UBC Faculty Research and Publications

Water and sewage systems, socio-demographics, and duration of residence associated with endemic intestinal… Teschke, Kay; Bellack, Neil; Shen, Hui; Atwater, Jim; Chu, Rong; Koehoorn, Mieke; MacNab, Ying C; Schreier, Hans; Isaac-Renton, Judith L Dec 16, 2010

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
52383-12889_2010_Article_2678.pdf [ 967.86kB ]
Metadata
JSON: 52383-1.0223829.json
JSON-LD: 52383-1.0223829-ld.json
RDF/XML (Pretty): 52383-1.0223829-rdf.xml
RDF/JSON: 52383-1.0223829-rdf.json
Turtle: 52383-1.0223829-turtle.txt
N-Triples: 52383-1.0223829-rdf-ntriples.txt
Original Record: 52383-1.0223829-source.json
Full Text
52383-1.0223829-fulltext.txt
Citation
52383-1.0223829.ris

Full Text

RESEARCH ARTICLE Open AccessWater and sewage systems, socio-demographics,and duration of residence associated with endemicintestinal infectious diseases: A cohort studyKay Teschke1*, Neil Bellack1, Hui Shen1, Jim Atwater2, Rong Chu1, Mieke Koehoorn1, Ying C MacNab1,Hans Schreier3, Judith L Isaac-Renton4AbstractBackground: Studies of water-related gastrointestinal infections are usually directed at outbreaks. Few haveexamined endemic illness or compared rates across different water supply and sewage disposal systems. Weconducted a cohort study of physician visits and hospitalizations for endemic intestinal infectious diseases in amixed rural and urban community near Vancouver, Canada, with varied and well-characterized water and sewagesystems.Methods: Cohort members and their disease events were defined via universal health insurance data from 1995through 2003. Environmental data were derived from municipal, provincial, and federal government sources.Logistic regression was used to examine associations between disease events and water and sewage systems,socio-demographic characteristics, and temporal factors.Results: The cohort included 126,499 individuals and approximately 190,000,000 person-days. Crude incidence rateswere 1,353 physician visits and 33.8 hospitalizations for intestinal infectious diseases per 100,000 person-years.Water supply chlorination was associated with reduced physician visit incidence (OR: 0.92, 95% CI 0.85-1.0). Twowater systems with the highest proportions of surface water had increased incidence (ORs: 1.57, 95% CI 1.39-1.78;and 1.45, 95% CI 1.28-1.64). Private well water and well depth were not associated with increased risk, likelybecause of residents’ awareness of and attention to water quality. There was increased crude incidence withincreasing precipitation in the population served by surface water supplies, but this trend did not remain withadjustment for other variables. Municipal sewer systems were associated with increased risk (OR: 1.26, 95% CI 1.14-1.38). Most socio-demographic variables had predicted associations with risk: higher rates in females, in the veryyoung and the elderly, and in residents of low income areas. Increased duration of area residence was associatedwith reduced risk (OR, duration ≥ 6 years: 0.69, 95% CI 0.60-0.80 vs. < 1 year: 1.16, 95% CI 1.03-1.30).Conclusions: This large cohort study, with objective data on exposures and outcomes, demonstrated associationsbetween endemic infectious intestinal diseases and factors related to water supply, sewage disposal, socio-demographics, and duration of residency. The results did not always follow prior expectations based on studiesexamining outbreaks and single systems, and underscore the importance of studying factors associated withendemic disease across water and sewage system types.* Correspondence: kay.teschke@ubc.ca1School of Population and Public Health, University of British Columbia,Vancouver, CanadaFull list of author information is available at the end of the articleTeschke et al. BMC Public Health 2010, 10:767http://www.biomedcentral.com/1471-2458/10/767© 2010 Teschke et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative CommonsAttribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction inany medium, provided the original work is properly cited.BackgroundMost studies of water and enteric infections havefocused on disease outbreaks in single water systems [1].High profile examples in North America include the1993 Milwaukee Cryptosporidium outbreak whereapproximately 100 people died and a further 400,000became ill [2], and the 2000 Walkerton Escherichia coliO157 and Campylobacter outbreak that caused 7 knowndeaths and about 2,300 illnesses [3-5].Few studies have examined non-epidemic illness. Cana-dian surveillance data suggest that potentially waterborneintestinal infectious diseases account for about 20% ofreported communicable disease cases [6,7]. A number ofinvestigations have examined factors influencing endemicintestinal disease within water systems, including turbid-ity, chlorination, residence time, and system maintenance[8-14]. Studies of endemic disease across multiple wateror sewage systems are very rare [15-17].We investigated the association between the incidenceof intestinal infectious diseases and environmental factorspotentially contributing to drinking water quality in theTownship of Langley in the Metro Vancouver area of Brit-ish Columbia (BC), Canada. The Township was selectedfor study, not because of concerns about endemic diseasein the area, but because it encompasses rural and urbanareas with varied and well-characterized water supply andsewage disposal systems. In addition, individual-level datawere available about physician visits and hospitalizations,and about socio-demographic characteristics of the popu-lation that might also be associated with disease events.MethodsThe study methods were reviewed and approved by theUniversity of British Columbia Behavioural ResearchEthics Board and by the British Columbia Ministry ofHealth Services data steward.Cohort identificationThe cohort was identified by Population Data BC [18].Its data holdings include the Client Registry of the pro-vincial Medical Services Plan (universal health insurance,estimated to enumerate over 95% of the population), allhospital discharge records, and billing records for visitsto physicians. The cohort included all individuals wholived in the Township for at least 6 months betweenJanuary 1, 1995 and December 31, 2003.Disease eventsTwo types of disease events were identified by linkingcohort members to administrative records: a) Physicianvisit, a record in the Medical Services Plan Billings Filewith a 3-digit International Classification of Diseases, 9thRevision (ICD-9) [19] “diagnostic code” listed in Table1; and b) Hospitalization, a record in the HospitalDischarge Records File with a “most responsible diagno-sis” or “primary diagnosis” indicating a 3- to 5-digitICD-9 code in the range listed in Table 1.The ICD-9 codes were selected in consultationsbetween the study medical microbiologist (JLI-R) andDr. Robert Fisk of the BC Ministry of Health Services[personal communication, September 2003] to includeintestinal infectious diseases with potential to be water-borne and, where possible within the constraints of dis-ease coding, to exclude diseases known not to beassociated with waterborne disease transmission inCanada, such as cholera, typhoid fever, and amoebiasis,and gastrointestinal illnesses known to be predominantlyfood-borne (e.g., staphylococcal food poisoning, botu-lism, enteritis necroticans).Socio-demographic and temporal variablesData on sex and birth date were abstracted from the Cli-ent Registry. Season was defined as spring (March 1 toMay 31), summer (June 1 to August 31), fall (September1 to November 30) and winter (December 1 to