Stef van Buuren

Nederlandse Organisatie voor Wetenschappelijk Onderzoek, 's-Gravenhage, South Holland, Netherlands

Are you Stef van Buuren?

Claim your profile

Publications (169)365.94 Total impact

  • [Show abstract] [Hide abstract]
    ABSTRACT: Monitoring overweight in risk groups is necessary. Our aim is to assess the trend in overweight and obesity in Turkish and Moroccan children in the Netherlands since 1997 and to monitor the levels of lifestyle-related behaviours in 2009. We selected cross-sectional data of Turkish and Moroccan children aged 2-18 years from two national Growth Studies performed in 1997 and 2009 in the Netherlands. Lifestyle-related behaviours were obtained in the 2009 study by questionnaire. In 2009, 31.9% of Turkish and 26.6% of Moroccan children had overweight, whereas this was, respectively, 26.7% and 19.6% in 1997. Already at 2 years, 21.1% in Turkish and 22.7% in Moroccan children had overweight in 2009. The prevalence of obesity was above 4% from 3 years onwards. High (i.e. ≥25%) prevalence rates of unhealthy lifestyle-related behaviours were found for not having breakfast (26-49%) among Turkish and Moroccan adolescent (i.e. 15-18 years) girls, consuming no fruit (29-45%) and watching TV/PC ≥2 h (35-72%) among all Turkish and Moroccan adolescents, no walking/cycling to school/day care among preschool children (2-4 years) (28-56%) and adolescents (34-94%), drinking ≥2 glasses of sweet beverages (44-74%) and being <1 h physically active (29-65%) among all children. An upward trend of overweight and obesity occurred in Turkish and Moroccan children. Already at 2 years of age, one out of five Turkish and Moroccan children had overweight, which calls for early prevention with attention to specific lifestyle-related behaviours. © The Author 2015. Published by Oxford University Press on behalf of the European Public Health Association. All rights reserved.
    The European Journal of Public Health 04/2015; DOI:10.1093/eurpub/ckv051 · 2.46 Impact Factor
  • Source
  • Source
  • Source
  • [Show abstract] [Hide abstract]
    ABSTRACT: Individual participant data meta-analyses (IPD-MA) are increasingly used for developing and validating multivariable (diagnostic or prognostic) risk prediction models. Unfortunately, some predictors or even outcomes may not have been measured in each study and are thus systematically missing in some individual studies of the IPD-MA. As a consequence, it is no longer possible to evaluate between-study heterogeneity and to estimate study-specific predictor effects, or to include all individual studies, which severely hampers the development and validation of prediction models. Here, we describe a novel approach for imputing systematically missing data and adopt a generalized linear mixed model to allow for between-study heterogeneity. This approach can be viewed as an extension of Resche-Rigon's method (Stat Med 2013), relaxing their assumptions regarding variance components and allowing imputation of linear and nonlinear predictors. We illustrate our approach using a case study with IPD-MA of 13 studies to develop and validate a diagnostic prediction model for the presence of deep venous thrombosis. We compare the results after applying four methods for dealing with systematically missing predictors in one or more individual studies: complete case analysis where studies with systematically missing predictors are removed, traditional multiple imputation ignoring heterogeneity across studies, stratified multiple imputation accounting for heterogeneity in predictor prevalence, and multilevel multiple imputation (MLMI) fully accounting for between-study heterogeneity. We conclude that MLMI may substantially improve the estimation of between-study heterogeneity parameters and allow for imputation of systematically missing predictors in IPD-MA aimed at the development and validation of prediction models. Copyright © 2015 John Wiley & Sons, Ltd. Copyright © 2015 John Wiley & Sons, Ltd.
