Optimizing Epidemiology–Laboratory Collaborations

Although these isolated quotes make Louis Pasteur and Alexander Langmuir seem to have opposing ideas about the role of the laboratory in outbreak investigations, each quote has some validity. In the late nineteenth century, Pasteur’s laboratory data supporting the “germ theory” of disease not only led to pasteurization and vaccination (1), but also provided the evidence that swayed the court of scientific opinion toward accepting the theory that microorganisms are a cause of disease. This shift in scientific focus shed light on the then relatively unknown work of Dr. James Lind, Dr. Ignaz Semmelweiss, and Dr. John Snow, whose elegant maps and statistics introduced epidemiology as a science. Nearly a century later, at the time of Dr. Langmuir’s comment (2), a field investigation team, armed only with epidemiology tools, arrived in Pontiac, Michigan, and determined the chain of infection in a point-source outbreak associated with an unknown microbial agent growing in a new reservoir with a new mode of transmission (3). Importantly, the epidemiology results were used to halt the Pontiac fever outbreak—8 years before the laboratory of the Center for Disease Control (later Centers for Disease Control and Prevention [CDC]) identified Legionella spp. as the causative agent (4). These examples demonstrate that comparing the relative importance of epidemiology and laboratory science to a field investigation is similar to the age-old debate about the order of appearance of the chicken or the egg—it is circular and obscures more fundamental truths. What both Pasteur and Langmuir believed—and what history has shown—is that both epidemiologists and laboratory scientists can make independent discoveries that have significant scientific impact, but collaboration across these disciplines has a synergistic effect, yielding public health data that are stronger than either discipline can provide alone (2).

Modern Role of Public Health Laboratories in Field Investigations

During the early 1980s, revolutionary developments in molecular biology and computer science began to be reflected in technologic advances in laboratory science. As a result, investigators can now accomplish sample collection, electronic sample receipt and tracking, sample processing, patient diagnostics, chemical identification and quantification, microbial identification, and antimicrobial susceptibility testing more rapidly, safely, and accurately than ever before. Laboratory technology—including whole-genome sequencing, bioinformatics, and other technologies associated with advanced molecular detection—continues to develop at a phenomenal pace.

As laboratory technology has expanded over time, so has the laboratory role in field investigations. In addition to the traditional roles of providing causative agent and point-source identification and case confirmation, current laboratory results can be leveraged to inform all aspects of an outbreak investigation. Critical field investigation goals that have been informed by laboratory data include

• Etiologic agent identification and characterization (5,6).
• Case determination and detection (7,8).
• Point-source identification (9).
• Clinical care guidance and case management (10–12).
• Root cause analysis and intervention design (13).
• Outbreak control (14–16).
• Chain of infection definition (17,18).

Improving a multidisciplinary understanding of epidemiologic needs and requirements, as well as ways to best leverage advanced diagnostic laboratory capabilities, will continue to strengthen the performance of future outbreak investigation teams.

Chapter Goals

This chapter provides general guidance and recommendations for the field investigation team. While the chapter’s primary focus is infectious disease outbreaks, laboratory-related information for nonbiological exposures also is included. These instructions, although not necessarily universal, are intended to help the team consider how to integrate local, state, regional, national, and international laboratory expertise and capacity throughout the outbreak investigation process. Online resource links for the specific laboratory-related tasks involved in a field investigation are included to provide access to the most recently updated information.

These recommendations represent a best-case scenario. Adjustments to laboratory-related plans often are required because of resource or time constraints, emergencies, and local considerations. Local considerations can include policies, regulations, environment, climate, culture, infrastructure, and socioeconomics. In many cases, the field investigation team needs to use judgment to adapt these recommendations to the unique local circumstances—but only after careful consideration of their effects on safety, patient care, and outbreak control. This analysis should involve collaboration with field epidemiologists, laboratory experts, and local stakeholders.

Step 1. Involve the Laboratory Early and Often

Because new technologies are regularly added to the arsenal of available laboratory tests, listing them is impractical. This evolving catalog of tests means that optimal sample types also can change, thus dictating changes in collection protocols. Therefore, it is vital that outbreak investigators contact the relevant laboratories as soon as possible, preferably before deploying to the field investigation site. Local health department, state public health, and contract laboratories can perform some testing, whereas CDC subject-matter experts and laboratories can support investigations as needed through consultation, testing, and collaboration.

