LYMEPOLICYWONK: 10 things you should know about new IDSA guidelines
The Infectious Diseases Society of America (IDSA) just released its new Lyme guidelines. They are 48 pages long and will take time to digest, but I want to share my initial impressions. These guidelines are in many ways a walk down memory lane – not much has changed – and what has changed has gotten more entrenched.
The guidelines use the GRADE approach to evidence assessment that the National Academy of Science recommends . There are now three sets of Lyme guidelines that use GRADE, including the International Lyme and Associated Diseases Society (ILADS) and the guidelines of NICE (the UK health agency).
The three sets of guidelines vary dramatically in their recommendations of key areas that Lyme patients care about—particularly in their assessment of how to diagnose and treat non-specific symptoms of Lyme disease and whether or not to retreat patients who remain ill.
Along the continuum of allowing for clinical judgment and consideration of patient values, the IDSA guidelines are by far the most restrictive of the three. Both the NICE and ILADS guidelines allow more flexibility in the exercise of clinical judgment and the use of shared medical decision-making between patients and clinicians based on individual circumstances.
The broad curtailment of clinical judgment by the IDSA here means that diagnosis, treatment, and retreatment are highly restricted and individualized care is replaced with a “one-size-fits-all” approach. It also means that for the most part individualized assessment of the risks and benefits for the individual patient have been hijacked by the IDSA, without examining or knowing the patient’s clinical history, circumstances, severity of illness, or values.
Denies persistent infection
Another key difference is that both ILADS and NICE recognize the potential for persistent infection, which the IDSA denies entirely.
To my eyes, the IDSA guidelines preserve deeply held biases of the former guidelines (many of whom are authors here). They also make a mockery of both the spirit and the rigor that GRADE is intended to instill in the guideline process. The goal does not appear to be to help patients get well, but rather to “game the system” and continue a pattern of systematically denying diagnosis and treatment to patients.
Process irregularities abound. They include using token patients, giving lip-service only to shared medical decision-making, inflating the importance of evidence base where it suits the authors, and using treatment outcomes the authors value instead of outcomes that are important to patients.
These guidelines deny care wherever they can. They do this by making it harder to be diagnosed and treated, by limiting treatment duration, by curtailing retreatment for most patients all together, and by not providing for the exercise of clinical judgment or shared medical decision-making in either the diagnosis or treatment of Lyme disease.
They have abandoned the use of the term Post Treatment Lyme Disease Syndrome (PTLDS) –at least in the guidelines. The importance of this is unknown.
The “chronic Lyme disease” section seems more designed to address legal concerns than patient care. (The IDSA has been subject to two legal actions for anti-competitive conduct.)
Some of the key points from the guidelines are highlighted below.
1. There was no representation of chronic Lyme patients on the guideline panel.
The IDSA says that the panel had “three patient representatives.” But the IDSA will not tell us their names. Representation needs to authentically reflect the patient community interests. It needs to be meaningful rather than token.
This means that a patient who is claimed to be a representative must be empowered to speak for the community with some form of accountability to the community. Anonymous patients cannot represent chronic Lyme disease patients because the community doesn’t even know who they are or if they are capable to fill the role—and there’s no way for them to be accountable. This is simply a form of tokenism .
2. The guidelines do not provide for shared-medical decision-making despite their lip service to the contrary.
Shared decision making in its broadest form is a process by which the clinician ensures that the voice of the patient is represented in the healthcare decision that is being made. It comes into play when more than one treatment option exists. Prostate cancer is the example most often given where patients can choose among four options that have different risk/benefit trade-offs.
In Lyme disease, the treatment options offered by the International Lyme and Associated Diseases Societies (ILADS) differ from those of the IDSA. Patients are entitled to know this. According to Professor Dorothy Fried at Yale, “virtually all patients . . . want to know . .. what other options are available” . Yet, the IDSA guidelines do not even mention the ILADS guidelines.
I could not find any recommendations where the guidelines recommend that the clinician and patient discuss the risks and benefits of a treatment to determine the best course of action (which is what shared decision making requires). Shared-medical decision making does not require that IDSA physicians provide the treatment to the patient if they do not believe the treatment will be effective, but patients should be advised that there are different treatment approaches so that the patient can seek out a physician who might provide that treatment.
3. The IDSA guidelines do not use treatment outcomes important to patients as GRADE requires.
Under the GRADE evaluation scheme which the IDSA says it has used,, the ranking of outcomes is supposed to be based on the importance patients place on them . Instead, the IDSA guidelines say that potential adverse events are more important than potential treatment benefits. Obviously, patients who are very ill or disabled would disagree. Both the ILADS and the NICE guidelines recognize this and rank potential treatment benefits first as patients would.
To understand the significance of this, you need to realize that all medical interventions have potential side effects or adverse effects. In addition, no treatments are universally effective. Whenever the evidence of treatment benefits was uncertain, the IDSA used ranking of adverse events above treatment benefits to deny treatment.
