Children

The possibility for undue influence must be considered separately for the child participating in research and the parent providing permission.

• Children. The payment plan should consider the age of the child and their ability to assess the value of the payment against the research contribution they are making.
  In general, a toy may be a more appropriate incentive or compensation for younger subjects while a gift card or cash may be more appropriate for an adolescent. Incentives and compensation for the child should generally be given directly to the child, rather than the parent.
• Parents. It may be acceptable to reimburse or compensate parents for the costs and burden experienced by facilitating their child’s participation in research, if the amounts are not likely to cloud the parent’s judgement about what is best for the child. Incentive payments to parents who are not also study subjects are generally not appropriate.

Prisoners

There is wide variation in whether federal, state, and local institutions consider it appropriate to provide research compensation or incentives to incarcerated individuals.

The UW IRB allows compensation and incentives to incarcerated subjects, if appropriate and allowed by the institution(s). However, compensation and incentives may be more likely to be unduly influential because of the limited opportunities prisoners have to earn money and to obtain specific items or services as well as the inherent limitations on their autonomy resulting from their incarceration.

Although there are no quantitative studies measuring actual undue influence in incarcerated subjects, several qualitative studies have been conducted asking prisoners about their experience with research and their perceptions about undue influence related to payment. These studies found that prisoners think they should be compensated for their contribution, that payment is a motivator for enrolling but is generally not the only or the primary motivator, and that the prisoners would likely still enroll in a study even without compensation. Of concern, one study found that the potential for undue influence was significantly higher among prisoners with lower educational attainment and those who had spent a longer time in prison. So, just like with all other populations, the likelihood of payment unduly influencing incarcerated subjects to enroll in research will vary depending on the subject and the study specifics.

Other groups

Vulnerable populations are groups of subjects who may be at higher risk of undue influence or who may be unable to easily understand the consent information. This may include, but is not limited to: individuals who are decisionally-impaired; individuals of lower socio-economic status; illiterate individuals; individuals with low educational attainment; non-English speakers or speakers of English as a second language; undocumented migrants; and subjects with addiction issues. The particulars of how to minimize the likelihood of undue influence for a vulnerable population will depend on the specifics of the study and subject population.

Elements of a payment plan

A well-designed payment plan should specify these elements:

• Reimbursement, compensation and/or incentives. Which of these categories of payment will be offered? Reimbursement and compensation are nationally recognized to be an issue of fairness to subjects and an ethical best practice that is especially encouraged for funded clinical research. For some studies, a single payment may be intended to address all three categories of payment. For other studies, there may be incentive payments that are separate from reimbursement or compensation.
• Type (method) of payment. How will the payment be provided? Examples: check, cash, gift card, toy, drawing for a prize, class credit. Note that the options for reimbursement and compensation are somewhat limited (e.g., cash, check, gift card) whereas there are many possible options for incentives.
• Amount. How much will be provided? The amount should not be unduly influential for the circumstances and study population. Factors to consider:
  • Reimbursement. Consider the amount of out-of-pocket expenses for participants, including (for clinical research) any health insurance co-pays or deductibles. Which will be reimbursed? Will reimbursement be for actual expenses, or for an average/typical amount? Will participation expenses be balanced by other payments associated with the research (e.g., compensation; a single combined payment intended to provide reimbursement plus some incentive) or by study-delivered services or return of individual study results?
  • Compensation. Consider the time and effort (burden) associated with participation. What will be the basis for determining the amount of compensation? Examples: an hourly rate based on the average hourly wage in the research location; a set amount for the time and effort associated with specific procedures or study visits.
  • Relationship with risk. Federal guidance states that it is acceptable to consider the study risks when deciding on the incentive amount. Empirical studies (e.g., see References) have consistently shown that large incentives are generally considered an indicator of study risks by participants. If risk level is part of the rationale for a large incentive, then it may help minimize the possibility of undue influence to state in the consent process that the size of the incentive was chosen to be appropriate for the risks of the study.
  • Impact on enrollment goals. Payment may affect whether enrollment goals are met (and therefore affect the overall benefit of conducting the study).
  • Impact on population diversity. Payment amounts may influence the diversity of the enrolled participants and therefore the generalizability (societal benefits) of the research. For example, if unreimbursed expenses are significant, individuals with lower income and/or inflexible working arrangements may not be able to afford to participate. For some research, this could result in a significantly unrepresentative study population.
  • Potential exploitation may be an ethical concern if subjects are not offered adequate compensation or other benefits compared to the value of their contribution. When the costs and burden of volunteering for research are high, higher payments may be warranted.
• Pro-rating. Prorating means dividing up the total payment into two or more payments across time. This reduces the possibility that subjects will stay in a study just to receive the payment when they would otherwise choose to withdraw from it. If appropriate, the payment schedule should reflect that subjects will receive reimbursement and compensation payments as the study progresses and for all procedures completed, regardless of whether a subject completes the study, withdraws, or is withdrawn. For studies of short duration (e.g., a few weeks) it may be acceptable to pay only at the end of the subject’s participation, but for studies of longer duration (e.g., several months), there should be multiple payment points (e.g., monthly or after some set number of sessions). Pro-rating is an important way to reduce the possibility of undue influence that the UW IRB strongly encourages and may require.

