Assign. 2

Snapple_girl1

06CH_Community_Bell.pdf

Home >History homework help >Assign. 2

6 Epidemiology

Microgen/iStock/Thinkstock

Learning Outcomes

After reading this chapter, you should be able to

• Explain epidemiology and its use in public health.

• Outline methods for disease surveillance.

• Compare descriptive and analytic epidemiology.

• Apply the 13 epidemiological steps to investigations.

182

Section 6.1 What Is Epidemiology?

Understanding how, when, and why disease occurs is crucial for successful public health ini- tiatives. This chapter will demonstrate how studying the incidence, distribution, and control of disease contributes to the field of community health. Epidemiology is important because many circumstances that produce adverse health effects among community residents occur at the population level; thus, it is vital to take a population perspective when examining indi- vidual health outcomes in the community. For example, people in communities surrounded by high traffic volume are likely to suffer from respiratory issues because of the exhaust par- ticles in the air. A city susceptible to frequent cloud cover and very few sunny days, such as Seattle, Washington, might struggle more with mental health/depression. The study of how, when, and why disease occurs focuses on the health of populations, and, in this respect, it dif- fers from clinical medicine’s involvement with individual patients.

In fact, epidemiology provides a method- ological foundation for the entire public health field by embracing a spectrum of tools for studying health and illness. These methodologies include natural experiments, descriptive and analytic study designs (e.g., cross-sectional, case-control, cohort, and experimental), and mapping technologies. Epidemiologic research findings help develop hypotheses that can be applied to the health of the community and the study of potential causal relationships.

Epidemiologic research is likened to detective work because the causes of many diseases— especially when they first appear—are

unknown. Some examples are hantavirus in national parks, periodic episodes of foodborne illnesses, West Nile virus, and the resurgence of whooping cough (pertussis). This chapter presents epidemiologic procedures and methodologies that aid in unraveling the causes of mysterious disease outbreaks and health conditions that can afflict community members.

6.1 What Is Epidemiology? Epidemiology is the study of the occurrence and distribution of illnesses, injuries, and dis- eases in specific populations. It also includes the study of the factors that influence illnesses, diseases, and injuries in an effort to help reduce or eliminate the problem.

Epidemiology is a discipline that describes, quantifies, and finds possible causes, or deter- minants, for health phenomena in populations. Determinants are also known as etiological, or causal, factors. Recall that a determinant of health is a factor that affects health either negatively or positively. For example, economic status is considered a determinant. Those with more money seem to have better health outcomes than those with less money, making income also a determinant of health. Other examples of determinants or etiological factors are behavioral, such as smoking (negative) or physical activity (positive). Smoking is a factor in the development of arteriosclerosis, which is a poor health outcome. Physical activity can ward off obesity and obesity-related diseases, which is a good health outcome.

PBFloyd/iStock/Thinkstock Epidemiologists study the how, when, and why of a disease outbreak by using different methods of investigation, including experiments and mapping.

183

Section 6.1 What Is Epidemiology?

The results of epidemiologic studies aid public health practitioners in their quest to control health problems such as disease outbreaks. Like much of public health, epidemiology is inter- disciplinary because it operates within other fields, including clinical medicine, biostatistics, toxicology, the social sciences, and genetics. One example of this interdisciplinary approach is the application of genetics and the Human Genome Project to epidemiology in order to con- duct research on the genetic bases for breast and ovarian cancer. The Human Genome Project began in 1990 as an international effort to understand the sequence of chemicals in DNA and to identify and map the thousands of genes in the human genome. Researchers delivered a complete map in 2003, propelling further genomic research and explora- tion into medical applications. Epidemiologists are using that information to uncover the link between genes and certain cancers, such as breast cancer.

History The history of epidemiology extends over many centuries, beginning with the contributions of the ancient Greeks during the classical period. Some of the landmarks in the history of epi- demiology overlap with the history of public health. Recall the historically noteworthy and frightening epidemics that threatened the very existence of humanity, which were detailed in Chapter 1. These epidemics included the Black Death, which occurred between 1346 and 1352, and the great influenza pandemic that coincided with World War I early in the 20th century. Significant historical developments in epidemiology included Edward Jenner’s devel- opment of an effective vaccine against smallpox, and Robert Koch’s work that led to the iden- tification of microbial agents in human disease. In 1928, Alexander Fleming discovered that certain molds had antibiotic properties. As a result of this research, the antibiotic penicillin became available at the end of World War II. Table 6.1 lists several other important mile- stones in the history of epidemiology.

Jupiterimages/Stockbyte/Thinkstock Physical activity is an example of a determinant or etiological factor that can positively affect one’s health.

Table 6.1: Key events in the history of epidemiology

Date Event Significance

400 BCE Hippocrates publishes On Airs, Waters, and Places

Implies that the physical environment is associated with human illness

1662 John Graunt publishes Natural and Political Observations

Pioneers work in vital statistics (data that pertain to births and deaths)

Mid-1800s John Snow investigates a cholera outbreak in London

Uses the method of natural experiment to examine the causes of cholera

1964 The surgeon general’s report Smoking and Health is published

States that smoking is the cause of lung cancer

1974 The Lalonde Report, A New Perspective on the Health of Canadians, is published

Initiates large community health studies; community interventions are implemented

184

Section 6.1 What Is Epidemiology?

Hippocrates Links Environment and Illness In 400 BCE, the Greek physician Hippocrates suggested in his treatise On Airs, Waters, and Places that disease could be linked with a person’s physical environment—marking perhaps the earliest instance that logic, rather than mysticism, was used to explain an illness’s origin (Friis & Sellers, 2009). Hippocrates wrote that the seasons, the influence of the sun, the qual- ity of water, and the elevation at which people live were factors in health.

John Graunt Describes Trends in Vital Statistics In 1662, John Graunt published Natural and Political Observations Mentioned in a Following Index, and Made Upon the Bills of Mortality, more commonly referred to as simply Observa- tions. Graunt described various details of birth and death data, including seasonal variations and infant mortality. Public health historians regard Graunt’s work as among the first to organize mortality data in tables in order to discern trends in births and deaths from spe- cific causes (Friis & Sellers, 2009). Refer to Spotlight on Public Health Figures for more about Graunt’s contributions to public health.

Spotlight on Public Health Figures: John Graunt (1620–1674)

Who is John Graunt? John Graunt was born in 1620 to a storekeeper in England. He was the oldest of eight children. He initially worked as a shopkeeper like his father but eventually became involved in politics for the city of London. Despite his lack of formal education, he found mortality statistics interesting and wrote a book called Natural and Political Observations Mentioned in a Following Index, and Made Upon the Bills of Mortality. In this book, Graunt analyzed London’s “bills of mortality”—basically a list of the dead—and was the first to use this kind of data to identify trends and draw conclusions.

What was the political climate at the time? During Graunt’s adulthood, Britain was prospering. The economy was strong, and trade and colonization were extending across the globe. It was also considered the age of intellectual advances, and the time period produced a number of thinkers such as poets John Milton and John Dryden, philosophers such as John Locke and Thomas Hobbes, and scientists such as Isaac Newton and Robert Boyle.

(continued)

evryka23/iStock/Thinkstock John Graunt was fascinated by statistics. He wrote Observations, which helped set up a framework for tracking vital population statistics.

185

Section 6.1 What Is Epidemiology?

John Snow Maps Cholera Outbreak Chapter 1 described the work of John Snow, who, during the mid-19th century, investigated a cholera epidemic in London, eventually linking the epidemic to contaminated water sup- plies (Snow, 1855). His classic observational study reflects many of the features of modern epidemiologic inquiry, including mapping the location of instances of disease and tabulating fatalities. The natural experiment described in Snow on Cholera (Frost, 1936/1965), which includes two papers by John Snow and a biographic memoir, reported how changing to a cleaner water source reduced the occurrence of cholera. Despite the absence of knowledge about the nature of microbial agents during his time, Snow made numerous insightful discov- eries that could be applied subsequently to the control of epidemics.

Smoking and Lung Cancer Linked The 1964 surgeon general’s report Smoking and Health stated that cigarette smoking is a cause of lung cancer in men (U.S. Department of Health and Human Services, 1964). This report caused a global reaction when it first appeared. The report listed five “criteria of judgment”

Spotlight on Public Health Figures: John Graunt (1620–1674) (continued)

What was his contribution to public health? Long before epidemiology was a science, the field of statistical collection was known as the science of demography. John Graunt is considered the founder of this field, making him the greatest demographer of his time. His Observations was the first organized analysis of vital statistics in London, which had been collected over a 70-year period following the Black Plague. Graunt soon realized that his research could be used to make projections of population figures. Graunt was the first to publish the fact that more boys than girls are born to a population, but the mortality rate for males is far greater. This was the beginning of the use of vital statistics in research and later in epidemiology and public health.

What motivated him? Because the political climate of the time in Britain was one of intellectual advancement, Graunt was strongly encouraged to pursue whatever field interested him. He was fascinated with statistics—in particular, the number of deaths from the Great Plague, which ended before his birth. It was his fascination with those deaths that prompted him to keep a count of them and other incidents of mortality around the nation.

Sources: John Graunt facts. (2010). In Encyclopedia of World Biography. Retrieved from http://biography.yourdictionary.com /john-graunt Sommerville, J. P. (2018). Britain in the seventeenth century. Retrieved from https://faculty.history.wisc.edu/sommerville/351 /england.htm

http://biography.yourdictionary.com/john-graunt

https://faculty.history.wisc.edu/sommerville/351/england.htm

186

Section 6.1 What Is Epidemiology?

that were used to judge the statistical causal significance of the association between smok- ing and lung cancer. These criteria were strength of association, time sequence, consistent relationship upon repeated exposure to smoking, specificity of association, and coherence of explanation. The authors of the report said that accumulated research conducted up to the time of the report tended to support these five criteria of judgment in testing causal associa- tions between this health behavior and morbidity due to lung cancer. Refer to Spotlight on Public Health Figures for more about Luther Terry, the surgeon general when the Smoking and Health report was published.

