Open Collections

UBC Faculty Research and Publications

Determinants of hospitalization for a cutaneous injection-related infection among injection drug users:… Lloyd-Smith, Elisa; Wood, Evan; Zhang, Ruth; Tyndall, Mark W; Sheps, Sam; Montaner, Julio S; Kerr, Thomas Jun 9, 2010

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

Item Metadata


52383-12889_2009_Article_2238.pdf [ 522kB ]
JSON: 52383-1.0223858.json
JSON-LD: 52383-1.0223858-ld.json
RDF/XML (Pretty): 52383-1.0223858-rdf.xml
RDF/JSON: 52383-1.0223858-rdf.json
Turtle: 52383-1.0223858-turtle.txt
N-Triples: 52383-1.0223858-rdf-ntriples.txt
Original Record: 52383-1.0223858-source.json
Full Text

Full Text

Lloyd-Smith et al. BMC Public Health 2010, 10:327 AccessR E S E A R C H  A R T I C L EResearch articleDeterminants of hospitalization for a cutaneous injection-related infection among injection drug users: a cohort studyElisa Lloyd-Smith1,2, Evan Wood1,2,3, Ruth Zhang1,2, Mark W Tyndall1,2, Sam Sheps2, Julio SG Montaner1,2,3 and Thomas Kerr*1,2,3AstractBackground: Cutaneous injection-related infections (CIRI) are a primary reason individuals who inject drugs (IDU) are hospitalized. The objective of this study was to investigate determinants of hospitalization for a CIRI or related infectious complication among a cohort of supervised injection facility (SIF) users.Methods: From 1 January 1 2004 until 31 January 2008, using Cox proportional hazard regression, we examined determinants of hospitalization for a CIRI or related infectious complication (based on ICD 10 codes) among 1083 IDU recruited from within the SIF. Length of stay in hospital and cost estimates, based on a fully-allocated costing model, was also evaluated.Results: Among hospital admissions, 49% were due to a CIRI or related infectious complication. The incidence density for hospitalization for a CIRI or related infectious complication was 6.07 per 100 person-years (95% confidence intervals [CI]: 4.96 - 7.36). In the adjusted Cox proportional hazard model, being HIV positive (adjusted hazard ratio [AHR] = 1.79 [95% CI: 1.17 - 2.76]) and being referred to the hospital by a nurse at the SIF (AHR = 5.49 [95% CI: 3.48 - 8.67]) were associated with increased hospitalization. Length of stay in hospital was significantly shorter among participants referred to the hospital by a nurse at the SIF when compared to those who were not referred (4 days [interquartile range {IQR}: 2-7] versus 12 days [IQR: 5-33]) even after adjustment for confounders (p = 0.001).Conclusions: A strong predictor of hospitalization for a CIRI or related infectious complication was being referred to the hospital by a nurse from the SIF. This finding indicates that nurses not only facilitate hospital utilization but may provide early intervention that prevents lengthy and expensive hospital visits for a CIRI or related infectious complication.BackgroundCutaneous injection-related infections (CIRI), whichinclude cellulitis and abscesses, are among the primarycauses of hospitalization among individuals who injectdrugs (IDU) [1-4]. Hospitalizations due to CIRI carryconsiderable economic burden [3,5,6]. Complications ofCIRI that are more likely to require hospitalizationinclude, but are not limited to: osteomyelitis [7], bactere-mia and sepsis [8,9], endocarditis [10,11], septic arthritis[7,12], ulcer [9], thrombophlebitis [13,14] and myositis[9].A recent report by Hope et al. (2008) suggested thathealthcare associated costs for CIRI among IDU in Eng-land were substantial, ranging from £15.5 million to £30.0million per annum [5] ($27.1 to $52.4 million Canadiandollars). In 2001, Palepu et al. reported that of the IDUseen at St. Paul's Hospital, an urban hospital in Vancou-ver, Canada, 35% had been hospitalized and a third ofthese hospitalizations were due to a CIRI or related infec-tious complication [1]. Hospitalization was expensivewith hospital utilization cost per day reported to be $610Canadian dollars (C$610) (95% confidence interval {CI}:* Correspondence: uhri-tk@cfenet.ubc.ca1© 2010 Lloyd-Smith et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Com-mons Attribution License (, which permits unrestricted use, distribution, and reproduc-tion in any medium, provided the original work is properly cited.C$C576- C$645) [1]. British Columbia Centre for