The role of informed consent

Designing a consent form and process that supports and promotes autonomous and informed decision-making can help minimize the likelihood that prospective subjects will ignore or misunderstand study risks or other key information because of their interest in receiving payment. This includes an ongoing consent process that ensures subjects comprehend risks throughout the study and are updated about any changes to risks and benefits. An adequate consent process allows participants to decline participation in research that would be subjectively harmful according to their own judgement of their interests, values, and obligations.

Payment information to provide as part of the consent process

Per nationally recognized best practice and UW IRB policy, the consent process (or consent form, if there is one) must provide the following payment information:

• The total amount and method of reimbursement, compensation, incentives and/or combined payment; if appropriate, reimbursement and/or compensation may be described separately from incentives
• If appropriate, a description of any pro-rating and when payment will occur for the procedures that have been completed if subjects withdraw or are withdrawn
• Approximately how long it will take to receive payments. One of the most frequent subject complaints received by HSD is about the timing of payments; subjects typically expect that they will be paid as soon as they finish the study or promptly thereafter, but that rarely occurs.
• If is is possible or likely that a subject participating in U.S. research will earn $600 or more in payments (not including reimbursements) for all UW studies in which they participate during the calendar year, those subjects must be informed that it is considered Miscellaneous Income and will be reported to the Internal Revenue Service for federal income tax purposes. See the FAQs for more information about federal income tax and subject payments.
• Consider the study population when deciding whether participants should be informed that Social Security Numbers (SSNs) need to be collected for compensation and incentive payments (but not reimbursement). This might be of concern for populations such as gig workers, undocumented individuals, or foreign nationals. Note that HSD and the UW IRB can appropriately advise researchers about including this information in the consent process but cannot offer advice about when collection of SSNs is, or is not, required. UW Financial Management is responsible for compliance and advice about this requirement: https://finance.uw.edu/ps/how-pay/research-subjects.
• If the size of any incentives is related to the study risks, consider whether subjects should be informed of the relationship by including a statement such as “This amount of money was selected as being appropriate for the level of risk to study participants”. Some controlled experimental research showed that such statements make participants more vigilant about seeking and viewing study risk information.
• If applicable, a statement about anything for which subjects will not be compensated. For example, if biospecimens collected for research may result in commercial profit, subjects should be informed if they will not share in those potential profits.

Advertising of IRB-approved payment is acceptable

Subjects must be made aware of the information so that it can facilitate enrollment, as it is intended to do. Payment information in recruiting materials should be truthful, clear, and not emphasized at the expense of other important information.

“At the expense of” means that the payment information is presented in the absence of other important information (e.g., research purpose; time commitment; brief description of procedures) or in a way that is likely to cause individuals to neglect or overlook the other information provided. In many cases, it is acceptable for payment information to be the featured or most prominent information so long as other important information is not missing or provided in hard-to-perceive ways.

Which study information (and how much) is included in recruiting materials is a function of the specific recruiting method and its information capacity (e.g., a small flyer posted on a bulletin board; a “cold contact” letter; a study website; Craig’s List posting)

Criteria for IRB approval

The IRB considers incentives, reimbursement, and compensation in the context of several IRB criteria for approval (and the Belmont ethical principles underlying those criteria):

• Undue influence. Incentives should not be so large as to create undue influence on the decision to enroll and to remain enrolled (see What is Undue Influence, above). This is a difficult concern to address. For example, there is some evidence that rather than blinding subjects to risks or impairing comprehension, incentives (especially large ones) may increase levels of caution and perceptions of risk.
• Vulnerable populations. An incentive that is not unduly influential for a general population may have a larger impact on specific populations (see Population Specific Issues).
• Equitable selection of subjects. If the IRB limits payments in order to protect prospective subjects of lower socioeconomic status (SES) from undue influence, these individuals may still be encouraged by the payment to enroll but would receive a lower payment for their contribution. This strategy could also result in fewer subjects of a higher SES. An imbalanced representation of SES in the subject population could be contrary to the Belmont Report ethical principle of Justice with its emphasis on equal distribution of research risks, burdens, and benefits across populations.
• Research benefits are likely to justify the risks.
  The benefit of most research is the knowledge or societal good that is likely to be gained. Limiting incentives for participation may be a hinderance to recruitment. If a study does not meet its recruitment goals and is therefore underpowered or terminated, it is unlikely that the research question can be answered. If the research question cannot be answered, then the IRB’s assessment that risks were reasonable in relation to anticipated benefits may no longer be appropriately balanced and subjects may have been exposed to risks for limited or no societal benefit. For example, low response rates are a known and common problem for electronic surveys. Low response rates can introduce bias, compromise the validity of the study, and increase the cost of conducting the study.There is some evidence that larger incentives may also result in a more diverse pool of research participants, which in turn may result in more representative and generalizable results as well as a more equal distribution of the potential benefits (individual and societal) of the research.
• Identifying the procedures as research rather than other activities (such as clinical care or classroom instruction). Some recent evidence suggests that incentives may draw attention to the distinctions between research and clinical care. For example, if there is an additional, research blood draw associated with a clinical visit, compensating for the research blood draw may help draw attention to the fact that the blood draw is a research procedure instead of a clinical procedure.

