Prediction of<i>Chlamydia trachomatis</i>infection to facilitate selective screening on population and individual level: a cross-sectional study of a population-based screening programme

Klaveren, David van; Götz, Hannelore M; Coul, Eline Lm Op de; Vergouwe, Yvonne

doi:10.1136/sextrans-2015-052048

Cited by 6 publications

(6 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Previous studies also found clustering of Ng within low SES regions and among migrant populations 9–11 16 18. Neighbourhood and regional SES had no influence on regional variance in Ct positivity, as is also described previously 19. However, regional degree of urbanisation was an important contributor to regional variance in Ct.…”

Section: Discussionsupporting

confidence: 76%

Regional differences in chlamydia and gonorrhoeae positivity rate among heterosexual STI clinic visitors in the Netherlands: contribution of client and regional characteristics as assessed by cross-sectional surveillance data

et al. 2019

Self Cite

View full text Add to dashboard Cite

ObjectivesTo assess to what extent triage criteria, client and regional characteristics explain regional differences in Chlamydia trachomatis (Ct) and Neisseria gonorrhoeae (Ng) positivity in sexually transmitted infection (STI) clinics.DesignRetrospective cross-sectional study on the Dutch STI surveillance database of all 24 STI clinics.ParticipantsSTI clinic visits of heterosexual persons in 2015 with a Ct (n=101 495) and/or Ng test (n=101 081).Primary outcome measureCt and Ng positivity and 95% CI was assessed for each STI clinic. Two-level logistic regression analyses were performed to calculate the percentage change in regional variance (PCV) after adding triage criteria (model 1), other client characteristics (model 2) and regional characteristics (model 3) to the empty model. The contribution of single characteristics was determined after removing them from model 3.ResultsCt positivity was 14.9% and ranged from 12.6% to 20.0% regionally. Ng positivity was 1.7% and ranged from 0.8% to 3.8% regionally. For Ct, the PCV was 11.7% in model 1, 32.2% in model 2% and 59.3% in model 3. Age, notified for Ct (triage), level of education (other characteristics) and regional degree of urbanisation (region) explained variance most. For Ng, the PCV was 38.7% in model 1, 61.2% in model 2% and 69.1% in model 3. Ethnicity (triage), partner in risk group, level of education and neighbourhood (other characteristics) and regional socioeconomic status (SES) explained variance most. A significant part of regional variance remained unexplained.ConclusionsRegional variance was explained by differences in client characteristics, indicating that triage and self-selection influence positivity rates in the surveillance data.Clustering of Ng in low SES regions additionally explained regional variance in Ng; targeted interventions in low SES regions may assist Ng control. Including educational level as triage criterion is recommended. Studies incorporating prevalence data are needed to assess whether regional clustering underlies unexplained regional variance.

show abstract

Section: Discussionsupporting

confidence: 76%

Regional differences in chlamydia and gonorrhoeae positivity rate among heterosexual STI clinic visitors in the Netherlands: contribution of client and regional characteristics as assessed by cross-sectional surveillance data

et al. 2019

Self Cite

View full text Add to dashboard Cite

show abstract

“…For example, a recently developed prediction model to screen for Chlamydia trachomatis infection was developed in a dataset that was clustered within neighborhoods. 41 Given that the intended user of the model screens individuals from various neighborhoods, population-level performance measures were of interest.…”

Section: Declaration Of Conflicting Interestsmentioning

confidence: 99%

Does ignoring clustering in multicenter data influence the performance of prediction models? A simulation study

Wynants

Vergouwe

Huffel

et al. 2016

Stat Methods Med Res

View full text Add to dashboard Cite

Clinical risk prediction models are increasingly being developed and validated on multicenter datasets. In this article, we present a comprehensive framework for the evaluation of the predictive performance of prediction models at the center level and the population level, considering population-averaged predictions, center-specific predictions, and predictions assuming an average random center effect. We demonstrated in a simulation study that calibration slopes do not only deviate from one because of over- or underfitting of patterns in the development dataset, but also as a result of the choice of the model (standard versus mixed effects logistic regression), the type of predictions (marginal versus conditional versus assuming an average random effect), and the level of model validation (center versus population). In particular, when data is heavily clustered (ICC 20%), center-specific predictions offer the best predictive performance at the population level and the center level. We recommend that models should reflect the data structure, while the level of model validation should reflect the research question.

show abstract

“…Two types of prediction tools for binary outcomes can be distinguished: (1) a tool that can be used to predict an individual's probability of the presence of disease at the moment of prediction (i.e., a diagnostic prediction model) and (2) one that can be used to predict the probability of the future occurrence of an event (i.e., a prognostic prediction model). An example of the former is a model to estimate the probability of Chlamydia trachomatis infection to aid selective screening of youth at high risk of an infection [3]. An example of a prognostic prediction model to estimate an individual's probability of a future event is a model that estimates the probability of a successful vaginal birth after previous caesarean section, which is subsequently included in a decision aid to discuss the intended mode of delivery [4,5].…”

Section: Introductionmentioning

confidence: 99%

Preparing Data for Predictive Modelling

Kuijk

Dankers

Traverso

et al. 2018

Fundamentals of Clinical Data Science

View full text Add to dashboard Cite

This is the first chapter of five that cover an introduction to developing and validating models for predicting outcomes for the individual patient. Such prediction models can be used for predicting the occurrence or recurrence of an event, or of the most likely value on a continuous outcome. We will mainly focus on the prediction of binary outcomes, such as the occurrence of a complication, recurrence of disease, the presence of metastases, remission, survival, etc. This chapter deals with the selection of an appropriate study design for a study on prediction, and on methods to manipulate the data before the statistical modelling can begin.

show abstract

Prediction ofChlamydia trachomatisinfection to facilitate selective screening on population and individual level: a cross-sectional study of a population-based screening programme

Abstract: A registry-based prediction model can facilitate selective Ct screening at population level, with further refinement at the individual level by including questionnaire risk factors.

Cited by 6 publications

References 26 publications

Regional differences in chlamydia and gonorrhoeae positivity rate among heterosexual STI clinic visitors in the Netherlands: contribution of client and regional characteristics as assessed by cross-sectional surveillance data

Regional differences in chlamydia and gonorrhoeae positivity rate among heterosexual STI clinic visitors in the Netherlands: contribution of client and regional characteristics as assessed by cross-sectional surveillance data

Does ignoring clustering in multicenter data influence the performance of prediction models? A simulation study

Preparing Data for Predictive Modelling

Contact Info

Product

Resources

About