PBH 330 Determinants of Health and Health Equity

Original research studies are published in scholarly journals. There are many different types of research methodologies with varying levels of quality of evidence (see the evidence pyramid at the bottom of this page). Researchers sometimes do their own experiments in a laboratory or a real-world setting, and they might gather their own observational or empirical data; other times they analyze datasets that have been collected by other institutions (i.e. government or nonprofit agencies).

Here are some different research study methodologies with basic definitions.

1. Case report / Case study - This type of study focuses on one individual case. One major shortcoming of this type of study is that its findings cannot be generalized.

2. Case control studies - These studies compare two different case groups with one control group (for instance, one case group with a disease versus a control group that does not have a disease). These studies do not follow study subjects over time.

3. Cross-sectional study design - This common research method involves data collection at a single point in time. These most often take the form of survey data research. Their main limitation is that their findings are limited to only one point in time. 

4. Cohort studies - Two groups (cohorts) of people are studied, evaluating behaviors of those people who share some type of common characteristic or profile over multiple points in time (this is a type of longitudinal study). An example might be studying (at multiple points in time) a group of people in a certain demographic with one group of cigarette smokers and the other group who do not smoke.

5. Randomized controlled trial - This experimental study seeks to evaluate "the effectiveness of treatments [or interventions] especially treatments not administered in a clinical or laboratory setting, such as educational or economic interventions" (Elliot, et al. 2016).[1] As the name indicates, patients are randomly assigned to either a control group or a treatment group. These studies have high standards for the evaluation of the effectiveness of treatments (see the evidence pyramid below) and are also useful for the cost-benefit analysis of an intervention.

Definitions adapted or directly quoted from 

1. Elliot, M. Mark J., Fairweather, I., Olsen, W. K., & Pampaka, M. (Eds.). (2016). A dictionary of social research methods. Oxford University Press 

2. FIU Libraries. (n.d.). Research methods help guide. https://library.fiu.edu/researchmethods/studytypes

A systematic review is a type of scholarly article. It is not an original study, but rather, entails the analysis of the existing research on a research question. Some criteria for a systematic review are outlined below, using the Cochrane Handbook for Systematic Reviews of Intervention (2019, p. 4).

A systematic review:

1. is narrow and clear in its scope, focusing on "a specific research question" that it attempts to answer 

2. is comprehensive in its coverage in that it "attempts to collate all the empirical evidence that fits pre-specified eligibility criteria in order to answer [that question]"

3. is expected to address any "issues of bias in included studies"

4. analyzes "the included studies in order to draw conclusions"

5. outlines the research process so that it is clearly documented and capable of being reproduced by others.

A meta-analysis can be included in a systematic review or it can be considered it own type of review article. A meta-analysis involves the researcher's analysis/synthesis of statistical data to provide "comparisons of interventions, such as a new intervention versus a control intervention . . . or the comparison of two competing interventions" (2019, p. 17). 

A literature review (or narrative review) can be similar to a systematic review is some ways, but there are also notable differences. This type of review is also not an original study--though it can be included as a part of an original study as a kind of introductory overview of the literature related to the study. A narrative literature review differs from a systematic review in some of the following ways:

1. It is not expected to be comprehensive

2. The focus can be a general topic (or a broader research question)

3. The choice of which studies to include are determined by the researcher and therefore, not expected to be reproducible 

4. Provides a summary, evaluation, and synthesized analysis of the chosen literature but does not include substantial data analysis

5. Researcher can speculate on potential interventions or future studies

6. Because there is no defined methodology for inclusion/exclusion criteria, these tend to be less rigorous and have more potential for bias   

7. A brief literature review is sometimes included as a small section of an original research study to provide an overview of prior research

Reference: 

Ferrari, R. (2015). Writing narrative style literature reviews. Medical Writing, 24(4), 230–235. https://doi.org/10.1179/2047480615Z.000000000329

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International Lice from Duke University Medical Center Library and Archives.