February29) to correspond more closely to local weather, employ-ment, and school attendance patterns than classical sea-sonal definitions, and thus to control for seasonalpatterns of interpersonal contact that might influenceperson-to-person transmission (e.g., indoor living, schoolyear). Duration of residence in the Township was categor-ized as < 1 year, 1 - < 2 years, 2 - < 3 years, 3 - < 6 years,≥ 6 years, and unknown. Household income quintile ofthe neighborhood was derived from Statistics Canada2001 census data available for each dissemination area(one or more neighboring blocks with a population ofTable 1 Number of subjects with physician visits andhospitalizations for intestinal infectious diseases, 1995 to2003 inclusive, in a mixed rural-urban community inMetro Vancouver.Number of subjects with aICD-9Code*Diseasedescriptionphysician visit hospitalization003 Other Salmonellainfections65 0004 Shigellosis 20 0007 Other protozoal intestinaldiseases (includesbalantidiasis,giardiasis, coccidiosis,intestinal trichomoniasis,cryptosporidiosis,cyclosporiasis)21 0008 Intestinal infections dueto other organisms721 †180009 Ill-defined intestinalinfections6190 0Total all diseases above 7017 180* International Classification of Diseases, 9th Revision [19]Teschke et al. BMC Public Health 2010, 10:767http://www.biomedcentral.com/1471-2458/10/767Page 2 of 13400 - 600 persons), and assigned via residential postalcode. Population density was calculated as the averagenumber of persons per hectare per year at each property.Distance to nearest hospital was calculated from the cen-ter of each property to the nearest of two local hospitals,and classified as < 1 km, 1 - < 3 km, and 3 - < 8 km.Environmental variables potentially contributing todrinking water qualityEnvironmental variables were linked geographically viaspatial link functions in ArcGIS 9 (ESRI, Redlands, CA)using latitude and longitude coordinates, unique propertyidentifiers, or street addresses, then linked to subjects viastreet addresses at the Ministry of Health Services.Water systemAll addresses supplied by municipal water systems wereidentified via Township tax records of water connection,further specified as one of 7 categories by location. Thisallowed distinctions between municipal systems servedby wells only (further categorized by the numbers ofhomes served as small, with < 100 connections, or largesystems), and those with mixed water systems suppliedfrom both Township wells and protected surface waterreservoirs in the North Shore Mountains (further cate-gorized by relative proportions of surface water in themix). The municipal wells and distribution systems aremanaged and tested by the Township, and the reservoirsare managed and tested by the Greater VancouverWater District. Addresses with water supplied by com-munity systems were identified via Fraser HealthAuthority records. These systems are managed privatelyby each community, but tested by health authority offi-cials. All remaining addresses were designated as havingprivate systems (served by on-property wells, with nopublic health monitoring).Drinking water disinfectionDates of implementation of chlorination systems wereused to indicate presence or absence of chlorine disin-fection. Data were provided by the Township for themunicipal systems and by Fraser Health Authority forthe community systems. All private water supplies weredesignated as having no disinfection.Sewage disposalAll residential addresses connected to the municipal sew-age system were identified via Township tax records ofsanitary sewer connection. All other addresses weredesignated as having private on-property sewage disposal.Land useProperty land uses were classified as agricultural or resi-dential, via BC Assessment records.Well depthData were available for a subset of wells from a databasebased on voluntary reporting by well drillers to the BCMinistry of Environment.PrecipitationDaily precipitation data were downloaded from theEnvironment Canada National Climate Data for a sta-tion in the Coquitlam watershed in the North ShoreMountains, and assigned to properties receiving surfacewater from the mountain reservoirs.Survey of Township householdsTo gather descriptive data not available from adminis-trative sources, from fall 2006 to summer 2007, we con-ducted a survey of a random sample (N = 1000) ofhouseholds with Township postal codes, selected fromthe CanPages electronic directory. A self-administeredquestionnaire was mailed, followed by one telephoneand two mailed reminders. It included questions abouttype of water supply, type of sewage disposal, use of tapwater for cooking and drinking, filtration of tap water,and for appropriate subsets, questions about well depthand testing of water quality. Agreement between thesurvey and administrative data on water system type(municipal, community, or private) and sewage systemtype (municipal or private) was examined by calculatingunweighted kappa statistics.Data analysisAll statistical analyses were done using SAS 9.1 (SASInstitute, Cary, NC). The unit of analysis was person-day, as most variables were time-varying and cohortmembership was dynamic, with individuals movingwithin and in and out of the Township during the studyperiod.Initial examinations of the data included identificationof intestinal infectious disease events, plots of theiroccurrence over the study period to check for evidenceof epidemics, and examination of multiple event recordsper subject to ascertain the time periods betweenrecords.Two sets of analyses were conducted, one for physi-cian visits, the other for hospitalizations. Only the firstphysician visit and first hospitalization for each subjectwere included in analyses, to exclude the possibility thata subsequent event might be related to the first. Person-days after the first event were excluded. Disease eventsand person-days within the first 2 months after entry tothe cohort (i.e., January 1 1995 for most subjects or thedate of moving to the Township if after that date) werealso excluded, to allow an incubation period sufficientfor any associations to be plausibly related to exposuresin the Township.Person-time and crude disease incidence rates with95% confidence intervals (CI) were calculated for eachcategory of each independent variable. The statisticalestimation and inference reported here are based on fit-ting the following unconditional logistic regressionTeschke et al. BMC Public Health 2010, 10:767http://www.biomedcentral.com/1471-2458/10/767Page 3 of 13model, using the full dataset and the PROC LOGISTICprocedure.log ( )it P Xit j jitjM= +=∑ 01(1)wherelog ( ) logit PPPititit=−1, and Pit = the probability ofa physician visit (or hospitalization) for an intestinalinfectious disease for subject i on day t, and Xjit =value for subject i on day t for each of M variables.