Spotlight on Public Health Figures: Luther Terry (1911–1985)

Who is Luther Terry? Luther Terry was born in 1911 in Alabama. He earned a bachelor’s degree from Birmingham- Southern College in 1931 and his medical degree from Tulane University, Louisiana, in 1935. John F. Kennedy appointed him U.S. surgeon general in 1961, and shortly after that, Terry made it his life mission to end the nation’s smoking habit. He had been a smoker himself, quitting in 1963. A year later, he delivered a staggering health report on the effects of smoking.

What was the political climate at the time? The 1960s were a time of political change and revolution. The decade ushered in an era of protest, including the civil rights movement, female activism, and protests against the war in Vietnam. The American political system was also shaken when John F. Kennedy was assassinated in 1963. There was a significant amount of unrest. Terry’s tackling of the smoking habit was nothing short of a miracle. Given the unrest and changes occurring in the nation, it was difficult for contemporary politicians to focus on one particular issue. Terry’s points about smoking made it through the noise and began an era of reduced tobacco use.

What was his contribution to public health? Terry was the first to conclusively determine that smoking caused various diseases. Since his report, smoking has been recognized as a risk factor for various diseases, including illnesses from secondhand smoke. Terry’s actions led to a number of other reports on the dangers of smoking. His efforts also spurred the development of the Federal Cigarette Labeling and Advertising Act of 1965, which required warning labels on cigarettes, and the Public Health Cigarette Smoking Act of 1969, which modified that label to read, “Warning: Excessive Cigarette Smoking Is Dangerous to Your Health,” in addition to other terminology currently seen on tobacco package labels. This was a major milestone, as smoking was viewed and advertised as glamorous at the time.

(continued)

Associated Press Luther Terry was a surgeon general. He devoted much of his professional career to reducing tobacco use among the population.

187

Section 6.1 What Is Epidemiology?

Levels of Disease The history of epidemiology portrays the humble beginnings of how humans have discovered diseases and learned to identify and track those diseases. Public health is so intertwined with identifying, tracking, and stopping both infectious and chronic diseases that epidemiology has grown into its own field.

Epidemiologists use a variety of terms to refer to outbreaks. A sporadic outbreak occurs when diseases break out infrequently and irregularly, cluster outbreaks are the constant presence of a disease or infectious agent within a population, and hyperendemic outbreaks are a high level of disease occurrence (CDC, 2012a).

The following terms identify the levels to which disease can spread:

• An epidemic is a widespread occurrence of a disease. Typically, epidemics occur when a disease-causing agent is present in adequate numbers to cause illness or disease across a large geographic area. A good example is seasonal influenza in the United States.

• A pandemic refers to an epidemic that has spread across several countries or conti- nents, affecting significant numbers of people. The Black Death (bubonic plague of the 15th century) was a pandemic.

• An endemic is the constant presence of a disease in a population within a specific geographic area. One example is the Ebola outbreak in West Africa. The disease was prevalent in West Africa, but not elsewhere in the world.

Spotlight on Public Health Figures: Luther Terry (1911–1985) (continued)

What motivated him? His own battle with nicotine addiction was the most motivating factor for Terry. After he quit, he wanted to urge the rest of the nation’s smokers to do the same. In his famous surgeon general’s report Smoking and Health, he conclusively determined that smoking caused cardiovascular disease, emphysema, and lung cancer. He based his conclusions on more than 7,000 peer-reviewed articles on the topic, asserting that they provided more than sufficient evidence to conclude smoking was the cause.

Sources: Campbell, S. L. (2018). Social climate of the 1960s in America. Retrieved from https://classroom.synonym.com/social-climate -1960s-america-22162.html Flannery, M. A. (2017). Luther Terry. Retrieved from http://www.encyclopediaofalabama.org/article/h-1241 Society and life in the 1960s. (n.d.). In English Online. Retrieved from http://www.english-online.at/history/1960s/society-and-change -in-the-sixties.htm U.S. National Library of Medicine. (n.d.). The reports of the surgeon general: The 1964 report on smoking and health. Bethesda, MD: Author. Retrieved from https://profiles.nlm.nih.gov/ps/retrieve/Narrative/NN/p-nid/60

https://classroom.synonym.com/social-climate-1960s-america-22162.html

http://www.encyclopediaofalabama.org/article/h-1241

http://www.english-online.at/history/1960s/society-and-change-in-the-sixties.htm

https://profiles.nlm.nih.gov/ps/retrieve/Narrative/NN/p-nid/60

188

Section 6.1 What Is Epidemiology?

Many current and historical examples of epidemics exist, including recent epidemics of influ- enza, whooping cough, and foodborne illnesses. In order to monitor the presence of epidemic disease, local and state public health departments and the CDC have implemented surveil- lance systems. These involve the periodic reporting of conditions, known as reportable and notifiable diseases, and compiling statistics on their occurrence. When the incidence of a reportable disease such as influenza exceeds statistical limits, this event suggests that an epi- demic is underway.

Epidemiological Measures Public and community health epidemiologists as well as researchers use measures of disease occurrence for a variety of purposes. These include identifying epidemics and other health problems, assessing the effectiveness of prevention programs or interventions, finding health disparities, and showing associations between exposures and health outcomes. Fundamental measures discussed in this section are summarized in Table 6.2.

Table 6.2: Review of common epidemiologic measures

Measure Description

Count Total number of cases of the disease or other health phenomenon being studied

Ratio The relationship between two comparable amounts

Proportion A type of ratio; may be expressed as a percentage, indicating the part of a whole

Prevalence A measure of disease occurrence meaning the total number of cases of disease

Point prevalence The frequency of disease or other condition at a given point in time

Case fatality rate Number of deaths caused by disease among those who have the disease during a specified period

Incidence A count of new cases of disease among a group within a specified period

Rate A measure that includes time as a part of the denominator

Incidence rate A disease’s rate of development within a group during a specified period

Mortality rate The number of deaths during a given year divided by the size of a reference population (or denominator) during the middle of the year in question

General Measures The simplest and most commonly used quantitative measure in epidemiology is a count, which is the total number of cases of the disease or other health phenomenon being studied. One example could be the number of falls in a nursing home during a 1-month period.

A ratio shows the relationship between two comparable amounts. For example, a sex ratio would compare the number of male cases to female cases. A ratio can also be expressed as a value, when the first quantity is divided by the second. For example, the sex ratio of births in the United States is greater than 1, indicating that more boys than girls are born.

189

Section 6.1 What Is Epidemiology?

A proportion is a type of ratio and may be expressed as a percentage, indicating the part of a whole. For example, only a small proportion of graduates fails to find employment.

Prevalence and Incidence Two examples of more complex epidemiologic measures are prevalence and incidence—two terms that are often used incorrectly in the professional literature. Prevalence is a measure of disease occurrence, meaning the total number of cases of a disease. Point prevalence is the frequency of a disease or other condition at a given point in time. For example, the preva- lence of influenza or pertussis is the number of cases of that disease in a population. Point prevalence is a proportion. A point prevalence example would be the number of influenza cases reported on December 15, 2017. If there were 50 cases of influenza reported on that date and there were 1,000 total cases of influenza reported for the winter of 2017–2018, the point prevalence is 50/1,000 for December 15.

Incidence refers to a count of new cases of disease among a group within a specified period. An example of incidence is the number of new cases of influenza that occur within a time interval such as a 2-week period.

Rates A rate is a measure that includes time as a part of the denominator. It consists of a numerator (the frequency of new cases of disease during a specified period) and a denominator, which is a unit size of population. The population could be of a community, a state, or an entire country.

To calculate a rate, one must consider two points in time: the beginning of the period and the end of the period during which the new cases occur. Rates are expressed as numbers per unit size of population (e.g., 10 per 1,000 persons in a population or some other number per unit size of population). Inci- dence, for example, can be expressed as a rate. If 10 people in a given population get the flu in November (which has 30 days), the incidence of flu is 10:30, or 10 cases every 30 days, which is equal to 1 case every 3 days. (Note that prevalence is never a rate.)

An incidence rate describes a disease’s rate of development within a group during a specified period. This rate uses only the frequency of new cases that occur during a time period in the numer- ator. Consequently, individuals who already have the disease are not included in the numerator. The denominator for incidence rates is the population at risk: those who are capable of developing the disease either because they are not immune or for some other reason. An incidence rate includes (1) a numerator—the number of new cases, (2) a

Karen Kasmauski/SuperStock To determine the case fatality rate for a disease, such as hantavirus, which is thought to be spread by deer mice, divide the number of infected patients by the number of infected patient deaths, and then multiply that number by 100.

190

Section 6.2 Disease Surveillance

denominator—the population at risk, and (3) time—the period during which the cases amass. Again, imagine a population of 100 people. During influenza season 2017–2018, 20 people from September through February get sick. In this case, then, the rate of influenza for that season was 20:100, or 1:5.

The mortality rate is the number of deaths during a given year divided by the size of a refer- ence population (or denominator) during the middle of the year in question. For example, if there is a community with 1,000 people and 10 died in 2017, then the mortality rate was 10:1,000, or 1 in 100.

The case fatality rate refers to the number of deaths caused by a disease among those who have the disease during a specified period. To better understand this concept, consider this example: If 50 people became infected with hantavirus and 8 of them died, the case fatality rate would be (8 ÷ 50) × 100 = 16%.

6.2 Disease Surveillance There are several data sources for tracking and monitoring diseases across the United States. These sources provide information on a variety of levels to help monitor the health of the nation. While there are numerous sources available for epidemiology, this section will focus on these key resources: the National Notifiable Diseases Surveillance System, Vital Statistics Reports, morbidity surveys, U.S. Census data, and case registries.

The National Notifiable Diseases Surveillance System By legal statute, physicians and other health care providers must report cases of diseases known as reportable and notifiable diseases. The reporting individuals send the information to local agencies such as health departments, from which it flows to state and federal levels. Notifiable disease reporting at the community level protects the public’s health by ensuring the proper identification and follow-up of cases of disease. These diseases are usually infec- tious and communicable ones that might endanger a population. The reportable and notifi- able diseases in the United States are the following:

• Anthrax • Arboviral diseases • Botulism • Cholera • Hepatitis • Human immunodeficiency virus (HIV) • Meningococcal disease

191

Section 6.2 Disease Surveillance

• Rabies • Salmonellosis • Tuberculosis

Reportable and notifiable disease reports contain official statistics in tabular and graphical form. Statistics are collected and compiled from reports sent by state health departments and U.S. territories to the National Notifiable Diseases Surveillance System, which is operated by the CDC in collaboration with the Council of State and Territorial Epidemiologists.

