SI Research Methods - Primary Literature

Categorization of Research Methods

Categorization of Research Methods

1. Evaluative Research

2. Descriptive Research

Evaluative Research

Primary goal is to answer a question

Two types: Experimental, Observational

Evaluative Research: Experimental

Measures interventions

Types:

• clinical trials

• meta-analysis

• pharmacoeconomics

• health related quality of life

Evaluative Research: Experimental --> clinical trials

Purpose: evaluate cause and effect

• "true experiments" (randomly assigned) OR quasi (not randomly assigned)

• prospective

• involve some type of intervention (pharmacotherapeutric, pharmacokinetic, or pharmacodynamic) which are measured by the ECHO outcomes

• statistical analysis: alpha value and beta value

Alpha Value

probability of incorrectly rejecting the null hypothesis

p< 0.05 means data is significant (NOT DUE TO CHANCE)

This represents the probability of obtaining your results due to chance.

beta value

β (Beta) is the probability of Type II error in any hypothesis test-incorrectly failing to reject the null hypothesis.

Types of Data

nominal, ordinal, interval, ratio

Nominal Data

data of categories only.

Data cannot be arranged in an ordering scheme. (Gender, Race, Religion)

Ordinal Data

data exists in categories that are ordered but differences cannot be determined or they are meaningless. (Example: 1st, 2nd, 3rd)

Interval Data

Differences between values can be found, but there is no absolute 0. (Temp. and Time)

Ratio Data

a type of numerical data in which the difference between numbers is significant, but there is a fixed non-arbitrary zero point associated with the data

Pros/ Cons of Clinical Trails

Advantages: purest way to measure cause and effect

Disadvantages: time intensive, expensive, unethical for certain instances (testing med effect during pregnancy)

Evaluative Research: Experimental --> meta-analysis

• Studies of studies

• hypothesis and combine states from other studies

• findings are substantiated by the detemination of statistical relationships

• two reviewers search for relevant articles based on comparision of the METHOD SECTION ONLY. Inclusion and exclusion parameters set beforehand.

• pooling occurs if data is similar enough

Meta-analysis:Pooling

• combining results from included studies

• statistical data classification -

• nominal: compares effects NOT participants

• results are presented as Odds Ratios with 95% confidence intervals

• sensitivity analysis

• logistic regression

Pros/Cons of Meta-analysis

Advantage: Method to further evaluate previously identified associations, enhance power (significance), provide direction for future research intiatives

Con: Retrospective, does not confirm cause and effect (but encourages more research), only uses published research (so suspectible to publication bias)

Evaluative Research: Observational

examine specific data sources for the purpose of identifying relationships surrounding specific events or outcomes

Types

• cohort

• case-control

• Cross-section/ survey

Evaluative Research: Observational --> Cohort

Purpose: to determine relationship between risk factor exposure and outcomes -establish causation between risk and outcome

• prospective (start with a risk and determine its effect)

• LONG: need enough time and participants to see the outcome

• must have comparable groups for controls (not exposed to risk) and cases (those exposed to risk)

• Statistics: Relative risk (RR)

Pros/Cons of Cohort Studies

Advantages: the investigators control sample characteristics and evaluation parameters Disadvantages: long duration, drop out rates, biases (is outcome due to risk soley), surveillance

Evaluative Research: Observational --> Case-Control Analyses

Purpose: to determine which prior risk factor may be contributing to an observed outcome

• retrospective.

• The collector of data should be blinded to the study hypothesis.

• examine prevalence

• cases and controls are comparable (demographics, access to health care, and their RECALL ability)

• Statistics: Odds Ratio and regression analysis

Pros/Cons of Case-Control

Advantages: useful for evaluating rare diseases, diseases that progress slowly, ethically sensitive issues. Because the outcome is already known: its cheaper, quicker, and able to identify multiple risk factor association

Disadvantages: rely on the accuracy of accuracy and completeness of selected databases (medical records), identifying relationships may be challenging, bias (recall, surveillance, protopathic)

Evaluative Research: Observational --> Cross sectional/ Analysis

Purpose: to evaluate opinions, environmental circumstances, disease states, or interventions at a single, specified period in order to DETERMINE PREVALENCE

use of surveys

pros and cons of cross-sectional studies

Advantages: quicker to perform, less expensive

Disadvantages: transient effect (observation or outcome that occured by chance may not be appreciated if analysis is repeated), unable to determine cause and effect

Article sections: Evaluative Research

• Title

• authors

• abstract

• introduction/ background

• methods: Sample (demographics, inclusion/exclusion, size), Blinding (single, double, open-label), Control groups (placebo vs active), allocation of intervention, Number of sites (single vs multiple) -

• results

• discussion

• conclusion

• references

Descriptive Research

The primary goal is to report individual experiences

Lack a hypothesis

Findings often lead to the development of evaluative studies

Types:

• case studies and case reports -overlap with evaluative observation (cohort,case-control, cross-section/ survey)

Case report or Case series Study

retrospective study, single subject with novel clinical features and/or outcomes described, potentially generates new research questions; more than one subject case comprises a case series study -possible topics: adverse event, teratogenicity, new treatment options -No formal study design, no formal presentation format, limited statistical analysis, limited ability to make population inferences

Article Sections: Descriptive Research

Title Author Abstract Introduction Observations: Combination of methods and results because there is no clear hypothesis in Descriptive research Discussion Conclusion References

Outcomes (ECHO Model)

•Economic: Costs as a result of the treatment or intervention

•Clinical: Health outcomes that occur as a result of the treatment or intervention

•Humanistic: health-related quality of life consequences of the disease or its treatment as reported by the patient or caregiver

•Operational

Costs

-Direct -Indirect -Intangible

Direct costs

medical: monies spent on healthcare Nonmedical: movies spent while obtaining healthcare -time off work -transportation

Indirect Cost

Monies lost secondary to morbidity and mortality

intangible costs

pain, suffering, distress

Clinical outcomes

describe the expected status of health issues at certain points in time, after treatment is complete. They address whether the problems are resolved or to what degree they are improved. -laboratory and physical findings

humanistic outcomes

consequences of disease or treatment on patient functional status, or quality of life, measured along several dimensions, e.g., physical functioning, social functioning, general health perceptions and well-being. -uses scales/surveys to assess quality of life and satisfaction

Operational Outcomes

Assessment of how productivity, quality, service, innovation, and overall operational performance -example: how did a new safety protocol increase safety in ER