This transcript has been edited for clarity.
Terri Wilder: First of all, thanks for being with me.
Maile Young Karris: Thank you for the invitation to be here.
Wilder: During your presentation, you stated that up to 85 percent of people living with HIV experience chronic pain, and that feels really high to me. Is this high compared to people who don’t have HIV?
Karris: Yes, 85 percent is high. And The reason why we phrased it up to 85 percent is that, depending on what population of people living with HIV you’re looking at, the prevalence differs. For example, that specific paper where 85 percent of people living with HIV reported living with chronic pain was actually a paper that looked at an indigent population of people living with HIV.
If you look at the general population, including younger persons with HIV and older persons with HIV, it’s probably closer to something like 30 to 40 percent. In our specific cohort, only looking at older adults with HIV—and we classify older adults as people living with HIV who are 50 years and older—66 percent of them reported pain.
But regardless of what age you’re looking at, people living with HIV do appear to have more pain than HIV seronegative persons. For example, in general, chronic pain is believed to impact about 20 percent of the general population. In older adults, it’s roughly 38 percent. So, 66 percent is much greater than the estimate of 38 percent.
Wilder: How is pain defined? Is this physical pain, emotional pain, or both?
Karris: In our paper, we described pain as physical pain, and it was self-reported. The cohort that we looked at reported their other medical morbidities beyond HIV, or their comorbidities. We specifically asked them about whether or not they had back pain, hip pain, joint pain, or muscle pain. We also gave them options to write in other self-reported pain conditions. We included all of those self-reported conditions, plus if they indicated that they had any other types of pain.
Behind the Study: How Do We Assess Pain, and Pain Medication Use, Among Older People Living With HIV?
Wilder: Can you tell me about your study, and whether your research identified specific psychosocial and clinical factors associated with chronic pain?
Karris: It may surprise you because this is such a prevalent condition, but pain in HIV is pretty understudied, as a comorbidity goes. We do a lot of work in cancer and heart disease and hepatitis, but not nearly as much work in pain.
I just want to amplify a colleague—Jessica Merlin has really led the way in evaluating pain in people living with HIV. This is one of the first papers that specifically focuses on the population of older adults living with HIV.
As a clinician, I take care of people living with HIV in my own clinic. One of the things that I am interested in doing from a research standpoint is evaluating the areas in critical care where I feel either undertrained or unsatisfied with the current standard of care—and chronic pain management is difficult. It’s become more and more difficult, particularly with the CDC guidance documents that encourage pushing more nonpharmacologic approaches to care and minimizing opioid use in people, in general.
People with HIV, particularly older adults with HIV, are a fairly unique population, in the sense that HIV care started off as hospice care. Subsequently, many older adults with HIV who are often survivors of the AIDS crisis are and have been on really high doses of opioids for many, many, many years. And telling them now that “the government is saying I need to decrease your opioids. I need to make you do all these other things before we do increase opioids. Sorry you’re going to suffer,” was just not satisfying to me as a physician.
One of the blessings in my mind, being a critical researcher, is that I can take those frustrations in clinical care and then seek out funding—and thankfully we had funding to do this type of work—and try to better understand what can I change? How can I do this better?
So, this question about looking at chronic pain in older adults really came out of a personal frustration of my own, in my own clinic, and a desire to do better by the people that I care for. I happen to be already involved in this very unique cohort called ADHOC, and that stands for Aging with Dignity, Health, Optimism, and Community. It’s a new cohort, unlike some of our other well-known HIV cohorts like the MACS and the WIHS and CNICS. It was specifically designed to focus on issues that impact older adults living with HIV.
I approached them with this question and said, “Hey, chronic pain is really common. We really don’t know anything about how it impacts older adults and what factors contribute to chronic pain and subsequent chronic pain use. Can we use the cohort and evaluate this question?” Knowing what factors impact chronic pain and pain medication use is a starting point to then figuring out what we can do about this and how we can do this better. How can we manage pain better?