Vital Statistics Reports The types of information in this category include data collected on vital events (births and deaths) by the vital registration system of the United States. Statistics derived from the vital registration system are computed from the data that are collected routinely on all births and deaths that occur in the United States. For instance, birth certificate data are needed to calculate birth rates. In addition, this information may also contain data about a range of conditions that could affect newborn children, including conditions present during preg- nancy, congenital malformations, obstetric procedures, birth weight, length of gestation, and the demographic background of mothers. Death certificate data in the United States include demographic information about the deceased person and information about the cause of death, including the immediate cause and contributing factors (Friis & Sellers, 2009).

Morbidity Surveys for the Population Morbidity surveys are procedures for collecting information on the health status of a pop- ulation group by using self-administered questionnaires, interviews, and direct examina- tions of participants (Friis & Sellers, 2009). They are designed to determine the frequency of chronic and acute diseases and disability, to obtain measurements of bodily characteristics, to conduct physical examinations and laboratory tests, and to probe other health-related char- acteristics of special concern to those who sponsor the survey. Two examples are the National Health Interview Survey and the Behavioral Risk Factor Surveillance System.

Conducted by the National Center for Health Statistics, the National Health Interview Sur- vey (NHIS) is a household health interview survey that has been collecting data on a broad range of health topics since 1957 (CDC, 2017q).

The Behavioral Risk Factor Surveillance System (BRFSS) is the world’s largest ongoing telephone health survey system. BRFSS began collecting information on risk behaviors and tracking health conditions in the United States in 1984 (CDC, 2017q). Results from the BRFSS have provided data to track a community’s health status, access to health care, and progress toward achieving state and national health objectives.

192

Section 6.2 Disease Surveillance

U.S. Census Data The Census Bureau is an invaluable resource for characteristics of the United States population. Cen- sus data provide information about the denomina- tors used in the calculation of measures of morbid- ity and mortality. Recall that a denominator is the number of parts into which the whole is divided. For example, it is important to know how many people exist in a population to determine mortality and incidence rates. If 10 people died of lung can- cer in a city but the actual population of the city is unknown, then that number doesn’t reveal much. However, if 10 died of lung cancer in a population of 1 million, then we know that the rate is 10:1 mil- lion, which is relatively low. But, if the city under

review has 50 total residents, then 10 deaths out of 50 is significant because 1 in 5 would have died of lung cancer. That would raise a red flag for epidemiologists to investigate such a high rate of lung cancer.

The decennial (10-year) census counts every resident in the country. Census 2010, the most recent census to date, provides detailed information about the country’s entire population, including the variables of age, sex, race, and ethnicity, as well as housing characteristics and household information (United States Census Bureau, 2010). Researchers can extract this information from census data sets in order to describe the sociodemographic and other char- acteristics of a specific community.

Case Registries A registry is a centralized database for collection of information about a particular disease. Registries are used commonly for the compilation of statistical data on cancer, although other types of disease registries exist. Two examples of population-based registries include the Sudden Unexpected Infant Death (SUID) Case Registry and the Sudden Death in the Young (SDY) Case Registry. Epidemiologists use census data combined with registry data to deter- mine significance. If a population of 100 reported 20 deaths to the SUID registry, then the rate is 20:100 or 1:5. However, if those 20 SUID reports came from a population of 1.2 million, the rate is 20:1,200,000 or 1:60,000, a rate that is not nearly as significant as 1 in 5. The SUID registry has 18 monitored sites across the United States (Figure 6.1).

jcamilobernal/iStock/Thinkstock The United States Census occurs every 10 years. The census helps collect valuable information about population variables such as age, sex, and housing characteristics.

193

Section 6.3 Descriptive and Analytic Epidemiology

6.3 Descriptive and Analytic Epidemiology Epidemiology is divided into two categories. Descriptive epidemiology uses descriptive stud- ies or cases to describe the occurrence of disease, and analytic epidemiology uses analytic studies or cases to identify the causes of disease.

Descriptive Epidemiology One of the most useful applications of epidemiology, descriptive epidemiology is the char- acterization of disease occurrence in populations according to classifications by person, place, and time variables. It identifies that something has occurred among a specific group in a specific location, but not necessarily how or why it occurred. It has three broad objectives:

Figure 6.1: SUID Registry monitoring sites

The CDC routinely monitors sudden unexpected infant deaths and sudden death in the young through case registries.

Note: CDC supports SUID monitoring at 18 awardee sites, covering 30% of all SUID cases in the United States.

Source: Adapted from “SUID and SDY Case Registries,” by Centers for Disease Control and Prevention, 2018 (https://www.cdc.gov /sids/CaseRegistry.htm).

AZ NM OK

KSCO

WI MN

IL IA

LA MS AL GA

SC NCTN

IN OH PA

WV VA

MA NH

DE NJ CT

MEVT

States that participate in the SUID Case Registry

States and cities/counties that participate in both the SUID and SDY Case Registries

https://www.cdc.gov/sids/CaseRegistry.htm

194

Section 6.3 Descriptive and Analytic Epidemiology

1. Permit evaluation and comparisons of trends in health and disease among popu- lations or subpopulations (these can include countries and subgroups within countries)

2. Provide a basis for planning, provision, and evaluation of health services in order to facilitate the efficient allocation of resources

3. Identify problems to be studied by analytic methods and to suggest areas that may be fruitful for investigation

Person Variables Person variables are among the most important descriptive epidemiologic variables. Age, one such type of variable, is useful for showing the distribution of health outcomes. One way to classify age is according to 10-year intervals; larger age categories are used as well (e.g., chil- dren and teenagers, young adults, older adults, and the elderly). Age is related to many health outcomes, including morbidity and mortality from infectious diseases, chronic diseases, unin- tentional injuries, and disabilities. For example, chronic diseases (cancer, heart disease, and diabetes) affect the elderly more frequently than the young. Another example of age effects is the tendency for cancer mortality to increase linearly with age. In comparison, the leading cause of mortality among younger individuals is unintentional injuries. From this informa- tion, epidemiologists can see that younger individuals die far more than any other age divi- sion of unintentional injuries. This provides information on “who.” Once this type of data is known, more investigation can commence on the “why.”

Many outcomes in morbidity and mortality reflect sex or gender differences. One example of sex differences is in overall mortality, which is higher for males than for females for all causes. Men also have higher mortality rates than women for cancer of the lung and bronchus. Although men have higher rates of mortality, the reverse is true for morbidity rates, which for many acute and chronic conditions are higher for women than for men. An example of sex differences in deaths from lightning strikes in the United States shows a significantly higher number of male deaths versus female. Figure 6.2 breaks this down not only by gender, but also by age, state, day of the week, and month. The more detail an epidemiologist can acquire, the more information can be determined on potential risk factors, which may include catego- ries such as age, location, and day/week of the month. In some reports, data are broken down by the hour of the day of lightning strike occurrences.

One of the more well-developed systems for classifying race has been implemented by the United States Census Bureau. This system uses five categories plus a “some other race” cat- egory. Respondents to the 2010 Census questionnaire could self-identify their racial group membership and also select more than one racial group. The racial categories used in the 2010 questionnaire were White, Black or African American, American Indian or Alaska native, Asian, and native Hawaiian or other Pacific Islander. Respondents were also requested to indicate whether they were of Hispanic origin; persons of Hispanic origin could be of any race. For example, Hispanics can also be classified as Black, White, or Asian depending on the race with which they identify themselves.

195

Section 6.3 Descriptive and Analytic Epidemiology

Figure 6.2: Lightning strike fatalities demographics

Gender differences are often visible when examining morbidity and mortality rates, such as those presented in these lightning strike fatality demographics. Epidemiologists can use data to determine patterns and identify potential risk factors.

* Due to rounding, percentages may not add up to 100%. ** Known fatalities to date. Note: Monthly averages are based on a 10-year average from 2007 to 2016. Due to rounding, the sum of monthly averages may not meet the yearly averages.

Source: Adapted from “U.S. Lightning Deaths in 2017,” by National Weather Service, 2017 (http://www.lightningsafety.noaa.gov /fatalities.shtml).

1 (6%)

15 (94%)

Age*

2017 Lightning Fatality Demographics

Day of week*

Jan

Feb

Mar

Apr

May

Jun

Jul

Aug

Sep

Oct

Nov

Dec

Year

2017 0

16**

16 0

15 0

14 0

13 0

12 0

11 0

10 0

09 0

08 1

07 0

10 Yr Avg 0

31*

By month

GenderState 0–9

10–19

20–29

30–39

40–49

50–59

60–69

70–79

80–89

1 (6%)

3 (19%)

6 (38%)

2 (13%)

0 (0%)

2 (13%)

0 (0%)

1 (6%)

2 (13%)

4 (25%)

1 (6%)

2 (13%)

1 (7%)

Sun

Mon

Tues

Wed

Thu

Fri

Sat

http://www.lightningsafety.noaa.gov/fatalities.shtml

196

Section 6.3 Descriptive and Analytic Epidemiology

The United States is becoming increasingly diverse with respect to race and ethnicity. Racial characteristics are associated with substantial variations in several indices of morbidity and mortality, such as birth rates. One application of data on racial differences in health outcomes is to identify health disparities and develop community programs for groups at greatest risk of adverse health outcomes. For example, statistics in Figure 6.3 show that Black people are killed far more through homicide than all other races. Of interest, through 2014, there was a general decline in homicide trends for non-Hispanic White, non-Hispanic Black, and Hispanic populations (Minino, 2017). Then, a significant increase occurred between 2014 and 2015. In 2015, homicide rates were 5.7 deaths per 100,000 for the total population, 20.9 for non- Hispanic Blacks, 4.9 for Hispanics, and 2.6 for non-Hispanic Whites (Minino, 2017). Epidemi- ologists use this type of information to track health disparities and racial discriminations. In light of the recent homicides of Black people that have been in the news over the past several years, this information becomes critical not only for health research but also to assist law enforcement in determining a focus of racial tensions.

