State health department response to disease cluster reports: a protocol for investigation

Cancer clusters in the USA: what do the last twenty years of state and federal investigations tell us?

Background

Cancer clusters garner considerable public and legislative attention, and there is often an expectation that cluster investigations in a community will reveal a causal link to an environmental exposure. At a 1989 national conference on disease clusters, it was reported that cluster studies conducted in the 1970s and 1980s rarely, if ever, produced important findings. We seek to answer the question: Have cancer cluster investigations conducted by US health agencies in the past 20 years improved our understanding of cancer etiology, or informed cancer prevention and control?

Methods

We reviewed publicly available cancer cluster investigation reports since 1990, obtained from literature searches and by canvassing all 50 states and the District of Columbia. Investigations were categorized with respect to cancer type(s), hypothesized exposure, whether perceived clusters were confirmed (e.g. by elevated incidence), and conclusions about a link between cancer(s) of concern and hypothesized environmental exposure(s).

Results

We reviewed 428 investigations evaluating 567 cancers of concern. An increase in incidence was confirmed for 72 (13%) cancer categories (including the category “all sites”). Three of those were linked (with variable degree of certainty) to hypothesized exposures, but only one investigation revealed a clear cause.

Conclusions

It is fair to state that extensive efforts to find causes of community cancer clusters have not been successful. There are fundamental shortcomings to our current methods of investigating community cancer clusters. We recommend a multidisciplinary national dialogue on creative, innovative approaches to understanding when and why cancer and other chronic diseases cluster in space and time.

Bayesian versus frequentist statistical inference for investigating a one-off cancer cluster reported to a health department

Background

The problem of silent multiple comparisons is one of the most difficult statistical problems faced by scientists. It is a particular problem for investigating a one-off cancer cluster reported to a health department because any one of hundreds, or possibly thousands, of neighbourhoods, schools, or workplaces could have reported a cluster, which could have been for any one of several types of cancer or any one of several time periods.

Methods

This paper contrasts the frequentist approach with a Bayesian approach for dealing with silent multiple comparisons in the context of a one-off cluster reported to a health department. Two published cluster investigations were re-analysed using the Dunn-Sidak method to adjust frequentist p-values and confidence intervals for silent multiple comparisons. Bayesian methods were based on the Gamma distribution.

Results

Bayesian analysis with non-informative priors produced results similar to the frequentist analysis, and suggested that both clusters represented a statistical excess. In the frequentist framework, the statistical significance of both clusters was extremely sensitive to the number of silent multiple comparisons, which can only ever be a subjective "guesstimate". The Bayesian approach is also subjective: whether there is an apparent statistical excess depends on the specified prior.

Conclusion

In cluster investigations, the frequentist approach is just as subjective as the Bayesian approach, but the Bayesian approach is less ambitious in that it treats the analysis as a synthesis of data and personal judgements (possibly poor ones), rather than objective reality. Bayesian analysis is (arguably) a useful tool to support complicated decision-making, because it makes the uncertainty associated with silent multiple comparisons explicit.

Adequacy of state capacity to address noncommunicable disease clusters in the era of environmental public health tracking

OBJECTIVES

We sought to assess the capacity of state public health agencies to address noncommunicable disease clusters (NCCs) and to develop recommendations to enhance agencies' NCC response capacity.

METHODS

We conducted an inventory of state public health agency Web sites and administered a Web-based survey of state health agency personnel to examine NCC capacity with respect to responsibility and authority, scope, protocols, trends in NCC investigations, and desired assistance.

RESULTS

Twenty-six of the state agency Web sites listed an NCC contact, and 12 mentioned a cluster response team. Thirty-seven states completed the Web-based survey, all addressed cancer clusters, and 30 also responded to other NCCs, such as multiple sclerosis, amyotrophic lateral sclerosis, and autism.

CONCLUSIONS

NCCs are of key concern to communities, and all of the survey respondents indicated that citizen reports were an impetus for investigations; yet, state-level capacity to address NCCs was inconsistent and disjointed. State agency personnel were committed to responding to NCC inquiries but were hampered by lack of personnel, resources, and prescribed protocols, as well as inadequate interagency communication. We offer recommendations to address these challenges.