Excellence in HIV/AIDS, Providence Health Care, 608-1081 Burrard Street, Vancouver, V6Z 1Y6, CanadaFull list of author information is available at the end of the articleLloyd-Smith et al. BMC Public Health 2010, 10:327 2 of 7North America's first supervised injection facility (SIF)opened in Vancouver's Downtown Eastside (DTES) in2003. Within the SIF, IDU can inject pre-obtained drugsunder the supervision of nurses. Individuals visiting theSIF are provided with sterile injecting equipment andemergency intervention in the event of an overdose, aswell as primary medical care and addiction treatment,either on site or through referral. While several studieshave pointed to the positive impact of the SIF on publicdisorder [15], HIV risk behaviour [16], management ofoverdose [17,18] and use of addiction treatment [19], itsrole with regard to hospitalization for CIRI remainsunknown. However, recent research highlights that CIRIare a medical issue of concern among users of the SIF.Over a two year period, 6-10 per cent of IDU had a CIRIat time of interview [20]. In the present analysis, usinglongitudinal data, we examined predictors and cost ofhospitalization for a CIRI or related infectious complica-tion among IDU using the SIF.MethodsDesign and participantsThe SIF in Vancouver is being evaluated through the Sci-entific Evaluation of Supervised Injection (SEOSI)cohort, which has been described in detail [21]. Briefly,the cohort was assembled through random recruitmentof IDU from within the SIF. Random recruitment is basedon inviting users of the SIF to be referred to the researchstudy during random blocks of time. Among individualswho were recruited, a venous blood sample was drawnand an interviewer-administered questionnaire was con-ducted at baseline and at semi-annual follow-up visits.The informed consent agreement, obtained for all partic-ipants, included a request to link the SIF evaluation withadministrative health databases. In Vancouver, hospitalsare equipped with a database that tracks patient admis-sion. The SIF is also equipped with a similar database. Atthe SIF, nursing care includes wound care, dressingchanges, and measuring temperature. However, there areno physicians at the SIF. If microbiological investigationor medical treatment, including antibiotic prescription orintravenous antibiotic therapy is required, individuals atthe SIF must be referred to the hospital or a medicalclinic. In this study, a linkage of SEOSI participant data,SIF data and St. Paul's Hospital inpatient data was per-formed. St. Paul's Hospital is the major urban hospitalserving the DTES community, one of Canada's poorestpostal codes. The University of British Columbia-Provi-dence Health Care Research Ethics Board approved thepresent study.CIRI or related infectious complication. The infectiouscomplications included were based on previous literature[8,9]. The definition of the reason for hospitalization wasbased on International Classification of Diseases (ICD)10 codes on patients' hospital records and included:abscess (G061, G062, L020, L021, L022, L024, J851), cel-lulitis (L0300, L0310, L0311, L032, L0335, L038), osteo-myelitis (M4620, M4625, M4629, M8617, M8618, M8661,M8663, M8666, M8681, M8691, M8695), staphylococcalinfection 161 {(A490, A499, B956) including, septicaemia(A410, A412, A419) and Methicillin-Resistant Staphylo-coccus aureus (MRSA), (U000)}, endocarditis (I330), sep-tic arthritis (M0000, M0002, M0004, M0005, M0006,M0008, M0009), ulcer (L089, L979), thrombophlebitis(I802, I808) and myositis (M6005, M6008). A few ICD-10codes chosen suggest but do not require infectious etiolo-gies (e.g., thrombophlebitis, myositis, and ulcer). Thenumber of events refers to a CIRI or related infectiouscomplication as a primary, secondary, tertiary, quater-nary, or quinary diagnoses according to relevant ICD 10codes unless otherwise specified.We first examined the distribution and frequency of aCIRI or related infectious complication. Then, we evalu-ated length of stay in hospital among study participantsand examined this outcome as a continuous variable in alinear model that adjusted for the following confoundingvariables: age, sex, HIV serostatus and SIF nurse referral.We then considered the cost of hospitalization, associ-ated with CIRI, which was estimated at C$712 per hospi-tal