Why do our Interventions Result in Similar Outcomes?

By: Chad Cook PT, PhD, FAPTA; Derek Clewley PT, PhD, FAAOMPT

If you’ve seen the movie, Oppenheimer, you may remember him discussing the paradoxical wave-particle duality. This revolved around the finding that light exhibits both wave-like and particle-like properties. In fact, in certain experiments, light behaves more like a wave, whereas in others, it behaves more like a particle. Oppenheimer was perplexed because light shouldn’t have both properties, properties that seem to “depend” on how they are tested.

When you read comparative analyses involving two markedly different treatments that yield similar outcomes, it is likely that you are just as perplexed as Oppenheimer. As we’ve stated before in papers and blogs on this website and others, most musculoskeletal treatments result in similar overall outcomes [1]. In truth, it’s become the norm versus an exception. We could manage this using the current “circular firing squad” method of badmouthing the interventions we don’t like and supporting those we do, OR we can try to better understand why we are experiencing this. We chose the latter. The purpose of this blog is to provide possible reasons we see similar outcomes across studies involving different interventions.

Reason One: Our patient-reported outcomes measures, measure outcomes, not the results of a treatment. You likely remember this as the title of a paper by the illustrious Robert Herbert. Herbert and colleagues [2] suggested that a good outcome does not necessarily indicate that intervention was effective; that good outcome may have occurred even without intervention. Vice versa, the bad outcome may have occurred regardless of the intervention. When you package these variabilities into a trial, it makes it difficult to show differences.

Reason Two: Our patients’ outcomes are influenced by factors outside the treatment we are delivering. Clinical outcomes are influenced by many factors other than intervention, including the natural course of the condition, statistical regression, placebo effects, and so on [2]. Our previous work addressing social factors show that there is often an artificial ceiling on the progress one makes with their outcomes, regardless of the treatment [3,4]. Others have suggested the importance of behavioral factors on outcomes [5].

Reason Three: Our patient-reported outcomes measures may not represent every patients’ disability or recovery status in the same way. Content validity is the extent to which the items on a scale represent the entire domain the scale seeks to measure. Content validity may be the most important element of an outcomes scale, and it is also the most lacking, especially in legacy measures [6]. We’ve all experienced a patient’s completion of a patient reported outcome measure, only for them to tell us that the questions didn’t represent their condition at all.

Reason Four: Patients with similar diagnoses experience pain in different ways [7]. One’s pain experience is influenced by a complex interplay of factors, such as the physical nature, intensity, and location of the pain, psychological factors such as prior experience, fear, patient expectations, cultural and social influences, and the context in which a person experiences pain. Further, differences in pain mechanisms may result in differences in pain experience.

Reason Five: Patients respond differently to efficacious interventions, which is a source of consideration consternation among clinicians. We’d like to think there is consistency in how a person responds to an intervention. Unfortunately, this isn’t the case. Even pharmacological agents [8] have different levels of effectiveness because of the unique way patients respond to different treatments.

Reason Six: The context of how care is delivered plays a large role in the effectiveness of the care provided [9]. Contextual factors encompass patient and provider personal (e.g., race/ethnicity, genetic variables, expectations, values and preference), historical (e.g., clinical history, prior experiences), cultural (e.g., social norms, spirituality/religion and power differentials), environmental (e.g., settings and rituals), physical (e.g., sensorial perception, clinical examination and modalities in which the therapy is delivered), and rhetorical (e.g., verbal and non-verbal communication) dimensions around the therapeutic encounter and the patient-clinician interaction influencing moderators/mediators of therapeutic mechanisms and the response to any interventions/treatments and ultimately, the overall clinical outcomes. Can we realistically assume a similar context with the variety of patients that we treat?

Reason Seven: Unlike mortality, which is a clear and concise measure, morbidity is much murkier, since it is reported differently across individuals [10]. By definition, morbidity reflects the extent of suffering from a disease or medical condition. As we’ve discussed previously in this blog, people have differences in pain experiences and respond differently to interventions. We’ve all experienced numerous cases where individuals with the same diagnosis, the same mechanisms of injury, and the same apparent limitations, respond notably different in their recovery cycle.

