14 Applied epidemiology

Ms. Shumayla and Meenal Dhall

Contents:

    1.             Concept and Definition of Epidemiology

2.             Historic Evolution of Epidemiology

3.             Measuring Health and Disease

3.1 What is a case?

3.2  Incidence measures all new cases during a period of time

3.3  Prevalence measures all cases at one point in time

3.4  Difference between Incidence and Prevalence

3.5 Number and Rates

4               Approaches to epidemiological study

4.1 Ecological Studies

4.2   Cross-sectional Studies

4.3   Case Control Studies

4.4   Cohort Studies

4.5   Intervention Studies

4.5.1. Randomized Controlled Trials (RCT)

5               Uses of Epidemiology

5.1   Assessing the Community Heath

5.2    Completing the clinical picture

5.3      Making individual decisions

5.4      Searching for causes

6               Functions of Epidemiology in Public Health

6.1 Public health surveillance

6.2   Field Investigation

6.3   Linkages

6.4 Policy Development

Summary

    Learning Objective:

• To explore the field of epidemiology, its principles and the use in public health.

    1. Concept and Definition of Epidemiology

The word “epidemiology” comes from two Greek words “epi”, means on or upon, demos, means people, and logos means the study of. In other words, we can say that epidemiology has its roots in the study of what befalls a population. Numerous definitions have been proposed, but the following definition seize the “underlying principles” and “public health spirit” of epidemiology:

“Epidemiology is the study of the distribution, frequency and determinants of health-related problems or events in a specified population, and the application of this type of study is to control health problems (Last, 2001.”

The purpose of epidemiology is to obtain, interpret and use health information to promote health and reduce disease. The basic concepts of epidemiology are highly practical and are relevant not only for the members of the district health management team but also for all health workers. Epidemiology is a scientific discipline with comprehensive methods of scientific inquiry with its foundation. Epidemiology is a data-driven methods which relies on a systematic and impartial approach to collection data to analyze, and interpret it. Basic epidemiologic methods rely on careful observation and also used for valid comparison of groups to assess whether what was observed like as the number of cases in a particular area during particular time period or the occurrence of an exposure among persons with disease, may differs from what is expected. Epidemiology also draws methods from other scientific fields which includes biostatistics and informatics along with biologic, economic, behavioral and social sciences.

2. Historic Evolution of Epidemiology

In 400 B.C, Hippocrates attempted to define the disease occurrence from a rational viewpoint rather than supernatural. Hippocrates in his essay “On Airs, Waters and Places”, suggested that host factors like behavior and environmental factors may influence the development of disease. John Graunt was also one of the early contributor in epidemiology when he published mortality data in 1662. This was the first publication which quantify the patterns of death, birth and disease occurrence, noticing disparities between females and males, high infant mortality, seasonal variations and urban/rural differences (Snow, 1936). John Snow, an anesthesiologist in mid 1800s conducted studies on Cholera outbreaks to discover the cause of disease and also to prevent its recurrence. This was done 20 years before the development of microscope. He was also considered as the” father of field epidemiology”. The studies of Doll and Hill linking the cancer of lung to smoking (Doll and Hill, 1950) and study among residents of Framingham in Massachusetts about cardiovascular disease (Kennel, 2000) are two pioneer research examples in epidemiological field.

During 1969s and early 1970s health workers used epidemiologic methods to eliminate naturally occurring smallpox around the world (Fenner et al., 1988). In 1980s, epidemiology extended to studies about injuries and violence as well. In 1990s, the new field of epidemiology i.e. molecular and genetic epidemiology has been come into play. Meanwhile, the infectious diseases continued to be a challenge for epidemiologists with emergence of new infectious agents (Ebola virus, HIV/ AIDS, etc). Today, all the public health workers throughout the world accept and use epidemiology on regular basis to characterize health of their communities and also to solve small and large day-to-day problems.

3. Measuring Health and Disease

Measuring health and disease is fundamental for the practice of epidemiology. In epidemiology, a practical definitions of disease and health that emphasize the aspects of health which are easily measurable and acquiescent to improvement is needed. The two main measures of the frequency of diseases, health problems and the use made of health services are incidence and prevalence. It is important to be clear about which is being used. Before describing the measure of health and disease, it is important to know about a case.

