Immune response-eliciting exposure to Campylobacter vastly exceeds the incidence of clinically overt campylobacteriosis but is associated with similar risk factors: A nationwide serosurvey in the Netherlands

Susana Monge, Peter Teunis, Ingrid Friesema, Eelco Franz, Wim Ang, Wilfrid van Pelt, Lapo Mughini-Gras

Research output: Contribution to journalArticleAcademicpeer-review

Abstract

Background: We aimed to estimate population-level exposure to Campylobacter and associated risk factors, using three approaches for serological data analysis. Methods: Nationwide, population-based serosurvey in the Netherlands (Feb 2006–Jun 2007). Anti-Campylobacter IgG, IgM and IgA were measured using ELISA, and analysed via: a) seroincidence estimation, using reference values of antibody peak levels and decay rates over-time after Campylobacter exposure; b) two normal distributions of true positives/negatives fitted to the IgG distribution to derive seroprevalence and individual probability of being positive/negative; and c) IgG levels. Risk factors were analysed using multiple linear regressions. Results: From 1559 respondents, seroincidence was estimated at 1.61 infections/person-year (95%CI:1.58–1.64) and seroprevalence at 68.1% (65.4–70.9). The three approaches identified similar risk factors, although seroincidence had higher power and results were interpretable as risk: seroincidence was higher in females [exp(b) = 1.07(1.04–1.11)], older ages [vs. 15–34 years; for < 5, 5–14, 35–54 and 55–70 years: 0.60(0.58–0.63), 0.74(0.71–0.78), 1.08(1.03–1.13) and 1.08(1.01–1.16), respectively], non-Dutch background [Moroccan/Turkish: 1.25(1.14–1.37); Caribbean: 1.14(1.03–1.25)], low socioeconomic status [1.05(1.01–1.10)], traveling outside Europe [1.05(1.01–1.09)], and eating undercooked meat [1.04(1.01–1.08)]. Conclusion: Campylobacter exposure is much higher than clinical infection rates, but risk factors are similar to those previously described.Seroincidence is a powerful measure to study Campylobacter epidemiology, and is preferred over other methods.
Original languageEnglish
Pages (from-to)171-177
JournalJournal of Infection
Volume77
Issue number3
DOIs
Publication statusPublished - 2018

Cite this

Monge, Susana ; Teunis, Peter ; Friesema, Ingrid ; Franz, Eelco ; Ang, Wim ; van Pelt, Wilfrid ; Mughini-Gras, Lapo. / Immune response-eliciting exposure to Campylobacter vastly exceeds the incidence of clinically overt campylobacteriosis but is associated with similar risk factors: A nationwide serosurvey in the Netherlands. In: Journal of Infection. 2018 ; Vol. 77, No. 3. pp. 171-177.
@article{3512f9c6563c4135b7280c2690da608c,
title = "Immune response-eliciting exposure to Campylobacter vastly exceeds the incidence of clinically overt campylobacteriosis but is associated with similar risk factors: A nationwide serosurvey in the Netherlands",
abstract = "Background: We aimed to estimate population-level exposure to Campylobacter and associated risk factors, using three approaches for serological data analysis. Methods: Nationwide, population-based serosurvey in the Netherlands (Feb 2006–Jun 2007). Anti-Campylobacter IgG, IgM and IgA were measured using ELISA, and analysed via: a) seroincidence estimation, using reference values of antibody peak levels and decay rates over-time after Campylobacter exposure; b) two normal distributions of true positives/negatives fitted to the IgG distribution to derive seroprevalence and individual probability of being positive/negative; and c) IgG levels. Risk factors were analysed using multiple linear regressions. Results: From 1559 respondents, seroincidence was estimated at 1.61 infections/person-year (95{\%}CI:1.58–1.64) and seroprevalence at 68.1{\%} (65.4–70.9). The three approaches identified similar risk factors, although seroincidence had higher power and results were interpretable as risk: seroincidence was higher in females [exp(b) = 1.07(1.04–1.11)], older ages [vs. 15–34 years; for < 5, 5–14, 35–54 and 55–70 years: 0.60(0.58–0.63), 0.74(0.71–0.78), 1.08(1.03–1.13) and 1.08(1.01–1.16), respectively], non-Dutch background [Moroccan/Turkish: 1.25(1.14–1.37); Caribbean: 1.14(1.03–1.25)], low socioeconomic status [1.05(1.01–1.10)], traveling outside Europe [1.05(1.01–1.09)], and eating undercooked meat [1.04(1.01–1.08)]. Conclusion: Campylobacter exposure is much higher than clinical infection rates, but risk factors are similar to those previously described.Seroincidence is a powerful measure to study Campylobacter epidemiology, and is preferred over other methods.",
author = "Susana Monge and Peter Teunis and Ingrid Friesema and Eelco Franz and Wim Ang and {van Pelt}, Wilfrid and Lapo Mughini-Gras",
year = "2018",
doi = "10.1016/j.jinf.2018.04.016",
language = "English",
volume = "77",
pages = "171--177",
journal = "The Journal of Infection",
issn = "0163-4453",
publisher = "W.B. Saunders Ltd",
number = "3",

}

Immune response-eliciting exposure to Campylobacter vastly exceeds the incidence of clinically overt campylobacteriosis but is associated with similar risk factors: A nationwide serosurvey in the Netherlands. / Monge, Susana; Teunis, Peter; Friesema, Ingrid; Franz, Eelco; Ang, Wim; van Pelt, Wilfrid; Mughini-Gras, Lapo.

