Recently I was engaged in a conversation related to the ability of a speech pathologist to clinically determine swallowing pathophysiology. A therapist responded she had paid a great deal of money for her degree and if she could not rely on her “clinical judgement” to accurately diagnose patients with dysphagia she had wasted her money. I was not sure how to respond. However, after some reflection, here is my thoughtful response.
First I will say, I hope that in every impairment based course offered in school emphasized the need for objective assessments. When a child is referred for articulation disorders we administer the Goldman-Fristoe, Khan-Lewis, or some other such assessment—or these are some of the tested I used in the Dark Ages—to quantify the impairment even though the deficits are quite clear from simply listening to the child speak. When a child is referred for intellectual disabilities do we simply spend a bit of time with the child and try to quantify the level of disability? No, an IQ test is required. If a child has been identified with a hearing loss we do just give a label and move on? No, a referral is made for audiological testing. The same holds true for childhood language disorders, aphasia, cognition, fluency, etc. I am not sure where this concept was lost when it comes to dysphagia.
Please know, this is not just a problem for speech pathologists treating dysphagia. It is a universal issue in all of medicine. “Dual Processing Theory” is a widely accepted as a dominant explanation of cognitive processes that characterizes human decision-making. This theory postulates that reasoning and decision-making can be described as a function of both an intuitive, experiential, affective system (System I) and/or an analytical, deliberative (System II) processing system (Stanovich 2011, Croskerry 2009, Evans 2007, Stanovich 2000). When providing a diagnosis using System I processing alone, medical professionals are influenced by factors that may be irrelevant such as gender (Borkhoff 2008), race (Green 2007), obesity (Hebl 2001), history of psychiatric illness (Daumit 2006), and age (Podplsky 1993). Typically when making clinical decisions no account is taken of ambient conditions, such as other cases being managed concurrently, team dynamics, fatigue, sleep deprivation, location, and other variables critical to performance (Croskerry 2009).
This concept could be applied to dysphagia. Imagine, for instance, a clinician taking a break in the middle of a work day to have lunch with a friend. During the course of lunch the friend takes a drink and begins to cough. Both people laugh and make a joke about aspiration, and the meal continues without concern. Upon returning to work, a nurse comes to the clinician reporting that patient X coughed during lunch. An order is requested for a speech evaluation, and in the interim, for safety, the patient is placed on nectar thick liquids. Why were to two situations handled differently? There are many possible answers. However, much of it has to do with biases of location, prescribed roles, etc.
The “Cognitive Response Test” described by Frederick (2005) found that errors are quite predictable, even when people are presented with simple questions. These predictable errors can be explained, in part, by the brain’s “cognitive miser” function (Croskerry 2009). The tendency to limit cognitive effort in reasoning as a kind of energy saving technique, is an often overlooked property of the brain (Krueger 2004).
In contrast to System I processing, System II processing is a robust decision making paradigm that is more analytical than intuitive. It is a systematic approach that leads to making effective decisions. It is typically analytical, slow and resource intensive. However, it is more likely to end with a correct diagnosis/decision (Croskerry 2009).
Let’s consider some additional applications of the Dual Processing Theory to dysphagia treatment.
Example 1:
Vose (2018) provided SLPs with video clips of one swallow. Clinicians were asked to identify the swallow impairments. In one clip the obvious abnormal physiology was a delay in the pharyngeal trigger (20+ seconds). Although the swallow delay was quite compelling, 33% of the respondents did not identify the delay as the primary impairment. 67% of therapists identified the delay as the primary issue, but only 58% said this would be the focus of treatment. In addition, 77% of respondents indicated there were 5 -9 impairments and 27% indicated there were 10 or more issues. In this case the use of System I processing would have caused the patient to be both misdiagnosed and given inappropriate treatment. If a System II approach had been used, the clinician would use quantifiable measures to analyze the videos, employing a systematic approach such as measuring the delay using a frame by frame analysis. Once the pathophysiology was determined, the clinician would reference literature (or rely on having referenced literature previously) to determine possible treatments.
Example 2:
When performing a swallow study, the clinician observes the patient has laryngeal penetration at a PAS of 2 and 3. When documenting the events the therapist transcribes, “The patient displayed silent penetration.” The term “silent penetration” would suggest that not responding to laryngeal penetration is pathological. In fact a review of the afferent innervation of the airway teaches that the hypopharynx is innervated by the internal branch of the superior laryngeal nerve. When stimulated the nerve facilitates a swallow, not a cough (Mazzone 2016). This causes the material to be ejected from the laryngeal vestibule disallowing aspiration. In a System I approach the patient might be placed on altered liquids due to the concern of aspiration “risk” even though their body acted in a healthy manner. Investigation of the literature concerning basic neuroanatomy would yield a different result (System II approach).