day, based on a fully-allocated costing model for theprovince of British Columbia from 2001 [1]. This esti-mate was updated to Canadian dollars in 2005 [22]. Fully-allocated costing includes costs associated with nursingcare, medications, investigations, physician visits andlength of stay as well as overhead, opportunity cost ofhospital resources and a 5% depreciation of capital equip-ment [1]. Potential healthcare savings were estimated bymultiplying the cost per day value (C$712) by the differ-ence in number of days hospitalized among individualswith CIRI who were referred by a nurse within the SIFand those who were self referred to the hospital.We investigated baseline characteristics stratified byhospitalization or not bivariately. Using Cox proportionalhazard regression, we examined factors potentially asso-ciated with hospitalization. Variables considered for ouranalyses included: age; sex at birth (female vs. male); cur-rent residence in DTES (yes vs. no); living in unstablehousing (yes vs. no); daily cocaine injection (yes vs. no);daily heroin injection (yes vs. no); daily speedball injec-tion (yes vs. no); and HIV serostatus (positive vs. nega-tive). As used previously, unstable housing was defined asMeasurementsThe start point for these analyses was enrollment into theSEOSI cohort and the endpoint was hospitalization for aliving in a single room occupancy (SRO) hotel, shelter,recovery or transition house, jail, on the street, or havingno fixed address [23]. Variables from the semi-annualLloyd-Smith et al. BMC Public Health 2010, 10:327 3 of 7questionnaire referred to behaviour that occurred in thelast six months unless otherwise specified. We also exam-ined whether a SIF nurse referral to the hospital was asso-ciated with hospitalization and, if so, whether length ofstay was different given referral versus self referral. Forthis task, we conducted a record linkage matching theirSEOSI identifying code to each participant's record in theSIF database to determine nurse referral. Then, we linkedhis or her SEOSI identifying code with his or her uniquepersonal health number to examine hospital recordsprior to the censor or event date.Variable selection was based on previously publishedliterature on a CIRI or related infectious complicationand hospitalization among IDU [1,5,8,9,20]. Variablesconsidered associated with hospitalization were analyzedin unadjusted analyses and adjusted Cox proportionalhazard regression model. Time zero was defined as thedate of recruitment into the SEOSI study for all partici-pants and participants not hospitalized at St. Paul's Hos-pital were censored as of 31 January 2008. All behavioralvariables were treated as time-updated covariates basedon semi-annual follow-up data. The multivariate modelwas fit using a fixed model whereby we included all vari-ables that were statistically significant at the p < 0.05 levelin univariate analyses. All statistical analyses were per-formed using SAS 8.0 (Cary, NC) and all p-values weretwo-sided.ResultsDuring the study period (1 January 2004 to 31 January2008), 1083 individuals were recruited into the SEOSIcohort and 901 (83%) reported at least one follow-upvisit. The median age among SEOSI participants was 38.4years (interquartile range {IQR}: 32.7-44.3) and 314 (29%)were female. The median follow-up duration afterrecruitment into the cohort was 21.4 months (IQR: 13.1-24.6). During the study period, 99 (9%) participants wereadmitted to St. Paul's Hospital, yielding an incidence den-sity of hospitalization for a CIRI or related infectiouscomplication of 6.07 per 100 person-years (95% CI: 4.96 -7.36). Among all hospital admissions among SEOSI par-ticipants, 49% (216 of 442) were related to a CIRI orrelated infectious complication. A total of 216 hospital-ization events occurred among 99 individuals for a CIRIor related infectious complication. Among the 99 hospi-talized SEOSI participants, 47 received CIRI care at theSIF during the study period. A total of 145 SEOSI partici-pants were referred by a nurse at the SIF while 63 partici-pants were referred and visited the ED. In addition, 115SEOSI participants were referred but subsequently nothospitalized. Importantly, 60% (28/47) of those whoreferred by a nurse were more likely to be hospitalizedwithin three days (p = 0.011). Further, those