The general types of laboratory activities performed by CDC include

• Outbreak investigation.
• Emergency response.
• Population health studies.
• Laboratory quality improvement.
• Culture-based pathogen detection and characterization.
• Molecular-based pathogen detection and characterization.
• Detection of high-consequence pathogens.
• Genetic studies.
• Biomonitoring.
• Vaccine development.
• Pathogen discovery.
• Newborn screening.
• Occupational health.
• Chemical exposure testing.
• Assessment of environmental exposure.

A public health professional conducting an outbreak investigation should use the following checklist to enhance the effectiveness of these early communications with the laboratory receiving samples. This early communication is especially important in outbreaks of unknown etiology to ensure that samples are collected in a manner that enables rapid laboratory testing of a variety of potential etiologic hypotheses. It is also important to remain aware that etiologic hypotheses based on clinical signs and symptoms should include consideration of multiple factors (both biological and abiological) (19).

• If the investigation involves a suspected pathogen (e.g., bacteria, viruses, parasites, fungi) or biological toxin as an etiologic agent, then consult local laboratory resources and appropriate clinical and laboratory guidance (20). Pathogen-specific tests and contacts for CDC resources also can be found in the CDC Laboratory Test Directory (https://www.cdc.gov/laboratory/specimen-submission/list.html).
• If the investigation involves a suspected chemical or radiologic agent as an etiologic agent or if the outbreak involves a noninfectious disease or if the event might have been caused by biological, chemical, or radiologic terrorism, then contact the local or state public health department. CDC’s National Center for Environmental Health, Division of Laboratory Services (NCEH-DLS) (https://www.cdc.gov/nceh/dls/) also can offer guidance on options for testing and epidemiologic data, local laboratory resources, and relevant studies (21–25) and should guide testing during the investigation.
• Consult test descriptions relevant for the suspected agent and collaborate with the appropriate laboratory to determine the tests to perform based on the needs of the field investigation.
• Inform the laboratory of the purpose of the samples so that resources can be prioritized. The various purposes of sample submittal can include management of patient care, outbreak source identification, surveillance, research, law enforcement, and definition of cases.
• Gather information about the turnaround time for the selected test to manage expectations during the field investigation. Some tests can take up to 2 months to yield data.

Special Consideration: Test Selection

Test selection is one of the most important tasks to complete as early in the investigation as possible. A large number of tests are available to identify and characterize microorganisms, and serologic and antimicrobial susceptibility tests are available to provide a better understanding of potential exposures and inform treatment and infection control decisions. Local, state, and federal public health laboratories, as well as private hospital and contract laboratories, are available testing resources. However, it is important that the field investigator inquire about the accreditations or quality management programs of the laboratories to ensure use of the best resources for each field investigation. CDC laboratories can perform more than 300 infectious disease tests for a wide variety of microorganisms; the current list of available tests can be found at https://www.cdc.gov/laboratory/specimen-submission/list.html. This list can be used to select the appropriate tests for the suspected organism during a field investigation or to find CDC contact information for subject-matter experts on a particular organism.

Not all field investigations involve infectious microorganisms. For those investigations of noninfectious sources, some local and state health departments and NCEH-DLS perform testing in the areas of chronic disease markers, chemical and radiologic threat agents, environmental chemicals, newborn screening, and nutritional markers of disease. NCEH-DLS also can provide the field investigation team with appropriate sampling containers and protocols to minimize sample contamination from extraneous sources, as well as personnel to assist the field team. Contact NCEH-DLS directly through its website (https://www.cdc.gov/nceh/dls/).

To ensure the most rapid turnaround of accurate test samples, laboratories need lead time to order surge supplies, rearrange workloads, implement or modify protocols, and train additional staff. Laboratory experts can also help with some of the more difficult aspects of test selection. For example, specific symptoms (e.g., atypical pneumonia) can be associated with exposure to multiple biological and abiological etiologic agents; similar organisms (e.g., Chlamydia spp.) can cause different symptoms. In addition, multiple types of tests are available for most organisms, and not all of them give results that can be interpreted in the same way (i.e., a positive reverse-transcription polymerase chain reaction [RT-PCR] is not the same as a positive culture because RT-PCR tests for presence of nucleic acids and culture tests for viable organisms). Endemic diseases at the field investigation site also can affect testing for specific organisms (e.g., dengue, chikungunya, and Zika viruses), and some infections might require testing across multiple labs with different sample collection and submission requirements (e.g., healthcare-associated fungal infections might need testing by an environmental laboratory and a mycology laboratory). The setting (e.g., hospital), geography (e.g., Old World vs. New World hantaviruses), disease (e.g., genital ulcer disease), circumstances (e.g., biodefense), agent grouping (e.g., respiratory agents), sample type (e.g., whole blood), or purpose (e.g., surveillance) of the samples also can form the basis of test selection instead of, or in addition to, the suspected organism. Finally, some tests (especially chemical, radiologic, and molecular) are often exquisitely sensitive, and great care must be taken when selecting these tests to ensure appropriate sampling.

