Adverse Childhood Experiences and Autoimmune Disease — Is There a Link?

By Jim Jarvis, MD
Dept of Pediatrics and Genetics, Genomics, & Bioinformatics Program,
University at Buffalo Jacobs School of Medicine & Biomedical Sciences

Have you ever wondered, even if just a little bit, why it is that for every complex genetic trait, whether it’s coronary artery disease, type 2 diabetes, or preterm labor, prevalence rates are always higher in economically challenged, socially marginalized, and historically traumatized populations?  I have been pondering this question for more than 15 years.  As a Native American person, I can tell you that our communities are pretty tired of being told, “It’s because you have bad genes,” and we’re not buying that answer anyway.

Last month, I was part of a small group invited by our Association of American Indian Physicians to assist in developing culturally appropriate ways to inform tribal governments and physicians caring for American/Indian Alaskan Native (AI/AN) people of the results and implications of the Adverse Childhood Experiences  (ACE) study.  During our discussions, I was asked whether there was any known link between ACEs and rheumatic diseases, a reasonable question since both rheumatoid arthritis and systemic lupus show prevalence rates that are 5-10x higher in our AI/AN communities than in the broader population.  Well, it turns out that there is a link.

The study, Cumulative childhood stress and autoimmune diseases in adults (Psychosom Med 2009; 71: 243-250.  doi:  10.1097/PSY.0b013e3181907888) is actually 8 years old, and I’m surprised I hadn’t noticed it sooner.  In this paper, the authors queried ACE data from  >15,000 adults in the San Diego Kaiser Health Foundation cohort who were part of Dr. Vince Fellitti’s original study (which, in case you’ve been residing on Mars for the past decade, links adult health outcomes to specific traumatic childhood experiences).  Their outcome variable was hospitalization for any of 21 different autoimmune/rheumatic diseases, which, of course, would under-detect an effect of ACEs on an illness like rheumatoid arthritis, which is seldom a primary cause for hospitalization. Compared to people with ACE scores of 0, people with ACE scores of > 2 showed a 100% increased risk for hospitalizations related to rheumatic diseases. Given that ACE scores of > 5 are twice as common in our AI/AN communities than in the broader population, no one should be surprised that we also have high rates of rheumatic disease.

The important question is whether the association is actually causal.  As we learn more about the effects of trauma and toxic stress on the epigenomes of cells of the central nervous system and the immune system, a causal connection becomes quite plausible.  Furthermore, we know from animal data that epigenetic changes associated with trauma responses can be transmitted across the generations (see the fascinating paper in Nature Neurosciences 2014; doi:10.1038/nn.3594).  To AI/AN people, the idea that the multiple generations in which we have faced cultural loss, poverty, racism, violence, and social marginalization has led to the observed high rates of rheumatic diseases rings true. It certainly provides a much better explanation than “you have bad genes.”

The good news, both for AI/AN people and for rheumatologists, is that the epigenome is wonderfully plastic.  Indeed, we have shown in my laboratory that one of the primary things that successful therapy for JIA does is reorganize the epigenomes of multiple different cells types.  Furthermore, multiple studies have demonstrated effective interventions that can attenuate the effects of chronic, intergenerational trauma (see Science 2014; DOI: 10.1126/1248429).

So what’s the take-home in all this?  For me, these studies are a reminder that, even though we are pediatric subspecialists, we are still pediatricians.  We are predisposed by both inclination and training to understand the social determinants of health.  Could there be a link between ACEs and JIA?  Could the traumas of our patients’ parents, grandparents, or great grandparents be manifesting themselves in the children we care for?  In our Mohawk culture, we speak of the seven generations to come who will be affected by what we do and say, or by what others do and say to us. It is my fondest hope that somewhere in CARRA, right now, there’s a young fellow or junior faculty member willing to ask the kinds of questions that emerge from the ACE study and its applicability to our own subspecialty. The answers may affect the next seven generations.


What Caught My Eye… and Provoked My Paranoia and Propensity to Rant

By Kenneth N. Schikler, MD
Chief, Divisions of Pediatric Rheumatology & Adolescent Medicine
University of Louisville School of Medicine
Kosair Children’s Hospital


In the January 2017 issue of Journal of Pediatrics, Morris et al report on their findings in 1584 Australian youth age 14 years who were part of a longitudinal pregnancy cohort between 1989 and 1991 who were evaluated using the Beighton scoring system and a range of other factors including musculoskeletal pain status in “Hypermobility and Musculoskeletal Pain in Adolescents”.

