Population structure, case clusters, and genetic lesions associated with Canadian Salmonella 4,[5],12:i:- isolates

Monophasic Salmonella 4,[5]:12:i:- are a major public health problem because they are one of the top five Salmonella serotypes isolated from clinical cases globally and because they can carry resistance to multiple antibiotics. A total of 811 Salmonella 4,[5]:12:i:- and S. Typhimurium whole genome sequences (WGS) were generated. The various genetic lesions causing the Salmonella 4,[5]:12:i:- genotype were identified and assessed with regards to their distribution in the population of 811 Salmonella 4,[5]:12:i:- and S. Typhimurium isolates, their geographical and temporal distribution, and their association with non-human sources. Several clades were identified in the population structure, and the largest two were associated almost exclusively with a short prophage insertion and insertion of a mobile element carrying loci encoding antibiotic and mercury resistance. IS26-mediated deletions and fljB point mutants appeared to spread clonally. 'Inconsistent' Salmonella 4,[5]:12:i:- isolates associated with specific, single amino acid changes in fljA and hin were found in a single clade composed of water, shellfish, and avian isolates. Inclusion of isolates from different case clusters identified previously by PFGE validated some of the clusters and invalidated others. Some wgMLST clusters of clinical isolates composed of very closely related isolates contained an isolate(s) with a different genetic lesion, suggesting continuing mobility of the implicated element responsible. Such cases may need to be left out of epidemiological investigations until sufficient numbers of isolates are included that statistical significance of association with sources is not impaired. Non-human sources were frequently found in or near clinical case clusters. Prospective surveillance and WGS of non-human sources and retrospective analysis by WGS of isolates from existing culture collections provides data critical for epidemiological investigations of food- and waterborne outbreaks.

Emergence of a Novel Salmonella enterica Serotype Reading Clonal Group Is Linked to Its Expansion in Commercial Turkey Production, Resulting in Unanticipated Human Illness in North America

Two separate human outbreaks of Salmonella enterica serotype Reading occurred between 2017 and 2019 in the United States and Canada, and both outbreaks were linked to the consumption of raw turkey products. In this study, a comprehensive genomic investigation was conducted to reconstruct the evolutionary history of S. Reading from turkeys and to determine the genomic context of outbreaks involving this infrequently isolated Salmonella serotype. A total of 988 isolates of U.S. origin were examined using whole-genome-based approaches, including current and historical isolates from humans, meat, and live food animals. Broadly, isolates clustered into three major clades, with one apparently highly adapted turkey clade. Within the turkey clade, isolates clustered into three subclades, including an "emergent" clade that contained only isolates dated 2016 or later, with many of the isolates from these outbreaks. Genomic differences were identified between emergent and other turkey subclades, suggesting that the apparent success of currently circulating subclades is, in part, attributable to plasmid acquisitions conferring antimicrobial resistance, gain of phage-like sequences with cargo virulence factors, and mutations in systems that may be involved in beta-glucuronidase activity and resistance towards colicins. U.S. and Canadian outbreak isolates were found interspersed throughout the emergent subclade and the other circulating subclade. The emergence of a novel S Reading turkey subclade, coinciding temporally with expansion in commercial turkey production and with U.S. and Canadian human outbreaks, indicates that emergent strains with higher potential for niche success were likely vertically transferred and rapidly disseminated from a common source.IMPORTANCE Increasingly, outbreak investigations involving foodborne pathogens are difficult due to the interconnectedness of food animal production and distribution, and homogeneous nature of industry integration, necessitating high-resolution genomic investigations to determine their basis. Fortunately, surveillance and whole-genome sequencing, combined with the public availability of these data, enable comprehensive queries to determine underlying causes of such outbreaks. Utilizing this pipeline, it was determined that a novel clone of Salmonella Reading has emerged that coincided with increased abundance in raw turkey products and two outbreaks of human illness in North America. The rapid dissemination of this highly adapted and conserved clone indicates that it was likely obtained from a common source and rapidly disseminated across turkey production. Key genomic changes may have contributed to its apparent continued success in commercial turkeys and ability to cause illness in humans.

