The use of different outcomes across clinical trials on the same health condition may hamper comparability of findings and conduction of meta-analysis. This problem can be dealt with the development of a core set of outcome domains and measures. A core outcome domain is an aspect of health that should be measured and reported in every single trial assessing health interventions for the same health condition. A core outcome measure is a tool that should be used to measure and report on a core outcome domain. Deciding ‘what’ to measure should have priority over determining ‘how’ to measure, therefore core outcome domains should be determined before determining the core outcome measures [1,2]. Consensus should be achieved on both domains and measures by all relevant stakeholders (i.e. researchers, clinicians, patients, and others) [1,2]. Core domains or measures do not necessarily represent the primary outcomes of each single clinical trial, but outcomes to be collected in each trial to facilitate comparison of findings alongside trial-specific outcomes.
The problem of outcome inconsistency was identified already 20 years ago for low back pain (LBP, the most burdensome musculoskeletal condition!), when a group of experienced researchers proposed a set of core outcome domains and measures for clinical research [3]. In more recent years, developments in the methodology on how to establish core outcome domains and measures [1,2,4,5,6] have convinced another group of researchers (including some veterans of the ‘original’ set and two patient consumers) to update this 20-year old work [7]. A consensus exercise led to agreement on 3 core outcome domains for clinical trials in non-specific LBP: physical functioning, pain intensity, and health-related quality of life [8]. These results have been presented in this blog, see above link for a recap.
Four steps were taken to reach consensus on core outcome measures for LBP [9]. First, patient-reported outcome measures with potential (in terms of validity and feasibility) to measure each core outcome domain were selected. Second, three systematic reviews were conducted on the measurement properties (i.e. validity, reliability and responsiveness) of these measures in patients with LBP; one of these reviews was also recently published [10]. Third, a 2-round Delphi consensus exercise was performed to seek consensus and input from researchers, clinicians and patients on the measures. Fourth, the results of the Delphi were discussed by the group of researchers overseeing this international initiative to formulate recommendations.
LBP core outcome measures were defined and recently presented in the journal Pain [9]. The recommendation is to use:
- The Oswestry Disability Index version 2.1a or the 24-item Roland Morris Disability Questionnaire for physical functioning;
- An 11-point Numeric Rating Scale referring to average low back pain intensity over the last week for pain intensity;
- the Short Form Health Survey 12 or the 10-item PROMIS Global Health form for health-related quality of life [9].
It goes without saying that, to avoid redundancy, only one measure per domain should be adopted. Trial teams can choose the measure they prefer when there is a double choice. Thus, for example, the core outcome measures to be used in every LBP clinical trial could be Roland Morris, Numeric Rating Scale and PROMIS Global Health, especially if funds are tight as these are free. Hopefully, this brief set of recommendations will greatly facilitate LBP trials comparison in the years to come.
One may wonder whether these recommendations for LBP clinical trials are also applicable to LBP clinical practice… well, why not?! These measures were selected amongst those most studied, those with better measurement properties and those for which there is consensus for use across relevant stakeholders [10]. These are good reasons that they be incorporated into clinical practice. Indeed, the same recommendations (with some small amendments) were also presented in an article included in the back pain special issue of Best Practice & Research: Clinical Rheumatology [11].
Clinically, there are three other domains of high importance: work, psychological functioning and pain interference [11]. Consensus was almost reached to include these as core outcome domains [6]. Three questions, on current work ability, current work status and return to work are suggested for measuring work [11]. For psychological functioning, the Hospital Anxiety and Depression Scale is suggested because it has undergone extensive testing and can measure both anxiety and depression with relatively few (14!) items [11]. The pain interference subscale of the Brief Pain Inventory or the pain interference items of the Multidimensional Pain Inventory could be used to capture pain interference [11]. Our suggestion is that clinicians consider adding these three key domains and measures to their assessment of patients with LBP.
A last but important consideration that emerged from the consensus process on core outcome measures is that good quality head-to-head comparison studies for measures of the same domains are lacking [9, 11]. Studies that compare different measures of the same domain will help to determine if there is really a best measure for each domain (as this is not clear so far). Since there is lack of high quality evidence on outcome measures in patients with LBP [10], these studies will also help to fill this gap! Less known or more recently developed (but not recommended) tools should also be included in these comparison studies. For now, let’s build an understanding of LBP using the recent recommendations to facilitate between trial comparisons [9]. As Prof. Douglas Altman reported in the 2016 meeting of the Core Outcome Measures in Effectiveness Trials (COMET) initiative: “Trials should build on what’s already done; it’s not the place to be too novel; use a core outcome set if it exists; too much originality won’t help patients”.