We evaluated this cohort of roughly 1,000 or so people at the time. There are 11 different clinics, both private and university clinics, involved in this cohort—that’s also a little bit different. It spans across the United States. Specifically, there are HIV clinics in California, Florida, Texas, Washington, North Carolina, Chicago, Illinois, Wisconsin, and Washington, D.C. So, it’s a pretty broad group of people who are aging with HIV.
We applied a couple of different approaches to analysis, looking at pain and pain medication use. First, we started with some bivariate analyses to look at differences between older adults with pain and without pain and with pain medication use and without pain medication use. Then we applied some multivariate logistic progression models to evaluate what factors were actually statistically significant and associated with both of those conditions.
Wilder: Can you tell me about the demographics of the study participants? Is it fair to say that it was an overwhelmingly 50-plus white gay male cohort?
Karris: Yes. Unfortunately, that is true. The mean age of the cohort is 60 years. We have a pretty decent range, anywhere between 50 to 89. A little over half of the participants are between 50 to 59. So, this cohort does contain a significant proportion of older-older adults living with HIV. It’s not just looking at the 55-year-old.
It is, unfortunately, predominantly male. So, 85 percent are male, 83 percent identify as LGBT, and about 69 percent are white. It is overall still an older male white LGBT population.
Wilder: Can you talk about the specific areas of the body that your study asked about and why those areas were asked about?
Karris: We asked about back pain, hip pain, joint pain, and muscle pain. We chose those specifically because they are common areas of pain that people report. We did not specifically ask, unfortunately, about neuropathic pain. I think if I were to design this study myself, I would have put in other specific conditions up front that we know are commonly reported in older adults with HIV. So, headache is another pain condition that is very commonly reported that we did not include, and neuropathic pain, as well.
That’s one of the limitations of the study—that it was done within a cohort that was not specifically designed to answer this question. So, we may not have captured everybody’s pain. Although we’re hoping that if there was a pain condition that significantly impacted someone’s life, and it was something that they viewed as a primary comorbidity, that they would have self-reported it.
Study Results: Factors Associated With Pain Medication Use Among Older PLWH
Wilder: Let’s talk about the results of the study. What did you find?
Karris: I’m specifically going to talk about the multivariate regression model. I’ll briefly mention that when we did the bivariate analysis, multiple factors ended up looking like it was associated with pain, including things like annual household income less than $50,000, unemployment, and multimorbidity, which we defined as having six or more medical conditions. That’s a little bit more specific than the general multimorbidity. Usually when people talk about multimorbidity they say two or more. We actually looked at six or more. We looked at tobacco use and low or no alcohol use.
Bivariate analysis also showed that pain was associated with anxiety, depression, loneliness, lower social wellbeing, and lower cognitive function. But when we put it into a logistic regression model, the only things that remained significant were male gender, Black race, and higher cognitive function. These were actually protective. So, those three conditions decrease the odds of having pain. Multimorbidity and tobacco use increase the odds of having pain.
We found pretty similar findings, as one would expect, when we looked at what are the factors and conditions that impacted the use of pain medication. So, again, bivariate analysis showed that pain medication was higher among heterosexuals, persons with lower educational levels, annual incomes less than $50,000, unemployment, multimorbidity, and low or no alcohol use, and also with anxiety and depression.
Here we saw lower interpersonal support and lower cognitive function as also being associated with pain medication use. But when we put it into multivariate logistic regression models, only multimorbidity and low annual household income were associated with pain medication use.
Clinical Analysis: How Treatment Access and Inequities Limit a Person’s Pain Care Options
Wilder: When you think about the findings, what do you think the results of this study can do to inform clinical practice, in terms of interventions to either prevent or to treat pain?
Karris: What I was really hoping to get out of this is to identify potential factors that are modifiable. There are some things that are difficult to modify. It’s hard to modify whether or not you’re male. It’s hard to modify your annual income. These are things that are very difficult.
But we can target tobacco use. And we can identify some other factors that maybe might help to mitigate some health inequities. When I think about why did we find some association, at least early on, with lower income and lower education, I don’t necessarily think that those things cause pain, but I think they’re just more reflective of the health inequities that people that maybe make less money and are underemployed or are not as highly educated struggle with.