(Primary models for a priori hypotheses included thevariables sex, age, season, duration of residence,household income quintile, drinking water disinfec-tion, water system, sewage disposal, and land use;post-hoc analyses added distance to nearest hospitaland population density to the primary models).Log odds ratios and their standard errors were esti-mated using maximum likelihood methods, and 95%confidence limits were based on the asymptotic normal-ity of parameter estimators. Calendar year was alsoincluded in all models, because including only firstevents meant that they were weighted somewhat moreheavily to the initial years. This variable was categorizedas 1995-6, 1997-8, 1999-2001, and 2002-3, groupingyears with nearly identical crude event rates.For certain environmental variables apropos only tosubsets of the data, secondary analyses were conducted.These analyses were restricted to physician visits only,because there were so few hospitalizations, and includedall variables in the primary analyses. In one model, theanalysis was restricted to those living at propertieswhose well depth was known, and well depth wasFigure 1 Maps showing geographic distributions of the water and sewage systems. Data for parcels in a mixed rural-urban community inthe Metro Vancouver area of Canada, on July 1 1999, the mid-point of the study period (of a total of 29,458 parcels). Uncolored areas areparcels not included in the study, either because they were outside the Township or did not house a study subject on that date (e.g., industrial,commercial or park land, or vacant property).Teschke et al. BMC Public Health 2010, 10:767http://www.biomedcentral.com/1471-2458/10/767Page 4 of 13Table 2 Proportions of person-time and crude incidence rates of intestinal infectious diseases, 1995 to 2003 inclusive,stratified by categories of each socio-demographic, temporal, and environmental variable.Variable & Category Proportion of person-time in each categoryCrude rate of physician visitsaper 100,000 person-years[95% CI]Crude rate of hospitalizationsbper 100,000 person-years[95% CI]SexFemale 50.4% 1,398 [1351, 1351] 35.8 [28.6, 43.0]Male 49.6% 1,310 [1265, 1265] 31.8 [24.9, 38.5]Age< 1 0.3% 7,848 [6484, 6484] 119 [0, 284]1 - 4 5.0% 6,607 [6286, 6286] 315 [ 249, 381]5 - 9 7.4% 2,164 [2014, 2014] 55.8 [33.0, 78.5]10 - 19 16.2% 902 [838, 838] 21.9 [12.0, 31.6]20 - 29 11.5% 1,548 [1448, 1448] 11.3 [3.0, 19.9]30 - 39 15.2% 1,190 [1114, 1114] 6.2 [0.8, 11.6]40 - 49 17.5% 770 [713, 713] 8.8 [2.7, 14.6]50 - 59 12.2% 653 [591, 591] 7.7 [1.0, 14.7]60 - 69 7.2% 756 [667, 667] 24.1 [8.3, 39.8]≥ 70 7.5% 1,095 [991, 991] 38.3 [19.0, 57.8]SeasonSpring 25.5% 1,566 [1496, 1496] 56.2 [43.5, 68.7]Summer 25.6% 1,288 [1229, 1229] 21.2 [13.5, 29.0]Fall 25.2% 1,128 [1070, 1070] 19.3 [11.9, 26.7]Winter 23.7% 1,438 [1369, 1369] 38.7 [27.9, 49.7]Duration of Residence in the Township< 1 year 15.0% 2,325 [2219, 2219] 65.7 [47.8, 83.5]1 - 2 years 17.2% 1,745 [1657, 1657] 46.0 [32.2, 60.0]2 - 3 years 15.3% 1,226 [1150, 1150] 33.2 [20.7, 45.7]3 - 6 years 30.8% 923 [876, 876] 21.9 [14.6, 28.9]≥ 6 years 15.5% 712 [652, 652] 13.1 [5.4, 21.0]Unknown 6.2% 1,938 [1786, 1786] 37.2 [16.1, 58.0]Household income quintile of the neighborhoodLow 4.7% 1,482 [1329, 1329] 63.5 [32.5, 94.8]Medium Low 10.1% 1,559 [1452, 1452] 40.9 [23.7, 57.7]Medium 24.1% 1,256 [1194, 1194] 33.6 [23.6, 43.7]Medium High 43.1% 1,391[1340, 1340] 29.6 [22.5, 36.6]High 18.0% 1,252 [1178, 1178] 32.1 [20.9, 43.6]Drinking Water DisinfectionChlorination 38.7% 1,343 [1293, 1293] 27.7 [20.7, 35.1]None 61.3% 1,358 [1319, 1319] 37.6 [30.9, 44.2]Water System or SubsystemMunicipal A*: mixed surface (66-96%) &well27.2% 1,905 [1832, 1832] 28.8 [20.0, 37.3]Municipal B*: mixed surface (66-96%) &well8.6% 1,741 [1619, 1619] 29.9 [14.3, 45.8]Municipal C*: mixed surface (12-63%) &well16.2% 949 [882, 882] 23.4 [13.1, 33.6]Municipal D*: well, with added surfacewater in emergencies & summer8.8% 1,252 [1150, 1150] 43.1 [24.1, 61.7]Municipal E*: well, with added surfacewater in emergencies & summer1.5% 829 [627, 627] 50.7 [1.0, 100]Municipal F: well, large systems 14.9% 1,340 [1257, 1257] 54.4 [38.1, 70.6]Municipal G: well, small systems 0.4% 412 [142, 142] 0 [0, 216]Community well 1.7% 949 [747, 747] 11.3 [0, 33.5]Private well 20.7% 949 [887, 887] 32.9 [22.1, 43.5]Teschke et al. BMC Public Health 2010, 10:767http://www.biomedcentral.com/1471-2458/10/767Page 5 of 13included as an additional variable. In another set ofmodels, analyses were restricted to those living at prop-erties receiving municipal water from surface waterreservoirs in the North Shore Mountains, and precipita-tion was included as an additional variable. Precipitationwas calculated as a continuous variable then categorizedinto 6 categories for modelling (0, 1 to < 10, 10 to < 25,25 to < 100, 100 to < 250, and ≥ 250 mm). It was calcu-lated in several ways (accumulated millimetres of rainover one- and two-week periods, with no lag and 2-, 5-and 10-day lags), each offered in a separate model.A number of alternative analysis methods, includingCox proportional hazards and Poisson regression mod-els (classical cohort analysis methods), as well asmixed effects logistic regression and logistic regressionwith generalized estimating equations (to account forrepeated measures on subjects), were considered in theinitial stages of data analysis. Due to computationalintensity, most of the alternative models were fit on a10% (1% for the mixed effects model) subsample of theoriginal data. The resulting risk estimates and inferen-tial results remained relatively unchanged, with theexception of some categories in the Poisson regression(likely due to rare events and small person-day countsfor these categories in the data subsamples). Of parti-cular note is that random intercept logistic regression,which allows for over-dispersed and correlated out-comes arising from repeated measures and clustereddesign, was fit on a 1% subsample of the data usingPROC GLIMMIX. The estimated variance of the ran-dom effects was near 0, which indicates that the mixedeffects logistic regression degenerated to regular logis-tic regression with no random effects. We were unableto run the mixed effects model on the whole data oron a 10% subsample, due to insufficient memory. TheGEE logistic regression, which accounts for within-sub-ject correlation in repeated measurement designs, wasfit using PROC GENMOD for the whole physicianvisit dataset, assuming exchangeable within-subjectcorrelation. The estimated effects (the point and inter-val estimates of the beta coefficients) were nearly iden-tical to those under a comparable unconditionallogistic regression model fit using PROC LOGISTIC;Table 2 Proportions of person-time and crude incidence rates of intestinal infectious diseases, 1995 to 2003 inclusive,stratified by categories of each socio-demographic, temporal, and environmental variable. (Continued)Sewage DisposalMunicipal sewer user 52.8% 1,701 [1653, 1653] 39.4 [32.3, 46.9]Private sewer user 47.3% 971 [930, 930] 27.0 [20.7, 33.6]Land UseAgricultural 8.3% 1,059 [962, 962] 32.1 [15.2, 48.7]Residential 91.7% 1,380 [1347, 1347] 33.9 [28.8, 39.1]a Number of physician visits = 7,017 in 189,234,765 person-days of observation.b Number of hospitalizations = 180 in 194,552,383 person-days of observation.