The nature of the study population: What would be an incentive for them?The possible cost and effort required for the subjects to make use of the incentive.

Age appropriateness

Cultural appropriateness

Privacy and confidentiality issues. There are many ways in which privacy or confidentiality concerns could arise with respect to incentives. The relevance of these to the specific study circumstances should be considered. Some examples:

• Subjects may be placed at risk if others observe them receiving cash or a check.
• Incentives that are mailed to the subjects’ homes may go astray or be intercepted.
• Checks, and incentives requiring an online account, may create risks for undocumented individuals in this country.

Many payment methods require the involvement of finance and other administrative staff that are not part of the study team (e.g., department administrator, Shared Environment specialist, Tax Office). This is not considered a significant privacy and confidentiality risk, as the roles and responsibilities of these positions require discretion, confidentiality, and appropriate security measures and the information accessed is limited to that necessary to facilitate payment.

For researchers processing payments via Workday, you can minimize the association of private identifiable study data to the participant by ensuring use of a non-descript worktag name. Avoid using worktag names that include a study acronym that is searchable on the web or has the condition under study (e.g., GR12346 HIV Study). If a particular study population may have concerns that their financial information is accessible to others outside the study team, you can provide additional disclosure information as part of the consent process.

The logistics associated with each type of incentive, including department and researcher responsibilities. For example, giving checks to subjects requires the establishment, management, and auditing of a checking account as well as independent financial controls. Department financial administrators should be able to answer questions about whether a specific method is possible or preferred.

The research budget

Legal and tax issues. There may be legal and tax issues associated with specific types/amounts of payments and specific study populations. The Human Subjects Division and the IRB cannot provide advice on these issues. Consult UW Financial Management through email or their Pay Research Subjects website. Examples of some considerations:

• Payments made to foreign nationals in the U.S. must be made through a specific mechanism. In addition, those payments are automatically subject to 30% income tax withholding. In other words, paying subjects $100 means that foreign national subjects would actually receive a check for $70.
• Payments to an individual must be reported by the UW to the U.S. Internal Revenue Service as Miscellaneous Income if the payments total $600 or more annually. This means that subject Social Security Numbers must be collected.
• Questions about payment methods, income tax reporting, and collection of Social Security Numbers should be directed to UW Procurement Customer Service at pcshelp@uw.edu or 206.543.4500.

Minimizing the possibility of undue influence

Subjects should be informed of the value of the item, the approximate chance of winning (in the context of the expected enrollment), and the timing of the drawing. There should be a reasonable balance between the cash value of the prize and the number of prizes offered. The IRB may ask for the researcher’s justification if one item of significant value (e.g., an electronic tablet, a new smart phone) is being offered instead of several items of smaller value.

The Belmont principle of justice

The Belmont principle of justice requires equal treatment of all participants, which is not necessarily the same as equal payment. With drawings, each participant is treated equally in being given the same opportunity to win a prize.

Legal issues

HSD and the IRB cannot offer guidance about any legal issues or requirements associated with drawings.

Minimizing the possibility of undue influence

It is UW IRB policy that a non-research alternative must be offered to earn the same extra credit. The non-research alternative must not involve more time or risk than the research participation.

Other issues to consider

The researcher and the IRB should consider the possibility of coercion when a study is offered to a specific class, especially if researcher is the teacher. The circumstances should be managed so that students aren’t penalized for not taking part. This is typically accomplished by providing some way to keep the teacher/researcher from knowing who did or did not participate until after the course has been completed and grades have been assigned.

Minimizing the possibility of undue influence

Because these are of low monetary value, the possibility of unduly influencing subjects to participate is low. However, there may be special considerations associated with specific subjects populations such as prisoners.

Other issues to consider

The cultural, and age appropriateness of the items should be considered. For example, items for infants or young children should be not be too small in size, have small removable parts, or made of materials that could be toxic if ingested (e.g., lead paint, or food with nuts that could be dangerous to children with nut allergies).

Minimizing the possibility of undue influence

As with other types of subject payment, the permissibility of this method depends on whether the amount paid for each referral, as well as the total amount a subject may receive for peer recruitment, is appropriate given the subject’s time and effort. The IRB must also assess the potential for coercion or undue influence that the enrolled subject may direct against the individuals they attempt to recruit. Researchers should provide guidelines to peer recruiters in order to minimize this possibility.

Other issues to consider

This is an incentive for assisting researchers in conducting their research, rather than incentive for participation in the research as a subject. Depending on the study, it may raise significant privacy and confidentiality concerns that the researcher and IRB need to manage.