    Statistics in Medicine 02/2015; 34(11). DOI:10.1002/sim.6451 · 2.04 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Accurate calculations of testicular volume standard deviation (SD) scores are not currently available. We constructed LMS-smoothed age-reference charts for testicular volume in healthy boys. The LMS method was used to calculate reference data, based on testicular volumes from ultrasonography and Prader orchidometer of 769 healthy Dutch boys aged six months to 19 years. We also explored the association between testicular growth and pubic hair development and data were compared to orchidometric testicular volumes from the 1997 Dutch nationwide growth study. The LMS-smoothed reference charts showed that no revision of the definition of normal onset of male puberty - from nine to 14-years-of-age - was warranted. In healthy boys, the pubic hair stage SD scores corresponded with testicular volume SD scores (r=0.394). However, testes were relatively small for pubic hair stage in Klinefelter syndrome and relatively large in immunoglobulin superfamily member 1 deficiency syndrome. The age-corrected SD scores for testicular volume will aid in the diagnosis and follow up of abnormalities in the timing and progression of male puberty and in research evaluations. The SD scores can be compared with pubic hair SD scores to identify discrepancies between cell functions that result in relative microorchidism or macroorchidism. This article is protected by copyright. All rights reserved. This article is protected by copyright. All rights reserved.
    Acta paediatrica (Oslo, Norway: 1992). Supplement 02/2015; DOI:10.1111/apa.12972
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: To derive regional weight-for-age growth references to help optimize age-based dosing of antimalarials in Africa, the Americas, South-East Asia and the Western Pacific. A weight-for-age database was constructed from pre-existing population-based anthropometric data obtained from household surveys and research groups. It contained data collected between 1995 and 2012 on 1 263 119 individuals (909 368 female, 353 751 male) older than 14 days and younger than 50 years in 64 malaria-endemic countries. Regional growth references were generated using a generalized additive model for location, scale and shape by combining data with varying distributions from a range of sources. Countries were weighted by their population at risk of malaria to enable references to be used in optimizing the dosing of antimalarials. Large differences in weight-for-age distributions existed between the regions and between the regions and global growth standards. For example, the average adult male from the Americas weighed 68.1 kg - 6.0 kg more than males in South-East Asia and the Western Pacific (average: 62.1 kg). For adult women, the difference was over 10.4 kg: the average was 60.4 kg in the Americas and 50.0 kg in South-East Asia and the Western Pacific. There were substantial variations in weight-for-age growth curves between malaria-endemic areas. The growth reference charts derived here can be used to guide the evidence-based optimization of aged-based dosing regimens for antimalarials and other drugs often prescribed by age.
    Bulletin of the World Health Organisation 02/2015; 93(2):74-83. DOI:10.2471/BLT.14.139113 · 5.11 Impact Factor
  • Source
    Gerko Vink, Stef van Buuren
    [Show abstract] [Hide abstract]
    ABSTRACT: Current pooling rules for multiply imputed data assume infinite populations. In some situations this assumption is not feasible as every unit in the population has been observed, potentially leading to over-covered population estimates. We simplify the existing pooling rules for situations where the sampling variance is not of interest. We compare these rules to the conventional pooling rules and demonstrate their use in a situation where there is no sampling variance. Using the standard pooling rules in situations where sampling variance should not be considered, leads to overestimation of the variance of the estimates of interest, especially when the amount of missingness is not very large. As a result, populations estimates are over-covered, which may lead to a loss of statistical power. We conclude that the theory of multiple imputation can be extended to the situation where the sample happens to be the population. The simplified pooling rules can be easily implemented to obtain valid inference in cases where we have observed essentially all units and in simulation studies addressing the missingness mechanism only.
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Although children both at the upper and lower tail of the body mass index (BMI) distribution are at greater health risk, relatively little is known about the development of thinness prevalence rates in developed countries over time. We studied trends in childhood thinness and assessed changes in the BMI distribution since the onset of the obesity epidemic.
    The European Journal of Public Health 08/2014; DOI:10.1093/eurpub/cku130 · 2.46 Impact Factor
  • Gerko VINK, Stef VAN BUUREN
    [Show abstract] [Hide abstract]
    ABSTRACT: We propose a new multiple imputation technique for imputing squares. Cur- rent methods yield either unbiased regression estimates, or preserve data re- lations. No method, however, seems to deliver both, which limits researchers in the implementation of regression analysis in the presence of missing data. Besides, current methods only work under a MCAR mechanism. Our method for imputing squares uses a polynomial combination. The proposed method yields both unbiased regression estimates, whilst at the same time preserving the quadratic relations in the data for both MAR and MCAR mechanisms.
