Abstract
Despite its widespread use and low complication rates, arthroscopic meniscectomy has not been uniformly successful in all patients, especially in those with concurrent osteoarthritis. The Patient-Reported Outcomes Measurement Information System (PROMIS) is an initiative funded by the National Institutes of Health to develop and validate patient-reported outcomes for clinical research and practice. PROMIS has shown the ability to enhance and standardize measurement of a variety of health domains affecting musculoskeletal function and in discriminating between various orthopaedic procedures through the use of computer adaptive testing. Preoperative PROMIS scores are valid predictors of postoperative minimal clinically important difference in patients undergoing arthroscopic meniscectomy based on preoperative decreased physical function and increased pain interference. PROMIS score cutoffs may be used by arthroscopic surgeons to counsel patients considering arthroscopic meniscectomy.
While the application of patient-reported outcome measures (PROMs) to various clinical settings is well documented across a number of medical disciplines, there is still a need to determine the clinical significance of PROMs and to consider how they can potentially affect clinical decision making. The minimal clinically important difference (MCID) is a common metric for generating clinical significance from patient-reported data and is useful for detecting the smallest difference in PROM scores that are meaningful to a number of stakeholders.
1
MCID can be defined from the patient’s perspective (i.e., a reduction in pain or improvement in function), from the physician’s perspective (i.e., an improvement in prognosis), or from society’s perspective (i.e., a faster return to employment or less expensive intervention). Therefore, given the diversity of perspectives, definitions of the MCID may be contradictory and mutually exclusive. From an individual clinical perspective, rather than focusing solely on whether a treatment outcome is merely statistically significant, MCID incorporates patients’ attitudes and priorities on their disease or clinical state to predict which patients will benefit from a particular intervention. With increasing momentum toward linking PROMs to reimbursement,2
clinical trials and evidence-based treatment algorithms, clinicians, investigators, and policy makers are increasingly interested in whether a particular treatment is actually meaningful and worthwhile to patients. This idea is especially true of arthroscopic meniscectomy—one of the most commonly performed orthopaedic procedures around the world.3
, 4
, 5
In their study entitled “Preoperative Patient-Centric Predictors of Postoperative Outcomes in Patients Undergoing Arthroscopic Meniscectomy,” Franovic, Kuhlman, Pietroski, Schlosser, Page, Okoroha, Moutzouros, and Makhni report their use of Patient-Reported Outcomes Measurement Information System and computer-adaptive testing (PROMIS-CAT)
6
scores to predict those patients who have an increased probability of improving from an arthroscopic partial meniscectomy. This study advances the PROMIS literature by establishing the clinical significance of PROMs and connecting them to clinical decision-making to predict operative benefit based on preoperative data. PROMIS-CAT provides real-time patient-reported outcomes in approximately 70 outcome domains. PROMIS relies on the paradigm of item response theory, which is a collection of mathematical models that assumes that responses to a set of items or questions are related to an unmeasured “trait” or domain.7
An example of such domains may be physical function (PF) or pain interference (PI). Item response theory assumes a person’s level of a specific domain will predict that person’s probability of endorsing each specific item related to that domain.7
CAT selects the best items to refine the estimate of a person’s level of function based on responses to previous questions. The resulting questionnaire is efficiently customized to ask the fewest questions that will accurately arrive at the person’s level of a given domain.8
These surveys give immediate outcomes regarding a patient’s disabilities or limitations caused by injury or illness and are standardized against the national population. A PROMIS score of 50 represents the average for the general population in the United States, with each 10-point increment indicating a difference of 1 standard deviation. In regard to meniscectomy, data have shown PROMIS-CAT is able to detect early changes in patient function, pain, and depression and provides prognostic value in predicting early outcomes after the procedure.9
It has been described as a powerful tool that provides highly responsive, clinically relevant outcome data.9
In their study, Franovic et al.