About Alessandro Chiarotto
After working six years as a physiotherapist managing patients with musculoskeletal complaints, Alessandro decided that it was timely to move to clinical research. He has now just finished his PhD on the topic of outcome measurement for low back pain and will continue his work as a post-doctoral researcher at the Department of Epidemiology and Biostatistics of the VU University Medical Center in Amsterdam (Netherlands). His work is affiliated to two research institutes: Amsterdam Public Health and Amsterdam Movement Sciences. His main research interest are pain/musculoskeletal health and methodology. Being a dreamer, he would really like to make a substantial contribution to improve healthcare solutions for patients with musculoskeletal complaints. What he likes the most about his job is the possibility to take a step back and to look more objectively at the wide range of available therapeutic options, and to try to help clinicians and patients on which option to prefer over others.
References
[1] Boers M, Kirwan JR, Wells G, Beaton D, Gossec L, d’Agostino MA, Conaghan PG, Bingham CO, Brooks P, Landewè R, March L, Simon LS, Singh JA, Strand V, Tugwell P. Developing core outcome measurement sets for clinical trials: OMERACT filter 2.0. J Clin Epidemiol 2014;67(7):745-53. Doi: 10.1016/j.jclinepi.2013.11.013
[2] Williamson PR, Altman DG, Blazeby JM, Clarke M, Devane D, Gargon E, Tugwell P. Developing core outcome sets for clinical trials: issues to consider. Trials 2012;13:132. Doi: 10.1186/1745-6215-13-132
[3] Deyo RA, Battie M, Beurskens AJ, Bombardier C, Croft P, Koes B, Malmivaara A, Roland M, von Korff M, Waddell G. Outcome measures for low back pain research. A proposal for standardized use. Spine (Phila Pa 1976) 1998;23(18):2003-13
[4] Prinsen CA, Vohra S, Rose MR, Boers M, Tugwell P, Clarke M, Williamson PR, Terwee CB. How to select outcome measurement instruments for outcomes included in a “Core Outcome Set” – a practical guideline. Trials 2016;17(1):449. Doi: 10.1186/s13063-016-1555-2
[5] Williamson PR, Altman DG, Bagley H, Barnes KL, Blazeby JM, Brookes ST, Clarke M, Gargon E, Gorst S, Harman N, Kirkham JJ, McNair A, Prinsen CAC, Schmitt J, Terwee CB, Young B. Trials 2017;18(Suppl 3):280.
[6] Boers M, Kirwan JR, Tugwell P, Beaton D, Bingham II CO, Conaghan PG, D’Agostino MA, de Wit M, Gossec L, March L, Simon LS, Singh JA, Strand V, Wells GA. The OMERACT handbook 2017. Available at: https://www.dropbox.com/s/m5or42vadlwvbgz/OMERACT%20Handbook%20Oct%2012%202017.pdf?dl=0
[7] Chiarotto A, Terwee CB, Deyo RA, Boers M, Lin CW, Buchbinder R, Corbin TP, Costa LO, Foster NE, Grotle M, Koes BW, Kovacs FM, Maher CG, Pearson AM, Peul WC, Schoene ML, Turk DC, van Tulder MW, Ostelo RW. A core outcome set for clinical trials on non-specific low back pain: study protocol fort he development of a core domain set. Trials 2014;15;511. Doi: 10.1186/1745-6215-15-511
[8] Chiarotto A, Deyo RA, Terwee CB, Boers M, Buchbinder R, Corbin T, Costa LO, Foster NE, Grotle M, Koes BW, Kovacs FM, Lin CWC, Maher CG, Pearson AM, Peul WC, Schoene ML, Turk DC, van Tulder MW, & Ostelo RW. Core outcome domains for clinical trials in non-specific low back pain Eur Spine J 2015;1127-42. Doi: 10.1007/s00586-015-3892-3
[9] Chiarotto A, Boers M, Deyo RA, Buchbinder R, Corbin TP, Costa LOP, Foster NE, Grotle M, Koes BW, Kovacs FM, Lin CWC, Maher CG, Pearson AM, Peul WC, Schoene ML, Turk DC, van Tulder MW, Terwee CB, Ostelo RW. Core outcome measurement instruments for clinical trials in non-specific low back pain. Pain 2017; [Epub ahead of print]. Doi: 10.1097/j.pain.0000000000001117
[10] Chiarotto A, Ostelo RW, Boers M, Terwee CB. A systematic review highlights the need to investigate the content validity of patient-reported outcome measures for physical functioning in low back pain. J Clin Epidemiol 2017; [Epub ahead of print]. Doi: 10.1016/j.jclinepi.2017.11.005
[11] Chiarotto A, Terwee CB, Ostelo RW. Choosing the right outcome measurement instruments for patients with low back pain. Best Pract Res Clin Rheumatol 2016;30(6):1003-20. Doi: 10.1016/j.berh.2017.07.001