who werereferred to the hospital by a nurse at the SIF were signifi-cantly more likely to have been seen by a nurse at the SIFfor a CIRI (8 nurse visits for a CIRI [IQR: 3-16]) com-pared to those that were not referred to a hospital (1nurse visit for a CIRI [IQR: 0-4]) (p < 0.001). Virtually allpatients are admitted to the hospital from the EmergencyDepartment (ED).Tables 1, 2 and 3 display the frequency of a CIRI orrelated infectious complication. Cellulitis was the mostcommon reason for hospitalization. Fifteen persons hadmissing data on HIV serostatus and were excluded fromanalyses. Therefore, all results on HIV were based on asample size of 1068. Participants who had been referredto the hospital by an SIF nurse had a significantly shorterlength of stay in hospital as compared to those who werenot referred (4 days [IQR: 2-7], 12 days [IQR: 5-33]). Theeight day reduction in the length of hospital stay remainedsignificant (p = 0.001) after adjustment for confoundingvariables. Considering the total potential healthcare sav-ing based on a fully allocated hospital cost per day calcu-lation, each referral from the SIF would have resulted in asaving of C$5,696 (IQR: C$2, 136 - C$18, 512).The factors associated with an increased risk of hospi-talization after recruitment into the SEOSI cohort areshown in Table 4. In the multivariate model being HIVpositive (adjusted hazard ratio [AHR] = 1.79 [95% CI:1.16 - 2.75]) and being referred to the hospital by a nurseat the SIF (AHR = 5.38 [95% CI: 3.39 - 8.55]) were posi-tively and independently associated with an increasedlikelihood of hospitalization.Table 1: Description of hospitalizations for a cutaneous injection-related infection or related infectious complication among Scientific Evaluation of Supervised Injection participants.Classification† First event Total eventsn (%) n (%)n = 99 n = 216Cellulitis 33 (33) 59 (27)Abscess 14 (14) 26 (12)Osteomyelitis 10 (10) 39 (18)Staphylococcal infection 22 (22) 42 (17)Endocarditis 9 (9) 24 (12)Septic arthritis 7 (7) 12 (6)Ulcer 1 (1) 6 (3)Thrombophlebitis 1 (1) 4 (2)received CIRI care at the SIF were referred to the hospitalby a nurse at the SIF and subsequently sought treatmentat the hospital. Individuals with an ED visit who wereMyositis 2 (2) 4 (2)† Classification is based on International Classification of Diseases 10 codesLloyd-Smith et al. BMC Public Health 2010, 10:327 4 of 7DiscussionIn the present study, being referred by a nurse at a SIFwas independently associated with an elevated rate ofhospitalization. Importantly, participants who had beenreferred to the hospital by a nurse at the SIF had a signifi-cantly shorter length of stay in hospital despite adjust-ment for HIV infection and other potential confounders.This finding indicates that nurses facilitate hospital utili-Table 2: Distribution of hospitalizations for a cutaneous injection-related infection or related infectious complication*.Primary diagnosis +1 +2 +3 +4 +5 +6 +7 +8 +9 +1&2 +1&41cellulitis 45 - 3 0 5 0 0 2 1 0 - -2abscess 9 1 - 0 5 0 0 0 0 0 - 33osteomyelitis 14 0 5 - 5 0 0 0 1 2 0 14staph.infection 18 3 0 1 - 4 0 0 0 0 1 -5endocarditis 12 0 1 0 5 - 0 0 0 0 0 16septic arthritis 8 0 2 0 1 0 - 0 0 0 0 17ulcer 3 0 0 0 1 0 0 - 0 0 0 08thrombophlebitis 3 1 0 0 0 0 0 0 - 0 0 09myositis 1 0 0 0 3 0 0 0 0 - 0 0+1&7 +2&4 +3&4 +3&7 +4&7 +5&4 +9&4 + more than two diseases1cellulitis - 0 0 1 2 0 0 02abscess 0 - 2 0 0 1 1 43osteomyelitis 1 3 - - 2 0 0 54staph.infection 0 - - 0 - - - 155endocarditis 0 0 0 0 0 - 0 56septic arthritis 0 0 0 0 0 0 0 07ulcer - 0 0 - - 0 0 28thrombophlebitis 0 0 0 0 0 0 0 09myositis 0 0 0 0 0 0 - 0*According to ICD 10 codes. ICD 10 codes categorized into 9 diseases and listed from 1 to 9. First column refers to when disease listed as primary diagnosis. Other columns refer to multiple diseases reported, irrespective of whether event was primary, secondary, tertiary, quaternary, or quinary.Table 3: Frequency of hospitalizations for a cutaneous injection-related infection or related infectious complication*.Once Twice Three times Four times Five plus times Total1cellulitis 19 7 3 0 2 592abscess 10 5 2 0 0 263osteomyelitis 5 2 2 0 3 394staph.infection 10 2 0 3 2 425endocarditis 11 0 1 0 1 246septic arthritis 6 3 0 0 0 127ulcer 1 0 1 0 1 68thrombophlebitis 1 0 1 0 0 49myositis 