6
used the PROMIS domains of physical function (PROMIS-PF), pain interference (PROMIS-PI), and depression (PROMIS-D) surveys to measure patient outcomes both preoperatively and at 3 months postoperatively. All 3 domains showed statistically significant improvement at the 3-month follow-up. While multiple PROMIS domains may be considered, it has been previously suggested that PF is most specific and relevant to patients who undergo meniscal surgery.10
It is important to consider which domain is most specific to the intervention and desired outcome to avoid questionnaire fatigue whenever assessing PROM for either clinical care or research purposes. Not surprisingly, this study also found that patients who never smoked and who had minimal knee osteoarthritis had a greater likelihood of achieving PROMIS-PF MCID. Obesity (defined as a body mass index ≥30) was associated with significantly lower preoperative and postoperative PROMIS-PF and PI scores. However, both patients with and without obesity improved to a similar degree postoperatively. Given the influence of body weight on lower-extremity function, it is not surprising that patients with obesity had lower preoperative and postoperative scores compared to patients without obesity.Franovic et al.
6
used the distribution-based method to determine MCID values for each domain in contrast to an anchor-based approach. Anchor-based methods compare the change in patient-reported outcome to a second, external measure of change, or anchor that can be objective (i.e., number of pain pills used) or subjective (i.e., improvement in pain).11
Anchor-based calculations are not as suitable for populations where high post-treatment satisfaction is expected12
(such as following an arthroscopic partial meniscectomy) and they are susceptible to discordant results depending on whether the patient’s data were collected prospectively or retrospectively.13
Distribution-based methods rely on various statistical calculations of the obtained responses, which can result in different definitions of the MCID.14
Distribution-based methods can support estimates from anchor-based approaches and can be used in situations in which anchor-based estimates are unavailable.15
Given that each domain significantly improved after surgery, it is understandable why Franovic et al. chose this approach.The benefit of the current study is in the interpretation of what their results mean in a clinical scenario involving a patient presenting with a meniscal injury. Assuming the patient completed a baseline PROMIS survey, the surgeon would be able to discuss his or her probability of achieving a clinically significant result from surgery in relation to physical function, pain interference, or depression. This information would benefit the patient and surgeon in deciding whether to proceed with surgery based on relevant preoperative patient-specific characteristics.
This study builds on previous research on this topic and shows that clinically meaningful improvement in patient-derived outcomes is highly likely following an arthroscopic meniscectomy. Okoroha et al.
10
found that the MCID for PROMIS-PF was 2.09 in 73 patients undergoing arthroscopic meniscectomy. This was a smaller, less-comprehensive study than the investigation by Franovic at al.,6
who followed 135 patients and evaluated 3 PROMIS domains (PF, PI, and Dn) rather than just PF. It should also be noted that 44% of Okoroha et al.’s patients achieved MCID at 6 months compared with just 3 months in Franovic et al.’s study.6
Our group at Washington University demonstrated statistically significant improvements in PROMIS-PF, PROMIS-PI, and PROMIS-D in 75 patients undergoing arthroscopic meniscectomy.9
Preoperatively, our patients had decreased PF and increased PI in excess of 1 standard deviation from the general population. At the 6-week postoperative visit, patients had significant improvements in both PF and PI. Two findings of our study merit discussion: female patients had significantly less improvement in PROMIS-PF and PROMIS-D scores postoperatively compared with male patients, and patients with high-grade articular cartilage lesions had less improvement in both domains compared with patients with low-grade or no chondral lesion. We concluded that both PF and pain could be improved as early as 6 weeks following an arthroscopic meniscectomy, although patient and surgeon expectations should be tempered in female patients and in those with associated chondral lesions.Franovic et al.