Special Consideration: Classic Versus Molecular Tests

Advanced molecular technologies backed by bioinformatics provide powerful new molecular detection systems that enable public health agencies to conduct surveillance, identify pathogens, recognize outbreaks, track transmission of a pathogen, detect antimicrobial resistance, and identify better ways to prevent disease. Genomics is central to many advanced molecular detection systems, and proteomics and transcriptomics are becoming important tools for public health. Molecular testing often can yield results more quickly than tests based on classical culture-based microbiology or serology/immunology, making them attractive alternatives during a field investigation. These tests play an increasingly important role in the general trend toward culture-independent diagnostic tests, which have the advantage of quickly and simultaneously testing for multiple pathogens within one sample. However, classical tests remain critical for completely characterizing disease-causing organisms and host responses and identifying new pathogens and new methods of antimicrobial resistance because detecting the presence of organism DNA/RNA by PCR is not the same as detecting a viable organism. Thus, investigation planning should include sample collection protocols suitable for the tests deemed most effective after consultation with the receiving laboratory.

Rapid molecular tests, such as real-time PCR, can yield definitive results within hours of sample receipt in the laboratory, but these use primers and probes designed to detect likely pathogens, which means these assays might miss an unsuspected or new pathogen. Technologies based on mass spectrometry (e.g., matrix-assisted laser desorption ionization–time of flight) are similarly limited in that the signal is compared with a reference library of likely pathogens. In contrast, untargeted nucleic acid sequencing (e.g., microbiome sequence analysis) can theoretically identify all organisms in a sample but present formidable analytic challenges because of the computational resources required to filter out irrelevant signals from host genome and benign commensal organisms. Even when a rapid molecular test identifies a pathogen, additional characterization using classical protocols often is needed to present a complete picture of the epidemiologic situation. For example, real-time PCR might identify the causative pathogen in an outbreak of meningitis as Neisseria meningitides, but the picture would be incomplete without also determining the serotype, which requires classical culture techniques.

Deciding on the tests to be performed on the samples before embarking on an investigation is necessary. The types of samples suitable for culture-independent diagnostic tests, which focus on the genotype (sequencing or PCR methods) or physical characteristics (mass spectroscopy) of the organism, can be different from those sample types needed for classical culture methods, which focus on the phenotype of the organism through methods such as serotyping and antimicrobial susceptibility testing. Protocols for culture-independent diagnostic tests often destroy the sample during nucleic acid or other target extraction, leaving no viable organisms to culture. Ideally, multiple samples should be collected, enabling both molecular and classical testing. However, if samples are in short supply or packaging and shipping limitations are a factor, testing and sample collection decisions should be made in advance in collaboration with the laboratory to determine whether molecular or classic tests would generate the most useful information.

For all tests, consult with the laboratory as early in the investigation as possible to address the following criteria for appropriate test selection:

• Preapproval.
• Supplemental information.
• Supplemental forms.
• Sample types.
• Acceptable samples.
• Minimum volumes required.
• Storage and preservation of samples.
• Sample transport medium.
• Sample labeling.
• Shipping instructions.
• Sample handling requirements.
• Testing methods.
• Testing turnaround time.
• Test interferences and limitations.
• Additional information.
• Laboratory points of contact.

Step 2. Collaborate on the Planning and Execution of Field Sample Collection

Collaborate with the laboratory before collecting samples to ensure samples are collected safely and are acceptable for the selected test. Accidental exposures during sample collection could result in severe illness, and unacceptable sample collection can result in missed opportunities for testing. For example, collecting blood from 20 people with potential exposures to an unknown pathogen and using the wrong anticoagulant can result in delayed microbial identification and delayed treatment, which could have serious consequences. The laboratory scientist can also provide ecology, growth, transmission, and pathogenesis expertise about microorganisms, as well as chemical and radiologic expertise. This expertise can support the investigation in many ways, some of which include

• A risk assessment of the planned sampling activities to create a safe environment for the work. At a minimum, the risk assessment should identify the potential biological, chemical, radiologic, and physical hazards and plan appropriate mitigations, including the use of personal protective equipment (PPE), to minimize exposure to hazards.
• A sampling plan, which includes preliminary hypotheses and ways the laboratory can assist in testing those hypotheses through targeted sampling.
• Sample collection methods, which need to be appropriate and sufficient for the specific tests (e.g., immunologic assays require specific timing of sample collection for diagnostic testing to be performed).
• Sample collection training, including assessing whether all the field investigators have adequate sampling experience and training (including training in PPE use) or whether laboratory personnel should deploy to the field to collect samples.
• Sample transport. Ideally, sampling activities should be planned to ensure that shipments arrive at the receiving laboratory on a weekday. Some shipments cannot be accepted on weekends.