* = Systems receiving both well water from municipal wells in the Township and surface water from North Shore Mountain reservoirs. Areas A, B, and C receivemixed water throughout the year; the estimated proportions of surface water listed above are based on 2009 total, peak and minimum flows at centralmonitoring stations. Areas D and E receive surface water only in emergencies (e.g., system maintenance work) and in peak summer months when well waterlevels are low. Township personnel indicated that the water distribution network design suggests additional differences between areas that cannot be quantified,such that area A is believed to receive more surface water than area B, and area D is believed to receive more surface water than area E. The proportions ofsurface water in all five mixed systems vary with rainfall, system water use, and system maintenance.Figure 2 Crude intestinal infectious disease incidence rates vs.age. Data from 1995 to 2003 inclusive: a) N = 7,017 physician visitsin 189,234,765 person-days of observation (top). b) N = 180hospitalizations in 194,552,383 person-days (bottom).Teschke et al. BMC Public Health 2010, 10:767http://www.biomedcentral.com/1471-2458/10/767Page 6 of 13the estimated exchangeable working correlation wasclose to 0.Note that in this cohort analysis, the follow-up oneach individual ended at first observation of a diseaseevent or over the available follow-up years for no eventever observed; this may be the reason for observingsmall and negligible within-subject correlation in themixed effects and GEE analyses. The computing timefor GEE using PROC GENMOD on the whole physicianvisit data was about 15 days, while the computing timeusing PROC LOGISTIC was about 2 days. Therefore,the final models were developed using PROC LOGIS-TIC, which was computationally most efficient amongall methods considered.ResultsThe average population of the Township during thestudy period was 90,273. The study cohort included126,499 individuals who resided in the Township for atleast 6 months between 1995 and 2003. The analysisdataset for physician visits included 7,017 cases with anintestinal infectious disease and 189,234,765 person-days.The analysis dataset for hospitalizations included 180cases and 194,552,383 person-days. Person-days differedbecause the number of days removed after subjects’ firstevents differed. Almost all events had general diagnosticcodes not specific to a single organism (Table 1). Plots ofdisease events by day and week over the study periodshowed no temporal patterns indicative of outbreaks.The Township database included 29,458 unique par-cels of land (Figure 1). Of the parcels, 30.9% wererecorded as private water users only, 66.3% wererecorded as municipal water users only, though 69.1%were connected to the municipal water supply at somepoint during the study period. At the beginning of thestudy period, no parcels received chlorinated waterwhile over two-thirds (20,387) did by the end of thestudy, including municipal and community water sup-plies with surface and well water sources. About half theparcels (49.4%) had private septic systems only, 48.9%were connected to the municipal sewer system only, and50.6% were connected to the municipal sewer systemsometime during the study period. There was a slightdecrease in the number of parcels with an agriculturaluse code from the beginning of the study (8.1%) to theend (7.0%).Table 2 lists the proportion of person-time for eachcategory of each variable. These differ from the propor-tions of parcels outlined above, because the numbers ofpeople living at each parcel were not equally distributedby category.Crude incidence rates of intestinal infectious diseaseswere 1,353 physician visits per 100,000 person-years(95% CI: 1,322-1,385/100,000) and 33.8 hospitalizationsper 100,000 person-years (95% CI: 28.8-38.7/100,000).Table 2 lists the crude incidence rates for each categoryof each variable. These rates exclude potentially inde-pendent subsequent disease episodes after an initial epi-sode. Among 7,017 subjects with a physician visit for anintestinal infectious disease, 23% (1,592/7,017) had mul-tiple billing records for such diseases. Most (1,410/1,592) had the same ICD-9 codes for the initial andrepeat visits, and more than half (835) had consecutiverecords less than 3 months apart making it very likelythe records were for the same episode. Only 6.5% ofsubjects with a physician visit (454/7,017) had consecu-tive billing records more than one year apart. Among180 subjects hospitalized for an intestinal infectious dis-ease, 3.3% (6/180) had multiple hospital dischargerecords for such diseases, all with the same ICD-9 codesfor the initial and repeat visits. Five of the 6 subjectswith repeat records had second records less than amonth after the initial hospitalization, and one had asecond hospitalization more than a year later.Variability in crude incidence rates was greatest for age(Table 2 Figure 2). The pattern of higher rates at theextremes of age was similar for physician visits and hos-pitalizations The rates were consistently high amongthose under 5, whereas those in the oldest ages had morevariable rates due to small numbers, especially amongthose over 90 years old for physician visits and over 70years for hospitalizations (Figure 2). Physician visit rateswere also higher among young adults (ages ~ 20-35), butthis pattern was not observed for hospitalizations.Patterns of incidence rates were similar for both phy-sician visits and hospitalizations for many of the otherindependent variables (Table 2). Seasonal rates werehighest in the spring and lowest in the fall. Those whohad resided in the Township for fewer years had higherrates. Those living in neighborhoods with the two lowesthousehold income quintiles had higher rates. Rates werelower with chlorinated water, with private sewage dispo-sal systems, and agricultural land uses. There was noclear pattern related to water system, except that thosewith wells connected to a small number of propertiesand run either by a private community or by the muni-cipality had the lowest rates.Table 3 summarizes the results of logistic regressionsexamining associations between intestinal infectious dis-eases and the independent variables representing socio-demographic, temporal, and environmental factors. Allvariables, except land use, were significant in the analy-sis of physician visits, and supported the patternsobserved for the crude incidence rates. The only vari-ables with significant relationships to hospitalizationswere age and season. Though not statistically significant,the patterns of the odds ratio estimates for most vari-ables (except socioeconomic status, water system, andTeschke et al. BMC Public Health 2010, 10:767http://www.biomedcentral.com/1471-2458/10/767Page 7 of 13Table 3 Odds ratiosa and 95% confidence intervals for associations between physician visits or hospitalizations forintestinal infectious diseases and socio-demographic, temporal, and environmental variables.Physician Visits HospitalizationsWald 95%ConfidenceLimitsWald 95%ConfidenceLimitsVariable & Category OddsRatioLower Upper OddsRatioLower UpperSexFemale 1.08 1.03 1.13 1.18 0.88 1.58Male 1.00 - - 1.00 - -Age< 1 5.22 4.26 6.40 2.75 0.60 12.61 - 4 4.83 4.32 5.39 6.54 3.64 11.85 - 9 1.81 1.61 2.04 1.33 0.69 2.5810 - 19 0.80 0.71 0.90 0.54 0.27 1.0720 - 29 1.31 1.17 1.47 0.26 0.11 0.6430 - 39 0.96 0.86 1.08 0.14 0.05 0.4040 - 49 0.69 0.61 0.78 0.22 0.09 0.5150 - 59 0.63 0.55 0.72 0.20 0.07 0.5660 - 69 0.69 0.59 0.80 0.62 0.27 1.42≥ 70 1.00 - - 1.00 - -SeasonSpring 1.08 1.01 1.15 1.45 1.01 2.07Summer 0.89 0.83 0.95 0.55 0.35 0.87Fall 0.77 0.72 0.83 0.50 0.31 0.81Winter 1.00 - - 1.00 - -Duration of Residence in the Township< 1 year 1.16 1.03 1.30 1.04 0.46 2.351 - 2 years 0.98 0.88 1.09 0.89 0.40 1.952 - 3 years 0.80 0.70 0.91 0.73 0.30 1.783 - 6 years 0.73 0.64 0.82 0.62 0.26 1.49≥ 6 years 0.69 0.60 0.80 0.55 0.20 1.57Unknown 1.00 - - 1.00 - -Household income quintile of the neighborhoodLow 1.18 1.03 1.34 1.17 0.58 2.37Medium Low 1.19 1.07 1.32 0.91 0.49 1.69Medium 1.06 0.98 1.15 0.85 0.51 1.42Medium High 1.02 0.95 1.09 0.89 0.57 1.39High 1.00 - - 1.00 - -Drinking Water DisinfectionChlorination 0.92 0.85 1.00 0.89 0.53 1.48None 1.00 - - 1.00 - -Water System or SubsystemMunicipal A: mixed surface (66-96%) & well 1.57 1.39 1.78 0.54 0.24 1.20Municipal B: mixed surface (66-96%) & well 1.45 1.28 1.64 0.66 0.29 1.52Municipal C: mixed surface (12-63%) & well 1.01 0.90 1.12 0.71 0.36 1.40Municipal D: well, with surface water added in emergencies & summer 1.00 0.87 1.15 0.83 0.37 1.87Municipal E: well, with surface water added in emergencies & summer 0.89 0.69 1.15 1.35 0.45 4.03Municipal F: well, large systems 0.96 0.85 1.08 0.93 0.45 1.94Municipal G: well, small systems 0.51 0.27 0.99 0 0 > 999Community well 1.04 0.83 1.30 0.34 0.05 2.49Private well 1.00 - - 1.00 - -Sewage DisposalTeschke et al. BMC Public Health 2010, 10:767http://www.biomedcentral.com/1471-2458/10/767Page 8 of 13land use) were the same for hospitalizations and physi-cian visits.Well depth was known for 3,815 wells. The mean depthwas 36.6 m, with a recorded range of 0 to 280 m. No rela-tionship between well depth and physician visits forintestinal infectious diseases was observed in crude inci-dence rates or in logistic regression modelling restrictedto those living at properties whose well depth wasknown. Crude incidence rates (and proportion of totalperson-time) in three well depth categories were: 1,049/100,000 person-years (2.0% of person-time) for welldepths up to 9 m; 1,089 (6.2%) for depths of greater than9 to 30 m; and 1,032 (8.2%) for depths greater than 30 m.The association between precipitation and physicianvisits for an intestinal infectious disease was examinedfor those residents receiving municipal water from theNorth Shore mountain surface water reservoirs (62.1% ofperson-time). No relationship was observed for one-weekrainfall accumulations. For two-week accumulations,crude rates suggested a pattern of increasing illness withincreasing rainfall, particularly with a 10-day lag (Figure3a). However, in logistic regressions adjusting for allother variables, the trends did not hold (Figure 3b).The response rate for the survey of Township house-holds was 59% (N = 546 respondents of 926 eligible; N =74 were ineligible, including businesses, vacant addresses,mail returned to sender, etc.). Of those responding, 62%received municipal water, 5% community water, and 33%private well water. Almost all households used tap waterfor cooking (92%, 95% CI: 90-94%), most also used it fordrinking (79%, CI: 76-82%), and a large proportion filteredtheir tap water (45%, CI: 41-49%). These proportions didnot vary by water source. Most respondents with privatewells tested their water for coliform bacteria (71%, CI: 67-75%) and for nitrates (68%, CI: 64-72%). Almost all (88%,CI: 85-91%) knew their well depth, with reported depthsranging from 3 to 152 m and a mean of 39 m, very similarto the mean for wells in the Ministry of Environment data-base. About 35% (CI: 31-39%) of respondents with privatewells reported problems with their water. Most wererelated to minerals and sediment (reported for 17% of allwells) or the well running dry (5%), but 12% reported bac-teriological contamination at some point. Contaminatedwells were treated, about half with bleach and the otherhalf by installing ultraviolet light disinfection systems. Ofthose with private sewage systems (49%; CI: 45-53%), 97%reported that the system was a septic tank and drain field.The few other types included engineered systems with bio-filters or small treatment plants.DiscussionDisease ratesThe crude incidence rates of intestinal infectious dis-eases found in this study (1,353 physician visits and 33.8Table 3 Odds ratiosa and 95% confidence intervals for associations between physician visits or hospitalizations forintestinal infectious diseases and socio-demographic, temporal, and environmental variables. (Continued)Municipal sewer 1.26 1.14 1.38 1.42 0.77 2.62Private sewage system 1.00 - - 1.00 - -Land UseAgricultural 1.06 0.95 1.19 0.95 0.51 1.80Residential 1.00 - - 1.00 - -a Based on unconditional logistic regression, also adjusted for calendar year. Statistically significant odds ratios in bold.- Reference value, no confidence limits calculated.Figure 3 Physician visits for intestinal infectious diseases vs.precipitation accumulated over prior 2 weeks. Data from 1995to 2003 inclusive; restricted to subpopulation (N = 4,868 physicianvisits and 117,543,309 person-days of observation) that receivedunfiltered surface water from North Shore Mountain reservoirs. a)Crude incidence rates per 100,000 person-years (top). b) Odds ratiosadjusted for sex, age, calendar year, season, duration of residence inthe Township, neighborhood household income quintile, drinkingwater disinfection, water system, sewage disposal, and land use(bottom)Teschke et al. BMC Public Health 2010, 10:767http://www.biomedcentral.com/1471-2458/10/767Page 9 of 13hospitalizations per 100,000 person-years) are difficult tocompare to those reported elsewhere, because of differ-ent case definitions and identification methods. Themost similar data are for enteric disease hospitalizationsin the US from 1998 to 2006; Christensen et al. [20]found a rate of 56.6/100,000, somewhat higher than inthis study, likely because they included more ICD-9codes (001-009, 022.2). Tinker et al. [13] reported emer-gency department visits for gastrointestinal illnesses inAtlanta and found a rate of about 800/100,000. In asummary of several of his Canadian and US studies,Payment [21] estimated a self-reported gastrointestinalillness (nausea, vomiting, diarrhea) rate of 70,000/100,000. In a modeling exercise to estimate the rate ofacute gastrointestinal illness specifically due to drinkingwater in the US, Messner et al. [22] estimated a rate of6,000/100,000.Environmental factorsThe main focus of this study was whether environmen-tal variables related to water quality were associatedwith intestinal infectious diseases. Chlorination wasassociated with lower physician visit and hospitalizationrates (latter not statistically significant). Even the smallreduction in risk suggested by these data may have pub-lic health importance, since this is one of the dominantmethods of water treatment. Although chlorination hasbeen shown to be effective in preventing enteric diseaseoutbreaks [23], the few reports investigating endemicdisease have been less clear. Hellard et al. [11] examineda children’s hospital emergency visits (numerator dataonly) before and after chlorination of the Melbourne,Australia water supply, but found no change. Odoi et al.[15] found no difference in giardiasis rates betweenareas in Ontario, Canada with and without chlorination.This study was able to examine numerous types ofwater systems, and to examine sub-classifications ofthose systems. We expected that those with private wellwater would have higher enteric disease rates than thosewith municipal and community well water, as found byothers [16,24-26], though not all [17]. Beyond the effectof chlorination, this was not the case. Two small muni-cipal well water supplies each serving fewer than 100homes had physician visit rates that were about halfthose of the private wells, but other municipal and com-munity well water sources had rates similar to privatewells. Well depth was not associated with differences inendemic disease rates, lending support to lack of influ-ence of private well water on disease rates, though it isimportant to note that the subset of wells with knowndepth was likely to have been correctly constructed toprevent contamination, since well depths were volunta-rily reported by professional well drillers. Evidence fromthe survey suggests that most Township residents withprivate wells were well informed about their water sys-tems and conscientious about testing them and correct-ing problems.We also expected that those served by municipal sur-face water supplies would have higher disease rates thanthose served exclusively by municipal well water sys-tems, as found elsewhere [15,17,24,26,27]. There wassome evidence of this in the physician visit but not thehospitalization data. Of the 5 systems