{"id":106,"date":"2024-09-16T16:39:18","date_gmt":"2024-09-16T16:39:18","guid":{"rendered":"https:\/\/pb-sandbox.library.illinois.edu\/introductiontoepidemiology\/chapter\/chapter-5-descriptive-and-analytical-epidemiological-study-designs\/"},"modified":"2024-09-29T18:57:20","modified_gmt":"2024-09-29T18:57:20","slug":"chapter-5-descriptive-and-analytical-epidemiological-study-designs","status":"publish","type":"chapter","link":"https:\/\/pb-sandbox.library.illinois.edu\/introductiontoepidemiology\/chapter\/chapter-5-descriptive-and-analytical-epidemiological-study-designs\/","title":{"raw":"Chapter 5: Descriptive and Analytical Epidemiological Study Designs","rendered":"Chapter 5: Descriptive and Analytical Epidemiological Study Designs"},"content":{"raw":"<div class=\"chapter-5:-descriptive-and-analytical-epidemiological-study-designs\">\r\n<h2>Overview<\/h2>\r\n<p class=\"import-NormalWeb\">Public health issues are often complex and involve analyzing the distribution, patterns, mechanisms, and dynamics of health-related states or events within a population. The first step towards better understanding of these issues is to formulate a specific question and then operationalize the answers to it through a scientific process. As an integral component of public health, the core function of epidemiology is to provide evidence-based information on health-related states or events to support decision-making in public health<sup>1,2<\/sup>. This chapter will present the key features of common epidemiology study designs, including their main characteristics, strengths, limitations, types of information generated, and applications in public health practice for both descriptive and analytic epidemiology study designs.<\/p>\r\n\r\n<div class=\"textbox textbox--learning-objectives\"><header class=\"textbox__header\">\r\n<p class=\"textbox__title\">Learning Objectives<\/p>\r\n\r\n<\/header>\r\n<div class=\"textbox__content\">\r\n<p class=\"import-Normal\">By the end of this chapter, you will be able to:<\/p>\r\n\r\n<ul>\r\n \t<li>Define the core elements and features used to classify epidemiological study designs.<\/li>\r\n \t<li>Differentiate descriptive and analytical epidemiology study designs.<\/li>\r\n \t<li>Describe the strengths and limitations of epidemiological study designs.<\/li>\r\n \t<li>Discuss the application of epidemiological study designs in public health practice.<\/li>\r\n<\/ul>\r\n<h2><\/h2>\r\n<\/div>\r\n<\/div>\r\n&nbsp;\r\n<h2><span style=\"font-size: 1.602em\">Epidemiological Study designs<\/span><\/h2>\r\n<p class=\"import-Normal\">Epidemiological study designs can be defined as a set of structured scientific methods and procedures aimed at operationalizing a well-designed research question. In other words, selecting an appropriate study design is a function of the research question. Therefore, researchers\/epidemiologists should consider the nature of <strong><em>what<\/em><\/strong> (i.e., case definition), <strong><em>who<\/em><\/strong> (i.e., people), <strong><em>where<\/em><\/strong> (i.e., place), and <strong><em>when<\/em><\/strong> (i.e., time) regarding health-related states or events, as well as test hypothesis to answer the <strong><em>why<\/em><\/strong> (i.e., cause, risk factors) and <strong><em>how<\/em><\/strong> (i.e., mechanism) they occur. Additionally, it is important to note that each epidemiological study design has intrinsic strengths and limitations related to its characteristics, including the processes of selecting, collecting, analyzing, and reporting data. The nature of the epidemiological study designs can be primarily classified as descriptive or analytical based on the purpose of the research question. As we covered in previous chapters, descriptive epidemiology focuses on describing the distribution and patterns of health-related outcomes in terms of people, place, and time, while analytic epidemiology allows for the assessment of hypotheses regarding the associations between exposures and outcomes.<sup>3<\/sup> Since disparities in the distribution of rates serve to generate hypotheses, descriptive epidemiology designs can also be referred to as hypothesis-generating studies. Before we delve into more details of the common epidemiological study designs, it is important to identify and describe key features of the core categories and elements often used to classify the types and subtypes of epidemiological study designs. Table 1 summarizes the characteristics, elements, and the key features for epidemiological study design classification.<\/p>\r\n<p class=\"import-NormalWeb\">Table 1. Characteristics, elements, and key features of epidemiological study design classification.<\/p>\r\n\r\n<div style=\"margin: auto\">\r\n<table style=\"width: 472.5pt\">\r\n<tbody>\r\n<tr class=\"TableGrid-R\" style=\"height: 0\">\r\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\r\n<p class=\"import-NormalWeb\" style=\"text-align: center\"><strong>C<\/strong><strong>haracteristics<\/strong><\/p>\r\n<\/td>\r\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\r\n<p class=\"import-NormalWeb\" style=\"text-align: center\"><strong>Elements<\/strong><\/p>\r\n<\/td>\r\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\r\n<p class=\"import-NormalWeb\" style=\"text-align: center\"><strong>Key feature<\/strong><strong>s<\/strong><\/p>\r\n<\/td>\r\n<\/tr>\r\n<tr class=\"TableGrid-R\" style=\"height: 5.3pt\">\r\n<td class=\"TableGrid-C\" style=\"background-color: #f2f2f2;vertical-align: middle;border: solid windowtext 0.5pt\" rowspan=\"2\">\r\n<p class=\"import-NormalWeb\" style=\"text-align: center\">Researcher\/<\/p>\r\n<p class=\"import-NormalWeb\" style=\"text-align: center\">epidemiologist role<\/p>\r\n<\/td>\r\n<td class=\"TableGrid-C\" style=\"background-color: #f2f2f2;vertical-align: middle;border: solid windowtext 0.5pt\">\r\n<p class=\"import-NormalWeb\" style=\"text-align: center\">Observational<\/p>\r\n<p class=\"import-NormalWeb\" style=\"text-align: center\">(Non-experimental)<\/p>\r\n<\/td>\r\n<td class=\"TableGrid-C\" style=\"background-color: #f2f2f2;border: solid windowtext 0.5pt\">\r\n<p class=\"import-NormalWeb\">Researcher does not manipulate any study factors (i.e., intervention, random allocation).<\/p>\r\n<p class=\"import-NormalWeb\">The distribution, pattern and\/or association between exposure(s) and outcome(s) are assessed without any researcher manipulation.<\/p>\r\n<\/td>\r\n<\/tr>\r\n<tr class=\"TableGrid-R\" style=\"height: 5.3pt\">\r\n<td class=\"TableGrid-C\" style=\"background-color: #f2f2f2;vertical-align: middle;border: solid windowtext 0.5pt\">\r\n<p class=\"import-NormalWeb\" style=\"text-align: center\">Experimental<\/p>\r\n<\/td>\r\n<td class=\"TableGrid-C\" style=\"background-color: #f2f2f2;border: solid windowtext 0.5pt\">\r\n<p class=\"import-NormalWeb\">Researcher introduces intervention, experiment, and or manipulate study factor(s).<\/p>\r\n<\/td>\r\n<\/tr>\r\n<tr class=\"TableGrid-R\" style=\"height: 14.35pt\">\r\n<td class=\"TableGrid-C\" style=\"vertical-align: middle;border: solid windowtext 0.5pt\" rowspan=\"2\">\r\n<p class=\"import-NormalWeb\" style=\"text-align: center\">Unit of observation<\/p>\r\n<\/td>\r\n<td class=\"TableGrid-C\" style=\"vertical-align: middle;border: solid windowtext 0.5pt\">\r\n<p class=\"import-NormalWeb\" style=\"text-align: center\">Individual level<\/p>\r\n<\/td>\r\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\r\n<p class=\"import-NormalWeb\">Observation, measure, or data collection is a person.<\/p>\r\n<\/td>\r\n<\/tr>\r\n<tr class=\"TableGrid-R\" style=\"height: 13.45pt\">\r\n<td class=\"TableGrid-C\" style=\"vertical-align: middle;border: solid windowtext 0.5pt\">\r\n<p class=\"import-NormalWeb\" style=\"text-align: center\">Aggregate<\/p>\r\n<\/td>\r\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\r\n<p class=\"import-NormalWeb\">Observation, measure, or data collection is an entire group, community, population.<\/p>\r\n<\/td>\r\n<\/tr>\r\n<tr class=\"TableGrid-R\" style=\"height: 13.45pt\">\r\n<td class=\"TableGrid-C\" style=\"background-color: #f2f2f2;vertical-align: middle;border: solid windowtext 0.5pt\" rowspan=\"2\">\r\n<p class=\"import-NormalWeb\" style=\"text-align: center\">Unit of<\/p>\r\n<p class=\"import-NormalWeb\" style=\"text-align: center\">analysis<\/p>\r\n<\/td>\r\n<td class=\"TableGrid-C\" style=\"background-color: #f2f2f2;vertical-align: middle;border: solid windowtext 0.5pt\">\r\n<p class=\"import-NormalWeb\" style=\"text-align: center\">Individual<\/p>\r\n<\/td>\r\n<td class=\"TableGrid-C\" style=\"background-color: #f2f2f2;border: solid windowtext 0.5pt\">\r\n<p class=\"import-NormalWeb\">Individual is the primary focus of the analysis.<\/p>\r\n<\/td>\r\n<\/tr>\r\n<tr class=\"TableGrid-R\" style=\"height: 9.4pt\">\r\n<td class=\"TableGrid-C\" style=\"background-color: #f2f2f2;vertical-align: middle;border: solid windowtext 0.5pt\">\r\n<p class=\"import-NormalWeb\" style=\"text-align: center\">Group or community<\/p>\r\n<\/td>\r\n<td class=\"TableGrid-C\" style=\"background-color: #f2f2f2;border: solid windowtext 0.5pt\">\r\n<p class=\"import-NormalWeb\">Group or community is the primary focus of the analysis.<\/p>\r\n<\/td>\r\n<\/tr>\r\n<tr class=\"TableGrid-R\" style=\"height: 15.25pt\">\r\n<td class=\"TableGrid-C\" style=\"vertical-align: middle;border: solid windowtext 0.5pt\" rowspan=\"3\">\r\n<p class=\"import-NormalWeb\" style=\"text-align: center\">Temporality (exposure and outcome)<\/p>\r\n<\/td>\r\n<td class=\"TableGrid-C\" style=\"vertical-align: middle;border: solid windowtext 0.5pt\">\r\n<p class=\"import-NormalWeb\" style=\"text-align: center\">Present<\/p>\r\n<\/td>\r\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\r\n<p class=\"import-NormalWeb\">Exposure(s) and outcome(s) are assessed at the same point in time.<\/p>\r\n<\/td>\r\n<\/tr>\r\n<tr class=\"TableGrid-R\" style=\"height: 29.65pt\">\r\n<td class=\"TableGrid-C\" style=\"vertical-align: middle;border: solid windowtext 0.5pt\">\r\n<p class=\"import-NormalWeb\" style=\"text-align: center\">Present past<\/p>\r\n<\/td>\r\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\r\n<p class=\"import-NormalWeb\">Outcome(s) is (are) measured in the present and exposure(s) is (are) measured retrospectively.<\/p>\r\n<\/td>\r\n<\/tr>\r\n<tr class=\"TableGrid-R\" style=\"height: 25.6pt\">\r\n<td class=\"TableGrid-C\" style=\"vertical-align: middle;border: solid windowtext 0.5pt\">\r\n<p class=\"import-NormalWeb\" style=\"text-align: center\">Present future<\/p>\r\n<\/td>\r\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\r\n<p class=\"import-NormalWeb\">Outcome(s) and exposure(s) can be measured both at the present time and during the follow-up.<\/p>\r\n<\/td>\r\n<\/tr>\r\n<tr class=\"TableGrid-R\" style=\"height: 15.25pt\">\r\n<td class=\"TableGrid-C\" style=\"background-color: #f2f2f2;vertical-align: middle;border: solid windowtext 0.5pt\" rowspan=\"2\">\r\n<p class=\"import-NormalWeb\" style=\"text-align: center\">Randomization<\/p>\r\n<\/td>\r\n<td class=\"TableGrid-C\" style=\"background-color: #f2f2f2;vertical-align: middle;border: solid windowtext 0.5pt\">\r\n<p class=\"import-NormalWeb\" style=\"text-align: center\">Yes<\/p>\r\n<\/td>\r\n<td class=\"TableGrid-C\" style=\"background-color: #f2f2f2;border: solid windowtext 0.5pt\">\r\n<p class=\"import-NormalWeb\">Individuals or clusters are randomized in different groups.