Answering cluster investigation requests: the value of simple simulations and statistical tools

Cluster investigations remain an important public health issue as the number of reported clusters and public concern increase. This study shows how statistical considerations and a simulation tool may be helpful in providing communities with proper answers to the questions usually raised in such situations: How surprising is an observed childhood cancer excess? What could be learned from a statistical test? What could be learned from a case-control study? Using real demographic and incidence-rate data together with simulations based on the hypothesis that incidence rates are homogeneous, the probabilities of observing given situations were estimated. A number of real situations have been used as examples. The results of the simulation study showed, in detail, that no reliable information on the reality of an observed excess could be obtained a posteriori from a statistical test. A cluster of the same size may or may not be surprising, depending on the spatial area and time window to which the cases are related (i.e., the expected number of cases), and depending on the size of the referential territory to which this area is associated. The lack of power of a case-control study if no particular unusual exposure is present is also addressed. The approach described in this paper can easily be reproduced and adapted to many situations. It may be of assistance to health departments conducting cluster investigations and communicating with the public.

Wisconsin's environmental public health tracking network: information systems design for childhood cancer surveillance

In this article we describe the development of an information system for environmental childhood cancer surveillance. The Wisconsin Cancer Registry annually receives more than 25,000 incident case reports. Approximately 269 cases per year involve children. Over time, there has been considerable community interest in understanding the role the environment plays as a cause of these cancer cases. Wisconsin's Public Health Information Network (WI-PHIN) is a robust web portal integrating both Health Alert Network and National Electronic Disease Surveillance System components. WI-PHIN is the information technology platform for all public health surveillance programs. Functions include the secure, automated exchange of cancer case data between public health-based and hospital-based cancer registrars; web-based supplemental data entry for environmental exposure confirmation and hypothesis testing; automated data analysis, visualization, and exposure-outcome record linkage; directories of public health and clinical personnel for role-based access control of sensitive surveillance information; public health information dissemination and alerting; and information technology security and critical infrastructure protection. For hypothesis generation, cancer case data are sent electronically to WI-PHIN and populate the integrated data repository. Environmental data are linked and the exposure-disease relationships are explored using statistical tools for ecologic exposure risk assessment. For hypothesis testing, case-control interviews collect exposure histories, including parental employment and residential histories. This information technology approach can thus serve as the basis for building a comprehensive system to assess environmental cancer etiology.

Methods for a public health response to birth defects clusters

Few resources are available to guide public health officials in investigations of reported birth defects clusters. The majority of published resources focus on the investigation of cancer and infectious disease clusters and do not address clinical and epidemiologic concerns specific to birth defects research. This document aims to address these concerns, discuss the needs of the affected community, and provide suggestions for the development of a standardized protocol to be used as a guide in the investigation of birth defects clusters. We suggest that health departments and birth defects registries that may receive reports of birth defects clusters establish a protocol for responding that includes the following steps: develop a proactive plan for future birth defects cluster reports (step I), receive report of a birth defects cluster (step II), verify diagnoses and complete case ascertainment (step III), compare the observed rate to a reference rate (step IV), ascertain exposures among cases from available records (step V), interview case mothers (step VI), initiate further epidemiologic study-selection of controls (step VII), and communicate results to the community (step VIII). Specific criteria for continuing or terminating an investigation should be established before receiving cluster reports. The recommendations in this report should be carefully considered to ensure that the specific needs of the region, agency and affected community are met.

Things to know and do about cancer clusters

Perceived cancer clusters present difficulties and opportunities for clinicians and public health officials alike. Public health officials receive reports of perceived cancer clusters, evaluate the validity of these reports, and/or launch investigations to identify potential causes. Clinicians interact directly with the affected patients, families, or community representatives who question the occurrence of cancer and the underlying causes. Clinicians may identify cancer clusters when they question the unusual occurrence of a rare form of cancer within their practice or community. In addition, clinicians may be asked to discuss cancer clusters and inform local debates. In this paper, we describe the public health practice experience with cancer clusters and identify cancer prevention and control opportunities for clinicians and public health officials. Scientific investigations of cancer clusters rarely uncover new knowledge about the causes of cancer. However, a set of common characteristics, unique to etiologic cluster investigations have uncovered new information about the causes of cancer or demonstrated a preventable link to a known carcinogen. These characteristics may provide useful clues for sorting out the small number of clusters worthy of further scientific investigation. Public awareness of cancer clusters may promote an opportunity to inform and motivate people about the preventable causes of cancer and effective cancer screening methods.