that had mixedsurface and well water, the two with the highest propor-tions of surface water had the highest physician visitrates, about 50% higher than those served by most othersystems types, including those with smaller proportionsof surface water. We further investigated the systemsreceiving surface water, to examine the impact of recentprecipitation. While crude rates suggested increasedphysician visits with increased precipitation, this trendwas not robust to adjustment for other variables. Anumber of investigators have examined both precipita-tion and the turbidity it can induce. Curriero et al. [28]found that more than half of surface water-related out-breaks were preceded by extreme precipitation in themonth prior. Aramini et al. [10] studied Vancouver,which receives all its drinking water from the same pro-tected North Shore mountain watersheds that partiallyserve areas of the Township. They found associationsbetween increased turbidity and endemic gastroenteritis,as measured by physician visits and hospitalizations.Others have observed similar relationships [8,14,29]. Inall studies, the associations were small, but significant.These studies did not include the range of variablesavailable about the Township and its residents, so it isunknown whether controlling for other factors wouldhave diminished the associations, as occurred in ouranalyses.Although the relationship between sewage contamina-tion and enteric disease outbreaks is well established[30], we identified only two other studies that examinedendemic disease risk by sewage system type. Denno et al.[16] found higher odds ratios for Salmonella infection forthose with private septic systems. Febriani et al. [17]found no difference in gastrointestinal illness prevalencebetween those with private and municipal sewage dispo-sal. We expected private septic systems might havedrain-field-to-well contamination that could result inhigher disease rates. Instead higher disease rates wereobserved among those with municipal sewer connectionsfor both physician visits and hospitalizations (the latternot statistically significant). This result was apparent inthe initial crude analyses and remained after adjustmentfor all other factors in the models, including the neigh-borhood-level adjustment allowed by the water systemvariable. To check whether municipal sewer might be asurrogate for other factors not considered a priori, weTeschke et al. BMC Public Health 2010, 10:767http://www.biomedcentral.com/1471-2458/10/767Page 10 of 13conducted post hoc analyses including population densityand distance to nearest hospital. Neither variable wasassociated with disease events and the higher odds ratiosfor municipal sewer remained stable. We also examineddisease rates for the interaction of sewage and water sys-tem types. There was no difference in rates between pri-vate and municipal sewage systems for those with privatewell water, but those with both municipal sewer andmunicipal water systems, whose pipes are likely to runalong similar paths, had higher disease rates. Althoughthe higher disease rates for municipal sewer users may bea chance finding or may be related to other factors notaccounted for in this study, it is prudent to consider pos-sible explanations. The Township does not have com-bined storm and sanitary sewer lines, so overflow is notan issue. Water lines are required to be separated fromsewer lines by 0.5 m vertically (sewer lines below) and 3m horizontally, and where these conditions cannot bemet, the water main is to be protected from infiltrationby wrapping the pipe joints with petrolatum tape.Although there is no comparable data from the Town-ship, some studies have shown possible modes of con-tamination despite the care taken when pipes are laid. AUS study described sources of sewer line leakage andconcluded exfiltration can occur where the groundwatertable is below the sewer lines (as is the case in the Town-ship) [31]. Others have found that soil and water samplesnext to water lines frequently have fecal contamination[32], and that breaks and maintenance work on waterlines [12,23] and low water pressure [33] are related togastrointestinal illness.Sociodemographic and temporal factorsSociodemographic and temporal factors were also asso-ciated with infectious intestinal disease rates in this study.Females had higher rates than males in our study, asfound elsewhere [20]. The age distributions of physicianvisits and hospitalizations largely followed a u-shaped pat-tern of high rates in the very young and elderly, and forphysician visits only, of increased rates in young adults aswell (an association often thought to be related to travel orparenting young children). Christensen et al. [20] reportedthe age distribution of hospitalizations for all infections(not just enteric), and the u-shaped pattern was similar tothat observed here. The demographic differences in dis-ease rates are likely, in part, to reflect differences in healthcare utilization across these characteristics.Hospitalizations and physician visits for enteric dis-eases were highest from December through May andlowest from June through November, the same patternobserved in a Quebec study of gastrointestinal illnessprevalence [17]. In a study of physician billings forenteric diseases in the neighboring Canadian province ofAlberta, the highest rates were observed from Novemberthrough March, and the lowest in September [34].Christensen et al. [20] examined seasonality for allinfectious diseases in the US and found the highest inci-dence from December through March inclusive and thelowest incidence from June to September. Slight offsetsin seasonality may in part reflect differences in climatesbetween the jurisdictions studied.Lower household income (as an ecological variable, foreach census dissemination area) was related to increasedphysician visits for infectious intestinal diseases, not sur-prising since income has been similarly associated withmany health indicators in Canada, including life expec-tancy [35]. Studies specific to enteric disease have alsofound increased rates for those with lower income [15].A study by Tam et al. [36] raises the question of theextent to which these results reflect differences in dis-ease rates or differences in likelihood of attending aphysician by socio-economic status.Finally, duration of residence was consistently relatedto disease rates. Those who lived in the Townshiplonger had lower rates of both physician visits and hos-pitalizations (the latter not statistically significant). Feb-riani et al. [17] identified a similar pattern. Frost et al.