Understanding Control Groups in Clinical Trials: Types and Selection

A control group is a set of participants in a clinical trial who do not receive the experimental intervention under investigation, serving as a benchmark against which the effects of that intervention can be measured. The purpose of a control group in an experiment is to isolate the true effect of the treatment from other factors such as natural disease progression, participant expectations, and the mere passage of time. This article explains the main types of control groups used in clinical research, including placebo, active, historical, waitlist, and attention controls, and provides a practical decision framework for selecting the appropriate control based on trial objectives, ethical constraints, and available data. The content is written for students, researchers, life-science professionals, and informed general readers who need to design, interpret, or evaluate clinical studies.

At a Glance: Control Group Types and Selection Criteria

The table below summarizes the primary control group types, their defining features, typical use cases, and key limitations. Use this table as a quick reference when planning a trial design.

The Core Purpose of a Control Group in Clinical Research

A control in a study provides the counterfactual, that is, what would happen to participants if they did not receive the experimental intervention. Without a control group, a researcher cannot determine whether an observed improvement is caused by the treatment or by other mechanisms. These mechanisms include the natural history of the condition, regression to the mean, the placebo effect, and the effects of being observed and cared for by health professionals.

Implementation science, which focuses on maximizing the adoption and appropriate use of effective clinical practices in real-world settings, often relies on randomized controlled trials to answer questions about intervention effectiveness. However, some implementation questions are better suited to quasi-experimental designs that estimate intervention effects without randomization, such as pre-post designs with a non-equivalent control group, interrupted time series, and stepped-wedge designs where all participants eventually receive the intervention in a staggered fashion. The choice between experimental and quasi-experimental designs depends on feasibility, ethical constraints, and the specific research question being asked.

The control group in a clinical trial serves several distinct functions. First, it provides a baseline for measuring the magnitude of the treatment effect. Second, it helps control for confounding variables that could otherwise explain observed differences. Third, it allows researchers to assess the safety profile of the intervention by comparing adverse event rates between groups. Fourth, it enables blinding, which distributes expectancy effects equally between treatment arms in theory.

Placebo Control Groups

A placebo control group receives an inert substance or sham procedure that is indistinguishable from the active intervention. The purpose of the placebo is to control for the psychological and physiological effects of receiving a treatment, including participant expectations of improvement. In a double-blind placebo-controlled trial, neither the participant nor the investigator knows which treatment is being administered, which helps ensure that outcomes are assessed objectively.

The distinction between having a placebo control group and conducting a genuinely placebo-controlled trial is important. A trial may formally include a placebo control group, but if blinding fails, participants may correctly guess their assignment, leading to an uneven distribution of expectancy effects between groups. This phenomenon, called activated expectancy bias, can inflate estimates of treatment effects and create false positive findings. Computational modeling has shown that weak blinding combined with positive treatment expectancy can produce this bias, and statistical tools such as the Correct Guess Rate Curve have been developed to estimate what the outcome of a perfectly blinded trial would be based on data from an imperfectly blinded trial.

Placebo controls are most appropriate when no effective treatment exists for the condition under study, when the condition is mild or self-limiting, or when the risk of harm from withholding treatment is minimal. In vaccine trials, the context and timing of placebo use matter considerably. During a pandemic or outbreak, for example, the ethical acceptability of a placebo control may change as effective vaccines become available and as the background risk of infection in the population shifts.

The use of placebo control groups raises specific ethical and practical problems, particularly in phase III trials where an established treatment already exists. In such cases, withholding effective treatment from placebo recipients may be considered unethical, and an active control design may be required instead.

Active Control Groups

An active control group receives an established, effective treatment instead of an inert substance. This design is used when a standard of care exists and the research question concerns whether a new intervention is at least as good as the current treatment, or whether it offers advantages in terms of safety, tolerability, cost, or convenience.

Active control trials can be designed to demonstrate superiority, where the new treatment is shown to be better than the active control, or non-inferiority, where the new treatment is shown to be no worse than the active control within a predefined margin. Non-inferiority trials typically require larger sample sizes than placebo-controlled superiority trials because the margin of equivalence must be estimated with precision.

