National Databases for Neurosurgical Outcomes Research: Options, Strengths, and Limitations

Karhade, Aditya V.; Larsen, Alexandra Giantini; Coté, David J.; Dubois, Heloise; Smith, Timothy R.

doi:10.1093/neuros/nyx408

Cited by 77 publications

(34 citation statements)

References 99 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We refer the reader to the excellent recent review by Karhade et al [39] of the various databases used in pediatric neurosurgical clinical research. Studies seeking to investigate relatively rare pediatric diseases or conditions would be relevant applications of big data as these inquiries would be supported by much larger sample sizes than are obtainable by individual centers.…”

Section: Discussionmentioning

confidence: 99%

Big Data Research in Pediatric Neurosurgery: Content, Statistical Output, and Bibliometric Analysis

et al. 2019

View full text Add to dashboard Cite

Background/Aims: We sought to describe pediatric “big data” publications since 2000, their statistical output, and clinical implications. Methods: We searched 4 major North American neurosurgical journals for articles utilizing non-neurosurgery-specific databases for clinical pediatric neurosurgery research. Articles were analyzed for descriptive and statistical information. We analyzed effect sizes (ESs), confidence intervals (CIs), and p values for clinical relevance. A bibliometric analysis was performed using several key citation metrics. Results: We identified 74 articles, which constituted 1.7% of all pediatric articles (n = 4,436) published, with an exponential increase after 2013 (53/74, 72%). The Healthcare Cost and Utilization Project (HCUP) databases were most frequently utilized (n = 33); hydrocephalus (n = 19) was the most common study topic. The statistical output (n = 49 studies with 464 ESs, 456 CIs, and 389 p values) demonstrated that the majority of the ESs (253/464, 55%) were categorized as small; half or more of the CI spread (CIS) values and p values were high (274/456, 60%) and very strong (195/389, 50%), respectively. Associations with a combination of medium-to-large ESs (i.e., magnitude of difference), medium-to-high CISs (i.e., precision), and strong-to-very strong p values comprised only 20% (75/381) of the reported ESs. The total number of citations for the 74 articles was 1,115 (range per article, 0–129), with the median number of citations per article being 8.5. Four studies had > 50 citations, and 2 of them had > 100 citations. The calculated h-index was 16, h-core citations were 718, the e-index was 21.5, and the Google i10-index was 34. Conclusions: There has been a dramatic increase in the use of “big data” in the pediatric neurosurgical literature. Reported associations that may, as a group, be of greatest interest to practitioners represented only 20% of the total output from these publications. Citations were weighted towards a few highly cited publications.

show abstract

Section: Discussionmentioning

confidence: 99%

Big Data Research in Pediatric Neurosurgery: Content, Statistical Output, and Bibliometric Analysis

et al. 2019

View full text Add to dashboard Cite

show abstract

“…Registry databases may also underreport cases for which there are no clinical symptoms and therefore present with selection bias. 13 In the absence of a similarly large-scale database that includes these other factors, this analysis represents the largest available baseline from which to work, with acknowledgment of its limitations, and makes valuable suggestions about the treatment choices that physicians make. Moreover, given the limitations of using this type of registry data, we strongly suggest that further work be done in a prospective manner (while documenting a wide variety of clinically relevant cofactors) to assess the effectiveness of adjuvant radiotherapy use (while noting the specific type and dosing regimen) after resection of an atypical meningioma.…”

Section: Discussionmentioning

confidence: 99%

Determining the role of adjuvant radiotherapy in the management of meningioma: a Surveillance, Epidemiology, and End Results analysis

