Guidelines for Defining Research

From:	Guidelines for Defining Public Health Research & Public Health Non-Research, rev. 1-4-99. Published by the CDC. Adapted liberally.
Note:	this information was taken directly from CDC guidelines and does not incorporate state regulations or SKCDPH policies/philosophy.

This appendix is broken down into four categories: Attributes of Research; Surveillance; Emergency Responses and Evaluation.

Attributes of Research

Considerations	Research	Public Health Activities/Non-Research
Primary intent of activity	To generate or contribute to generalizable knowledge(1) to improve public health practice by a systematic investigation, including research development.	To prevent or control disease or injury and improve health or to improve a public health program or service. In some cases this knowledge may be generalizable.
Who benefits from study results?	May or may not include study participants, but always extend beyond study participants, usually to society	Clients participating in SKCDPH programs or for controlling a health problem in the population from which the information is generated.
Data Collection Purpose	Exceeds requirements for care of the study participants or extends beyond the scope of the activity	Data needed to access and/or improve the program or service, the health of the participants or the participants' community
How collected	Under systematic procedures that reduce bias, allowing it to be applied to populations and settings different from the ones from which it was collected	Knowledge generated does not extend beyond the scope of the activity
Role of experimental studies	Can be included	Would never include
Does subsequent analysis of PHI undertaken generate or contribute to generalizable knowledge?	Classify as research	It may start out in this category but becomes research that requires IRB review when it "crosses over" into generalizable knowledge.
Does the project include multiple components and at least one is designed to generate generalizable knowledge?	Classify as research	The entire project becomes research when this occurs
Items that have no bearing on how a study is categorized: Publication of Findings Methodological Design Subject Selection Hypothesis testing/ generating

Surveillance

Considerations	Research	Public Health Activities/Non-Research
A method for public health data collection	Involves collection & analysis of health related data conducted either to generate knowledge that is applicable to other populations and settings that the ones from which the data were collected or to contribute to new knowledge about the health condition.	Involves regular, ongoing collection and analysis or health related data conducted to monitor frequency & distribution in the population: Ability to invoke public health mechanisms to prevent or control disease or injury in response to an event
Intent	To generalize generalizable knowledge	To prevent or control disease or injury in a defined population by producing information about the population from whom the data were collected.
Data Use	Beyond SKCDPH, literature	Management of public health programs
Attributes Generally Found	In research design proposal	In state statute or regulation where the intent of the activity, its purposes, and uses of the data are specified. Need to give SKCDPH specific examples
Subject selection	Design	Via passive reporting systems
Hypothesis testing	Yes. Includes longitudinal data collection systems that allows for hypothesis testing	No
Scope	Broad. Etiologic analysis is likely to fall in this category.	Generally, lawful state disease reporting, monitoring requirements and other data collection activities conducted under state statute. When data collection extends beyond this, it is not automatically considered non-research.

Emergency Responses

Purpose

Research

Public Health Activities/Non-Research

To identify, characterize and solve an immediate health problem and the knowledge gained will directly benefit those involved in the investigation or their communities

May become research when the study or a component of it has:

samples which are stored for future use intended to create generalizable knowledge
additional analyses are conducted beyond those needed to solve the immediate health problem

This is almost always research when:

investigational new drugs are used or drugs are used off-label
medical devices are investigational or (?? Off-label apply here)

This is research when:

there is a systematic investigation of a non-standard intervention
there is a systematic comparison of standard interventions

These activities tend to fall into this category.

Evaluation or Program Evaluation

Definitions	Research	Public Health Activities/Non-Research
Evaluations (broad in meaning) refer to the systematic use of scientific methods to measure efficacy, implementation, utility and so on of a program in its entirety or its components.	Can be research when conditions are met.
Program Evaluations -- an essential organizational practice in public health using a systematic approach to improve and account for public health actions. These are a subset of evaluations.		Is not considered "research"
Formative Evaluation -- data collection activities that occur prior to the implementation of an intervention, service or program	If conducted prior to implementing a new, modified or previously untested intervention	If conducted to provide information on how to tailor a proven-effective intervention, service or program in a specific setting or context
Purpose	To test new, modified or previously untested intervention, service, or program to determine whether it is effective Systematic comparison of standard or non-standard interventions in an experimental-type design Note: in both cases the intent is to generate new knowledge or contribute to knowledge in the scientific literature and to apply this knowledge to other sites or populations.	When used as a management tool to monitor and improve the program: Is often a component of an on-going, regular program Information is not generalizable beyond the individual program Provides immediate benefit for clients Evaluations are not on anything experimental or new (they are known from empirical data or through consensus to be effective)

Further notes

There are some projects that do not fit easily into the categories described above, so the primary intent and elements of the project must be taken into consideration. See the table entitled "Attributes of Research" for guidance. An example would be when a public health department, under its public health authority provides an untested intervention in an outbreak situation. An evaluation component is then added. In this example, because the intervention and evaluation activities are undertaken with different intentions and are separable, the intervention activities are not research but the evaluation activities are research.

(1) Generalizable knowledge means new information that has relevance beyond the population or program from which it was collected, or information that is added to the scientific literature. This does not refer to the statistical concept of population estimation or to the traditional public health method of collecting information from a sample to understand health in the population from which the sample came.