Multistate Outbreak of Listeriosis Associated with Packaged Leafy Green Salads, United States and Canada, 2015-2016

We investigated an outbreak of listeriosis detected by whole-genome multilocus sequence typing and associated with packaged leafy green salads. Nineteen cases were identified in the United States during July 5, 2015–January 31, 2016; isolates from case-patients were closely related (median difference 3 alleles, range 0–16 alleles). Of 16 case-patients interviewed, all reported salad consumption. Of 9 case-patients who recalled brand information, all reported brands processed at a common US facility. The Public Health Agency of Canada simultaneously investigated 14 cases of listeriosis associated with this outbreak. Isolates from the processing facility, packaged leafy green salads, and 9 case-patients from Canada were closely related to US clinical isolates (median difference 3 alleles, range 0–16 alleles). This investigation led to a recall of packaged leafy green salads made at the processing facility. Additional research is needed to identify best practices and effective policies to reduce the likelihood of Listeria monocytogenes contamination of fresh produce.

Outbreak of Escherichia coli O157:H7 Infections Linked to Mechanically Tenderized Beef and the Largest Beef Recall in Canada, 2012

Contaminated beef is a known vehicle of Escherichia coli O157:H7 infection, although more attention is given to the control of E. coli O157:H7 in ground, rather than whole-cut, beef products. In September 2012, an investigation was initiated at an Alberta, Canada, beef plant after the detection of E. coli O157:H7 in two samples of trim cut from beef originating from this plant. Later in September 2012, Alberta Health Services identified five laboratory-confirmed infections of E. coli O157:H7, and case patients reported eating needle-tenderized beef steaks purchased at a store in Edmonton, Alberta, produced with beef from the Alberta plant. In total, 18 laboratory-confirmed illnesses in Canada in September and October 2012 were linked to beef from the Alberta plant, including the five individuals who ate needle-tenderized steaks purchased at the Edmonton store. A unique strain of E. coli O157:H7, defined by molecular subtyping and whole genome sequencing, was detected in clinical isolates, four samples of leftover beef from case patient homes, and eight samples of Alberta plant beef tested by industry and food safety partners. Investigators identified several deficiencies in the control of E. coli O157:H7 at the plant; in particular, the evaluation of, and response to, the detection of E. coli O157 in beef samples during routine testing were inadequate. To control the outbreak, 4,000 tons of beef products were recalled, making it the largest beef recall in Canadian history. This outbreak, in combination with similar outbreaks in the United States and research demonstrating that mechanical tenderization can transfer foodborne pathogens present on the surface into the interior of beef cuts, prompted amendments to Canada's Food and Drug Regulations requiring mechanically tenderized beef to be labeled as such and to provide safe cooking instructions to consumers. A detailed review of this event also led to recommendations and action to improve the safety of Canada's beef supply.

Shared genome analyses of notable listeriosis outbreaks, highlighting the critical importance of epidemiological evidence, input datasets and interpretation criteria

The persuasiveness of genomic evidence has pressured scientific agencies to supplement or replace well-established methodologies to inform public health and food safety decision-making. This study of 52 epidemiologically defined Listeria monocytogenes isolates, collected between 1981 and 2011, including nine outbreaks, was undertaken (1) to characterize their phylogenetic relationship at finished genome-level resolution, (2) to elucidate the underlying genetic diversity within an endemic subtype, CC8, and (3) to re-evaluate the genetic relationship and epidemiology of a CC8-delimited outbreak in Canada in 2008. Genomes representing Canadian Listeria outbreaks between 1981 and 2010 were closed and manually annotated. Single nucleotide variants (SNVs) and horizontally acquired traits were used to generate phylogenomic models. Phylogenomic relationships were congruent with classical subtyping and epidemiology, except for CC8 outbreaks, wherein the distribution of SNV and prophages revealed multiple co-evolving lineages. Chronophyletic reconstruction of CC8 evolution indicates that prophage-related genetic changes among CC8 strains manifest as PFGE subtype reversions, obscuring the relationship between CC8 isolates, and complicating the public health interpretation of subtyping data, even at maximum genome resolution. The size of the shared genome interrogated did not change the genetic relationship measured between highly related isolates near the tips of the phylogenetic tree, illustrating the robustness of these approaches for routine public health applications where the focus is recent ancestry. The possibility exists for temporally and epidemiologically distinct events to appear related even at maximum genome resolution, highlighting the continued importance of epidemiological evidence.