<\/p>\r\n<\/td>\r\n<\/tr>\r\n<tr class=\"TableGrid-R\" style=\"height: 31pt\">\r\n<td class=\"TableGrid-C\" style=\"background-color: #f2f2f2;vertical-align: middle;border: solid windowtext 0.5pt\">\r\n<p class=\"import-NormalWeb\" style=\"text-align: center\">No<\/p>\r\n<\/td>\r\n<td class=\"TableGrid-C\" style=\"background-color: #f2f2f2;border: solid windowtext 0.5pt\">\r\n<p class=\"import-NormalWeb\">Individuals or clusters aren\u2019t randomly allocated into groups.<\/p>\r\n<p class=\"import-NormalWeb\">Not applied since there is only one group.<\/p>\r\n<p class=\"import-NormalWeb\">Nor applied since there is a single individual.<\/p>\r\n<\/td>\r\n<\/tr>\r\n<tr class=\"TableGrid-R\" style=\"height: 10.6pt\">\r\n<td class=\"TableGrid-C\" style=\"vertical-align: middle;border: solid windowtext 0.5pt\" rowspan=\"2\">\r\n<p class=\"import-NormalWeb\" style=\"text-align: center\">Comparison<\/p>\r\n<\/td>\r\n<td class=\"TableGrid-C\" style=\"vertical-align: middle;border: solid windowtext 0.5pt\">\r\n<p class=\"import-NormalWeb\" style=\"text-align: center\">No groups<\/p>\r\n<\/td>\r\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\r\n<p class=\"import-NormalWeb\">There are no groups to be compared.<\/p>\r\n<\/td>\r\n<\/tr>\r\n<tr class=\"TableGrid-R\" style=\"height: 0\">\r\n<td class=\"TableGrid-C\" style=\"vertical-align: middle;border: solid windowtext 0.5pt\">\r\n<p class=\"import-NormalWeb\" style=\"text-align: center\">Groups<\/p>\r\n<\/td>\r\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\r\n<p class=\"import-NormalWeb\">There are two or more groups to be compared.<\/p>\r\n<\/td>\r\n<\/tr>\r\n<tr class=\"TableGrid-R\" style=\"height: 0\">\r\n<td class=\"TableGrid-C\" style=\"background-color: #f2f2f2;vertical-align: middle;border: solid windowtext 0.5pt\" rowspan=\"3\">\r\n<p class=\"import-NormalWeb\" style=\"text-align: center\">Research Source<\/p>\r\n<\/td>\r\n<td class=\"TableGrid-C\" style=\"background-color: #f2f2f2;vertical-align: middle;border: solid windowtext 0.5pt\">\r\n<p class=\"import-NormalWeb\" style=\"text-align: center\">Primary<\/p>\r\n<\/td>\r\n<td class=\"TableGrid-C\" style=\"background-color: #f2f2f2;border: solid windowtext 0.5pt\">\r\n<p class=\"import-NormalWeb\">New and original data collected from participants.<\/p>\r\n<\/td>\r\n<\/tr>\r\n<tr class=\"TableGrid-R\" style=\"height: 0\">\r\n<td class=\"TableGrid-C\" style=\"background-color: #f2f2f2;vertical-align: middle;border: solid windowtext 0.5pt\">\r\n<p class=\"import-NormalWeb\" style=\"text-align: center\">Secondary<\/p>\r\n<\/td>\r\n<td class=\"TableGrid-C\" style=\"background-color: #f2f2f2;border: solid windowtext 0.5pt\">\r\n<p class=\"import-NormalWeb\">Existing data or records from participants or groups.<\/p>\r\n<\/td>\r\n<\/tr>\r\n<tr class=\"TableGrid-R\" style=\"height: 0\">\r\n<td class=\"TableGrid-C\" style=\"background-color: #f2f2f2;vertical-align: middle;border: solid windowtext 0.5pt\">\r\n<p class=\"import-NormalWeb\" style=\"text-align: center\">Tertiary<\/p>\r\n<\/td>\r\n<td class=\"TableGrid-C\" style=\"background-color: #f2f2f2;border: solid windowtext 0.5pt\">\r\n<p class=\"import-NormalWeb\">Review and synthesis of existing literature.<\/p>\r\n<\/td>\r\n<\/tr>\r\n<tr class=\"TableGrid-R\" style=\"height: 10.6pt\">\r\n<td class=\"TableGrid-C\" style=\"vertical-align: middle;border: solid windowtext 0.5pt\" rowspan=\"2\">\r\n<p class=\"import-NormalWeb\" style=\"text-align: center\">Statistical approach<\/p>\r\n<\/td>\r\n<td class=\"TableGrid-C\" style=\"vertical-align: middle;border: solid windowtext 0.5pt\">\r\n<p class=\"import-NormalWeb\" style=\"text-align: center\">Measures of occurrence<\/p>\r\n<\/td>\r\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\r\n<p class=\"import-NormalWeb\">Quantify the frequency, distribution, and\/or pattern of health-related states or events within a population. (e.g. prevalence, incidence, etc.).<\/p>\r\n<\/td>\r\n<\/tr>\r\n<tr class=\"TableGrid-R\" style=\"height: 0\">\r\n<td class=\"TableGrid-C\" style=\"vertical-align: middle;border: solid windowtext 0.5pt\">\r\n<p class=\"import-NormalWeb\" style=\"text-align: center\">Measures of association<\/p>\r\n<\/td>\r\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\r\n<p class=\"import-NormalWeb\">Quantify the relationship between exposure(s) and outcome (e.g., Relative Risk, Odds ratio, Hazard ratio, etc.).<\/p>\r\n<\/td>\r\n<\/tr>\r\n<tr class=\"TableGrid-R\" style=\"height: 0\">\r\n<td class=\"TableGrid-C\" style=\"background-color: #f2f2f2;vertical-align: middle;border: solid windowtext 0.5pt\" rowspan=\"3\">\r\n<p class=\"import-NormalWeb\" style=\"text-align: center\">Type of information<\/p>\r\n<\/td>\r\n<td class=\"TableGrid-C\" style=\"background-color: #f2f2f2;vertical-align: middle;border: solid windowtext 0.5pt\">\r\n<p class=\"import-NormalWeb\" style=\"text-align: center\">Descriptive<\/p>\r\n<\/td>\r\n<td class=\"TableGrid-C\" style=\"background-color: #f2f2f2;border: solid windowtext 0.5pt\">\r\n<p class=\"import-NormalWeb\">Provides a description of a case report or case series.<\/p>\r\n<p class=\"import-NormalWeb\">Provides an overview of the distribution and patterns of health-related events in terms of place, people, and time.<\/p>\r\n<\/td>\r\n<\/tr>\r\n<tr class=\"TableGrid-R\" style=\"height: 0\">\r\n<td class=\"TableGrid-C\" style=\"background-color: #f2f2f2;vertical-align: middle;border: solid windowtext 0.5pt\">\r\n<p class=\"import-NormalWeb\" style=\"text-align: center\">Analytical<\/p>\r\n<\/td>\r\n<td class=\"TableGrid-C\" style=\"background-color: #f2f2f2;border: solid windowtext 0.5pt\">\r\n<p class=\"import-NormalWeb\">Assesses associations between exposure(s) and outcome(s) to identify risk factors and\/or causes.<\/p>\r\n<\/td>\r\n<\/tr>\r\n<tr class=\"TableGrid-R\" style=\"height: 0\">\r\n<td class=\"TableGrid-C\" style=\"background-color: #f2f2f2;vertical-align: middle;border: solid windowtext 0.5pt\">\r\n<p class=\"import-NormalWeb\" style=\"text-align: center\">Analytical (experimental)<\/p>\r\n<\/td>\r\n<td class=\"TableGrid-C\" style=\"background-color: #f2f2f2;border: solid windowtext 0.5pt\">\r\n<p class=\"import-NormalWeb\">Provides evidence on the effects and\/or the mechanisms of interventions or treatments.<\/p>\r\n<\/td>\r\n<\/tr>\r\n<tr>\r\n<td><\/td>\r\n<td><\/td>\r\n<td><\/td>\r\n<\/tr>\r\n<\/tbody>\r\n<\/table>\r\n<\/div>\r\n<p class=\"import-Normal\">After defining the research question, epidemiological study designs are primarily classified at the study design level based on the role of the researcher\/epidemiologist. In this sense, study designs can be classified as experimental or observational (non-experimental). Descriptive epidemiologic study designs are observational, and the type of information they generate is descriptive. Experimental study designs and analytic observational study designs are classified as having the highest strength for generating evidence; however, they may not always be appropriate for addressing specific public health research questions.<\/p>\r\n[h5p id=\"2\"]\r\n<h2>Descriptive Epidemiological Study designs<\/h2>\r\n<p class=\"import-NormalWeb\">Descriptive epidemiological study designs are crucial for public health for several reasons, including: (a) identifying the extent of a public health issue, (b) identifying at-risk groups, (c) monitoring trends over time, (d) allocating public health resources, (e) setting priorities, and (f) informing planning, policy, and evaluation. Since study design depend on the research questions and descriptive epidemiology focuses on the distribution and patterns of outcome(s) in terms of <strong>place<\/strong>, <strong>people<\/strong>, and <strong>time<\/strong>, the most common related questions, though not limited to these, are:<\/p>\r\n<p class=\"import-NormalWeb\">What is the case definition?<\/p>\r\n<p class=\"import-NormalWeb\">Is the defined case of public health relevance?<\/p>\r\n<p class=\"import-NormalWeb\">What is the disease distribution in the population?<\/p>\r\n<p class=\"import-NormalWeb\">What are the patterns of the outcome in different populations?<\/p>\r\n<p class=\"import-NormalWeb\">Are there seasonal or temporal patterns in disease occurrence?<\/p>\r\n<p class=\"import-NormalWeb\">How does the distribution and pattern vary over time?<\/p>\r\n<p class=\"import-NormalWeb\">Have the rates of disease changed over different time periods?<\/p>\r\n<p class=\"import-NormalWeb\">Are there emerging trends or new patterns in disease distribution?<\/p>\r\n<p class=\"import-Normal\">Descriptive epidemiological studies include: (1) case report, (2) case series, and (3) ecological studies. Although a cross-sectional study design can be both descriptive and analytical depending on its purpose, we will present it in the analytic section because most of the available cross-sectional studies include an analytical component. From a descriptive point of view, in a cross-sectional design, the prevalence of an outcome is assessed for a given population at that specific point in time. The following sub-section will define each common descriptive study design including the main characteristics, strengths, limitations, type of information generated, and applications in public health practice.<\/p>\r\n\r\n<h3>Case report and case series<\/h3>\r\n<p class=\"import-Normal\">A case report is a research design that provides a detailed description of the clinical aspects of a particular health-related state or event from a single person within their real-life context. A case series is an extension of case report where multiple individuals are classified as the same case. These type of study designs are appropriate for reporting unusual, unexpected, unfamiliar, or rare health-related outcomes. This includes, but is not limited to, observation of symptoms, clinical findings, natural history of disease, side effects, and complications of interventions<sup>4<\/sup>. Although a case report and case series are classified as generating low levels in the hierarchy of evidence<sup>5<\/sup>, they provide unique information for an initial understanding of a novel situation. An in-depth narrative of the potential relationship of exposures explaining the outcome occurrence is usually conducted, described, and explained in chronological order. Generally, a follow-up evaluation includes the retrospective collection of information regarding potential exposures that might have led to the occurrence. Therefore, both designs might generate a potential hypothesis and provide a valuable information for identifying additional cases. Table 2 summarizes the appropriate use of this approach, as well as the strengths and limitations of the case report and case series study designs.<\/p>\r\n<p class=\"import-Normal\">Table 2. Strengths, limitations, and when to use case report and case series designs.