3.1 What is a case?

Stipulation of diagnostic criteria is required for measuring disease frequency in a population. The definition of “a case” in a clinical setting generally assumes that, for a disease, people are divided into two discrete groups- the affected and the non-effected or ‘disease present’ or ‘disease absent’. To overcome problems in defining a definite case, it is good exercise for some diseases, to establish diagnostic criteria and classify cases into several groups, such as possible, probable or definite accordingly. For instance, a patient with fever, headache and body aches could be said to be a possible case of malaria; a probable case might be someone who also has responded well to antimalarial treatment. Only if there was a positive blood slide for malarial parasites might the case be called definite one. However, it still has to be recognized that the symptoms might be due to some other illness, particularly in individuals living in malarial endemic areas.

3.2 Incidence measures all new cases during a period of time:

Incidence measures the number of new cases. Episodes or events occurring over a defined period of time, commonly one year. Incidence is the most basic measure of frequency and is the best indicator of whether a condition is decreasing, increasing or remaining constant. It is therefore, the best measure to use in evaluating the effectiveness of health programmes. It is also used as a measure in surveillance systems and for analyzing how people are using the health services.

3.3 Prevalence measures all cases at one point in time:

Prevalence measures the total number of existing cases, episodes or events occurring at one point in time, commonly on a particular day. Prevalence may be more complicated to interpret than incidence because it depends upon the number of people who have developed their illness in the past and have continued to be ill to the present time. It is the combination of the previous incidence of a condition and its duration. Examples of frequency measured by prevalence are the total number of a particular disease patients on a register at the beginning of each month or the number of hospital beds occupied per day. Prevalence is very useful in chronic conditions while incidence is more useful for those diseases with a short average duration like measles, pneumonia and diarrhea.

In other words, the incidence of a disease represents the rate of occurrence of new cases rising in a given period of time in a specified population, whereas prevalence is simply the frequency of existing cases in distinct population at a given point of time. There may be high prevalence and low incidence of diabetes as per say and low prevalence and high incidence of common cold. This can be explained by saying that cold is a common and more frequent than diabetes but it may last for a short time, while diabetes is a lifelong illness.

Prevalence (P) of a specific disease is calculated as:

P= No. of people with disease at a specified time/ No. of people in the population at risk at specified time* 10/7

In many studies, the total population in study area is used as approximation as data on population at risk are not available most of the time.

3.4 Difference between Incidence and Prevalence

3.5 Number and Rates:

Incidence and prevalence are used in reporting health information and may be given as a whole number or a calculated rate.

The most readily available data are in absolute numbers. These are often used in monitoring the occurrence of important infectious disease, especially in outbreaks, when the populations involved are restricted in time and locality and the population structure can be assumed to be stable. When we have to look at trends over a period of time, or compare the frequency of disease between subgroups or communities, using the total number of cases can lead to invalid conclusions. The population size and age-sex structure of each group must also be considered before the groups can be compared and the information should than be expressed in terms of incidence or prevalence rate. Prevalence and incidence rates are calculated as:

Prevalence rate = New cases in specified period of time/ Total population at risk *X factor Incidence rate = Existing cases at specified point of time/ Total population at risk* X factor

The rate is multiplied by a factor, the size of which is chosen so as to enable the rate to be expressed as a suitable whole number; the factor used is commonly 100, 1000 or 10,000.

4. Approaches to epidemiological study

The two main approaches to epidemiological study are -descriptive and analytical.

Descriptive epidemiology provide information about the distribution of health outcomes by population type, age, geography and over time. Sources for descriptive data may include routine monitoring like registration at the time of births and deaths and hospital and clinical records etc. also population censuses may also provide data on births, deaths, and a variety of risk factors and there is an overlap with demography. Population health surveys evolved from census provide information on the use of health services, coverage of interventions and the frequency of specific outcomes.

On the other hand Analytical epidemiology aims to investigate which factors may be responsible for increasing or decreasing the probability (‘risk’) of an outcome. Identifying the cause of an outcome is not always simple, and can be described in terms of sufficient cause and component causes. Also we can say that Analytical methods may confirm an association between an exposure and outcome, but causality can only be inferred if alternative explanations, namely chance, bias and confounding, have been accounted for.