Evaluation of whole-genome sequencing for outbreak detection of Verotoxigenic Escherichia coli O157:H7 from the Canadian perspective

BACKGROUND

Rapid and accurate identification of Verotoxigenic Escherichia coli (VTEC) O157:H7 is dependent on well-established, standardized and highly discriminatory typing methods. Currently, conventional subtyping tests for foodborne bacterial pathogen surveillance are rapidly being replaced with whole-genome sequencing (WGS) in public health laboratories. The capacity of WGS to revolutionize global foodborne disease surveillance has positioned this tool to become the new gold standard; however, to ensure evidence standards for public health decision making can still be achieved, the performance of WGS must be thoroughly validated against current gold standard methods prior to implementation. Here we aim to verify the performance of WGS in comparison to pulsed-field gel electrophoresis (PFGE) and multiple-locus variable-number tandem repeat analysis (MLVA) for eight retrospective outbreaks of VTEC O157:H7 from the Canadian perspective. Since real-time implementation and routine use of WGS in public health laboratories is highly reliant on standardized data analysis tools, we also provide a comparative analysis of two popular methodologies for WGS analyses; an in-house developed single nucleotide variant phylogenomics (SNVPhyl) pipeline and the BioNumerics whole genome multilocus sequence typing (wgMLST) tool. To provide a useful and consistent starting point for examining laboratory-based surveillance data for VTEC O157:H7 in Canada, we also aim to describe the number of genetic differences observed among outbreak-associated isolates.

RESULTS

WGS provided enhanced resolution over traditional subtyping methods, and accurately distinguished outbreak-related isolates from non-outbreak related isolates with high epidemiological concordance. WGS also illuminated potential linkages between sporadic cases of illness and contaminated food, and isolates spanning multiple years. The topologies generated by SNVPhyl and wgMLST were highly congruent with strong statistical support. Few genetic differences were observed among outbreak-related isolates (≤5 SNVs/ < 10 wgMLST alleles) unless the outbreak was suspected to be multi-strain.

CONCLUSIONS

This study validates the superiority of WGS and indicates the BioNumerics wgMLST schema is suitable for surveillance and cluster detection of VTEC O157:H7. These findings will provide a useful and consistent starting point for examining WGS data for prospective laboratory-based surveillance of VTEC O157:H7, but however, the data will continue to be interpreted according to context and in combination with epidemiological and food safety evidence to inform public-health decision making in Canada.

Outbreak of Escherichia coli O157:H7 Infections Linked to Aged Raw Milk Gouda Cheese, Canada, 2013

Between 12 July and 29 September 2013, 29 individuals in five Canadian provinces became ill following infection with the same strain of Escherichia coli O157:H7 as defined by molecular typing results. Five case patients were hospitalized, and one died. Twenty-six case patients (90%) reported eating Gouda cheese originating from a dairy plant in British Columbia. All of the 22 case patients with sufficient product details available reported consuming Gouda cheese made with raw milk; this cheese had been produced between March and July 2013 and was aged for a minimum of 60 days. The outbreak strain was isolated from the implicated Gouda cheese, including one core sample obtained from an intact cheese wheel 83 days after production. The findings indicate that raw milk was the primary source of the E. coli O157:H7, which persisted through production and the minimum 60-day aging period. This outbreak is the third caused by E. coli O157:H7 traced to Gouda cheese made with raw milk in North America. These findings provide further evidence that a 60-day ripening period cannot ensure die-off of pathogens that might be present in raw milk Gouda cheese after production and have triggered an evaluation of processing conditions, physicochemical parameters, and options to mitigate the risk of E. coli O157:H7 infection associated with raw milk Gouda cheese produced in Canada.

Outbreak of Salmonella Typhimurium associated with feeder rodents

In December 2012, an increase in human Salmonella Typhimurium cases was identified in the province of Ontario, Canada launching an outbreak investigation. The outbreak spanned 3 years (2012-2014), with 134 cases reported from five Canadian provinces. There was a substantial burden of illness among children: 45% of cases were children 12 years old or under, and 23% of cases were under 5 years old. Epidemiologic, traceback and laboratory findings linked this outbreak to feeder rodents (used to feed snakes) supplied by a network of rodent breeders in Ontario. Cases likely acquired their illness through either direct or indirect contact with feeder rodents. This investigation not only contributes to the weight of evidence on the risk that feeder rodents pose, but also underscores the importance of investigating indirect animal contact and associated risks, especially for high-risk individuals.