<\/p>\r\n\r\n<table>\r\n<tbody>\r\n<tr class=\"TableGrid-R\">\r\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\r\n<p class=\"import-Normal\">When should these epidemiologic study designs be used?<\/p>\r\n<\/td>\r\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\r\n<ul>\r\n \t<li>When there is a unique, rare, unusual, or unexpected health-related state or event from a single person or few cases.<\/li>\r\n<\/ul>\r\n<\/td>\r\n<\/tr>\r\n<tr class=\"TableGrid-R\">\r\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\r\n<p class=\"import-Normal\">Strenghts<sup>4<\/sup><\/p>\r\n<\/td>\r\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\r\n<ul>\r\n \t<li>Provides in-depth information on a unique case or case series.<\/li>\r\n \t<li>Educational value.<\/li>\r\n \t<li>New hypotheses could be formulated.<\/li>\r\n \t<li>Help identify additional cases.<\/li>\r\n \t<li>Flexible structure.<\/li>\r\n<\/ul>\r\n<\/td>\r\n<\/tr>\r\n<tr class=\"TableGrid-R\">\r\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\r\n<p class=\"import-Normal\">Limitations<sup>4<\/sup><\/p>\r\n<\/td>\r\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\r\n<ul>\r\n \t<li>Lack of generalizability.<\/li>\r\n \t<li>Lack of internal validity<\/li>\r\n \t<li>Causal inference is not possible.<\/li>\r\n \t<li>No control groups.<\/li>\r\n \t<li>Selection bias.<\/li>\r\n<\/ul>\r\n<\/td>\r\n<\/tr>\r\n<tr>\r\n<td><\/td>\r\n<td><\/td>\r\n<\/tr>\r\n<\/tbody>\r\n<\/table>\r\n<h3>Ecological studies<\/h3>\r\n<p class=\"import-Normal\">An ecological study, often referred to as a correlation study<sup>2<\/sup>, is an observational research design that describes measures (e.g., rates, percentages, etc.) of exposures and outcome(s) at the population or group level rather than at the individual level. These groups can include geographical regions, communities, settings, or time periods. This is particularly important for public health planning because it allows the monitoring of population health and enables large-scale comparisons between and within groups over time. However, since data are collected and analyzed at group level, the generated information might not be true at individual level. The ecological fallacy is an error that occurs when conclusions about individual level are drawn from aggregate level. Table 3 summarizes the appropriate use of this study design, as well as the strengths and limitations.<\/p>\r\n<p class=\"import-Normal\" style=\"margin-left: 36pt;text-indent: 36pt\">Table 3. Strengths, limitations, and when to use ecological study design.<\/p>\r\n\r\n<table>\r\n<tbody>\r\n<tr class=\"TableGrid-R\">\r\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\r\n<p class=\"import-Normal\">When should this epidemiologic study design be used?<\/p>\r\n<\/td>\r\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\r\n<ul>\r\n \t<li>To identity and describe patterns and trends in health-related states or events across different populations, groups, or at different time points.<\/li>\r\n<\/ul>\r\n<\/td>\r\n<\/tr>\r\n<tr class=\"TableGrid-R\" style=\"height: 59.8pt\">\r\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\r\n<p class=\"import-Normal\">Strengths<\/p>\r\n<\/td>\r\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\r\n<ul>\r\n \t<li>Often require less time and resources compared to individual level studies.<\/li>\r\n \t<li>Useful for generating hypotheses.<\/li>\r\n \t<li>Useful for resource allocation and informing policy at the population level.<\/li>\r\n<\/ul>\r\n<\/td>\r\n<\/tr>\r\n<tr class=\"TableGrid-R\">\r\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\r\n<p class=\"import-Normal\">Limitations<\/p>\r\n<\/td>\r\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\r\n<ul>\r\n \t<li>Ecological fallacy.<\/li>\r\n \t<li>Information bias.<\/li>\r\n<\/ul>\r\n<\/td>\r\n<\/tr>\r\n<tr>\r\n<td><\/td>\r\n<td><\/td>\r\n<\/tr>\r\n<\/tbody>\r\n<\/table>\r\n<h2>Analytical Epidemiological Study designs<\/h2>\r\n<p class=\"import-NormalWeb\">Analytical epidemiological study designs focus on understanding the associations between exposure(s) and outcome(s). Thus, the common strategy of analytical designs for explaining why and\/or how a health outcome occurs involves testing hypotheses. This process often seeks to distinguish between causal and non-causal relationships as well as quantify the association between variables. Like descriptive study designs, each analytical study design serves a specific purpose and is chosen based on the research question and to provide specific answers for public health. This classification is important because it determines the strength of the evidence regarding the potential causation of health-related states and events. The most common related questions, though not limited to these, are:<\/p>\r\n<p class=\"import-NormalWeb\">What is the cause of a specific health outcome?<\/p>\r\n<p class=\"import-NormalWeb\">What is the mechanism underlying the causation of a specific health outcome?<\/p>\r\n<p class=\"import-NormalWeb\">What determinants and factors increase or decrease the risk of a particular health outcome?<\/p>\r\n<p class=\"import-NormalWeb\">Is there a dose-response relationship between the exposure and the health outcome?<\/p>\r\n<p class=\"import-NormalWeb\">How do relationships between exposures and outcomes differ among subgroups of the population?<\/p>\r\n<p class=\"import-NormalWeb\">How do health outcomes differ between groups with different exposure levels or characteristics?<\/p>\r\n<p class=\"import-NormalWeb\">What is the effectiveness of a specific intervention in reducing the incidence of a health condition?<\/p>\r\n<p class=\"import-NormalWeb\">How does the efficacy of an intervention translate to real-world settings?<\/p>\r\n<p class=\"import-Normal\">Analytical epidemiological study designs include: (1) Cross-sectional studies, (2) Case-control studies, (3) Cohort studies, and (4) Experimental studies. These studies are primarily classified based on the researcher role. In this sense, they can be classified as observational (non-experimental) or experimental, depending on whether the researcher manipulates the study factors (e.g., intervention, randomization) to investigate the relationship between exposure(s) and outcome(s). This classification is important because it determines the strength of the evidence regarding the potential causal relationship between the exposures and health-related states and events. The following sections will define each common analytical study designs including the main characteristics, strengths, limitations, types of information generated, and application in public health practice.<\/p>\r\n\r\n<h3>Cross-sectional studies<\/h3>\r\n<p class=\"import-Normal\">A cross-sectional study is an observational research design in which data are collected at the individual level. The main characteristic is that exposures and outcomes are measured simultaneously at a single point in time. In other words, cross-sectional studies are equivalent to a \u201cphoto\u201d of the health status of a given population. The type of information generated from an analytical perspective is that the association between exposure(s) and outcome(s) can be estimated; however, with a lower level of evidence compared to other analytic designs, and that is because exposure and outcomes are measured at the same time. For example, a cross-sectional study was conducted in May 2020 on the proportion of the population with Covid-19 antibodies according to socioeconomic status in Brazil<sup>11<\/sup>. Individuals were tested for Covid-19 and information on the family socioeconomic status was collected by questionnaire. The proportion of individuals with Covid-19 antibodies was significantly higher among the poorest quintile (2.1%) compared to the richest quintile (1.0%). In addition to assessing the burden of the health outcome, it helps public health with resource allocation. Table 4 summarizes the appropriate use of this design, as well as the strengths and limitations.<\/p>\r\n<p class=\"import-Normal\" style=\"margin-left: 36pt;text-indent: 36pt\">Table 4. Strengths, limitations, and when to use cross-sectional design.<\/p>\r\n\r\n<table>\r\n<tbody>\r\n<tr class=\"TableGrid-R\">\r\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\r\n<p class=\"import-Normal\">When should this epidemiological study design be used?<\/p>\r\n<\/td>\r\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\r\n<ul>\r\n \t<li>When researchers are interested to examine the prevalence and associate factors of a health-related outcome in a given population at a single point in time.<\/li>\r\n<\/ul>\r\n<\/td>\r\n<\/tr>\r\n<tr class=\"TableGrid-R\">\r\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\r\n<p class=\"import-Normal\">Strengths<\/p>\r\n<\/td>\r\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\r\n<ul>\r\n \t<li>Provides a \u201cphoto\u201d of the health-related outcome in a population.<\/li>\r\n \t<li>Less time-consuming to collect data.<\/li>\r\n \t<li>Multiple outcomes and exposures can be studied at a time.<\/li>\r\n \t<li>Useful for public health planning.<\/li>\r\n<\/ul>\r\n<\/td>\r\n<\/tr>\r\n<tr class=\"TableGrid-R\">\r\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\r\n<p class=\"import-Normal\">Limitations<\/p>\r\n<\/td>\r\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\r\n<ul>\r\n \t<li>Temporal limitation.<\/li>\r\n \t<li>Causal inference not possible.<\/li>\r\n \t<li>Selection bias.<\/li>\r\n<\/ul>\r\n<\/td>\r\n<\/tr>\r\n<tr>\r\n<td><\/td>\r\n<td><\/td>\r\n<\/tr>\r\n<\/tbody>\r\n<\/table>\r\n<h3>Case-control studies<\/h3>\r\n<p class=\"import-Normal\">A Case-control study is an observational research design aimed at investigating the potential causes of a health outcome by comparing two groups. The study groups are defined based on the presence or absence of the outcome variable under investigation. After identifying case and control groups, data on exposure(s) are collected retrospectively among cases (those with the outcome of interest) and controls. For example, 122 breast cancer patients and 121 controls were enrolled in a case-control study on the influence of lifetime physical activity on breast cancer risk<sup>12<\/sup>. Both cases and controls were asked about their lifetime physical activity involvement. Most physical activity reported by participants took place in the household domain. The study found no association between lifetime physical activity and breast cancer risk. On one hand, because the exposures are assessed retrospectively after identifying the cases and controls, this design is more prone to bias compared to designs that assess exposures before the occurrence of the outcome. On the other hand, when the outcome is rare, this design may be more appropriate since the cases have already been identified. Therefore, it is useful for public health, especially in early outbreak investigations. Table 5 summarizes the appropriate use of this design, as well as the strengths and limitations.<\/p>\r\n<p class=\"import-Normal\" style=\"margin-left: 36pt;text-indent: 36pt\">Table 5. Strengths, limitations, and when to use Case-control design.