To the top ^^ Close ^^

Fear Based vs Evidence Based Decision Making in the Radiology Suite: There is always more than one explanation for everything

By James L. Coyle, PhD, CCC-SLP, BCS-S, ASHA Fellow, University of Pittsburgh, and Jo Puntil, M.S., CCC-SLP, BCS-S, ASHA Fellow, St. George Regional Hospital

Published November 2023 in the ISHA Voice

ISHA’s Evidence-based Practice Committee invited Dr. Coyle and Ms. Puntil to share this article with the ISHA membership. The committee thanks them for the opportunity to do so.

Critical thinking is the disciplined process of clear, rational, open-minded thinking that is informed by evidence. When we think critically, we twist our beliefs to align with reality. We are skeptical, curious, and expect the evidence to support our methods and beliefs, not the other way around. Confirmation bias is the other way around. It is a bias in which we seek to support our beliefs by twisting reality to conform to our beliefs. Confirmation bias confirms whatever a person wants to think for whatever reason, and insulates them from addressing reality. In Rehabilitation Medicine, there is always a mechanistic source of a patient’s impairments, that has a specific mechanism of action by which it impairs patient activity and participation. While restoration is the goal or primary aim, medical professionals first work hard to identify that mechanism, and then identify treatments that align with the disease mechanisms of action, trying to reverse these impairments through their own physical actions upon the patient. Critical thinking enables the clinician to objectively focus on the impairment’s ingredients, how they alter normal function (mechanism of action), and then identify the treatment ingredients (what the SLP does) that move the case toward the goal or primary aim. But the treatment must align clearly with the disease that is causing the impairments by selecting the evidence-based proper treatment targets, understanding the mechanism by which the treatment is known to beneficially alter swallow physiology, and then list the ingredients of the treatment (Hart et al., 2019).

The Videofluoroscopic swallowing study (VFS) is a combined diagnostic examination and treatment efficacy trial (Logemann, 1993). As a diagnostician, the clinician reviews available records to understand the evolution and underlying physiologic nature of the patient’s problems, examines the patient to characterize oropharyngeal, facial and oropharyngeal sensorimotor function, observes the patient swallowing, and forms inferences about the swallowing impairments that are most likely to produce this cluster of signs. They form a list of feasible hypotheses regarding likely patterns of impairment (differential diagnosis), determine what needs to be done in a VFS study to address each hypothesis, and then during the clinical examination, samples the patient behaviors that may be necessary as compensations during the VFS study. They then systematically test these hypotheses using imaging, administering sets of trials with specific bolus conditions. As observations are made, the clinician treads both diagnostic and treatment waters, systematically employing one, after another, compensation based on their knowledge of the evidence supporting them in resolving a specific impairment. They record the efficacy results for each trial in real time (sometimes needing to be creative because our hypotheses were wrong!), finalizes the differential diagnosis or recommends tests that will sort out mimicking conditions. They formulate impressions about the nature of the problem, its likely natural history of evolution with or without treatment, propose logical restorative and compensatory treatments based on published peer-reviewed evidence of their efficacy with similar patients, and begins treatment.

It is highly recommended that Speech Pathologists access clinically relevant information about dysphagia by utilizing the ASHA Practice Portal.  It provides clinical topics with links to peer reviewed articles, competency tools to use, Client/Patient handouts and tools and templates for practical examples. The recommended standard of practice for dysphagia is to screen patients for swallowing, complete a clinical swallowing evaluation and if needed, proceed to instrumental testing (MBS/FEES).

Regarding Modified Barium Swallows (MBS):  It is recommended by CMS (Centers for Medicare/Medicaid Services) to establish and implement best practice on evidence-based medicine to implement a standardized tool for performing, documenting, and communicating modified barium swallows. This ensures patient safety by maximizing standardization, ensures the patients receive the highest quality of care and follows CMS guidelines. Merit Based Incentive Payment System is required by law to implement a quality payment incentive program.  It rewards value and outcomes in two ways:  Merit Based Incentive Payment System (MIPS) and Advanced alternative Payment Models (APM). MIPS is designed to tie payments to quality of efficient care, drive improvement in care processes and health outcomes.  The purpose of the MBS is to assess severity of the impairment, provide information regarding the sensorimotor function and physiologic impairment and to target compensatory strategies and/or treatment goals.  Executing a standardized Modified Barium Swallows allows clinicians to visualize and incorporate physiological response to various consistencies, communicate results verbally and in the ERM. Speech Pathologists should rely on evidence of what they see in the radiology suite, not opinion to drive decision making. And they should always perform good examinations and review all of the data, regardless of their experience level.

In 2020 CMS/Medicare looked at Modified Barium Swallows and saw the vast difference in practice patterns across centers and Speech Pathologists. Their goal is to improve outcomes by all Speech Pathologists to adhere to the research and execute, communicate, and document the MBS in an evidence based standardized manner. The lead measure is NQF 2624, quality measure #182. This measure ensures quality, improvement in activity or process, PROMOTING INTEROPERABILITY, and advancing care information.  All Speech Pathologists should be executing, interpreting, and documenting in a standardized fashion, and when that is not possible, in the most standardized manner possible. Sometimes, a case is so complex that an initial sub-goal is all that can possible be sorted in an initial examination; the clinician can apply the same systematicity to a limited set of trials as they do with standardized procedures. When a clinician does this, uses valid methods and measurements, keeps good data and decides to share their experiences with the rest of us in a publication, we call this an “N of One clinical trial”. 

The Modified Barium Swallow has evolved over the last 40 years including the original cookie swallow (Logemann, 1983) which evolved to be called the  to more detailed perceptual  measurements such as the Modified Barium Swallow Impairment Profile (MBSImP) (Martin-Harris et al., 2008), the Dynamic Grade of Swallowing Toxicity (DIGEST) (Hutcheson et al., 2017), Analysis of Swallowing Physiology, Events, Kinematics, and Timing for Use in Clinical Practice (ASPEKT) (Steele et al., 2019) and the BabyVFSSImP (Martin-Harris, Carson, Pinto, & Lefton-Greif, 2020).  CMS sees the objective nature of scoring and documenting MBS procedures as an essential process of standardization using one of these methods.  It is preferred to test all phases of the swallow with all consistencies of food and fluids in the lateral and then A/P view as the patient can tolerate, though the clinician needs to quickly and efficiently prioritize each test’s safety and aims.  Arguably a complex case would have more focused clinical questions that are not amenable to a standardized protocol.  Varibar (Bracco) is a barium formulation that was specifically formulated to NOT adhere to the mucosal lining (even the pudding) and is standardized for the MBS procedure and should be used to enable comparisons at different test points. There should be sufficient administration of all consistencies to establish the patient’s physiological response while also adhering to the radiation safety imperatives of ALARA. It is expected that ALL Speech Pathologists review the recordings for best practice as well as to measure the patient’s safety and efficiency of their swallowing mechanism. Vose and colleagues (2018) conducted a survey of SLP practice patterns in the US. More than 30% of the sample admitted to never conducting post-exam frame by frame review of the VFS data. We wonder what the actual number is.

What does this have to do with critical thinking and confirmation bias? In over sixty combined years of clinical practice, we have seen and heard many examples of confirmation bias amid the conduct of diagnostic services and especially the VFS examination. “I don’t see airway residue, but I know it is there.”  “Elimination of aspiration is the goal of dysphagia treatment.” “I know I can get this patient through a VFS without aspirating!” “Always rotate the head to the ‘damaged’ side.” “You aspirate thin liquids, so you need to drink thick liquids.” “Patient needs a feeding tube because aspiration is fatal.”  “No straws!!!”  “Free water protocols are completely harmless to everybody.”  “I don’t need to do a VFS study, I did a blue dye test.”  “I don’t need VFS, I use cervical auscultation.” I never prescribe a diet without a VFS study.” “I’ve seen hundreds of cases like this; no imaging needed, here is what we will do.”

