# Lesson 5: Research Ethics in Practice: TCPS 2

*Companion-podcast transcript, Sarah and Kiffer*

---

**Sarah:** Welcome back to Office Hours. I'm Sarah.

**Kiffer:** And I'm Kiffer. This week we're on Lesson five of Research Methods in Health Sciences, which is about research ethics in practice.

**Sarah:** I'll admit that ethics sounds like the week where we read rules instead of doing research. Is that unfair?

**Kiffer:** It's a common reaction. My answer is that the ethics of a study is made of the same decisions as its method. Who you recruit, how you approach them, what you tell them, what you collect and where you keep it are all design decisions. Each of them is also an ethical decision, and each will be read by a research ethics board before you can start.

**Sarah:** And in Canada those boards work from one document.

**Kiffer:** They do. It's the Tri-Council Policy Statement on ethical conduct for research involving humans. Everyone calls it TCPS two, because it's the second edition, and the current version dates from twenty twenty-two.

**Sarah:** Remind us of the running case and what it involves.

**Kiffer:** The Cedar Valley Social Connection Study is a fictional study planned by a small team at a British Columbia university, led by a fictional researcher, Dr. Maya Hart, with the fictional Cedar Valley Health Authority. It studies loneliness among adults aged sixty-five and older. There's a regional survey with one thousand six hundred completed responses, and people who agree can have their answers linked to records of their doctor visits, hospital stays and emergency department visits. The team will also review three hundred clinic charts, run twenty-four interviews and four focus groups, and work with a First Nations health centre.

**Sarah:** The lesson starts with history. Why open an ethics lesson with old cases?

**Kiffer:** Because every rule in the policy answers a specific failure. In HSCI one thirty, students met the Tuskegee study and the Belmont Report, and the eugenic sterilization laws in Alberta and British Columbia. This lesson returns to that history and asks, for each case, which rule would have prevented the harm and where that rule now sits in Canadian policy.

**Sarah:** Walk me through Tuskegee, for anyone who hasn't seen it.

**Kiffer:** From nineteen thirty-two to nineteen seventy-two, the United States Public Health Service followed about six hundred Black men in Alabama, most of whom had syphilis, to watch the course of the untreated disease. The men were told they were being treated for bad blood, and when penicillin became the standard treatment, they didn't receive it. The study ended only after a Public Health Service employee, Peter Buxtun, took his concerns to the press.

**Sarah:** Which principles did it break?

**Kiffer:** All three of the ones we'll name in a minute. The men were deceived, effective treatment was withheld, and the burden fell on poor, racialized men who had little power to refuse.

**Sarah:** The lesson also includes Canadian cases.

**Kiffer:** It does. The historian Ian Mosby documented federal nutrition experiments in the nineteen forties and fifties in First Nations communities in northern Manitoba and in several residential schools. Researchers studied malnutrition among people who were already hungry, some children were kept on inadequate diets as comparison groups, and some were denied dental care because the state of their teeth was a study measure. In another case, blood samples that members of the Nuu-chah-nulth Nations gave for a study of arthritis were later used for genetic research on population history without their knowledge or consent.

**Sarah:** That second case involves no physical harm at all.

**Kiffer:** That's why it's useful. The harm was to people's control over their own samples and to a community's trust. It's a case about secondary use, which comes back in Section four.

**Sarah:** Was all of this happening at the margins, where nobody was watching?

**Kiffer:** Much of it happened in mainstream medicine, which was the point of Henry Beecher's paper in nineteen sixty-six. Beecher was a Harvard anaesthesiologist, and he described twenty-two published studies from leading hospitals and journals that had exposed participants to risk, usually without telling them. His paper helped convince institutions that researchers couldn't be left to judge their own studies.

**Sarah:** How did we get from those cases to TCPS two?

**Kiffer:** In stages. The Nuremberg Code came out of the trial of Nazi doctors in nineteen forty-seven, and its first principle is that the voluntary consent of the human subject is absolutely essential. The Declaration of Helsinki followed in nineteen sixty-four, and the Belmont Report in nineteen seventy-nine. In Canada, the three federal research funding agencies issued a joint policy in nineteen ninety-eight, and the second edition came out in twenty ten, with revisions since.

**Sarah:** And who has to follow it?

**Kiffer:** Any institution that wants funding from those agencies agrees to apply it to all research involving humans done under its auspices, whatever the source of the money. Simon Fraser University is one of those institutions, so the policy covers a student project with no funding at all.

