency (p=.887). Possible reasons for the lack of significant difference between the two intervention groups include the short time between sessions, participants’ limited access to computers, and a preference for information from healthcare providers. In addition, the interventions may not have been sufficiently different to produce significantly different results.

There were, however, significant within group improvements from baseline to the final session in HIV health literacy (MEDLINE: p=.005; E-HELP: p=.045), ehealth (MEDLINE: p=.003; E-HELP: p=.000), and ability to identify reliable internet health information (MEDLINE: p=.000; E-HELP: p=.000), which indicates that some components of the interventions made a difference.

Participants’ demonstration of ehealth skills belied the lack of difference between groups on electronic health literacy and ability to identify reliable internet health information. Scores on the at home assignment, which required participants to navigate to an HIV health-related website, then find and identify the six website reliability criteria, were higher for the E-HELP group (received the teach-back session) compared with the MEDLINE group that just watched the video. When the investigator observed the participant find an HIV health-related website and identify the six website reliability criteria, there was an increase in ehealth for the E-HELP group compared to a decrease in ehealth for the MEDLINE group. Although further testing and refinement of the interventions are needed, evidence from this study indicates that either intervention may be a low cost and effective means to teach people how to identify reliable internet health information for self-care.