3 0 1 0 0 4*According to ICD 10 codes. Diseases listed from 1 to 9. ICD 10 codes categorized into 9 diseases and listed from 1 to 9. Data relates to number of times an individual developed a disease, irrespective of whether event was coded as primary, secondary, tertiary, quaternary, or quinary.Lloyd-Smith et al. BMC Public Health 2010, 10:327 5 of 7zation as well as providing early intervention that pre-vents lengthy and expensive hospital visits for a CIRI orrelated infectious complication. According to the costsavings calculation based on the length of stay reduction,SIF nurse referral resulted in a minimum saving ofC$2,136 per admission. However, it is important to notethat variability in treatment costs exists in the CIRI andrelated infectious complications included in this investi-gation.Using the same cohort, the risk factors for the outcomeof developing a CIRI have been reported to include beingfemale, living in unstable housing, borrowing syringes,requiring help injecting and injecting cocaine daily [20].Some factors differ from risk factors in other cities. Forexample, research from San Francisco reports an associ-ated between developing a CIRI and frequent black tarheroin injection [2] whereas in Vancouver, frequentcocaine injection has been associated with developing aCIRI [20]. The present study evaluated the predictors ofthe outcome of hospitalization for a CIRI or related infec-tious complication. The predictors of hospitalization for aCIRI or related infectious complication included beingHIV positive and referred to the hospital by a nurse at theSIF. Future research investigating outcomes, includingmortality from a CIRI or related infectious complicationamong those referred compared to those not referred tothe hospital by a nurse at the supervised injection facilitywould support our understanding of the potential impactof nurse referral. Interestingly, being HIV positive wasnot a risk factor for the development of CIRI but was apredictor of hospitalization for a CIRI or related infec-tious complication.As noted above, an increased risk of hospitalization fora CIRI or related infectious complication was observedamong HIV positive participants. This finding is consis-tent with previous research on hospitalization amongIDU in this setting [1]. Individuals with HIV may be morelikely to be treated as an inpatient as opposed to an out-patient (i.e., HIV positive patients with abscesses may bemore likely to get admitted to the hospital than HIV neg-ative patients with abscesses). A potentially elevated sus-ceptibility to bacterial infections [8,24] as well as knownhigh-risk drug injection practices in this subpopulation[25,26] may related to why HIV positive participants inthis study had elevated rates of hospitalization. Furtherelucidation of this finding with a larger sample of HIVpositive IDU is necessary to examine this importantresearch question.After controlling for factors that are known to be asso-ciated with hospitalization [1,27], referral from an SIFnurse remained a strong predictor of hospitalization.medical care among participants who have infections orinjuries of greater severity [28]. Hospital care for moresevere forms of CIRI is essential since local hospitals pro-vide treatment to CIRI not available at the SIF (e.g., inci-sion and drainage of an abscess, intravenous antibiotictherapy as well as diagnostic tools and therapy appropri-ate for complications). Further, given that the averagelength of stay in hospital for those referred to the hospitalby a nurse at the SIF was significantly shorter, it may bethat nurses at the SIF are helping reduce the incidence oflate presentation of complications, such as osteomyelitis,that are often lengthy and are particularly costly to thehealthcare system [7]. Although our study design doesnot allow us to infer causation, it is noteworthy that thisassociation persisted in multivariate analyses. This mayreflect prompt referral as well as the preventative effect ofthe SIF on serious infections [16].There are several examples of efficient and cost effec-tive community-based treatment services that mayinform changes in the provision of care that is required inorder to reduce the incidence of hospitalization of a CIRIor related infectious complication in our setting. In itsfirst year of