    XVIII ISA World Congress of Sociology; 07/2014
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Abstract Background: People from Asian populations are generally shorter than other ethnic groups. It is unknown if current universal height references are suitable for affluent South Asian children in the Netherlands. Aims: To develop height-for-age charts for contemporary South Asian children aged 0-20 years living in the Netherlands, to evaluate secular trends, and to compare the charts with current Asian Indian, Dutch and WHO references. Subjects and methods: A population-based study measured 3315 South Asian children aged 0-20 years between 2007-2010. Among this cohort, 6876 measurements were taken. Another 7388 measurements were taken of a historical cohort of 1078 children born between 1974-1976 (aged 0-18 years). Results: An upward trend in height was observed for South Asian children living in the Netherlands between 1992-2010. The height-for-age charts of the South Asian historical cohort were similar to current Asian Indian charts. South Asian children in the Netherlands were shorter than their Dutch contemporaries at every age; and these differences increased further during adolescence. Compared to the WHO height-for-age references, there were considerable discrepancies in height, with curves intersecting twice. Conclusion: The discrepancies between the South Asian and Dutch and WHO height-for-age references indicate differences in growth patterns between the source populations.
    Annals of Human Biology 06/2014; 42(1):1-7. DOI:10.3109/03014460.2014.926988 · 1.15 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: To develop a short instrument to measure determinants of innovations that may affect its implementation. We pooled the original data from eight empirical studies of the implementation of evidence-based innovations. The studies used a list of 60 potentially relevant determinants based on a systematic review of empirical studies and a Delphi study among implementation experts. Each study used similar methods to measure both the implementation of the innovation and determinants. Missing values in the final data set were replaced by plausible values using multiple imputation. We assessed which determinants predicted completeness of use of the innovation (% of recommendations applied). In addition, 22 implementation experts were consulted about the results and about implications for designing a short instrument. Eight innovations introduced in Preventive Child Health Care or schools in the Netherlands. Doctors, nurses, doctor's assistants and teachers; 1977 respondents in total. The initial list of 60 determinants could be reduced to 29. Twenty-one determinants were based on the pooled analysis of the eight studies, seven on the theoretical expectations of the experts consulted and one new determinant was added on the basis of the experts' practical experience. The instrument is promising and should be further validated. We invite researchers to use and explore the instrument in multiple settings. The instrument describes how each determinant should preferably be measured (questions and response scales). It can be used both before and after the introduction of an innovation to gain an understanding of the critical change objectives.
    BMJ quality & safety 06/2014; 22(Suppl 1). DOI:10.1093/intqhc/mzu060 · 3.28 Impact Factor
  • Source
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Morbid obesity can be a life threatening condition. The aim of our study is to assess the trend in morbid obesity in The Netherlands among children of Dutch origin since 1980, and among children of Turkish and Moroccan origin since 1997. Cross-sectional height and weight data of children of Dutch, Turkish and Moroccan origin aged 2-18 years were selected from three national Dutch Growth Studies performed in 1980, 1997 and 2009 (n = 54,814). Extended international (IOTF) cut-offs in childhood were used to define morbid obesity (obesity class II and III combined). The morbidity index for overweight was calculated as the prevalence of morbid obesity divided by the prevalence of overweight. Our study showed that the prevalence of morbid obesity in children of Dutch origin was 0.59% in boys and 0.53% in girls in 2009. Significant upward trends occurred since 1980 and 1997. The prevalence was three to four fold higher in Turkish children compared to Dutch children. The Turkish children also had an upward trend since 1997, but this was only statistically significant in boys. The prevalence of morbid obesity in Moroccan children was two to three fold higher than in Dutch children, but it remained almost stable between 1997 and 2009. The Dutch and Turkish children showed an upward trend in morbidity index for overweight since respectively 1980 and 1997, while the Moroccan children showed a downward trend since 1997. In 2009, children of low educated parents had the highest prevalence rates of morbid obesity; 1.06% in Dutch, 2.11% in Turkish and 1.41% in Moroccan children. An upward trend of morbid obesity in Dutch and Turkish children in The Netherlands occurred. Monitoring and reducing the prevalence of childhood morbid obesity is of high importance for these children, health care and the community.