Exposure to armament wastes and leukemia: a case-control study within a cluster of AML and CML in Germany

An unusually high incidence of acute myeloid leukemia (AML) and chronic myeloid leukemia (CML) concentrated in a specific locality of a region in Germany motivated a descriptive incidence study in that region which showed a near 10-fold increased risk of CML among males but not among females (Kolb G, Becker N, Scheller S, Zugmaier G, Pralle H, Wahrendorf J, Havemann K. Increased risk of acute myelogenous leukemia (AML) and chronic myelogenous leukemia (CML) in a County of Hesse, Germany, Soc Prev Med 1993;38:190-195). Since a serious environmental contamination of areas in this locality with armament wastes containing toluene-derivatives has been known for a long time, the hypothesis arose that TNT production and the related severe contamination of soil and water might be responsible for the observed higher risk. We carried out a case-control study within the cluster to test this hypothesis. Overall, the results do not confirm the hypothesis. There is an indication of a relationship of an increased odds ratio with the exposure for a small group of persons who lived at a particular site in one of the two communities involved during the peak phase of TNT production during the 1940s. However, this finding is spurious and cannot explain the large majority of cases which occurred in that area in the 1980s. At the moment, no other explanation can be given for the increased risk of leukemias in that area.

Public requests for cancer cluster investigations: a survey of state health departments

OBJECTIVES

This study examined the frequency of requests that state health departments investigate cancer clusters, the nature of those requests, and the resources available for the investigations.

METHODS

A mail survey was sent to state health departments requesting data for 1997.

RESULTS

Approximately 1100 cluster investigation requests were made in 1997. Most requests were made by citizens, and no pattern emerged for types of cancer or hazards suspected. States rate this work as average in importance and feel satisfied with the successfullness of their communication efforts.

CONCLUSIONS

Few cluster inquiries require further investigation. Nonetheless, this interaction represents resources well spent in terms of public service and education.

Investigation of clusters of adverse reproductive outcomes, an overview

Clusters of adverse reproductive outcomes are reported with increasing frequency to public health services, but it is likely that only a minority of clusters are caused by a common environmental teratogen or mutagen. Many guidelines and protocols have been developed for the investigation and the management of chronic disease clusters and can be applied to the study of adverse reproductive outcomes. Cluster investigation normally follows four successive phases: (i) the generation of one or more etiologic hypotheses, (ii) the confirmation of the hypotheses, (iii) an intervention aimed at reducing any dangerous exposure, and (iv) the evaluation of the effect of the intervention. However, each situation is unique and there is no standard recipe for conducting the epidemiologic investigation at each successive phase, for choosing the appropriate statistical technique and for communicating with interested parties. In order to minimize both the risk of pursuing a worthless investigation and the risk of ignoring a real excess of cases caused by an adverse environmental factor, it is imperative to rely on a team of specialists with expertise in epidemiology, statistics, toxicology, embryology, and communication.

Investigation of a cluster of leukaemia in the Illawarra region of New South Wales, 1989-1996

OBJECTIVES

To investigate a cluster of leukaemia among young people and assess the plausibility of a disease-exposure relationship.

DESIGN

Descriptive analysis of population-based leukaemia incidence data, review of evidence related to the causation of leukaemia, assessment of environmental exposures to known leukaemogens, and resulting risks of leukaemia.

SETTING

Illawarra region of New South Wales, Australia, focusing on suburbs between the Port Kembla industrial complex and Lake Illawarra (the Warrawong area).

MAIN OUTCOME MEASURES

Standardised incidence ratios (SIRs) for leukaemia; current measured and past estimated ambient air benzene concentrations; and expected leukaemia cases attributable to estimates of ambient air benzene concentrations.

RESULTS

In 1989-1996, 12 leukaemia cases among Warrawong residents aged less than 50 years were observed, more than the 3.49 cases expected from the rate in the rest of the Illawarra region (SIR, 343.8; 99% CI, 141.6-691.7). These people lived in suburbs immediately to the south-southwest of a coke byproducts plant (a major industrial source of benzene, one of the few known leukaemogens). The greatest excess was among 15-24-year-olds (SIR, 1085.6; 99% CI, 234.1-3072.4). In 1996, ambient air concentrations of benzene averaged less than 1 part per billion (ppb). Since 1970, ambient air concentrations of benzene were estimated to have averaged up to 3 ppb, about one-thousandth of the level at which leukaemia risk has been identified in occupational epidemiological studies. Using the risk assessment model developed by the US Environmental Protection Agency, we estimate that past benzene levels in the Warrawong area could have resulted in 0.4 additional cases of leukaemia in 1989-1996.