Observational and interventional or experimental are the two forms the analytical epidemiology can take. An interventional study involves an active attempt change the determinant of a disease. For instance an exposure or a behavior and sometimes the progress of treatment for a disease. The main experimental/interventional study designs are as follows:

1.      Clinical trials or randomized controlled trials (patients as subjects)

2.      Field trials (subjects are healthy individuals)

3.      Community trials (communities are the subject themselves)

  An observational study aims to compare the frequency of the outcome in groups or individuals with or without the exposure of interest. An intervention study is effectively an experiment, and therefore restricted to evaluating the effect of reducing a risk factor or increasing a protective factor on the frequency of an outcome. Following study designs form the core of epidemiological research:

4.1 Ecological studies:

These studies are also called correlational studies which are useful in generating hypothesis. In this type of study unit of analysis are groups and not an individual. For instance, there was a relationship found between occurrence of high number of asthma deaths and an average sale of an anti-asthmatic drug in different part of New-Zealand (Pearce, 1998). This observation need to be tested by keeping a control on all the possible confounders to eliminate the possibility that other characteristics like severity of the disease in different population did not account for this relationship.

Also, ecological studies can be done by comparing population in different places at same time or in a series of time, by comparing same population in one place at different times. Ecological studies are simple and easy to conduct and are sometimes difficult to interpret as it is rarely possible to examine the various potential explanations directly for the findings. One important feature of ecological studies is that data can be used from the populations with widely divergent characteristics from different data sources.

4.2 Cross-sectional studies:

These studies measures the prevalence of diseases. The measurement of exposure and the effect are made at same time in cross-sectional studies. It is hard to find the reasons for the association shown in these studies. These type of studies are inexpensive and relatively easy and are very useful in investigating the exposures that are fixed (characteristics) like blood group or ethnicity. A cross-sectional study could be a good measure to study the first step in investigating the cause of an outbreak of disease.

These studies are helpful in assessing the needs of population related to health care. Data from repeated surveys of cross-sectional studies which used independent random samples with standardized definitions and methods provide very useful indications of trends (Bonita et al., 2003; Tolonen et al., 2006). Each cross-sectional survey should have a clear purpose. A valid survey requires a well-designed questionnaire, a sufficient sample size and good response rate. Many of the countries have regular cross-sectional surveys focusing on demographic, personal, illness and health related characteristics on the representative samples of the population.

4.3 Case Control Studies:

These type of studies gives a relatively simple way to find out the cause of diseases, especially the rare diseases. Case control studies involve a person with a disease of interest and suitable control (reference) unaffected with the disease. It compares the possible causes of occurrence in both cases and controls. The data collector or the investigators collect data on the occurrence of disease at one point in time and the exposures at the previous point in time. These studies are longitudinal which is in contrast to cross-sectional studies

Since the investigators are looking backward from the disease for a possible cause, hence also called as retrospective studies. This may be confusing sometimes as the terms prospective and retrospective are also used to define the timing of data collection in relation with the current date. In such cases, a case control study may be a prospective one when data collection continues with passage of time or retrospective, when all data deal with past.

4.4 Cohort Studies:

Cohort studies or follow-up studies or incidence studies starts with a group of people who are not affected with the disease and are classified into the subgroups according to the exposure to a potential cause for the disease or outcome. The cohort study is different from experimental study as the investigator in cohort study only observes rather than determining the status of participant’s exposure. The investigator compares the rate of disease between exposed and unexposed groups over a period of time. The unexposed group is also known as comparison group, which provide estimate of baseline or expected rate of disease occurrence in community. If rate of disease is practically different in exposed group as compared to comparison group, the exposure is said to be related with illness.

The length of follow-up studies vary considerably. In case of an outbreak, public health public health professionals tend to conduct brief studies in an attempt to a quick response. But academicians and researchers are more likely to do study for a longer time which may last for years or decades. The Framingham study is an example of such cohort study in which 5000 individual living in Framingham, Massachusetts were followed since early 1950s to establish rates and risk factors for various heart disease (Kannel, 2000).

Retrospective study is another type of cohort study in which both the outcomes and exposure have already occurred. These type of studies are commonly conducted in a group of people who are easily identifiable ,for instance workers from a factory, student of a particular university or attendees of a wedding. In a retrospective cohort study in Pennsylvania (2004), the source of infection for cyclosporiasis (a parasitic disease) that caused and outbreak was determined. The result showed that consumption of snow peas was involved as the vehicle for cycloporiasis outbreak (CDC, 2004).