Their findings might make us reconsider what the appropriate Beighton score might be significant regarding the assignment of Generalized Joint Hypermobility (GJH), and if we might base it on gender and sexual maturity ratings (SMR). The authors also discuss how gender and SMR might alter parts of the scoring system (placing palms to floor).

The findings that provoked my paranoia and reach for my soap box though was the prevalence of GJH they found.  When a greater than or equal to “4” score was used, 60.6% of the girls and 36.7% of the boys had GJH. When a score of greater than or equal to “6” was used 26.1% of the girls and 11.5% of the boys qualified as having GJH. This prevalence is not as high as reported in some series, and higher than other reported findings, but clearly finding GJH is not rare.

Now, to my  issue; I am troubled by the number of parents and patients I see who either on their own (with the assistance of the internet) or on the basis of a diagnosis given to them by friend, neighbor or physician come in as an Ehlers-Danlos patient, based on their recognized hypermobility. They do, in fact, have GJH, but they have no abnormal skin translucency, extensibility, nor doughiness, “velvetiness”, no abnormal scars, no easy bruisability, no petechial lesions nor history of such, no scleral discoloration or periodontal disease.  They have no features other than GJH and usually features of musculoskeletal pain, G-I complaints, lightheadedness or other features that have been lumped into a Hypermobility Syndrome, that doesn’t give the impression of a DISEASE.

If we use only the prevalence numbers in this study , which are lower than the numbers in other studies, and we use the cut off of a Beighton score of 6 do we want to say that a quarter of 14 year old Australian girls and just over a tenth of 14 year old Australian boys have a “disease” diagnosis of Ehlers-Danlos and have these kids and their parents concerned about the potential problems of the forms of the Ehlers-Danlos Syndrome that have potential pathologic outcomes, and  contraindications for procedures (e.g. arteriography) and can be genetically confirmed ?  When is a relatively common physical finding that can be found as part of a disease a disease? Does every teenage female with acne have a form of polycystic ovary syndrome?

Now for the paranoia, pertinent to those of us in the United States.  If our government does away with the ACA and with it the protection on “NO PRE-EXISTING CONDITIONS” is lost, do we as pediatricians want to put diagnoses that are not completely merited on a patient’s medical record that might allow an insurer to deny them insurance or put them in a high risk pool based on the diagnosis. As my paranoia and cynicism expand, as we now have made it clear that “Obesity” in childhood and adolescence is a risk factor for adult disease, will that also be a reason for difficulty in obtaining reasonable health insurance. As we now are being made aware of the confluence of Adverse Childhood Events (ACES) and epigenetic influences on future health outcomes, are we doing out patients any favors by documenting rather than being aware of these factors ?

OK, I have completed my rant and can take a breath and get back to work.

Doctors Facing Borders

A couple weeks ago, our administrator received an email that sent our division into panic. The title of the email was “URGENT,” in all caps, followed by three exclamation points. It was a warning of a possible draft of a new executive order that would limit entry into the US of individuals from yet another list of countries, this one including the country of origin of our first-year fellow. The past several weeks of drama and confusion over the recent executive order for a 90-day ‘travel ban’ of citizens from seven countries in the Middle East and Africa has put into relief an important sector of medical providers in the US: international medical graduates or IMGs.

National interest in a case of a Cleveland Clinic doctor, who was barred and then allowed re-entry into the States and a handful of other similar stories, have led to discussions both in and out of medical communities about the role of IMGs in the American medical system.

The New England Journal of Medicine ran two pieces back to back, that described the vast number of international residents and faculty that contribute to patient care and biomedical research in the US, the arduous and costly path they take to get to their positions, and the impact that barring these people from entry to the US might have on the future of American medicine. It turns out that last year, remarkably greater than 50% of matched internal medicine residents were IMGs.

Data from the American Board of Pediatrics shows that 22% of the pediatric residents who took the general boards were IMGs. IMGs are more likely to go into pediatric subspecialties than American medical graduates, and 41% of physicians that sat for the pediatric rheumatology boards were IMGs.