A public health professional conducting an outbreak investigation should follow these generalized recommendations during sampling:

• Identify and obtain appropriate PPE in sufficient quantities before deploying to the field and double-check it for completeness at the sample collection location. Ensure the team has been trained in donning, doffing, cleaning/disinfection, storage, and proper disposal of the designated PPE.
• Review all relevant safety, infection control, and patient management guidelines before, and adhere to them during, sample collection. For instance, identify and maintain a specific area for donning and doffing the designated PPE and have a plan for managing sample collection waste.
• Collect an appropriate volume of sample.
• Label each clinical sample with at least two identifiers that link the sample to the patient, with the expectation that personal identifying information will not be used.
• Label each nonclinical sample (e.g., environmental, animal) with at least two identifiers that enable linking of the sample with the most pertinent organism, place, or thing (e.g., OrganismID and LocationID, SampleTypeID and MedicalDeviceID).
• Coordinate with local and state labs and clinicians to obtain samples. Do not collect samples without specific training in the collection procedure because the generalized guidance in this chapter might not be appropriate for a specific requested test.
• Review the special considerations discussed later and contact the laboratory for specific guidance before sampling.

Special Consideration: Risk Assessment

A risk is the possibility that an undesired event will occur (i.e., a function of the likelihood and consequences of a particular undesired event). Before conducting a field investigation or any laboratory activity, assess the risks associated with that activity. The epidemiologists and the laboratory scientists should conduct the risk assessment for a field investigation jointly, with assistance from other subject-matter experts including local and state public health laboratory scientists and epidemiologists, clinicians, appropriate facility staff, and appropriate emergency response planners and responders. Share the results of the risk assessment among the investigation team members so that everyone understands the risks involved. Use the results of the risk assessment as the basis for determining how to mitigate those risks. During the investigation, routinely monitor and reassess operations to identify and mitigate additional risks and to account for any new information or circumstance associated with particular activities. After the field investigation, review operations to identify ways to improve future field investigation risk assessment.

The principles of risk governance (26,27) articulate that a risk assessment should follow three general steps:

1. Define the situation: What work will occur?
2. Define the risks in that situation: What can go wrong?
3. Characterize each of the risks: How likely is each risk to occur? What would be the consequences of each risk?

Start the risk assessment process by thoroughly defining the situation and the activity. Particularly important is where the work will take place, who will conduct it (including their knowledge, skills, and abilities), what equipment they will use (including sampling and PPE), and what hazards they will encounter. A hazard is something that has the potential to cause harm, such as a sharp object or a biological agent. Considerations of risk should address the most obvious agent-related hazards of a field investigation (e.g., unintended exposure, physical injury). However, risk assessments should also encompass the less obvious hazards that could result in negative consequences. For each field investigation, in addition to agent-related hazards, also consider any investigation activities– related hazards that might result in negative outcomes (e.g., poor sample collection yielding incorrect or ambiguous laboratory results; shipping hazard resulting in delayed or incorrect case management; data management issues with patient privacy or consent; inadvertent violation of facility rules or local, state, or federal regulations; miscommunications that might erode collaborative relationships).

In a field investigation, the hazards are multifactorial, diverse, unique, and potentially of major consequence. To mitigate the risks, everyone involved in an investigation must consider everything that might go wrong during the various stages of that activity and then evaluate each risk from the perspective of its likelihood of occurrence and the consequences. After prioritizing those risks from highest to lowest, review the use of specific mitigations measures to reduce those identified risks. Before work begins, agree that the mitigated risks are acceptable. If the risks cannot be adequately mitigated and remain unacceptable, do not undertake the work.

Special Consideration: Etiologic Agents of Disease Syndromes

The same disease syndrome (e.g., respiratory) can be caused by one or more of many pathogens (e.g., influenza virus, Legionella spp., or hantaviruses) or by an abiological chemical or radiation exposure (e.g., chemical-induced acute respiratory distress syndrome). It is also possible for clinical presentations to be atypical, which can complicate field investigations (28–30). To maximize the effectiveness of investigations of outbreaks of unknown etiology, investigators need to collaborate with laboratory and epidemiology subject matter experts representing a diverse range of potential etiologic agents. Table 9.1 can be used as a tool in these larger discussions to help formulate hypotheses about the etiologic agents of various disease syndromes and to determine appropriate samples to collect for testing. Additional online resources for investigations of outbreaks of unknown etiology based on disease syndrome are also identified.

