Ethical Issues in Clinical Trials and Protocol Design
Have you ever found yourself in a situation where you’d like to confirm if your trial design ethically sound?
In this article, I’ll share with you the ethical questions you should be asking yourself when planning and conducting a clinical trial.
You’ll commonly hear people say:
- Why is the trial not moving forward faster?
- It’s been months and we still have not finalized the protocol
- Why is site activation slow?
Any delay or slowdown is directly or indirectly related to ethics in research.
The matter of ethics is of utmost importance in the field of clinical research and clinical trial management.
Someone somewhere is concerned about how the informed consent template is worded or how the protocol is designed. These are all ethical issues.
Now let’s uncover these ethical considerations together in this two-part article.
In the first part, we’ll focus on ethical issues in clinical trials.
In the second part, we’ll cover specific ethical topics as they relate to protocol design.
Part I: Ethical Issues in Clinical Trials
The questions below will serve as an ethical framework when you are confronted with ethical issues or concerns as they relate to your clinical trial.
1. Is there a need for a clinical trial?
Yes, for new medical products, it is generally necessary and required to conduct a clinical trial. However, you should always ask the question, “Is there a need for a clinical trial?”
It may be entirely possible for you to gather the clinical data via other means such as retrospective data collection i.e. review of medical records of patients who have already undergone treatment, discussion with medical experts or literature search.
There is one other key issue with starting a clinical trial when you don’t need one.
A Sponsor can get into legal issues with the government if they start a trial to boost their medical product sales under the guise of a clinical trial.
For instance, if you pay a clinical trial site $5000 per patient enrollment. However, on the backend, you are charging the site $10,000 to buy your medical product. This means you are profiting from your clinical trial.
The goal of conducting a clinical trial must be a scientific one and must not be guided by commercial interests.
2. Is it ethical to include a placebo treatment group?
When there is no alternative treatment available to the patient, it may be acceptable to have a placebo control arm.
However, when alternative treatment is available, having a placebo control may not be viewed as ethical by certain physicians, patients or regulatory agencies unless the available treatment is highly toxic.
Another ethical consideration is whether the patient is relatively healthy or extremely sick.
According to the FDA guidance, E10 Choice of Control Group and Related Issues in Clinical Trials, short term placebo-controlled trial is generally acceptable in a trial where the patients are not experiencing severe health problems.
Long term placebo-controlled trials would not acceptable in patients in sicker patients with severe health problems.
3. Is your choice of an active control acceptable?
In clinical trials with a control arm, the choice of the control device or drug matters.
For example, let’s say a medical device sponsor is comparing the safety and effectiveness of their marketed device with the next-generation device.
If there are unanticipated adverse events in the marketed device, this would have negative financial consequences for the Sponsor.
The Sponsor may need to recall the marketed device and stop or hold future sales.
For this reason, Sponsors may not conduct clinical trials of their new medical products against already marketed and commercialized products.
4. Has the clinical trial eliminated obvious bias and deception?
It is easy to introduce bias in the trial.
Bias can take many forms.
Bias is best illustrated using examples of operational bias, scientific bias, data collection and cleaning bias.
Depending on the nature of the bias the validity of your clinical trial data may be questioned.
- Operational bias
You are in the process of selecting sites for a new surgical clinical trial.
One of the key questions in the site assessment is the number of surgical procedures performed at the potential clinical trial site.
You set a threshold of 400 procedures per year. However, there is one site that your senior management really wants you to include in the trial.
But this senior management recommended site only performs 20 procedures per year, which is significantly less than your threshold of 400 procedures per year.
Would you select this site?
If you choose to select the site with less surgical procedure experience, you are introducing operational bias in your trial.
- Scientific or Statistical bias
As a sponsor, your statistical programmer generates a table for your final trial report that lists how many patients have missed visits by the site.
However, he does not validate the programming (software code) of the table.
Now the unvalidated table gets sent to the FDA.
You later find out, through a round of FDA questions, that your report has erroneous data and fails to capture several missed visits.
Such matters will make the FDA question the validity of your entire dataset and delay your product approval.
- Data collection and cleaning bias
You are a research coordinator responsible for ensuring patients return for their follow-up visits.
Some patients will not show up for their follow-up visits.
You have an established process of following up with such patients three times via email, phone and/or certified letter before you consider the patient as “lost to follow-up.”
However, halfway through the trial, your research coordinator workload has increased significantly.
She no longer has the bandwidth of following up with patients three times as per your process.
She is only able to follow up once and if she does not hear back from the patient, the patient is marked as “lost to follow-up”
By changing your data collection process halfway through the trial, you’ve introduced bias in the trial.
