Geographic aspects of enterobiasis in Estonia

Date

2009-07-02T11:23:10Z

Journal Title

Journal ISSN

Volume Title

Publisher

Abstract

This investigation focused on the prevalence, prediction, risk factors and geographical aspects of enterobiasis in Estonia. Enterobius vermicularis is a widespread parasitic risk factor for a healthy environment, in particular for children. The correct diagnosis of pinworm is complicated and the elimination of the parasite from a family group or institution often poses significant problems though its treatment is straightforward. There was not reliable up-to-date information about the occurrence of E. vermicularis in Estonia. The investigation was conducted among nursery school children in the capital Tallinn and seven counties. The anal swab examination involved 3131 children from 279 groups in 80 nursery schools. The data on possible risk factors were obtained from questionnaires for children's parents, observations of nursery schools and structured interviews with school staff. Geographical samples were formed according to the three principles: 1) six regional units according to geographical location; 2) according to the size of the settlement: large towns, small towns, rural settlements; 3) the most detailed division into seventeen small geographical units combining the size of settlement and geographical district. In order to model the risk of enterobiasis the similarity-based machine learning and prediction software Constud was compared with data mining methods available in the Statistica 8 Data Miner (Statsoft Inc) software package. The main results of this study are as follows. The prevalence of enterobiasis among nursery school children is high in Estonia. This can be over 40% if to take into account the results from the multiple examination and the calculated predictions. 74% of nursery groups were infected. 83% of studied children belonged to the infected groups. The mean prevalence of enterobiasis according to settlement type varied being the lowest in Tallinn (and other large towns) and the highest in rural settlements. Variables as: region according to the detailed division, the age of the child, the mean age of the children in the nursery group were among the more indicative features for risk modelling. From risk modelling methods, Constud, random forest classification, and boosting classification trees from the Statistica 8 Data Miner were successful. Käesolev uuring keskendus enterobiaasi esinemissageduse, prognoosimisvõimaluste, riskitegurite ja leviku geograafiliste aspektide väljaselgitamisele. Naaskelsaba (Enterobius vermicularis) on maailmas, sealhulgas paraskliima vööndis, laialtlevinud kontakthelmint, kellega nakatuvad eelkõige lapsed. Naaskelsaba diagnoosimine ja nakatatuse õige taseme kindlakstegemine on sageli aeganõudev. Ravi on küll lihtne, kuid naaskelsabade tõrjumine perekonnast või lasterühmadest võib palju vaeva nõuda. Tõesed andmed naaskelsaba levikust Eestis puudusid. Uuring toimus Tallinnas ja seitsmes maakonnas. Uuringusse kaasati 3131 last 80 lasteaia 279 rühmast, neilt kõigilt koguti perianaalkaaped. Riskitegurite kindlakstegemiseks anketeeriti lapsevanemaid, viidi läbi vaatlus lasteaedade rühmaruumides ja struktureeritud intervjuu personaliga. Kogu valim jagati geograafilisteks valimiteks kolmel viisil: kuueks regionaalseks üksuseks lähtuvalt geograafilisest paiknemisest; kolmeks asumi tüübi alusel: suured linnad, väikesed linnad, maapiirkond; seitsmeteistkümneks väikeseks üksuseks kombineerides asumitüüpi geograafilise paiknemisega. Enterobiaasi hindamisel kasutati ja tõhususes võrreldi Statsoft Statistica andmekaevandamispaketti kuuluvaid meetodeid ja Tartu Ülikooli geoinformaatika õppetooli juures arendatud tarkvara Constud. Uuringu tulemusena leiti, et enterobiaasi esinemissagedus lasteaialaste seas on kõrge. Kui arvestada kordusuuringute tulemusi ja arvutuslikult saadud hinnanguid, võib see olla üle 40%. 74% rühmadest olid nakatunud ja 83% uuritud lastest kuulus nakatunud rühmadesse. Asumitüüpide alusel saadud valimites oli keskmine nakatatus erinev: madalaim Tallinnas ja teistes suurtes linnades ja kõrgeim maapiirkonnas. Võimalikest indikaatortunnustest olid parimad detailne piirkond, lapse vanus ja laste keskmine vanus lasteaiarühmas. Uuritud meetoditest olid rühma nakatatuse indiviidipõhisel hindamisel tõhusamad random forest classification, boosting classification trees ja Constud.

Description

Keywords

Citation