Special Consideration: Sample Collection

Table 9.2 shows sample collection supplies, a basic collection procedure, and sampling considerations for various clinical sample types for infectious disease testing. For similar information for environmental toxicant testing, see Table 9.3. These guidelines are presented to help in planning for sample collection in the field but should always be discussed with the laboratory before sample collection takes place.

Special Consideration: Potential Sampling Pitfalls

One reason it is important to contact the laboratory before sampling is because of sampling pitfalls. The pitfalls can include issues with sampling tools (e.g., inhibition of tests by certain swab materials), sample collection technique (e.g., hemolysis of blood samples), sample storage (e.g., degradation of RNA), or sample timing (e.g., matched sera). Table 9.4 describes some specific pitfalls to avoid, but it is not a comprehensive list, so consult with the laboratory prior to collecting samples.

Special Consideration: Personal Protective Equipment

PPE is specialized clothing or equipment used to protect against exposure to hazards that can cause serious injury or illness. Exposures can result from contact with chemical, radiologic, physical, electrical, mechanical, or other hazards. PPE may include items such as gloves, safety glasses and shoes, earplugs or muffs, hard hats, respirators, or coveralls, vests, and full-body suits. PPE selection should be tailored to the specific risks associated with each individual field investigation. The Occupational Safety and Health Administration has created multiple online resources that can be used to help select appropriate PPE and identify additional safety-related information for specific hazards (https://www.osha.gov/dts/osta/oshasoft/index.html). Additional Occupational Safety and Health Administration (https://www.osha.gov/Publications/osha3151.pdf) and National Institute for Occupational Safety and Health (https://www.cdc.gov/niosh/docs/2005100/pdfs/2005-100.pdf) guidance is also available.

In general, there are three major questions to consider when selecting PPE. First, what is the anticipated exposure type (splash, spill, spray), volume (large, small), and source (chemical, biological, or radiologic agent)? Second, what PPE is durable enough and appropriate for the task (protect against fluids, powders, gases)? Third, how will the PPE affect movement and work (appropriate size, not too hot)? Some PPE considerations are shown in Table 9.5, but investigation teams should consult with appropriate laboratory scientists on the types of PPE that are most effective for any particular field investigation as biological, chemical, and radiologic hazards each require specialized PPE.

Step 3. Collaborate with Laboratory for Storage and Shipment of Samples

The sender is responsible for ensuring that samples are stored and transported to the laboratory under appropriate conditions. Shipping requirements for infectious or potentially harmful samples submitted for diagnostic or investigational purposes must be packaged and shipped in compliance with appropriate regulations. Before shipping any samples from the field investigation, consult with the receiving laboratory and appropriate shipping experts to ensure adherence to relevant regulations and best practices. The information and links given here provide a starting point to navigating the storage, submission, and shipping requirements you may need to address.

• Samples to be sent to CDC for infectious disease testing:
  • Storage requirements: See the specific storage requirements for each test at https://www.cdc.gov/laboratory/specimen-submission/list.html
  • Submission forms: https://www.cdc.gov/laboratory/specimen-submission/form.html
  • Shipping instructions: https://www.cdc.gov/laboratory/specimen-submission/shipping-packing.html
• Samples to be sent to CDC for chemical or radiologic testing:
  • Radiologic: https://emergency.cdc.gov/radiation/labinfo.asp
  • Chemical: https://emergency.cdc.gov/chemical/lab.asp

• Samples to be sent to a state, regional, or local laboratory
  • Work with the state, regional, local, or facility representative to contact the appropriate laboratory.
  • In the event of an emergency in the field, contact the state public health laboratory. A list of emergency contacts for state public health laboratories can be found at http://www.aphl.org/programs/preparedness/Crisis-Management/Pages/Emergency-Lab-Contacts.aspx

The following checklist provides some practical advice about packaging and shipping samples.