Figure 6.3: Age-adjusted rates for homicides by race/ethnicity, U.S. 1999–2015

Displaying the data in this graph by race/ethnicity makes it evident that homicides of Black individuals spiked between 2014 and 2015. Epidemiologists can use this type of data to identify racial differences in health outcomes, identify risk factors, and share their research with other organizations to help communities address potential issues.

Source: Adapted from “QuickStats: Age-Adjusted Rates for Homicides by Race/Ethnicity – United States, 1999–2015,” by A. Minino, 2017, Morbidity and Mortality Weekly Report, 66(31), 839 (https://www.cdc.gov/mmwr/volumes/66/wr/mm6631a9.htm).

1999

2001

Year

R at

e p

er 1

0 0,

0 0

0 st

an d

ar d

p o

p u

la ti

o n

2003 2005 20092007 2011 2013 2015

All races/origins White, non-Hispanic Black, non-Hispanic Hispanic

The person variable of socioeconomic status is defined by type of occupation, income level, and amount of formal education. Individuals who have higher income and education levels and are employed in occupations such as the learned professions have a higher socioeco- nomic status on average than those who are lower with respect to these three characteristics. Low socioeconomic status is associated with higher rates of morbidity from infectious and

https://www.cdc.gov/mmwr/volumes/66/wr/mm6631a9.htm

197

Section 6.3 Descriptive and Analytic Epidemiology

chronic conditions and generally a greater frequency of adverse health outcomes in compari- son with high socioeconomic status. The association between low socioeconomic status and health disparities is discussed in Chapter 2.

Marital status—whether one is single or nonmarried, married, or unmarried but living with a partner—has been found to be a factor in health. Rates of morbidity and mortality vary according to marital status, and evidence has suggested that married adults tend to be health- ier overall than nonmarried adults.

Another person variable is immigrant status. Migration is related to patterns of chronic dis- ease and life expectancy. When persons immigrate to a new country, often they acquire the health-related characteristics of the inhabitants of the host country, perhaps as a result of changes in diet and lifestyle. The process of adopting the cultural practices of the host coun- try is known as acculturation. For example, Japanese immigrants (a low-risk group for coro- nary heart disease in their home country) who relocated to the United States and became acculturated had a higher incidence of coronary heart disease than immigrants who were less acculturated. The health-related aspects of immigration are significant for the many Ameri- can communities that have large immigrant populations.

Finally, religious background affects personal lifestyle characteristics and consequently is associated with variations in morbidity and mortality. For example, the adherents of some religious groups (e.g., Seventh-day Adventists who practice vegetarianism) have lower rates of cancer and chronic disease mortality in comparison with the general population.

Place Variables The variable of place includes international, within-country, regional, and local patterns in health outcomes. Within-country variations can be subdivided further into urban and rural comparisons. The United States Census Bureau defines urbanization by using geographic locations called metropolitan statistical areas (MSAs), which are urbanized areas based on the number of inhabitants in a particular area. An MSA is a geographical region with a rela- tively high population density at its core and close economic ties throughout the area. It is typically centered around a single city that wields substantial influence on the region (e.g., Chicago).

Conditions that show regional variations include Lyme disease, obesity, HIV infections, and many others. Among the causes of these variations are differences in climate, local environ- mental conditions, socioeconomic status, lifestyle and cultural factors, and availability of health care services. For example, local physical environmental conditions may be conducive to the survival of microbes and disease vectors. By examining location, epidemiologists can begin identifying similar characteristics within a certain region to determine the cause or fac- tors connected with an outbreak. Another example is the distribution of obesity in the United States. Figure 6.4 shows that obesity is concentrated in states in the deep South, including Arkansas, Louisiana, Mississippi, and Alabama. This information shows epidemiologists a regionalized view of factors that can contribute to obesity. Knowing about these concentra- tions in specific Southern states, epidemiologists can examine social, environmental, and economic factors associated with those areas—such as culture, lifestyle, climate, and health resources—to determine how to tackle the problem.

198

Section 6.3 Descriptive and Analytic Epidemiology

Time Variables The epidemiologic variable of time includes the following time trends: secular (long term), seasonal, clustering, and epidemic. Investigations of time variation help to identify changing patterns in morbidity and mortality, demonstrate the success of preventive efforts for chronic and other diseases, and signal the occurrence of epidemics. Secular trends are the gradual change in the frequency of a disease over long periods of time. A great example is polio. After the introduction of the vaccination against the disease, the frequency over time dropped. This is an example of a time epidemic. For disease investigations, epidemiologists can look at the incidence rates over a period before or after an incident to determine if the incident was the cause. Chapter 2 discussed this in relation to the Flint water crisis. Over time, there was an increase in waterborne illnesses after Flint switched its water source to the Flint River.

Cycle fluctuations also occur in terms of time. Typically, cyclical fluctuations reflect seasonal patterns in the occurrence of diseases such as allergies or seasonal influenza. Tracking a dis- ease by time can provide insights into an epidemiological investigation that can lead to a cause and, eventually, a resolution.

Analytic Epidemiology Analytic epidemiology is the process of using data gathered by descriptive experts to study patterns suggesting causes of diseases and other health conditions. While descriptive

Figure 6.4: Adult obesity by state, 2016

Obesity is greater in the Southern states, giving epidemiologists a geographic focal point to review factors in the area that might influence obesity.

Source: “Adult Obesity in the United States,” by Trust for America’s Health and Robert Wood Johnson Foundation, 2017 (https:// stateofobesity.org/adult-obesity/). Copyright 2017, Robert Wood Johnson Foundation. Adapted with permission from the Robert Wood Johnson Foundation.

AZ NM OK

KS CO

WI MN

IL IA

LA MS AL GA

NCTN

IN OH PA

WV VA

MA NH

DE NJ CT

MD DC

MEVT

20%–24.9%

25%–29.9%

30%–34.9%

35%+

https://stateofobesity.org/adult-obesity/

199

Section 6.4 Applying Epidemiology to Community Health

epidemiology tells who, what, where, and when, analytic data are used to discover the “how” and “why” factors in investigation. Why did a disease occur? How did it happen? Analytic epi- demiologists might ask questions like:

• Do toxic chemicals cause cancer? • What is the effect of diet on heart disease and diabetes?

For the most part, analytic epidemiologic studies use observational studies such as case- control studies and cohort studies. In a case-control study, subjects are enrolled on the basis of whether they have had the disease to determine if there is an association between the disease and potential causal factors. Cohort study subjects are enrolled based on their mem- bership in a controlled subpopulation or category of the population (e.g., racial groupings). Analytic epidemiologists also make use of intervention studies, which test the effectiveness of a particular program or product.

An example of an analytic epidemiological study may involve taking a group of students and measur- ing their BMI and dietary habits for breakfast. Then, an intervention that includes a healthier option for the school breakfast program is added, and the researchers monitor BMI and dietary habits again after a period of time to see if changes occurred. They are seeking to determine whether the inter- vention makes a difference.

It is important to realize that one of the goals of epi- demiologic research is to portray the frequency and patterns of disease occurrence in the population and link them with specific exposures. An exposure (sometimes called an independent variable) is the potential causal factor in an epidemiologic study. Exposures can include many factors, including environmental, lifestyle, and economic categories, as well as medical treatments and genetic traits. In order to examine the occurrence of disease outbreaks and environmentally caused diseases in the population, the field of epidemiology uses several characteristic study designs. Because this text does not focus on public health research, study methodologies will not be discussed. For investigative techniques used by epidemiologists, it is important to have a basic but not necessarily a full working knowledge of the terminology.

6.4 Applying Epidemiology to Community Health The British epidemiologist Jerry Morris wrote the Uses of Epidemiology in 1957. This classic book was a product of a time in Britain’s history when there was growing recognition of the importance of social and economic factors in health. Morris’s visionary account of the seven uses of epidemiology (Table 6.3) remains relevant to modern community health issues.

Wavebreakmedia/iStock/Thinkstock Data gathered during an analytic epidemiology study that examines how breakfast affects children’s BMI measurements could be used to determine if changes to breakfast dietary habits affect BMI measurements.

200

Section 6.4 Applying Epidemiology to Community Health

Table 6.3: The seven uses of epidemiology*

1. Study the history of the health of populations

2. Diagnose the health of the community

3. Study the working of health services

4. Estimate individual risks

5. Identify syndromes

6. Complete the clinical picture

7. Search for causes

* Green shaded items are discussed in the text.

Source: Adapted from “Uses of Epidemiology,” by J. N. Morris, 2007, in International Journal of Epidemiology, 36, 1165–1172.

Four Uses of Epidemiology Four of Morris’s seven uses of epidemiology are directly relevant to the field of community and public health. The other three revolve around medical care and health services, which are outside of the scope of public health. These four uses of epidemiology in the community and public health realm include studying historical trends in health, describing the health of the community, assessing individual risks, and examining the causes of disease in the community.

Studying Historical Trends The historical use of epidemiology refers to the study of time trends in health and illness. One example of the historical use of epidemiology is the study of changes in disease frequency, or secular trends, over long time periods (Morris, 2007). In general, chronic conditions have replaced acute infectious diseases as the major causes of morbidity and mortality in contem- porary industrialized societies. With respect to the health of the community, it is apparent that chronic health conditions such as obesity, nutritional deficiency, and heart disease have now become major challenges to the population’s health, although infectious diseases remain important causes of morbidity and mortality.

Describing the Health of the Community Describing the health of a community pertains to the identification of demographics, behav- ioral risks, and environmental factors in a community that affect health-related outcomes. Using epidemiology to describe the health of a community relates to Healthy People 2020’s overarching goals:

1. Attain high-quality, longer lives free of preventable disease, disability, injury, and premature death.

2. Achieve health equity, eliminate disparities, and improve the health of all groups. 3. Create social and physical environments that promote good health for all. 4. Promote quality of life, healthy development, and healthy behaviors across all life

stages. (HealthyPeople.gov, 2018)

201

Section 6.4 Applying Epidemiology to Community Health

Categories of variables that affect community health include demographic and social compo- sition, community infrastructure, and environmental characteristics. Refer to A Closer Look for more detailed information regarding these variables and how they could affect the health of a community.

