Clinical predictive modelling of post-surgical recovery in individuals with cervical radiculopathy: a machine learning approach.
Journal
Scientific reports
ISSN: 2045-2322
Titre abrégé: Sci Rep
Pays: England
ID NLM: 101563288
Informations de publication
Date de publication:
08 10 2020
08 10 2020
Historique:
received:
13
05
2020
accepted:
18
09
2020
entrez:
9
10
2020
pubmed:
10
10
2020
medline:
15
12
2020
Statut:
epublish
Résumé
Prognostic models play an important role in the clinical management of cervical radiculopathy (CR). No study has compared the performance of modern machine learning techniques, against more traditional stepwise regression techniques, when developing prognostic models in individuals with CR. We analysed a prospective cohort dataset of 201 individuals with CR. Four modelling techniques (stepwise regression, least absolute shrinkage and selection operator [LASSO], boosting, and multivariate adaptive regression splines [MuARS]) were each used to form a prognostic model for each of four outcomes obtained at a 12 month follow-up (disability-neck disability index [NDI]), quality of life (EQ5D), present neck pain intensity, and present arm pain intensity). For all four outcomes, the differences in mean performance between all four models were small (difference of NDI < 1 point; EQ5D < 0.1 point; neck and arm pain < 2 points). Given that the predictive accuracy of all four modelling methods were clinically similar, the optimal modelling method may be selected based on the parsimony of predictors. Some of the most parsimonious models were achieved using MuARS, a non-linear technique. Modern machine learning methods may be used to probe relationships along different regions of the predictor space.
Identifiants
pubmed: 33033308
doi: 10.1038/s41598-020-73740-7
pii: 10.1038/s41598-020-73740-7
pmc: PMC7545179
doi:
Types de publication
Journal Article
Randomized Controlled Trial
Research Support, Non-U.S. Gov't
Langues
eng
Sous-ensembles de citation
IM
Pagination
16782Références
Murray, C. J. et al. The state of US health, 1990–2010: burden of diseases, injuries, and risk factors. J. Am. Med. Assoc. 310, 591–608. https://doi.org/10.1001/jama.2013.13805 (2013).
doi: 10.1001/jama.2013.13805
Radhakrishnan, K., Litchy, W. J., O’Fallon, W. M. & Kurland, L. T. Epidemiology of cervical radiculopathy. A population-based study from Rochester, Minnesota, 1976 through 1990. Brain 117(Pt 2), 325–335 (1994).
doi: 10.1093/brain/117.2.325
Thoomes, E. J., Scholten-Peeters, W., Koes, B., Falla, D. & Verhagen, A. P. The effectiveness of conservative treatment for patients with cervical radiculopathy: a systematic review. Clin. J. Pain 29, 1073–1086. https://doi.org/10.1097/AJP.0b013e31828441fb (2013).
doi: 10.1097/AJP.0b013e31828441fb
pubmed: 23446070
Bono, C. M. et al. An evidence-based clinical guideline for the diagnosis and treatment of cervical radiculopathy from degenerative disorders. Spine J. 11, 64–72. https://doi.org/10.1016/j.spinee.2010.10.023 (2011).
doi: 10.1016/j.spinee.2010.10.023
pubmed: 21168100
Hemingway, H. et al. Prognosis research strategy (PROGRESS) 1: a framework for researching clinical outcomes. BMJ 346, e5595. https://doi.org/10.1136/bmj.e5595 (2013).
doi: 10.1136/bmj.e5595
pubmed: 23386360
pmcid: 3565687
Steyerberg, E. W. et al. Prognosis research strategy (PROGRESS) 3: prognostic model research. PLOS Med. 10, e1001381. https://doi.org/10.1371/journal.pmed.1001381 (2013).
doi: 10.1371/journal.pmed.1001381
pubmed: 23393430
pmcid: 3564751
da Costa, M. C. L. et al. The prognosis of acute and persistent low-back pain: a meta-analysis. CMAJ Can. Med. Assoc. J. journal de l’Association medicale canadienne 184, E613-624. https://doi.org/10.1503/cmaj.111271 (2012).
doi: 10.1503/cmaj.111271
da Silva, T. et al. Risk of recurrence of low back pain: a systematic review. J Orthop Sports Phys Ther 47, 305–313. https://doi.org/10.2519/jospt.2017.7415 (2017).
