Social Support and Health Status: Model Predictor of Social Support Levels and its Future Implications

Chidinma Iheanyi-Okeahialam

Authors:  Chidinma O. Iheanyi-Okeahialam, Alyssa M. Falise, Catherine W. Striley

Faculty Mentor:  Catherine Striley

College:  College of Public Health and Health Professions


Background: Evidence suggests increased hospital readmissions are associated with weaker social relationships. We hypothesized that mid-to-older adults with poorer health would rate their social support lower than those with better health.

Methods: University of Florida’s HealthStreet community engagement program surveys people about their health in areas where they live and work and refers them to services. Logistic regression was used to analyze predictors of social support using SAS 9.4.

Results: Among participants 50+ years old (n= 2,238), 40.6% identified as African American, with 62.6% female, 72.5% unmarried. the sample had an average of 2.13 lifetime chronic conditions (SD = 1.44 and 44.2% categorized their health as fair/poor. Low social support predictors included being 50-64 years old (aOR = 1.40 95%CI: 1.12-1.76), being male (aOR = 1.29, 95% CI: 1.05-1.58), not being married (aOR = 2.33 95% CI: 1.79 – 3.02), identifying as Black/African American (aOR = 0.63 95% CI: 0.50-0.78), having fair/poor health (aOR = 1.42 95% CI: 1.15-1.76), and having 3 or more chronic conditions(aOR = 1.38 95% CI: 1.11-1.71).

Conclusion: Results indicated poorer health is associated with lower social support. Future research should continue to consider the importance of low social support and how to improve it.

Poster Pitch

Click the video below to view the student's poster pitch.


Click the image to enlarge.
6 Responses
  1. Sarah Peeling

    Hi Chidinma,

    Great work, this was really interesting! Do you think that the direction of causality goes both ways? Or do you think that it is mainly either poor health leading to poor social outcomes, or poor social outcomes leading to poor health?

    1. Chidinma Iheanyi

      Hi Sarah,

      Thank you for visiting my page. Based on our data, we primarily looked at poor social outcomes leading to poor health/more comorbid illnesses. However, during the formation of our project, we did investigate the association of poor health leading to social outcomes. It is possible that the association can be made but we did not feel that our data would best be used for that particular investigation.

  2. Sheila Bosch

    Interesting research! I may have missed it, but are you considering the number of co-morbid conditions to be the indicator of “poor health”? Did you account for the fact that the severity of those co-morbidities differs between patients? Are these conditions self-reported, or do you have data documenting them? Your intro mentions hospital readmissions. Was that part of your analysis at all? Thanks for doing this important work!

  3. Corinne Evans

    Thank you for pursuing research in this topic. Your results are very interesting and extremely clinically-relevant. I agree that this research is very timely given the effects of COVID-19 on mental health and other health outcomes. I found your results regarding race to be some of the most interesting findings. Do you have any hypotheses as to why you found identification as a Native Hawaiian/Pacific Islander and American Indian/Alaskan Native to be a predictor of social support in this sample?

  4. Chidinma Iheanyi

    Hello Shelia,

    Thank you for visiting my page. The number of comorbidities is one of the indicators but is not alone an indicator of poor-health. There were other indicators such as being male and Native American, for example, that we also determined to be indicators. The conditions were self-reported, but we did not account for the severity of these conditions, as the data we used did not include them. We did not analyze hospital readmission but we took the data we found on readmissions as a baseline for background information for this investigation, fencing making readmissions our personal indication of poorer health. This information was one of the base tenets of our research.

  5. Chidinma Iheanyi

    Hi Corinne,

    Thank you for your kind remarks. My hypothesis would be that these two groups tend to be the least represented and hence may have limited access to health resources. The lack of resources can lead to strain on the lives of those with comorbid conditions and those who take care of them. Additionally, more extensive research with these two groups may lead to more definitive associations, especially regarding Native Americans because there is less research available on them particularly.