1.4: How is Biological Aging Studied?

Two methods are commonly used to study biological aging, and each has advantages and disadvantages. The most reliable conclusions regarding aging are those supported by both types of study or a combination of the two.

Cross-Sectional Method

A cross-sectional study starts with a group of people of different ages who are placed into age categories. In some cases, each category may contain all individuals who have reached the same age in years; in other cases, each category may contain all individuals whose age in years falls within a selected range. For example, each range may span five years. Thus, one category may include all those between the ages of 45 and 49; the next category may include all those between the ages of 50 and 54; and so on. Alternatively, the age ranges may be of different sizes, such as all those age 50 through 59 and all those age 60 and above.

Once the categories have been established, the researchers measure characteristics such as an intelligence, muscle strength, or heart rate for each individual. The data for individuals in each category are then compared with the data from individuals in the other categories. In this way, correlations between differences in characteristics and increases in age can be identified. If a trend is observed, the researchers conclude that it is caused by increasing age.

Cross-sectional studies are very popular for several reasons. First, they can be done quickly, and so there is no need to wait for years while the subjects in the study age. Second, since each subject needs to be evaluated only once, many subjects can be tested and then released from the study. Therefore, this procedure is relatively inexpensive. Third, since many subjects are included in the study, the results are statistically reliable. Finally, these studies largely eliminate the problem of a period effect. A period effect is the influence of events or conditions during the study on the people being studied. For example, changes in the employment status of the subjects during an economic depression or a war cause period effects.

There are several drawbacks to cross-sectional studies. A very important one is that such studies do not really measure changes that occur as time passes. It is only inferred that the differences among the age groups result from the passage of time. These differences could be caused by other factors that affected the subjects before the study. This flaw in the basic design of cross-sectional studies is called a birth-cohort effect. For example, the individuals in certain age categories might have been different from those in the other age categories at birth. This could have resulted from immigration or from relocation of large segments of the population. Thus, one age category may have an overly large representation of individuals of one nationality. These individuals could be genetically different from individuals in another age category composed largely of people with a different nationality and genetic makeup. As another example, some cross-sectional studies show that there is a decrease in intelligence with aging. This difference may be due not to aging but to less opportunity and encouragement for those in the old-age categories to have attended school in their youth. This last problem can be identified by performing a time-lag study. It carries out the same cross-sectional study procedure after many years and makes comparisons between two groups of the same age category. For example, measurements of people who are 65 years old in 1990 could be compared with measurements of people who are 65 years old in 2010. Differences between these groups would reveal effects from differences in historical conditions.

Another design flaw in these studies is called differential mortality. It means that because of inborn differences in susceptibility to certain causes of death (e.g., certain infectious diseases), specific groups of individuals who would have been included in certain age categories have been inadvertently selected out of their categories because they died before the study began. Thus, there is a built-in bias among the age categories that has nothing to do with aging. Another problem with cross-sectional studies is that they measure only average changes. They cannot detect change in a single individual.

Overall, though cross-sectional studies sometimes detect true age changes, investigators using this technique may believe that they have found an age change where none exists. They also may conclude that an age change occurs faster or slower than it truly does.

Longitudinal Method

Another method for studying biological aging is the longitudinal study, in which a group of individuals of similar or identical chronological age is selected. Each individual is evaluated for the characteristics that are to be studied. Then, at specified intervals, the same individuals are evaluated in the same ways for the same characteristics. The intervals may be short, such as 1 year, or longer.

Longitudinal studies have several advantages over cross-sectional studies. First, they actually measure changes that occur as time passes; the relationship of the changes to aging is not simply inferred. Second, though they establish averages for a group as cross-sectional studies do, longitudinal studies can also detect age changes within the individual and can even establish the rate of change for each person. As a result, longitudinal studies reveal that different individuals age at different rates. As we will see later in this chapter, this is a very important finding.