These are just some examples to illustrate bias. You can have bias throughout the trial from site start-up to close-out. Your job should be to reduce bias as much as possible.
5. Has sufficient attention been given to sample size?
When designing a clinical trial, it would be unethical to guess the total number of patients you need to enroll.
Your enrollment calculations are based on assumptions such as the prevalence of the disease, previous clinical data, regulatory requirements, and medical products.
It is important to come up with a large enough sample size such that the trial results can be replicated in the real world. The goal is to enroll just the right number of patients to make your trial results relevant to the larger target population.
If your sample size is too small, you run the ethical risk of enrolling patients in a trial where the results will have little or no scientific value.
You’re also like to have false positives (a result which incorrectly indicates that a particular condition is present) or false negative (a result which incorrectly indicates that a particular condition is absent).
6. Is the statistical power of a trial adequate to demonstrate efficacy i.e. medical treatment effective?
If you’re comparing a new treatment versus an existing standard of care, you want to have the greatest probability that there is a difference between the two treatments. This is a known “power”.
Generally, a power of 80% is sufficient to detect treatment differences. Anything less than 80% power could mean that your trial is not sufficiently powered.
Enrolling patients in an underpowered trial is considered to be unethical, especially for industry-sponsored research or a multi-center approval trial.
7. Are the chances of the patient receiving the active medical product acceptable?
For randomized control studies with a placebo control, the chances of a patient receiving the active drug or device should be equal to or greater than 50%.
For example, if a patient is in pain and her chances of receiving the medicine are 20% (one in five enrolled patients), she is less likely to consent to participate in the trial.
8. What is the safety of the patients enrolled in the clinical trial?
It is important to ensure that the healthy volunteers and patients are safely enrolled and managed during the course of a clinical trial.
Let’s take my personal experience participating in a clinical trial as a healthy volunteer.
The protocol required the doctor has to perform a fat biopsy. Unfortunately, my bleeding would not stop after the fat biopsy.
The research team did the following to ensure I was safe:
- The doctor would not discharge me until the bleeding has stopped
- The nurse provided me with medical supplies to control the bleeding, should it start again after I left the clinic
- Two days after my visit, research coordinator followed up with me via email to confirm if the bleeding had stopped
For this research team, patient safety was their top priority at all times.
When designing any research protocol, you want to ensure there are provisions in place to ensure patient safety at all times.
9. What types of patients are to be entered into a clinical trial?
Every trial must have a clearly outlined inclusion/ exclusion criteria.
For example, it may be unethical to enroll healthy volunteers in a trial when it is known that healthy patients will not tolerate the trial medication.
Or in the case of a medical device trial, you cannot enroll a healthy patient in a heart stent trial when the patient does not have heart disease required to qualify for the trial.
Great caution must be taken when enrolling infants, children, pregnant women, lactating mothers, prisoners, racial or ethnic minorities, elderly, socioeconomically disadvantaged, underinsured or those with certain medical conditions.
Such patients are generally unable to protect their own rights.
These types of individuals would not be able to consent freely and it would be unethical to enroll them in trials where their rights are compromised.
10. What should patients be paid for their participation?
There has been confusion among clinical research professionals about whether and how much should research participants be paid for participating in clinical research.
As a research professional, your goal must be to ensure that payments to trial patients do not lead to undue influence or coercion. You can learn more about this from Dr. Lindsay McNair’s podcast episode on Bioethics in Clinical Research.
If you wish to explore how to best answer this question for your research, consider reading, Paying Research Participants: Regulatory Uncertainty, Conceptual Confusion, and a Path Forward
Part II: Ethical Considerations in Protocol Design
Once you have addressed the above ethical questions, you’ll be confronted with a different set of considerations during protocol design.
In this second part, we’ll cover the most important sections of a clinical protocol (also known as the investigational plan) where you need to pay close attention to from research ethics point of view.
1. Inclusion/ Exclusion Criteria
Vulnerable Populations
According to FDA ICH-GCP E6(R2) guidelines, vulnerable population is defined as “Individuals whose willingness to volunteer in a clinical trial may be unduly influenced by the expectation, whether justified or not, of benefits associated with participation, or of a retaliatory response from senior members of a hierarchy in case of refusal to participate.
Examples are members of a group with a hierarchical structure, such as medical, pharmacy, dental, and nursing students, subordinate hospital and laboratory personnel, employees of the pharmaceutical industry, members of the armed forces, and persons kept in detention. Other vulnerable subjects include patients with incurable diseases, persons in nursing homes, unemployed or impoverished persons, patients in emergency situations, ethnic minority groups, homeless persons, nomads, refugees, minors, and those incapable of giving consent.”