The American College of Rheumatology’s 2015 Workforce Report paints a similar picture: 43% of trainees in pediatric rheumatology are IMGs. What also bears mentioning, is that they project that by 2030 we will need twice as many pediatric rheumatologists in the US to cover the demand. Almost half (18 out of 40) of the fellowship positions in pediatric rheumatology went unmatched in 2015, with only 27 applicants going into the match. In a survey from the ACR Fellows-in-Training committee, 76% of IMG pediatric rheumatology fellows planned to continue to practice in the US.

For those trainees who eventually return to their countries of origin, in some cases the value of their contribution may be arguably greater. If our goal is to produce practitioners for children across the globe with rheumatic diseases, ponder for a second that as of 2014 there were only 5 pediatric rheumatologists in South Africa and fewer than 10 in all of Sub-Saharan Africa.

Regardless of your feelings about how the borders of the United States should be controlled, the inescapable fact is that these doctors play a vital role in serving US patients, particularly in fields such as pediatric rheumatology, where shortages exist and are only projected to get worse. And regardless of whether this or any new revised executive order on immigration (reportedly in the works) will impact IMGs to the US, I think it is worthwhile to start a dialogue about how, as a community, we will press forward with our mission of collaborative research and care and our vision of “a world free of limitations from pediatric rheumatic diseases” in a world where borders may be more difficult to traverse.

Share your thoughts about What Caught Our Eye in the comments section!


You’re Asking ME??!?

Harry L. Gewanter, MD, FAAP, FACR

By Harry L. Gewanter, MD, FAAP, FACR

“Isn’t there something natural or a diet we can use instead of these poisons?”

This is a question all of us hear on a regular basis. And while we may believe we are recommending treatments utilizing the best evidence, that may not be the perception of some families. It is understandable that the parental desire to protect their child results in appropriate anxiety and fear over what they may believe to be overaggressive, or even experimental, treatment regimens. Similarly, anecdotes and subconscious beliefs may drive decision-making in these stressful situations more than evidence.

Our children do not respond well to our answers of “Because I told you so” and neither do our patients and their families. On the other hand, it is also difficult to answer beliefs just with facts. Like our offspring, patients and their families must often “discover” what we already know on their own. It is therefore important to have an understanding of their perspectives and provide valid information in an approachable format.

One problem with answering their questions is that there may not be sufficient (or any) comparable evidence to cite about the specific complementary, alternative or integrative intervention the families may be interested in trying. Responding with “more sleep and exercise are excellent natural interventions” or “work on reducing your stress”, while true, just ain’t gonna cut it. And essentially every article about both allopathic as well as non-allopathic interventions ends with the statement: “Ask your physician.”

So … what is “your physician” to do?

What caught my eye were recent posts by the NIH’s National Center on Complementary and Integrative Health (NCCIH). They have been working on trying to develop and/or report better evidence on integrative practices since their inception in 1998. While this will not necessarily provide you with the data you may want, it is an excellent place for valid information for both you and your patients. They have a clinical digest that they post monthly and the reports on “Musculoskeletal Inflammation and Natural Products” and “Complementary Approaches to Chronic Pain” are particularly relevant to our field. One can also subscribe to their Clinical Digest and other informative items, join or view their webcasts and utilize their site as a place families can go for more information. And, by the way, they also fund research projects.

These two reports can give you at least some evidence and suggestions that you can give to your families and you can direct them to the NCCIH website. In addition, MedlinePlus has material ( as does the UK’s National Health Service ( and many US medical centers now have Integrative Medicine programs ( The American Academy of Pediatrics also has a Section on Integrative Medicine ( ) and its webpage within the AAP and/or its members can be a useful resource. For example, there is a handout for families ( that can be personalized for your practice.

Beyond sending your families to the web for more information, here are a few things you can point out (beyond the obvious fact that there is even less data in children than adults):

  • Please tell me about everything your child/you are taking! We need to know so we can be sure we don’t think there is a positive or negative issue is from the wrong intervention(s).
  • There is good evidence that mind-body techniques help reduce pain and improve quality of life for people with rheumatic diseases and other chronic illnesses.
  • There is good evidence that sleep and activity is good for everyone. (Ideally everyone should try to “Eat, sleep & sweat” daily.)
  • Acupuncture has been shown to be effective in a number of musculoskeletal conditions, especially painful ones, but we are still learning about its impact on inflammatory conditions.
  • There is developing evidence on the effect of various diets on inflammatory conditions, but it is early and we have much to learn.
  • We are learning about the interactions between natural substances and our current therapies and there are situations where the combination can be helpful or cause problems.
  • While many of these interventions are helpful, there is not yet sufficient information to say they work as well in reducing the inflammation and preventing the damage as what I am recommending.