**Sarah:** Then the principles themselves.

**Kiffer:** The policy starts from one underlying value, respect for human dignity, and expresses it through three core principles. Respect for persons means respecting the autonomy of people who can decide for themselves and protecting people whose autonomy is reduced, mainly through consent. Concern for welfare means protecting the quality of people's lives, including their privacy, by aiming for a favourable balance of harms and benefits. Justice means treating people fairly and equitably, so that no group carries the burdens of research or is shut out of its benefits.

**Sarah:** Give me Cedar Valley versions of each.

**Kiffer:** For respect for persons, the team presumes older adults can decide for themselves, because age by itself is never evidence of reduced capacity. For concern for welfare, it prepares for sadness during interviews, protects the linked health data and avoids reports that stigmatize a named small town. For justice, the survey is offered on paper, with a telephone help line and with interpreters, so the oldest residents and people without internet access can take part.

**Sarah:** The lesson says the principles sometimes pull against each other.

**Kiffer:** That's where the real work is. The most common tension is between protection and inclusion. A team worried about welfare may want to exclude people it sees as fragile, like very old adults or people in care homes. Excluding them also denies them the benefits of the research, which is a justice problem. Chapter four of the policy says researchers shouldn't exclude people on attributes like age, language or disability without a valid reason.

**Sarah:** How did the Cedar Valley team handle it?

**Kiffer:** Their first plan was an online survey in English. The advisory group of six older adults pointed out that many of the loneliest people in the region don't use the internet, live in long-term care, or speak another language at home. So the team now offers a paper version and a telephone help line that arranges interpreters. It still excludes two groups and states the reason for each. Care home residents live in a different social setting that the team plans to study separately, and people who lack the capacity to consent are excluded because the loneliness scale asks people to report their own feelings, and nobody else can report those for them.

**Sarah:** So the test is whether the reason is scientific or merely convenient.

**Kiffer:** That's a fair summary. A board will accept a stated reason tied to the research question much more readily than an exclusion that saves the team some data entry, and the team records the exclusion as a limitation.

**Sarah:** Section two moves to procedure. What's the first question a researcher should ask?

**Kiffer:** Whether the project needs review at all. The policy defines research as an undertaking intended to extend knowledge through a disciplined inquiry or systematic investigation. Human participants are people whose data, biological materials or responses are relevant to the research question. If you have both, you need approval from a research ethics board, which I'll call an REB, before the research begins, and recruitment counts as beginning.

**Sarah:** Are there activities that look like research but don't need review?

**Kiffer:** Chapter two lists several, each under defined conditions. Research that relies only on publicly available information, like published Statistics Canada tables, is one. Observation in public places is another, as long as people have no reasonable expectation of privacy and nobody is identified. Secondary use of anonymous information is outside review when nothing in the process creates identifiable information. Quality improvement and program evaluation used only inside the organization fall outside review, and so does an artist's creative process.

**Sarah:** Quality improvement sounds like a grey zone.

**Kiffer:** It is. A clinic auditing its own referral times doesn't need review, and if it later wants to publish the audit as research, it should ask the REB first. My general advice is to avoid deciding on your own that something falls outside review, and to ask the ethics office for a written decision.

**Sarah:** Then risk. How does the lesson define it?

**Kiffer:** A harm is anything that has a negative effect on a participant's welfare. It can be physical, psychological, social, economic or legal, and in survey and interview research many harms are informational, which means something a participant shared ends up with someone who shouldn't have it. Risk combines how likely a harm is with how serious it would be.

**Sarah:** And the threshold everyone talks about is minimal risk.

**Kiffer:** The policy defines minimal risk as research in which the probability and magnitude of possible harms are no greater than those participants meet in the parts of their everyday life that relate to the research. Asking an older adult how often they feel left out is close to what a family doctor might ask, so a short loneliness questionnaire is usually minimal risk. Detailed questions about suicidal thoughts go further than everyday life, and the risk may be above minimal.

**Sarah:** The lesson includes a risk table for Cedar Valley.

**Kiffer:** It lists each component with its harms and the steps that reduce them, which the policy calls mitigation. Every survey comes with a resource sheet listing the provincial health line, HealthLink BC, at eight one one, and the bc two one one community services line at two one one. The linkage is optional and analyzed only in a secure research environment, and interviews have a distress protocol.

**Sarah:** How does risk connect to the level of review?

**Kiffer:** Through what the policy calls a proportionate approach, where the higher the risk, the closer the review. Research above minimal risk gets full REB review at a convened meeting of the board. Minimal-risk research can get delegated review, where one or more board members or other qualified reviewers assess the application and can approve it, ask for changes, or send it to the full board. The researcher can propose a level, and the REB decides.

