Karel G M Moons

University Medical Center Utrecht, Utrecht, Utrecht, Netherlands

Are you Karel G M Moons?

Claim your profile

Publications (379)2045.08 Total impact

  • [Show abstract] [Hide abstract]
    ABSTRACT: Hybrid single photon emission computed tomography (SPECT)/coronary computed tomography angiography (CCTA) has only been evaluated for its diagnostic accuracy as a single test in patients suspected of significant coronary artery disease (CAD). Added value of hybrid SPECT/CCTA beyond usual clinical work-up, or use of each of these tests separately, remains unclear. We evaluated the added value of hybrid myocardial perfusion SPECT (SPECT) and CCTA, beyond pre-test likelihood and exercise stress ECG (X-ECG), in the diagnosis of CAD.
    European Heart Journal – Cardiovascular Imaging 07/2014; · 2.39 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: To examine how authors explore and report on sources of heterogeneity in systematic reviews of diagnostic accuracy studies.
    Journal of clinical epidemiology. 07/2014;
  • Source
    Ray Moynihan, David Henry, Karel G M Moons
    [Show abstract] [Hide abstract]
    ABSTRACT: Ray Moynihan and colleagues outline suggestions for improving the way that medical evidence is produced, analysed, and interpreted to avoid problems of overdiagnosis and overtreatment. Please see later in the article for the Editors' Summary.
    PLoS Medicine 07/2014; 11(7):e1001655. · 15.25 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: George Peat and colleagues review and discuss current approaches to transparency and published debates and concerns about efforts to standardize prognosis research practice, and make five recommendations. Please see later in the article for the Editors' Summary.
    PLoS Medicine 07/2014; 11(7):e1001671. · 15.25 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: It is uncertain whether screening of older persons for chronic obstructive pulmonary disease (COPD) is worthwhile because the effects on patient management and prognosis are unknown. We aimed to assess the short-term consequences of detecting COPD in frail elderly subjects with dyspnoea, considering pulmonary drug use, hospitalisations and all-cause mortality. Community-dwelling frail elderly subjects, aged 65 years and older, with dyspnoea, participating in a screening study on COPD and heart failure were included. Final diagnoses were assigned by an expert panel based on all data from the screening strategy, including spirometry. Follow-up data were collected from the general practitioners. Of the 386 patients, 84 (21.8%) were received a new diagnosis of COPD. Overall, changes in pulmonary drug prescription during 6 months of follow-up were infrequent (n = 53, 13.7%; among new cases of COPD, 15 (17.9%) out of 84). Of all participants, 25.9% were hospitalised in the first year of follow-up, with the highest rate in patients with newly detected COPD (32.1%). Many new cases of COPD could be detected by screening frail elderly subjects with dyspnoea, but the impact on patient management seems limited. Our study underlines the importance of obtaining follow-up data to assess the true impact of a (screen-detected) diagnosis of COPD on patient management and outcome.
    The European respiratory journal. 06/2014;
  • Source
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: An important question in the evaluation of an additional risk prediction marker is how to interpret a small increase in the area under the receiver operating characteristic curve (AUC). Many researchers believe that a change in AUC is a poor metric because it increases only slightly with the addition of a marker with a large odds ratio. Because it is not possible on purely statistical grounds to choose between the odds ratio and AUC, we invoke decision analysis, which incorporates costs and benefits. For example, a timely estimate of the risk of later non-elective operative delivery can help a woman in labor decide if she wants an early elective cesarean section to avoid greater complications from possible later non-elective operative delivery. A basic risk prediction model for later non-elective operative delivery involves only antepartum markers. Because adding intrapartum markers to this risk prediction model increases AUC by 0.02, we questioned whether this small improvement is worthwhile. A key decision-analytic quantity is the risk threshold, here the risk of later non-elective operative delivery at which a patient would be indifferent between an early elective cesarean section and usual care. For a range of risk thresholds, we found that an increase in the net benefit of risk prediction requires collecting intrapartum marker data on 68 to 124 women for every correct prediction of later non-elective operative delivery. Because data collection is non-invasive, this test tradeoff of 68 to 124 is clinically acceptable, indicating the value of adding intrapartum markers to the risk prediction model. Copyright © 2014 John Wiley & Sons, Ltd.