Multi-provincial Salmonellosis Outbreak Related to Newly Hatched Chicks and Poults: A Genomics Perspective

BACKGROUND

A multi-provincial outbreak of Salmonella enterica serovar Enteritidis was linked to newly hatched chicks and poults from a single hatchery during the spring of 2015. In total, there were 61 human cases that were epidemiologically confirmed to be linked to the chicks and poults and the outbreak was deemed to have ended in the summer of 2015.

METHODS

PulseNet Canada, in coordination with the affected provinces, used genome sequencing of human and agricultural Salmonella Enteritidis isolates to aid in the epidemiological investigation, while also using traditional typing methods such as phagetyping and pulsed-field gel electrophoresis (PFGE).

RESULTS

All human outbreak cases, except one, were Phage Type (PT) 13a. Single nucleotide variant analysis (SNV) was able to provide a level of resolution commensurate with the results of the epidemiological investigation. SNV analysis was also able to separate PT13a outbreak-related isolates from isolates not linked to chicks or poults, while clustering some non-PT13a agricultural strains with the outbreak cluster.

CONCLUSIONS

Based on conventional typing methods (phagetyping or PFGE), clinical and agricultural PT13a SE isolates would have been considered as part of a related cluster. In contrast, phagetyping would have led to the exclusion of several non- PT13a strains that clustered with the outbreak isolates using the genome sequence data. This study demonstrates the improved resolution of genome sequence analysis for coordinated surveillance and source attribution of both human and agricultural SE isolates.

The utility of multiple molecular methods including whole genome sequencing as tools to differentiate Escherichia coli O157:H7 outbreaks

A standardised method for determining Escherichia coli O157:H7 strain relatedness using whole genome sequencing or virulence gene profiling is not yet established. We sought to assess the capacity of either high-throughput polymerase chain reaction (PCR) of 49 virulence genes, core-genome single nt variants (SNVs) or k-mer clustering to discriminate between outbreak-associated and sporadic E. coli O157:H7 isolates. Three outbreaks and multiple sporadic isolates from the province of Alberta, Canada were included in the study. Two of the outbreaks occurred concurrently in 2014 and one occurred in 2012. Pulsed-field gel electrophoresis (PFGE) and multilocus variable-number tandem repeat analysis (MLVA) were employed as comparator typing methods. The virulence gene profiles of isolates from the 2012 and 2014 Alberta outbreak events and contemporary sporadic isolates were mostly identical; therefore the set of virulence genes chosen in this study were not discriminatory enough to distinguish between outbreak clusters. Concordant with PFGE and MLVA results, core genome SNV and k-mer phylogenies clustered isolates from the 2012 and 2014 outbreaks as distinct events. k-mer phylogenies demonstrated increased discriminatory power compared with core SNV phylogenies. Prior to the widespread implementation of whole genome sequencing for routine public health use, issues surrounding cost, technical expertise, software standardisation, and data sharing/comparisons must be addressed.

The Impact of Multilocus Variable-Number Tandem-Repeat Analysis on PulseNet Canada Escherichia coli O157:H7 Laboratory Surveillance and Outbreak Support, 2008-2012

The lack of pattern diversity among pulsed-field gel electrophoresis (PFGE) profiles for Escherichia coli O157:H7 in Canada does not consistently provide optimal discrimination, and therefore, differentiating temporally and/or geographically associated sporadic cases from potential outbreak cases can at times impede investigations. To address this limitation, DNA sequence-based methods such as multilocus variable-number tandem-repeat analysis (MLVA) have been explored. To assess the performance of MLVA as a supplemental method to PFGE from the Canadian perspective, a retrospective analysis of all E. coli O157:H7 isolated in Canada from January 2008 to December 2012 (inclusive) was conducted. A total of 2285 E. coli O157:H7 isolates and 63 clusters of cases (by PFGE) were selected for the study. Based on the qualitative analysis, the addition of MLVA improved the categorization of cases for 60% of clusters and no change was observed for ∼40% of clusters investigated. In such situations, MLVA serves to confirm PFGE results, but may not add further information per se. The findings of this study demonstrate that MLVA data, when used in combination with PFGE-based analyses, provide additional resolution to the detection of clusters lacking PFGE diversity as well as demonstrate good epidemiological concordance. In addition, MLVA is able to identify cluster-associated isolates with variant PFGE pattern combinations that may have been previously missed by PFGE alone. Optimal laboratory surveillance in Canada is achieved with the application of PFGE and MLVA in tandem for routine surveillance, cluster detection, and outbreak response.

