1.1 Why Research Involving Humans Has Rules

Health research asks people to give their time, their information and sometimes their bodies so that others may benefit from new knowledge. Participants trust that researchers will tell them the truth about a study, will avoid exposing them to needless harm, and will treat what they share with care. The rules of research ethics exist because that trust has been broken many times, often by respected scientists in respected institutions, and often at the expense of people who had the least power to refuse.

HSCI 130 introduced some of these cases: the Tuskegee study and the Belmont Report in its lesson on sexual and reproductive health, and the eugenic sterilization laws of Alberta and British Columbia in its lesson on genetics. This lesson returns to that history with a practical question for each case: what rule would have prevented the harm, and where does that rule now appear in Canadian research policy? Click each card to read a short account.

Unethical research also occurred in mainstream academic medicine. In 1966 the anaesthesiologist Henry K. Beecher described 22 published studies in which participants had been exposed to risk, usually without being told (Beecher, 1966). His examples came from leading hospitals and journals, and his paper helped persuade institutions that investigators could not be left to judge their own studies.

The cases share several features. The participants were often poor, institutionalized, racialized, disabled or Indigenous, and had little power to refuse. They were told little about the research or were misled about it. Their welfare was treated as a cost worth paying for knowledge that would mainly benefit others. In almost every case, nobody independent of the research team reviewed the study before it began. Modern research ethics policy responds to each of these features.

1.2 From Codes to TCPS 2

The Nuremberg Code (1947) made voluntary consent the first condition of ethical experimentation. The World Medical Association’s Declaration of Helsinki (1964), revised several times since, set out principles for medical research, and its later versions call for protocols to be reviewed by an independent ethics committee. In the United States, public reaction to the Tuskegee study led to the Belmont Report (1979), which named three principles: respect for persons, beneficence and justice.

In Canada, the Medical Research Council issued guidelines in 1978 and 1987. In 1998 the three federal research funding agencies, the Canadian Institutes of Health Research (CIHR), the Natural Sciences and Engineering Research Council of Canada (NSERC) and the Social Sciences and Humanities Research Council of Canada (SSHRC), issued the first Tri-Council Policy Statement. Its second edition, TCPS 2, appeared in 2010 and was revised in 2014, 2018 and 2022. This lesson uses TCPS 2 (2022), cited as CIHR, NSERC and SSHRC (2022). The Panel on Research Ethics develops, interprets and updates the policy on the agencies’ behalf.

Institutions that wish to receive agency funding agree to apply TCPS 2 to all research involving humans conducted under their auspices, whatever the source of funding for a particular project. Simon Fraser University is one of these institutions, so the policy applies to research by SFU faculty, staff and students, including unfunded student research. TCPS 2 operates alongside the law, including provincial privacy legislation, and researchers must meet both.

1.3 Respect for Human Dignity and the Three Core Principles

TCPS 2 rests on one underlying value, respect for human dignity, which requires that research involving humans be conducted in a way that is sensitive to the inherent worth of all human beings and the respect and consideration they are due. The policy expresses this value through three core principles: respect for persons, concern for welfare and justice (CIHR, NSERC & SSHRC, 2022, Chapter 1). The principles are applied together and in context, and the rest of the policy can be read as instructions for putting them into practice.

Respect for persons

Respect for persons recognizes the intrinsic value of human beings. It carries two obligations: to respect the autonomy of people who can make their own decisions, and to protect people whose autonomy is developing, impaired or diminished. Autonomy means the ability to deliberate about a decision and to act on it. The main practical expression of the principle is consent that is free, informed and ongoing, which Section 3 develops. For people who lack the capacity to decide, someone legally authorized decides on their behalf, and the person’s own wishes are still sought and respected.

In the Cedar Valley Social Connection Study (a fictional study used throughout this course), respect for persons means that older adults are presumed capable of deciding for themselves, because age alone is never evidence of reduced capacity. It also means that the team needs a plan for potential participants who live with dementia or another condition that affects decision-making.