CONCLUSIONS

The excess occurrence of leukaemia in the Warrawong area in 1989-1996 is highly unusual. Current environmental benzene exposure and the reconstructed past environmental benzene exposure level are too low to explain the large excess of leukaemia. The cause of the cluster is uncertain.

Weighing the risks and benefits of vaccination

The following summarizes this author's current thoughts regarding veterinary vaccines and their safety: 1. Every licensed animal vaccine is probably effective, but also produces some adverse effects. 2. Prelicensing studies of vaccines are not specifically designed to detect adverse vaccine reactions. 3. An improved system of national postmarketing surveillance is required to identify most adverse vaccine reactions that occur at low and moderate frequency. 4. Even a good postmarketing surveillance system is unlikely, however, to detect delayed adverse vaccine reactions, and the longer the delay the less likely they will be associated with vaccination. 5. Analytic epidemiologic (field) studies are the best way to link vaccination with delayed adverse reactions, but these are often hindered by incomplete vaccination histories in medical records in veterinary practice and by a lack of veterinarians in industry trained in epidemiologic methods. 6. Each licensed veterinary vaccine should be subjected to a quantitative risk assessment, and these should be updated on a regular basis as new information becomes available. 7. Risk assessment should be used to identify gaps in information regarding the safety and efficacy of vaccines, and appropriate epidemiologic studies conducted to fill these gaps that contribute to the uncertainty in risk estimates. 8. Risk assessment is an analytical process that is firmly based on scientific considerations, but it also requires judgments to be made when the available information is incomplete. These judgments inevitably draw on both scientific and policy considerations. 9. Representatives from industry, government, veterinary medicine, and the animal-owning public should be involved in risk management, that is, deciding between policy options. The controversy regarding vaccine risks is intensifying to the point that some animal owners have stopped vaccinating their animals. They offer as justification the belief that current vaccines are "just too dangerous." Some owners report that since they completely stopped vaccinating their animals, they have been healthy. What they fail to realize is that a high percentage of animal owners are responsible and do vaccinate their animals, thus providing "herd immunity" protection to the unvaccinated animals whom they contact. The solution to the vaccine controversy is not to abandon vaccination as an effective means of disease prevention and control, but rather to encourage vaccine research to answer important questions regarding safety and to identify the biological basis for adverse reactions. Key questions to be answered include these: What components of vaccines are responsible for adverse reactions? What is the genetic basis for susceptibility to adverse health effects in animals? How can susceptible individuals be identified? Do multivalent vaccines cause a higher rate of adverse reactions than monovalent vaccines? Is administration of multiple doses of monovalent vaccines really any safer than administering a single multivalent vaccine? These and other vaccine-related questions deserve our attention as veterinarians so we can fulfill our veterinary oath to relieve animal suffering and "above all else, do no harm."

Responding to reported clusters of common diseases: the case of multiple sclerosis

Reports of disease clustering are becoming ever more common, and there is increasing pressure on public health agencies to respond rapidly and appropriately to these reports. We investigated a cluster of five cases of MS occurring in a small multidisciplinary hospital department. Data were collected by a variety of methods, including measurement and description of the workplace, completion of survey instruments by department staff, and construction of case histories of persons with MS. The results indicated that the department comprised a high-risk population and that only one case of MS could have any possible etiologic significance. Investigators should consider a number of factors when evaluating disease clusters, including the accuracy of diagnosis, the induction period and cause of the disease, and possible biases in the population at risk. Additionally, boundaries should not encircle the cases that led to identification of the cluster and should reflect environmental significance.