The selection of the active comparator is a critical decision. The comparator should be the current standard of care, ideally one that has demonstrated efficacy in rigorous trials and is widely used in clinical practice. The dose, route of administration, and treatment schedule of the active control should match its approved or recommended use to ensure a fair comparison.

Evidence from randomized controlled trials suggests that complex intervention programs integrating multiple intervention strategies are not necessarily more effective than active control groups. This finding underscores the importance of carefully designing the control condition in behavioral and complex intervention trials, because a poorly chosen active control may either overestimate or underestimate the true effect of the experimental intervention.

In trials of physical activity interventions, such as yoga, matching the exercise volume in the active control group is essential. If the control group receives a different amount or intensity of physical activity than the intervention group, any observed differences could be attributed to the dose of exercise instead of to the specific components of the yoga intervention being tested.

Historical Control Groups

A historical control group uses data from patients who were treated in the past, instead of enrolling a concurrent control group. This approach is used when randomization is impractical or unethical, when the condition is rare, or when the intervention is a surgical procedure or device that cannot easily be blinded.

Historical controls are common in single-arm trials, where all enrolled participants receive the experimental intervention and their outcomes are compared to previously collected data from a similar patient population. For example, a prospective single-arm trial of indocyanine green administration for bile leak detection after hepatectomy planned to compare outcomes in 40 patients receiving the intervention to 45 historical controls who underwent surgery without the imaging agent. The primary endpoint was the bilirubin concentration in the drainage fluid on postoperative day three.

The main limitation of historical controls is the potential for bias due to changes over time. Medical care evolves, diagnostic criteria change, patient populations shift, and supportive care improves. These factors can make historical data incomparable to current outcomes, even when the same inclusion and exclusion criteria are applied. In a study defining an optimal historical control group for a phase 1 trial of mesenchymal stromal cell delivery during pediatric cardiac surgery, researchers found that postoperative scores improved in more recent years at their center despite no differences in surgical factors between era groups. This observation, consistent with improved postoperative care over time, illustrates the need to assess era effects when using historical controls.

When selecting a historical control group, researchers should apply the same inclusion and exclusion criteria used in the current trial, match the population on key prognostic variables, and assess whether outcomes have changed over the period covered by the historical data. The more recent the historical data, the more likely it is to be comparable to current outcomes.

Waitlist Control Groups

A waitlist control group consists of participants who are randomly assigned to receive the intervention after a delay, typically after the intervention group has completed treatment and follow-up assessments. During the waiting period, these participants serve as controls, allowing comparison of outcomes between those who received the intervention immediately and those who did not.

Waitlist controls are commonly used in trials of behavioral, educational, and internet-based interventions where a placebo is difficult to construct and where withholding the intervention entirely may be considered unethical. For example, a randomized controlled trial of an internet-based self-help intervention for procrastination enrolled 160 participants in the intervention group and 160 in a waitlist control group. The control group received access to the intervention after 12 weeks, and follow-up assessments were scheduled at 6 and 12 weeks after baseline.

The main advantage of the waitlist design is that all participants eventually receive the intervention, which can facilitate recruitment and address ethical concerns about denying treatment. However, waitlist controls have important limitations. Participants in the waitlist group may seek other forms of help while waiting, which can dilute the observed difference between groups. The waitlist design also cannot control for the effects of attention and expectation in the same way that a placebo or attention control can. Additionally, the waitlist period is typically short, limiting the ability to assess long-term outcomes.

Attention Control Groups

Attention control groups are designed to control for the benefits of interpersonal interaction and engagement that may come from behavioral interventions. Participants in an attention control group receive the same dose and format of contact as intervention participants, but the content of that contact does not include the active elements of the intervention.

In a randomized controlled trial of an aging-in-place intervention for older adults, the intervention group received goal-directed visits facilitated by an occupational therapist, nurse, and handyman, while the attention control group received visits from a lay person. The attention control participants most often chose conversation and playing games as visit activities, and the majority reported a great deal of perceived benefit from the visits. This finding demonstrates that attention control visits can be an appropriate comparison in studies of behavioral interventions for community-dwelling older adults.