Multi-Province Listeriosis Outbreak Linked to Contaminated Deli Meat Consumed Primarily in Institutional Settings, Canada, 2008

A multi-province outbreak of listeriosis occurred in Canada from June to November 2008. Fifty-seven persons were infected with 1 of 3 similar outbreak strains defined by pulsed-field gel electrophoresis, and 24 (42%) individuals died. Forty-one (72%) of 57 individuals were residents of long-term care facilities or hospital inpatients during their exposure period. Descriptive epidemiology, product traceback, and detection of the outbreak strains of Listeria monocytogenes in food samples and the plant environment confirmed delicatessen meat manufactured by one establishment and purchased primarily by institutions was the source of the outbreak. The food safety investigation identified a plant environment conducive to the introduction and proliferation of L. monocytogenes and persistently contaminated with Listeria spp. This outbreak demonstrated the need for improved listeriosis surveillance, strict control of L. monocytogenes in establishments producing ready-to-eat foods, and advice to vulnerable populations and institutions serving these populations regarding which high-risk foods to avoid.

Isolation and detection of Shiga toxin-producing Escherichia coli in clinical stool samples using conventional and molecular methods

The isolation of Shiga toxin-producing Escherichia coli (STEC) other than serogroup O157 from clinical stool samples is problematic due to the lack of differential phenotypic characteristics from non-pathogenic E. coli. The development of molecular reagents capable of identifying both toxin and serogroup-specific genetic determinants holds promise for a more comprehensive characterization of stool samples and isolation of STEC strains. In this study, 876 stool samples from paediatric patients with gastroenteritis were screened for STEC using a cytotoxicity assay, commercial immunoassay and a conventional PCR targeting Shiga-toxin determinants. In addition, routine culture methods for isolating O157 STEC were also performed. The screening assays identified 45 stools presumptively containing STEC, and using non-differential culture techniques a total of 20 O157 and 22 non-O157 strains were isolated. These included STEC serotypes O157 : H7, O26 : H11, O121 : H19, O26 : NM, O103 : H2, O111 : NM, O115 : H18, O121 : NM, O145 : NM, O177 : NM and O5 : NM. Notably, multiple STEC serotypes were isolated from two clinical stool samples (yielding O157 : H7 and O26 : H11, or O157 : H7 and O103 : H2 isolates). These data were compared to molecular serogroup profiles determined directly from the stool enrichment cultures using a LUX real-time PCR assay targeting the O157 fimbrial gene lpfA, a microsphere suspension array targeting allelic variants of espZ and a gnd-based molecular O-antigen serogrouping method. The genetic profile of individual stool cultures indicated that the espZ microsphere array and lpfA real-time PCR assay could accurately predict the presence and provide preliminary typing for the STEC strains present in clinical samples. The gnd-based molecular serogrouping method provided additional corroborative evidence of serogroup identities. This toolbox of molecular methods provided robust detection capabilities for STEC in clinical stool samples, including co-infection of multiple serogroups.

Tracking clinical trial accrual strategies and barriers via a Web-based screening tool

6586

Background: The National Cancer Institute Community Cancer Centers Program (NCCCP) has developed a web based tracking tool to monitor screening of patients for clinical trial participation. The system is designed to collect data from multiple research sites throughout the geographically diverse group. The log blindly collects patient demographics, successful trial entry, and reasons for trial enrollment failure. Methods: Seven NCCCP sites utilized the log during the 60 day open accrual period for the Wake Forrest WFU 07–02–03 cancer control trial (chronic lymphocytic leukemia COLD- fX) in Novermber 2008 and December 2008. Results: 327 chronic lymphocytic leukemia patients were screened mostly by chart review. 162 (50%) were eligible for study entry. 45 (14%) entered the trial. 103 (31%) eligible patients declined entry. Of the 103, 72 (70%) did not wish to participate in research, 15 (15%) had travel constraints, 5 (5%) perceived excess toxicity, 9 (9%) gave no reason. 14 (3%) eligible had their physician decline entry. Overall, 28% of eligible patients entered the trial. 165 (50%) were trial ineligible. 82 (50%) for prior therapy, 30 (18%) for co morbid conditions, 10 (6%) missed an entry time requirement, 39 (24%) have incomplete data. Conclusions: The NCCCP trial screening log successfully captures real time clinical trial participation data from a diverse network of clinical trial sites. This data base will allow future analysis of individual clinical trial and site accrual barriers and empower strategies to increase trial participation.