**Sarah:** And after approval, is the REB finished with you?

**Kiffer:** Approval starts what the policy calls continuing review. Any change to the study, even a new survey question, needs approval before you use it. Unexpected problems are reported promptly, and at a minimum you file an annual status report for projects longer than a year and an end-of-study report.

**Sarah:** Let's talk about Simon Fraser University specifically.

**Kiffer:** I'll keep this general, because forms and procedures change. At Simon Fraser, the Office of Research Ethics administers the review of research involving human participants, and the university's Research Ethics Board makes the decisions. Applications are submitted online, and students doing research for their degree normally apply with their faculty supervisor. The office's website has the current forms and guidance, and students should check it when they apply.

**Sarah:** What does an application ask for?

**Kiffer:** Forms differ between institutions, but you always write a lay summary and describe recruitment, risks and benefits, the consent process, privacy and data management, community engagement where it applies, and conflicts of interest. Then you attach the questionnaires, interview guides, recruitment materials and consent forms.

**Sarah:** And ethics approval is only one approval.

**Kiffer:** Often it's one of several. A study in a health authority's facilities usually needs the health authority's operational approval. A study with a First Nation may need the Nation's approval, often set out in a research agreement. Administrative health data need approval from the data stewards, which Lesson nine covers, and clinics that open their charts sign their own agreements.

**Sarah:** How did the Cedar Valley application go?

**Kiffer:** In our illustrative version, the REB reviewed it at a full board meeting, because it combined linked health data, a request to use charts without consent and a First Nations partnership. Another board might reasonably have used delegated review, because each component was judged minimal risk. The board approved the study on condition of six changes.

**Sarah:** What were they?

**Kiffer:** The board wanted the linkage section of the consent form in plain language. It asked that physicians stop approaching their own patients, so reception staff now hand out study cards. It asked for a distress protocol, and for a justification of the chart review against each condition in the policy. It wanted to know where data would be stored and who holds the key linking study numbers to names. It also wanted evidence of engagement with the First Nations health centre.

**Sarah:** That sounds like a lot of revision.

**Kiffer:** It's normal. Reply in a memo that quotes each comment and says what you changed. Keep your numbers identical across the application, the protocol and the consent form, because inconsistencies are one of the most common reasons for a request for revision.

**Sarah:** Section three is consent, and the policy uses three words for it.

**Kiffer:** Free, informed and ongoing. The policy treats consent as a process that starts at the first contact and continues until the person's involvement ends, and a signature records one moment in that process.

**Sarah:** What threatens consent being free?

**Kiffer:** The main threat is undue influence, which happens when someone with authority over a person does the recruiting, like their physician, instructor or employer, and the person feels unable to say no. Coercion is the extreme form, a threat of harm or punishment for refusing. A modest payment that recognizes people's time is acceptable, while a payment large enough to lead someone to accept risks they would otherwise refuse is a problem. Cedar Valley sends a small grocery gift card, and people keep it even if they stop early.

**Sarah:** Then informed.

**Kiffer:** Article three point two of the policy lists what people must be told. It covers the purpose, procedures, foreseeable risks and benefits, the right to withdraw and any limits on removing data, privacy protections, conflicts of interest, payments, how results will be shared, and contacts both on the team and outside it for ethical concerns.

**Sarah:** Is telling people enough?

**Kiffer:** Telling them is half the job, because they also have to understand. One technique I like is teach-back, where you ask the person to explain the study in their own words. If someone can't tell you what happens when they stop partway through, you haven't finished explaining.

**Sarah:** The lesson also mentions the therapeutic misconception.

**Kiffer:** That's the belief that a study is designed to help the participant personally, first described in psychiatric research in the early nineteen eighties. Some Cedar Valley participants might assume that reporting loneliness on a survey will connect them to services, so the form says plainly that it won't and lists resources separately. Consent is also ongoing, which means sharing new information that might change someone's mind and asking again at new stages, like before a second interview.

**Sarah:** Does consent always mean a signature?

**Kiffer:** The policy allows consent to be recorded in a signed form or by other appropriate means. Signed consent suits interviews and anything with identifiers. Oral consent, recorded in a log, suits telephone interviews and situations where a signature would add risk. Implied consent suits anonymous questionnaires, where returning the survey after reading an information letter shows agreement. Electronic consent works on survey platforms like REDCap.