Concern for welfare

TCPS 2 defines the welfare of a person as the quality of that person’s experience of life in all its aspects, including physical, mental and spiritual health and physical, economic and social circumstances. Concern for welfare asks researchers and REBs to protect participants’ welfare and, where possible, to promote it, through a favourable balance of foreseeable harms and potential benefits. Privacy is part of welfare, so information risks count as seriously as physical ones. The principle extends to groups: findings about a small town or a First Nation can affect everyone associated with it.

For the Cedar Valley team, questions about loneliness may bring up sadness or grief. Linking survey answers to records of physician and emergency department visits creates a privacy risk if the data were ever exposed. A report that names “the loneliest community in the region” could stigmatize the residents of that town.

Justice

Justice is the obligation to treat people fairly and equitably. Fairness means treating all people with equal respect and concern. Equity means distributing the benefits and burdens of research participation so that no segment of the population is unduly burdened by its harms or denied the benefits of the knowledge it generates. Justice also directs attention to power, between researcher and participant and between institutions and communities. TCPS 2 treats vulnerability as a matter of circumstance: a person may be vulnerable when a clinician who controls their care asks them to join a study, and less so in other settings.

In Cedar Valley, a survey offered only online and only in English would exclude many of the oldest residents, people with limited literacy, and people who speak another language at home. Recruiting only in Cedar City would leave the region’s small towns and First Nations communities without evidence about their own needs.

1.4 When the Principles Pull in Different Directions

The hard cases arise when the principles point in different directions. The most common tension in health research is between protection and inclusion. A team concerned for welfare may decide to exclude groups it sees as fragile, such as very old adults, people with dementia or people living in care homes. Exclusion protects them from the burdens of research and also denies them its benefits, which raises a question of justice. Chapter 4 of TCPS 2 states that researchers should not exclude people on the basis of attributes such as culture, language, religion, race, disability, sexual orientation, ethnicity, linguistic proficiency, gender or age unless there is a valid reason for the exclusion.

Other tensions arise between respect for persons and concern for welfare, as when a participant wishes to share an identifiable story that could harm them or others, and between individual and collective interests, which Section 4 discusses for research with First Nations, Inuit and Métis communities. In each case, TCPS 2 asks researchers to reason through the competing principles in context, record the reasoning, and let the REB judge the balance.

The revised plan shows the reasoning REBs look for. The team identified the principles in tension, changed the design to reduce the tension, kept two exclusions for stated reasons, and acknowledged their cost. An exclusion based on convenience alone, such as offering only an online survey because paper forms take more work to enter, would be much harder to defend.

2.1 Deciding Whether a Project Needs REB Review

The first practical question is whether a project needs review at all. TCPS 2 answers it with two definitions. Research is an undertaking intended to extend knowledge through a disciplined inquiry or systematic investigation. Human participants are individuals whose data, biological materials, or responses to interventions, stimuli or questions by the researcher are relevant to answering the research question. Research involving living human participants or human biological materials requires approval by a research ethics board (REB) before it begins, and the research begins with recruitment. A team that posts a recruitment flyer before approval has started its study without approval.

Chapter 2 of TCPS 2 also describes activities that do not require REB review. Open each item for the conditions and an example.

Researchers should avoid deciding alone that a project falls outside review. When the answer is unclear, institutions generally ask researchers to request a written determination from their research ethics office. The flowchart summarizes the sequence of questions.

2.2 Thinking About Risk

A harm is anything that has a negative effect on the welfare of participants. Health research can cause physical harms, such as injury from a procedure; psychological harms, such as distress or embarrassment; social harms, such as stigma; economic harms, such as lost wages; and legal harms, such as exposure to legal action if sensitive information is disclosed. Many harms in survey and interview research are informational: they arise when what participants shared reaches someone it should not reach. Risk combines the probability that a harm will occur with its magnitude, or seriousness, if it does.