A Closer Look: Variables and Community Health

Demographic and social variables: Those in poorer communities often suffer greater ill health and shorter life expectancies. These variables include the following:

• Age and sex distribution • Socioeconomic status • Family structure, including marital status and number of single-parent families • Racial, ethnic, and religious composition

Variables related to community infrastructure: Poorer facilities and physical environments directly and indirectly affect health behaviors and subsequently the risk of developing chronic disease. Some of these variables include the following:

• Availability of social and health services, including hospitals, emergency rooms, and community clinics

• Quality of housing stock, including presence of lead-based paint and asbestos • Social stability (residential mobility), such as community policing and employment

opportunities

Health-related outcome variables: These outcome variables measure dimensions of health status in the community. These include the following:

• Homicide and suicide rates • Infant mortality rate • Mortality from selected conditions • Scope of chronic infectious diseases • Alcoholism and substance abuse rates • Teenage pregnancy rates • Occurrence of sexually transmitted diseases • Birthrate

Environmental variables: The physical environment can provide factors that adversely affect or improve the health of a population. These include the following:

• Air pollution from stationary and mobile sources • Access to parks and recreational facilities • Availability of clean water • Availability of markets that supply healthful groceries • Number of liquor stores and fast food outlets • Nutritional quality of foods and beverages vended to schoolchildren • Soil levels of radon

Source: Adapted from Epidemiology for Public Health Practice (4th ed.), by R. H. Friis and T. A. Sellers, 2009, Sudbury, MA: Jones & Bartlett.

202

Section 6.4 Applying Epidemiology to Community Health

Estimating Individual Risk Epidemiologic methods are linked closely with the field of risk assessment, which seeks to identify and analyze potential hazards and fallout should a hazard occur. In environmental risk assessment, epidemiological studies can be used to estimate potential health risk factors in the environment and how they are related, or associated. An example of this is projections about the probability of developing lung cancer among smokers versus nonsmokers.

Searching for Causes Among the most important uses of epidemiology is uncovering the causes, or etiology, of disease. When describing the causes of disease, epidemiologists employ exposures and risk factors. Epidemiologic research explores possible associations among exposures and health outcomes through the appli- cation of appropriate study designs, measures, and other methodologies. The correct study design can provide insight into whether an observed associa- tion is due to chance or is a causal association. A classic example of using epidemiology to study the etiology of disease is research on smoking and lung cancer. Smoking is a risk factor for lung cancer, and understanding the disease etiology will help a cli- nician prescribe the most effective preventive or therapeutic interventions.

The findings of epidemiologic research also aid in policy development by providing meth- odological skill sets and contributing to the fund of information needed to guide informed decision-making. Backed by supportive, evidence-based epidemiologic research, policy actors are able to introduce health-related policies that protect the community’s health. Legislators and government officials are charged with the responsibility of creating policies and enacting and enforcing many laws that have substantial impacts on public health. Moreover, following the adoption of desired health policies and programs, public health professionals can apply epidemiologic methods to the evaluation of program effectiveness.

An example of health policies that have been implemented in a community comes from the California city of Long Beach, which enacted a no-smoking ordinance, LBMC 8.68 (City of Long Beach, 2013). Within the past decade, the city of Long Beach has created smoke-free indoor public areas, including bars and restaurants, pool halls, office buildings, and elevators and laundry rooms. No-smoking areas include parks, beaches, bus stops, farmer’s markets, and areas within 20 feet of state, county, and city buildings. The rationale for this ordinance stemmed from evidence that linked adverse health effects with secondhand cigarette expo- sure. In 2011, the state of California passed Senate Bill (SB) 332. This bill “prohibits any per- son from smoking a cigarette, cigar, or other tobacco-related product, or from disposing of cigarette butts, cigar butts, or any other tobacco-related waste, within a playground” (Califor- nia Legislation Senate Bill 332, 2011, para. 1). Furthermore, SB 332 authorizes landlords to prohibit smoking on, in, or near their buildings.

Kallista Images/SuperStock The pink in this CT scan represents cancerous growth in the lungs of a 54-year-old man. In the quest to discover the etiology of lung cancer, epidemiological researchers could identify smoking as a major risk factor.

203

Section 6.4 Applying Epidemiology to Community Health

Epidemiological Investigations There are 13 critical steps that epidemiologists take when working on an outbreak investi- gation (Table 6.4). Investigations take time because epidemiologists must be thorough and include every element possible for a solid conclusion. If they are wrong, then the disease continues to spread and lives are at stake.

Prepare for Fieldwork Field investigators must have the appropriate scientific knowledge, supplies, and equipment to go into the field prior to embarking on an investigation outside of the office. If material samples such as blood samples or soil samples are required, the investigator must take collection units as well as storage cases for these elements. Other such equipment may include personal protec- tive devices such as gloves, face masks, and heavy clothing to protect themselves against con- tracting a viral agent. Finally, epidemiologists need to have an action plan before they enter the field. Typically, managers and operations directors will help the field worker with such a plan; however, the field investigator must be a good manager as well. Most investigations are done as a team. Each role must be clearly outlined and differentiated so there is no duplication of efforts. This will also help to ensure every element on the investigation plan is covered.

Table 6.4: Epidemiological steps of an outbreak investigation 1. Prepare for fieldwork

2. Establish the existence of an outbreak

3. Verify the diagnosis

4. Construct a working case definition

5. Find cases systematically and record information

6. Perform descriptive epidemiology

7. Develop hypotheses

8. Evaluate hypotheses epidemiologically

9. As necessary, reconsider, refine, and re-evaluate hypotheses

10. Compare and reconcile with laboratory and/or environmental studies

11. Implement control and prevention measures

12. Initiate or maintain surveillance

13. Communicate findings

Source: Adapted from “Lesson 6: Investigating an Outbreak,” by Centers for Disease Control and Prevention, in Principles of Epidemiology in Public Health Practice: An Introduction to Applied Epidemiology and Biostatistics (3rd ed.), 2016 (https://www.cdc.gov/ophss/csels/dsepd/ss1978 /lesson6/section2.html).

https://www.cdc.gov/ophss/csels/dsepd/ss1978/lesson6/section2.html

204

Section 6.4 Applying Epidemiology to Community Health

Establish the Existence of an Outbreak One of the first tasks of the field investigator is to determine if there truly is an outbreak and/ or epidemic. This is typically done by looking at descriptive information that has been pre- sented, such as numbers of people affected, ages, gender, and geographic locations. As noted earlier, an epidemic is when the occurrence of a disease is more than expected in a given area or among a specific population. An epidemic may also occur if there is an outbreak of an unknown illness that has yet to be diagnosed. While the latter may not be a formal epidemic, caution is always advised, as emerging diseases could cause epidemics or even pandemics. A good example is the Zika virus, a newly emerging disease that U.S. epidemiologists investi- gated in order to protect the nation’s health. Once the establishment of an outbreak is final- ized, a field investigation is launched.

Verify the Diagnosis In many cases, this step is performed at the same time as Step 2 (establishing that an out- break actually occurred). Investigators typically review any clinical or laboratory results and consult with laboratory technicians on accuracy of testing and final outcomes. Then, inves- tigators will often visit one of the individuals with the disease, if that person is alive. Talking directly with the victims of an outbreak can help provide a better understanding of how and where they were affected.

Construct a Working Case Definition A case definition is a standard set of criteria for deciding whether someone actually has the illness or disease under review. This means that the definition would include the descriptive ele- ments of person, place, and time, as described earlier in this chapter. In addition, the symptoms of the victim are noted and compared with those of others who have the disease to review simi- larities or differences (if any). For example, the case definition of meningococcal disease (menin- gitis) is listed as “an illness with sudden onset of fever (>38.0 degrees C) and one or more of the following: neck stiffness, altered consciousness, other meningeal sign or petechial or puerperal rash” (CDC, 2016f, Step 4, para. 4). Any time an epidemiologist goes out to investigate any out- break, the case definition must be determined and understood so as to identify all victims.

Find Cases Systematically and Record Information With a solid case definition, it is much easier for epidemiologists to identify new cases when they are in the field. Usually, this case definition assists health care providers such as doctors and nurses in the identification process. Often, this definition is publicized to the media to help locate individuals who may not realize they have the disease. Recording the informa- tion becomes more important for documentation purposes and may help with the location of Patient 0, or the first person to have contracted the disease. Information collected from each person who falls within the case definition includes name, address, phone number, age, sex, race, occupation, signs and symptoms displayed, and risk factor information. Risk factor information includes where patients were over a specific time frame, what they ate or drank, and what elements they were exposed to, such as pollutants. This information can provide common variables among all affected to help home in on the root cause or to find Patient 0.

205

Section 6.4 Applying Epidemiology to Community Health

Perform Descriptive Epidemiology In this step, the investigator writes and maps out the information that has been given during the prior stages. This is a criti- cal step because after summarizing the col- lected data, the epidemiologist has a com- prehensive characterization of the outbreak, trends over time, location of the disease outbreak, and who specifically was affected. This characterization often provides clues about the cause of the disease. The measles outbreak at Disneyland in California was discovered based on descriptive epidemiol- ogy. Every person affected by this outbreak had a connection to the amusement park, but it wasn’t until the field epidemiologist summarized all of the investigative materi- als that this connection was discovered.

Develop Hypotheses Hypotheses, proposed explanations that are a starting point for further investigation, are gen- erated in a number of ways. At this stage, the investigator considers the disease characteris- tics itself. If it is a known agent, such as measles, the investigator will review the disease’s normal transmission, the hosts, the agents, and the risk factors for contracting the disease. In the Disneyland case, descriptive epidemiology was able to pinpoint where the exposure occurred, and the hypothesis assisted in determining why it happened: lack of herd immunity.