Each of these statements reflect some degree of confirmation bias, because each of them is false or objectively invalidated or unsupported in the literature or assuages a clinician’s fear of litigation. And in each of these situations the decisions have been pre-emptively limited by bias. None of these statements are valid or supported by evidence, and, in the case of structuring a VFS study to ensure there is no aspiration in a suspected aspirator, just plain bad clinical practice. (The authors’ replies to the statements above appear below).

Being a clinician who performs treatment with people who have disabilities requires a specific set of skills that most SLP’s possess. But some of those “therapist” skills are misplaced in the other half of our world – the medical side. Here physicians expect us to be mid-level practitioners and literally inform them what needs to be done. In that scenario, we perform the consult in the same manner as a physician to specify the problem list. Separating the diagnostic and treatment roles of the medical SLP can be a challenge for some individuals because their belief system intrudes into the clinical process. And in some cases, knowing one fact, but not the others, can be problematic and contribute to confirmation bias. Consider an SLP whose family member underwent dysphagia management after a stroke. The clinician is certainly deeply affected by their loved-one’s situation. They may believe that any aspiration can cause death, or that tubes eliminate aspiration, because they are not trained and educated in pulmonary science. This clinician’s beliefs may intrude into their decision-making in a way that does not align with the patient’s values and expectations, and lead to inadvertent coercion.

Many instances of confirmation bias are rooted in defensive clinical practice by the clinician. The clinician fears doing harm so much that they exclude perfectly reasonable interventions from the treatment plan. “My patient had a PAS score of 6 during the VFS. NPO is recommended.” This sort of decision-making reflects a substantial lack of critical thinking and directly leads to harm, regardless of the clinician’s motivations, and the diagnostic process has turned into a screen. The clinician believes that any aspiration will cause permanent and irreversible harm, though the risk is very low for the vast majority of people. Recall that from Robbins et al (2008), the incidence of pneumonia among 515 institutionalized, frail aspirators, all on oral diets, was less than 11%. What is your threshold for risk? If it is close to zero, this sort of decision-making would be common in your practice.

Full Disclosure: early in their practice in the 80’s, the authors would literally never give water to a patient during a bedside exam. Why? Because at that time, all we knew was about aspiration and how to stop it. We now know that it is but one small factor in the pathogenesis of pneumonia. Now, water is the go-to! Likewise, we, too, in the 80’s, tended to be very cautious. Aspiration led to NPO status, but science knew little about aspiration pneumonia. Times have changed.

When we exit graduate school, we are storing a lot of knowledge. The teachers’ and supervisors’ opinions are absorbed by the students and trainees. Can the trainee/student now adapt as an independent clinician? Can they (and we) keep up with rapidly changing evidence? Can the teacher leave their biases at the door when they are in the teacher/supervisor role?  A clinician who does not deploy critical thinking does a disservice to our patients if they are training other clinicians by spreading bad practice. They confirm their bias with their impressions and eminence and do not seek to prove themselves wrong. They often create “new” methods to replace standardized methods which were painstakingly developed to provide reliable and stable measurements.

Unfortunately, proving ourselves wrong is the job of a diagnostician and therapist. This is top-of-license thinking. Critical thinking is “what have I missed?” It is “is there an alternate explanation for what I am observing?” Overconfidence can also shroud critical thinking. Why? Because there ALWAYS at least two feasible explanations for every natural event. One may be less likely or extremely unlikely. The clinician that goes into a diagnostic examination knowing what they are looking for, will find it. And they will then stop looking for anything another explanation. There are always at least two plausible explanations for everything. Challenge yourself by trying to prove you are wrong. It is the most effective way for an individual to identify personal sources of bias influencing clinical decision-making while constantly improving their practice effectiveness. Developing evidence-based confidence overcomes the fear.

Here are the answers to the scenarios and invalid statements above. 