By evaluating people periodically, longitudinal studies can also identify and measure the influence on aging of sudden events such as an accident or of long-term treatments or diseases. Alternatively, these studies can investigate the effects of aging on the course of a disease. Through careful analysis, longitudinal studies can establish the complicated interactive effects of several variables, such as the effects of changes in body weight on the way in which exercise affects the regulation of blood glucose. Finally, these studies can discover the predictive value of conditions present at one period of life on parameters such as future health and time of death.

Despite their many advantages, longitudinal studies on humans are not done as frequently as are cross-sectional studies because longitudinal studies have several negative characteristics. Of prime importance is the length of time needed to carry out such a study. It may be necessary to evaluate subjects over a period of many years. For example, if the study attempts to measure certain age changes from age 50 to age 80, the study must be conducted continuously for 30 years. During this period, many subjects may lose interest in the study, move away from the area where it is conducted, or die. The investigators themselves also face these problems. In addition, to achieve scientific reliability, the techniques for performing measurements of the characteristics of interest must remain basically the same despite technological advances. These factors cause a second drawback: Longitudinal studies usually cost a great deal more than do cross-sectional studies. Because of the expense, longitudinal studies usually include fewer subjects. Thus, after all the work, the results are not as statistically valid as those garnered from cross-sectional studies.

Longitudinal studies also contain certain design flaws. One is the period effect. For example, the results from a longitudinal study during a time of economic prosperity may be quite different from those obtained during a period of economic hardship. There is even a birth-cohort effect. This effect can be substantially reduced in longitudinal studies, but only by extending the studies over much longer periods.

Thus, while longitudinal studies can provide more and better information about biological aging than can cross-sectional studies, they do so only at great human and financial expense. Therefore, few long-term longitudinal studies have been conducted.

The BLSA started in 1958. At first it had only a few hundred male subjects, most of whom were at or beyond middle age. It now includes more than 700 volunteer subjects, both female and male, ranging in age from the twenties through age 90. Subjects receive a thorough evaluation, including numerous biological and psychological characteristics, every two to three years.

A third study method combines cross-sectional and longitudinal studies, forming a cross-sequential study. In this method, a cross-sectional study is performed and is then repeated after some years have passed. For example, people in five-year age categories extending from ages 40 to 70 could be evaluated in 1990, in 2000, and in 2010. Separate and combined comparisons could then be made among the groups at each time and among the three times. Using this complex method helps reduce problems from period effects, birth cohort effects, and differential mortality.

Nonhuman Studies

Besides being studied in humans, biological aging is studied in many animals, including mice, rats, flies, and worms. In addition, individual cells, both human and animal, are grown in nutrient materials to study biological aging. Though many of these studies have little or no immediate application to biological aging in humans, many others are used either as preliminary studies for future human studies or as experiments to support the results from human studies.

Animal and cell studies are very useful and important in the investigation of biological aging in humans since studying humans presents several problems. One is the genetic heterogeneity among people. This high degree of intrinsic variability results in considerable difficulty when one is interpreting data. Obviously, one cannot selectively breed people to achieve more genetic similarity among them, but selective breeding can be done with animals. Furthermore, environmental factors such as diet, temperature, and exercise can be controlled in animals to a degree that would be impossible in people. Even more control and freedom for study and experimentation are possible when one is dealing with individual cells.

Other factors make studying animals and cells desirable. Laboratory animals and cells take up less space, are less expensive to maintain, and have much shorter life spans than do humans. Unlike humans, certain lower animals (e.g., flies, worms) and cells apparently are not affected by psychological and emotional factors. Animals and cells can be humanely sacrificed for detailed anatomic and chemical analysis. Finally, unlike animals and cells, many elderly people have diseases and receive treatments that would affect the outcome of studies in which they might be subjects.

Still, humans must be studied directly if we are to increase our knowledge and understanding of the biology of human aging. It is known that aging in animals and individual cells differs from aging in humans in a variety of ways. For example, as some commonly used laboratory animals get older, they develop specific types of cancer and kidney disease that are not found in humans. Also, as some animals age, they have changes in hormone levels that are not seen in humans. The hormones involved may alter the aging processes in many parts of the body. Finally, certain chemicals and dietary substances are known to affect aging in animals differently from the way they affect aging in humans.