1
COMMUNITY DENTISTRY YEAR I: 2002 INTRODUCTION TO EPIDEMIOLOGY AND POPULATION HEALTH Objectives 1. to indicate the nature and scope of epidemiology; 2. to indicate its links to public health and dental practice Two types of health care practice A good starting point is to distinguish between two different types of health care practice. The first, and probably most familiar, is clinical practice. Here a clinician treats individuals who have one or more physical or psychological disorders in order to restore that person to a state of health. The second, probably less familiar, is population health practice. Here, a public health specialist directs interventions at populations with the aim of preventing disease and maintaining health. For example, a dentist treats dental caries by removing diseased tissue and restoring a tooth. A dental public health specialist will implement water fluoridation or provide dental health education to groups or populations in order to prevent the onset of dental caries. The distinction is not of course absolute. Clinicians do use preventive interventions and educate patients, and public health specialists do develop and deliver programs that treat patients. However, the fundamental role of the clinician is treating disease while that of the public health specialist is preventing disease. A good definition of population health practice is the development of organized community efforts to prevent disease and promote health. Epidemiology is the core scientific discipline of public and population health practice and has an important role in clinical practice.
Definition Epidemiology is the scientific study of the frequency, distribution and determinants of health and disease in human populations. Oral epidemiology is that branch of the discipline that studies oral health and disease. Epidemiology has its roots in the study of epidemics. As such, it is an academic discipline that contributes to both public health practice and clinical practice. It does this through its central concern with causation and the relationships between various exposures or interventions and their outcomes. Understanding causal mechanisms allows us to identify ways of trying to control the events in the natural and social worlds. All decisions that health professionals make about disease and its treatment involve assumptions about cause and effect relationships.
2
Causal relationships in health care There are three types of causal relationships of concern to epidemiology: Etiology – epidemiology is concerned with the factors that cause the disease. This forms the basis of prevention. Eliminating or reducing these factors leads to a reduction in the frequency of disease in a population. Therapy – treatment is intended to produce an improvement in a patient’s condition. Epidemiology can identify whether or not a given treatment is effective or which of several treatments are most effective for a patient with a given condition. Health service delivery - the provision of certain types of health services or facilities is intended to improve the health of the community served. Epidemiology is used to determine if this is the case. Why study epidemiology? All health professionals need to be familiar with epidemiology, its principles and procedures, for the following reasons: 1. It provides for a comprehensive understanding of health and disease in individuals and populations, and the forces and factors which influence them. This is consistent with the mission of the health care system and health care professionals which is to eliminate disease from and improve the health of individuals and populations; 2. It provides for an understanding of the scientific methods used to produce the knowledge base on which health care practice if founded; 3. Epidemiological principles are coming to play an important role in clinical decisions for individual patients.
Clinical practice used to be based on clinical experience and clinical opinion. While these are still relevant, decisions regarding a patient’s health problems and how to treat them are increasingly based on scientific evidence. Understanding the methods used to study health and disease is a prerequisite for critical appraisal of the scientific literature. In turn, the ability to distinguish good from poor science leads to evidence based care, which is the process of applying credible scientific evidence to the problems that patients present.
3
Characteristics of epidemiology as a discipline Epidemiology as a field of enquiry has the following characteristics: 1) First, it is a science. What do we mean by science? A good definition is: a systematic set of methods and procedures for developing knowledge about events and the relationships between them which eliminates chance, bias and error. 2) Second, it is not concerned with individuals but with populations and groups. 3) Third, it asks questions such as how much disease is there in a population, who has it and why. Key concepts in epidemiology are prevalence, incidence and risk. Other important characteristics of epidemiology as a discipline can be illustrated by looking at some classic investigations of epidemics and the way in which they were controlled as a result of epidemiological investigations. 1. London cholera epidemics in the 19th century A physician named John Snow investigated the cholera epidemics which plagued London during the mid 19th century. He did not believe that current theories of transmission (the effluvia theory, in which cholera was transmitted by breathing vapours emitted by cholera victims or their corpses) fitted his observations about the distribution of the disease. Rather, his observations lead him to believe that cholera was transmitted by contaminated water involving some unknown mechanism. In 1848 there was a major epidemic in Soho, a part of London. Eighty-three deaths occurred over ten days within a very small geographic area. He plotted on a map the houses where deaths occurred and ascertained where they obtained their water. All took their water from a public pump in Broad Street. Few deaths occurred in a workhouse in the neighbourhood that had its own well, or among workers in a local brewery where the workers drank beer rather than water. Using this evidence Snow persuaded the local authorities to remove the handle from the pump. They did and the epidemic subsided. Snow also investigated an epidemic in 1854 and noticed that deaths from cholera were more frequent in houses whose water was supplied by the Southark and Vauxhall Company which drew its water from a part of the river Thames heavily contaminated with sewage. Deaths from cholera were much less frequent in houses supplied by the Lambeth Company which drew its water from areas of the river not so contaminated. Again these observations were consistent with the idea that cholera was transmitted by water by some, as then, unknown mechanism. This identification of the cause of cholera epidemics happened some 40 years before the germ theory of disease emerged and the bacterial origins of cholera became known.