In: Journal of Infection, Vol. 77, No. 3, 2018, p. 171-177.

Research output: Contribution to journalArticleAcademicpeer-review

TY - JOUR

T1 - Immune response-eliciting exposure to Campylobacter vastly exceeds the incidence of clinically overt campylobacteriosis but is associated with similar risk factors: A nationwide serosurvey in the Netherlands

AU - Monge, Susana

AU - Teunis, Peter

AU - Friesema, Ingrid

AU - Franz, Eelco

AU - Ang, Wim

AU - van Pelt, Wilfrid

AU - Mughini-Gras, Lapo

PY - 2018

Y1 - 2018

N2 - Background: We aimed to estimate population-level exposure to Campylobacter and associated risk factors, using three approaches for serological data analysis. Methods: Nationwide, population-based serosurvey in the Netherlands (Feb 2006–Jun 2007). Anti-Campylobacter IgG, IgM and IgA were measured using ELISA, and analysed via: a) seroincidence estimation, using reference values of antibody peak levels and decay rates over-time after Campylobacter exposure; b) two normal distributions of true positives/negatives fitted to the IgG distribution to derive seroprevalence and individual probability of being positive/negative; and c) IgG levels. Risk factors were analysed using multiple linear regressions. Results: From 1559 respondents, seroincidence was estimated at 1.61 infections/person-year (95%CI:1.58–1.64) and seroprevalence at 68.1% (65.4–70.9). The three approaches identified similar risk factors, although seroincidence had higher power and results were interpretable as risk: seroincidence was higher in females [exp(b) = 1.07(1.04–1.11)], older ages [vs. 15–34 years; for < 5, 5–14, 35–54 and 55–70 years: 0.60(0.58–0.63), 0.74(0.71–0.78), 1.08(1.03–1.13) and 1.08(1.01–1.16), respectively], non-Dutch background [Moroccan/Turkish: 1.25(1.14–1.37); Caribbean: 1.14(1.03–1.25)], low socioeconomic status [1.05(1.01–1.10)], traveling outside Europe [1.05(1.01–1.09)], and eating undercooked meat [1.04(1.01–1.08)]. Conclusion: Campylobacter exposure is much higher than clinical infection rates, but risk factors are similar to those previously described.Seroincidence is a powerful measure to study Campylobacter epidemiology, and is preferred over other methods.

AB - Background: We aimed to estimate population-level exposure to Campylobacter and associated risk factors, using three approaches for serological data analysis. Methods: Nationwide, population-based serosurvey in the Netherlands (Feb 2006–Jun 2007). Anti-Campylobacter IgG, IgM and IgA were measured using ELISA, and analysed via: a) seroincidence estimation, using reference values of antibody peak levels and decay rates over-time after Campylobacter exposure; b) two normal distributions of true positives/negatives fitted to the IgG distribution to derive seroprevalence and individual probability of being positive/negative; and c) IgG levels. Risk factors were analysed using multiple linear regressions. Results: From 1559 respondents, seroincidence was estimated at 1.61 infections/person-year (95%CI:1.58–1.64) and seroprevalence at 68.1% (65.4–70.9). The three approaches identified similar risk factors, although seroincidence had higher power and results were interpretable as risk: seroincidence was higher in females [exp(b) = 1.07(1.04–1.11)], older ages [vs. 15–34 years; for < 5, 5–14, 35–54 and 55–70 years: 0.60(0.58–0.63), 0.74(0.71–0.78), 1.08(1.03–1.13) and 1.08(1.01–1.16), respectively], non-Dutch background [Moroccan/Turkish: 1.25(1.14–1.37); Caribbean: 1.14(1.03–1.25)], low socioeconomic status [1.05(1.01–1.10)], traveling outside Europe [1.05(1.01–1.09)], and eating undercooked meat [1.04(1.01–1.08)]. Conclusion: Campylobacter exposure is much higher than clinical infection rates, but risk factors are similar to those previously described.Seroincidence is a powerful measure to study Campylobacter epidemiology, and is preferred over other methods.

UR - https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85047192274&origin=inward

UR - https://www.ncbi.nlm.nih.gov/pubmed/29746943

U2 - 10.1016/j.jinf.2018.04.016

DO - 10.1016/j.jinf.2018.04.016

M3 - Article

VL - 77

SP - 171

EP - 177

JO - The Journal of Infection

JF - The Journal of Infection

SN - 0163-4453

IS - 3

ER -