When you define inclusion/ exclusion criteria for a trial, it is extremely important to ensure the rights of the vulnerable population is protected.
If possible, it may be best to exclude vulnerable patients from your trial.
Patient Age
It is important to carefully choose the age of potential trial patients. Depending on the medical product and it’s intended use, you will need to make a decision of the age range for potential research patients.
Children (minors) are considered a vulnerable population as noted above. For this reason, it would be necessary to have additional protections when enrolling children in a clinical trial.
Similarly, it is necessary to ensure the rights of older patients, who may not be in their best of health, remaining protected during consenting and trial conduct.
Pregnant and Lactating Women
The efficacy and safety of most pharmaceutical products in clinical trials remains unknown in pregnant and lactating women.
We don’t know whether a test drug will have a harmful effect on the fetus.
Similarly, we don’t know if a lactating mother taking the test drug can cause a safety risk to the child who is dependent on the mother’s milk.
For this reason, most studies will and should exclude pregnant and lactating mothers.
It is also common for trial patients to undergo pregnancy tests prior to and during the trial to confirm they are not pregnant. If a woman is pregnant, her participation in the trial could immediately be discontinued.
Furthermore, researchers will need to monitor the mother during her entire pregnancy for safety reasons.
2. Safety and Efficacy Endpoints
When making a decision on safety and efficacy endpoints, you want to develop trial endpoints based on a comprehensive literature search including national guidelines and previous clinical studies in the specific therapeutic area and disease.
The goal of innovative medical products may be to change the medical guidelines.
In such cases, patient safety must be prioritized keeping in mind the current medical guidelines and “gold standard” treatment for the target patient population.
3. Staff credentials and training
To ensure patients are ethically treated in a clinical trial, it is necessary for medical personnel to be adequately trained on the trial protocol and any trial-specific procedures.
Physicians and other medical professionals need to have an active medical license to provide medical care for the patient.
Furthermore, individuals performing data collection and analysis must also have the relevant experience (or oversight) and training in clinical trial management.
4. Schedule of Events
Schedule of Events is a section of the protocol, usually in tabular format, that outlines the trial visits and the requirements for each visit.
When designing a Schedule of Events it is necessary to map out how the trial will unfold in a clinical setting once a potential patient is identified for the trial.
Let’s say, in one heart failure trial, patients were asked to walk for six minutes without any assistance immediately following their medical treatment.
Given the heart-failure patients were still recovering from their treatment, it was extremely difficult for them to walk for six minutes without any assistance from a caregiver.
Such requirements generally lead to multiple protocol deviations.
For a drug trial, this could mean that any adjustments to the dosing schedule must be documented in the Schedule of Events.
It would be unethical for a trial design to require a patient who has experienced adverse events to continue on a fixed dosing schedule.
5. Medication Compliance
It is not uncommon for trial patients to be sent home and asked to take medications as per trial protocol.
However, medication compliance can vary greatly among patients.
Some patients may be disciplined and follow the doctor’s instructions. Others may forget to take their medication or choose to ignore the doctor’s instructions.
In order to ensure medication compliance, clinical trials may have monitoring systems to randomly check patient adherence to the trial protocol.
This type of monitoring can have an impact on patient privacy. For this reason, it is important to include any such monitoring systems in the informed consent form.
6. Lost to Follow-up
Is it acceptable to contact the patient for additional information related to their participation in the trial after he or she is considered “lost to follow-up”?
Protocols usually include a definition for patients lost to follow-up.
For example, if a patient misses two consecutive visits, the patient may be considered lost to follow-up. The ethical question is whether the patient can be contacted after they have been flagged as lost to follow-up.
Also, data collection can be inconsistent across sites if the protocol lacks a clear definition of lost to follow-up.
For example, sites may be required to make three attempts to get in touch with the patient before her or she is considered lost to follow-up.
If all participating sites are not trained on follow-up requirements, patient follow-up compliance rates can vary greatly amongst sites.
7. Patient Discontinuation
When a patient experience a serious adverse event, the patient may choose to discontinue from the clinical trial.
However, it is the site investigator’s responsibility to ensure the patient is safe and receives the necessary medical care after such discontinuation.
In some situations, the patient may decide to join the trial again after discontinuation. If this happens, it would be necessary to re-consent the patient.
8. Early Trial Termination
Some clinical trials may end earlier than expected.
Reasons for stopping studies early include positive or negative results, extremely slow enrollment, protocol non-compliance, and research staff changes.
If positive results are discovered during an interim analysis of a randomized control trial, it may become ethically necessary to provide control patients with the new device or drug.
If negative results such as higher than anticipated rate of serious adverse events can lead the trial being terminated early.