  1. If you don’t see it, it isn’t there. It is a rule for visual perception-based instruments.
  2. You cannot eliminate aspiration in anybody, without surgery. You aspirate. A PAS score of 2 is normal for everybody (Robbins, Coyle, Rosenbek, Roecker, & Wood, 1999).
  3. Structuring a VFS so that the patient does not aspirate fails to identify the problem. You can’t fix something unless you observe it. 
  4. Sometimes the damaged side is less compliant than the opposite side. Know your spastic/flaccid taxonomy and direct boluses to the more compliant side. 
  5. There is no superior outcome cited in the literature, to support prescribing dietary thick liquids, in cases of thin liquid aspiration. Logemann et al. (2008), in an investigation of more than 700 elderly, frail patients with thin liquid aspiration on the VFS, found thickened liquids were superior at eliminating aspiration during a VFS study, than a chin-down posture.  More than 500 of these same patients, all aspirators, were followed for 3 months post VFS and randomized to receive thin liquids with a chin-down posture, or one of the two common thickened liquid textures during all meals. In this study, Robbins et al. (2008) found no statistically significant difference in pneumonia rates among patients randomized to thin liquids or the two different common thickened liquid textures.  Although not significant, patients who developed pneumonia while drinking thickened liquids, presumably aspirating less, when hospitalized for pneumonia, remained in hospital two to three-times longer than patients in the other groups.
  6. Aspiration is not fatal unless it is an airway obstruction. You aspirate. And feeding tubes do not eliminate aspiration – they are a second source of aspiration! Always use caution with immunocompromised people, especially lung transplantation. The rules are different.
  7. Intraoral seal is made anteriorly with lips and posteriorly with linguavelar closure, to enable subatmospheric pressure to develop in the mouth as the mandible depresses and the tongue flattens and retracts (Boyle’s Law!). To pull water uphill from a straw, these valved must be working. Linguavelar closure protects the oropharynx! Do Not Use in patients with vagal impairments – they may have to inhale to develop intraoral negative pressure because of palatoglossus paresis.
  8. Becker and colleagues (2008), in an unpublished study presented at the ASHA convention, had two deaths in their “Free Water Protocol” treatment group (2008) and none in the standard treatment group
  9. A screen is not a diagnostic test; they only identify risk. Blue dye tests are about 50% accurate (Brady, Hildner, & Hutchins, 1999).
  10. A stethoscope does not transmit the frequency range of sounds emanating from the throat during swallowing. The human auditory system cannot interpret the clicks and squirts heard in a stethoscope, reliably. That means that 2 clinicians would form different impressions based on the same observation (Coyle & Sejdić, 2020). Multiple studies have failed to identify agreement among listeners of the same recordings (Leslie, Drinnan, Finn, Ford, & Wilson, 2004).
  11. What is the clinical question? If it is “should I do a VFS for all diet upgrades?” and not “what would make this diet less safe or efficient given patient’s swallow physiology?” then they need to confirm their bias by repeating the VFS. Most diet changes are incremental.
  12. If you’ve seen one case with [insert diagnosis and problem list here], you have seen one case with [enter same diagnosis here]. No two patients are the same.

References

Becker, D. L., Tews, L. K., & Lemke, J. H. (2008, 2008/11/21/2008 ASHA Convention Technical Presentation). An oral water protocol in rehabilitation patients with dysphagia for liquids.

Brady, S. L., Hildner, C. D., & Hutchins, B. F. (1999). Simultaneous videofluoroscopic swallow study and modified Evans blue dye procedure: An evaluation of blue dye visualization in cases of known aspiration. Dysphagia., 14(3), 146-149.