doi: 10.2519/jospt.2017.7415
pubmed: 28355981
Kelly, J., Ritchie, C. & Sterling, M. Clinical prediction rules for prognosis and treatment prescription in neck pain: a systematic review. Musculoskelet. Sci. Pract. 27, 155–164. https://doi.org/10.1016/j.math.2016.10.066 (2017).
doi: 10.1016/j.math.2016.10.066
pubmed: 27852530
Wong, J. J., Cote, P., Quesnele, J. J., Stern, P. J. & Mior, S. A. The course and prognostic factors of symptomatic cervical disc herniation with radiculopathy: a systematic review of the literature. Spine J. 14, 1781–1789. https://doi.org/10.1016/j.spinee.2014.02.032 (2014).
doi: 10.1016/j.spinee.2014.02.032
pubmed: 24614255
Sleijser-Koehorst, M. L. S. et al. Clinical course and prognostic models for the conservative management of cervical radiculopathy: a prospective cohort study. Eur. Spine J. 27, 2710–2719. https://doi.org/10.1007/s00586-018-5777-8 (2018).
doi: 10.1007/s00586-018-5777-8
pubmed: 30327908
Peolsson, A., Vavruch, L. & Öberg, B. Predictive factors for arm pain, neck pain, neck specific disability and health after anterior cervical decompression and fusion. Acta Neurochir. 148, 167–173. https://doi.org/10.1007/s00701-005-0660-x (2006).
doi: 10.1007/s00701-005-0660-x
pubmed: 16341632
Peolsson, A. & Peolsson, M. Predictive factors for long-term outcome of anterior cervical decompression and fusion: a multivariate data analysis. Eur. Spine J. 17, 406–414. https://doi.org/10.1007/s00586-007-0560-2 (2008).
doi: 10.1007/s00586-007-0560-2
pubmed: 18084782
Chester, R., Jerosch-Herold, C., Lewis, J. & Shepstone, L. Psychological factors are associated with the outcome of physiotherapy for people with shoulder pain: a multicentre longitudinal cohort study. BJSM Online 52, 269. https://doi.org/10.1136/bjsports-2016-096084 (2018).
doi: 10.1136/bjsports-2016-096084
15Harrell, F. E., Jr. In Regression modeling strategies. With applications to linear models, logistic and ordinal regression, and survival analysis. Springer series in statisticsCh. 4, 67 (2015).
Tibshirani, R. Regression shrinkage and selection via the Lasso. J. R. Stat. Soc. Ser. B Methodol. 58, 267–288 (1996).
Friedman, J., Hastie, T. & Tibshirani, R. Additive logistic regression: a statistical view of boosting (with discussion and a rejoinder by the authors). Ann. Statist. 28, 337–407. https://doi.org/10.1214/aos/1016218223 (2000).
doi: 10.1214/aos/1016218223
Friedman, J. H. Multivariate adaptive regression splines. Ann. Stat. 19, 1–67. https://doi.org/10.1214/aos/1176347963 (1991).
doi: 10.1214/aos/1176347963
Sanchez-Pinto, L. N., Venable, L. R., Fahrenbach, J. & Churpek, M. M. Comparison of variable selection methods for clinical predictive modeling. Int. J. Med. Inform. 116, 10–17. https://doi.org/10.1016/j.ijmedinf.2018.05.006 (2018).
doi: 10.1016/j.ijmedinf.2018.05.006
pubmed: 29887230
pmcid: 6003624
Churpek, M. M. et al. Multicenter comparison of machine learning methods and conventional regression for predicting clinical deterioration on the wards. Crit. Care Med. 44, 368–374. https://doi.org/10.1097/CCM.0000000000001571 (2016).
doi: 10.1097/CCM.0000000000001571
pubmed: 4736499
pmcid: 4736499
Weng, S. F., Reps, J., Kai, J., Garibaldi, J. M. & Qureshi, N. Can machine-learning improve cardiovascular risk prediction using routine clinical data?. PLoS ONE 12, e0174944. https://doi.org/10.1371/journal.pone.0174944 (2017).
doi: 10.1371/journal.pone.0174944
pubmed: 28376093
pmcid: 5380334
Young, I. A., Cleland, J. A., Michener, L. A. & Brown, C. Reliability, construct validity, and responsiveness of the neck disability index, patient-specific functional scale, and numeric pain rating scale in patients with cervical radiculopathy. Am. J. Phys. Med. Rehabil. 89, 831–839. https://doi.org/10.1097/PHM.0b013e3181ec98e6 (2010).
