Social Isolation and Diabetes Risk in Older Adults
At ENDO 2025, Samiya Khan, MD, a third-year internal medicine resident at the University of Southern California, presented findings on the link between social isolation and diabetes in older adults. Using data from the National Health and Nutrition Examination Survey (2003–2008), her team found that socially isolated individuals aged 60 to 84 had 34% higher odds of having diabetes and 75% higher odds of poor glycemic control (A1C >8%), even after adjusting for age, comorbidities, and behavioral factors. Dr Khan emphasized the urgent need for clinicians to screen for social isolation—a modifiable risk factor—as part of routine care, especially in the post-pandemic era, and called for further prospective research to explore causality and the distinct impact of loneliness.
Additional Resource:
-
Khan S, Khan F, Lampe Dominguez A. Social isolation is associated with diabetes and poor glycemic control among older adults. Presented at: ENDO 2025; July 12–15, 2025; San Francisco, CA. https://www.endocrine.org/meetings-and-events/endo-2025
-
Sun H, Saeedi P, Karuranga S, Pinkepank M, et al. IDF Diabetes Atlas: global, regional and country-level diabetes prevalence estimates for 2021 and projections for 2045. Diabetes Res Clin Pract. 2022;183:109119. doi:10.1016/j.diabres.2021
-
Holt-Lunstad J, Smith TB, Layton JB. Social relationships and mortality risk: a meta-analytic review. PLoS Med. 2010;7(7):e1000316. doi:10.1371/journal.pmed.1000316
TRANSCRIPTION
Samiya Khan, MD: My name is Samiya Khan. I am a third year internal medicine resident at the University of Southern California affiliated with the Internal Medicine Department.
Consultant360: What are the key findings that you're presenting at Endo 2025?
Dr Khan: Yeah, so the findings that I'm really excited to be presenting at this year's conference included the relationship on social isolation, diabetes and poor glycemic control. Our study found positive relationships between social isolation and diabetes, as well as social isolation and poor glycemic control, specifically in the older adult population, ages 60 to 84. We utilized data from the National Health and Nutrition Examination Survey from the years 2003 to 2008, and we utilized this data set, which fortunately is very robust and is a nationally representative data set. So the way this data is collected is done through a very multi-stage complex design and in a way that can be representative of the national population. The findings that we were able to extrapolate from our research can be applied on a national level and can be interpreted on a national level, which is one of the strengths of our projects.
And then most notably, our findings were also significant after controlling for all of our covariates, such as age sociodemographic factors and comorbidities like hypertension, hyperlipidemia, and behavioral factors like smoking and alcohol use, which is really notable as well because we know that insulin sensitivity reduces as we get older. But this relationship was controlled after we controlled for age was still significant, for example. And then to be more specific, we found at 34% higher odds of having diabetes for older adults that were socially isolated and a 75% higher odds of having poor glycemic control in older adults that were socially isolated compared to non-isolated individuals. And we defined poor glycemic control as an A1C as greater than 8%.
C360: Why do you feel it's especially important to focus on this link between social isolation and health outcomes like this in the post-pandemic era?
Dr Khan: I think one of the biggest reasons why this research is so important, number one, is how prevalent diabetes is. I was reading some literature and I saw that globally in 2021 diabetes affected about 10.5% of the world. And despite having more knowledge and awareness of the pathogenesis of diabetes about genetic links for diabetes, the prevalence of diabetes is expected to increase by 12.2% later in the future. And so really, not only is it such a big illness that affects so many people worldwide, but it's in fact expected to increase.
And I think after the COVID-19 pandemic is when we really saw social isolation become such an issue at the forefront. Even though social connections were on the decline before the COVID-19 pandemic, that really increased this problem. Social isolation really increased after COVID-19 due to issues like social distancing, loss of loved ones, and even death of loved ones as death and disability. So another thing that I want to highlight is understanding that if we can really intervene on social isolation potentially earlier for patients, then we may be able to hopefully prevent diabetes complications and really prevent the diagnosis of diabetes for older adults where this illness of diabetes is really devastating and can be catastrophic is basically the main finding.
C360: What practical implications do your findings have for clinicians caring for older adults with or at risk for diabetes?
Dr Khan: I think screening for social isolation is going to be something that is going to become more important for clinicians to do. There was the surgeon general's statement in 2023 and a really pivotal meta-analysis that was conducted in 2010 by Holt Lundestad, which showed that social isolation can be as deadly as smoking 15 cigarettes a day—can be even equivalently deadly as physical inactivity, obesity, and alcohol. And many doctors, or mostly all physicians screen for these issues like alcohol, like smoking, physical activity. And so if they can screen for these issues that are equivalently detrimental as social isolation, it really begs the question, should we also be screening for this? Especially if it's such a modifiable risk factor for patients. And the reason I think it's modifiable is that it's something that potentially physicians can work on and connect patients for resources to improve their social connections.
It's not really an easy feat to do, really requires a lot of coordination and it's something that I think we need to mobilize as a public to really help try to solve. But the first step of course would be screening. And then if once clinicians feel ready, they have the resources that they feel can help these patients, they can refer these patients potentially to community social workers, patient navigators, virtual support groups or support groups. And these patients may even need closer follow-up visits if they are particularly socially isolated because for a lot of patients, their primary care doctor, it may be their only visit that month or that couple of months. So there's definitely a lot of steps to be made. But I think the first step and probably the most important step is to really screen for this issue so that we can address this issue and then second, coming together as a public, as different universities, even on a national level, to come up with policies and programs that can potentially improve and really implement measures that address social isolation.
C360: What further research do you feel is needed around this topic?
Dr Khan: Yeah, I think there are definitely more studies that need to be done on this topic. One limitation of our study was that it was cross-sectional. So although we were able to see that social isolation was associated with diabetes and poor glucose control, there still needs to be further studies to examine and see if there's any causality. So does social isolation cause diabetes or cause poor glycemic control? And this can be best be done through prospective studies that follow patients over time, patients that are not socially isolated over time, seeing if the socially isolated patients develop diabetes and see how it impacts their A1C levels. Of course, randomized control trials would also be beneficial. So trials that implement measures that combat social isolation and seeing if there's any change to A1C with these specific measures. In addition, other studies that would be helpful as well is studies that assess loneliness.
And I say this because another limitation of the study was that while we focused on social isolation, loneliness and social isolation are not the same thing. Social isolation refers to more of an objective number of social contacts that a person has. They would have very few social contacts, whereas loneliness refers to patients that feel lonely. And so patients, especially older adults, can feel lonely even despite having a moderate number of social contacts. I think it would be really interesting to see, does this relationship also hold true for patients that are lonely? And we weren't able to assess for this just because of the limitation of the dataset, but fortunately there are also validated scoring systems for loneliness that I think would be really interesting to look at.
The main takeaway is really educating clinicians and the general public on how important it is to think about social isolation, and asking our patients questions surrounding their support system and where they live so that we can really target these patients better to come up with the best support for them.