operation, the Integrated Soft Tissue Infec-tion Services (ISIS) Clinic in San Francisco, California,resulted in a 47% decrease in surgical service admissionsand an estimated savings of over $8.0 (C$8.4) million forcosts related to CIRI [29]. In addition, a wound manage-ment clinic operated in conjunction with a syringeexchange program in Oakland, California [30]. It wasfound that the average cost per individual treated at thiswound management clinic was $5.0 (C$5.2), substantiallylower than equivalent hospital costs of $185.0 (C$193.4)and $360.0 (C$376.3) [30]. Given the success other citieshave had in implementing additional medical treatmentfor CIRI in the community, we recommend initiating pro-vision of on-site incision and drainage for abscesses andadministration of antibiotic therapy at the SIF or in anearby community setting. Further, to prevent develop-ment of CIRI, it may be of value to screen and treat forMRSA at the SIF or at the SEOSI research site. In our set-ting, expanding the capacity to provide primary care in anintegrated manner is warranted.There are limitations of the present study to considerwhen interpreting this data. Firstly, St. Paul's Hospital wasthe only facility linkage that was conducted. However,this is the primary hospital serving the SIF catchmentarea and is accessed extensively by the IDU population inthe DTES [1,27,31]. Secondly, our study relies on selfreport to obtain drug use and other behavioural variables.However, self report among IDU is considered valid [32]and hospital utilization and nurse referral were accessedFindings from this study confirm existing qualitativeresearch indicating that nurses at the SIF provide CIRI-related care and play a key role in enhancing access todirectly from databases within the hospital and at the SIF,respectively. Thirdly, a limitation of this analysis is that alow sample size of HIV positive participants precludedLloyd-Smith et al. BMC Public Health 2010, 10:327 6 of 7our ability to conduct further analyses on this association.Additional research is required with a larger sample ofHIV positive individuals to better elucidate this finding.Fourthly, additional research is required to better under-stand why a shorter length of stay is seen among SEOSIparticipants referred by a nurse at the SIF to the hospital.It is possible that nurses may be referring to the hospitalSEOSI participants that require shorter lengths of stay(e.g., abscess versus osteomyelitis). However, in our study,individuals with an ED visit who were referred by a nursewere significantly more likely to be hospitalized withinthree days. Similarly, Binswanger et al. found ED use forCIRI to be associated with hospitalization and also death[33].ConclusionIn summary, we found high levels of hospitalization for aCIRI or related infectious complication among local IDU.Being HIV positive and being referred to the hospital by anurse at the SIF, as opposed to self referral to the hospital,were both independently and positively associated withan increased likelihood of hospitalization. Participantswho had been referred to the hospital by a nurse at thering individuals who require hospitalization for a CIRI orrelated infectious complication to the hospital, whichmay result in shorter and less expensive hospital visits.Expanded management of CIRI in the community mayreduce the need for referral to the hospital, further reduc-ing the cost of caring for this common clinical problemamong IDU.Competing interestsThe authors declare that they have no competing interests.Authors' contributionsELS conceived and designed the study and drafted the manuscript. RZ and ELSperformed the statistical analyses. ELS, TK, EW, MT contributed to the designand coordination of the study. All authors provided assistance with interpreta-tion of the results to the drafts of the manuscript and read and approved thefinal manuscript.AcknowledgementsThe authors wish to thank the participants in SEOSI and the staff at Insite, the Portland Hotel Society, and Vancouver Coastal Health (Chris Buchner, David Marsh and Heather Hay.) We also thank all current and past SEOSI staff. We would also like to thank Deborah Graham, Leslie Rae, Caitlin Johnston, Steven Kain and Calvin Lai for their research assistance. Viviane Dias Lima and Karissa Johnston contributed statistical expertise. The evaluation of the SIF was origi-nally made possible through a financial contribution from Health Canada, although the views expressed here do not reflect the official policies of Health Canada. The evaluation is currently supported by the Canadian Institutes of Table 4: Univariate and multivariate Cox proportional hazard analyses of hospitalizations for a cutaneous injection-related infection or related infectious complication among among Scientific Evaluation of Supervised Injection participants.Unadjusted AdjustedHazard Ratio (HR) Hazard Ratio (AHR)Variable HR (95% CI) p-value AHR (95% CI) p-valueAge(per year older) 0.98 (0.96 - 1.01) 0.146Sex(Female vs Male) 1.59 (1.07 - 2.39) 0.024 1.36 (0.90 - 2.05) 0.139Unstable housing*(Yes vs No) 1.65 (1.08 - 2.53) 0.021 1.26 (0.79 - 2.02) 0.328Cocaine injection*(Daily vs Not daily) 1.75 (1.17 - 2.62) 0.006 1.46 (0.94 - 2.25) 0.090Speedball injection*(Daily vs Not daily) 1.90 (1.15 - 3.14) 0.012 1.19 (0.69 - 2.07) 0.528HIV serostatus(Positive vs Negative) 2.12 (1.39 - 3.24) <0.001 1.79 (1.16 - 2.75) 0.008Hospital referral†(Yes vs No) 2.41 (1.55 - 3.77) <0.001 5.38 (3.39 - 8.55) <0.001*Behaviour refers to activities in the last 6 months. †Indicates data derived from SIF database and by a study nurse. CI = confidence interval.SIF had a significantly shorter length of stay in hospital ascompared to those who were not. These findings indicatethat nurses at the SIF play a critical role in terms of refer-Health Research (grants HPR-85526 and RAA-79918) and Vancouver Coastal Health. TK, MT and ELS are supported by the Michael Smith Foundation for Health Research; TK and ELS are supported by the Canadian Institutes of Health Research.Lloyd-Smith et al. BMC Public Health 2010, 10:327 7 of 7Author Details1British Columbia Centre for Excellence in HIV/AIDS, Providence Health Care, 608-1081 Burrard Street, Vancouver, V6Z 1Y6, Canada, 2School of Public and Population Health, University of British Columbia, 2206 East Mall, Vancouver, V6T 1Z3, Canada and 3Department of Medicine, University of British Columbia, 10203-2275 Laurel Street, Vancouver, V5Z 1M9References1. Palepu A, Tyndall MW, Leon H, Muller J, O'Shaughnessy MV, Schechter MT, et al.: Hospital utilization and costs in a cohort of injection drug users.  CMAJ 2001, 165:415-420.2. Murphy E, DeVita D, Liu H, Vittinghoff E, Leung P, Ciccarone DH, et al.: Risk factors for skin and soft-tissue abscesses among injection drug users: a case-control study.  Clin Infect Dis 2001, 33:35-40.3. Ciccarone D, Bamberger JD, Kral AH, Edlin BR, Hobart CJ, Moon A, et al.: Soft tissue infections among injection drug users - San Francisco, California, 1996-2000.  MMWR Morb Mortal Wkly Rep 2001, 50:381-384.4. Takahashi TA, Baernstein A, Binswanger I, Bradley K, Merrill JO: Predictors of hospitalization for injection drug users seeking care for soft tissue infections.  J Gen Intern Med 2007, 22:382-388.5. Hope V, Kimber J, Vickerman P, Hickman M, Ncube F: Frequency, factors and costs associated with injection site infections: findings from a national multi-site survey of injecting drug users in England.  BMC Infect Dis 2008, 8:.6. DiNubile MJ, Lipsky BA: Complicated infections of the skin and skin structures: when the infection is more than skin deep.  J Antimicrob Chemother 2004, 53:ii37-ii50.7. Kak V, Chandrasekar PH: Bone and joint infections in injection drug users.  Infect Dis Clin North Am 2002, 16:681-695.8. Gordon RJ, Lowy FD: Bacterial infections in drug users.  N Engl J Med 2005, 353:1945-1954.9. del Giudice P: Cutaneous complications of intravenous drug abuse.  Br J Dermatol 2004, 150:1-10.10. DiNubile MJ: Short-course antibiotic therapy for right-sided endocarditis caused by Staphylococcus aureus in injection drug users.  Ann Intern Med 1994, 121:873-876.11. Ngaage DL, Cowen ME: Right ventricular needle embolus in an injecting drug user: the need for early removal.  Emerg Med J 2001, 18:500-501.12. Ebright JR, Pieper B: Skin and soft tissue infections in injection drug users.  Infect Dis Clin North Am 2002, 16:697-712.13. Schulz S, Beckenbach C, Philipp M, Hengstmann J: Color coded duplex sonography of inguinal vessels in i.v. drug addicts.  Vasa 2002, 31:7-13.14. Roszler MH, McCarroll KA, Donovan KR, Rashid T, Kling GA: The groin hit: complications of intravenous drug abuse.  