<\/p>\r\n\r\n<table>\r\n<tbody>\r\n<tr class=\"TableGrid-R\">\r\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\r\n<p class=\"import-Normal\">When should this epidemiologic study design be used?<\/p>\r\n<\/td>\r\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\r\n<ul>\r\n \t<li>To compare individuals with the outcome (case) to those without (controls) to identify differences in exposures.<\/li>\r\n<\/ul>\r\n<\/td>\r\n<\/tr>\r\n<tr class=\"TableGrid-R\">\r\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\r\n<p class=\"import-Normal\">Strengths<\/p>\r\n<\/td>\r\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\r\n<ul>\r\n \t<li>Cost-effective.<\/li>\r\n \t<li>Useful for rare outcomes.<\/li>\r\n \t<li>Useful for prolonged exposure.<\/li>\r\n<\/ul>\r\n<\/td>\r\n<\/tr>\r\n<tr class=\"TableGrid-R\">\r\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\r\n<p class=\"import-Normal\">Limitations<\/p>\r\n<\/td>\r\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\r\n<ul>\r\n \t<li>Recall bias.<\/li>\r\n \t<li>Selection bias.<\/li>\r\n \t<li>Not good for rare exposure.<\/li>\r\n<\/ul>\r\n<\/td>\r\n<\/tr>\r\n<tr>\r\n<td><\/td>\r\n<td><\/td>\r\n<\/tr>\r\n<\/tbody>\r\n<\/table>\r\n<h3>Cohort studies<\/h3>\r\n<p class=\"import-Normal\">Cohort studies are equivalent to a \u201cvideo\u201d of the health status of a given population. In cohort studies, data on exposure(s) is collected at baseline and participants are followed-up over time to monitor the presence or absence of outcomes. The longitudinal characteristics of this design and the assessment of exposures before the outcomes provide a better understanding of how several exposures influence the occurrence of health outcomes. This design allows researchers to establish temporal relationships between exposures and outcomes in a natural context, as there is no manipulation (i.e., intervention, random allocation). From a public health perspective, it is crucial for planning since evidence on potential causation and the risk of developing certain conditions over time is available. Table 6 summarizes the appropriate use of this design, as well as the strengths and limitations.<\/p>\r\n<p class=\"import-Normal\" style=\"margin-left: 36pt;text-indent: 36pt\">Table 6. Strengths, limitations, and when to use Cohort design.<\/p>\r\n\r\n<table>\r\n<tbody>\r\n<tr class=\"TableGrid-R\">\r\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\r\n<p class=\"import-Normal\">When should this epidemiologic study design be used?<\/p>\r\n<\/td>\r\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\r\n<ul>\r\n \t<li>When researchers are interested in establishing a temporal relationship between exposures and outcomes in a natural context without research manipulation.<\/li>\r\n<\/ul>\r\n<\/td>\r\n<\/tr>\r\n<tr class=\"TableGrid-R\">\r\n<td class=\"TableGrid-C\" style=\"background-color: transparent;border: solid windowtext 0.5pt\">\r\n<p class=\"import-Normal\">Strengths<\/p>\r\n<\/td>\r\n<td class=\"TableGrid-C\" style=\"background-color: transparent;border: solid windowtext 0.5pt\">\r\n<ul>\r\n \t<li>Temporal relationship<\/li>\r\n \t<li>Assessment of multiple exposures and outcomes<\/li>\r\n \t<li>Ethical advantage as compared to experimental studies<\/li>\r\n<\/ul>\r\n<\/td>\r\n<\/tr>\r\n<tr class=\"TableGrid-R\">\r\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\r\n<p class=\"import-Normal\">Limitations<\/p>\r\n<\/td>\r\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\r\n<ul>\r\n \t<li>Time-consuming<\/li>\r\n \t<li>Loss to follow-up<\/li>\r\n \t<li>Resource intensive<\/li>\r\n<\/ul>\r\n<\/td>\r\n<\/tr>\r\n<tr>\r\n<td><\/td>\r\n<td><\/td>\r\n<\/tr>\r\n<\/tbody>\r\n<\/table>\r\n<h3>Intervention studies<\/h3>\r\n<p class=\"import-Normal\">Intervention studies, also referred to as experimental studies, are designed to evaluate the effect of specific interventions on health outcomes. These studies can be classified into different types and sub-types depending on the research question (e.g., testing efficacy or effectiveness), the nature of the intervention, and the context (e.g., clinical trials, field trials) in which the intervention is applied. Unlike observational studies, the association between exposures and outcomes in intervention studies is evaluated under controlled conditions. By doing this, researchers can control the type, timing, and degree of the exposure and quantify the observed effects on health outcomes, ensuring that these effects are due to the intervention rather than other variables. One important aspect of intervention studies that distinguishes the quality of evidence in terms of establishing potential causal relationships is whether the study design includes randomization (a type of study factor manipulation). For example, in randomized controlled trials (RCTs), participants are randomly assigned to either the intervention group or the control group. This process helps minimize selection bias and balance potential confounders between groups. A quasi-experimental design is a type of intervention where researchers do not randomize participants into groups. Intervention studies play a vital role in public health for several reasons, but not limited to: (a) evaluating program effectiveness, (b) testing new treatments, and (c) Guiding policy decisions. Table 7 summarizes the appropriate use of this design, as well as the strengths and limitations.<\/p>\r\n<p class=\"import-Normal\" style=\"margin-left: 36pt;text-indent: 36pt\">Table 7. Strengths, limitations, and when to use Intervention design.<\/p>\r\n\r\n<table>\r\n<tbody>\r\n<tr class=\"TableGrid-R\">\r\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\r\n<p class=\"import-Normal\">When should this epidemiologic study design be used?<\/p>\r\n<\/td>\r\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\r\n<ul>\r\n \t<li>When researchers are interested to test efficacy and\/ or evaluate the effectiveness of an intervention on a specific outcome.<\/li>\r\n<\/ul>\r\n<\/td>\r\n<\/tr>\r\n<tr class=\"TableGrid-R\">\r\n<td class=\"TableGrid-C\" style=\"background-color: transparent;border: solid windowtext 0.5pt\">\r\n<p class=\"import-Normal\">Strengths<\/p>\r\n<\/td>\r\n<td class=\"TableGrid-C\" style=\"background-color: transparent;border: solid windowtext 0.5pt\">\r\n<ul>\r\n \t<li>Temporal relationship<\/li>\r\n \t<li>Potential causal relationship<\/li>\r\n \t<li>Control over exposure and\/or groups<\/li>\r\n<\/ul>\r\n<\/td>\r\n<\/tr>\r\n<tr class=\"TableGrid-R\">\r\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\r\n<p class=\"import-Normal\">Limitations<\/p>\r\n<\/td>\r\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\r\n<ul>\r\n \t<li>Cost and time<\/li>\r\n \t<li>Resource intensive<\/li>\r\n \t<li>Ethical concerns<\/li>\r\n<\/ul>\r\n<\/td>\r\n<\/tr>\r\n<tr>\r\n<td><\/td>\r\n<td><\/td>\r\n<\/tr>\r\n<\/tbody>\r\n<\/table>\r\n<div class=\"textbox textbox--key-takeaways\"><header class=\"textbox__header\">\r\n<p class=\"textbox__title\">Key Takeaways<\/p>\r\n\r\n<\/header>\r\n<div class=\"textbox__content\">\r\n<p class=\"import-Normal\">In this chapter, we covered common epidemiology studies designs, highlighting the main characteristics, strengths, limitations, and applications for public health practices. The decision about what study design to use relies on the research question and hypotheses the researcher would like to test. Descriptive designs are vital for identifying public health issues, occurrence, patterns, and trends in terms of people, place and time, while analytic designs provide insights into causal and non-causal relationships between exposures and outcomes in a population. Understanding these study designs enhances the ability to address complex public health issues and informs decision-making processes in public health practice.<\/p>\r\n\r\n<\/div>\r\n<\/div>\r\n&nbsp;\r\n<p class=\"import-Normal\">References<\/p>\r\n\r\n<ul>\r\n \t<li>Holland, W.W., Olsen, J., Florey, C.D.V., et al. (eds.) (2007). The Development of Modern Epidemiology: Personal Reports from Those Who Were There. Oxford: Oxford University Press.<\/li>\r\n \t<li><span lang=\"pt-BR\" xml:lang=\"pt-BR\">Bonita R, Beaglehole R, Kjellstr\u00f6m T. Basic Epidemiology. <\/span>World Health Organization; 2006. <a class=\"rId8\" href=\"https:\/\/apps.who.int\/iris\/bitstream\/handle\/10665\/43541\/9241547073_eng.pdf\"><span class=\"import-Hyperlink\">https:\/\/apps.who.int\/iris\/bitstream\/handle\/10665\/43541\/9241547073_eng.pdf<\/span><\/a>. Accessed July 18, 2028.<\/li>\r\n \t<li>Sapkota, K. (2023). Descriptive and Analytical Epidemiology. In Statistical Approaches for Epidemiology: From Concept to Application (pp. 1-18). Cham: Springer International Publishing.<\/li>\r\n \t<li>Nissen T, Wynn R. The clinical case report: a review of its merits and limitations. BMC Res Notes. 2014 Apr 23;7:264. doi: 10.1186\/1756-0500-7-264.<\/li>\r\n \t<li>Guyatt, G. H., Oxman, A. D., Vist, G. E., Kunz, R., Falck-Ytter, Y., Alonso-Coello, P., &amp; Sch\u00fcnemann, H. J. (2008). GRADE: an emerging consensus on rating quality of evidence and strength of recommendations. Bmj, 336(7650), 924-926.<\/li>\r\n \t<li>Holland, W.W., Olsen, J., Florey, C.D.V., et al. (eds.) (2007). The Development of Modern Epidemiology: Personal Reports from Those Who Were There. Oxford: Oxford University Press.<\/li>\r\n \t<li><span lang=\"pt-BR\" xml:lang=\"pt-BR\">Bonita R, Beaglehole R, Kjellstr\u00f6m T. Basic Epidemiology. <\/span>World Health Organization; 2006. <a class=\"rId9\" href=\"https:\/\/apps.who.int\/iris\/bitstream\/handle\/10665\/43541\/9241547073_eng.pdf\"><span class=\"import-Hyperlink\">https:\/\/apps.who.int\/iris\/bitstream\/handle\/10665\/43541\/9241547073_eng.pdf<\/span><\/a>. Accessed July 18, 2028.<\/li>\r\n \t<li>Sapkota, K. (2023). Descriptive and Analytical Epidemiology. In Statistical Approaches for Epidemiology: From Concept to Application (pp. 1-18). Cham: Springer International Publishing.<\/li>\r\n \t<li>Nissen T, Wynn R. The clinical case report: a review of its merits and limitations. BMC Res Notes. 2014 Apr 23;7:264. doi: 10.1186\/1756-0500-7-264.<\/li>\r\n \t<li>Guyatt, G. H., Oxman, A. D., Vist, G. E., Kunz, R., Falck-Ytter, Y., Alonso-Coello, P., &amp; Sch\u00fcnemann, H. J. (2008). GRADE: an emerging consensus on rating quality of evidence and strength of recommendations. Bmj, 336(7650), 924-926.<\/li>\r\n \t<li>Hallal PC, Hartwig FP, Horta BL, et al. SARS-CoV-2 antibody prevalence in Brazil: results from two successive nationwide serological household surveys. Lancet Glob Health. 2020;8(11):e1390-e1398. doi:10.1016\/S2214-109X(20)30387-9<\/li>\r\n \t<li>Yen SH, Knight A, Krishna M, Muda W, Rufai A. Lifetime Physical Activity and Breast Cancer: a Case-Control Study in Kelantan, Malaysia. Asian Pac J Cancer Prev. 2016;17(8):4083-4088.