6
correctly acknowledge some limitations of their study. These include those limitations inherent to PROMs, such as selection and response bias, where scores are likely to vary from sample to sample based on disease severity, general health, expectations, and sociodemographic characteristics.16
For example, this study excluded patients who were younger than 40 years of age, those undergoing a concomitant procedure, and patients who were unable to communicate in English. Outcomes also likely differ at time points other than at 3 months postoperatively, as evidenced by the results of Franovic et al.,6
Okoroha et al.,10
and our study.9
As mentioned earlier, Franovic et al.6
used a distribution-based method to calculate MCID versus an anchor-based method. Future studies involving PROMIS-derived MCID values based on different calculation methods should provide further refinement of this research tool. The effect of patient sex on PROMs also should be considered, since we have previously shown that female patients have less improvement (at least following an arthroscopic meniscectomy) than male patients. Research in sex-based differences following the treatment of a number of orthopaedic conditions is currently lacking but will likely expand as the use of PROMIS-CAT is more widely used.In conclusion, we congratulate Franovic et al. on presenting a valuable contribution to PROMIS research as it relates to arthroscopic partial meniscectomy.
6
This study adds to the growing body of outcomes literature by defining MCID values for 3 PROMIS-CAT domains—PF, PI, and D—and by prospectively identifying those patients who have a high probability of achieving clinically significant improvement from surgery based on their preoperative PROMIS scores. These findings are another step in applying PROMs to therapeutic decision-making. There are clearly more opportunities to expand the work in this field, such as establishing a consensus MCID for arthroscopic meniscectomy by comparing various calculation methods, increasing the generalizability of results to broader patient groups, and comparing the results following arthroscopic meniscectomy with physical therapy or other interventions. Given the current emphasis on clinical outcomes based on the patient’s perspective, use of preoperative predictors refined by CAT will predictably increase in all areas of medicine to guide treatment decisions. We look forward to future work by these authors in the expanding application of PROMIS-CAT to the field of arthroscopic surgery.Supplementary Data
- ICMJE author disclosure forms
References
- Measurement of health status.Controlled Clinical Trials. 1989; 10: 407-415
- The role of patient-reported outcome measures in value-based payment reform.Value Health. 2017; 20: 834-836
- Incidence and risk factors associated with meniscal injuries among active-duty US military service members.J Athletic Training. 2012; 47: 67-73
- Epidemiology of acute knee injuries: A prospective hospital investigation.J Trauma. 1991; 31: 1644-1648
- Top orthopedic sports medicine procedures.J Orthop Surg Res. 2018; 13: 190
- Preoperative patient-centric predictors of postoperative outcomes in patients undergoing arthroscopic meniscectomy.Arthroscopy. 2021; 37: 964-971
- Fundamentals of item response theory. Vol 2. Sage Publications, Inc., Newbury Park, CA1991
- Improving patient reported outcomes using item response theory and computerized adaptive testing.J Rheumatol. 2007; 34: 1426-1431
- Patient reported outcomes measurement information system scores are responsive to early changes in patient outcomes following arthroscopic partial meniscectomy.Arthroscopy. 2018; 34: 1113-1117
- How should we define clinically significant improvement on patient-reported outcomes measurement information system test for patients undergoing knee meniscal surgery?.Arthroscopy. 2020; 36: 241-250
- Interpretation of quality of life changes.Qual Life Res. 1993; 2: 221-226
- The minimal clinically important difference of the Michigan hand outcomes questionnaire.J Hand Surg Am. 2009; 34: 509-514
- Comparing retrospective and prospective anchors for identifying minimally important differences.Arthritis Res Ther. 2015; 17: 143
- Understanding the minimum clinically important difference: A review of concepts and methods.Spine J. 2007; 7: 541-546
- Recommended methods for determining responsiveness and minimally important differences for patient-reported outcomes.J Clin Epidemiol. 2008; 61: 102-109
- Editorial Commentary: Defining improvement after arthroscopic meniscal surgery—how much of a difference does a difference make?.Arthroscopy. 2020; 36: 251-252
Article info
Footnotes
See related article on page 964
The authors report the following potential conflicts of interest or sources of funding: M.J.M. reports other from Arthrex, Breg, Schwartz Biomedical, and Heron Therapeutics; and personal fees from Elite Orthopedics, outside the submitted work. Full ICMJE author disclosure forms are available for this article online, as supplementary material.
Identification
Copyright
© 2020 by the Arthroscopy Association of North America