[37] showed that episodes of gastrointestinal illnesseswere 35 to 60% lower among subjects with antigens toCryptosporidium parvum, suggesting that an explanationof this finding may be increasing immunity to localpathogens over time. It may also be possible that indivi-duals are likely to seek medical advice for enteric ill-nesses when they first move to an area, but may be lesslikely to do once they are established in the community.Strengths and limitationsThis study had the following advantages: a large cohortwith almost a decade of follow-up; a cohort designwhich allowed calculation of disease incidence rates;objective data on dependent and independent variablesof interest; and survey data about water-related beha-viors in the study area.There were also limitations. The administrative data ondisease events were provided by physicians, not labora-tories, therefore clinical suspicion was used to assign gen-eral rather than specific disease coding, preventinganalyses of individual diseases such as giardiasis or cryp-tosporidiosis. Because it is difficult to specify a between-event period to define second incident events within anindividual, we included only first events. The incidencerates for physician visits and hospitalizations are thereforeunderestimates, though the distribution of subsequentevents suggests that this is a small problem, representingan underestimate of 10% or less. Intestinal infectious dis-eases do not always result in contacts with the health caresystem, so the incidence rates calculated here omit lesssevere events and events among people less inclined toTeschke et al. BMC Public Health 2010, 10:767http://www.biomedcentral.com/1471-2458/10/767Page 11 of 13visit a physician [36]. If the propensity to visit a physicianis related to the main environmental variables of interest,the effect estimates could be biased. We did post-hoc ana-lyses to adjust for factors thought to be related to the pro-pensity for physician visits and hospitalizations (distanceto the nearest hospital, population density) and these didnot change the results for the environmental variables. Infact, the physician visit results for all independent vari-ables, except land use and precipitation, were stable fromthe crude rates to the adjusted analyses.Many of the independent variables, including environ-mental variables, household income quintile, and dura-tion of residence in the Township, were dependent onthe temporal and spatial accuracy of residential addressrecords held by the Client Registry of the Ministry ofHealth Services. Since address records are updated whenthe health care system is used, the timeliness of addresschanges may be related to overall health status. Wetested the reliability of selected administrative data viacomparisons to data collected in the survey. Despite the3-year offset between these two data sources and poten-tial errors in the survey data, kappas for agreementbeyond chance were 0.76 for water system type and 0.78for sewage system type, indicating very good reliability.Finally, details on certain features were not available, forexample well and septic system ages, and water qualityat the tap of individual residences.ConclusionsThe study is one of few that have examined risk of phy-sician visits and hospitalizations for endemic infectiousintestinal diseases across an array of water and sewersystem types. Chlorination of water supplies was shownto be associated with lower risks. Surface water wasassociated with higher risks in two of five systems. Pri-vate well water was not associated with increased risk,likely because of awareness of water quality and quantityissues by Township residents. Those with municipalsewer systems appeared to have increased disease risk,though this result needs to be viewed with caution sinceit was not anticipated and has not been observed else-where. Further studies to examine disease incidenceamong populations with municipal versus private sewagesystems are warranted. Most socio-demographic vari-ables had predicted associations, with higher physicianvisit and hospitalization rates in females, in the veryyoung and very old, and in those in low income areas.Increased duration of residence in the Township wasassociated with reduced risk, perhaps due to increasingimmunity to local pathogens over time.For public health and civil engineering personnel,these results reinforce confidence in chlorination as ameans of reducing enteric disease risk, and indicate thatprivate well water users who understand their systemsand know how to respond to water quality issues identi-fied during monitoring can minimize disease risks. Theresults also convey cautions: they support previousresearch indicating that surface water sources, even withprotected watersheds, deserve special attention; and pre-sent new data suggesting municipal sewer systems mayrequire more scrutiny than previously thought.The results of this study, which did not always followprior expectations, underscore the importance of study-ing factors associated with endemic disease across waterand sewage system types.List of AbbreviationsBC: Canadian province of British Columbia; CI: confidence interval; ICD-9:International Classification of Diseases, 9th edition; OR: odds ratio.AcknowledgementsWe greatly appreciate the work of Antigone Dixon-Warren, Meaghan NortonDaniel, Dean Scovill and the many other personnel of the Township ofLangley and Fraser Health Authority who helped with introducing theproject to the region, assembling the data sets, and interpreting the results.We thank Negar Elmieh, AJ Gunson, Jennifer Hinton, Elizabeth Matovinovic,Karen McCaig, John Rowse, Carolina da Silva, and Sylvia Struck for their workin the initial stages of this project, John Spinelli and Nhu Le for analyticaladvice, Suhail Marino and Saleema Dhalla for their work on the opinionsurvey, and members of the Program on Water Governance for their helpfulcomments about the results. The study was funded in part by the CanadianInstitutes of Health Research, the Canadian Water Network, and the BridgeProgram at the University of British Columbia.Author details1School of Population and Public Health, University of British Columbia,Vancouver, Canada. 2Department of Civil Engineering, University of BritishColumbia, Vancouver, Canada. 3Institute for Resources, Environment andSustainability, University of British Columbia, Vancouver, Canada.4Department of Pathology and Laboratory Medicine, University of BritishColumbia and BC Centre for Disease Control, Vancouver, Canada.Authors’ contributionsKT led the design, conduct, and analysis of the study and drafted the paper.NB led the environmental data gathering, conducted the ArcGIS linkage,liaised with the Ministry of Health Services for the linkage of theenvironmental data to the cohort data, and participated in interpretation ofresults and manuscript review. H Shen conducted the data analyses andparticipated in interpretation of results and various stages of the manuscriptreviews and revisions. JA had the initial idea for the study, and participatedin the design, analyses, interpretation, and manuscript revisions. RCmanaged the linked dataset, screened the cohort for eligibility, and usedMinistry billings data to define cases. MK led the application to PopulationData BC, participated in data analyses, interpretation, and manuscript review.YCM participated in the study design, analyses, interpretation and in variousstages of the manuscript reviews and revisions. H Schreier, and JLI-Rparticipated in the study design, analyses, interpretation and manuscriptreview. All authors have read and approved the final manuscript.Competing interestsThe authors declare that they have no competing interests.Received: 24 February 2010 Accepted: 16 December 2010Published: 16 December 2010References1. Yoder J, Roberts V, Craun GF, Hill V, Hicks LA, Alexander NT, Radke V,Calderon RL, Hlavsa MC, Beach MJ, Roy SL: Surveillance for waterborne diseaseand outbreaks associated with drinking water and water not intended fordrinking - United States, 2005-2006. MMWR Surveill Summ 2008, 57:39-62.Teschke et al. BMC Public Health 2010, 10:767http://www.biomedcentral.com/1471-2458/10/767Page 12 of 132. MacKenzie WR, Hoxie NJ, Proctor ME, Gradus MS, Blair KA, Peterson DE,Kazmierczak JJ, Addiss DG, Fox KR, Rose JB, Davis JP: A massive outbreakin Milwaukee of Cryptosporidium infection transmitted through thepublic water supply. New England J Med 1994, 331:161-167.3. Krewski D, Balbus J, Butler-Jones D, Haas C, Isaac-Renton J, Roberts KJ,Sinclair M: Managing health risks from drinking water - a report to theWalkerton inquiry. J Toxicol Environ Health A 2002, 65:1635-823.4. O’Connor D: Report of the Walkerton Commission of Inquiry, Parts 1 and2. Ontario Ministry of the Attorney General; 2002 [http://www.attorneygeneral.jus.gov.on.ca/english/about/pubs/walkerton/], (AccessedFeb. 