<\/li>\r\n<\/ul>\r\n<p class=\"import-Normal\"><\/p>\r\n\r\n<\/div>","rendered":"<div class=\"chapter-5:-descriptive-and-analytical-epidemiological-study-designs\">\n<h2>Overview<\/h2>\n<p class=\"import-NormalWeb\">Public health issues are often complex and involve analyzing the distribution, patterns, mechanisms, and dynamics of health-related states or events within a population. The first step towards better understanding of these issues is to formulate a specific question and then operationalize the answers to it through a scientific process. As an integral component of public health, the core function of epidemiology is to provide evidence-based information on health-related states or events to support decision-making in public health<sup>1,2<\/sup>. This chapter will present the key features of common epidemiology study designs, including their main characteristics, strengths, limitations, types of information generated, and applications in public health practice for both descriptive and analytic epidemiology study designs.<\/p>\n<div class=\"textbox textbox--learning-objectives\">\n<header class=\"textbox__header\">\n<p class=\"textbox__title\">Learning Objectives<\/p>\n<\/header>\n<div class=\"textbox__content\">\n<p class=\"import-Normal\">By the end of this chapter, you will be able to:<\/p>\n<ul>\n<li>Define the core elements and features used to classify epidemiological study designs.<\/li>\n<li>Differentiate descriptive and analytical epidemiology study designs.<\/li>\n<li>Describe the strengths and limitations of epidemiological study designs.<\/li>\n<li>Discuss the application of epidemiological study designs in public health practice.<\/li>\n<\/ul>\n<h2><\/h2>\n<\/div>\n<\/div>\n<p>&nbsp;<\/p>\n<h2><span style=\"font-size: 1.602em\">Epidemiological Study designs<\/span><\/h2>\n<p class=\"import-Normal\">Epidemiological study designs can be defined as a set of structured scientific methods and procedures aimed at operationalizing a well-designed research question. In other words, selecting an appropriate study design is a function of the research question. Therefore, researchers\/epidemiologists should consider the nature of <strong><em>what<\/em><\/strong> (i.e., case definition), <strong><em>who<\/em><\/strong> (i.e., people), <strong><em>where<\/em><\/strong> (i.e., place), and <strong><em>when<\/em><\/strong> (i.e., time) regarding health-related states or events, as well as test hypothesis to answer the <strong><em>why<\/em><\/strong> (i.e., cause, risk factors) and <strong><em>how<\/em><\/strong> (i.e., mechanism) they occur. Additionally, it is important to note that each epidemiological study design has intrinsic strengths and limitations related to its characteristics, including the processes of selecting, collecting, analyzing, and reporting data. The nature of the epidemiological study designs can be primarily classified as descriptive or analytical based on the purpose of the research question. As we covered in previous chapters, descriptive epidemiology focuses on describing the distribution and patterns of health-related outcomes in terms of people, place, and time, while analytic epidemiology allows for the assessment of hypotheses regarding the associations between exposures and outcomes.<sup>3<\/sup> Since disparities in the distribution of rates serve to generate hypotheses, descriptive epidemiology designs can also be referred to as hypothesis-generating studies. Before we delve into more details of the common epidemiological study designs, it is important to identify and describe key features of the core categories and elements often used to classify the types and subtypes of epidemiological study designs. Table 1 summarizes the characteristics, elements, and the key features for epidemiological study design classification.<\/p>\n<p class=\"import-NormalWeb\">Table 1. Characteristics, elements, and key features of epidemiological study design classification.<\/p>\n<div style=\"margin: auto\">\n<table style=\"width: 472.5pt\">\n<tbody>\n<tr class=\"TableGrid-R\" style=\"height: 0\">\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\n<p class=\"import-NormalWeb\" style=\"text-align: center\"><strong>C<\/strong><strong>haracteristics<\/strong><\/p>\n<\/td>\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\n<p class=\"import-NormalWeb\" style=\"text-align: center\"><strong>Elements<\/strong><\/p>\n<\/td>\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\n<p class=\"import-NormalWeb\" style=\"text-align: center\"><strong>Key feature<\/strong><strong>s<\/strong><\/p>\n<\/td>\n<\/tr>\n<tr class=\"TableGrid-R\" style=\"height: 5.3pt\">\n<td class=\"TableGrid-C\" style=\"background-color: #f2f2f2;vertical-align: middle;border: solid windowtext 0.5pt\" rowspan=\"2\">\n<p class=\"import-NormalWeb\" style=\"text-align: center\">Researcher\/<\/p>\n<p class=\"import-NormalWeb\" style=\"text-align: center\">epidemiologist role<\/p>\n<\/td>\n<td class=\"TableGrid-C\" style=\"background-color: #f2f2f2;vertical-align: middle;border: solid windowtext 0.5pt\">\n<p class=\"import-NormalWeb\" style=\"text-align: center\">Observational<\/p>\n<p class=\"import-NormalWeb\" style=\"text-align: center\">(Non-experimental)<\/p>\n<\/td>\n<td class=\"TableGrid-C\" style=\"background-color: #f2f2f2;border: solid windowtext 0.5pt\">\n<p class=\"import-NormalWeb\">Researcher does not manipulate any study factors (i.e., intervention, random allocation).<\/p>\n<p class=\"import-NormalWeb\">The distribution, pattern and\/or association between exposure(s) and outcome(s) are assessed without any researcher manipulation.<\/p>\n<\/td>\n<\/tr>\n<tr class=\"TableGrid-R\" style=\"height: 5.3pt\">\n<td class=\"TableGrid-C\" style=\"background-color: #f2f2f2;vertical-align: middle;border: solid windowtext 0.5pt\">\n<p class=\"import-NormalWeb\" style=\"text-align: center\">Experimental<\/p>\n<\/td>\n<td class=\"TableGrid-C\" style=\"background-color: #f2f2f2;border: solid windowtext 0.5pt\">\n<p class=\"import-NormalWeb\">Researcher introduces intervention, experiment, and or manipulate study factor(s).<\/p>\n<\/td>\n<\/tr>\n<tr class=\"TableGrid-R\" style=\"height: 14.35pt\">\n<td class=\"TableGrid-C\" style=\"vertical-align: middle;border: solid windowtext 0.5pt\" rowspan=\"2\">\n<p class=\"import-NormalWeb\" style=\"text-align: center\">Unit of observation<\/p>\n<\/td>\n<td class=\"TableGrid-C\" style=\"vertical-align: middle;border: solid windowtext 0.5pt\">\n<p class=\"import-NormalWeb\" style=\"text-align: center\">Individual level<\/p>\n<\/td>\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\n<p class=\"import-NormalWeb\">Observation, measure, or data collection is a person.<\/p>\n<\/td>\n<\/tr>\n<tr class=\"TableGrid-R\" style=\"height: 13.45pt\">\n<td class=\"TableGrid-C\" style=\"vertical-align: middle;border: solid windowtext 0.5pt\">\n<p class=\"import-NormalWeb\" style=\"text-align: center\">Aggregate<\/p>\n<\/td>\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\n<p class=\"import-NormalWeb\">Observation, measure, or data collection is an entire group, community, population.<\/p>\n<\/td>\n<\/tr>\n<tr class=\"TableGrid-R\" style=\"height: 13.45pt\">\n<td class=\"TableGrid-C\" style=\"background-color: #f2f2f2;vertical-align: middle;border: solid windowtext 0.5pt\" rowspan=\"2\">\n<p class=\"import-NormalWeb\" style=\"text-align: center\">Unit of<\/p>\n<p class=\"import-NormalWeb\" style=\"text-align: center\">analysis<\/p>\n<\/td>\n<td class=\"TableGrid-C\" style=\"background-color: #f2f2f2;vertical-align: middle;border: solid windowtext 0.5pt\">\n<p class=\"import-NormalWeb\" style=\"text-align: center\">Individual<\/p>\n<\/td>\n<td class=\"TableGrid-C\" style=\"background-color: #f2f2f2;border: solid windowtext 0.5pt\">\n<p class=\"import-NormalWeb\">Individual is the primary focus of the analysis.<\/p>\n<\/td>\n<\/tr>\n<tr class=\"TableGrid-R\" style=\"height: 9.4pt\">\n<td class=\"TableGrid-C\" style=\"background-color: #f2f2f2;vertical-align: middle;border: solid windowtext 0.5pt\">\n<p class=\"import-NormalWeb\" style=\"text-align: center\">Group or community<\/p>\n<\/td>\n<td class=\"TableGrid-C\" style=\"background-color: #f2f2f2;border: solid windowtext 0.5pt\">\n<p class=\"import-NormalWeb\">Group or community is the primary focus of the analysis.<\/p>\n<\/td>\n<\/tr>\n<tr class=\"TableGrid-R\" style=\"height: 15.25pt\">\n<td class=\"TableGrid-C\" style=\"vertical-align: middle;border: solid windowtext 0.5pt\" rowspan=\"3\">\n<p class=\"import-NormalWeb\" style=\"text-align: center\">Temporality (exposure and outcome)<\/p>\n<\/td>\n<td class=\"TableGrid-C\" style=\"vertical-align: middle;border: solid windowtext 0.5pt\">\n<p class=\"import-NormalWeb\" style=\"text-align: center\">Present<\/p>\n<\/td>\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\n<p class=\"import-NormalWeb\">Exposure(s) and outcome(s) are assessed at the same point in time.<\/p>\n<\/td>\n<\/tr>\n<tr class=\"TableGrid-R\" style=\"height: 29.65pt\">\n<td class=\"TableGrid-C\" style=\"vertical-align: middle;border: solid windowtext 0.5pt\">\n<p class=\"import-NormalWeb\" style=\"text-align: center\">Present past<\/p>\n<\/td>\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\n<p class=\"import-NormalWeb\">Outcome(s) is (are) measured in the present and exposure(s) is (are) measured retrospectively.<\/p>\n<\/td>\n<\/tr>\n<tr class=\"TableGrid-R\" style=\"height: 25.6pt\">\n<td class=\"TableGrid-C\" style=\"vertical-align: middle;border: solid windowtext 0.5pt\">\n<p class=\"import-NormalWeb\" style=\"text-align: center\">Present future<\/p>\n<\/td>\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\n<p class=\"import-NormalWeb\">Outcome(s) and exposure(s) can be measured both at the present time and during the follow-up.<\/p>\n<\/td>\n<\/tr>\n<tr class=\"TableGrid-R\" style=\"height: 15.25pt\">\n<td class=\"TableGrid-C\" style=\"background-color: #f2f2f2;vertical-align: middle;border: solid windowtext 0.5pt\" rowspan=\"2\">\n<p class=\"import-NormalWeb\" style=\"text-align: center\">Randomization<\/p>\n<\/td>\n<td class=\"TableGrid-C\" style=\"background-color: #f2f2f2;vertical-align: middle;border: solid windowtext 0.5pt\">\n<p class=\"import-NormalWeb\" style=\"text-align: center\">Yes<\/p>\n<\/td>\n<td class=\"TableGrid-C\" style=\"background-color: #f2f2f2;border: solid windowtext 0.5pt\">\n<p class=\"import-NormalWeb\">Individuals or clusters are randomized in different groups.<\/p>\n<\/td>\n<\/tr>\n<tr class=\"TableGrid-R\" style=\"height: 31pt\">\n<td class=\"TableGrid-C\" style=\"background-color: #f2f2f2;vertical-align: middle;border: solid windowtext 0.5pt\">\n<p class=\"import-NormalWeb\" style=\"text-align: center\">No<\/p>\n<\/td>\n<td class=\"TableGrid-C\" style=\"background-color: #f2f2f2;border: solid windowtext 0.5pt\">\n<p class=\"import-NormalWeb\">Individuals or clusters aren\u2019t randomly allocated into groups.<\/p>\n<p class=\"import-NormalWeb\">Not applied since there is only one group.<\/p>\n<p class=\"import-NormalWeb\">Nor applied since there is a single individual.<\/p>\n<\/td>\n<\/tr>\n<tr class=\"TableGrid-R\" style=\"height: 10.6pt\">\n<td class=\"TableGrid-C\" style=\"vertical-align: middle;border: solid windowtext 0.5pt\" rowspan=\"2\">\n<p class=\"import-NormalWeb\" style=\"text-align: center\">Comparison<\/p>\n<\/td>\n<td class=\"TableGrid-C\" style=\"vertical-align: middle;border: solid windowtext 0.5pt\">\n<p class=\"import-NormalWeb\" style=\"text-align: center\">No groups<\/p>\n<\/td>\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\n<p class=\"import-NormalWeb\">There are no groups to be compared.<\/p>\n<\/td>\n<\/tr>\n<tr class=\"TableGrid-R\" style=\"height: 0\">\n<td class=\"TableGrid-C\" style=\"vertical-align: middle;border: solid windowtext 0.5pt\">\n<p class=\"import-NormalWeb\" style=\"text-align: center\">Groups<\/p>\n<\/td>\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\n<p class=\"import-NormalWeb\">There are two or more groups to be compared.<\/p>\n<\/td>\n<\/tr>\n<tr class=\"TableGrid-R\" style=\"height: 0\">\n<td class=\"TableGrid-C\" style=\"background-color: #f2f2f2;vertical-align: middle;border: solid windowtext 0.5pt\" rowspan=\"3\">\n<p class=\"import-NormalWeb\" style=\"text-align: center\">Research Source<\/p>\n<\/td>\n<td class=\"TableGrid-C\" style=\"background-color: #f2f2f2;vertical-align: middle;border: solid windowtext 0.5pt\">\n<p class=\"import-NormalWeb\" style=\"text-align: center\">Primary<\/p>\n<\/td>\n<td class=\"TableGrid-C\" style=\"background-color: #f2f2f2;border: solid windowtext 0.5pt\">\n<p class=\"import-NormalWeb\">New and original data collected from participants.<\/p>\n<\/td>\n<\/tr>\n<tr class=\"TableGrid-R\" style=\"height: 0\">\n<td class=\"TableGrid-C\" style=\"background-color: #f2f2f2;vertical-align: middle;border: solid windowtext 0.5pt\">\n<p class=\"import-NormalWeb\" style=\"text-align: center\">Secondary<\/p>\n<\/td>\n<td class=\"TableGrid-C\" style=\"background-color: #f2f2f2;border: solid windowtext 0.5pt\">\n<p class=\"import-NormalWeb\">Existing data or records from participants or groups.<\/p>\n<\/td>\n<\/tr>\n<tr class=\"TableGrid-R\" style=\"height: 0\">\n<td class=\"TableGrid-C\" style=\"background-color: #f2f2f2;vertical-align: middle;border: solid windowtext 0.5pt\">\n<p class=\"import-NormalWeb\" style=\"text-align: center\">Tertiary<\/p>\n<\/td>\n<td class=\"TableGrid-C\" style=\"background-color: #f2f2f2;border: solid windowtext 0.5pt\">\n<p class=\"import-NormalWeb\">Review and synthesis of existing literature.<\/p>\n<\/td>\n<\/tr>\n<tr class=\"TableGrid-R\" style=\"height: 10.6pt\">\n<td class=\"TableGrid-C\" style=\"vertical-align: middle;border: solid windowtext 0.5pt\" rowspan=\"2\">\n<p class=\"import-NormalWeb\" style=\"text-align: center\">Statistical approach<\/p>\n<\/td>\n<td class=\"TableGrid-C\" style=\"vertical-align: middle;border: solid windowtext 0.5pt\">\n<p class=\"import-NormalWeb\" style=\"text-align: center\">Measures of occurrence<\/p>\n<\/td>\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\n<p class=\"import-NormalWeb\">Quantify the frequency, distribution, and\/or pattern of health-related states or events within a population. (e.g. prevalence, incidence, etc.).<\/p>\n<\/td>\n<\/tr>\n<tr class=\"TableGrid-R\" style=\"height: 0\">\n<td class=\"TableGrid-C\" style=\"vertical-align: middle;border: solid windowtext 0.5pt\">\n<p class=\"import-NormalWeb\" style=\"text-align: center\">Measures of association<\/p>\n<\/td>\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\n<p class=\"import-NormalWeb\">Quantify the relationship between exposure(s) and outcome (e.g., Relative Risk, Odds ratio, Hazard ratio, etc.).<\/p>\n<\/td>\n<\/tr>\n<tr class=\"TableGrid-R\" style=\"height: 0\">\n<td class=\"TableGrid-C\" style=\"background-color: #f2f2f2;vertical-align: middle;border: solid windowtext 0.5pt\" rowspan=\"3\">\n<p class=\"import-NormalWeb\" style=\"text-align: center\">Type of information<\/p>\n<\/td>\n<td class=\"TableGrid-C\" style=\"background-color: #f2f2f2;vertical-align: middle;border: solid windowtext 0.5pt\">\n<p class=\"import-NormalWeb\" style=\"text-align: center\">Descriptive<\/p>\n<\/td>\n<td class=\"TableGrid-C\" style=\"background-color: #f2f2f2;border: solid windowtext 0.5pt\">\n<p class=\"import-NormalWeb\">Provides a description of a case report or case series.<\/p>\n<p class=\"import-NormalWeb\">Provides an overview of the distribution and patterns of health-related events in terms of place, people, and time.<\/p>\n<\/td>\n<\/tr>\n<tr class=\"TableGrid-R\" style=\"height: 0\">\n<td class=\"TableGrid-C\" style=\"background-color: #f2f2f2;vertical-align: middle;border: solid windowtext 0.5pt\">\n<p class=\"import-NormalWeb\" style=\"text-align: center\">Analytical<\/p>\n<\/td>\n<td class=\"TableGrid-C\" style=\"background-color: #f2f2f2;border: solid windowtext 0.5pt\">\n<p class=\"import-NormalWeb\">Assesses associations between exposure(s) and outcome(s) to identify risk factors and\/or causes.<\/p>\n<\/td>\n<\/tr>\n<tr class=\"TableGrid-R\" style=\"height: 0\">\n<td class=\"TableGrid-C\" style=\"background-color: #f2f2f2;vertical-align: middle;border: solid windowtext 0.5pt\">\n<p class=\"import-NormalWeb\" style=\"text-align: center\">Analytical (experimental)<\/p>\n<\/td>\n<td class=\"TableGrid-C\" style=\"background-color: #f2f2f2;border: solid windowtext 0.5pt\">\n<p class=\"import-NormalWeb\">Provides evidence on the effects and\/or the mechanisms of interventions or treatments.<\/p>\n<\/td>\n<\/tr>\n<tr>\n<td><\/td>\n<td><\/td>\n<td><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/div>\n<p class=\"import-Normal\">After defining the research question, epidemiological study designs are primarily classified at the study design level based on the role of the researcher\/epidemiologist. In this sense, study designs can be classified as experimental or observational (non-experimental). Descriptive epidemiologic study designs are observational, and the type of information they generate is descriptive. Experimental study designs and analytic observational study designs are classified as having the highest strength for generating evidence; however, they may not always be appropriate for addressing specific public health research questions.<\/p>\n<div id=\"h5p-2\">\n<div class=\"h5p-iframe-wrapper\"><iframe id=\"h5p-iframe-2\" class=\"h5p-iframe\" data-content-id=\"2\" style=\"height:1px\" src=\"about:blank\" frameBorder=\"0\" scrolling=\"no\" title=\"When should epidemiologic study design be used?\"><\/iframe><\/div>\n<\/div>\n<h2>Descriptive Epidemiological Study designs<\/h2>\n<p class=\"import-NormalWeb\">Descriptive epidemiological study designs are crucial for public health for several reasons, including: (a) identifying the extent of a public health issue, (b) identifying at-risk groups, (c) monitoring trends over time, (d) allocating public health resources, (e) setting priorities, and (f) informing planning, policy, and evaluation. Since study design depend on the research questions and descriptive epidemiology focuses on the distribution and patterns of outcome(s) in terms of <strong>place<\/strong>, <strong>people<\/strong>, and <strong>time<\/strong>, the most common related questions, though not limited to these, are:<\/p>\n<p class=\"import-NormalWeb\">What is the case definition?<\/p>\n<p class=\"import-NormalWeb\">Is the defined case of public health relevance?<\/p>\n<p class=\"import-NormalWeb\">What is the disease distribution in the population?<\/p>\n<p class=\"import-NormalWeb\">What are the patterns of the outcome in different populations?<\/p>\n<p class=\"import-NormalWeb\">Are there seasonal or temporal patterns in disease occurrence?<\/p>\n<p class=\"import-NormalWeb\">How does the distribution and pattern vary over time?<\/p>\n<p class=\"import-NormalWeb\">Have the rates of disease changed over different time periods?<\/p>\n<p class=\"import-NormalWeb\">Are there emerging trends or new patterns in disease distribution?<\/p>\n<p class=\"import-Normal\">Descriptive epidemiological studies include: (1) case report, (2) case series, and (3) ecological studies. Although a cross-sectional study design can be both descriptive and analytical depending on its purpose, we will present it in the analytic section because most of the available cross-sectional studies include an analytical component. From a descriptive point of view, in a cross-sectional design, the prevalence of an outcome is assessed for a given population at that specific point in time. The following sub-section will define each common descriptive study design including the main characteristics, strengths, limitations, type of information generated, and applications in public health practice.<\/p>\n<h3>Case report and case series<\/h3>\n<p class=\"import-Normal\">A case report is a research design that provides a detailed description of the clinical aspects of a particular health-related state or event from a single person within their real-life context. A case series is an extension of case report where multiple individuals are classified as the same case. These type of study designs are appropriate for reporting unusual, unexpected, unfamiliar, or rare health-related outcomes. This includes, but is not limited to, observation of symptoms, clinical findings, natural history of disease, side effects, and complications of interventions<sup>4<\/sup>. Although a case report and case series are classified as generating low levels in the hierarchy of evidence<sup>5<\/sup>, they provide unique information for an initial understanding of a novel situation. An in-depth narrative of the potential relationship of exposures explaining the outcome occurrence is usually conducted, described, and explained in chronological order. Generally, a follow-up evaluation includes the retrospective collection of information regarding potential exposures that might have led to the occurrence. Therefore, both designs might generate a potential hypothesis and provide a valuable information for identifying additional cases. Table 2 summarizes the appropriate use of this approach, as well as the strengths and limitations of the case report and case series study designs.<\/p>\n<p class=\"import-Normal\">Table 2. Strengths, limitations, and when to use case report and case series designs.<\/p>\n<table>\n<tbody>\n<tr class=\"TableGrid-R\">\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\n<p class=\"import-Normal\">When should these epidemiologic study designs be used?<\/p>\n<\/td>\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\n<ul>\n<li>When there is a unique, rare, unusual, or unexpected health-related state or event from a single person or few cases.<\/li>\n<\/ul>\n<\/td>\n<\/tr>\n<tr class=\"TableGrid-R\">\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\n<p class=\"import-Normal\">Strenghts<sup>4<\/sup><\/p>\n<\/td>\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\n<ul>\n<li>Provides in-depth information on a unique case or case series.<\/li>\n<li>Educational value.<\/li>\n<li>New hypotheses could be formulated.<\/li>\n<li>Help identify additional cases.<\/li>\n<li>Flexible structure.<\/li>\n<\/ul>\n<\/td>\n<\/tr>\n<tr class=\"TableGrid-R\">\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\n<p class=\"import-Normal\">Limitations<sup>4<\/sup><\/p>\n<\/td>\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\n<ul>\n<li>Lack of generalizability.<\/li>\n<li>Lack of internal validity<\/li>\n<li>Causal inference is not possible.<\/li>\n<li>No control groups.<\/li>\n<li>Selection bias.<\/li>\n<\/ul>\n<\/td>\n<\/tr>\n<tr>\n<td><\/td>\n<td><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<h3>Ecological studies<\/h3>\n<p class=\"import-Normal\">An ecological study, often referred to as a correlation study<sup>2<\/sup>, is an observational research design that describes measures (e.g., rates, percentages, etc.) of exposures and outcome(s) at the population or group level rather than at the individual level. These groups can include geographical regions, communities, settings, or time periods. This is particularly important for public health planning because it allows the monitoring of population health and enables large-scale comparisons between and within groups over time. However, since data are collected and analyzed at group level, the generated information might not be true at individual level. The ecological fallacy is an error that occurs when conclusions about individual level are drawn from aggregate level. Table 3 summarizes the appropriate use of this study design, as well as the strengths and limitations.<\/p>\n<p class=\"import-Normal\" style=\"margin-left: 36pt;text-indent: 36pt\">Table 3. Strengths, limitations, and when to use ecological study design.<\/p>\n<table>\n<tbody>\n<tr class=\"TableGrid-R\">\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\n<p class=\"import-Normal\">When should this epidemiologic study design be used?<\/p>\n<\/td>\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\n<ul>\n<li>To identity and describe patterns and trends in health-related states or events across different populations, groups, or at different time points.<\/li>\n<\/ul>\n<\/td>\n<\/tr>\n<tr class=\"TableGrid-R\" style=\"height: 59.8pt\">\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\n<p class=\"import-Normal\">Strengths<\/p>\n<\/td>\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\n<ul>\n<li>Often require less time and resources compared to individual level studies.<\/li>\n<li>Useful for generating hypotheses.<\/li>\n<li>Useful for resource allocation and informing policy at the population level.<\/li>\n<\/ul>\n<\/td>\n<\/tr>\n<tr class=\"TableGrid-R\">\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\n<p class=\"import-Normal\">Limitations<\/p>\n<\/td>\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\n<ul>\n<li>Ecological fallacy.<\/li>\n<li>Information bias.<\/li>\n<\/ul>\n<\/td>\n<\/tr>\n<tr>\n<td><\/td>\n<td><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<h2>Analytical Epidemiological Study designs<\/h2>\n<p class=\"import-NormalWeb\">Analytical epidemiological study designs focus on understanding the associations between exposure(s) and outcome(s). Thus, the common strategy of analytical designs for explaining why and\/or how a health outcome occurs involves testing hypotheses. This process often seeks to distinguish between causal and non-causal relationships as well as quantify the association between variables. Like descriptive study designs, each analytical study design serves a specific purpose and is chosen based on the research question and to provide specific answers for public health. This classification is important because it determines the strength of the evidence regarding the potential causation of health-related states and events. The most common related questions, though not limited to these, are:<\/p>\n<p class=\"import-NormalWeb\">What is the cause of a specific health outcome?<\/p>\n<p class=\"import-NormalWeb\">What is the mechanism underlying the causation of a specific health outcome?<\/p>\n<p class=\"import-NormalWeb\">What determinants and factors increase or decrease the risk of a particular health outcome?<\/p>\n<p class=\"import-NormalWeb\">Is there a dose-response relationship between the exposure and the health outcome?<\/p>\n<p class=\"import-NormalWeb\">How do relationships between exposures and outcomes differ among subgroups of the population?<\/p>\n<p class=\"import-NormalWeb\">How do health outcomes differ between groups with different exposure levels or characteristics?<\/p>\n<p class=\"import-NormalWeb\">What is the effectiveness of a specific intervention in reducing the incidence of a health condition?<\/p>\n<p class=\"import-NormalWeb\">How does the efficacy of an intervention translate to real-world settings?<\/p>\n<p class=\"import-Normal\">Analytical epidemiological study designs include: (1) Cross-sectional studies, (2) Case-control studies, (3) Cohort studies, and (4) Experimental studies. These studies are primarily classified based on the researcher role. In this sense, they can be classified as observational (non-experimental) or experimental, depending on whether the researcher manipulates the study factors (e.g., intervention, randomization) to investigate the relationship between exposure(s) and outcome(s). This classification is important because it determines the strength of the evidence regarding the potential causal relationship between the exposures and health-related states and events. The following sections will define each common analytical study designs including the main characteristics, strengths, limitations, types of information generated, and application in public health practice.<\/p>\n<h3>Cross-sectional studies<\/h3>\n<p class=\"import-Normal\">A cross-sectional study is an observational research design in which data are collected at the individual level. The main characteristic is that exposures and outcomes are measured simultaneously at a single point in time. In other words, cross-sectional studies are equivalent to a \u201cphoto\u201d of the health status of a given population. The type of information generated from an analytical perspective is that the association between exposure(s) and outcome(s) can be estimated; however, with a lower level of evidence compared to other analytic designs, and that is because exposure and outcomes are measured at the same time. For example, a cross-sectional study was conducted in May 2020 on the proportion of the population with Covid-19 antibodies according to socioeconomic status in Brazil<sup>11<\/sup>. Individuals were tested for Covid-19 and information on the family socioeconomic status was collected by questionnaire. The proportion of individuals with Covid-19 antibodies was significantly higher among the poorest quintile (2.1%) compared to the richest quintile (1.0%). In addition to assessing the burden of the health outcome, it helps public health with resource allocation. Table 4 summarizes the appropriate use of this design, as well as the strengths and limitations.<\/p>\n<p class=\"import-Normal\" style=\"margin-left: 36pt;text-indent: 36pt\">Table 4. Strengths, limitations, and when to use cross-sectional design.<\/p>\n<table>\n<tbody>\n<tr class=\"TableGrid-R\">\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\n<p class=\"import-Normal\">When should this epidemiological study design be used?<\/p>\n<\/td>\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\n<ul>\n<li>When researchers are interested to examine the prevalence and associate factors of a health-related outcome in a given population at a single point in time.<\/li>\n<\/ul>\n<\/td>\n<\/tr>\n<tr class=\"TableGrid-R\">\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\n<p class=\"import-Normal\">Strengths<\/p>\n<\/td>\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\n<ul>\n<li>Provides a \u201cphoto\u201d of the health-related outcome in a population.<\/li>\n<li>Less time-consuming to collect data.<\/li>\n<li>Multiple outcomes and exposures can be studied at a time.<\/li>\n<li>Useful for public health planning.<\/li>\n<\/ul>\n<\/td>\n<\/tr>\n<tr class=\"TableGrid-R\">\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\n<p class=\"import-Normal\">Limitations<\/p>\n<\/td>\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\n<ul>\n<li>Temporal limitation.<\/li>\n<li>Causal inference not possible.<\/li>\n<li>Selection bias.<\/li>\n<\/ul>\n<\/td>\n<\/tr>\n<tr>\n<td><\/td>\n<td><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<h3>Case-control studies<\/h3>\n<p class=\"import-Normal\">A Case-control study is an observational research design aimed at investigating the potential causes of a health outcome by comparing two groups. The study groups are defined based on the presence or absence of the outcome variable under investigation. After identifying case and control groups, data on exposure(s) are collected retrospectively among cases (those with the outcome of interest) and controls. For example, 122 breast cancer patients and 121 controls were enrolled in a case-control study on the influence of lifetime physical activity on breast cancer risk<sup>12<\/sup>. Both cases and controls were asked about their lifetime physical activity involvement. Most physical activity reported by participants took place in the household domain. The study found no association between lifetime physical activity and breast cancer risk. On one hand, because the exposures are assessed retrospectively after identifying the cases and controls, this design is more prone to bias compared to designs that assess exposures before the occurrence of the outcome. On the other hand, when the outcome is rare, this design may be more appropriate since the cases have already been identified. Therefore, it is useful for public health, especially in early outbreak investigations. Table 5 summarizes the appropriate use of this design, as well as the strengths and limitations.<\/p>\n<p class=\"import-Normal\" style=\"margin-left: 36pt;text-indent: 36pt\">Table 5. Strengths, limitations, and when to use Case-control design.<\/p>\n<table>\n<tbody>\n<tr class=\"TableGrid-R\">\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\n<p class=\"import-Normal\">When should this epidemiologic study design be used?<\/p>\n<\/td>\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\n<ul>\n<li>To compare individuals with the outcome (case) to those without (controls) to identify differences in exposures.<\/li>\n<\/ul>\n<\/td>\n<\/tr>\n<tr class=\"TableGrid-R\">\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\n<p class=\"import-Normal\">Strengths<\/p>\n<\/td>\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\n<ul>\n<li>Cost-effective.<\/li>\n<li>Useful for rare outcomes.<\/li>\n<li>Useful for prolonged exposure.<\/li>\n<\/ul>\n<\/td>\n<\/tr>\n<tr class=\"TableGrid-R\">\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\n<p class=\"import-Normal\">Limitations<\/p>\n<\/td>\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\n<ul>\n<li>Recall bias.<\/li>\n<li>Selection bias.<\/li>\n<li>Not good for rare exposure.<\/li>\n<\/ul>\n<\/td>\n<\/tr>\n<tr>\n<td><\/td>\n<td><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<h3>Cohort studies<\/h3>\n<p class=\"import-Normal\">Cohort studies are equivalent to a \u201cvideo\u201d of the health status of a given population. In cohort studies, data on exposure(s) is collected at baseline and participants are followed-up over time to monitor the presence or absence of outcomes. The longitudinal characteristics of this design and the assessment of exposures before the outcomes provide a better understanding of how several exposures influence the occurrence of health outcomes. This design allows researchers to establish temporal relationships between exposures and outcomes in a natural context, as there is no manipulation (i.e., intervention, random allocation). From a public health perspective, it is crucial for planning since evidence on potential causation and the risk of developing certain conditions over time is available. Table 6 summarizes the appropriate use of this design, as well as the strengths and limitations.<\/p>\n<p class=\"import-Normal\" style=\"margin-left: 36pt;text-indent: 36pt\">Table 6. Strengths, limitations, and when to use Cohort design.<\/p>\n<table>\n<tbody>\n<tr class=\"TableGrid-R\">\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\n<p class=\"import-Normal\">When should this epidemiologic study design be used?<\/p>\n<\/td>\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\n<ul>\n<li>When researchers are interested in establishing a temporal relationship between exposures and outcomes in a natural context without research manipulation.<\/li>\n<\/ul>\n<\/td>\n<\/tr>\n<tr class=\"TableGrid-R\">\n<td class=\"TableGrid-C\" style=\"background-color: transparent;border: solid windowtext 0.5pt\">\n<p class=\"import-Normal\">Strengths<\/p>\n<\/td>\n<td class=\"TableGrid-C\" style=\"background-color: transparent;border: solid windowtext 0.5pt\">\n<ul>\n<li>Temporal relationship<\/li>\n<li>Assessment of multiple exposures and outcomes<\/li>\n<li>Ethical advantage as compared to experimental studies<\/li>\n<\/ul>\n<\/td>\n<\/tr>\n<tr class=\"TableGrid-R\">\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\n<p class=\"import-Normal\">Limitations<\/p>\n<\/td>\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\n<ul>\n<li>Time-consuming<\/li>\n<li>Loss to follow-up<\/li>\n<li>Resource intensive<\/li>\n<\/ul>\n<\/td>\n<\/tr>\n<tr>\n<td><\/td>\n<td><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<h3>Intervention studies<\/h3>\n<p class=\"import-Normal\">Intervention studies, also referred to as experimental studies, are designed to evaluate the effect of specific interventions on health outcomes. These studies can be classified into different types and sub-types depending on the research question (e.g., testing efficacy or effectiveness), the nature of the intervention, and the context (e.g., clinical trials, field trials) in which the intervention is applied. Unlike observational studies, the association between exposures and outcomes in intervention studies is evaluated under controlled conditions. By doing this, researchers can control the type, timing, and degree of the exposure and quantify the observed effects on health outcomes, ensuring that these effects are due to the intervention rather than other variables. One important aspect of intervention studies that distinguishes the quality of evidence in terms of establishing potential causal relationships is whether the study design includes randomization (a type of study factor manipulation). For example, in randomized controlled trials (RCTs), participants are randomly assigned to either the intervention group or the control group. This process helps minimize selection bias and balance potential confounders between groups. A quasi-experimental design is a type of intervention where researchers do not randomize participants into groups. Intervention studies play a vital role in public health for several reasons, but not limited to: (a) evaluating program effectiveness, (b) testing new treatments, and (c) Guiding policy decisions. Table 7 summarizes the appropriate use of this design, as well as the strengths and limitations.<\/p>\n<p class=\"import-Normal\" style=\"margin-left: 36pt;text-indent: 36pt\">Table 7. Strengths, limitations, and when to use Intervention design.<\/p>\n<table>\n<tbody>\n<tr class=\"TableGrid-R\">\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\n<p class=\"import-Normal\">When should this epidemiologic study design be used?<\/p>\n<\/td>\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\n<ul>\n<li>When researchers are interested to test efficacy and\/ or evaluate the effectiveness of an intervention on a specific outcome.<\/li>\n<\/ul>\n<\/td>\n<\/tr>\n<tr class=\"TableGrid-R\">\n<td class=\"TableGrid-C\" style=\"background-color: transparent;border: solid windowtext 0.5pt\">\n<p class=\"import-Normal\">Strengths<\/p>\n<\/td>\n<td class=\"TableGrid-C\" style=\"background-color: transparent;border: solid windowtext 0.5pt\">\n<ul>\n<li>Temporal relationship<\/li>\n<li>Potential causal relationship<\/li>\n<li>Control over exposure and\/or groups<\/li>\n<\/ul>\n<\/td>\n<\/tr>\n<tr class=\"TableGrid-R\">\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\n<p class=\"import-Normal\">Limitations<\/p>\n<\/td>\n<td class=\"TableGrid-C\" style=\"border: solid windowtext 0.5pt\">\n<ul>\n<li>Cost and time<\/li>\n<li>Resource intensive<\/li>\n<li>Ethical concerns<\/li>\n<\/ul>\n<\/td>\n<\/tr>\n<tr>\n<td><\/td>\n<td><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<div class=\"textbox textbox--key-takeaways\">\n<header class=\"textbox__header\">\n<p class=\"textbox__title\">Key Takeaways<\/p>\n<\/header>\n<div class=\"textbox__content\">\n<p class=\"import-Normal\">In this chapter, we covered common epidemiology studies designs, highlighting the main characteristics, strengths, limitations, and applications for public health practices. The decision about what study design to use relies on the research question and hypotheses the researcher would like to test. Descriptive designs are vital for identifying public health issues, occurrence, patterns, and trends in terms of people, place and time, while analytic designs provide insights into causal and non-causal relationships between exposures and outcomes in a population. Understanding these study designs enhances the ability to address complex public health issues and informs decision-making processes in public health practice.<\/p>\n<\/div>\n<\/div>\n<p>&nbsp;<\/p>\n<p class=\"import-Normal\">References<\/p>\n<ul>\n<li>Holland, W.W., Olsen, J., Florey, C.D.V., et al. (eds.) (2007). The Development of Modern Epidemiology: Personal Reports from Those Who Were There. Oxford: Oxford University Press.<\/li>\n<li><span lang=\"pt-BR\" xml:lang=\"pt-BR\">Bonita R, Beaglehole R, Kjellstr\u00f6m T. Basic Epidemiology. <\/span>World Health Organization; 2006. <a class=\"rId8\" href=\"https:\/\/apps.who.int\/iris\/bitstream\/handle\/10665\/43541\/9241547073_eng.pdf\"><span class=\"import-Hyperlink\">https:\/\/apps.who.int\/iris\/bitstream\/handle\/10665\/43541\/9241547073_eng.pdf<\/span><\/a>. Accessed July 18, 2028.<\/li>\n<li>Sapkota, K. (2023). Descriptive and Analytical Epidemiology. In Statistical Approaches for Epidemiology: From Concept to Application (pp. 1-18). Cham: Springer International Publishing.<\/li>\n<li>Nissen T, Wynn R. The clinical case report: a review of its merits and limitations. BMC Res Notes. 2014 Apr 23;7:264. doi: 10.1186\/1756-0500-7-264.<\/li>\n<li>Guyatt, G. H., Oxman, A. D., Vist, G. E., Kunz, R., Falck-Ytter, Y., Alonso-Coello, P., &amp; Sch\u00fcnemann, H. J. (2008). GRADE: an emerging consensus on rating quality of evidence and strength of recommendations. Bmj, 336(7650), 924-926.<\/li>\n<li>Holland, W.W., Olsen, J., Florey, C.D.V., et al. (eds.) (2007). The Development of Modern Epidemiology: Personal Reports from Those Who Were There. Oxford: Oxford University Press.<\/li>\n<li><span lang=\"pt-BR\" xml:lang=\"pt-BR\">Bonita R, Beaglehole R, Kjellstr\u00f6m T. Basic Epidemiology. <\/span>World Health Organization; 2006. <a class=\"rId9\" href=\"https:\/\/apps.who.int\/iris\/bitstream\/handle\/10665\/43541\/9241547073_eng.pdf\"><span class=\"import-Hyperlink\">https:\/\/apps.who.int\/iris\/bitstream\/handle\/10665\/43541\/9241547073_eng.pdf<\/span><\/a>. Accessed July 18, 2028.<\/li>\n<li>Sapkota, K. (2023). Descriptive and Analytical Epidemiology. In Statistical Approaches for Epidemiology: From Concept to Application (pp. 1-18). Cham: Springer International Publishing.<\/li>\n<li>Nissen T, Wynn R. The clinical case report: a review of its merits and limitations. BMC Res Notes. 2014 Apr 23;7:264. doi: 10.1186\/1756-0500-7-264.<\/li>\n<li>Guyatt, G. H., Oxman, A. D., Vist, G. E., Kunz, R., Falck-Ytter, Y., Alonso-Coello, P., &amp; Sch\u00fcnemann, H. J. (2008). GRADE: an emerging consensus on rating quality of evidence and strength of recommendations. Bmj, 336(7650), 924-926.<\/li>\n<li>Hallal PC, Hartwig FP, Horta BL, et al. SARS-CoV-2 antibody prevalence in Brazil: results from two successive nationwide serological household surveys. Lancet Glob Health. 2020;8(11):e1390-e1398. doi:10.1016\/S2214-109X(20)30387-9<\/li>\n<li>Yen SH, Knight A, Krishna M, Muda W, Rufai A. Lifetime Physical Activity and Breast Cancer: a Case-Control Study in Kelantan, Malaysia. Asian Pac J Cancer Prev. 2016;17(8):4083-4088.<\/li>\n<\/ul>\n<p class=\"import-Normal\">\n<\/div>\n","protected":false},"author":345,"menu_order":5,"template":"","meta":{"pb_show_title":"on","pb_short_title":"","pb_subtitle":"","pb_authors":["pedro-rodrigues-curi-hallal","rafael-miranda-tassitano"],"pb_section_license":""},"chapter-type":[],"contributor":[64,65],"license":[],"class_list":["post-106","chapter","type-chapter","status-publish","hentry","contributor-pedro-rodrigues-curi-hallal","contributor-rafael-miranda-tassitano"],"part":3,"_links":{"self":[{"href":"https:\/\/pb-sandbox.library.illinois.edu\/introductiontoepidemiology\/wp-json\/pressbooks\/v2\/chapters\/106","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/pb-sandbox.library.illinois.edu\/introductiontoepidemiology\/wp-json\/pressbooks\/v2\/chapters"}],"about":[{"href":"https:\/\/pb-sandbox.library.illinois.edu\/introductiontoepidemiology\/wp-json\/wp\/v2\/types\/chapter"}],"author":[{"embeddable":true,"href":"https:\/\/pb-sandbox.library.illinois.edu\/introductiontoepidemiology\/wp-json\/wp\/v2\/users\/345"}],"version-history":[{"count":9,"href":"https:\/\/pb-sandbox.library.illinois.edu\/introductiontoepidemiology\/wp-json\/pressbooks\/v2\/chapters\/106\/revisions"}],"predecessor-version":[{"id":134,"href":"https:\/\/pb-sandbox.library.illinois.edu\/introductiontoepidemiology\/wp-json\/pressbooks\/v2\/chapters\/106\/revisions\/134"}],"part":[{"href":"https:\/\/pb-sandbox.library.illinois.edu\/introductiontoepidemiology\/wp-json\/pressbooks\/v2\/parts\/3"}],"metadata":[{"href":"https:\/\/pb-sandbox.library.illinois.edu\/introductiontoepidemiology\/wp-json\/pressbooks\/v2\/chapters\/106\/metadata\/"}],"wp:attachment":[{"href":"https:\/\/pb-sandbox.library.illinois.edu\/introductiontoepidemiology\/wp-json\/wp\/v2\/media?parent=106"}],"wp:term":[{"taxonomy":"chapter-type","embeddable":true,"href":"https:\/\/pb-sandbox.library.illinois.edu\/introductiontoepidemiology\/wp-json\/pressbooks\/v2\/chapter-type?post=106"},{"taxonomy":"contributor","embeddable":true,"href":"https:\/\/pb-sandbox.library.illinois.edu\/introductiontoepidemiology\/wp-json\/wp\/v2\/contributor?post=106"},{"taxonomy":"license","embeddable":true,"href":"https:\/\/pb-sandbox.library.illinois.edu\/introductiontoepidemiology\/wp-json\/wp\/v2\/license?post=106"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}