The natural history of disease refers to the progression of a disease from its preclinical state (prior to symptoms) to its clinical state (from onset of symptoms to cure, control, disability, or death). Disease is not something that occurs suddenly, but rather, it is a multifactorial process that is dynamic and occurs over time. It evolves and changes and is sometimes initiated by events that take place years, even decades, before symptoms first appear. Many diseases have a natural life history that can extend over a very long period of time. The natural history of disease is described in stages. Understanding the different stages allows for a better understanding of the approach to the prevention and control of disease.

Stage of Susceptibility

The stage of susceptibility refers to the time prior to disease development. In the presence of certain risk factors, genetics, or environment, disease may develop and the severity can vary among individuals. Risk factors are those factors that are associated with an increased likelihood of disease developing over time. The idea that individuals could modify “risk factors” tied to heart disease, stroke, and other diseases is one of the key findings of the Framingham Heart Study (National Heart, Lung, and Blood Institute and Boston University, 2010). Started in 1948 and still in operation, this study is one of the most important population studies ever carried out in the United States. Before Framingham, for example, most healthcare providers believed that atherosclerosis was an inevitable part of the aging process. Although not all risk factors are amenable to change (e.g., genetic factors), the identification of risk factors is important and fundamental to disease prevention.

Preclinical Stage of Disease

During the preclinical phase, the disease process has begun but there are no obvious symptoms. Although there is no clear manifestation of disease, because of the interaction of biological factors, changes have started to occur. During this stage, however, the changes are not always detectable. Screening technologies have been developed to detect the presence of some diseases before clinical symptoms appear. The Papanicolaou (Pap) smear is an example of an effective screening method for detecting cancer in a premalignant state to improve mortality related to cervical cancer. The use of the Pap smear as a screening tool facilitates early detection and treatment of premalignant changes of the cervix prior to development of malignancy.

Clinical Stage of Disease

The Nonclinical Disease Stage

This nonclinical or unapparent disease stage can be broken into four subparts. The first subpart is the preclinical stage, which, as mentioned earlier, is the acquisition of disease prior to development of symptoms and is destined to become disease. The second subpart is the subclinical stage that occurs when someone has the disease but it is not destined to develop clinically. The third subpart is the chronic or persistent stage of disease, which is disease that persists over time. And finally, there is the fourth subpart or latent stage in which one has disease with no active multiplication of the biologic agent (Gordis, 2014).

The Iceberg Phenomenon

For most health problems, the number of identified cases is exceeded by the number of unidentified cases. This occurrence, referred to as the “iceberg phenomenon,” makes it difficult to assess the true burden of disease. Many diseases do not have obvious symptoms, as stated earlier, and may go unrecognized for many years. Unrecognized diseases, such as diabetes, hypertension, and mental illness, create a significant problem with identifying populations at risk and estimating service needs. Complications also arise when patients are not recognized or treated during an early stage of a disease when interventions are most effective. Additionally, patients who do not have symptoms or do not recognize their symptoms do not seek medical care and, in many cases, even if they do have a diagnosis, will not take their medications as they perceive that they are healthy when they are asymptomatic.

PREVENTION

Primary Prevention

Primary prevention refers to the process of altering susceptibility or reducing exposure to susceptible individuals and includes general health promotion and specific measures designed to prevent disease prior to a person getting disease. Interventions designed for primary prevention are carried out during the stage of susceptibility and can include such things as providing immunizations to change a person’s susceptibility. Actions taken to prevent tobacco usage are another example of primary prevention. Tobacco use is one of the 12 leading health indicators used by Healthy People 2020 to measure health. Cigarette smoking is the leading cause of preventable morbidity and mortality in the United States (Centers for Disease Control and Prevention [CDC], 2014a), and prevention or cessation of smoking can reduce the development of many smoking-related diseases. Taxes on cigarettes, education programs, and support groups to help people stop smoking and the creation of smoke-free zones are all examples of primary prevention measures. The CDC linked a series of tobacco control efforts by Minnesota to a decrease in adult smoking prevalence rates. From 1999 to 2010, Minnesota implemented a series of antismoking initiatives, including a statewide smoke-free law, cigarette tax increases, media campaigns, and statewide cessation efforts. Adult smoking prevalence decreased from 22.1% in 1999 to 16.1% in 2010 (CDC, 2011). This is an excellent example of a statewide primary prevention effort to reduce smoking prevalence through a variety of initiatives.

Secondary Prevention

The early detection and prompt treatment of a disease at the earliest possible stage are referred to as secondary prevention. The goals of secondary prevention are to either identify and cure a disease at a very early stage or slow its progression to prevent complications and limit disability. Secondary prevention measures are carried out during the preclinical or presymptomatic stage of disease. Screening programs are designed to detect specific diseases in their early stages while they are curable and to prevent or reduce morbidity and mortality related to a later diagnosis of disease. Examples of secondary prevention include the Pap smear, mentioned earlier, as well as annual testing of cholesterol levels, mammography, and rapid HIV testing of asymptomatic individuals.

Tertiary Prevention

CAUSATION

The Epidemiological Triangle

The relationship between risk factors and disease is complex. Research studies may describe a relationship between a risk factor and disease, but how do we know this relationship is causal? An understanding of causation is important if APRNs want to effectively impact the health of populations. The epidemiological triangle is a model that has historically been used to explain causation. The model consists of three interactive factors: the causative agent (those factors for which presence or absence cause disease—biologic, chemical, physical, nutritional), a susceptible host (such things as age, gender, race, immune status, genetics), and the environment (including such diverse elements as water, food, neighborhood, pollution). A change in the agent, host, and environmental balance can lead to disease (Harkness, 1995). The underlying assumptions of this model are that causative factors can be both intrinsic and extrinsic to the host and that the cause of disease is related to interaction among these three factors. This model initially was developed to explain the transmission of infectious diseases and was particularly useful when the focus of epidemiology was on acute diseases. It is less helpful for understanding and explaining the more complicated processes associated with chronic disease. With the rise of chronic diseases as the primary cause of morbidity and mortality, a model that recognizes multiple causative factors was needed to better understand this complex interaction.