Some studies are extremely slow to enroll due to extremely tight inclusion and exclusion criteria or lack of commitment from the research staff.
For slow enrolling trials, it may be necessary to terminate the trial early and transfer enrolled patients to another nearby site for their safety.
If major audit findings such as protocol non-compliance are discovered, the regulatory agency or Sponsor may decide to stop the trial early.
All these scenarios require careful planning for successful protocol design and trial conduct.
9. Medical Emergency
In the case of a medical emergency, the site must be able to contact the Sponsor safety representative to discuss the adverse reactions and potential treatment.
In double-blinded protocols where the patient and the investigator do not know the treatment arm, a process must be established on how the blind will be broken in case of an emergency.
10. Amount and Type of Data Collection
Once enrollment commences, clinical data needs to be collected and entered in the case report forms.
If the amount of clinical data being collected is vast, researchers may end up with a partially complete dataset at the time of analysis.
Missing or incomplete can pose a serious issue to provide the scientific value of the trial.
Conversely, if the trial fails to collect the required critical fields, the scientific community will find it difficult to accept or interpret the trial results.
11. Operational Instructions
Operational instructions to patients such as consent form and procedure brochures must be reviewed and approved by the ethics committee.
Patients must also be provided clear instructions on whom to contact in case of an emergency or follow-up questions after their visit.
Aside from patient instructions, sites must have access to Sponsor name and contact information when dealing with a medical emergency.
12. Continuation Protocol
There may be a lag between trial completion and regulatory review of the clinical data. During this period, the Sponsor may decide to implement a continuation protocol.
One objective of a continuation protocol is to ensure that the patients have access to the trial drug after trial completion.
For device trials, the Sponsor may want to collect additional safety and efficacy data while the trial data is under regulatory review.
For control patients, the question arises whether they should be receiving the trial drug or device, while the regulatory review is underway.
This sought of ethical concern needs to be clearly thought through when designing a continuation protocol.
13. Committees in Multicenter Trials
For multi-center studies, centralized committees are generally ongoing clinical data monitoring.
For example, certain medical device clinical studies may be required to appoint a Data Safety and Monitoring Board for patient safety.
During the course of the trial, if this committee is concerned about the data being collected, it has the authority to ask the Sponsor to stop or pause the trial due to ethical and patient safety concerns.
14. Data Confidentiality
Due to privacy laws, it goes without saying that Personal Health Information (PHI) must remain protected at all times and de-identified as necessary.
Hence, the protocol and patient consent form must clearly outline how patient confidentiality will be ensured during the trial conduct.
Clinical trial data is generally presented as an aggregate and the identity of no individual patient is ever revealed.
15. Publications
Once the clinical trial is complete and the data is analyzed, it is ethically necessary to publish the trial results, irrespective of whether the trial outcome is favorable or not.
For large multicenter trials, individual sites, in most cases, will not be able to publish data on their own without Sponsor approval.
Furthermore, individual sites may need to wait until the entire trial is complete and the results are published before publishing results based on their site-specific partial dataset.
Including a high-level summary of the publication plan within the protocol can help Sponsor personnel and clinical trial sites stay aligned on when and how trial results will be published.
Summary – Here’s What’ve Covered on Clinical Research Ethics
In part 1, we covered the following ethical questions you need to ask when developing a clinical trial strategy.
- Is there a need for a clinical trial?
- Is it ethical to include a placebo treatment group?
- Is your choice of an active control acceptable?
- Has the clinical trial eliminated obvious bias and deception?
- Has sufficient attention been given to sample size?
- Is the statistical power of a trial adequate to demonstrate efficacy i.e. medical treatment effective?
- What is the safety of the patients enrolled in the clinical trial?
- What is the safety of the patients enrolled in the clinical trial?
- What types of patients are to be entered into a clinical trial?
- What should patients be paid for their participation?
In part 2, we focused on the following protocol sections that require us to stop and think about research ethics.
- Inclusion/ Exclusion Criteria
- Safety and Efficacy Endpoints
- Staff credentials and training
- Schedule of Events
- Medication Compliance
- Lost to Follow-up
- Patient Discontinuation
- Early Trial Termination
- Medical Emergency
- Amount and Type of Data Collection
- Operational Instructions
- Continuation Protocol
- Committees in Multicenter Trials
- Data Confidentiality
- Publications
As a general rule of thumb, always remember what matters most to regulatory authorities such as the FDA and ethics committees (also known as Institutional Review Board) are patient safety, patient privacy and whether benefits to patients outweigh the risks.
When dealing with ethical considerations, your goal should be to keep the regulatory perspective in mind at all times.
After reading this article, if you a question or comment on ethics in research, please let me know by leaving a comment below.
