The use and reporting of multiple imputation in medical research – a review

Mackinnon, Andrew

doi:10.1111/j.1365-2796.2010.02274.x

Cited by 175 publications

(153 citation statements)

References 20 publications

Supporting

Mentioning

144

Contrasting

Unclassified

Order By: Relevance

“…HRs are adjusted for all explanatory variables included in the multivariable Cox regression analysis. We assume missing values-BMI (21%, 197 of 927 patients) and hemoglobin (6%, 59 of 927 patients)-to be random and used multiple imputation to replace missing values 40 times based on the remaining explanatory variables [22].…”

Section: Discussionmentioning

confidence: 99%

2015 Marshall Urist Young Investigator Award: Prognostication in Patients With Long Bone Metastases: Does a Boosting Algorithm Improve Survival Estimates?

Janssen

Heijden

Dijke

et al. 2015

Clin Orthop Relat Res

View full text Add to dashboard Cite

Background Survival estimation guides surgical decisionmaking in metastatic bone disease. Traditionally, classic scoring systems, such as the Bauer score, provide survival estimates based on a summary score of prognostic factors. Identification of new factors might improve the accuracy of these models. Additionally, the use of different algorithms-nomograms or boosting algorithms-could further improve accuracy of prognostication relative to classic scoring systems. A nomogram is an extension of a classic scoring system and generates a more-individualized survival probability based on a patient's set of characteristics using a figure. Boosting is a method that automatically trains to classify outcomes by applying classifiers (variables) in a sequential way and subsequently combines them. A boosting algorithm provides survival probabilities based on every possible combination of variables. Questions/purposes We wished to (1) assess factors independently associated with decreased survival in patients with metastatic long bone fractures and (2) compare the accuracy of a classic scoring system, nomogram, and boosting algorithms in predicting 30-, 90-, and 365-day survival. Methods We included all 927 patients in our retrospective study who underwent surgery for a metastatic long bone fracture at two institutions between January 1999 and December 2013. We included only the first procedure if patients underwent multiple surgical procedures or had more than one fracture. Median followup was 8 months (interquartile range, 3-25 months); 369 of 412 (90%) patients who where alive at 1 year were still in followup. Multivariable Cox regression analysis was used to identify One author (SJ) certifies that he has received, an amount less than USD 10,000, from the Anna Foundation (Oegstgeest, The Netherlands), an amount less than USD 10,000 from the De Drie Lichten Foundation (Hilversum, The Netherlands), an amount less than USD 10,000 from the KWF Kankerbestrijding (Amsterdam, The Netherlands), and an amount less than USD 10,000, from the Michael van Vloten Foundation (Rotterdam, The Netherlands). One author (JS) certifies that he or a member of his family has or may receive payments or benefits, an amount less than USD 10,000, from Stryker (Kalamazoo, MI, USA), and an amount less than USD 10,000, from Biom'up (Saint-Priest, Lyon, France). All ICMJE Conflict of Interest Forms for authors and Clinical Orthopaedics and Related Research 1 editors and board members are on file with the publication and can be viewed on request. Each author certifies that his or her institution approved the human protocol for this investigation, that all investigations were conducted in conformity with ethical principles of research. This work was performed at Massachusetts General Hospital, Boston, MA, USA. clinical and laboratory factors independently associated with decreased survival. We created a classic scoring system, nomogram, and boosting algorithms based on identified variables. Accuracy of the algorithms was assessed using area unde...

show abstract

Section: Discussionmentioning

confidence: 99%

2015 Marshall Urist Young Investigator Award: Prognostication in Patients With Long Bone Metastases: Does a Boosting Algorithm Improve Survival Estimates?

Janssen

Heijden

Dijke

et al. 2015

Clin Orthop Relat Res

View full text Add to dashboard Cite

show abstract

“…Missing data rarely occur completely at random and, therefore, imputation using multiple regression techniques was applied as recommended by many methodologists [9,10].…”

Section: Discussionmentioning

confidence: 99%

Diagnosing pneumonia in patients with acute cough: clinical judgment compared to chest radiography

et al. 2013

View full text Add to dashboard Cite

Pneumonia is often diagnosed and treated empirically. We set out to determine the diagnostic accuracy of clinical judgment based on signs and symptoms to detect radiographic pneumonia in patients presenting with acute cough in primary care.In 2810 European patients with acute cough, general practitioners (GPs) recorded whether they considered pneumonia to be present (“yes” or “no”) immediately after history and physical examination. Chest radiography was performed within 1 week by local radiologists blind to other patient characteristics.140 patients had radiographic pneumonia (5%), of whom 41 (29%) had been diagnosed as such. 31 (1%) patients had a clinical diagnosis that was not confirmed by radiography (n=2670). In clinically suspected pneumonia, 57% of subjects were subsequently diagnosed with radiographic pneumonia. Negative predictive value (NPV), sensitivity and specificity of GPs’ clinical judgment were 96%, 29% and 99%, respectively. Compared to patients with a clinical diagnosis of pneumonia, less severe symptoms were found in radiographic pneumonia cases not suspected clinically (p<0.05).The predictive values of GPs’ clinical judgment, particularly the high NPVs, are helpful in routine care. Nonetheless, the majority of diagnoses of radiographic pneumonias was not suspected on clinical grounds. There is a need to further support the detection of clinically relevant pneumonia in primary care.

show abstract

“…19 Missing data were estimated by the multiple imputation technique. 20 Two-tailed values of p<0.05 were considered significant. Data were tabulated in MS Excel 2003® and analyzed using SPSS 17.0 ® software.…”

Section: Methodsmentioning

confidence: 99%