TCPS 2 sets the threshold for a lighter review with its definition of minimal risk: research in which “the probability and magnitude of possible harms implied by participation in the research is no greater than those encountered by participants in those aspects of their everyday life that relate to the research” (CIHR, NSERC & SSHRC, 2022, Chapter 2). The comparison is with the everyday lives of the participants themselves. Being asked how often one feels left out resembles the questions an older adult might hear from a family physician or discuss with a friend, so a short loneliness questionnaire is usually minimal risk. Detailed questions about suicidal thoughts, or questions to people who use drugs about illegal activity, go beyond everyday experience and may raise the risk above minimal.

Researchers must also weigh benefits. Direct benefits to participants are uncommon in survey and interview research, and consent documents should say so plainly; the main benefit is usually knowledge that helps others. Harms and benefits can also fall on groups: a finding that one small town has high rates of loneliness may help it obtain services and may also stigmatize its residents. Researchers reduce risk through mitigation, the steps that make a harm less likely or less serious, such as a distress protocol, secure storage, or combining small areas in reports.

2.3 Levels of REB Review

TCPS 2 asks REBs to take a proportionate approach: the more invasive or risky the research, the greater the care in assessing it. The REB first determines the level of risk and then the level of review. Research above minimal risk receives full REB review, which is the default. Minimal-risk research may receive delegated review. The researcher may propose a level, and the REB decides.

TCPS 2 sets the minimum composition of an REB: at least five members, including at least two with expertise in relevant research disciplines and methods, at least one knowledgeable in ethics, at least one community member with no affiliation with the institution, and, for biomedical research, at least one knowledgeable in the relevant law. Institutions must give REBs the independence to approve, reject, propose modifications to, or stop research. REB members manage their own conflicts of interest, for example by leaving the room when a colleague’s application is discussed.

When a study involves more than one institution, Chapter 8 of TCPS 2 allows several models of review, from separate review by each institution’s REB to arrangements in which one REB’s review is accepted by the others. British Columbia has a provincial harmonized process for health research involving more than one of its universities and health authorities, so a study like Cedar Valley can go through one coordinated review.

2.4 Applying Through SFU’s Office of Research Ethics

At Simon Fraser University, the Office of Research Ethics administers the ethics review of research involving human participants, and the university’s Research Ethics Board makes the review decisions. Faculty, staff and students apply through this office, and students conducting research for their degree normally apply with their faculty supervisor, who shares responsibility for the project. Applications are submitted online. The office’s website lists current forms, guidance documents, templates, submission procedures and information about review times. These details change, so check the website when you prepare an application instead of relying on a colleague’s old forms. No recruitment or data collection may begin until approval is granted.

Application forms differ between institutions, but they ask about the same matters, because those matters come from TCPS 2.

Ethics approval is often one of several approvals. Research in a health authority’s facilities usually also requires the health authority’s operational or institutional approval. Research with a First Nation may require the approval of the Nation’s leadership or health organization, often set out in a research agreement. Access to administrative health data requires approval from the data stewards responsible for each data set (Lesson 9), and clinics that allow chart abstraction usually sign an agreement. Project plans (Lesson 6) should allow time for each.

Several habits make applications easier to review. Write for a reviewer outside your field. Keep numbers and procedures identical across the application, protocol and consent form, because inconsistencies are a common reason for requests for revision. Put a version number and date on every document. When the REB asks for changes, reply in a memo that quotes each comment, states what changed, and shows where.

3.1 Consent as a Process

TCPS 2 treats consent as a process that begins with the first contact between researcher and participant and continues until the participant’s involvement ends. A signature records one moment in that process. The policy describes the consent it requires as free, informed and ongoing (CIHR, NSERC & SSHRC, 2022, Chapter 3).

Free: voluntary consent

Consent must be given voluntarily and can be withdrawn at any time. Voluntariness is threatened by undue influence, which arises when people are recruited by someone with authority over them, such as their physician, instructor or employer, and feel unable to refuse. Coercion is a more extreme form, involving a threat of harm or punishment for refusing. Modest incentives that recognize participants’ time are acceptable, while an incentive large enough to lead people to accept risks they would otherwise refuse compromises voluntariness. In Cedar Valley, clinic reception staff hand out study cards so that physicians never ask their own patients, and everyone keeps the thank-you gift card even if they stop early.