Investigation of a cluster of ten cases of Hodgkin's disease in an occupational setting

The objective of this study was to identify occupational exposures that might be etiologically linked to an unusual cluster of ten cases of Hodgkin's disease. The cases were identified within the active workforce of a large chemical manufacturing firm over a 23-year period by the medical director of the facilities. Based on comparison with regional cancer incidence rates, the standardized incidence ratio for Hodgkin's disease was 497 (95% confidence interval: 238-915) for the period from the construction of the facilities in 1966 through early 1992. A nested case-control study was undertaken with 200 controls selected according to case-cohort sampling. Simultaneously, efforts were initiated to confirm and characterize each case more fully. Occupational exposures were identified and categorized using process, work history, medical record, and industrial hygiene data. Tissue slides were available for eight cases and a second review confirmed the diagnosis of Hodgkin's disease. For one case, a final diagnosis of large-cell anaplastic lymphoma was determined after histology review. Among 214 different chemical agents studied, eight were identified to which three or more of the cases had been exposed prior to the date of their initial diagnosis. Exposure odds ratios were statistically elevated for five of these agents; dose-response evaluations for two of the agents, ethylene oxide and benzene, failed to provide additional support for a causal relationship. In conclusion, although several statistical associations were identified, no substance emerged as a likely candidate for explaining the observed Hodgkin's disease cluster.

Apparently normal frequency of congenital anomalies in the highly polluted town of Cubatão, Brazil

Anencephaly is Suspected to cluster in the municipality of Cubatão (Brazil), and is attributed to industrial pollution. We surveyed malformations prospectively in 10,000 births, in 3 consecutive series, corresponding to 3 maternity hospital in Cubatão, under working definitions and norms taken from the ECLAMC (Latin American Collaborative Study of Congenital Malformations). The sample from Cubatão had a higher proportion of black racial ancestry, and lower frequencies for stillbirths and for immediate neonatal death, than in the rest of South America. The birth prevalence of malformations varied between the 3 hospitals in Cubatão, due to differences in ascertainment of minor defects. A higher-than-expected prevalence rate in Cubatão was observed only for postaxial polydactyly, probably due to a higher proportion of black racial ancestry, while lower-than-expected rates were obtained for congenital dislocation of the hip, and for babies with multiple congenital anomalies patterns, both of them attributed to underascertainment of minor defects. These observations do not support the suspicion of an elevated birth prevalence rate for anencephaly, neural tube defects, or any other major congenital malformation in Cubatão.

Assessment of cancer clusters using limited cohort data with spreadsheets: application to a leukemia cluster among rubber workers

This paper presents methods for initial investigation of occupational cancer clusters using limited data. Phase 1 of the methods developed uses basic data from the cluster of cases, but with cohort data limited to the size of the workforce at the plant. Phase 2 of the methods requires knowledge of the number of workers entering and leaving the workforce in each year. In the absence of data concerning age, the spreadsheet programs explore a variety of worker age distributions in synthetic cohort analyses. The methods were used to assess a cluster of six cases of leukemia with an average duration of work of 11 years in a tire manufacturing plant. It was concluded that the relative risk for the age range 30-50 was at least 7, providing further evidence for the association of leukemia with work in tire manufacture. The use of spreadsheet programs can provide a valuable first step assessment of apparent workplace cancer clusters.

Investigation of a spontaneous abortion cluster: lessons learned

An investigation of a reported spontaneous abortion excess in an office environment was undertaken employing a multidisciplinary approach, including (1) an epidemiologic/validation step; (2) an industrial hygiene survey, including electromagnetic field measurements and indoor air quality determinations; and (3) a risk perception/risk communication component. This approach was needed because there are numerous chemical and physical agents and psychosocial stressors that may potentially impact the reproductive status of female office workers. Although video display terminals (VDTs) are typically the focus of spontaneous abortion (SAB) investigations, one cannot ignore other stressors in the environment. Magnetic field exposures within a payroll area were determined using a hand-held survey meter and data logging dosimeter. On average, the full shift time-weighted average exposures of workers to extremely low frequency (ELF) magnetic fields in the payroll office area ranged from 1.0 to 5.6 mG. Influencing the investigation's protocol design were the Centers for Disease Control's (CDC's) recent Guidelines for Investigating Clusters of Health Events. Although these guidelines grew primarily out of cancer cluster investigations, we applied them in this instance and found them to be generalizable to reproductive hazards investigations. A spontaneous abortion excess was validated over a 2-year period among 26 women with 32 reproductive events, with rates 1.5-2.5 times the expected, depending on comparison figures used. Lessons learned in the investigation, including the applicability of the CDC's Cluster Investigation Protocol and the enormous importance of risk perception and risk communication, are described.