4.5 Intervention Studies:

The association between an exposure and outcome to a specific intervention has been measured in intervention studies. Interventions may be focusing on a preventive measure to an exposure in those who are at risk of the outcome (example: insecticide treated mosquitoes nets against malaria) or the treatment to reduce the severity or mortality in those who are already have an outcome (example: interferon treatment of individuals with chronic Hepatitis B for the reduction of progression to liver cancer). It is also used to reduce or remove an assumed risk factor (example: Health related education to  reduce  smoking),  also  it  may  increase  or  introduce  an  assumed  protective  factor  (example: vaccination against polio-virus).

Unlike observational studies, intervention studies are experimental, as they are able to intervene. The investigators in this design define that which subjects are “exposed” and which are “unexposed”. For some ethical issues, these studies are restricted only to evaluating exposures that may reduce the frequency of negative outcomes.

The intervention studies are long and quite expensive to carry out. These may require large study teams and extensive follow ups to identify sufficient subjects with an outcome. There may also be an additional cost of intervention as well as cost relating to concealing the allocation and monitoring safety of the interventions.

Within interventional studies there is a study called Plausibility study which evaluate the effectiveness of interventions by comparing with a geographical, historical or opportunistic control group with no randomization. These studies are appropriate in following situations (Victora et al., 2004)

1. If an intervention is so complex that Randomized Controlled Trial (RCT) results will be excessively artificial.
2. If an intervention is effective in small-scale studies but its effectiveness on large scale must be demonstrated.
3. If ethical concerns prevent use of an RCT.

    4.5.1. Randomized Controlled Trials (RCT)

When the effectiveness studies may be subjected to ethical or practical design limitations then an efficacy study is conducted ideally as a RCT. The trials in which interventions are not allocated randomly are more likely to suffer for selection bias and confounding. But may be useful in providing preliminary evidences prior to the more rigorous study. Randomized Controlled Trials are defined by:

a.) Presence of an existing comparison group of study individuals who do not receive any intervention, known as “control” arm,

b.) Random allocation of study individuals for intervention and control arms.

5. Uses of Epidemiology

Epidemiology and the information engendered by using epidemiologic methods have been used for many purposes (Morris, 1957). Some of them are as follows:

5.1 Assessing the Community Heath

Public health personnel are responsible for policy making, implementing and evaluating. They use epidemiological information and methods as a factual framework for making decision. To assess health of a community or population, relevant sources of data should be identified and analyzed by the person, place and time:

• What are the potential and actual problems related to health in a community?
• Where is the region of occurrence?
• Which population are at risk?
• Which problem has declined over time?
• Which problems are increasing or have potential to increase?
• How  these  patterns  are  related  to  the  level  and  distribution  of  the  public  health  services available?

  Majority of Public health professionals used epidemiological data and methods for the identification of baselines and to set health goals for our nation in 2000 and 2010, and to monitor the progress to these goals (HHS, 1991,2000a, 2000b).

5.2 Completing the clinical picture

Epidemiologists rely on laboratory technicians and health-care providers to establish proper diagnosis of an individual when investigating a disease outbreak. They also contribute to the physicians to understand the clinical picture and the history of disease. Lately, clinicians, researchers and epidemiologists have collaborated to characterize SARS (a disease caused by a new type of coronavirus which emerged in China in late 2002) ,Kamps (2003). Epidemiologists are also instrumental in describing many non-acute diseases like many numerous conditions associated with smoking of cigarette, from pulmonary and heart disease to throat, lip and lung cancer.

5.3 Making individual decisions

Some individuals use epidemiologic methods and information to make their daily decision related to their health unknowingly. Suppose when an individual decides to quit smoking, or climb the stairs instead of taking elevator, eating of salad rather than any fast food for lunch, using a condom etc. these conditions are influenced consciously or unconsciously by epidemiologists assessment of risk. In mid 80s epidemiologists identified an increased risk of HIV infection associated with some certain behavior associated with sex and drugs. These and other hundreds of other epidemiological findings are directly or indirectly associated with the choices people make on day-day basis, choices that affect their health over a period of time.