    Statistics in Medicine 05/2014; · 2.04 Impact Factor
  • Canadian Medical Association Journal 04/2014; · 6.47 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: The value of measuring levels of glycated hemoglobin (HbA1c) for the prediction of first cardiovascular events is uncertain. To determine whether adding information on HbA1c values to conventional cardiovascular risk factors is associated with improvement in prediction of cardiovascular disease (CVD) risk. Analysis of individual-participant data available from 73 prospective studies involving 294,998 participants without a known history of diabetes mellitus or CVD at the baseline assessment. Measures of risk discrimination for CVD outcomes (eg, C-index) and reclassification (eg, net reclassification improvement) of participants across predicted 10-year risk categories of low (<5%), intermediate (5% to <7.5%), and high (≥7.5%) risk. During a median follow-up of 9.9 (interquartile range, 7.6-13.2) years, 20,840 incident fatal and nonfatal CVD outcomes (13,237 coronary heart disease and 7603 stroke outcomes) were recorded. In analyses adjusted for several conventional cardiovascular risk factors, there was an approximately J-shaped association between HbA1c values and CVD risk. The association between HbA1c values and CVD risk changed only slightly after adjustment for total cholesterol and triglyceride concentrations or estimated glomerular filtration rate, but this association attenuated somewhat after adjustment for concentrations of high-density lipoprotein cholesterol and C-reactive protein. The C-index for a CVD risk prediction model containing conventional cardiovascular risk factors alone was 0.7434 (95% CI, 0.7350 to 0.7517). The addition of information on HbA1c was associated with a C-index change of 0.0018 (0.0003 to 0.0033) and a net reclassification improvement of 0.42 (-0.63 to 1.48) for the categories of predicted 10-year CVD risk. The improvement provided by HbA1c assessment in prediction of CVD risk was equal to or better than estimated improvements for measurement of fasting, random, or postload plasma glucose levels. In a study of individuals without known CVD or diabetes, additional assessment of HbA1c values in the context of CVD risk assessment provided little incremental benefit for prediction of CVD risk.
    JAMA The Journal of the American Medical Association 03/2014; 311(12):1225-33. · 29.98 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Before considering whether to use a multivariable (diagnostic or prognostic) prediction model, it is essential that its performance be evaluated in data that were not used to develop the model (referred to as external validation). We critically appraised the methodological conduct and reporting of external validation studies of multivariable prediction models. We conducted a systematic review of articles describing some form of external validation of one or more multivariable prediction models indexed in PubMed core clinical journals published in 2010. Study data were extracted in duplicate on design, sample size, handling of missing data, reference to the original study developing the prediction models and predictive performance measures. 11,826 articles were identified and 78 were included for full review, which described the evaluation of 120 prediction models. in participant data that were not used to develop the model. Thirty-three articles described both the development of a prediction model and an evaluation of its performance on a separate dataset, and 45 articles described only the evaluation of an existing published prediction model on another dataset. Fifty-seven percent of the prediction models were presented and evaluated as simplified scoring systems. Sixteen percent of articles failed to report the number of outcome events in the validation datasets. Fifty-four percent of studies made no explicit mention of missing data. Sixty-seven percent did not report evaluating model calibration whilst most studies evaluated model discrimination. It was often unclear whether the reported performance measures were for the full regression model or for the simplified models. The vast majority of studies describing some form of external validation of a multivariable prediction model were poorly reported with key details frequently not presented. The validation studies were characterised by poor design, inappropriate handling and acknowledgement of missing data and one of the most key performance measures of prediction models i.e. calibration often omitted from the publication. It may therefore not be surprising that an overwhelming majority of developed prediction models are not used in practice, when there is a dearth of well-conducted and clearly reported (external validation) studies describing their performance on independent participant data.