Reddy¹,

Ryoo²,

Denyer³

et al. 2019

Neurosurgical Focus

View full text Add to dashboard Cite

OBJECTIVEThe aim of this study was to illustrate the demographic characteristics of meningioma patients and observe the effect of adjuvant radiation therapy on survival by using the Surveillance, Epidemiology, and End Results (SEER) database. More specifically, the authors aimed to answer the question of whether adjuvant radiotherapy following resection of atypical meningioma confers a cause-specific survival benefit. Additionally, they attempted to add to previous characterizations of the epidemiology of primary meningiomas and assess the effectiveness of the standard of care for benign and anaplastic meningiomas. They also sought to characterize the efficacy of various treatment options in atypical and anaplastic meningiomas separately since nearly all other analyses have grouped these two together despite varying treatment regimens for these behavior categories.METHODSSEER data from 1973 to 2015 were queried using appropriate ICD-O-3 codes for benign, atypical, and anaplastic meningiomas. Patient demographics, tumor characteristics, and treatment choices were analyzed. The effects of treatment were examined using a multivariate Cox proportional hazards model and Kaplan-Meier survival analysis.RESULTSA total of 57,998 patients were included in the analysis of demographic, meningioma, and treatment characteristics. Among this population, cases of unspecified WHO tumor grade were excluded in the multivariate analysis, leaving a total of 12,931 patients to examine outcomes among treatment paradigms. In benign meningiomas, gross-total resection (HR 0.289, p = 0.013) imparted a significant cause-specific survival benefit over no treatment. In anaplastic meningioma cases, adjuvant radiotherapy imparted a significant survival benefit following both subtotal (HR 0.089, p = 0.018) and gross-total (HR 0.162, p = 0.002) resection as compared to gross-total resection alone. In atypical tumors, gross-total resection plus radiotherapy did not significantly change the hazard risk (HR 1.353, p = 0.628) compared to gross-total resection alone. Similarly, it was found that adjuvant radiation did not significantly benefit survival after a subtotal resection (HR 1.440, p = 0.644).CONCLUSIONSThe results of this study demonstrate that the role of adjuvant radiotherapy, especially after the resection of atypical meningioma, remains somewhat unclear. Thus, given these results, prospective randomized clinical studies are warranted to provide clear information on the effects of adjuvant radiation in meningioma treatment.

show abstract

“…Databases, preferably big data analytics like multi-institutional clinical registries 36,37,38 have been used for developing the predictive algorithms. The de-identi ed data from large national database usually get exemption from review by institutional review board.…”

Section: Study Design and Data Sourcesmentioning

confidence: 99%

Machine Learning Algorithms Using Logistic Regression for Predicting Neurosurgical Outcomes

Fatima¹

2020

Preprint

View full text Add to dashboard Cite

Background: Preoperative prognostication of clinical and surgical outcome in patients with neurosurgical diseases can improve the risk stratification, thus can guide in implementing targeted treatment to minimize these events. Therefore, the author aims to highlight the development and validation of predictive models determining neurosurgical outcomes through machine learning algorithms using logistic regression.Methods: Logistic regression (enter, backward and forward) and least absolute shrinkage and selection operator (LASSO) method for selection of variables from selected database can eventually lead to multiple candidate models. The final model with a set of predictive variables must be selected based upon the clinical knowledge and numerical results.Results: The predictive model which performed best on the discrimination, calibration, Brier score and decision curve analysis must be selected to develop machine learning algorithms. Logistic regression should be compared with the LASSO model. Usually for the big databases, the predictive model selected through logistic regression gives higher Area Under the Curve (AUC) than those with LASSO model. The predictive probability derived from the best model could be uploaded to an open access web application which is easily deployed by the patients and surgeons to make a risk assessment world-wide.Conclusions: Machine learning algorithms provide promising results for the prediction of outcomes following cranial and spinal surgery. These algorithms can provide useful factors for patient-counselling, assessing peri-operative risk factors, and predicting post-operative outcomes after neurosurgery.

show abstract

National Databases for Neurosurgical Outcomes Research: Options, Strengths, and Limitations

Abstract: The improvement of existing large national databases and the establishment of new registries will be crucial to the future of neurosurgical outcomes research.

Cited by 77 publications

References 99 publications

Big Data Research in Pediatric Neurosurgery: Content, Statistical Output, and Bibliometric Analysis

Big Data Research in Pediatric Neurosurgery: Content, Statistical Output, and Bibliometric Analysis

Determining the role of adjuvant radiotherapy in the management of meningioma: a Surveillance, Epidemiology, and End Results analysis

Machine Learning Algorithms Using Logistic Regression for Predicting Neurosurgical Outcomes

Contact Info

Product

Resources

About