• Double-check that all sample containers are closed and intact.
• Disinfect the outside of the sample container before transport, storage, or shipment. Be sure to maintain the integrity of the labels.
• Transport samples in their primary container (e.g., tube, bottle, sample container) placed into a sealed and leak-proof secondary container (plastic bag, plastic container) prior to placing in the outer container (shipping envelope, shipping box).
• Ensure samples are cushioned to prevent breakage.
• Create a line-list of the samples with all appropriate information (e.g., sample site, sample type, patient identifier, device identifier, environment location, suspected source of sample).
• In some investigations (e.g., when criminal activity is suspected), a formal chain of custody form may be needed. Consult your local public health officials, shipping experts, and the receiving laboratory to obtain appropriate forms.
• Determine the specific shipping requirements of the samples (https://www.iata.org/whatwedo/cargo/dgr/Documents/infectious-substance-classification-DGR56-en.pdf).
• Provide the tracking number of the shipment to the laboratory.

Step 4. Collaborate on the Interpretation of Laboratory Test Results

Laboratory tests are more complicated than ever to interpret, and the subsequent conclusions from—and uses of—laboratory data from any field investigation are most effective and reliable when collaboration is strong among the following:

• Laboratory scientists, who can explain the language of the laboratory report and any specifics of the test or organism.
• Epidemiologists, who interpret the tests in the context of the field investigation.
• Clinicians, who interpret the tests in the context of patient management.

Because each of these specialties has different knowledge and serves different purposes, laboratory data are most effective when interpretation occurs in collaboration. This ensures that data limitations and strengths are understood and that the potential occurrence and consequences of false positives or negatives, undetermined results due to improper sample collection or insufficient sample volume, and nonreportable results due to values below the limit of detection or other confounders found during testing are minimized. Issues to consider when interpreting laboratory results may include the following:

• Laboratory tests for field investigations should be interpreted in the context of properly framed epidemiologic hypotheses. Investigators should always consider why the test was chosen and what was being asked.
• Interpretation of test results depends on sensitivity and specificity of the selected test. Consider how the prevalence of disease will affect the predictive value of the test.
• Consider what population the test characteristics are derived from and how that might differ from the population being tested.
• If a sample is negative on nonspecific media, the sample cannot be interpreted as negative unless it is also negative on a suspected agent-specific medium.
• In field investigations that involve emerging pathogens, Koch’s postulates form the basis of proof that an emergent agent is the etiologic agent. Therefore, the interpretation should consider the successful fulfillment of each of Koch’s postulates. Just because an agent is found does not necessarily mean it caused the disease.
• Consider the status of the patient because pathogens behave differently in different hosts. Patient factors to consider include immunologic status, treatments administered, age, physiologic status, sex, and race.
• Molecular typing and other relatedness studies can confirm the relatedness of isolates to support the epidemiologic hypotheses generated by the field investigation. Interpretation of relatedness is specific for the typing assay used.
• Reference ranges are not precise and can vary by laboratory, depend on the test, and are typically selected to contain 95% of healthy persons. However, correlation between out-of-range values and illness is not always clear. Consider the sample size used to collect the values that established the reference range, the demographics of the population, and the reference range sample population.

Interpreting test results in collaboration with the laboratory is a standard best practice. Table 9.6 shows some considerations for interpretation of laboratory results; these can be used as a guide to facilitate collaborations with clinicians, epidemiologists, and laboratory scientists.

Step 5. Continue Laboratory Collaboration Through Publication of Findings

Any field investigation could lead to discovery of a new pandemic pathogen (31), identification of a product or device hazard (32), discovery of an old pathogen in a new place (33), identification of a new risk to public health (34), or even arrest of a criminal (35). Given the significance of field investigations, accuracy and completeness are critical. History has shown that when public health recommendations must (for political reasons or in emergency situations) be based solely on epidemiologic data, unintended consequences can occur (36). Similarly, poor laboratory diagnostic capabilities also can create unnecessary difficulties in patient care (37). Therefore, the most effective and reliable field investigation teams are built on a strong collaboration between epidemiology and laboratory science. Sustaining that cooperation through all stages of the scientific process, including analysis of the data, formulation of conclusions, and presentation or publication of results, is also necessary. Several general considerations can help inform that collaboration:

• Investigators should understand the tests and how to interpret them, including knowing the test limitations and considering those limitations in the context of the hypotheses.
• Laboratory staff should conduct data analysis from laboratory samples using appropriate biostatistics and reference standards.
• Field investigators should collaborate with the laboratory staff to determine whether laboratory results support or refute the epidemiologic hypotheses.
• Field investigators (both epidemiologists and laboratory scientists) should collaborate to design laboratory or epidemiologic studies that can further develop the field investigation findings into public health guidance.
• Field investigators and laboratory staff should jointly pursue publication of results, as deemed appropriate by the investigation and in accordance with all appropriate author guidelines.