Usually, the decision of whether the risks of treatment outweigh the benefits is made by the patient and their clinician in the context of shared medical decision-making based on individual circumstances. For example, how ill is the patient, have they been responsive to treatment before, how severe are the side effects for a particular treatment, how frequent are they, is this a patient who commonly has side effects or not? The IDSA guidelines do not recognize outcomes that are important to patients or provide for individualized care in the context of shared medical decision-making.
4. The IDSA guidelines set the evidence bar too high by requiring that studies be done before any treatment is appropriate.
In a disease that is research-disadvantaged like Lyme disease, that is a bar that cannot be overcome in the lifetime of sick patients. Clinical trials take a long time and no trials for the treatment of chronic Lyme disease have been funded by the NIH in over 20 years. When the IDSA guidelines say “there is no convincing evidence” or “no causal association has been found” or trials have not been done, they are saying that the needs of patient for care today should be deferred until clinical trials are conducted.
But patients can’t wait. As Deborah Zarin, director of ClinicalTrials.gov. at the National Institute of Health explains:
“Clinical decisions are driven by the current reality. You can’t say to someone who has a medical need right then and there, ‘hold on we’ll do more clinical trials and get back to you in two years.’ You have to make decisions based on the best information available .”
The IDSA guidelines are also using average treatment effects from studies. That means that patients on average have to benefit. This approach does not work if patients vary in their response to treatment. For example, if one patient improves and another does not, on average there is no benefit even though one patient has improved. Precision medicine and individualized care recognize this and look to whether subgroups of patients improved. The NIH trials were too small to allow subgroup analysis. Studies of MyLymeData patients have shown that patients vary widely in their response to treatment and that a substantial portion improve with treatment [5,6].
5. Patients who don’t present with objective signs of early Lyme (an erythema migrans rash or Bell’s palsy) will have a difficult time getting diagnosed.
There is no diagnostic approach provided for patients who do not present with an EM rash. Although the guidelines seem to acknowledge that early Lyme can occur in the form of a flu-like illness without a rash, it is not separately addressed, and the guidelines do not provide a means of diagnosing it.
Nor are clinicians told how to diagnose any other form of early Lyme disease that manifests as non-specific symptoms. For example, the guidelines could say when clinicians have a high clinical suspicion of Lyme disease, they should test. But they do not say this. Some might argue that these diagnostic gaps will be filled in by clinicians in the trenches, I think it is more that patients without an EM rash will not be diagnosed.
6. Patients who don’t present with objective signs of late Lyme disease or neuroborreliosis will have a difficult time getting diagnosed.
The guidelines strongly recommend against “routine” testing for disease in patients with:
- Typical amyotrophic lateral sclerosis (ALS),
- Relapsing-remitting multiple sclerosis (MS),
- Parkinson’s disease,
- Dementia, or cognitive decline,
- New-onset seizures,
- Psychiatric illness, and
- Children with developmental disorders.
One could argue that recommending against “routine” testing does not prohibit testing where clinical impressions or patient history suggest Lyme disease. But it seems more likely to be interpreted by rushed clinicians as a recommendation that they should not test these patients at all.
Frankly, it is also hard for me to see why we would not routinely screen these patients for a disease that may be treatable, like Lyme disease. Many of these diseases are progressive neurologic diseases with no hope of cure. All of these conditions involve treatments. Some treatments are merely palliative (designed to treat symptoms rather than the cause) and often must be taken for life. All treatments have side effects – most far more serious than the side effects associated with oral antibiotics. For example, many anti-depressants have side effects of weight gain or sexual impairment. This is not to say that anti-depressant should not be taken, but let’s not say we should not test these patients for Lyme disease.
They also strongly recommend against testing for Lyme disease in patients with non-specific neurological symptoms in the absence of a history of other clinical or epidemiologic support for the diagnosis of Lyme disease. It’s hard to say what the effect of this recommendation will be. The guidelines do not provide any basis for supporting a diagnosis that does not have objective manifestations (e.g. EM rash) and epidemiologic support is largely restricted to endemic areas that are mainly on the east coast.
In MyLymeData, 70% of patients report that they were not diagnosed until late stage (six months or more after symptoms onset). Symptoms reported by most of these patients are non-specific neurologic symptoms. Most of these patients are not on the east coast. I think these patients will have a tough time getting diagnosed under the new IDSA guidelines.
7. Retreatment for early and late Lyme disease is very restrictive generally because the possibility of persistence of infection is denied across the board.
The IDSA guidelines regard all animal studies as “highly heterogeneous and hav[ing] limited generalizability to natural human infection.” The exclusion of all animal evidence (which is widely recognized in other diseases) raises the evidence bar too high because human evidence generally is not obtainable. The fact is that persistent infection has been demonstrated in humans who are undergoing other medical procedures where a biopsy or tissue collection is required . But these types of invasive procedures cannot be used commonly. Clinical trials targeted toward finding answers would be both not feasible and unethical. Limited retreatment exceptions are made for arthritis, meningitis, or neuropathy. As noted earlier, both the NICE and the ILADS guidelines accept the possibility of persistent infection.