Evaluate Hypotheses Epidemiologically Sometimes, more than one hypothesis is determined. At this time, the field investigator must critically examine the hypotheses and evaluate the plausibility of each one. In the measles out- break at Disneyland, the investigators likely considered the plausibility of a measles outbreak. They would have taken childhood vaccinations into consideration as well as laws and opt-out laws. Because of information available to epidemiologists, such as the numbers of unvacci- nated children, and percentages of children congregating at Disneyland, epidemiologists could have actually predicted an outbreak at some point. However, because measles had been con- sidered eradicated, it was not on anyone’s radar at the time of the epidemic. In other cases, circumstances are not as straightforward. The investigators must look at observed patterns from past outbreaks (if they exist) and how the disease transmission occurs. By comparing past experiences, epidemiologists can focus on the plausibility of their hypotheses. Sometimes there is time to conduct a study to further test the hypotheses that are suggested; however, when dealing with viral diseases (diseases with a fast spread rate or that are deadly), there is often no time. Researching the literature for past experiences is the best option at this point.

Marc Rasmus/imageBROKER/SuperStock Descriptive epidemiology helped investigators determine that all of the patients affected by the 2014–2015 measles outbreak had recently visited the Disneyland Resort in California.

206

Section 6.4 Applying Epidemiology to Community Health

Reconsider, Refine, and Re-evaluate Hypotheses In some cases, when the hypotheses lead to a roadblock, epidemiologists need to start over and perhaps look in a different direction. In analytical epidemiology, this is often the case. Sometimes, meeting with patients again, visiting their homes, and rehashing details can reveal new informa- tion. Field investigators must continuously focus on the questions that remain unanswered in order to get to the root problem. It is much like unpeeling an onion one layer at a time.

Compare and Reconcile With Laboratory and Environmental Studies Once an epidemiologist has a working hypothesis, the next step is to return to the laboratory or bring in scientific specialists to verify and test the hypothesis. Consider John Snow and the cholera outbreak in London. If waterborne illness is suspected, then a laboratory can confirm that information. In Snow’s time, individuals were clearly affected with cholera, so he would have been the one to suggest testing the main water source (the Broad Street Pump). This would have been solid proof that everyone who drank the water coming from that pump was getting sick, causing the outbreak. As noted in Chapter 1, there were no tests of the pump; however, had such an outbreak occurred today, laboratory testing would have taken place during this phase of the process.

Implement Control and Prevention Measures At this stage, epidemiologists have now identified the disease and the cause and understand what is needed to control it. The goal is to control the outbreak so that no other individu- als become infected. In John Snow’s time, the control was the removal of the pump handle. Today, if there were a cholera outbreak, the entire water source would be shut off. The point is the same: control and stop the outbreak. Then, put prevention measures into place. Dur- ing the Flint, Michigan, water crisis, the control was to stop using water from the Flint River. The prevention was the return to using the Detroit water system. Some interventions are not that easy to realize. Blocking the transmission of an infectious disease seems easier in today’s world; however, what about the epidemic of obesity? There is not a one-size-fits-all control and prevention measure for this type of epidemic. That is why the crisis still exists today.

Initiate or Maintain Surveillance Once control and prevention measures have been implemented, actions must be monitored to ensure they worked. In the case of the Flint, Michigan, water crisis, the water system from Detroit was reinstated, but the water quality is still being monitored for disease. In addition, cleanup efforts of the Flint River are part of the surveillance project. It is also important to note that even if a root cause is found and a control is put into place, that does not mean the disease is eradicated. For the Disneyland measles outbreak, epidemiologists are still monitor- ing the nation’s incidence of the disease beyond California. It is easier to halt a disease when it is discovered early than to wait until it becomes a pandemic.

207

Summary & Resources

Communicate Findings The last step involves communicating the investigation results and protective measures to all stakeholders. Stakeholders include those affected by the disease, those who may have been in the community at the time of the outbreak but did not contract the disease, the health care providers who treated the illnesses, and outside communities. Investigators put together a full written report on their investigation and their final tasks for control and prevention. From there, it is typically the public relations personnel from the investigating agency (typically the CDC) who put together the communications to the public and others who need to have the details.

Now that the 13 steps of the epidemiological investigation process have been outlined, exam- ine the reading located at the end of this chapter to follow an actual case where the steps were used. Can you spot the process of the 13 steps along the way?

Summary & Resources

Chapter Summary Epidemiology is a fundamental discipline of public health focusing on the study of diseases and disease outbreaks. As early as the 400s BCE, humans were showing signs of understanding the interconnectedness of the environment and disease in human beings. Hippocrates wrote that the seasons, the influence of the sun, the quality of water, and the elevation at which people live are factors in health. The field became more recognizable after the work of John Snow on the mitiga- tion of cholera in London in the mid-1800s. Epidemics are widespread occurrences of a disease, such as annual influenza. An endemic is specific to a population, such as the Ebola outbreak in West Africa. Pandemics are diseases that rage across several countries.

Public health professionals have a variety of tools that help with disease surveillance. These sys- tems provide data on incidence rates of a plethora of diseases as well as statistics such as birth and death rates. U.S. Census data is used for more than just counting the population. The information is also used by epidemiologists to help identify descriptive statistics such as gender, race, and age across the country. Other sources of data used for disease surveillance include the National Notifi- able Diseases Surveillance System, the National Health Interview Survey, and case registries.

Descriptive epidemiology characterizes the disease occurrence by person, place, and time variables (i.e., the who, what, where, and when concepts). Analytic epidemiology takes these descriptive elements and analyzes them to determine trends or patterns of a disease within a population (i.e., the how and why concepts). Using this information, epidemiologists convert data to various proportions and rates. The measures used to determine statistical data and its significance in a population include count, ratio, proportion, prevalence, and incidence. There are seven uses of epidemiology, four of which are particularly relevant to public health: studying historical trends, describing the health of a community, estimating individual risk, and searching for causes.

208

Summary & Resources

Epidemiological sources and measurements are applied to actual investigations of diseases and outbreaks. There are 13 steps followed by all epidemiological investigators: prepare for fieldwork, establish the existence of an outbreak, verify the diagnosis, construct a working case definition, find cases systematically and record information, perform descriptive epide- miology, develop hypotheses, evaluate those hypotheses, refine the hypotheses, compare and reconcile with laboratory tests and/or environmental studies, implement control and pre- vention measures, maintain or initiate surveillance, and then communicate the findings. The process seems tedious; however, it is effective in determining causes and resolutions.

Critical Thinking and Review Questions

1. Describe the field of epidemiology and its importance to public health. 2. Describe two surveillance systems that epidemiologists use to obtain information to

track diseases. 3. Consider the annual influenza strain in the United States, using the steps of the epi-

demiological investigation. How would you determine the spread of the virus? 4. What would the Behavioral Risk Factor Surveillance System be used for? 5. Explain John Graunt’s role in vital statistics. 6. State the seven uses of epidemiology and then describe the four key uses outlined in

this chapter. Why are they important for epidemiologists? 7. Consider your own community. What types of determinants in the environment

would cause poor health outcomes? What about positive health outcomes? 8. What types of diseases would epidemiologists be interested in investigating by con-

currently factoring in variables of both time and age? 9. In the “11 blue men” reading, describe the key symptoms of the men. How did

knowledge of these symptoms help with the investigation? 10. In the “11 blue men” reading, what was the final culprit, and how was it finally

discovered?

Additional Resources

Principles of Epidemiology in Public Health Practice, Third Edition

https://www.cdc.gov/ophss/csels/dsepd/ss1978/index.html This source offers a self-guided course on epidemiology from the Centers for Disease Control and Prevention.

Survey and data collection systems at the Centers for Disease Control and Prevention

https://www.cdc.gov/nchs/surveys.htm Review national surveys collected routinely by the CDC.

https://www.cdc.gov/ophss/csels/dsepd/ss1978/index.html

https://www.cdc.gov/nchs/surveys.htm

209

Summary & Resources

Key Terms analytic epidemiology The process of using data gathered by descriptive experts to study patterns suggesting causes of diseases and other health conditions.

Behavioral Risk Factor Surveillance Sys- tem The world’s largest ongoing telephone health survey system. BRFSS began collect- ing information on risk behaviors and track- ing health conditions in the United States in 1984.

case definition A standard set of criteria for deciding whether someone actually has the illness or disease under review.

case fatality rate The number of deaths caused by disease among those who have the disease during a time period.

case-control study A type of study where subjects are enrolled on the basis of whether they have had a disease to determine whether there is an association between the disease and exposure.

cluster outbreaks A constant presence of a disease or infectious agent within a population.

cohort study A type of study where sub- jects are enrolled based on their member- ship in a controlled subpopulation or cat- egory of the population.

count The total number of cases of the disease or other health phenomenon being studied.

descriptive epidemiology The character- ization of disease occurrence in populations according to classifications by person, place, and time variables.

epidemiology The study of the occurrence and distribution of illnesses, injuries, and diseases in specific populations.

exposure The potential causal factor in an epidemiologic study (sometimes called an independent variable).

hyperendemic outbreaks Outbreaks that involve a high level of disease occurrence.

incidence A count of new cases of disease among a group within a specified period.

Flu activity and surveillance at the Centers for Disease Control and Prevention

https://www.cdc.gov/flu/weekly/fluactivitysurv.htm This source offers weekly reports on influenza surveillance activities from epidemiologists at the CDC.

Health in disadvantaged neighborhoods

https://youtu.be/e48K4RN2nrI Watch this California Newsreel video for a closer look at how one’s neighborhood can influ- ence his or her overall health.

https://www.cdc.gov/flu/weekly/fluactivitysurv.htm

https://youtu.be/e48K4RN2nrI

210

Summary & Resources

incidence rate Describes a disease’s rate of development within a group during a speci- fied period.

intervention studies Types of studies that test the effectiveness of a particular program or product.

morbidity surveys Means of collecting information on the health status of a popula- tion group by using self-administered ques- tionnaires, interviews, and direct examina- tions of participants.

National Health Interview Survey A household health interview survey that has been collecting data on a broad range of health topics since 1957. A type of morbidity survey.

Patient 0 The first person to have con- tracted a disease.

point prevalence The frequency of disease or other condition at a given point in time.

prevalence A measure of disease occur- rence meaning the total number of cases of disease.

proportion A type of ratio. May be expressed as a percentage, indicating the part of a whole.

rate A measure that includes time as a part of the denominator.

ratio The relationship between two compa- rable amounts.

secular trends Changes in disease fre- quency over long periods of time.

sporadic outbreak An outbreak of a dis- ease that occurs infrequently and irregularly.