How Can We Incorporate this Knowledge to Clinical Practice? Sometimes, understanding is the first step towards change. By recognizing that there are multiple reasons to see variation in patient outcomes, we can better recognize how to navigate our increasingly challenging caseloads. A “one size fits all” care approach will not likely work, nor should we expect it to. We didn’t delve into the likelihood that differences in prescribed dosages of our treatments and how adherent patients are to each likely also leads to outcome variability. There are likely many additional reasons we see similar outcomes in how we measure our patient’s treatment effectiveness, and we invite comments below. These factors suggest that the patient in front of you won’t automatically improve from one treatment, or another, but may need a combination of unique treatment options to optimize improvement. This requires an adaptive, lateral-thinking clinician, one who can work with what they have.

References

1. Cook C. (Blog) Thorough research questions should have layers. https://sites.duke.edu/cemmt/2022/07/25/thorough-research-questions-should-have-layers/.
2. Herbert R, Jamtvedt G, Mead J, Hagen KB. Outcome measures measure outcomes, not effects of intervention. Aust J Physiother. 2005;51(1):3-4. doi: 10.1016/s0004-9514(05)70047-7.
3. Buck E, Rethorn ZD, Garcia AN, Cook CE, Gottfried O. The Collective Influence of Social Determinants of Health on Individuals Who Underwent Lumbar Spine Revision Surgeries: A Retrospective Cohort Study. World Neurosurg. 2022 Sep;165:e619-e627.
4. Rethorn ZD, Cook CE, Park C, Somers T, Mummaneni PV, Chan AK, Pennicooke BH, Bisson EF, Asher AL, Buchholz AL, Bydon M, Alvi MA, Coric D, Foley KT, Fu KM, Knightly JJ, Meyer S, Park P, Potts EA, Shaffrey CI, Shaffrey M, Than KD, Tumialan L, Turner JD, Upadhyaya CD, Wang MY, Gottfried O. Social risk factors predicting outcomes of cervical myelopathy surgery. J Neurosurg Spine. 2022 Jan 28:1-8.
5. McGinnis JM, Williams-Russo P, Knickman JR. The case for more active policy attention to health promotion. Health Aff (Millwood). 2002 Mar-Apr;21(2):78-93.
6. Pellekooren S, Ostelo R, Pool A, van Tulder M, Jansma E, Chiarotto A. Content Validity of Patient-Reported Outcome Measures of Satisfaction With Primary Care for Musculoskeletal Complaints: A Systematic Review. J Orthop Sports Phys Ther. 2021 Mar;51(3):94-102. doi: 10.2519/jospt.2021.9788.
7. Fillingim RB. Individual differences in pain: understanding the mosaic that makes pain personal. Pain. 2017 Apr;158 Suppl 1(Suppl 1):S11-S18.
8. Edwards RR, Dworkin RH, Turk DC, Angst MS, Dionne R, Freeman R, Hansson P, Haroutounian S, Arendt-Nielsen L, Attal N, Baron R, Brell J, Bujanover S, Burke LB, Carr D, Chappell AS, Cowan P, Etropolski M, Fillingim RB, Gewandter JS, Katz NP, Kopecky EA, Markman JD, Nomikos G, Porter L, Rappaport BA, Rice ASC, Scavone JM, Scholz J, Simon LS, Smith SM, Tobias J, Tockarshewsky T, Veasley C, Versavel M, Wasan AD, Wen W, Yarnitsky D. Patient phenotyping in clinical trials of chronic pain treatments: IMMPACT recommendations. Pain. 2016 Sep;157(9):1851-1871.
9. Cook CE, Bailliard A, Bent JA, Bialosky JE, Carlino E, Colloca L, Esteves JE, Newell D, Palese A, Reed WR, Vilardaga JP, Rossettini G. An international consensus definition for contextual factors: findings from a nominal group technique. Front Psychol. 2023 Jul 3;14:1178560.
10. Audibert M. Issues and Challenges of Measurement of Health: Implications for Economic Research. 2011. ffhalshs-00554267f