    BMC Medical Research Methodology 03/2014; 14(1):40. · 2.21 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Background: chronic dyspnoea is common in older people and is often of cardiac or pulmonary aetiology. Information on the exact prevalence and distribution of underlying causes is scarce. Our aim was to review the literature on prevalence and underlying causes of dyspnoea in the older population.Methods: two MEDLINE searches were conducted: the first on studies on the prevalence of dyspnoea in older persons aged ≥65 years using the Medical Research Council (MRC) dyspnoea scale and the second on the underlying causes of dyspnoea in this population. Quality assessment was performed for all included studies. Random effects models based on the logit transformed prevalences were used to calculate pooled prevalence with 95% confidence intervals (95% CI).Results: a total of 21 articles from 20 different populations reported the prevalence in the general older population with a median sample size of 600 (Interquartile range 262-1289). The pooled prevalence was 36% (95% CI: 27-47%) for an MRC of ≥2, 16% (95% CI: 12-21%) for an MRC of ≥3 and 4% (95% CI: 2-9%) for an MRC of ≥4. Prevalence rates were higher in women than in men.Only one article investigated the underlying causes of dyspnoea in older persons; in 70% of these patients, the dyspnoea was considered to be of cardiac or pulmonary origin.Conclusion: dyspnoea is very common in older people, but estimates vary considerably between studies. Only one study describes the underlying causes.
    Age and Ageing 01/2014; · 3.82 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Published clinical prediction models are often ignored during the development of novel prediction models despite similarities in populations and intended usage. The plethora of prediction models that arise from this practice may still perform poorly when applied in other populations. Incorporating prior evidence might improve the accuracy of prediction models and make them potentially better generalizable. Unfortunately, aggregation of prediction models is not straightforward, and methods to combine differently specified models are currently lacking. We propose two approaches for aggregating previously published prediction models when a validation dataset is available: model averaging and stacked regressions. These approaches yield user-friendly stand-alone models that are adjusted for the new validation data. Both approaches rely on weighting to account for model performance and between-study heterogeneity but adopt a different rationale (averaging versus combination) to combine the models. We illustrate their implementation in a clinical example and compare them with established methods for prediction modeling in a series of simulation studies. Results from the clinical datasets and simulation studies demonstrate that aggregation yields prediction models with better discrimination and calibration in a vast majority of scenarios, and results in equivalent performance (compared to developing a novel model from scratch) when validation datasets are relatively large. In conclusion, model aggregation is a promising strategy when several prediction models are available from the literature and a validation dataset is at hand. The aggregation methods do not require existing models to have similar predictors and can be applied when relatively few data are at hand. Copyright © 2014 John Wiley & Sons, Ltd.
    Statistics in Medicine 01/2014; · 2.04 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: To assess the accuracy of the Wells rule for excluding deep vein thrombosis and whether this accuracy applies to different subgroups of patients. Meta-analysis of individual patient data. Authors of 13 studies (n=10 002) provided their datasets, and these individual patient data were merged into one dataset. Studies were eligible if they enrolled consecutive outpatients with suspected deep vein thrombosis, scored all variables of the Wells rule, and performed an appropriate reference standard. Multilevel logistic regression models, including an interaction term for each subgroup, were used to estimate differences in predicted probabilities of deep vein thrombosis by the Wells rule. In addition, D-dimer testing was added to assess differences in the ability to exclude deep vein thrombosis using an unlikely score on the Wells rule combined with a negative D-dimer test result. Overall, increasing scores on the Wells rule were associated with an increasing probability of having deep vein thrombosis. Estimated probabilities were almost twofold higher in patients with cancer, in patients with suspected recurrent events, and (to a lesser extent) in males. An unlikely score on the Wells rule (≤1) combined with a negative D-dimer test result was associated with an extremely low probability of deep vein thrombosis (1.2%, 95% confidence interval 0.7% to 1.8%). This combination occurred in 29% (95% confidence interval 20% to 40%) of patients. These findings were consistent in subgroups defined by type of D-dimer assay (quantitative or qualitative), sex, and care setting (primary or hospital care). For patients with cancer, the combination of an unlikely score on the Wells rule and a negative D-dimer test result occurred in only 9% of patients and was associated with a 2.2% probability of deep vein thrombosis being present. In patients with suspected recurrent events, only the modified Wells rule (adding one point for the previous event) is safe. Combined with a negative D-dimer test result (both quantitative and qualitative), deep vein thrombosis can be excluded in patients with an unlikely score on the Wells rule. This finding is true for both sexes, as well as for patients presenting in primary and hospital care. In patients with cancer, the combination is neither safe nor efficient. For patients with suspected recurrent disease, one extra point should be added to the rule to enable a safe exclusion.