**Sarah:** The Cedar Valley survey uses implied consent, but some parts need more.

**Kiffer:** That's where tiered consent comes in. The survey itself collects no names, so returning it is enough. The two optional parts, linkage and future contact for an interview, need identifiers, so each has its own yes and no boxes and a signature. Saying no to either one doesn't exclude anyone from the survey.

**Sarah:** And the documents have to be readable.

**Kiffer:** A good form uses short sentences, questions as headings, and the word you instead of the participant. Many REBs suggest about a grade six to eight reading level. The original Cedar Valley draft said that survey responses would be linked with administrative health data. The revision says that, with your permission, the team will connect your answers to records of your doctor visits, hospital stays and emergency department visits. It's longer, and far more people understand it.

**Sarah:** Capacity is next. What does the policy mean by it?

**Kiffer:** Capacity is the ability to understand the relevant information about a study and to appreciate the consequences of deciding to take part or not. It's specific to the decision, so someone might be able to consent to a short survey and be unable to weigh a complex trial. It can also change over time. Adults are presumed to have capacity unless there's reason to think otherwise.

**Sarah:** And when someone lacks capacity?

**Kiffer:** The policy allows research only under conditions. They include that the question can be answered only with people from that group, that consent comes from an authorized third party who isn't on the research team, and that research with no prospect of direct benefit stays at minimal risk. Provincial law decides who the authorized third party is. If the person can understand something of the study, you ask for their assent, and if they dissent, they don't take part.

**Sarah:** What does Cedar Valley do?

**Kiffer:** The interviews may include people with mild cognitive impairment who can consent for themselves. Before each interview, the interviewer asks three teach-back questions about the purpose, the right to stop and the recording. If the person can't answer after a second explanation, the interviewer ends the session kindly and uses none of the data.

**Sarah:** The last part of the consent section is about changing consent requirements. That sounds alarming.

**Kiffer:** It's meant to be rare. Imagine a study of how pharmacists counsel customers, where the customers are actually researchers. Article three point seven A lets an REB approve partial disclosure or deception like that only when it documents that every condition holds. The research has to be minimal risk, and the change has to be unlikely to harm participants' welfare. It has to be impossible or impracticable to do the research with consent in advance, the change has to be precisely defined, and there has to be a suitable debriefing plan.

**Sarah:** Impracticable is doing a lot of work there.

**Kiffer:** It is. The policy defines it as a degree of hardship that jeopardizes the research, so inconvenience doesn't qualify.

**Sarah:** The section ends with a full consent form.

**Kiffer:** It does, and it's the part of the lesson I'd most like students to spend time with. It's the revised form for the Cedar Valley survey, with a note beside every part explaining which requirement it meets. It opens with key information: this is research, taking part is voluntary, saying no won't affect your care, and the optional parts are separate. It also says the health authority will never see individual answers.

**Sarah:** Why include that last point?

**Kiffer:** Because older adults may worry that their care provider will read what they say about loneliness or their health. Answering it directly makes consent more genuinely free. The form is also honest that people probably won't benefit directly, and it explains that answers given without a name can't be found once submitted, while people who gave their names can withdraw until a set date.

**Sarah:** Section four is about information. Start with the vocabulary.

**Kiffer:** There are three terms to keep apart. Privacy is a person's right to be free from intrusion, including some control over information about themselves. Confidentiality is the researcher's duty to safeguard information that people entrust to them. Security is the set of measures that make that duty real, like locked cabinets, access rules, passwords and encryption.

**Sarah:** Can a researcher promise complete confidentiality?

**Kiffer:** Usually they can't, and the form has to say so. In British Columbia, anyone who believes a child needs protection has to report it. In focus groups, the team can ask members to keep what they hear private, and it still can't guarantee that they will.

**Sarah:** The lesson then sorts information by how identifiable it is.

**Kiffer:** The policy describes five categories. Directly identifying information includes names and health numbers. Indirectly identifying information can identify someone through a combination of details, like date of birth and place of residence. Coded information has had identifiers replaced with a code, and someone holds the key. Anonymized information has had identifiers removed for good, with no key kept and a low risk of re-identification. Anonymous information never had identifiers at all.

**Sarah:** And the small-community problem?

**Kiffer:** That's where rural research gets difficult. In Cedar City, a seventy-eight-year-old woman who lives alone is one of many. In Kestrel Lake, a woman over ninety who lives alone and belongs to a First Nation may be the only person who fits, so someone who knows her might recognize her answers even without a name. The team reports age in bands, combines small towns into sub-regions, and never reports a number smaller than five.