Example 3:
When Plowman (2018) provided clinicians with five swallows that had been recorded on videofluoroscopy, clinicians were asked to determine if the swallow was “normal” or “abnormal.” As an average, 34% of clinicians labeled each of the 5 swallows as “impaired” (range 54% to 6%). In fact, all of the swallows were performed by healthy graduate students. In a System I approach the clinicians most likely hypothesized that some of the swallows were impaired and randomly assigned impairments. This is the same rationale the SLP may employ when a patient is referred for a swallow study. Instead of simply reviewing anatomy and physiology, the therapist is looking for something wrong. A critical analysis of a swallow requires employing quantifiable measures of both temporal and kinematic events and recognizing the internal bias to “find” issues that do not actually exist. This approach would lead to a more reliable diagnosis.
Ahh, I bet you thought I was going to indicate imaging reflected a System II approach. I would say it only becomes a System II approach if appropriate analysis is applied. Imagining alone is grossly insufficient. If the therapist simply views imaging without the use of quantifiable measures the information could be no more accurate than a clinical swallow assessment (Brates 2018).
In all three examples, System I processing led to errant decision making. However, the facility level clinician is not alone. This occurs at all levels of expertise.
Lee (2017) studied the ability of experts in the area of swallowing and swallowing disorders to rate impairments in swallow physiology when only viewing and rating modified barium swallow studies verses using quantifiable measures. Evaluators correctly classified only 61.5% of MBSS videos as normal or abnormal. Only 28% of the time did evaluators unanimously agree on a correct interpretation of a MBSS. This indicates that lack of exposure or experience is not the issue. The issue is that the physiology of the swallow is too fast and there is a vast range of “normal” associated with the small movements, making it impossible for the human eye to make accurate assessments when watching it in real time.
To drive home the point of the overreliance on System I processing, a study investigated how clinicians make decisions concerning treatment (McCurtin 2017). Respondents explained treatment is determined based on the following:
|
Reason |
% of therapists responding |
|
It is suitable for my clients |
31.8% |
|
Based on my clinical experience |
28.2% |
|
It is theoretically sound |
8.3% |
|
Experts recommend |
5.1% |
|
I learned it at university |
4.9% |
Note that a total of 13.4% are suggesting the use of System II processing.
When analyzing a list of the recommendations these same therapists most often utilize, the twelve most employed interventions were not exercise based, leading the authors to conclude:
“Reasoning, rather than being complex, appears relatively simplified, with a limited range of reasoning displayed generally and irrespective of the respondent characteristics. The original categories which guided survey development can be refined to a few core reasons centered on client suitability and knowledge. Knowledge reflects both absent knowledge, evident for example in lack of training or accumulated knowledge evident in clinical experience… Evidence-based practice is a model which does not accurately reflect the clinical reasoning underpinning treatment selections…”
One of the most compelling concerns related to reliance on System I processing comes from Croskerry 2009:
“Autopsy findings have consistently shown a 20% to 40% discrepancy with the antemortem diagnosis, and a third of these autopsies would not have taken place if the true diagnosis had been known. Despite improved technology and an improved evidence base in medicine, the misdiagnosis rate detected through autopsy studies has not changed significantly during the last century”
Think of all the medical advances in the past 100 years. Now consider that the over use of System I processing continues to override these advances when patients are being diagnosed.
Plowman (2018) enumerates some possible reasons we depend of System I processing. (This could be another blog post; instead I will simply list the reasons and think about expanding on them another day.)
- Education has a focus on the disordered system leaving clinicians with a poor understanding of a “normal” swallow.
- Clinicians are trying to conserve cognitive energy (System I processing is easier and faster).
- Swallowing is complex and the consequences of swallowing impairment are more complicated than we understand.
- The inability to visualize the swallowing processes, clinically and the limited exposure to “normal” when performing imaging.
So, is there a role for clinical judgement? Absolutely! We need to employ sound clinical judgement, based in science—not what someone else taught, or our “gut”—to formulate questions to shape our investigations. We can accumulate knowledge, across time, to help guide us. We should also superimpose a strong knowledge of normal anatomy and physiology into our observations/assessments. At the end of the day, we must acknowledge the literature is quite clear that relying on our clinical judgement alone can lead to negative consequences for our patients. System II processing requires additional time and energy. It requires stepping back and analyzing the data using measureable and systematic methods. This is time well spent because, in the end, we owe it to our patients to get it right!