    BMJ (online) 01/2014; 348:g1340. · 17.22 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: In Asia, up to 25% of breast cancer patients present with distant metastases at diagnosis. Given the heterogeneous survival probabilities of de novo metastatic breast cancer, individual outcome prediction is challenging. The aim of the study is to identify existing prognostic models for patients with de novo metastatic breast cancer and validate them in Asia. We performed a systematic review to identify prediction models for metastatic breast cancer. Models were validated in 642 women with de novo metastatic breast cancer registered between 2000 and 2010 in the Singapore Malaysia Hospital Based Breast Cancer Registry. Survival curves for low, intermediate and high-risk groups according to each prognostic score were compared by log-rank test and discrimination of the models was assessed by concordance statistic (C-statistic). We identified 16 prediction models, seven of which were for patients with brain metastases only. Performance status, estrogen receptor status, metastatic site(s) and disease-free interval were the most common predictors. We were able to validate nine prediction models. The capacity of the models to discriminate between poor and good survivors varied from poor to fair with C-statistics ranging from 0.50 (95% CI, 0.48-0.53) to 0.63 (95% CI, 0.60-0.66). The discriminatory performance of existing prediction models for de novo metastatic breast cancer in Asia is modest. Development of an Asian-specific prediction model is needed to improve prognostication and guide decision making.
    PLoS ONE 01/2014; 9(4):e93755. · 3.73 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Objective To review the prognostic value of cardiac magnetic resonance (CMR) imaging findings for future cardiovascular events in patients with a recent myocardial infarction (MI) and patients with suspected or known coronary artery disease (CAD). Background Although the diagnostic value of CMR findings is established, the independent prognostic association with future cardiovascular events remains largely unclear. Methods Studies published until February 2013 identified by systematic MEDLINE and EMBASE searches were reviewed for associations between CMR findings (left ventricular ejection fraction [LVEF], (inducible) wall motion abnormalities [WMA], abnormal myocardial perfusion, microvascular obstruction [MVO], late gadolinium enhancement, edema, intramyocardial haemorrhage), and hard events (all-cause mortality, cardiac death, cardiac transplantation, and myocardial infarction) or major adverse cardiovascular events (MACE: hard events and other cardiovascular events defined by the authors of the evaluated articles) were included. Results Fifty-six studies (25,497 patients) were evaluated. For patients with recent MI, too few patients were evaluated to establish associations between CMR findings and hard events. LVEF (range of adjusted Hazard Ratios (adjHRs): 1.03-1.05 per % decrease) was independently associated with MACE. In patients with suspected or known CAD, WMA (adjHR: 1.87-2.99), inducible perfusion defects (adj HR: 3.02-7.77), LVEF (adjHR: 0.72-0.82 per 10% increase), and infarction (adjHR: 2.82-9.43) were independently associated with hard events, and presence of inducible perfusion defects was associated with MACE (adjHRs 1.76-3.21). Conclusion Independent predictors of future cardiovascular events were LVEF for patients with a recent MI and WMA, inducible perfusion defects, LVEF and presence of infarction for patients with suspected or known CAD.
    Journal of the American College of Cardiology 01/2014; · 14.09 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: This study aimed to gather insights in physicians' considerations for decisions to either refer for- or to withhold additional diagnostic investigations in nursing home patients with a suspicion of venous thromboembolism. Our study was nested in an observational study on diagnostic strategies for suspected venous thromboembolism in nursing home patients. Patient characteristics, bleeding-complications and mortality were related to the decision to withhold investigations. For a better understanding of the physicians' decisions, 21 individual face-to-face in-depth interviews were performed and analysed using the grounded theory approach. Referal for additional diagnostic investigations was forgone in 126/322 (39.1%) patients with an indication for diagnostic work-up. 'Blind' anticoagulant treatment was initiated in 95 (75.4%) of these patients. The 3month mortality rates were higher for patients in whom investigations were withheld than in the referred patients, irrespective of anticoagulant treatment (odds ratio 2.45; 95% confidence interval 1.40 to 4.29) but when adjusted for the probability of being referred (i.e. the propensity score), there was no relation of non-diagnosis decisions to mortality (odds ratio 1.75; 0.98 to 3.11). In their decisions to forgo diagnostic investigations, physicians incorporated the estimated relative impact of the potential disease; the potential net-benefits of diagnostic investigations and whether performing investigations agreed with established management goals in advance care planning. Referral for additional diagnostic investigations is withheld in almost 40% of Dutch nursing home patients with suspected venous thromboembolism and an indication for diagnostic work-up. We propose that, given the complexity of these decisions and the uncertainty regarding their indirect effects on patient outcome, more attention should be focused on the decision to either use or withhold additional diagnostic tests.