The design of attention control activities requires careful consideration. The dose of contact, including the number and length of visits, should match the intervention group. The activities offered should be engaging enough to maintain participant interest and retention but should not include components that could plausibly produce the same effects as the intervention. Unfortunately, few publications provide detailed information about attention control activities, which limits the ability of other researchers to replicate or evaluate these designs.

No-Treatment and Usual Care Control Groups

A no-treatment control group receives no intervention at all and is followed over the same period as the intervention group. This design is used when a placebo is not feasible and when the natural course of the condition needs to be documented. No-treatment controls are common in educational research, where a sham intervention may be difficult to construct.

Usual care control groups receive whatever standard care is normally provided to patients with the condition, without any additional intervention from the research team. Usual care controls are frequently used in trials of physiotherapy, rehabilitation, and complex interventions where a placebo is impractical. However, the reporting of usual care in control groups is often poor. A systematic review of physiotherapy trials for multiple sclerosis found that interventions described as usual care were underdescribed compared to experimental treatments, affecting the validity, generalizability, and interpretability of the results. Researchers should document the components of usual care as thoroughly as they document the experimental intervention, using frameworks such as the Template for Intervention Description and Replication checklist.

Case-Control Studies and Non-Randomized Comparisons

A case-control study is an observational design that compares individuals with a condition to individuals without the condition, instead of a randomized trial with a control group. For example, a study compared the serum level of vascular endothelial growth factor in patients with active pulmonary tuberculosis to a control group of individuals without the disease. Case-control studies are useful for studying rare conditions and for generating hypotheses, but they are subject to selection bias and confounding because participants are not randomly assigned to groups.

In contrast, a control group in a randomized clinical trial is created through random assignment, which helps ensure that known and unknown confounding variables are distributed equally between groups. The distinction between observational comparisons and randomized controlled comparisons is fundamental to the interpretation of study results.

Decision Framework for Selecting a Control Group

The selection of a control group type should be guided by the trial objectives, the availability of effective treatments, ethical considerations, and practical constraints. The following steps provide a structured approach to this decision.

Step 1: Define the Primary Research Question

Determine whether the trial aims to demonstrate superiority of the intervention over no treatment, superiority over an existing treatment, or non-inferiority to an existing treatment. A placebo or no-treatment control is appropriate for superiority trials when no effective treatment exists. An active control is required for non-inferiority trials and for superiority trials where an effective standard of care exists.

Step 2: Assess Ethical Constraints

Review whether withholding treatment from control participants is ethically acceptable. If an effective treatment exists and the condition is serious, a placebo control may be unethical and an active control should be used. Consult relevant ethical guidelines and obtain approval from an institutional review board or research ethics committee.

Step 3: Evaluate Feasibility of Randomization

Consider whether randomization is practical and acceptable to participants and clinicians. In surgical trials, blinding and randomization may be difficult. In rare diseases, enrolling enough participants for a concurrent control group may not be feasible. In such cases, a historical control or single-arm design may be the only option, with the understanding that the evidence will be weaker than that from a randomized trial.

Step 4: Consider the Nature of the Intervention

For behavioral interventions, an attention control may be necessary to distinguish the specific effects of the intervention from the general benefits of attention and engagement. For drug trials, a placebo control is standard when no active comparator exists. For complex interventions, the control condition should be designed to match the intervention on non-specific elements such as contact time, format, and setting.

Step 5: Determine the Need for Blinding

Assess whether blinding is possible and whether it is likely to be successful. Blinding is easier for drug trials using identical placebo formulations than for behavioral or surgical interventions. If blinding is likely to fail, consider whether the resulting expectancy bias could threaten the validity of the trial and whether statistical adjustments or alternative designs are needed.

Step 6: Review Available Historical Data

If a historical control is being considered, evaluate the quality and comparability of available data. Apply the same inclusion and exclusion criteria, match on prognostic variables, and assess whether outcomes have changed over time. The more recent and complete the historical data, the more credible the comparison.

Step 7: Document the Control Condition Thoroughly

Regardless of the control group type selected, document the control condition in detail. Include the content, dose, format, and timing of any interventions or contacts received by control participants. This documentation supports replication, interpretation, and comparison across studies.