doi: 10.1097/PHM.0b013e3181ec98e6
pubmed: 20657263
Breiman, L. Statistical modeling: the two cultures (with comments and a rejoinder by the author). Stat. Sci. 16, 199–231. https://doi.org/10.1214/ss/1009213726 (2001).
doi: 10.1214/ss/1009213726
Berk, R., Brown, L., Buja, A., Zhang, K. & Zhao, L. Valid post-selection inference. Ann. Stat. 41, 802–837. https://doi.org/10.1214/12-AOS1077 (2013).
doi: 10.1214/12-AOS1077
Rügamer, D. & Greven, S. Selective inference after likelihood- or test-based model selection in linear models. Stat. Prob. Lett. 140, 7–12. https://doi.org/10.1016/j.spl.2018.04.010 (2018).
doi: 10.1016/j.spl.2018.04.010
Peolsson, A., Hedlund, R. & Vavruch, L. Prediction of fusion and importance of radiological variables for the outcome of anterior cervical decompression and fusion. Eur. Spine J 13, 229–234. https://doi.org/10.1007/s00586-003-0627-7 (2004).
doi: 10.1007/s00586-003-0627-7
pubmed: 14714242
pmcid: 3468143
Buckland, S. T., Burnham, K. P. & Augustin, N. H. Model selection: an integral part of inference. Biometrics 53, 603–618. https://doi.org/10.2307/2533961 (1997).
doi: 10.2307/2533961
Zhao, J. et al. Learning from longitudinal data in electronic health record and genetic data to improve cardiovascular event prediction. Sci. Rep. 9, 717. https://doi.org/10.1038/s41598-018-36745-x (2019).
doi: 10.1038/s41598-018-36745-x
pubmed: 30679510
pmcid: 6345960
Breiman, L. Bagging predictors. Mach. Learn. 24, 123–140. https://doi.org/10.1023/A:1018054314350 (1996).
doi: 10.1023/A:1018054314350
Schapire, R. E. The strength of weak learnability. Mach. Learn. 5, 197–227. https://doi.org/10.1007/BF00116037 (1990).
doi: 10.1007/BF00116037
Molgaard Nielsen, A. et al. Exploring conceptual preprocessing for developing prognostic models: a case study in low back pain patients. J Clin Epidemiol 122, 27–34. https://doi.org/10.1016/j.jclinepi.2020.02.005 (2020).
doi: 10.1016/j.jclinepi.2020.02.005
pubmed: 32097713
Vernon, H. The neck disability index: state-of-the-art, 1991–2008. J. Manip. Physiol. Ther. 31, 491–502. https://doi.org/10.1016/j.jmpt.2008.08.006 (2008).
doi: 10.1016/j.jmpt.2008.08.006
Sterling, M., Hendrikz, J. & Kenardy, J. Compensation claim lodgement and health outcome developmental trajectories following whiplash injury: a prospective study. Pain 150, 22–28. https://doi.org/10.1016/j.pain.2010.02.013 (2010).
doi: 10.1016/j.pain.2010.02.013
pubmed: 20307934
Sturgeon, J. A. et al. Nonlinear effects of noxious thermal stimulation and working memory demands on subjective pain perception. Pain Med. 16, 1301–1310. https://doi.org/10.1111/pme.12774 (2015).
doi: 10.1111/pme.12774
pubmed: 25929747
Peolsson, A., Vavruch, L. & Öberg, B. Disability after anterior decompression and fusion for cervical disc disease. Adv. Physiother. 4, 111–124. https://doi.org/10.1080/140381902320387531 (2002).
doi: 10.1080/140381902320387531
Anderson, P. A., Subach, B. R. & Riew, K. D. Predictors of outcome after anterior cervical discectomy and fusion: a multivariate analysis. Spine (Phila Pa 1976) 34, 161–166. https://doi.org/10.1097/BRS.0b013e31819286ea (2009).
doi: 10.1097/BRS.0b013e31819286ea
Peolsson, A., Hedlund, R., Vavruch, L. & Oberg, B. Predictive factors for the outcome of anterior cervical decompression and fusion. Eur. Spine J. 12, 274–280. https://doi.org/10.1007/s00586-003-0530-2 (2003).