Informed: disclosure and understanding

Article 3.2 of TCPS 2 lists the information researchers must provide. Grouped for practical use, it covers the topics in the table.

Disclosure alone does not make consent informed; people must also understand what they have been told. In the teach-back technique, the researcher asks the person to explain the study in their own words, for example “What happens if you decide to stop partway through?”, and clarifies anything missed. Researchers also watch for the therapeutic misconception, the belief that a study is designed to benefit the participant personally (Appelbaum et al., 1982). Some Cedar Valley participants may assume that reporting loneliness will connect them to services, so the form states that it will not and lists resources separately.

Ongoing: new information and withdrawal

Researchers must tell participants about changes or new information that might affect their decision to continue, and must seek consent again at new stages, such as a second interview. Participants who withdraw may ask for their data to be removed, and the consent form must explain any limits, for example that anonymous answers cannot be found once submitted.

3.2 Consent Documents and Formats

Most studies use an information letter, which explains the study, and a consent form, which records the decision, often combined in one document. TCPS 2 allows consent to be documented in a signed form or by other appropriate means that the researcher records, which gives researchers a choice of formats.

Studies with optional components use tiered consent, with a separate yes or no choice for each component, such as data linkage, audio recording, quotations or future contact. Declining an optional component should never exclude a person from the main study. Consent documents should be written in plain language, with short sentences, questions as headings, the reader addressed as “you”, technical terms defined, and a large font for older readers. Many REBs recommend about a grade 6 to 8 reading level, and the best test is to ask members of the participant population to read a draft and explain it back. The Cedar Valley advisory group helped produce these rewrites.

3.3 Capacity

TCPS 2 defines capacity as the ability of prospective or actual participants to understand relevant information about a research project and to appreciate the potential consequences of their decision to participate or not. Capacity is specific to the decision and can change over time: a person may be able to consent to a short survey and unable to weigh a complex trial, or able to decide on one day and unable on another. Adults are presumed to have capacity unless there is reason to think otherwise.

TCPS 2 permits research with people who lack capacity only under conditions, including that the research question can be addressed only with participants from that group, that consent comes from an authorized third party who is not on the research team, and that research without the prospect of direct benefit involves no more than minimal risk. Provincial law determines who counts as an authorized third party. Where the person can understand something of the research, researchers also seek their assent, and their dissent means they do not take part. People who regain capacity are asked for their own consent to continue, and TCPS 2 recognizes research directives, in which a person who expects to lose capacity records their wishes about future research.

The Cedar Valley interviews may include people with mild cognitive impairment who can consent for themselves. Before each interview, the interviewer reviews the information sheet and asks three teach-back questions about the study’s purpose, the right to stop, and the recording. If a person cannot answer them after explanation, the interviewer ends the session kindly and uses no data. The study does not seek third-party consent, consistent with the exclusion explained in Section 1.

3.4 When Consent Requirements Can Be Altered

Some studies cannot be done if every participant consents fully in advance. A study of how pharmacists counsel customers might use researchers posing as customers, because pharmacists who knew they were observed would change their behaviour. Article 3.7A of TCPS 2 allows an REB to approve research without prior consent, or with an alteration of consent requirements such as partial disclosure or deception, only when the REB is satisfied, and documents, that all of these conditions hold: the research involves no more than minimal risk; the alteration is unlikely to adversely affect participants’ welfare; it is impossible or impracticable to carry out the research and answer the question properly with prior consent; the precise nature and extent of the alteration is defined; and any debriefing plan meets the policy’s requirements. TCPS 2 defines impracticable as a degree of hardship or onerousness that jeopardizes the conduct of the research, a higher bar than inconvenience. A debriefing explains afterwards what was withheld and why, and may offer the chance to withdraw data.

Two situations are easily confused with an alteration. Choosing oral or implied consent changes only how consent is documented. Using identifiable information collected for another purpose, such as clinic charts, without consent falls under the secondary use provisions of Chapter 5, which Section 4 applies to the Cedar Valley chart review.