Epidemiology is the scientific foundation of public health. However, like any foundation, it must rest on solid ground—in this case, a composite of scientific disciplines. Local, state, and federal public health laboratories have experts in a wide variety of technical areas, including microbiology, parasitology, mycology, statistics, bioinformatics, molecular biology, toxicology, ecology, chemistry, occupational health, microbial ecology, laboratory science, environmental health, biosafety, biosecurity, clinical management, and medical technology. Therefore, creating stable and sustainable collaborations with the laboratory will improve the practice of epidemiology and, in turn, improve public health.

1. Berche, P., Louis Pasteur, from crystals of life to vaccination. Clin Microbiol Infect. 2012;18 Suppl 5:1–6.
2. Dowdle WR, Mayer LW, Steinberg KK, Ghiya ND, Popovic T; CDC. Laboratory contributions to public health. MMWR Suppl. 2011;60:27–34.
3. Glick TH, Gregg MB, Berman B, Mallison G, Rhodes WW Jr, Kassanoff I. Pontiac fever. An epidemic of unknown etiology in a health department: I. Clinical and epidemiologic aspects. Am J Epidemiol. 1978;107:149–60.
4. Kaufmann AF, McDade JE, Patton CM, et al. Pontiac fever: isolation of the etiologic agent (Legionella pneumophilia) and demonstration of its mode of transmission. Am J Epidemiol. 1981;114:337–47.
5. Jernigan DB, Raghunathan PL, Bell BP, et al. Investigation of bioterrorism-related anthrax, United States, 2001: epidemiologic findings. Emerg Infect Dis. 2002;8:1019–28.
6. Gieraltowski L, Higa J, Peralta V, et al. National outbreak of multidrug resistant Salmonella Heidelberg infections linked to a single poultry company. PLoS One. 2016;11:e0162369.
7. CDC. Severe acute respiratory syndrome (SARS) and coronavirus testing—United States, 2003. MMWR. 2003;52:297–302.
8. Moturi E, Mahmud A, Kamadjeu R, et al. Contribution of contact sampling in increasing sensitivity of poliovirus detection during a polio outbreak—Somalia, 2013. Open Forum Infect Dis. 2016;3:ofw111.
9. Hoffmaster AR, Fitzgerald CC, Ribot E, Mayer LW. Molecular subtyping of Bacillus anthracis and the 2001 bioterrorism-associated anthrax outbreak, United States. Emerg Infect Dis. 2002;8:1111–6.
10. Nakao JH, Talkington D, Bopp CA, et al. Unusually high illness severity and short incubation periods in two foodborne outbreaks of Salmonella Heidelberg infections with potential coincident Staphylococcus aureus intoxication. Epidemiol Infect. 2018;146:19–27.
11. Folster JP, Grass JE, Bicknese A, Taylor J, Friedman CR, Whichard JM. Characterization of resistance genes and plasmids from outbreaks and illness clusters caused by Salmonella resistant to ceftriaxone in the United States, 2011–2012. Microb Drug Resist. 2017;23:188–93.
12. de Oliveira AM, Skarbinski J, Ouma PO, et al. Performance of malaria rapid diagnostic tests as part of routine malaria case management in Kenya. Am J Trop Med Hyg. 2009;80:470–4.
13. Tyndall JA, Gerona R, De Portu G, et al. An outbreak of acute delirium from exposure to the synthetic cannabinoid AB-CHMINACA. Clin Toxicol (Phila). 2015;53:950–6.
14. Flint M, Goodman CH, Bearden S, et al. Ebola virus diagnostics: the US Centers for Disease Control and Prevention laboratory in Sierra Leone, August 2014 to March 2015. J Infect Dis. 2015;212 Suppl 2:S350–8.
15. Jelden KC, Iwen PC, Herstein JJ, et al. U.S. Ebola treatment center clinical laboratory support. J Clin Microbiol. 2016;54:1031–5.
16. McCarty CL, Basler C, Karwowski M, et al. Response to importation of a case of Ebola virus disease—Ohio, October 2014. MMWR. 2014;63:1089–91.
17. CDC. Spinal and paraspinal infections associated with contaminated methylprednisolone acetate injections—Michigan, 2012–2013. MMWR. 2013;62:377–81.
18. Lockhart SR, Pham CD, Gade L, et al. Preliminary laboratory report of fungal infections associated with contaminated methylprednisolone injections. J Clin Microbiol. 2013;51:2654–61.
19. CDC. Recognition of illness associated with exposure to chemical agents—United States, 2003. MMWR. 2003;52:938–40.