Borkhoff, C. M., Hawker, G. A., Kreder, H. J., Glazier, R. H., Mahomed, N. N., & Wright, J. G. (2008). The effect of patients' sex on physicians' recommendations for total knee arthroplasty. CMAJ, 178(6), 681-687. doi:10.1503/cmaj.071168
Borkhoff, C. M., Hawker, G. A., Kreder, H. J., Glazier, R. H., Mahomed, N. N., & Wright, J. G. (2013). Influence of patients' gender on informed decision making regarding total knee arthroplasty. Arthritis Care Res (Hoboken), 65(8), 1281-1290. doi:10.1002/acr.21970
Brates, D., Molfenter, S. M., & Thibeault, S. L. (2018). Assessing Hyolaryngeal Excursion: Comparing Quantitative Methods to Palpation at the Bedside and Visualization During Videofluoroscopy. Dysphagia. doi:10.1007/s00455-018-9927-2
Croskerry, P. (2009a). A universal model of diagnostic reasoning. Acad Med, 84(8), 1022-1028. doi:10.1097/ACM.0b013e3181ace703
Croskerry, P. (2009b). Clinical cognition and diagnostic error: applications of a dual process model of reasoning. Adv Health Sci Educ Theory Pract, 14 Suppl 1, 27-35. doi:10.1007/s10459-009-9182-2
Daumit, G. L., Pronovost, P. J., Anthony, C. B., Guallar, E., Steinwachs, D. M., & Ford, D. E. (2006). Adverse events during medical and surgical hospitalizations for persons with schizophrenia. Arch Gen Psychiatry, 63(3), 267-272. doi:10.1001/archpsyc.63.3.267
Evans, J. (2007) Hypothetical thinking. Dual processes in reasoning and judgment. Psychology Press: Taylor and Francis Group, New York.
Fredrick, S. (2005). Cognitive reflection and decision.
Green, A. R., Carney, D. R., Pallin, D. J., Ngo, L. H., Raymond, K. L., Iezzoni, L. I., & Banaji, M. R. (2007). Implicit bias among physicians and its prediction of thrombolysis decisions for black and white patients. J Gen Intern Med, 22(9), 1231-1238. doi:10.1007/s11606-007-0258-5
Hebl, M. R., & Xu, J. (2001). Weighing the care: physicians' reactions to the size of a patient. Int J Obes Relat Metab Disord, 25(8), 1246-1252. doi:10.1038/sj.ijo.0801681
Krueger, J. I., & Funder, D. C. (2004a). Towards a balanced social psychology: causes, consequences, and cures for the problem-seeking approach to social behavior and cognition. Behav Brain Sci, 27(3), 313-327; discussion 328-376.
Krueger, J. I., & Funder, D. C. (2004b). Towards a balanced social psychology: causes, consequences, and cures for the problem-seeking approach to social behavior and cognition. Behav Brain Sci, 27(3), 313-327; discussion 328-376.
Lee, J. W., Randall, D. R., Evangelista, L. M., Kuhn, M. A., & Belafsky, P. C. (2017). Subjective Assessment of Videofluoroscopic Swallow Studies. Otolaryngol Head Neck Surg, 156(5), 901-905. doi:10.1177/0194599817691276
Mazzone, S. B., & Undem, B. J. (2016). Vagal Afferent Innervation of the A
irways in Health and Disease. Physiol Rev, 96(3), 975-1024. doi:10.1152/physrev.00039.2015
McCurtin, A., & Healy, C. (2017). Why do clinicians choose the therapies and techniques they do? Exploring clinical decision-making via treatment selections in dysphagia practice. Int J Speech Lang Pathol, 19(1), 69-76. doi:10.3109/17549507.2016.1159333
Plowman, E. K., & Humbert, I. A. (2018). Elucidating inconsistencies in dysphagia diagnostics: Redefining normal. Int J Speech Lang Pathol, 20(3), 310-317. doi:10.1080/17549507.2018.1461931
Podolsky, D., & Silberner, J. (1993). How medicine mistreats the elderly. US News World Rep, 114(2), 72-76, 78-79.
Stanovich, K., (2011). Rationality and the reflective mind. Oxford University Press, Oxford.
Stanovich, K. E., & West, R. F. (2000). Individual differences in reasoning: implications for the rationality debate? Behav Brain Sci, 23(5), 645-665; discussion 665-726.
Vose, A., Nonnenmacher, J., Singer, M. L., & González-Fernández, M. (2014). Dysphagia Management in Acute and Sub-acute Stroke. Curr Phys Med Rehabil Rep, 2(4), 197-206. doi:10.1007/s40141-014-0061-2