3.5 Worked Example: The Annotated Cedar Valley Consent Form

The form below is the revised information and consent form for the Cedar Valley regional survey, written with the advisory group and amended after REB review. The survey uses implied consent, and two optional parts, linkage to health records and future contact, use signed tiered consent. Contact details appear as placeholders because the study is fictional. Each part of the form is paired with a note on the requirement it meets.

The finished form runs to about 1,100 words, longer than the team first wanted. The advisory group judged that every paragraph answered a question older adults would ask, so the team shortened sentences instead of removing content, used a 14-point font on paper, and placed the key information first.

4.1 Privacy, Confidentiality and Security

Three related terms organize Chapter 5 of TCPS 2. Privacy is an individual’s right to be free from intrusion or interference by others, which includes some control over information about oneself. Confidentiality is the obligation of a person or organization to safeguard information entrusted to it, protecting it from unauthorized access, use, disclosure, modification, loss or theft. Security refers to the measures used to protect information: physical safeguards such as locked cabinets, administrative safeguards such as confidentiality agreements and access rules, and technical safeguards such as passwords and encryption. In the Cedar Valley study, an older adult’s privacy is the interest at stake, the team’s promise in the consent form creates its duty of confidentiality, and the locked cabinet and password-protected server are its security measures.

TCPS 2 requires researchers to safeguard information entrusted to them and not to misuse or wrongfully disclose it. Confidentiality also has limits, and participants must be told about those that are reasonably foreseeable. Provincial law may require disclosure in some circumstances; in British Columbia, for example, anyone who has reason to believe a child needs protection must report it. In focus groups, researchers can ask members to keep what is said confidential but cannot guarantee it. Privacy law also applies alongside TCPS 2: British Columbia’s Freedom of Information and Protection of Privacy Act governs public bodies such as universities and health authorities, which shapes where and how they may store personal information.

4.2 How Identifiable Is the Information?

Privacy risk depends on how easily information can be linked to a person. TCPS 2 describes five categories, which can be pictured on a spectrum from most to least identifiable.

Directly identifying information identifies a specific person through direct identifiers such as a name, social insurance number or personal health number. Indirectly identifying information can reasonably be expected to identify a person through a combination of indirect identifiers, such as date of birth, place of residence or a unique personal characteristic. Coded information has had its direct identifiers removed and replaced by a code, and whoever holds the key linking codes to names can re-identify participants. Anonymized information has been irrevocably stripped of direct identifiers, no key is kept, and the risk of re-identification from the remaining details is low or very low. Anonymous information never had identifiers associated with it, and its risk of identification is also low or very low.

Indirect identifiers deserve particular attention in rural research. In Cedar City, a 78-year-old woman who lives alone is one of many. In Kestrel Lake, a woman aged 90 or older who lives alone and belongs to a First Nation may be the only person who fits that description, and her survey answers could be recognized by anyone who knows her. Survey answers collected without names are therefore anonymous in the TCPS 2 sense only if the remaining details carry a low risk of identification. Researchers reduce the risk by grouping ages into bands, reporting geography at a larger scale, combining or suppressing small counts, and reviewing interview quotations for recognizable details. Lesson 9 develops these de-identification techniques and the Five Safes framework for secure data access.

4.3 Secondary Use of Data and Data Linkage

Secondary use is the use in research of information originally collected for another purpose, such as clinical care, health system administration or an earlier study. Secondary use of anonymous information that produces no identifiable output falls outside REB review (Section 2). Secondary use of identifiable information requires either the consent of the people it describes or REB approval to proceed without consent. Article 5.5A of TCPS 2 allows the second route only when the REB is satisfied that six conditions are met: identifiable information is essential to the research; use without consent is unlikely to adversely affect the welfare of the people concerned; the researchers will take appropriate measures to protect privacy and safeguard the information; the researchers will comply with any known preferences previously expressed by those people about the use of their information; it is impossible or impracticable to seek their consent; and the researchers have obtained any other necessary permission for the secondary use.