    PLoS ONE 01/2014; 9(3):e90395. · 3.73 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Accurate visualization of the distal ascending aorta (DAA) can guide the surgical management and hence prevent dislodgment of atherogenic emboli during cardiac surgery. Conventional transoesophageal echocardiography (TEE) has a poor sensitivity; modified TEE was previously shown to accurately visualize atherosclerosis of the DAA. We studied the added value of modified TEE beyond the patient history and TEE screening. Included were 421 patients from a previous diagnostic study, which compared the diagnosis of severe atherosclerosis with modified TEE and epiaortic ultrasound (EUS; reference test). We fitted three models, which predicted presence of atherosclerosis Grade ≥3 of the DAA. Model 1 included preoperative patient characteristics; in Model 2 conventional TEE was added; Model 3 additionally included modified TEE results. For each model, the area under the receiver-operating curve (AUC), the 'net reclassification improvement' (NRI) and the 'integrated discrimination improvement' (IDI) were determined. Missing data were imputed.The AUCs of Models 1, 2, and 3 were 0.73 (95% CI: 0.68-0.78), 0.80 (95% CI: 0.76-0.85), and 0.93 (95% CI: 0.90-0.96), respectively. Comparing Model 3 with Model 2, the AUC was significantly higher (P < 0.001), the NRI was 0.60 (95% CI: 0.54-0.66; P < 0.001), and the IDI was 0.30 (95% CI: 0.28-0.32; P < 0.001), indicating that visualization of the DAA with modified TEE significantly improved reclassification. Visualization of atherosclerosis of the DAA with modified TEE provided information beyond patient history and conventional TEE screening, which resulted in an improved diagnosis of atherosclerosis.
    European heart journal cardiovascular Imaging. 12/2013;
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Bronchopulmonary dysplasia (BPD) is a common complication of preterm birth. Very different models using clinical parameters at an early postnatal age to predict BPD have been developed with little extensive quantitative validation. The objective of this study is to review and validate clinical prediction models for BPD. We searched the main electronic databases and abstracts from annual meetings. The STROBE instrument was used to assess the methodological quality. External validation of the retrieved models was performed using an individual patient dataset of 3229 patients at risk for BPD. Receiver operating characteristic curves were used to assess discrimination for each model by calculating the area under the curve (AUC). Calibration was assessed for the best discriminating models by visually comparing predicted and observed BPD probabilities. We identified 26 clinical prediction models for BPD. Although the STROBE instrument judged the quality from moderate to excellent, only four models utilised external validation and none presented calibration of the predictive value. For 19 prediction models with variables matched to our dataset, the AUCs ranged from 0.50 to 0.76 for the outcome BPD. Only two of the five best discriminating models showed good calibration. External validation demonstrates that, except for two promising models, most existing clinical prediction models are poor to moderate predictors for BPD. To improve the predictive accuracy and identify preterm infants for future intervention studies aiming to reduce the risk of BPD, additional variables are required. Subsequently, that model should be externally validated using a proper impact analysis before its clinical implementation.