Coyle, J. L., & Sejdić, E. (2020). High-Resolution Cervical Auscultation and Data Science: New Tools to Address an Old Problem. 29(2S), 992-1000. doi:doi:10.1044/2020_AJSLP-19-00155

Hart, T., Dijkers, M. P., Whyte, J., Turkstra, L. S., Zanca, J. M., Packel, A., . . . Chen, C. (2019). A theory-driven system for the specification of rehabilitation treatments. Arch Phys Med Rehabil, 100(1), 172-180. doi:10.1016/j.apmr.2018.09.109

Hutcheson, K. A., Barrow, M. P., Barringer, D. A., Knott, J. K., Lin, H. Y., Weber, R. S., . . . Lewin, J. S. (2017). Dynamic Imaging Grade of Swallowing Toxicity (DIGEST): Scale development and validation. Cancer, 123(1), 62-70. doi:10.1002/cncr.30283

Leslie, P., Drinnan, M. J., Finn, P., Ford, G. A., & Wilson, J. A. (2004). Reliability and validity of cervical auscultation: a controlled comparison using videofluoroscopy. Dysphagia, 19(4), 231-240.

Logemann, J. A. (1983). Evaluation and Treatment of Swallowing Disorders. San Diego, CA: College Hill Press.

Logemann, J. A. (1993). The dysphagia diagnostic procedure as a treatment efficacy trial. Clinics in Communication Disorders, 3(4), 1-10.

Logemann, J. A., Gensler, G., Robbins, J., Lindblad, A. S., Brandt, D., Hind, J. A., . . . Miller Gardner, P. J. (2008). A randomized study of three interventions for aspiration of thin liquids in patients with dementia or Parkinson's disease. Journal of Speech Language & Hearing Research, 51(1), 173-183.

Martin-Harris, B., Brodsky, M. B., Michel, Y., Castell, D. O., Schleicher, M., Sandidge, J., . . . Blair, J. (2008). MBS measurement tool for swallow impairment—MBSImp: Establishing a standard. Dysphagia, 23(4), 392-405. doi:10.1007/s00455-008-9185-9

Martin-Harris, B., Carson, K. A., Pinto, J. M., & Lefton-Greif, M. A. (2020). BaByVFSSImP© A Novel Measurement Tool for Videofluoroscopic Assessment of Swallowing Impairment in Bottle-Fed Babies: Establishing a Standard. Dysphagia, 35(1), 90-98. doi:10.1007/s00455-019-10008-x

Robbins, J., Coyle, J. L., Rosenbek, J. C., Roecker, E. B., & Wood, J. L. (1999). Differentiation of normal and abnormal airway protection during swallowing using the penetration-aspiration scale. Dysphagia, 14(4), 228-232.

Robbins, J., Gensler, G., Hind, J., Logemann, J. A., Lindblad, A. S., Brandt, D., . . . Miller Gardner, P. J. (2008). Comparison of 2 interventions for liquid aspiration on pneumonia incidence: a randomized trial. Annals of Internal Medicine, 148(7), 509-518.

Steele, C. M., Peladeau-Pigeon, M., Barbon, C. A. E., Guida, B. T., Namasivayam-MacDonald, A. M., Nascimento, W. V., . . . Wolkin, T. S. (2019). Reference Values for Healthy Swallowing Across the Range From Thin to Extremely Thick Liquids. J Speech Lang Hear Res, 62(5), 1338-1363. doi:10.1044/2019_jslhr-s-18-0448

Vose, A. K., Kesneck, S., Sunday, K., Plowman, E., & Humbert, I. (2018). A Survey of Clinician Decision Making When Identifying Swallowing Impairments and Determining Treatment. 61(11), 2735-2756. doi:doi:10.1044/2018_JSLHR-S-17-0212

Dr. Coyle is a medical SLP with 35 years' experience working with adults with dysphagia. He is professor of CSD, Otolaryngology, and Electrical and Computer Engineering at the University of Pittsburgh and an ASHA Fellow. He is funded by the NIH to investigate noninvasive technologies for use in dysphagia screening, assessment, and treatment.

Jo Puntil M.S. CCC-SLP BCS-S FASHA is a clinical specialist at St. George Regional Hospital. She has specialized in the management of dysphagia disorders, specifically in critical care for over 35 years. Jo Puntil has authored clinically based reference tools as well as chapters in college textbooks. She has developed numerous transdisciplinary dysphagia programs and has lectured nationally providing a more medically based understanding of the swallowing disordered patient.