No significant financial relationships to disclose.

Necessity of quality-controlled 16S rRNA gene sequence databases: identifying nontuberculous Mycobacterium species

The use of the 16S rRNA gene for identification of nontuberculous mycobacteria (NTM) provides a faster and better ability to accurately identify them in addition to contributing significantly in the discovery of new species. Despite their associated problems, many rely on the use of public sequence databases for sequence comparisons. To best evaluate the taxonomic status of NTM species submitted to our reference laboratory, we have created a 16S rRNA sequence database by sequencing 121 American Type Culture Collection strains encompassing 92 species of mycobacteria, and have also included chosen unique mycobacterial sequences from public sequence repositories. In addition, the Ribosomal Differentiation of Medical Microorganisms (RIDOM) service has made freely available on the Internet mycobacterial identification by 16S rRNA analysis. We have evaluated 122 clinical NTM species using our database, comparing >1,400 bp of the 16S gene, and the RIDOM database, comparing approximately 440 bp. The breakdown of analysis was as follows: 61 strains had a sequence with 100% similarity to the type strain of an established species, 19 strains showed a 1- to 5-bp divergence from an established species, 11 strains had sequences corresponding to uncharacterized strain sequences in public databases, and 31 strains represented unique sequences. Our experience with analysis of the 16S rRNA gene of patient strains has shown that clear-cut results are not the rule. As many clinical, research, and environmental laboratories currently employ 16S-based identification of bacteria, including mycobacteria, a freely available quality-controlled database such as that provided by RIDOM is essential to accurately identify species or detect true sequence variations leading to the discovery of new species.

Channel specificity and secondary structure of the glucose-inducible porins of Pseudomonas spp

The OprB porin-mediated glucose transport system was investigated in Pseudomonas chlororaphis, Burkholderia cepacia, and Pseudomonas fluorescens. Kinetic studies of [U-14C]glucose uptake revealed an inducible system of low Km values (0.3-5 microM) and high specificity for glucose. OprB homologs were purified and reconstituted into proteoliposomes. The porin function and channel preference for glucose were demonstrated by liposome swelling assays. Examination of the periplasmic glucose-binding protein (GBP) components by Western immunoblotting using P. aeruginosa GBP-specific antiserum revealed some homology between P. aeruginosa GBP and periplasmic proteins from P. fluorescens and P. chlororaphis but not B. cepacia. Circular dichroism spectropolarimetry of purified OprB-like porins from the three species revealed beta sheet contents of 31-50% in agreement with 40% beta sheet content for the P. aeruginosa OprB porin. These findings suggest that the high-affinity glucose transport system is primarily specific for glucose and well conserved in the genus Pseudomonas although its outer membrane component may differ in channel architecture and specificity for other carbohydrates.

Reasons for employment or non-employment during pregnancy

Although women's pattern of employment during pregnancy has markedly changed in the last 30 years, no research has explored the basis of contemporary women's decision making regarding employment during pregnancy. Conceptualizing pregnancy within a life span developmental perspective of adulthood, we used descriptive statistics and content analysis on data from 210 pregnant women to identify factors that influence employment decisions during pregnancy. Intrinsic reasons for working during pregnancy included self-actualization, escape from home, employment identity, and interpersonal contact. Extrinsic reasons for working included money, financial obligations, and employment setting compatibility. Intrinsic reasons for not working during pregnancy included pregnancy-related limitations, home identity, complexity, and family responsibilities. Extrinsic reasons for not working included financial juggle, job availability, and employment setting incompatibility. The results form a basis for anticipatory guidance recommendations for pregnant women.

Epidemiologic features of pain in pediatric cancer patients: a co-operative community-based study. North Central Cancer Treatment Group and Mayo Clinic)

The prevalence, etiology, and management of pain in pediatric cancer patients seen at the Mayo Clinic and member institutions of the North Central Cancer Treatment Group were assessed. Participating centers, including both primary care and referral institutions, surveyed all patients seen during a 1-week period (Monday through Friday); procedure-related pain was excluded. Of the 160 children surveyed, 28 reported pain of which 57.8% was related to a side effect of anticancer treatment, 21.1% was unrelated to the malignancy, and 21.1% arose directly from the cancer. Pain intensity assessment was performed by both health-care professional and patient using a variety of measurement tools. Correlation between assessors was close except in young children. The predominance of treatment-related rather than cancer-related pain differs from results in series in adult cancer patients.