2. Dental decay and fluoride in the water supply During the early part of the 20th century a US dentist named Frederick McKay investigated what was then called ‘mottled enamel’(now called dental fluorosis). This
4
was a condition in which the enamel was defective, deeply pitted and stained. He observed that it was common in areas where the water supply was obtained from deep artesian wells. He also observed that these areas had low rates of dental decay. It was not until 1931, however, that the water from the areas was shown to contain high levels of fluoride. During the 1930’s H.Trendley Dean undertook a series of studies (the most famous being the 21 cities study – see attached chapter from Burt and Eklund; pp279282) to test the hypothesis that there was an inverse relationship between rates of dental decay and the concentration of fluoride in the water supply. The hypothesis was confirmed; areas with high levels of fluoride had low levels of decay. This was followed in the 1940’s by community trials in which rates of dental decay were reduced by the addition of fluoride to the community water supply. Subsequently, water fluoridation has been adopted by many communities as a preventive public health intervention resulting in reductions in dental decay of 30 to 60%. This was achieved even though the mechanism by means of which fluoride prevented dental decay was not known. We now know that the effect of fluoride is mainly topical rather than systemic. That is, it promotes the remineralization of early carious lesions and does not become incorporated into developing enamel and reduce its solubility to acids as originally thought. Consequently, some authorities have now questioned whether it is necessary to maintain fluoride in the water supply since the ingestion of fluoride is not necessary for it to have a beneficial effect. 3. Toxic shock syndrome In the US in the late 1970’s, there was an epidemic of a disease in young women leading to a number of deaths which was subsequently identified as toxic shock syndrome, a disease induced by bacterial toxins. A comparison of women with and without the disease revealed that: 1) women with the disease were more likely to be menstruating, 2) they were more likely to use a particular type of hygiene product, and 3) they were more likely to use a particular brand of that product. In 1980, the product was with drawn and the epidemic subsided. It was only five years later that the mechanism underlying the disease became understood. It was the result of an interaction between the type of fibre used in the product, certain trace elements and a certain strain of bacteria which grew rapidly under those conditions.
These historic example illustrate a number of points: 1. Epidemiology is a logical discipline, which proceeds by way of a sequence of reasoning. 2. It is a comparative discipline in that the determinants of a disease are identified by comparing its frequency in different groups. 3. Epidemiological investigations enable us to reduce levels of disease in populations, even when the biological mechanisms involved are not known. This important principle is illustrated further in the Table 1 in the paper by Wynder (1994) attached.
5
The paper by Wynder (1994) addresses a debate concerning the relative merits of ‘nonmechanistic environmental’ and ‘mechanistic’ approaches to explaining and controlling disease. The former suggest that we should focus on the environmental, behavioural and lifestyle causes of disease while the latter argues that we should focus on research to uncover the biological and pathological mechanisms leading to disease. While both views have their merits (understanding mechanisms can lead to measures to treat disease) we have to acknowledge that major improvements in the health of modern populations have come about because of environment change rather than medical treatment. For example, in Europe death rates from tuberculosis declined dramatically prior to the discovery of the tubercle bacillus and well before the discovery of chemotherapy for the disease or BCG vaccination against the disease. The factors responsible were improvements in standards of living and nutrition. However, we should also note that smallpox and now measles have been eradicated from human populations by vaccination, an approach that requires the combination of basic science and epidemiology.
Sequence of epidemiological reasoning In determining the cause of disease epidemiology proceeds in the following way. An initial observation of the distribution of a disease in a population leads to the suspicion that a given factor is causally involved in the disease. (For example, in the 1920’s and 1930’s it was observed that rates of lung cancer were increasing as was the consumption of cigarettes.) This suspicion is stated as a formal hypothesis that is tested by means of an epidemiological study. This study tells us whether or not there is a valid statistical association between the factor and the disease. Since statistical association does not necessarily mean a cause and effect relationship, all available evidence has to be considered in order to make a casual inference. Once cause and effect has been established we can proceed to prevention by modifying the factor in question. TYPES OF EPIDEMIOLOGY Like other sciences, epidemiology is a dynamic discipline that has been developing continuously since its origins in the mid-nineteenth century. The discipline can be classified into several different types which appeared almost in historical sequence. Field epidemiology (1850) This investigates and identifies the causes of outbreaks of infectious disease. A good example is John Snow’s work in the Soho cholera epidemic. However, even today, epidemiologists are called in to find and remove the cause of outbreaks of disease such as food poisoning, Legionnaires disease, etc.
6
Descriptive epidemiology (1920) This describes the frequency and distribution of disease in populations and groups. In dentistry, systematic work of this kind appeared around the 1920’s and continues today monitoring the oral health of the population. This work enables us to estimate the need and demand for health services as well as providing important clues to the causes of disease. Analytic epidemiology (1950) This identifies the determinants or causes of disease. The methods used were developed in the 1950’s as part of the investigation of the role of smoking in lung cancer. Smoking as a cause of periodontal disease is now of considerable interest in dentistry. Experimental epidemiology (1970) This uses randomized controlled clinical trials to determine if new treatments are better than previous treatments in terms of controlling disease and improveing the ehalth of patients.
Clinical epidemiology (1980) This is the newest branch of epidemiology and the one likely to have the most immediate impact on dental practice. It involves the application of epidemiological principles and reasoning to the clinical problems involved in diagnosing and treating individual patients. More recently, this has come to be called evidence-based care. Molecular epidemiology (1990) This draws on advances in molecular biology and genetics to investigate biological parameters which are associated with an increased risk of disease. It also involves the use of biological indicators of the presence of disease and disease progression.
CURRENT APPLICATIONS OF EPIDEMIOLOGY Contemporary epidemiology is concerned with a number of important issues: 1. Monitoring the health of populations; 2. Investigating the natural history and cause of new diseases eg AIDS; 3. Identifying groups at high risk of disease so that treatment and prevention efforts can be targeted at those most in need;
7
4. Assessing the risks associated with exposure to harmful agents eg tobacco smoke, mercury, fluoride; 5. Assessing the effectiveness of treatment and preventive interventions; eg water fluoridation, topical fluoride treatments, sealants. 6. Assessing needs for health services, trends in use and ways of promoting access to care; 7. Assessing variations in diagnosis and treatment planning among health care practitioners.