5.4 Searching for causes

Majority of epidemiologic research is dedicated to searching of casual factors that effects one’s risk of disease. The goal for these researches is to find a cause so that an appropriate public health action should be taken. For instance, epidemiologists successfully identified a variety of risk factors for an outbreak of pneumonia among persons who attended the American Legion Convention in 1976, in Philadelphia. Even though there was no traces of Legionnaires’ bacillus in the laboratory from lung tissue of an individual who died because of this Legionnaires’ disease until nearly 6 months later (Fraser, 1977).

6. Functions of Epidemiology in Public Health

Five main tasks of epidemiology in public health in mid 1980s were identified as follows:

6.1 Public health surveillance

Public health surveillance is an ongoing, systematic collection, an analysis, interpretation and the dissemination of health data for helping public health decision makers to take action. Surveillance is quite similar to monitoring the pulse of the population or the community. The main purpose of the public health surveillance also called “information for action” (Orenstein, 1990) is to represent the current patterns of disease occurrence and potential disease so that the investigation, control and the prevention measures can be applied effectively and efficiently.

The common sources of the surveillance data for the local and the state health departments are mortality and morbidity reports. Health – care providers, laboratories or infection control practitioners generally submitted these reports to inform the health departments of any patient with a different, new or reportable disease like polio, AIDS, etc. Other sources of data which is used for surveillance includes reports from investigations of individual cases and a disease cluster, some public health program data like immunization coverage in a population or community, health surveys and disease registries.

Surveillance depends on the simple systems for the collection of limited amount of evidence about each case. Though each and every case is not registered, but health officials review these case reports on regular basis and also look for the patterns among them. These surveillance practices are important in detecting problems, evaluating the programs and also to guide public health action. From 2001, a wide range of systems that depends on electronic reporting have been installed, including those which reports the daily emergency department calls, trades of over-the-counter medicines and the worker absenteeism (Wagner, 2001; CDC, 2004b).

6.2 Field Investigation

As mentioned above , the surveillance gives information for action. The first step or the action that results from the surveillance case report is the investigation by public health departments. The investigation vary from a simple phone call to health care provider for clarifying the reported case to a field investigation that requires coordinated efforts from many people to characterize extent of the epidemic and to recognize the cause. Sometimes investigation leads to identification of many additional unrecognized or unreported cases, otherwise who might continue to spread the infection. For instance, in case of STD, the person is investigated by identifying the sexual partners and other contacts of the persons. When these persons are interviewed, many are found to be infected unknowingly and then are given treatment they did not realize, they needed. So, the identification and further treatment of these contacts prevent further spread of the disease.

In some diseases, through investigation, the vehicle or the source of infection that can be eliminated and controlled is identified. For instance, in an E.coli O157:H7, focus is to find out the source, often ground beef but rarely something more unusual like fruit juice may be the source. So by identifying the source, investigators may be able to conclude that how many other individuals might have already been exposed and how many are at risk to be exposed. Also if some commercial product is turned out to be the culprit, recalling the product and public announcements may prevent many of the additional cases.

6.3 Linkages

Epidemiologists who are working in public health field rarely act in isolation. Often the field of epidemiology is also called “team sport”. In an investigation an epidemiologist participate either as a member or leader of multidisciplinary team. Other team members may include laboratory technicians, nurses, sanitarians other clinical staff also computer information specialist sometimes. Often outbreaks are cross geographical or across jurisdictional lines, so the co-investigators may be from other local or state government, clinical facilities, academic institutions or some private sector.

To promote future and current collaboration, the epidemiologists sometimes need to maintain relationships with the staff of other institution and agencies as well. Mechanisms for supporting such connections includes official memoranda of understanding, sharing information for public health audience and outside partners and also informal networking which take place during professional meetings.

6.4 Policy Development

Generally, the definition of epidemiology ends with the phrase: “…and the application of this study to control of health problems.” But some epidemiologist involved in academics says that epidemiologists should stick to research only and should not get involved in policy making or even in recommendations (Rotham, 1993), public health epidemiologists do not have this type of luxury. Certainly, epidemiologists who understand the problem and the population in which the problem occurs are generally in a unique qualified position to recommend suitable interventions. As a results, the epidemiologists working in the public health sector provides input, testimonies and recommendations regularly regarding disease control strategies, reporting of disease regulations and also health-care policies.