211

Reading

Reading The following story is an actual case that occurred in 1944. The text is written in the first person by an epidemiologist (Berton Roueché) investigating the case of 11 men who came to a local hospital with unusual blue coloration. It provides a real-world example of how the steps of an epidemiological investigation are followed and how conclusions are drawn along the way. These steps are critical in order to come to an accurate conclusion about a potential epidemic or health crisis in a community.

11 BLUE MEN:

AN EPIDEMIOLOGICAL INVESTIGATION

At about eight o’clock on Monday morning, September 25, 1944, a ragged, aimless old man of 82 collapsed on the sidewalk on Dey Street, near the Hud- son Terminal. Innumerable people must have noticed him, but he lay there alone for several minutes, dazed, doubled up with abdominal cramps, and in an agony of retching. Then a policeman came along. Until the policeman bent over the old man, he may have supposed that he had just a sick drunk on his hands; wanderers dropped by drink are common in that part of town in the early morning. It was not an opinion that he could have held for long. The old man’s nose, lips, ears, and fingers were sky-blue. The policeman went to a telephone and put in an ambulance call to Beekman-Downtown Hospital, half a dozen blocks away. The old man was carried into the emergency room there at 8:30. By that time, he was unconscious and the blueness had spread over a large part of his body. The examining physician attributed the old man’s morbid color to cyanosis, a condition that usually results from an insufficient supply of oxygen in the blood, and also noted that he was diarrheic and in a severe state of shock. The course of treatment prescribed by the doctor was conventional. It included an instant gastric lavage, heart stimulants, bed rest, and oxygen therapy. Presently, the old man recovered an encouraging, if pain- ful, consciousness and demanded, irascibly and in the name of God, to know what had happened to him. It was a question that, at the moment, nobody could answer with much confidence.

For the immediate record, the doctor made a free-hand diagnosis of carbon- monoxide poisoning—from what source, whether an automobile or a gas pipe, it was, of course, pointless even to guess. Then, because an isolated instance of gas poisoning is something of a rarity in a section of the city as crammed with human beings as downtown Manhattan he and his colleagues in the emergency room braced themselves for at least a couple more victims. Their foresight was promptly and generously rewarded. A second man was rolled in at 10:25. Forty minutes later, an ambulance drove up with three more

212

Reading

11 Blue Men: An Epidemiological Investigation (continued)

men. At 11:20, two others were brought in. An additional two arrived during the next 15 minutes. Around noon, still another was admitted. All of the nine men were also elderly and dilapidated, all had been in misery for at least an hour, and all were rigid, cyanotic, and in a state of shock. The entire body of one, a bony, 73-year-old consumptive named John Mitchell, was blue. Five of the nine, including Mitchell, had been stricken in the Globe Hotel, a sunless, upstairs flophouse [cheap hotel] at 190 Park Row, and two in a similar place, called the Star Hotel, at 3 James Street. Another had been found slumped in the doorway of a condemned building on Park Row not far from City Hall Park, by a policeman. The ninth had keeled over in front of the Eclipse Cafeteria, at 6 Chatham Square. At a quarter to seven that evening, one more aged blue man was brought in. He had been lying, too sick to ask for help, on his cot in a cubicle in the Lion Hotel, another flophouse, at 26 Bowery, since 10 o’clock that morning. A clerk had finally looked in and seen him.

By the time this last blue man arrived at the hospital, an investigation of the case by the Department of Health, to which all outbreaks of an epidemiologi- cal nature must be reported, had been under way for 5 hours. Its findings thus far had not been illuminating. The investigation was conducted by two men. One was the Health Department’s chief epidemiologist, Dr. Morris Green- berg, a small, fragile, reflective man of 57, who is now acting director of the Bureau of Preventable Diseases; the other was Dr. Ottavio Pellitteri, a field epidemiologist, who, since 1946, has been administrative medical inspector for the Bureau. He is 36 years old, pale, and stocky, and has a bristling black mustache. One day, when I was in Dr. Greenberg’s office, he and Dr. Pellitteri told me about the case. Their recollection of it is, understandably, vivid. The derelicts were the victims of a type of poisoning so rare that only 10 previous outbreaks of it had been recorded in medical literature. Of these, two were in the United States and two in Germany; the others had been reported in France, England, Switzerland, Algeria, Australia, and India. Up to September 25, 1944, the largest number of people stricken in a single outbreak was four. That was in Algeria, in 1926.

The Beekman-Downtown Hospital telephoned a report of the occurrence to the Health Department just before noon. As is customary, copies of the report were sent to all the Department’s administrative officers. “Mine was on my desk when I got back from lunch,” Dr. Greenberg said to me. “It didn’t sound like much. Nine persons believed to be suffering from carbon-monoxide poi- soning had been admitted during the morning, and all of them said that they had eaten breakfast at the Eclipse Cafeteria, at 6 Chatham Square. Still, it was a job for us. I checked with the clerk who handles assignments and found that Pellitteri had gone out on it. That was all I wanted to know. If it amounted to

213

Reading

11 Blue Men: An Epidemiological Investigation (continued)

anything, I knew he’d phone me before making a written report. That’s an arrangement we have here. Well, a couple of hours later I got a call from him. My interest perked right up.”

“I was at the hospital,” Dr. Pellitteri told me, “and I’d talked to the staff and most of the men. There were 10 of them by then, of course. They were sick as dogs, but only one was in really bad shape.”

“That was John Mitchell,” Dr. Greenberg put in. “He died the next night. I understand his condition was hopeless from the start. The others, including the old boy who came in last, pulled through all right. Excuse me, Ottavio, but I just thought I’d get that out of the way. Go on.”

Dr. Pellitteri nodded. “I wasn’t at all convinced that it was gas poisoning,” he continued. “The staff was beginning to doubt it, too. The symptoms weren’t quite right. There didn’t seem to be any of the headache and general dopiness that you get with gas. What really made me suspicious was this: Only two or three of the men had eaten breakfast in the cafeteria at the same time. They had straggled in all the way from seven o’clock to 10. That meant that the place would have had to be full of gas for at least three hours, which is pre- posterous. It also indicated that we ought to have had a lot more sick people than we did. Those Chatham Square eating places have a big turnover. Well, to make sure, I checked with Bellevue, Gouverneur, St. Vincent’s, and the other downtown hospitals. None of them had seen a trace of cyanosis. Then I talked to the sick men some more.

“I learned two interesting things. One was that they had all got sick right after eating. Within 30 minutes. The other was that all but one had eaten oatmeal, rolls, and coffee. He ate just oatmeal. When 10 men eat the same thing in the same place on the same day and then all come down with the same illness . . . I told Greenberg that my hunch was food poisoning.”

“I was willing to rule out gas,” Dr. Greenberg said. A folder containing data on the case lay on the desk before him. He lifted the cover thoughtfully, then let it drop. “And I agreed that the oatmeal sounded pretty suspicious. That was as far as I was willing to go. Common, ordinary, everyday food poisoning— I gathered that was what Pellitteri had in mind—wasn’t a very satisfying answer. For one thing, cyanosis is hardly symptomatic of that. On the other hand, diarrhea and severe vomiting are, almost invariably. But they weren’t in the clinical picture, I found, except in two or three of the cases. Moreover, the incubation periods—the time lapse between eating and illness—were extremely short. As you probably know, most food poisoning is caused by

214

Reading

11 Blue Men: An Epidemiological Investigation (continued)

eating something that has been contaminated by bacteria. The usual offend- ers are the staphylococci—they’re mostly responsible for boils and skin infec- tions and so on—and the salmonella. The latter are related to the typhoid organism. In a staphylococcus case, the first symptoms rarely develop in under two hours. Often, it’s closer to five. The incubation period in the other ranges from 12 to 36 hours. But here we were with something that hit in 30 minutes or less. Why, one of the men had got only as far as the sidewalk in front of the cafeteria before he was knocked out. Another fact that Pellitteri had dug up struck me as very significant. All of the men told him that the ill- ness had come on with extraordinary suddenness. One minute they were feel- ing fine, and the next minute they were practically helpless. That was another point against the ordinary food-poisoning theory. Its onset is never that fast. Well, that suddenness began to look like a lead. It led me to suspect that some drug might be to blame. A quick and sudden reaction is characteristic of a great many drugs. So is the combination of cyanosis and shock.”

“None of the men were on dope,” Dr. Pellitteri said. “I told Greenberg I was sure of that. Their pleasure was booze.” “That was O.K.,” Dr. Greenberg said. “They could have got a toxic dose of some drug by accident. In the oatmeal, most likely. I couldn’t help thinking that the oatmeal was relevant to our prob- lem. At any rate, the drug idea was very persuasive.”

“So was Greenberg,” Dr. Pellitteri remarked with a smile.

“Actually, it was the only explanation in sight that seemed to account for every- thing we knew about the clinical and environmental picture.”

“All we had to do now was prove it,” Dr. Greenberg went on mildly. “I asked Pellitteri to get a blood sample from each of the men before leaving the hospi- tal for a look at the cafeteria. We agreed he would send the specimens to the city toxicologist, Dr. Alexander O. Gettler, for an overnight analysis. I wanted to know if the blood contained methemoglobin. Methemoglobin is a compound that’s formed only when any one of several drugs enters the blood. Gettler’s report would tell us if we were at least on the right track. That is, it would give us a yes-or-no answer on drugs. If the answer was yes, then we could go on from there to identify the particular drug. How we would go about that would depend on what Pellitteri was able to turn up at the cafeteria. In the mean- time, there was nothing for me to do but wait for their reports. I’d theorized myself hoarse.”

Dr. Pellitteri, having attended to his bloodletting with reasonable dispatch, reached the Eclipse Cafeteria at around five o’clock. “It was about what I’d

215

Reading

11 Blue Men: An Epidemiological Investigation (continued)

expected,” he told me. “Strictly a horse market, and dirtier than most. The sort of place where you can get a full meal for 15 cents. There was a grind house on one side, a cigar store on the other, and the ‘L’ overhead. Incidentally, the Eclipse went out of business a year or so after I was there, but that had noth- ing to do with us. It was just a coincidence. Well, the place looked deserted and the door was locked. I knocked, and a man came out of the back and let me in. He was one of our people, a health inspector for the Bureau of Food and Drugs, named Weinberg. His bureau had stepped into the case as a matter of routine, because of the reference to a restaurant in the notification report. I was glad to see him and to have his help. For one thing, he had put a tempo- rary embargo on everything in the cafeteria. That’s why it was closed up. His main job, though, was to check the place for violations of the sanitation code. He was finding plenty.”