    PLoS ONE 04/2014; 9(4):e94299. DOI:10.1371/journal.pone.0094299 · 3.53 Impact Factor
  • Source
    L. L. Doove, S. Van Buuren, E. Dusseldorp
    [Show abstract] [Hide abstract]
    ABSTRACT: Standard approaches to implement multiple imputation do not automatically incorporate nonlinear relations like interaction effects. This leads to biased parameter estimates when interactions are present in a dataset. With the aim of providing an imputation method which preserves interactions in the data automatically, the use of recursive partitioning as imputation method is examined. Three recursive partitioning techniques are implemented in the multiple imputation by chained equations framework. It is investigated, using simulated data, whether recursive partitioning creates appropriate variability between imputations and unbiased parameter estimates with appropriate confidence intervals. It is concluded that, when interaction effects are present in a dataset, substantial gains are possible by using recursive partitioning for imputation compared to standard applications. In addition, it is shown that the potential of recursive partitioning imputation approaches depends on the relevance of a possible interaction effect, the correlation structure of the data, and the type of possible interaction effect present in the data.
    Computational Statistics & Data Analysis 04/2014; 72:92–104. DOI:10.1016/j.csda.2013.10.025 · 1.15 Impact Factor
  • Source
    Stef van Buuren, Rom Perenboom
    [Show abstract] [Hide abstract]
    ABSTRACT: There are very few norms to evaluate and monitor the health and functioning of the elderly. This paper proposes a compact spatial representation of 25 health measurements of European citizens older than 50 years. Data from 44,285 unique individuals were obtained from the EU-wide Survey of Health, Ageing and Retirement in Europe(SHARE) data collected in 2004-2007 and were analyzed by homogeneity analysis, a form of non-linear principal components analysis. The resulting configuration of persons shows a remarkable three-dimensional shape that resembles a fountain. The three components explain 13.7, 5.8 and 4.6 percent of the total variation, respectively. Component 1 is driven by age and by the disabilities that come with old age. Component 2 portrays differences in health that are independent of age, with the high scores in relatively good health, given age. Component 3 distinguishes specific types of functional decline from general complaints that impact on daily life. The shape suggests that the elderly keep on maturing as they grow older, actually becoming more diverse as a group. We show how the solution may be used to develop and support profiles for the elderly. Another potential application is to track the individual development of the elderly, thereby objectifying personalized medicine.
    International Journal of Environmental Research and Public Health 04/2014; 11(4):4078-4090. DOI:10.3390/ijerph110404078 · 1.99 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: The association between poverty, malnutrition, illness and poor socioeconomic conditions on the one side, and poor growth and short adult stature on the other side, is well recognized. Yet, the simple assumption by implication that poor growth and short stature result from poor living conditions, should be questioned. Recent evidence on the impact of the social network on adolescent growth and adult height further challenges the traditional concept of growth being a mirror of health. Twenty-nine scientists met at Glücksburg castle, Northern Germany, November 15th - 17th 2013, to discuss genetic, endocrine, mathematical and psychological aspects and related issues, of child and adolescent growth and final height.
    Pediatric endocrinology reviews: PER 03/2014; 11(3):341-53.
  • P. van Dommelen, S. van Buuren, M. Roelants
    [Show abstract] [Hide abstract]
    ABSTRACT: InleidingGroei is een afspiegeling van de gezondheid van een populatie. Als we naar de Nederlandse geschiedenis kijken, zijn er al sinds het jaar 50 van onze jaartelling gegevens over lengte beschikbaar. Deze gegevens zijn veelal afgeleid uit metingen van pijpbeenderen of skeletten en gebaseerd op kleine aantallen en verschillende bevolkingsgroepen. Opvallend is dat de gemiddelde eindlengte van mannen tussen het jaar 50 en, ongeveer, het jaar 1300 lijkt af te nemen van 176 cm naar 170 cm.1 Factoren die een rol hebben gespeeld in deze verandering in eindlengte zijn de sociaal-economische condities, die onder andere de beschikbaarheid en kwaliteit van voeding en de hygiëne bepalen. In de periode tussen 1300 en 1825 bleef de eindlengte redelijk stabiel op 170 cm.1 In het jaar 1825 werd voor het eerst een groeistudie gehouden met levende mensen. Deze studie is uitgevoerd onder stadswachten uit zeven steden. De gemiddelde eindlengte was toen 169,0 cm.1 Mensen leefden in die tijd onder slech ...
