Health Services Research

Key Points

The origin of health services research is discussed.
Quantitative tools, including cross-sectional, cohort, case-control, and difference-in-difference study designs, are briefly reviewed.
Qualitative tools, such as focus groups, surveys, and interviews, are covered.
Examples of study designs in perioperative care are used to highlight how health services research can be transformational.

Introduction to Health Services Research

In the wake of World War II and the rapidly evolving landscape of America's identity, the 1946 Hill Burton Act helped to fund the development of hospitals across the country. During the expansion of access to hospital-based services, John Cronin, then chief of the Division of Hospital and Medical Facilities, recognized the need for a new body of research dedicated to better understanding healthcare delivery and access. He not only helped to birth the broad field of health services research but also secured over $10 million in federal funding to jumpstart the field's inception. As health care and its delivery have expanded in complexity over the last several decades, the field of health services research has grown exponentially in response. Now, health services research is defined by the Agency for Healthcare Research and Quality as a “multidisciplinary field of scientific investigation that studies how social factors, financing systems, organizational structures and processes, health technologies, and personal behaviors affect access to health care, the quality and cost of health care, and ultimately our health and well-being.”

Health services research is intimately linked to the improvement of healthcare delivery through the concept of transformational research. Transformational research is defined by the National Science Foundation as “research driven by ideas that have the potential to radically change our understanding of an important existing scientific or engineering concept leading to the creation of a new paradigm or field of science or engineering. Such research is also characterized by its challenge to current understanding, or its pathway to new frontiers.” As such, transformational research serves as the mechanism through which large advances in care and major paradigm shifts are realized.

In this chapter, we aim to first highlight study designs commonly used in the field of health services research. Subsequently, through the conceptual framework of the five phases of surgical care developed by the American College of Surgeons (ACS), we will provide examples of how these study designs have been successfully implemented as examples of transformational research. Finally, we will summarize the most frequently used standardized reporting guidelines and offer a review of how health services research differs from quality improvement (QI).

Common Study Designs

Observational Study Designs

Before identifying opportunities for improvement and change, researchers must first be equipped with the tools to better understand existing systems in place. Observational studies allow researchers to highlight the existence and magnitude of problems present and see possible correlations between outcomes and exposures. Examples of observational studies include cross-sectional, cohort, and case-control studies. More advanced analytic methods, such as difference-in-difference studies, aid in further exploring correlational relationships.

Cross-sectional studies are a useful descriptive tool to examine relationships between outcomes and exposures within a limited timeframe. They are frequently described as “snapshots” in time. These studies may be hypothesis generating for future investigations, providing a powerful tool that can be relatively quick and inexpensive to perform. Given the limited temporal scope of the data, however, cross-sectional studies are most limited by their inability to offer correlational or causal explanations.

Other tools, including cohort and case-control studies, allow for further examinations into the possible correlation between exposures and outcomes of interest. Cohort studies are either retrospective or prospective longitudinal examinations of two study groups. The two study groups consist of one that is associated with a predefined exposure and a second control group that is like the first in most regards except for the exposure variable. The researcher then measures rates of the outcome of interest across the two groups and calculates a relative risk of an outcome associated with that exposure. The utility of cohort studies is most limited when examining rare outcomes. Otherwise, cohort studies are subject to the inherent limitations of prospective and retrospective designs. Prospective cohort studies allow for targeted data collection and can examine multiple outcomes from the same exposure, but they are typically costly and may require a lengthy study period depending on the relationship between the study exposure and outcomes. Retrospective studies can be quickly performed using pre-existing data, but their application is limited by the manner and fidelity with which the data was originally collected. Case-control studies are retrospective designs that start with a group containing the outcome of interest and a matched control. The researcher than retrospectively searches and compares the prevalence of an exposure between the case and control groups. They can be relatively quick to perform and are particularly useful in studying relatively rare outcomes. Case-control studies are limited in their ability to study rare exposures and are unable to offer estimates of incidence.

Advanced quasi-experimental study designs exist for specific questions in healthcare delivery. One such design using observational data is the difference-in-difference study. Difference-in-difference studies examine two matched groups over time after an intervention that is limited to only one of the groups. The difference-in-difference analysis then compares the relative difference between respective changes in study outcomes across these two groups over time. In other words, it examines the difference between the respective differences within each study group over time, hence its namesake. Although this can mimic large pseudorandomized retrospective studies examining the isolated effects of individual interventions, difference-in-difference studies are most limited by their internal validity based upon adequate matching between the control and intervention groups. This study design is often used for the evaluation of health policies.

Qualitative Studies

Qualitative studies allow for a different approach by recognizing the inherent limitations of objective data to capture the intangible and subjective drivers of complex relationships. Qualitative study designs are typically time-intensive and require specialized training. Given the breadth of qualitative methods, we will only highlight interviews, focus groups, and surveys, which are commonly employed when preparing for the implementation of a quality or safety intervention.

Surveys can be administered through a variety of platforms, including in-person, telephone, video-call, paper-based, and on the Internet. The instrument or questionnaire can be used to obtain information from a broad audience in a standardized fashion. This allows for improved ease of analysis and interpretability. Although often quick to distribute and administer, response rates can be subjective to response bias based on the medium. Additionally, the restrictive format of surveys can prevent subjects from offering clarifying details.

One-to-one interviews can be an immensely powerful tool in exploring intricate details regarding an individual's experience. Themes discovered in these interviews can be hypothesis-generating and lead to future qualitative and quantitative studies. They are limited by their time-intensiveness and their generalizability given the often small sample size. Interviews can be either open-ended or semistructured, allowing for either exploratory discussions or more directed and efficient data collection.

Focus groups allow researchers to not only collect information from numerous individuals but also gain unique insights through group interactions. This additional dimension of complexity can be both beneficial and harmful. Although group interactions may facilitate the discovery of new ideas otherwise not available in one-to-one interviews, focus groups require expertise in navigating and leading a group. Without appropriate guidance, focus groups are subject to dominant opinions and groupthink, which can be counterproductive.

You're Reading a Preview

Become a Clinical Tree membership for Full access and enjoy Unlimited articles

Become membership

If you are a member. Log in here