“Let me read you a few of Weinberg’s findings,” Dr. Greenberg said, extracting a paper from the folder on his desk. “None of them had any direct bearing on our problem but I think they’ll give you a good idea of what the Eclipse was like—what too many restaurants are like. This copy of his report lists 15 specific violations. Here they are: ‘Premises heavily infested with roaches. Fly infestation throughout premises. Floor defective in rear part of dining room. Kitchen walls and ceiling encrusted with grease and soot. Kitchen floor encrusted with dirt. Refuse under kitchen fixtures. Sterilizing facilities inadequate. Sink defective. Floor and walls at serving tables and coffee urns encrusted with dirt. Kitchen utensils encrusted with dirt and grease. Storage- cellar walls, ceiling, and floor encrusted with dirt. Floor and shelves in cellar covered with refuse and useless material cellar ceiling defective. Sewer pipe leaking. Open sewer line in cellar.’ Well . . . .” He gave me a squeamish smile and stuck the paper back in the folder.

“I can see it now,” Dr. Pellitteri said. “And smell it. Especially the kitchen, where I spent most of my time. Weinberg had the proprietor and the cook out there and I talked to them while he prowled around. They were very coopera- tive. Naturally, they were scared to death. They knew nothing about gas in the place and there was no sign of any, so I went to work on the food. None of what had been prepared for breakfast that morning was left. That, of course, would have been too much to hope for. But I was able to get together some of the kind of stuff that had gone into the men’s breakfast, so that we could make a chemi- cal determination at the Department. What I took was ground coffee, sugar, a mixture of evaporated milk and water that passed for cream, some bakery rolls, a 5-pound carton of dry oatmeal, and some salt. The salt had been used in preparing the oatmeal. That morning, like every morning, the cook told me, he had prepared 6 gallons of oatmeal, enough to serve around 125 people.

216

Reading

11 Blue Men: An Epidemiological Investigation (continued)

To make it he used 5 pounds of dry cereal, 4 gallons of water—regular city water—and a handful of salt. That was his term—a handful. There was an open gallon can of salt standing on the stove. He said the handful he’d put in that morning’s oatmeal had come from that. He refilled the can on the stove every morning from a big supply can. He pointed out the big can—it was up on a shelf—and as I was getting it down to take with me, I saw another can, just like it, nearby. I took that one down, too. It was also full of salt, or, rather, something that looked like salt. The proprietor said it wasn’t salt. He said it was saltpetre—sodium nitrate—that he used in corning beef and in making pastrami. Well, there isn’t any harm in saltpetre; it doesn’t even act as an anti- aphrodisiac, as a lot of people seem to think. But I wrapped it up with the other loot and took it along, just for fun. The fact is, I guess, everything in that damn place looked like poison.”

After Dr. Pellitteri had deposited his loot with a Health Department chem- ist, Andrew J. Pensa, who promised to have a report ready by the following afternoon, he dined hurriedly at a restaurant in which he had confidence and returned to Chatham Square. There he spent the evening making the rounds of the lodging houses in the neighborhood. He had heard at Mr. Pensa’s office that an 11th blue man had been admitted to the hospital, and before going home he wanted to make sure that no other victims had been overlooked. By midnight, having covered all the likely places and having rechecked the downtown hospitals, he was satisfied. He repaired to his office and composed a formal progress report for Dr. Greenberg. Then he went home and to bed.

The next morning, Tuesday, Dr. Pellitteri dropped by the Eclipse, which was still closed but whose proprietor and staff he had told to return for question- ing. Dr. Pellitteri had another talk with the proprietor and the cook. He also had a few inconclusive words with the rest of the cafeteria’s employees—two dishwashers, a busboy, and a counterman. As he was leaving, the cook, who had apparently passed an uneasy night with his conscience, remarked that it was possible that he had absentmindedly refilled the salt can on the stove from the one that contained saltpetre. “That was interesting,” Dr. Pellitteri told me, “even though such a possibility had already occurred to me, and even though I didn’t know whether it was important or not. I assured him that he had nothing to worry about. We had been certain all along that nobody had deliberately poisoned the old men.” From the Eclipse, Dr. Pellitteri went on to Dr. Greenberg’s office, where Dr. Gettler’s report was waiting.

“Gettler’s test for methemoglobin was positive,” Dr. Greenberg said. “It had to be a drug now. Well, so far so good. Then we heard from Pensa.”

217

Reading

11 Blue Men: An Epidemiological Investigation (continued)

“Greenberg almost fell out of his chair when he read Pensa’s report,” Dr. Pel- litteri observed cheerfully.

“That’s an exaggeration,” Dr. Greenberg said. “I’m not easily dumfounded. We’re inured to the incredible around here. Why, a few years ago we had a case involving some numskull who stuck a fistful of potassium-thiocyanate crystals, a very nasty poison, in the coils of an office water cooler, just for a practical joke. However, I can’t deny that Pensa rather taxed our credulity. What he had found was that the small salt can and the one that was supposed to be full of sodium nitrate both contained sodium nitrite. The other food sam- ples, incidentally, were O.K.”

“That also taxed my credulity,” Dr. Pellitteri said.

Dr. Greenberg smiled. “There’s a great deal of difference between nitrate and nitrite,” he continued. “Their only similarity, which is an unfortunate one, is that they both look and taste more or less like ordinary table salt. Sodium nitrite isn’t the most powerful poison in the world, but a little of it will do a lot of harm. If you remember, I said before that this case was almost with- out precedent—only 10 outbreaks like it on record. Ten is practically none. In fact, sodium-nitrite poisoning is so unusual that some of the standard texts on toxicology don’t even mention it. So Pensa’s report was pretty startling. But we accepted it, of course, without question or hesitation. Facts are facts. And we were glad to. It seemed to explain everything very nicely. What I’ve been saying about sodium-nitrite poisoning doesn’t mean that sodium nitrite itself is rare. Actually, it’s fairly common. It’s used in the manufacture of dyes and as a medical drug. We use it in treating certain heart conditions and for high blood pressure. But it also has another important use, one that made its presence at the Eclipse sound plausible. In recent years, and particularly dur- ing the war, sodium nitrite has been used as a substitute for sodium nitrate in preserving meat. The government permits it but stipulates that the finished meat must not contain more than 1 part of sodium nitrite per 5,000 parts of meat. Cooking will safely destroy enough of that small quantity of the drug.” Dr. Greenberg shrugged. “Well, Pellitteri had had the cook pick up a handful of salt—the same amount, as nearly as possible, as went into the oatmeal—and then had taken this to his office and found that it weighed approximately a hundred grams. So we didn’t have to think twice to realize that the propor- tion of nitrite in that batch of cereal was considerably higher than 1 to 5,000. Roughly, it must have been around 1 to about 80 before cooking destroyed part of the nitrite. It certainly looked as though Gettler, Pensa, and the cafete- ria cook between them had given us our answer. I called up Gettler and told

218

Reading

11 Blue Men: An Epidemiological Investigation (continued)

him what Pensa had discovered and asked him to run a specific test for nitrites on his blood samples. He had, as a matter of course, held some blood back for later examination. His confirmation came through in a couple of hours. I went home that night feeling pretty good.”

Dr. Greenberg’s serenity was a fugitive one. He awoke on Wednesday morn- ing troubled in mind. A question had occurred to him that he was unable to ignore. “Something like 125 people ate oatmeal at the Eclipse that morning,” he said to me, “but only 11 of them got sick. Why? The undeniable fact that those 11 old men were made sick by the ingestion of a toxic dose of sodium nitrite wasn’t enough to rest on. I wanted to know exactly how much sodium nitrite each portion of that cooked oatmeal had contained. With Pensa’s help again, I found out. We prepared a batch just like the one the cook had made on Monday. Then Pensa measured out 6 ounces, the size of the average portion served at the Eclipse, and analyzed it. It contained two and a half grains of sodium nitrite. That explained why the 114 other people did not become ill. The toxic dose of sodium nitrite is three grains. But it didn’t explain how each of our 11 old men had received an additional half grain. It seemed extremely unlikely that the extra touch of nitrite had been in the oatmeal when it was served. It had to come in later. Then I began to get a glimmer. Some people sprinkle a little salt, instead of sugar, on hot cereal. Suppose, I thought, that the busboy, or whoever had the job of keeping the table salt shakers filled, had made the same mistake that the cook had. It seemed plausible. Pellitteri was out of the office—I’ve forgotten where—so I got Food and Drugs to step over to the Eclipse, which was still under embargo, and bring back the shakers for Pensa to work on. There were 17 of them, all good-sized, one for each table. Sixteen contained either pure sodium chloride or just a few inconsequential traces of sodium nitrite mixed in with the real salt, but the other was 0.37% nitrite. That one was enough. A spoonful of that salt contained a bit more than half a grain.”

“I went over to the hospital Thursday morning,” Dr. Pellitteri said. “Greenberg wanted me to check the table-salt angle with the men. They could tie the case up neatly for us. I drew a blank. They’d been discharged the night before, and God only knew where they were.”

“Naturally,” Dr. Greenberg said, “it would have been nice to know for a fact that the old boys all sat at a certain table and that all of them put about a spoonful of salt from that particular shaker on their oatmeal, but it wasn’t essential. I was morally certain that they had. There just wasn’t any other explanation. There was one other question, however. Why did they use so much salt? For

219

Reading

11 Blue Men: An Epidemiological Investigation (continued)

my own peace of mind, I wanted to know. All of a sudden, I remembered Pellit- teri had said they were all heavy drinkers. Well, several recent clinical studies have demonstrated that there is usually a subnormal concentration of sodium chloride in the blood of alcoholics. Either they don’t eat enough to get suf- ficient salt or they lose it more rapidly than other people do, or both. What- ever the reasons are, the conclusion was all I needed. Any animal, you know, whether a mouse or a man, tends to try to obtain a necessary substance that his body lacks. The final question had been answered.”