AcademyHealth: Privacy and Health Services Research

Overview: This document identifies the basic considerations relevant to obtain human subject approvals for a research project. Many of these elements derive from the US Federal Common Rule (45 CFR 46). Principle Investigators (PIs) must identify how Common Rule provisions are implemented and interpreted by local institutions and research partners. Researchers must also identify any additional federal, state, local, or institutional human subjects protection regulations or policies that may apply. Researchers should also consult with funders to determine their requirements for human subjects protection.

Institutional Review Board (IRB)

Overview: An IRB conforming to Federal regulations, State laws, and institutional policy, reviews research proposals involving human subjects to promote the ethical and equitable treatment of those subjects. Representatives of your research institution can direct you to the appropriate IRB. IRB approval is generally required for all human subjects research..

Informed Consent

Overview: PIs must develop procedures, protocols, and justifications for obtaining informed consent from research subjects. Generally, this includes written consent:

outlining the procedures for obtaining informed consent from research subjects
developing a protocol for documenting informed consent from research subjects
justifying the procedures and protocols in the context of the overall research project.

Your IRB and the IRBs of the institutions you are working with (including funding entities) will determine the specific procedures and protocols that are appropriate for your particular research project.

Written, Oral, and Waiver

Health policy and services researchers may seek written, oral, or waiver of informed consent and informed consent documentation. The propriety of written versus oral consent often depends on the research protocol, as detailed in table 1 below.

Table 1: Methods of obtaining and documenting informed consent

Purpose	Secondary Data Analysis or Analysis of Administrative Data	Telephone Interview or Survey	In-Person Interview or Survey	Direct Observation or Ethnography
Written	Difficult to obtain unless done at the time of original interaction	Usually not feasible	Usually required	Usually not required
Oral	Usually not considered adequate. Often not feasible	Often considered adequate. Obtain before starting interview	Possible in some very low risk projects.	Usual practice
Waiver	May be considered adequate. Consult PI Privacy Checklist for further information.	Usually not considered adequate.	Usually not considered adequate.	May be considered adequate, especially for observations in public setting

Confidentiality

Overview: Health policy and services research may put subjects at risk of a loss of confidentiality of information they or HIPAA would deem private. Your IRB and the IRBs of the institutions with which you are working (including funding entities) will determine the specific procedures and protocols that are considered adequate for protecting subject confidentiality. There are some general principles that health policy and services researchers should consider to protect subject confidentiality as well as some specific steps that can be taken depending on the research design, as detailed in table 2 below.

Table 2: Preserving research subject confidentiality

Secondary Data Analysis or

Analysis of Administrative Data

Telephone Interview or Survey

In-Person Interview or Survey

Direct Observation or Ethnography

Consult PI Privacy Checklist

1. separate individual identifiers from data in paper and electronic files

2. withhold identifying details of individuals and study sites in public reports and documents

3. destroy documents with individually-identified information as soon as possible

1. separate individual identifiers from data in paper and electronic files

2. withhold identifying details of individuals and study sites in public reports and documents

3. destroy documents with individually-identified information as soon as possible

1. use pseudonyms in field notes

2. don't reveal location of study sites

3. alter and/or withhold identifying details of individuals and study sites in public reports and documents

4. destroy primary documents if possible

1.maintain electronic records in password protected documents and computers

2. keep electronic files on secure computer networks

3. keep paper files in locked cabinets and offices

4. restrict file access to study team

Risks

Overview: Research subjects' perception of risk may vary, and subjects may experience health policy or services research as intrusive or invasive. In the context of the Common Rule, some IRBs may consider many health policy or services research projects to be “minimal risk”, i.e. not exceeding those encountered in everyday life (See CFR 46.102(i)). However, PIs can not assume that their research is minimal risk and must consult with their IRB.

The risks of most health policy or services research:

includes potential loss of confidentiality;
may include potential for embarrassment due to answering personal questions.

Steps to address and minimize these risks may include:

instituting appropriate measures to guard confidentiality and privacy (see table above);
informing subject that participating will not affect their medical care;
allowing respondent to terminate participation at any time.

Benefits

Usually there are no direct benefits to subjects from participating in a health policy or services research project. Usually there are some benefits to society at large due to new research knowledge.

Special Populations

Certain classes of subjects including minors, prisoners, pregnant women, human fetuses, neonates, and those unable to consent to research participation on their own benefit from special protections when it comes to health policy or services research. If your research project involves any of these special populations, you should consult with your IRB to learn how to include adequate human subject protections for these special populations.