Radiographics 1989, 9:487-508.15. Wood E, Kerr T, Small W, Palepu A, Tyndall MW: Changes in public order after the opening of a medically supervised safer injecting facility for illicit injection drug users.  CMAJ 2004, 169:759-763.16. Kerr T, Tyndall M, Li K, Montaner J, Wood E: Safer injection facility use and syringe sharing in injection drug users.  Lancet 2005, 366:316-318.17. Milloy MJS, Kerr T, Tyndall M, Montaner J, Wood E: Estimated drug overdose deaths averted by North America's first medically-supervised safer injection facility.  PLoS ONE 2008, 3:e3351.18. Milloy MJS, Kerr T, Mathias R, Zhang R, Montaner J, Tyndall M, Wood E: Non-fatal overdose among a cohort of active injection drug users recruited from a supervised injection facility.  Am J Drug Alcohol Abuse 2008, 34:499-509.19. Wood E, Tyndall MW, Zhang R, Stoltz JA, Montaner J, Kerr T: Attendance at supervised injection facilities and use of detoxification services.  N Engl J Med 2006, 354:2512-2514.20. Lloyd-Smith E, Wood E, Zhang R, Tyndall M, Montaner J, Kerr T: Risk factors for developing a cutaneous injection-related infection among injection drug users: a cohort study.  BMC Public Health 2008, 8:405.21. Wood E, Kerr T, Lloyd-Smith E, Buchner C, Marsh D, Montaner J, Tyndall 22. Organisation for Economic Co-operation and Development (OECD) Health data 2007.  2007.23. Corneil TA, Kuyper LM, Shoveller J, Hogg RS, Li K, Spittal P, Schechter M, Wood E: Unstable housing, associated risk behaviour, and increased risk for HIV infection among injection drug users.  Health Place 2006, 12:79-85.24. Brettle RP: Bacterial infections in HIV: the extent and nature of the problem.  Int J STD AIDS 1997, 5:5-15.25. Stein MD, Sobota M: Injection drug use: hospital care and charges.  Drug Alcohol Depend 2001, 64:117-120.26. Spijkerman IJ, van Ameijden EJ, Mientjes GH, Coutinho RA, van den Hoek A: Human immunodeficiency virus infection and other risk factors for skin abscesses and endocarditis among injection drug users.  J Clin Epidemiol 1996, 49:1149-1154.27. Palepu A, Strathdee SA, Hogg RS, Anis AH, Rae S, Cornelisse PG, Patrick DM, O'Shaughnessy MV, Schechter MT: The social determinants of emergency department and hospital use by injection drug users in Canada.  J Urban Health 1999, 76:409-418.28. Small W, Wood E, Lloyd-Smith E, Tyndall M, Kerr T: Accessing care for injection-related infections through a medically supervised injecting facility: a qualitative study.  Drug Alcohol Depend 2008, 98:159-162.29. Harris HW, Young DM: Care of injection drug users with soft tissue infections in San Francisco, California.  Arch Surg 2002, 137:1217-1222.30. Grau LE, Arevalo S, Catchpool C, Heimer R: Expanding harm reduction services through a wound and abscess clinic.  Am J Public Health 2002, 92:1915-1917.31. Kerr T, Wood E, Grafstein E, Ishida T, Shannon K, Lai C, Montaner JM, Tyndall MW: High rates of primary care and emergency department use among injection drug users in Vancouver.  J Public Health 2005, 27:62-66.32. Darke S: Self-report among injecting drug users: a review.  Drug Alcohol Depend 1998, 51:253-263.33. Binswanger IA, Takahashi TA, Bradley K, Dellit TH, Benton KL, Merrill JO: Drug users seeking emergency care for soft tissue infection at high risk for subsequent hospitalization and death.  J Stud Alcohol Drugs 2008, 69:924-932.Pre-publication historyThe pre-publication history for this paper can be accessed here: 10.1186/1471-2458-10-327Cite this article as: Lloyd-Smith et al., Determinants of hospitalization for a cutaneous injection-related infection among injection drug users: a cohort study BMC Public Health 2010, 10:327Received: 20 May 2009 Accepted: 9 June 2010 Published: 9 June 2010This article is available from:© 2010 Lloyd-Smith et al; licensee BioMed Central Ltd. is an Open Access rticl  distributed under th  erms of the Creative Commons Attribution License (, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.BMC Pub ic H alth 2010, 10:327MW: Methodology for evaluating Insite: Canada's first medically supervised safer injection facility for injection drug users.  Harm Reduct J 2004:1.


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



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"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items