**Sarah:** Then secondary use.

**Kiffer:** Secondary use means using information first collected for another purpose, like clinical care or an earlier study. If it's identifiable, you need either consent or the REB's approval to go ahead without consent. That approval depends on six conditions in Article five point five A. Identifiable information has to be essential, and using it without consent has to be unlikely to harm people's welfare. Privacy has to be protected, and preferences people have already expressed have to be respected. Seeking consent has to be impossible or impracticable, and any other permissions have to be in place.

**Sarah:** How does the Cedar Valley chart review meet them?

**Kiffer:** Trained abstractors review three hundred charts, fifty at each of six clinics. They have to open identifiable charts to read the clinical notes, but their forms carry only a random number, and nothing identifiable leaves the clinic. Patients aren't contacted, charts of people who've declined research use are skipped, and each clinic signs a data access agreement.

**Sarah:** And the impracticable condition?

**Kiffer:** That's the one boards examine hardest. The team's argument is that many older patients in the sample will have died, moved or become too unwell to reply, and that a sample of only those who consent would over-represent healthier patients and bias the results. In our illustrative case, the board accepted that, with one more safeguard: an abstractor who recognizes a patient they know personally skips that chart.

**Sarah:** And the linkage to health records works differently.

**Kiffer:** Yes, because it rests on consent. People who ticked yes in the optional part of the form agreed to it, so the six conditions don't apply. It still needs REB approval, because any linkage that could produce identifiable information does.

**Sarah:** Data management plans come next. Students might think of that as paperwork.

**Kiffer:** I think of it as the document that shows you've planned for the whole life of your data, from collection and storage to sharing, preservation and destruction. The federal agencies' Research Data Management Policy, from twenty twenty-one, requires these plans for some funding opportunities. A free online tool called the DMP Assistant, where DMP stands for data management plan, walks you through the standard questions. The Cedar Valley plan says that only one team member holds the key and sets retention at five years after the last publication.

**Sarah:** Which brings us to Chapter nine.

**Kiffer:** Chapter nine covers research involving First Nations, Inuit and Métis peoples, and it responds to the kind of history we started with, including the nutrition experiments and the Nuu-chah-nulth samples. Its central requirement is that when research is likely to affect the welfare of an Indigenous community, researchers have to seek engagement with that community.

**Sarah:** How do you know when that applies?

**Kiffer:** The chapter names situations. They include research 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 that refers to Indigenous communities or peoples. Cedar Valley recruits in First Nations communities and plans to analyze results by First Nations identity, so the chapter applies.

**Sarah:** What does engagement look like in practice?

**Kiffer:** It's often written into a research agreement. In Cedar Valley's illustrative agreement, the health centre has a seat on the steering committee, and analyses by First Nations identity need its approval. It reviews findings before publication and receives its own community report.

**Sarah:** Does community approval mean individuals don't need to consent?

**Kiffer:** Never. Each participant still decides for themselves. Community engagement adds respect at the level of the community, and it never replaces individual consent. Lesson four introduced the principles of ownership, control, access and possession, known as OCAP, and showed how to write an engagement plan. Chapter nine is where that plan meets ethics review.

**Sarah:** Before we finish, where should someone go next if they expect to work on a study involving people?

**Kiffer:** To the Course on Research Ethics, the free online tutorial known as CORE twenty twenty-two. Research ethics boards commonly ask team members to complete it, and its modules follow the order of this lesson's sections.

**Sarah:** And what makes a consent form work well?

**Kiffer:** The Cedar Valley form is the model. It covers every topic in the disclosure table, uses tiered consent for each optional part, explains the limits of confidentiality and withdrawal, and puts placeholders in square brackets where contact details would go, because the study is fictional.

**Sarah:** Any advice for testing a draft?

**Kiffer:** Give it to people from the participant population and ask them to explain the study back to you, then rewrite wherever they stumble. The Cedar Valley advisory group did exactly that. And when a study involves a First Nations, Inuit or Métis community, the form names a community contact, as the Cedar Valley form names the health centre's research liaison.

**Sarah:** And the form is only final once the REB approves it?

**Kiffer:** That's correct. The REB reviews the form with the rest of the application, and the approved version, with its version number and date, is the one participants see. Any later change to the form needs approval before it is used.

**Sarah:** Thanks, Kiffer.

**Kiffer:** Thanks, Sarah. Next week we move to Lesson six, on choosing an approach and managing a research project.