20. Baron EJ, Miller JM, Weinstein MP, et al. A guide to utilization of the microbiology laboratory for diagnosis of infectious diseases: 2013 recommendations by the Infectious Diseases Society of America (IDSA) and the American Society for Microbiology (ASM)(a). Clin Infect Dis. 2013;57:e22–e121.
21. Chen L, Brueck SE, Niemeier MT. Evaluation of potential noise exposures in hospital operating rooms. AORN J. 2012;96:412–8.
22. Dickerson AS, Rahbar MH, Han I, et al. Autism spectrum disorder prevalence and proximity to industrial facilities releasing arsenic, lead or mercury. Sci Total Environ. 2015;536:245–51.
23. Erck Lambert AB, Parks SE, Camperlengo L, et al. Death scene investigation and autopsy practices in sudden unexpected infant deaths. J Pediatr. 2016;174:84–90 e1.
24. Kennedy C, Lordo R, Sucosky MS, Boehm R, Brown MJ. Evaluating the effectiveness of state specific lead-based paint hazard risk reduction laws in preventing recurring incidences of lead poisoning in children. Int J Hyg Environ Health. 2016;219:110–7.
25. Miller CW1, Ansari A, Martin C, Chang A, Buzzell J, Whitcomb RC Jr. Use of epidemiological data and direct bioassay for prioritization of affected populations in a large-scale radiation emergency. Health Phys. 2011;101:209–15.
26. Renn O, Graham P. Risk governance: towards an integrative approach. Geneva: International Risk Governance Council; 2005:157.
27. Salerno RM, Gaudioso J. Laboratory biorisk management: biosafety and biosecurity. Boca Raton, FL: CRC Press; 2015.
28. Sejvar J, Lutterloh E, Naiene J, et al. Neurologic manifestations associated with an outbreak of typhoid fever, Malawi–Mozambique, 2009: an epidemiologic investigation. PLoS One. 2012;7:e46099.
29. Balestri R, Bellino M, Landini L, et al. Atypical presentation of enterovirus infection in adults: outbreak of ‘hand, foot, mouth and scalp disease’ in northern Italy. J Eur Acad Dermatol Venereol. 2017;32:e60–e61.
30. Majumdar R, Jana CK, Ghosh S, Biswas U. Clinical spectrum of dengue fever in a tertiary care centre with particular reference to atypical presentation in the 2012 outbreak in Kolkata. J Indian Med Assoc. 2012;110:904–6.
31. CDC. Pneumocystis pneumonia—Los Angeles. MMWR. 1981;30:250–2.
32. Hawley B, Casey ML, Cox-Ganser JM, Edwards N, Fedan KB, Cummings KJ. Notes from the field: respiratory symptoms and skin irritation among hospital workers using a new disinfection product—Pennsylvania, 2015. MMWR. 2016;65:400–1.
33. Duffy MR, Chen TH, Hancock WT, et al. Zika virus outbreak on Yap Island, Federated States of Micronesia. N Engl J Med. 2009;360:2536–43.
34. Cherry C, Leong K, Wallen R, Buttke D. Notes from the field: injuries associated with bison encounters—Yellowstone National Park, 2015. MMWR. 2016;65:293–4.
35. Istre GR, Gustafson TL, Baron RC, Martin DL, Orlowski JP. A mysterious cluster of deaths and cardiopulmonary arrests in a pediatric intensive care unit. N Engl J Med. 1985;313:205–11.
36. Sencer D. How should the federal government respond to the influenza problem caused by a new virus? In: Neustadt RE, Fineberg HV. The swine flu affair: decision-making on a slippery disease. Washington, DC: National Academies Press; 1978. https://www.ncbi.nlm.nih.gov/books/NBK219607/
37. Moore A, Nelson C, Molins C, Mead P, Schriefer M. Current guidelines, common clinical pitfalls, and future directions for laboratory diagnosis of Lyme disease, United States. Emerg Infect Dis. 2016;22.
38. World Health Organization. Guidelines for the collection of clinical specimens during field investigation of outbreaks. Geneva: World Health Organization; 2000.
39. Crump JA1, Corder JR, Henshaw NG, Reller LB. Development, implementation, and impact of acceptability criteria for serologic tests for infectious diseases. J Clin Microbiol. 2004;42:881–3.
40. Jorgensen JH, Ferraro MJ. Antimicrobial susceptibility testing: a review of general principles and contemporary practices. Clin Infect Dis. 2009;49:1749–55.
41. Empson MB. Statistics in the pathology laboratory: diagnostic test interpretation. Pathology. 2002;34:365–9.