No single medical philosophy has all the answers and we must remain open to new information and insights. Not too long ago we did not think much of an infectious cause for ulcers or that our gut flora could influence health. Who knows what we pooh-pooh now may be beneficial in the future.

Share your thoughts about What Caught Our Eye in the comments section!


Does Depression Cause Arthritis In Teenagers?

Mental illness is certainly an important aspect of many of our patients’ struggles, and I hope I’ve made the case earlier that it’s something pediatric rheumatologists should be paying attention to… There is a lot of good literature to show that patients with chronic disease, in general, are at higher risk for mental illness.

This makes sense to me, conceptually, that children and adolescents burdened with the stress and chaos of illness are at risk for psychological distress. Physical illness begets mental illness. I can also understand that this can be a bi-directional relationship, in which mental illness can in turn make physical illness worse. Perhaps this occurs in our patients when they are depressed if they are less likely to see their PCPs, or take their medications, or perhaps, as some have suggested, there are neuro-inflammatory effects that the brain has on the body that worsen physical inflammatory disease. But can mental illness cause inflammatory disease?

The authors of an article that recently came out in PLOS ONE suggest, yes. Specifically they suggest that affective disorders such as Major Depressive Disorder can cause arthritis in adolescents.

The article, more generally, is about links between mental and physical comorbidities, in which Tegethoff and her colleagues examined data from the Adolescent Supplement of the National Comorbidity Survey. In this survey, data was collected on demographics, mental illness, physical illness, and the reported timing of those.

The group previously published data showing associations between 5 different categories of mental illness and 8 different physical illnesses, in which there was a significant association between arthritis and mental illness, most notably anxiety (but also some association with affective disorder and substance abuse). In their most recent study, they examined timing to try to determine the direction of these associations. In terms of statistical significance and effect size, one of the most remarkable findings was the association of previous affective disorders with the development of arthritis, for which the hazard ratio was >3. In contrast, no significant hazard ratios were found the other way round, in terms of arthritis predicting mental illness.

While we’re cogitating on this, let’s review some important limitations to this study:

1) The presence and timing of physical illness was defined by a questionnaire administered to adolescent participants. The authors argue that this is a valid way of determining diagnosis, and they cite evidence showing that children are just as reliable as adults in describing their illness experience and health status. Perhaps one can use this to argue that they will be just as good at reporting physician-given diagnoses, but as one of their citations found, even in adults self-reported diagnoses agree well with physician diagnoses, except in the case of arthritis!

2) This brings up the question of how people interpret ‘arthritis’ (which is exactly as specific a terminology as was used in this study’s questionnaire ). To a rheumatologist this has a specific meaning and does not include joint pain without other findings, to a patient or even another researcher, this may be different. The US National Health Interview Survey used by the CDC collects data on arthritis with the following question: “Have you EVER been told by a doctor or other health professional that you have some form of arthritis, rheumatoid arthritis, gout, lupus, or fibromyalgia?”

Is it possible then, that the number of participants with JIA and other forms of chronic arthritis (what I thought of when I read “arthritis”) were few among this sample? And that others with joint pain or chronic pain syndromes were included and even drowned them out?

Looking at the authors’ previous paper about 2% of the total sample reported having had arthritis. This is at least 14 times what we would expect for the prevalence of chronic arthritis. On the other hand, juvenile fibromyalgia may be as prevalent as 6% and non-specific joint pain, more than that. Perhaps what is truly being observed here is the temporal relationship between affective disorders and musculoskeletal pain. This would be much less surprising and is congruent with all we know about amplified musculoskeletal pain syndromes, fibromyalgia, and their associations with mental illness.

So in my mind, the question of whether mental illness can cause chronic arthritis, whether JIA, lupus, or other, is still unanswered… but is it a far-fetched hypothesis? Certainly hypotheses that stress contributes to the development of autoimmune or inflammatory disease are not new, so perhaps this area is worthy of further research.

Share your thoughts about What Caught Our Eye in the comments section!