Chapter 13: Exploratory Research: Observational Designs

INTRODUCTION

The complexity of human behavior and clinical phenomena present a considerable challenge to the clinician and researcher. In many situations, the impetus for a research study is the need to understand how human attributes and environmental characteristics interact to control behavioral responses. As clinical scientists continue to question how we can achieve optimal outcomes under defined clinical conditions, we must examine the multiple factors that influence our patients' and clients' lives. Studies that help us gather this information are considered observational because data are collected as they naturally exist, rather than through manipulation of variables as in experiments. Observational studies may be considered descriptive or exploratory. Descriptive studies will be described in the next chapter.

Exploratory research is the systematic investigation of relationships among two or more variables. Researchers use this approach to predict the effect of one variable on another, or to test relationships that are supported by clinical theory. Diagnostic and prognostic factors are identified through exploration of their relationships with the results of specific tests and patient outcomes. This type of research is usually guided by a set of hypotheses that help to structure measurements and interpretation of findings. For example, researchers have used this approach to study the relationship between long-term medical conditions and depression,¹ and to demonstrate the association of decreased muscle mass and strength with loss of mobility and function in older men and women.² Miles and colleagues³ used a predictive model to identify outcomes and risks associated with different subgroups of autism, to assist with prognostication and counseling.

Observational studies may involve prospective data collection or retrospective analysis of existing data, and may be designed using longitudinal or cross-sectional methods. Correlational methods are generally used to develop predictive models, which may be used for diagnostic or prognostic purposes. Observational studies may be used to determine the risk of disease associated with specific exposures. Such designs may include an explicit comparison group, allowing the investigator to determine if the rate of disease or disability is different for those exposed or unexposed to the factor of interest. Case-control and cohort studies are observational analytic designs that are intended to study risk factors associated with disease or disability and specific exposures or conditions. Methodological research uses correlational methods to focus on reliability and validity of measurement tools. Historical research provides an opportunity to use data from the past to analyze current issues. The purpose of this chapter is to describe these research approaches and their various configurations.

RETROSPECTIVE AND PROSPECTIVE RESEARCH

Exploratory research can be carried out retrospectively or prospectively (see Figure 13.1). In prospective research variables are measured through direct recording in the present. The researcher follows subjects as they progress through their treatment or evaluation. The researcher is thereby able to identify the factors that precede given outcomes.