Analyzing Health Problems for Risk Factors

The ability to identify risk factors and pathways for causation is essential for rational public health decisions and actions to address important health problems in a population. First, however, it is necessary to define what is meant by health problem. Here, health problem means a condition of humans that can be represented in terms of measurable health status or quality-of-life indicators. Later in this course, additional dimensions will be added to this basic definition for the purposes of community problem solving and the development of interventions. This characterization of a health problem as something measured only in terms of outcomes is difficult for some to accept. They point to important factors such as access to care or poverty itself and feel that these should rightfully be considered as health problems. Important problems they may be, but if truly important in the causation of some unacceptable health outcome, they can be dealt with as related factors rather than health problems.

The factors linked with specific health problems are often generically termed risk factors and can exist at one of three levels. Those risk factors most closely associated with the health outcome in question are often termed determinants. Risk factors that play a role further back the chain of causation are called direct and indirect contributing factors. Risk factors can be described at either an individual or a population level. For example, tobacco use for an individual increases the chances of developing heart disease or lung cancer, and an increased prevalence of tobacco use in a population increases that population’s incidence of (and mortality rates from) these conditions.

Determinants are scientifically established factors that relate directly to the level of a health problem. As the level of the determinant changes, the level of the health outcome changes. Determinants are the most proximal risk factors through which other levels of risk factors act. The link between the determinant and the health outcome should be well established through scientific or epidemiological studies. For example, for analyzing neonatal mortality rates, two well-established determinants are the low birth weight rate (the number of infants born weighing less than 2500 grams, or about 5.5 pounds, per 100 live births) and weight-specific mortality rates. Improvement in the neonatal mortality rate cannot occur unless one of these determinants improves. Health outcomes can have one or many determinants.

Direct contributing factors are scientifically established factors that directly affect the level of a determinant. Again, there should be solid evidence that the level of the direct determinant affects the level of the determinant. For the neonatal mortality rate example, the prevalence of tobacco use among pregnant women has been associated with the risk of low birth weight. A determinant can have many direct contributing factors. For low birth weight, other direct contributing factors include low maternal weight gain and inadequate prenatal care.

Indirect contributing factors affect the level of the direct contributing factors. Though less proximal to the health outcome in question, these factors are often proximal enough to be modified. The indirect contributing factor affects the level of the direct contributing factor, which, in turn, affects the level of the determinant. The level of the determinant then affects the level of the health outcome. Many indirect contributing factors can exist for each direct contributing factor. For prevalence of tobacco use among pregnant women, indirect contributing factors might include easy access to tobacco products for young women, lack of health education, and lack of smoking-cessation programs.



Scientifically established factor that relates directly to the level of the health problem. A health problem may have any number of determinants identified for it.

Example: Low birthweight is a prime determinant for the health problem neonatal mortality.

Direct Contributing Factor

Scientifically established factor that directly affects the level of the determinant.

Example: Use of prenatal care is one factor that affects the low birth weight rate.

Indirect Contributing Factor

Community-specific factor that affects the level of a direct contributing factor. Such factors can vary considerably from one community to another.

Example: Availability of day care or transportation services within the community may affect the use of prenatal care services.

Source: Definitions adapted from Centers for Disease Control and Prevention Public Health Practice Program Office.

The health problem analysis framework begins with the identification of a health problem (defined in terms of health status indicators) and proceeds to establish one or more determinants; for each determinant, one or more direct contributing factors; and for each direct contributing factor, one or more indirect contributing factors. Intervention strategies at the community level generally involve addressing these indirect contributing factors. When completed, an analysis identifies as many of the causal pathways as possible in order to determine which contributing factors exist in the setting in which an intervention strategy is planned. The framework for this approach is presented below and in the text. This framework forms the basis for developing meaningful interventions; it is used in several of the processes and instruments to assess community health needs that are currently in wide use at the local level. Community needs assessment processes and tools will be further described in Parts 5 and 7 of the course.

CHART: Health Problem Analysis Model - wpe1.jpg (19432 bytes)

This framework can be applied to many different health problems. The following example represents a partial application of this framework to the health problem infant mortality. Prime determinants for infant mortality are low birthweight and birthweight-specific survival. for low birthweight, there are many direct contributing factors---two of which are maternal tobacco use and utilization of prenatal care. Two factors that might influence maternal tobacco use would be tobacco advertising in the community and the cost of cigarettes (including taxes). Two factors that might influence use of prenatal care services might be the availability of transportation and day care services so that a woman would have the ability to visit her health care provider.  Again, there are many other factors that might influence tobacco use and prenatal care utilization and these need to be considered, as well. However it should be clear that the factors that are less proximal to the actual problem often are factors that exist at a level (often in the community) where something can be done about them. Direct and indirect contributing factors for birthweight-specific survival are also presented in the second figure. It is interesting to note that most of the progress that has been made in reducing infant mortality---and especially its neonatal death component---has come through improving birthweight-specific survival rather than through reducing the prevalence of low birthweight infants among all births.

CHART: Infant Mortality Example - wpe3.jpg (18124 bytes)

Although this framework is useful, it does not fully account for the relationships among the various levels of risk factors. Some direct contributing factors may affect more than one determinant, and some indirect contributing factors may influence more than one direct contributing factor. For example, illicit drug use during pregnancy influences both the likelihood of low birth weight and birth weight-specific survival rates. To account fully for these interactions, some direct and indirect contributing factors may need to be included in several different locations on the worksheet.

CHSC 400 Part 2 Analysis last revised December 15, 2003 (epowell)