8. The treatment for early EM rash or flu-like symptoms is limited to 10-14 days of treatment.
In the absence of objective disease activity such as arthritis, meningitis, or neuropathy, no retreatment is permitted for patients who do not recover. The recommendation for no treatment here is inconsistent with underlying treatment trials the authors are relying on. The treatment trials for early Lyme disease commonly retreated patients who remained ill .
9. Chronic Lyme disease and persistent infection do not exist or at least should not be treated.
The reasoning for the section of the guidelines devoted to chronic Lyme disease is convoluted, hard to follow, and tortuous to read. It seems designed to address legal concerns rather than patient care. I will try to break it down piece-by-piece based on my read for you.
First, the IDSA guidelines state that there is no definition for chronic Lyme disease. This is not true as both ILADS and Aucott’s group have peer-reviewed publications that include the definition of chronic Lyme disease (essentially, patients who remain ill six or more months following treatment) [5,6] .
Second, the IDSA incorrectly characterize the four NIH trials as showing no treatment benefit when they had mixed results . Some of the trials showed no benefit while others showed benefit in certain domains. Two of the trials showed a benefit on improved fatigue. They refer to these trials as being prolonged treatment, when they were actually limited to 90 days and cannot apply to longer treatments.
They then say that there have been no high-quality studies of patients who have heterogeneous symptoms. This may be true because those patients were excluded from the clinical trials as part of the selection process.
Next, they say that patients with “heterogenous symptoms” should be evaluated and alternative diagnosis should be ruled out. But they then recommend against treating these patients because a) “prolonged” treatments don’t work for patients with persistent symptoms, and b) “by definition, these patients often have no compelling clinical or laboratory support for the diagnosis of ongoing or antecedent Lyme disease.”
That’s an awful lot of mental gymnastics to say don’t treat. And the reason given seems to be “because I said so.” Almost all patients in MyLymeData have clinical support for their diagnosis and most report supporting lab tests.
They proceed to identify as the sole evidence gap, the possibility that patients have “medically unexplained symptoms”—which is code for we don’t know, we don’t care, and not my problem.
10. The guidelines make no recommendation for or against the use of antibiotics to treat STARI—specifically say “no recommendation; knowledge gap.”
Patients with a rash in areas where both STARI and Lyme disease exist may be treated clinically for the rash. The fact that there is “no recommendation; knowledge gap” for how to provide for STARI generally may mean that these is room for clinical judgment even when there is no geographic overlap with Lyme disease.
Patients had hoped that the IDSA would take the GRADE guideline process seriously and address the extensive comments that were submitted to an earlier glimpse of the IDSA draft. However, these comments it seems were simply ignored. Instead the guidelines do not address patient concerns or improve their outcomes. While most of healthcare is embracing measures that matter to clinicians and patients, with these guidelines the IDSA continues to turn a blind eye to the plight of Lyme disease patients.
Lorraine Johnson, JD, MBA, is Chief Executive Officer of LymeDisease.org and Principal Investigator of MyLymeData. You can contact her at firstname.lastname@example.org. On Twitter, follow her @lymepolicywonk.
- National Academy of Medicine. Clinical Practice Guidelines We Can Trust. National Academies Press: Washington, DC, 2011; p 217.
- Johnson, L.; Smalley, J. Engaging the Patient: Patient-Centered Research; Hall, K., Vogel, A., Croyle, R., Eds.; Springer: Switzerland, 2019; Vol. Chapter 10, pp. 507.
- Kashef, Z. To treat or not to treat: making the tough medical decisions with patients. YaleNews Jan. 13, 2016, https://news.yale.edu/2016/01/13/treat-or-not-treat-making-tough-medical-decisions-patients.
- Cameron, D.J.; Johnson, L.B.; Maloney, E.L. Evidence assessments and guideline recommendations in Lyme disease: the clinical management of known tick bites, erythema migrans rashes and persistent disease. Expert Review Anti-Infective Therapy 2014, 12, 1103-1135, doi:10.1586/14787210.2014.940900 http://www.ncbi.nlm.nih.gov/pubmed/25077519.
- Shor, S.; Green, C.; Szantyr, B.; Phillips, S.; Liegner, K.; Burrascano, J.J., Jr.; Bransfield, R.; Maloney, E.L. Chronic Lyme Disease: An Evidence-Based Definition by the ILADS Working Group. . Antibiotics 2019, https://doi.org/10.3390/antibiotics8040269.
- Rebman, A.W.; Aucott, J.N. Post-treatment Lyme Disease as a Model for Persistent Symptoms in Lyme Disease. Front Med (Lausanne) 2020, 7, 57, doi:10.3389/fmed.2020.00057 https://www.ncbi.nlm.nih.gov/pubmed/32161761
- Fallon, B.A.; Petkova, E.; Keilp, J.; Britton, C. A reappraisal of the U.S. clinical trials of Post-Treatment Lyme Disease Syndrome. Open Neurology Journal 2012, 6, 79-87, doi:10.2174/1874205X01206010079 http://benthamscience.com/open/toneuj/articles/V006/SI0078TONEUJ/79TONEUJ.pdf.