Statistical Considerations in Control Group Selection

The statistical analysis of a clinical trial depends on the type of control group used. In randomized trials, the primary analysis typically compares outcomes between the intervention and control groups using regression models that adjust for baseline characteristics and stratification factors. Including a baseline assessment of the primary outcome as a covariate can reduce variation and increase statistical power.

A more advanced approach involves using a super-covariate, which is a patient-specific prediction of the control group outcome derived from historical data from other studies in similar patients. This prediction is included as a covariate in the analysis, not as an offset. The super-covariate approach has the potential to increase the power of clinical trials by using historical data, while avoiding the type I error inflation concerns associated with some Bayesian approaches. However, the prognostic models behind super-covariates must generalize well across different patient populations to be useful.

The choice of control group also affects sample size calculations. Placebo-controlled superiority trials generally require smaller sample sizes than active-controlled non-inferiority trials because the expected difference between groups is larger. Trials using historical controls may require smaller sample sizes than trials with concurrent controls, but the savings in sample size come at the cost of increased vulnerability to bias.

Common Failure Patterns in Control Group Design

Several recurring problems undermine the validity of control groups in clinical trials. Recognizing these patterns can help researchers avoid them and help readers evaluate the quality of published studies.

Inadequate Description of the Control Condition

Control conditions are often described in far less detail than experimental interventions. This asymmetry makes it difficult to determine what participants in the control group actually received and whether the control condition was appropriate. The systematic review of usual care in physiotherapy trials for multiple sclerosis found that control treatments were underdescribed compared to experimental treatments, affecting the validity, generalizability, and interpretability of results.

Failure to Match Non-Specific Elements

In behavioral and complex intervention trials, the control group may receive less attention, contact time, or engagement than the intervention group. Any observed difference between groups could then be attributed to these non-specific elements instead of to the specific components of the intervention. Attention control groups are designed to address this problem, but they require careful matching of dose and format.

Blinding Failure and Expectancy Bias

Blinding can fail when the intervention has noticeable effects or side effects that reveal group assignment. When blinding fails, expectancy effects may be distributed unevenly between groups, inflating estimates of treatment effects and creating false positive findings. Researchers should assess blinding integrity and consider statistical tools to estimate the outcome of a perfectly blinded trial.

Era Effects in Historical Controls

Historical control data may not be comparable to current outcomes because of changes in medical care, diagnostic criteria, and patient populations. Researchers should assess era effects by comparing outcomes across time periods within the historical data and should use the most recent data available.

Contamination and Co-Intervention

Participants in control groups may receive the experimental intervention from other sources, or they may receive additional treatments that affect outcomes. This contamination can dilute the observed difference between groups. Researchers should monitor and document co-interventions in both groups.

Records and Measurements for Control Group Monitoring

Maintaining accurate records of control group activities and outcomes is essential for trial integrity and interpretation. The following records should be maintained for each control group participant.

Participant-Level Records

For each participant, record the date of enrollment, group assignment, baseline characteristics, and all scheduled and unscheduled assessments. Document any protocol deviations, including missed visits, incomplete assessments, and receipt of prohibited co-interventions.

Intervention Delivery Records

For attention control and usual care groups, record the number and length of contacts, the content of each contact, and the personnel who delivered the contact. In the aging-in-place trial, attention visitor records documented the number and length of visits, the types of activities participants chose, and how much visit time was spent on each activity.

Blinding Assessment Records

If blinding was used, record participants guesses about their group assignment and whether these guesses were correct. This information allows researchers to assess blinding integrity and to estimate the potential impact of expectancy bias.

Adverse Event Records

Record all adverse events in both the intervention and control groups, including the severity, duration, and relationship to the study intervention. Adverse event monitoring in control groups is important for assessing the safety of the intervention and for detecting harms that may be masked by the treatment effect.

Quality and Welfare Considerations

The welfare of control group participants is a primary ethical concern in clinical research. Control participants may receive a placebo when effective treatment exists, may be asked to wait for treatment in a waitlist design, or may receive a less intensive intervention in an attention control design. Researchers have an obligation to minimize the risks and burdens of control participation and to ensure that participants understand the nature of the control condition before consenting to participate.