doi: 10.1007/s00586-003-0530-2
pubmed: 12687444
pmcid: 3615504
Burton, A. K. & Tillotson, K. M. Prediction of the clinical course of low-back trouble using multivariable models. Spine (Phila Pa 1976) 16, 7–14. https://doi.org/10.1097/00007632-199101000-00002 (1991).
doi: 10.1097/00007632-199101000-00002
Overas, C. K. et al. Association between objectively measured physical behaviour and neck- and/or low back pain: a systematic review. Eur. J. Pain https://doi.org/10.1002/ejp.1551 (2020).
doi: 10.1002/ejp.1551
pubmed: 32096285
Childs, J. D. et al. A clinical prediction rule to identify patients with low back pain most likely to benefit from spinal manipulation: a validation study. Ann. Intern. Med. 141, 920–928 (2004).
doi: 10.7326/0003-4819-141-12-200412210-00008
Lee, H. et al. How does pain lead to disability? A systematic review and meta-analysis of mediation studies in people with back and neck pain. Pain 156, 988–997. https://doi.org/10.1097/j.pain.0000000000000146 (2015).
doi: 10.1097/j.pain.0000000000000146
pubmed: 25760473
Moons, K. G. M. et al. Transparent reporting of a multivariable prediction model for individual prognosis or diagnosis (TRIPOD): explanation and elaboration. Ann. Intern. Med. 162, W1–W73. https://doi.org/10.7326/m14-0698 (2015).
doi: 10.7326/m14-0698
pubmed: 25560730
van der Ploeg, T., Austin, P. C. & Steyerberg, E. W. Modern modelling techniques are data hungry: a simulation study for predicting dichotomous endpoints. BMC Med. Res. Methodol. 14, 137–137. https://doi.org/10.1186/1471-2288-14-137 (2014).
doi: 10.1186/1471-2288-14-137
pubmed: 25532820
pmcid: 4289553
Wibault, J. et al. Structured postoperative physiotherapy in patients with cervical radiculopathy: 6-month outcomes of a randomized clinical trial. J. Neurosurg. Spine 28, 1–9. https://doi.org/10.3171/2017.5.spine16736 (2018).
doi: 10.3171/2017.5.spine16736
pubmed: 29087809
Wibault, J. et al. Neck-related physical function, self-efficacy, and coping strategies in patients with cervical radiculopathy: a randomized clinical trial of postoperative physiotherapy. J. Manip. Physiol. Ther. 40, 330–339. https://doi.org/10.1016/j.jmpt.2017.02.012 (2017).
doi: 10.1016/j.jmpt.2017.02.012
Peolsson, A. et al. Outcome of physiotherapy after surgery for cervical disc disease: a prospective randomised multi-centre trial. BMC Musculoskelet. Disord. 15, 34. https://doi.org/10.1186/1471-2474-15-34 (2014).
doi: 10.1186/1471-2474-15-34
pubmed: 24502414
pmcid: 3923245
Brooks, R. EuroQol: the current state of play. Health Policy (Amsterdam, Netherlands) 37, 53–72. https://doi.org/10.1016/0168-8510(96)00822-6 (1996).
doi: 10.1016/0168-8510(96)00822-6
Kuhn, M. & Johnson, K. Applied Predictive Modeling 487–519 (Springer, New York, 2013).
doi: 10.1007/978-1-4614-6849-3
van Buuren, S. & Groothuis-Oudshoorn, K. mice: multivariate imputation by chained equations in R. J. Stat. Softw. 1(3), 2011 (2011).
Hinkle, D., Wiersma, W. & Jurs, S. Applied Statistics for the Behavioral Sciences 5th edn. (Houghton Mifflin, Boston, 2003).
Tibshirani, R. Regression shrinkage and selection via the lasso. J. Roy. Stat. Soc. Ser. B (Methodol.) 58, 267–288. https://doi.org/10.1111/j.2517-6161.1996.tb02080.x (1996).
doi: 10.1111/j.2517-6161.1996.tb02080.x
Buhlmann, P. & Hothorn, T. Boosting algorithms: regularization, prediction and model fitting. Statist. Sci. 22, 477–505. https://doi.org/10.1214/07-STS242 (2007).
doi: 10.1214/07-STS242
James, G., Witten, D., Hastie, T. & Tibshirani, R. in An Introduction to Statistical Learning: With Applications in RVol. 1 (eds G James, D Witten, T Hastie, & R Tibshirani) Ch. 5, 175–201 (Springer, 2013).