    02/2014; 46(1):2-4. DOI:10.1007/s12452-014-0002-6
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Multiple imputation methods properly account for the uncertainty of missing data. One of those methods for creating multiple imputations is predictive mean matching (PMM), a general purpose method. Little is known about the performance of PMM in imputing non-normal semicontinuous data (skewed data with a point mass at a certain value and otherwise continuously distributed). We investigate the performance of PMM as well as dedicated methods for imputing semicontinuous data by performing simulation studies under univariate and multivariate missingness mechanisms. We also investigate the performance on real-life datasets. We conclude that PMM performance is at least as good as the investigated dedicated methods for imputing semicontinuous data and, in contrast to other methods, is the only method that yields plausible imputations and preserves the original data distributions.
    Statistica Neerlandica 02/2014; 68(1). DOI:10.1111/stan.12023 · 0.48 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: It is known that height and body mass index (BMI) are correlated in childhood. However, its impact on the (trend of) national prevalence rates of overweight and obesity has never been investigated. The aim of our study is to investigate the relation between height and national prevalence rates of overweight and obesity in childhood between 1980, 1997, and 2009, and to calculate which fixed value of p (2.0,2.1, …,3.0) in kg/m(p) during childhood is most accurate in predicting adult overweight. Cross-sectional growth data of children from three Dutch nationwide surveys in 1980, 1997, and 2009, and longitudinal data from the Terneuzen Birth Cohort and the Harpenden Growth Study were used. Relative risks (RR) and 95% confidence intervals (CI) were calculated. Our study showed that tall (>1 standard deviation (SD)) girls aged 5.0-13.9 y were more often overweight (RR = 3.5,95%CI:2.8-4.4) and obese (RR = 3.9,95%CI:2.1-7.4) than short girls (<-1 SD). Similar results were found in boys aged 5.0-14.9 y (RR = 4.4,95%CI:3.4-5.7 and RR = 5.3,95%CI:2.6-11.0). No large differences were found in the other age groups and in comparison with children with an average stature. Tall boys aged 2.0-4.9 y had a significantly higher positive trend in overweight between 1980 and 1997 compared to short boys (RR = 4.0,95%CI:1.38-11.9). For other age groups and in girls, no significant trends were found. The optimal Area Under the Curve (AUC) to predict adult overweight was found for p = 2.0. Tall girls aged 5.0-13.9y and tall boys aged 5.0-14.9y have much higher prevalence rates of overweight and obesity than their shorter peers. We suggest taking into account the impact of height when evaluating trends and variations of BMI distributions in childhood, and to use BMI to predict adult overweight.
    PLoS ONE 01/2014; 9(1):e85769. DOI:10.1371/journal.pone.0085769 · 3.53 Impact Factor

Publication Stats

5k Citations
365.94 Total Impact Points


  • 2010–2015
    • Nederlandse Organisatie voor Wetenschappelijk Onderzoek
      's-Gravenhage, South Holland, Netherlands
    • VU University Medical Center
      • Department of Public and Occupational Health
      Amsterdam, North Holland, Netherlands
  • 1996–2015
    • TNO
      Delft, South Holland, Netherlands
  • 1989–2014
    • Utrecht University
      • • Department of Methodology and Statistics
      • • Department of Radiology
      Utrecht, Utrecht, Netherlands
  • 2012
    • University of Applied Sciences Leiden
      Leyden, South Holland, Netherlands
    • University of Zurich
      • Centre for Evolutionary Medicine - EZM
      Zürich, Zurich, Switzerland
  • 2005–2011
    • Centraal Bureau voor de Statistiek
      's-Gravenhage, South Holland, Netherlands
    • Liverpool School of Tropical Medicine
      • Child and Reproductive Health Group
      Liverpool, England, United Kingdom
  • 2000–2010
    • Leiden University Medical Centre
      • Department of Pediatrics
      Leiden, South Holland, Netherlands
  • 2006
    • Harvard University
      Cambridge, Massachusetts, United States
    • University of Groningen
      • Department of Health Sciences
      Groningen, Groningen, Netherlands
  • 2004
    • Leiden University
      Leyden, South Holland, Netherlands
  • 2001
    • VU University Amsterdam
      Amsterdamo, North Holland, Netherlands