In trials of behavioral interventions for older adults, attention control visits were perceived as beneficial by the majority of participants, suggesting that well-designed control conditions can provide meaningful experiences even when they do not include the active intervention components. However, the perceived benefit of control conditions should not be assumed, and researchers should monitor participant satisfaction and distress throughout the trial.

For historical control designs, the welfare concern is different. Participants in the current trial all receive the experimental intervention, so there is no group that is denied treatment. However, the validity of the comparison depends on the quality of the historical data, and participants should be informed that their outcomes will be compared to those of past patients.

Safety and Regulatory Context

The regulatory framework for clinical trials varies by jurisdiction, but most authorities require that trials be conducted in accordance with good clinical practice and that the rights, safety, and well-being of participants be protected. The choice of control group is a key design element that is reviewed by ethics committees and regulatory authorities.

The World Health Organization provides guidance on laboratory quality management and biosafety that is relevant to the conduct of clinical trials involving biological samples. The Laboratory Quality Management System Handbook and the Laboratory Biosafety Manual offer frameworks for ensuring the reliability of laboratory data and the safety of personnel handling biological materials. These documents are relevant to trials that include laboratory-based outcome measures, such as biomarker assessments.

The U.S. Food and Drug Administration provides guidance on bioanalytical method validation, which is relevant to trials that measure drug concentrations or biomarkers in biological matrices. The Assay Guidance Manual from the National Center for Advancing Translational Sciences offers practical guidance on developing and validating assays used in drug discovery and development.

Researchers should consult the relevant regulatory guidance for their jurisdiction and should document how the chosen control group design complies with ethical and regulatory requirements.

Professional Escalation Criteria

Certain situations during the conduct of a trial warrant escalation to a data safety monitoring board, ethics committee, or regulatory authority. The following criteria indicate when escalation may be necessary.

Unexpected Serious Adverse Events in the Control Group

If control group participants experience unexpected serious adverse events, the trial sponsor and investigators should promptly review the events and determine whether they are related to the trial procedures or to the underlying condition. Escalation to the data safety monitoring board is appropriate if the events suggest that the control condition is causing harm.

Evidence of Harm From the Intervention

If interim analyses suggest that the intervention is causing harm, the trial should be stopped or modified. This decision should be made by an independent data safety monitoring board, not by the investigators alone.

Major Protocol Violations Affecting Control Group Integrity

If a substantial number of control group participants receive the experimental intervention, or if the control condition is not delivered as specified, the validity of the trial may be compromised. Escalation to the sponsor and ethics committee is appropriate to determine whether the trial should be modified or terminated.

Blinding Failure That Threatens Trial Validity

If blinding is found to be compromised to a degree that threatens the validity of the trial, investigators should consider whether statistical adjustments can address the bias or whether the trial should be redesigned.

Changes in Standard of Care During the Trial

If the standard of care for the condition changes during the trial, the appropriateness of the control group may be called into question. This situation is particularly relevant for active control and usual care designs. Escalation to the data safety monitoring board is appropriate to determine whether the trial design should be modified.

Limitations of Control Group Designs

Every control group design has limitations that should be acknowledged in trial reports and considered when interpreting results.

Placebo controls cannot distinguish the specific effects of the intervention from the effects of receiving any treatment. Active controls require a proven effective comparator and may not be available for all conditions. Historical controls are vulnerable to era effects and selection bias. Waitlist controls cannot provide long-term comparisons and may be affected by participants seeking other help. Attention controls require careful design to match non-specific elements and are often underreported.

The choice of control group also affects the generalizability of trial results. Trials with strict inclusion and exclusion criteria and intensive control conditions may produce results that do not apply to broader patient populations. Researchers should consider the trade-off between internal validity, which is maximized by rigorous control, and external validity, which is maximized by enrolling diverse participants and using pragmatic control conditions.

Frequently Asked Questions

What is the difference between a control group and a control in a study?