The Web of Causation

The dynamic nature of chronic diseases calls for a more sophisticated model for explaining causation than the epidemiological triangle. Introduction of the web of causation concept first appeared in the 1960s when chronic diseases overtook infectious diseases as the leading cause of morbidity and mortality in the United States. The foundation of this concept is that disease develops as the result of many antecedent factors and not as a result of a single, isolated cause. Each factor is itself the result of a complex pattern of events that can be best perceived as interrelated in the complex configuration of a web. The use of a web is helpful for visualizing how difficult it is to untangle the many events that can precede the onset of a chronic illness.

In order to compare rates in two or more groups, the events in the numerator must be defined in the same way, the time intervals must be the same, and the constant multiplier must be the same. Rates can be used to compare two different groups, or one group during two different time periods. Returning to the example about breastfeeding, the breastfeeding rates could be compared in the same hospital, but at two different times, before and after implementation of a planned intervention to increase breastfeeding rates.

An example of how prevalence rates are used in the literature is as follows:

Information on the rising prevalence of adult obesity in the United States has led to increased attention to factors that cause obesity (especially in children). This has led to the development of new programs aimed at primary and secondary prevention.

Mortality Rates

Mortality rates, also known as death rates, can be useful when evaluating and comparing populations. As stated earlier, there are many factors that can affect the natural history of disease, and measuring mortality allows investigators to compare death rates among and within populations. The formula for mortality rate is:

Standardization of crude rates is an important consideration when comparing mortality rates among populations. Standardization is used to control for the effects of age and other characteristics in order to make valid comparisons between groups. Age adjustment is an example of rate standardization and perhaps the most important one. No other factor has a larger effect on mortality than age. Consider the problem of comparing two communities with very different age distributions. One community has a much higher mortality rate for colon cancer than the other, leading investigators to consider a possible environmental hazard in that community, when in fact, that community’s population is older, which could account for the higher mortality. Direct age adjustment or standardization allows a researcher to eliminate the age disparities between two populations by using a standardized population. This allows the researcher to compare mortality or death rates between groups by eliminating age differences between populations and comparing actual age-adjusted mortality rates to determine whether age truly plays a role in the crude unadjusted mortality rates.

There are two methods of age adjustment: direct, as mentioned earlier, and indirect. The direct method applies observed age-specific mortality or death rates to a standardized population. The indirect method applies the age-specific rates of a standardized population to the age distribution of an observed population and is used to determine whether one population has a greater mortality because of an occupational hazard or risk compared to the general population (to learn how to perform age adjustment, refer to an advanced epidemiology text).

The CFR is usually expressed as a percentage; so in this case, one would multiply this rate by a constant multiplier of 100 to obtain the percentage of disease that is fatal. It is important in all of these rates to include those who have died from the disease in the denominator. Removing those who have died from the denominator falsely increases the CFR, making the disease appear more fatal or severe (Gordis, 2014).

The proportionate mortality ratio is useful for determining the leading causes of death. The formula for proportionate mortality ratio is as follows:

Before we discuss how to calculate and interpret survival data, we must touch on two important concepts, lead time bias and overdiagnosis bias. Lead time bias is a phenomenon whereby a patient is diagnosed earlier by screening and appears to have increased survival due to screening but rather dies at the same time he or she would regardless of screening. In other words, the time from which a patient is diagnosed earlier from screening is the lead time and the bias is the error that occurs as a result of concluding that screening leads to a longer survival after diagnosis. As can be seen in the following timeline, the survival time is longer when screening is implemented, but the ultimate time of death is unchanged. Although this is not true for all screening tests, it is important to recognize the phenomenon of lead time bias as it can affect the conclusions that are made regarding survival, which ultimately can affect a patient’s perceived prognosis.

Prognosis is calculated using survival rates. There are two methods of conducting survival analysis and estimating prognosis that are discussed. The first is the actuarial method, which measures the likelihood of surviving after each year of treatment (or a predetermined interval). This is calculated as follows: the probability of surviving 2 years if one survived 1 year, or the probability of surviving 3 years if one survived 2 years, and so on. Prognosis is most commonly described in the literature as the probability of surviving 1, 2, 3, or more years. Generally, survival is calculated as a probability P1, P2, P3, and so on. The survival after 1 year is designated as P1; if patients survived 1 year after treatment, those who survived to 2 years = P2; if patients survived 2 years after treatment, those who survived to 3 years = P3; and so on. To calculate P1, divide the number of survivors over the number of patients with the disease at the start of the study or treatment. It is important to note that those who are lost to follow-up (also known as withdrawals) or who are no longer studied must be removed from the denominator. When a study ends or is terminated, those patients are no longer followed and must be taken into consideration in your analysis, and this is called censorship. For simplicity, the following examples will assume no losses to follow-up, but it is important to recognize that those who are lost to follow-up or those who are censored must be taken into account in your calculations. Of note, in a more advanced epidemiology textbook, you will find that those who are lost to follow-up will be subtracted out of the denominator and multiplied by ½ to account for the chance they were at risk for half the interval. Again for the purposes of this text, we will use a hypothetical example in which no patients are lost to follow-up. By definition, here is how to calculate P1, P2, and so on:

To calculate the probability of surviving 1, 2, 3, or more years, the calculation is as follows:

In this example: