Ecological Studies & Secular Trends

Ecological Studies, Correlation Studies, and Secular Trends A. Ecological Studies The purpose of an ecological study is to correlate common characteristics of a population with a health problem over the same time period in multiple populations; or in the same population over different time periods. Ecological studies are groupbased units of analysis. B. Correlation Studies Correlation studies are another name for ecological studies An epidemiologic study that is an observational study. The unit of observation or analysis is the aggregate Here are some examples of aggregate sizes: 1. Aggregate measures that measure the mean, median or proportion of a set of individual values in a group e.g.: a. Rate of a disease; incidence, prevalence b. Average fat intake in a group of individuals or a community c. Value of program coverage d. The median income of a group of individuals 2. An aggregate measure of "environment" is a measure that represents the physical characteristics of an environment such as: a. Coverage value of healthy homes in an area b. Pollution intensity value in an area c. The value of larval density in an area The study population consists of the set of observation units from which the aggregate measure is measured. For example: the unit of observation for larva density, and DHF incidence is measured based on the working area of the health center, then the study population consists of the collection of health centers. C. Secular Trends Secular trends are changes in disease frequency over a long period of time. Factors to be considered in assessing secular trends: 1. Artifact (pseudo) changes a. Numerator  Changes in diagnosis techniques that lead to increased disease reporting  Changes in disease classification b. Denominator  Errors in enumerating the population (Increased accuracy in counting the population at risk) 2. True (actual) change  Changes in the age structure of the population  Changes in survivorship  Changes in disease incidence (due to environmental factors, occupation, etc.) D. Cross Sectional Survey Study 1. It is a study to find out the description and compare 2 variables, but does not test the cause-and-effect relationship. 2. Example: description of high obesity rates associated with cases of diabetes 3. Unit of observation: individual 4. Viewed simultaneously 5. Limitations:  Cannot determine causal relationship  Unknown variables precede or are preceded by other variables E. Quasi-Experiment A quasi-experiment is an experimental study in which non-randomization is used to control the research situation. This experiment is conducted as an alternative to randomization experiments, when the allocation of research factors to research subjects is impossible, unethical, or impractical to do with randomization, for example when the sample size is too small. This experiment is used to evaluate the impact of public health interventions to obtain an estimate of the true impact of the treatment, so researchers must choose a control group that has characteristics of confounding variables that are comparable (comparable) with the treatment group. Strengths and weaknesses of quasi-experiments: 1. Strengths a. Quasi-experiments are more feasible and less expensive than RCTs, when randomization is not feasible. 2. Weaknesses a. Since the allocation of treatments is not randomized, the researcher will be less able to control for confounding factors. b. Non-random allocation may result in biases that are difficult to control in data analysis