    BMC Pediatrics 12/2013; 13(1):207. · 1.98 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: To compare the accuracy of data from hospital administration databases and a national clinical cardiac surgery database and to compare the performance of the Dutch hospital standardised mortality ratio (HSMR) method and the logistic European System for Cardiac Operative Risk Evaluation, for the purpose of benchmarking of mortality across hospitals. Information on all patients undergoing cardiac surgery between 1 January 2007 and 31 December 2010 in 10 centres was extracted from The Netherlands Association for Cardio-Thoracic Surgery database and the Hospital Discharge Registry. The number of cardiac surgery interventions was compared between both databases. The European System for Cardiac Operative Risk Evaluation and hospital standardised mortality ratio models were updated in the study population and compared using the C-statistic, calibration plots and the Brier-score. The number of cardiac surgery interventions performed could not be assessed using the administrative database as the intervention code was incorrect in 1.4-26.3%, depending on the type of intervention. In 7.3% no intervention code was registered. The updated administrative model was inferior to the updated clinical model with respect to discrimination (c-statistic of 0.77 vs 0.85, p<0.001) and calibration (Brier Score of 2.8% vs 2.6%, p<0.001, maximum score 3.0%). Two average performing hospitals according to the clinical model became outliers when benchmarking was performed using the administrative model. In cardiac surgery, administrative data are less suitable than clinical data for the purpose of benchmarking. The use of either administrative or clinical risk-adjustment models can affect the outlier status of hospitals. Risk-adjustment models including procedure-specific clinical risk factors are recommended.
    Heart (British Cardiac Society) 12/2013; · 5.01 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Latent class models (LCMs) combine the results of multiple diagnostic tests through a statistical model to obtain estimates of disease prevalence and diagnostic test accuracy in situations where there is no single, accurate reference standard. We performed a systematic review of the methodology and reporting of LCMs in diagnostic accuracy studies. This review shows that the use of LCMs in such studies increased sharply in the past decade, notably in the domain of infectious diseases (overall contribution: 59%). The 64 reviewed studies used a range of differently specified parametric latent variable models, applying Bayesian and frequentist methods. The critical assumption underlying the majority of LCM applications (61%) is that the test observations must be independent within 2 classes. Because violations of this assumption can lead to biased estimates of accuracy and prevalence, performing and reporting checks of whether assumptions are met is essential. Unfortunately, our review shows that 28% of the included studies failed to report any information that enables verification of model assumptions or performance. Because of the lack of information on model fit and adequate evidence "external" to the LCMs, it is often difficult for readers to judge the validity of LCM-based inferences and conclusions reached.
    American journal of epidemiology 11/2013; · 5.59 Impact Factor

Publication Stats

8k Citations
2,045.08 Total Impact Points


  • 2000–2014
    • University Medical Center Utrecht
      • • Julius Center for Health Sciences and Primary Care
      • • Department of Anesthesiology
      Utrecht, Utrecht, Netherlands
  • 2013
    • Gelderse Vallei Hospital
      Ede, Gelderland, Netherlands
    • University of Birmingham
      • School of Health and Population Sciences
      Birmingham, ENG, United Kingdom
    • University of Exeter
      Exeter, England, United Kingdom
  • 2001–2013
    • Erasmus MC
      • Research Group for Public Health
      Rotterdam, South Holland, Netherlands
  • 2012
    • University of Amsterdam
      • Faculty of Medicine AMC
      Amsterdam, North Holland, Netherlands
  • 2009–2012
    • University of Oxford
      • Centre for Statistics in Medicine
      Oxford, ENG, United Kingdom
    • Medical Research Council (UK)
      • MRC Clinical Trials Unit
      London, ENG, United Kingdom
  • 2011
    • Sanquin Blood Supply Foundation
      Amsterdamo, North Holland, Netherlands
  • 2010
    • Certara
      San Luis, Missouri, United States
  • 1998–2010
    • Utrecht University
      Utrecht, Utrecht, Netherlands
  • 2006–2009
    • Kitasato University
      • Graduate School of Medical Sciences
      Edo, Tōkyō, Japan
  • 2008
    • Amphia Ziekenhui
      Breda, North Brabant, Netherlands
    • Wageningen University
      • Division of Human Nutrition
      Wageningen, Provincie Gelderland, Netherlands
  • 2007
    • Isala Klinieken
      Zwolle, Overijssel, Netherlands
  • 2000–2001
    • Het Oogziekenhuis Rotterdam
      Rotterdam, South Holland, Netherlands
  • 1997–1998
    • Erasmus Universiteit Rotterdam
      Rotterdam, South Holland, Netherlands