Data linkage joins two or more data sets so that information about the same person can be analyzed together. TCPS 2 requires REB approval before researchers carry out a linkage that could produce identifiable information, and the application must describe the data to be linked and the likelihood that identifiable information will be created. The Cedar Valley linkage to administrative health records rests on participants’ consent (Note 5 of the consent form in Section 3). The chart review does not, so the team had to justify it under Article 5.5A.

The REB in this illustrative case accepted the justification on condition that an abstractor who recognizes a patient they know personally skips that chart. The example shows how the conditions work as a checklist: each must be met, and a weak answer to any one of them, such as claiming impracticability when consent would only be inconvenient, can lead the REB to refuse.

4.4 Data Management Plans

A data management plan (DMP) is a living document that describes how a project’s data will be collected, documented, stored, protected, shared, preserved and eventually destroyed. Under the Tri-Agency Research Data Management Policy (2021), institutions develop research data management strategies, DMPs are required for certain funding opportunities, and data that support publications may need to be deposited in a repository. The policy also expects that data from research by and with First Nations, Inuit and Métis communities will be managed according to principles and approaches those communities accept. The DMP Assistant, a free bilingual online tool hosted by the Digital Research Alliance of Canada, guides researchers through a standard set of questions. REB applications ask many of the same questions, so a DMP written early serves both purposes. Lesson 6 develops the practical side of data management, including file organization and version control.

4.5 TCPS 2 Chapter 9: Research Involving First Nations, Inuit and Métis Peoples

Chapter 9 of TCPS 2 responds to a history in which research was done on Indigenous peoples without their consent or benefit, including the nutrition experiments and the Nuu-chah-nulth samples described in Section 1. It applies the core principles to communities as well as individuals and recognizes the right of First Nations, Inuit and Métis peoples to self-determination in research that affects them. Lesson 4 introduced the OCAP® principles of ownership, control, access and possession and showed how to write an engagement plan; this section explains what the ethics policy requires.

Where research is likely to affect the welfare of an Indigenous community to which prospective participants belong, researchers must seek engagement with that community (Article 9.1). Chapter 9 describes situations in which this applies, including research conducted on First Nations, Inuit or Métis lands, recruitment based on Indigenous identity, research about a community’s cultural heritage or traditional knowledge, analysis that uses Indigenous identity as a variable, and interpretation of results that refers to Indigenous communities or peoples. The nature of engagement should fit the community and the research, and it is often formalized in a research agreement. Chapter 9 also asks researchers to respect community customs and codes of research practice, to support collaborative approaches and capacity building, to give communities the opportunity to take part in interpreting the data and reviewing findings before publication, and to engage the community before secondary use of information that can identify it. Community engagement does not replace individual consent: each participant still decides for themselves.

The Cedar Valley study recruits in First Nations communities, plans to analyze results by First Nations identity, and partners with the Cedar Valley First Nations Health Centre, so Chapter 9 applies. Click each card for a component of the illustrative research agreement.

Bringing It All Together

This lesson treated research ethics as part of research method. The historical cases in Section 1 show what happens when consent is absent, harms are ignored and burdens fall on people with little power, and TCPS 2 (2022) responds with one underlying value, respect for human dignity, and three core principles: respect for persons, concern for welfare and justice. Section 2 turned those principles into procedure: deciding whether a project needs REB review, judging risk against the minimal risk standard, and preparing an application for delegated or full review.

Sections 3 and 4 dealt with the documents and safeguards that REBs read most closely. Consent is a process that must be free, informed and ongoing, with formats chosen to fit the study and plans for people whose capacity is reduced. Privacy protection depends on how identifiable the information is, secondary use of identifiable data without consent must meet six conditions, a data management plan sets out the life of the data, and Chapter 9 requires engagement with First Nations, Inuit and Métis communities whose welfare a study may affect. The annotated Cedar Valley consent form brought these requirements together in one document.

The final reflection asks you to read the Cedar Valley study as an REB member would and to apply the principles and requirements from all four sections.

Final Knowledge Assessment

Definitions of the research ethics terms, frameworks and people introduced in this lesson, based on TCPS 2 (2022) where the policy defines a term.