Working with community organizations to evaluate potential disease clusters

Increasingly, private citizens are organizing themselves to resist the intrusion of hazardous substances in their residential environments, or to promote the removal of the same. This movement is the source of the acronym (NIMBY-Not In My Back Yard). In North Carolina, the Cancer Surveillance Section has several experiences working with such community organizations. We have developed a commentary of our experience and observations on some of the characteristics that are more conductive to successful endeavors to resolve perception of increased health risk. Advantages and disadvantages of joint (health agency/concerned citizens group) studies or surveys are described. A case study of a successful joint survey to ascertain possible increased cancer incidence is described.

Approaches to disease cluster investigations in a state health department

Classical tests for clustering rarely have a major role in the investigation of disease clusters at the neighbourhood level performed by a health department. When examining a single cluster, the boundaries are often ill-defined, and the dimension (geographic, occupational, iatrogenic) is rarely obvious. Furthermore, using cluster tests to routinely scan disease registries looking for clusters is undesirable for several reasons. A better approach is to be vigilant for unusual environmental exposures, and to evaluate the impact they may have. In this context, cluster techniques may serve as part of a larger investigation that includes other epidemiologic approaches.

Solving the cluster puzzle: clues to follow and pitfalls to avoid

Dozens of methods have been proposed for the identification of disease clusters, although only a few are used routinely in published investigations. New methods, although designed to exploit some particular aspect of the data or use some specific statistical tool, are rarely compared thoroughly in terms of power or performance. Users, when confronted with the multitude of methods available, often select methods arbitrarily, basing choices on software availability, ease of implementation, or use experience rather than considerations of statistical power, possible alternative hypotheses (that is, cluster structure) and likely confounding. In this review, we extend our typology of disease clustering methods and apply it to many of the extant methods identifying strengths, weaknesses and unique features of the methods. We conclude with recommendations for which methods should be applied to which types of situations.

Cancer clusters: the importance of monitoring multiple geographic scales

Analysis of cancer incidence data at a variety of geographic scales provides surveillance information that can allay fears of the general public, prevent costly and unwarranted epidemiologic studies driven by political pressures, and target appropriate cases for further investigation. We systematically examined New Jersey Cancer Registry data (1979-1985) for childhood and young adult (0-24 years) cancers at multiple geographic scales--at the state level, then by degree of urbanization, county boundaries, and minor civil divisions. The state had increased rates for some cancers when compared to four other SEER (Surveillance, Epidemiology and End Results) states. No meaningful patterns at either the most urban/suburban/most rural scales or at the county level of analysis were found. At the minor civil division level, the Ederer, Myers and Mantel method found evidence of clustering of pediatric and young adult cases statewide. Stratification of cases by race yielded even stronger findings and indicated that whites had clustering of cases for several cancer types. In-depth analysis of individual cases yielded hypotheses for investigating identified clusters.

Response of a cancer registry to reports of disease clusters

A protocol has been developed to investigate and report perceived clusters of cancer using a population-based cancer registry. The protocol comprises a series of steps which lead to assessment of the cluster's importance on the basis of three criteria: (1) statistical evidence of clustering; (2) documentation of the existence of exposure to a carcinogen; and (3) biological plausibility of the relationship between the exposure and the cancer of interest. The evaluation of these criteria results in one of three recommendations: further study, surveillance only, or no action. The protocol provides a systematic approach for investigation, makes efficient use of available cancer registry data, and responds to public concerns. The protocol is demonstrated by its application to an inquiry concerning an apparent excess of lung cancer in a small Ontario town and the possible role of radon gas exposure. The public health importance and limitations of addressing perceived disease clusters are discussed.

Disease clusters in occupational medicine: a protocol for their investigation in the workplace

Disease clusters have been an important source of epidemiologic and medical information in the history of occupational medicine. Many accepted disease-exposure linkages were first observed and investigated as disease clusters in the workplace setting. Recent interest in disease cluster methodology has focused on traditional environmental settings. There has been very little work on a similar methodology for the investigation of disease clusters in the workplace, despite the many advantages of workplace cluster investigations for recognizing new etiologic associations. In this paper, a protocol is proposed and discussed which can be implemented in both acute and chronic disease cluster outbreaks in the workplace, where no obvious previously recognized cause is identified. A standardized approach to occupational disease cluster investigation will lead to increased efficiency, decreased social-political tensions, and a greater yield of scientific information.