8, 2010).5. Livernois J: The Economic Costs of the Walkerton Water Crisis The WalkertonInquiry Commissioned Paper 14. The Walkerton Inquiry: Toronto; 2001[http://www.uoguelph.ca/~live/Livernois_14%20Final%20Report.pdf],(Accessed Feb. 8, 2010).6. BC Centre for Disease Control: 2008 British Columbia Annual Summary ofReportable Diseases BC CDC: Vancouver; 2009 [http://www.bccdc.ca/NR/rdonlyres/59BFCFBB-933D-4337-9305-E3E5FF30D272/0/EPI_Report_CDAnnual2008_20091202.pdf], (Accessed Feb. 8, 2010).7. Public Health Agency of Canada: Notifiable Disease Incidence by Year,1989 to 2004. 2009 [http://dsol-smed.phac-aspc.gc.ca/dsol-smed/ndis/c_time-eng.php], (Accessed Feb. 8, 2010).8. Schwartz J, Levin R, Hodge K: Drinking water turbidity and pediatrichospital use for gastrointestinal illness in Philadelphia. Epidemiol 1997,8:615-620.9. Morris RD, Naumova EN, Griffiths JK: Did Milwaukee experiencewaterborne cryptosporidiosis before the large documented outbreak in1993? Epidemiol 1998, 9:264-270.10. Aramini J, McLean M, Wilson J, Holt J, Copes R, Allen B, Sears W: Drinkingwater quality and health-care utilization for gastrointestinal illness ingreater Vancouver. Can Commun Dis Rep 2000, 26:211-4.11. Hellard ME, Sinclair MI, Dharmage SC, Bailey MJ, Fairley CK: The rate ofgastroenteritis in a large city before and after chlorination. Int J EnvironHealth Res 2002, 12:355-360.12. Nygard K, Wahl E, Krogh T, Tveit OA, Bohleng E, Tverdal A, Aavitsland P:Breaks and maintenance work in the water distribution systems andgastrointestinal illness: a cohort study. Int J Epidemiol 2007, 36:873-880.13. Tinker SC, Moe CL, Klein M, Flanders WD, Uber J, Amirtharajah A, Singer P,Tolbert PE: Drinking water residence time in distribution networks andemergency department visits for gastrointestinal illness in Metro Atlanta,Georgia. J Water Health 2009, 7:332-43.14. Tinker SC, Moe CL, Klein M, Flanders WD, Uber J, Amirtharajah A, Singer P,Tolbert PE: Drinking water turbidity and emergency department visits forgastrointestinal illness in Atlanta, 1993-2004. J Expo Sci Environ Epidemiol2010, 20:19-28.15. Odoi A, Martin SW, Michel P, Holt J, Middleton D, Wilson J: Determinantsof the geographical distribution of endemic giardiasis in Ontario,Canada: a spatial modeling approach. Epidemiol Infect 2004, 132:967-76.16. Denno DM, Keene WE, Hutter CM, Koepsell JK, Patnode M, Flodin-Hursh D,Stewart LK, Duchin JS, Rasmussen L, Jnes R, Tarr PI: Tri-countycomprehensive assessment of risk factors for sporadic reportablebacterial enteric infection in children. J Infect Dis 2009, 199:467-76.17. Febriani Y, Levallois P, Gingras S, Gosselin P, Majowicz SE, Fleury MD: Theassociation between farming activities, precipitation, and the risk ofacute gastrointestinal illness in rural municipalities of Quebec, Canada: across-sectional study. BMC Public Health 2010, 10:48.18. Chamberlayne R, Green B, Barer ML, Hertzman C, Lawrence WJ, Sheps SB:Creating a population-based linked health database: A new resource forhealth services research. Can J Public Health 1998, 89:270-273.19. World Health Organization: International Classification of Diseases, 9thRevision World Health Organization: Geneva, Switzerland;1979-1998.20. Christensen KLY, Holman RC, Steiner CA, Sejvar JJ, Stoll BJ, Schonberger LB:Infectious disease hospitalizations in the United States. Clin Infect Dis2009, 49:1025-1035.21. Payment P: Epidemiology of endemic gastrointestinal and respiratorydiseases: incidence, fraction attributable to tap water and costs tosociety. Water Sci Tech 1997, 35:7-10.22. Messner M, Shaw S, Regli S, Rotert K, Blank V, Soller J: An approach todeveloping a national estimate of waterborne disease due to drinkingwater and a national estimate model application. J Water Health 2006,4(Suppl2):201-240.23. Tulchinsky TH, Burla E, Clayman M, Sadik C, Brown A, Goldberger S: Safetyof community drinking-water and outbreaks of waterborne entericdisease: Israel, 1976-97. Bull World Health Org 2000, 78:1466-73.24. Carrique-Ma J, Andersson Y, hjertqvist M, Svensson A, Torner A, Giesecke J:Risk factors for domestic sporadic campylobacteriosis among youngchildren in Sweden. Scand J Inf Dis 2005, 37:101-10.25. Roy SL, DeLong SM, Stenzel SA, Shiferaw B, Roberts JM, Khalakdina A,Marcus R, Segler SD, Shah DD, Thomas S, Vugia DJ, Zansky SM, Dietz V,Beach MJ: Emerging Infections Program FoodNet Working Group. Riskfactors for sporadic Cryptosporidiosis among immunocompetentpersons in the United States from 1999 to 2001. J Clin Microbiol 2004,42:2944-51.26. Dennis DT, Smith RP, Welch JJ, Chute CG, Anderson B, Herndon JL, vonReyn CF: Endemic Giardiasis in New Hampshire: A case-control study ofenvironmental risks. J Inf Dis 1993, 167:1391-5.27. Adak GK, Cowden JM, Nicholas S, Evans HS: The Public Health LaboratoryService national case-control study of primary indigenous sporadic casesof campylobacter infection. Epidemiol Infect 1995, 115:15-22.28. Curriero FC, Patz JA, Rose JB, Lele S: The association between extremeprecipitation and waterborne disease outbreaks in the United States,1948-1994. Am J Public Health 2001, 91:1194-9.29. Schwartz J, Levin R, Goldstein R: Drinking water turbidity andgastrointestinal illness in the elderly of Philadelphia. J EpidemiolCommunity Health 2000, 54:45-51.30. Schuster CJ, Ellis AG, Robertson WJ, Charron DF, Aramini JJ, Marshall BJ,Medeiros DT: Infectious disease outbreaks related to drinking water inCanada, 1974-2001. Can J Public Health 2005, 96:254-8.31. Selvakumar A, Field R, Burgess EH, Amick RS: Exfiltration in sanitary sewersystems in the US. Urban Water Journal 2004, 1:227-234.32. Karim MR, Abbaszadegan M, LeChevallier M: Potential for pathogenintrusion during pressure gradients. J Am Water Works Assoc 2003,95:134-46.33. Hunter PR, Chalmers RM, Hughes S, Syed Q: Self-reported diarrhea in acontrol group: A strong association with reporting of low-pressureevents in tap water. Clin Infect Dis 2005, 40:32-4.34. Yiannakoulias N, Svenson LW: Differences between notifiable andadministrative health information in the spatial-temporal surveillance ofenteric infections. Int J Med Inform 2009, 78:417-424.35. McIntosh CN, Fines P, Wilkins R, Wolfson MC: Income disparities in health-adjusted life expectancy for Canadian adults, 1991 to 2001. HealthReports 2009, 20:1-10.36. Tam CC, Rodrigues LC, O’Brien SJ: The study of infectious intestinaldisease in England: what risk factors for presentation to general practicetell us about potential for selection bias in case-control studies ofreported cases of diarrhoea. Int J Epidemiol 2003, 32:99-105.37. Frost FJ, Roberts M, Kunde TR, Craun G, Tollestrup K, Harter L, Muller T: Howclean must our drinking water be: The importance of protectiveimmunity. J Infect Dis 2005, 191:809-14.Pre-publication historyThe pre-publication history for this paper can be accessed here:http://www.biomedcentral.com/1471-2458/10/767/prepubdoi:10.1186/1471-2458-10-767Cite this article as: Teschke et al.: Water and sewage systems, socio-demographics, and duration of residence associated with endemicintestinal infectious diseases: A cohort study. BMC Public Health 201010:767.Teschke et al. BMC Public Health 2010, 10:767http://www.biomedcentral.com/1471-2458/10/767Page 13 of 13

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.52383.1-0223829/manifest

Comment

Related Items