Summary

At the basic level, epidemiology includes the study of frequency, patterns and the causes of health related problems in a population or a community, and the application of this study is to address the public health issues. Epidemiologists use an organized approach to assess who, what, where, why, when and how of these public health issues. The two most essentials of epidemiology are comparison and population. The task of epidemiologists are public health surveillance, research, evaluation, field investigation and policy development. To carry out these tasks, epidemiologists are almost and always a part of the team dedicated to protection and promotion of public health. Epidemiologists study the differences in problems among different populations to generate hypotheses about the risk factors and causes. Cohort and case-control studies are commonly used by them to evaluate the hypothesis. Knowledge regarding basic principles of occurrence of disease and its spread among a population is important for the implementation of effective control and measures for its prevention.

References

• Bonita R, Douglas K, Winkelmann R, De Courten M. The WHO STEPwise approach to surveillance (STEPS) of noncommunicable disease risk factors. In McQueen DV, Puska P eds. Global Risk Factor Surveillance. London, Kluwer Academic/Plenum Publishers, 2003:9-22.
• CDC a (Centers for Disease Control and Prevention). Framework for evaluating public health surveillance systems for early detection of outbreaks: recommendations from the CDC Working Group. MMWR May 7, 2004a; 53(RR05);1–11.
• CDC(Centers for Disease Control and Prevention). Outbreak of Cyclosporiasis associated with snow peas–Pennsylvania, 2004. MMWR 2004b;53 :876–8.
• Doll R, Hill AB. Smoking and carcinoma of the lung. Brit Med J 1950;2: 739–48.
• Fenner F, Henderson DA, Arita I, Jezek Z, Ladnyi D. Smallpox and its eradication. Geneva: World Health Organization; 1988.
• Fraser DW, Tsai TR, Orenstein W, Parkin WE, Beecham HJ, Sharrar RG, et al. Legionnaires’ disease: description of an epidemic of pneumonia. New Engl J Med 1977; 297:1189–97.
• HHS (U.S. Department of Health and Human Services). Healthy people 2000: national health promotion and disease prevention objectives. Washington, DC: HHS, Public Health Service; 1991.
• HHS (U.S. Department of Health and Human Services). Healthy people 2010. 2nd ed. Washington, DC: U.S. Government Printing Office (GPO); November 2000.
• HHS (U.S. Department of Health and Human Services). Tracking healthy people 2010. Washington, DC: GPO; November 2000.
• Kamps BS, Hoffmann C, editors. SARS Reference, 3rd ed. Flying Publisher, 2003.
• Kannel WB. The Framingham Study: its 50-year legacy and future promise. J Atheroscler Thromb 2000;6: 60–6.
• Last JM, Spasoff RA, Harris SS, Thuriaux MC. A dictionary of epidemiology. International Epidemiological Association, Inc.; 2001.
• Morris JN. Uses of epidemiology. Edinburgh: Livingstone; 1957.
• Orenstein WA, Bernier RH. Surveillance: information for action. Pediatr Clin North Am 1990; 37:709–34.
• Pearce N, Hensley MJ. Beta agonists and asthma deaths. Epidemiol Rev1998;20: 173-86
• Rothman KJ. Policy recommendations in epidemiology research papers. Epidemiol 1993; 4: 94–9.
• Snow J. Snow on cholera. London: Humphrey Milford: Oxford University Press; 1936.
• Tolonen H, Dobson A, Kulathinal S, Sangita A, for the WHO MONICA Project. Assessing the quality of risk factor survey data: lessons from the WHO MONICA Project. Eur J Cardiovasc Prev Rehabil 2006;13:104-14.
• Wagner MM, Tsui FC, Espino JU, Dato VM, Sittig DF, Caruana FA, et al. The emerging science of very early detection of disease outbreaks. J Pub Health Mgmt Pract 2001;6:51–9.

    Suggested Readings

• Rothman, K. J. (2012). Epidemiology: an introduction. Oxford university press.
• Ahlbom, A., & Norell, S. (1990). Introduction to modern epidemiology. Epidemiology Resources.
• Webb, P., & Bain, C. (2010). Essential epidemiology: an introduction for students and health professionals. Cambridge University Press.
• Janes, C., Stall, R., & Gifford, S. M. (Eds.). (1986). Anthropology and epidemiology: interdisciplinary approaches to the study of health and disease (Vol. 9). Springer Science & Business Media.