One health inequity that I think we see every day in the clinic is access. One of the frustrations—going back to why this project was kind of first born was the CDC saying you need to do all of these nonpharmacologic approaches to chronic pain management before you even consider doing pain medications or opioids—is that it’s really difficult to get people living with HIV who may be underfunded from an insurance standpoint access to some of these things.
At our university I can’t get anybody access to pain behavioral therapy. They just don’t have the capacity to do that. So, nobody can get CBT [cognitive behavioral therapy], or acceptance and commitment therapy, for chronic pain management at all. I can barely get them access to physical therapy. People can only do that if they have the means to drive to physical therapy. It’s incredibly difficult to get access to things like chiropractic care or massage, or even water physical therapy. Insurance companies have denied my patients that multiple times. One of them even used that other letter, telling me that the reason why my patient couldn’t get it is because he had HIV and hepatitis C, and that prevented him from being eligible for water physical therapy. Which is crazy.
Wilder: What does that mean? Why would that make the patient ineligible?
Karris: I think what that reflects is HIV stigma. There are some conditions, for example, if you have an open wound or chronic diarrhea, then your insurance company may not approve you to get water physical therapy, because you potentially could be contaminating the pool, potentially making other people who use the pool ill. That is not the case with HIV and hepatitis C. It’s not transmitted that way. So I read that as being biased against people living with HIV.
I wrote several letters, and they never got back to me. My poor patient just gave up and ended up pursuing other means to try and help manage his pain.
So, access is a huge issue. We know health inequities contribute to that, but there are ways that we might be able to creatively address that. One really cool thing that we are able to do at our HIV clinic at UC San Diego is provide acupuncture. We have a long relationship with the school of acupuncture locally, and commitment from one of their longtime instructors to come to our clinic, bring students so that they can learn in real time on real people. That allows our folks to be able to access this without any personal costs. Things like that may be something that we can consider doing for some of our folks who are living with HIV that maybe are struggling with these health inequities.
What this data might be giving us is evidence and ammunition to really argue for and fight for improved access to address some of these health inequities. In the long run, we are going to need to have some policy changes.
Next Steps: Developing New Pain-Management Interventions for Older Adults
Wilder: Can you talk about what further research needs to be conducted on pain in older adults with HIV?
Karris: I didn’t really talk about this originally, but one of the reasons why I think that focusing on older adults is particularly important is that the pain management in older adults is more difficult because of natural age-related changes in pharmacokinetics and pharmacodynamics in drugs. That impacts the risk-benefit ratio of many, many of the medications that we commonly use to manage pain.
For example, nonsteroidal anti-inflammatory drugs like ibuprofen and aspirin are staples of pain management. But they are associated with an increased risk of gastrointestinal bleed, higher in older adults. And the Beers Criteria, which is a recommendation from the American Geriatrics Society of medicines that we shouldn’t be using in older adults because the change in risk-to-benefit, does not recommend using NSAIDs for those reasons.
Other commonly used medicines like muscle relaxants, gabapentin, and sleep aids can also negatively impair cognition in older adults, and increase their risk of falls and subsequent fracture, and disability, and the inability to be independent. So, we’re ultimately left at the end of the day with using Tylenol, or acetaminophen, and low-dose opioids for the pain management of older adults. That to me is not sufficient, a not-sufficient-enough of a toolbox to use to manage their degree of chronic pain.
I really do think that we need to focus on developing nonpharmacologic approaches to pain management that are acceptable and accessible to older adults. I’m currently bringing a pilot study on acceptance and commitment therapy provided by non-trained therapists in the group setting to see if this is acceptable to older adults, and whether or not it benefits them, in terms of their pain.
There are opportunities for technology, particularly if we can improve access to technologies that may help enhance the safety of opioid prescriptions in older adults, as well as introduce some of these other novel treatments.
I recently wrote a couple of grants with some colleagues to use technology in different ways to enhance the management of chronic pain. One is a home-based neurostimulator device to manage chronic pain, and the other is to develop a smart pill dispenser to help improve the safety of chronic pain management, and couple that with some personalized pain management strategies like medication and mindfulness.
There’s a lot of work that we can do. We know that pain really impacts the quality of life of older adults living with HIV and that we should have more work in this area.