A control in a study is any comparison condition used to estimate what would happen to participants without the experimental intervention. A control group is the specific set of participants assigned to that comparison condition. The control group may receive a placebo, an active treatment, usual care, no treatment, or a delayed version of the intervention, depending on the trial design.

Why is a control group needed in a clinical trial?

A control group provides a benchmark for measuring the effect of the experimental intervention. Without a control group, researchers cannot determine whether observed improvements are caused by the treatment or by other factors such as natural disease progression, participant expectations, or the effects of being observed. The control group allows researchers to isolate the specific effect of the intervention.

What is the purpose of a placebo control group?

A placebo control group receives an inert substance or sham procedure that looks identical to the active intervention. The purpose is to control for the psychological and physiological effects of receiving a treatment, including participant expectations of improvement. In a double-blind placebo-controlled trial, neither the participant nor the investigator knows which treatment is being administered.

When should an active control group be used instead of a placebo control?

An active control group should be used when an effective standard treatment exists and withholding it from control participants would be unethical. Active controls are also required for non-inferiority trials, where the goal is to show that a new treatment is no worse than the existing treatment within a predefined margin.

What are the risks of using a historical control group?

Historical controls are vulnerable to bias because medical care, diagnostic criteria, and patient populations change over time. Outcomes in historical patients may not be comparable to current outcomes even when the same inclusion and exclusion criteria are applied. Researchers should assess era effects and use the most recent data available when selecting historical controls.

How does a waitlist control group work?

In a waitlist design, participants are randomly assigned to receive the intervention immediately or after a delay. During the waiting period, the delayed group serves as controls. After the waiting period ends, the control group receives the intervention. This design ensures that all participants eventually receive the intervention, which can facilitate recruitment and address ethical concerns.

What is an attention control group and when is it used?

An attention control group receives the same dose and format of contact as the intervention group, but the content of that contact does not include the active elements of the intervention. Attention controls are used in behavioral intervention trials to distinguish the specific effects of the intervention from the general benefits of attention and engagement.

How can researchers tell if blinding has failed in a placebo-controlled trial?

Researchers can assess blinding integrity by asking participants to guess their group assignment and comparing the accuracy of their guesses to what would be expected by chance. If participants can correctly identify their group assignment more often than chance, blinding may have failed. Statistical tools are available to estimate the outcome of a perfectly blinded trial based on data from an imperfectly blinded trial.

Related Articles

• DNA Ladder Selection Guide: Choosing the Right Size Marker for Your Gel
• Biosafety Cabinet Types and Selection Guide for Microbiology Laboratories
• Understanding Competent Cells: Types, Preparation, and Storage for Transformation
• Understanding Positive Controls in PCR: Purpose, Selection, and Interpretation
• SYBR Green vs TaqMan qPCR: Choosing the Right Chemistry

References and Further Reading

• Laboratory Quality Management System Handbook. World Health Organization.
• Laboratory Biosafety Manual. World Health Organization.
• Assay Guidance Manual. National Center for Advancing Translational Sciences.
• Bioanalytical Method Validation Guidance. U.S. Food and Drug Administration.
• NCBI Literature Resources. National Center for Biotechnology Information.
• PubMed. National Library of Medicine.
• Experimental and quasi-experimental designs in implementation research.. Psychiatry research, 2020.
• Internet-based self-help intervention for procrastination: randomized control group trial protocol.. Trials, 2023.
• "Super-covariates": Using predicted control group outcome as a covariate in randomized clinical trials.. Pharmaceutical statistics, 2023.
• Attention control group activities and perceived benefit in a trial of a behavioral intervention for older adults.. Research in nursing & health, 2019.
• Efficacy of indocyanine green systemic administration for bile leak detection after hepatectomy: a protocol for a prospective single-arm clinical trial with a historical control group.. BMJ open, 2023.
• Reporting of "usual care" as the control group in randomized clinical trials of physiotherapy interventions for multiple sclerosis is poor: a systematic review.. Neurological sciences : official journal of the Italian Neurological Society and of the Italian Society of Clinical Neurophysiology, 2022.
• Ethical considerations.. Epilepsy research. Supplement, 1993.
• Defining the optimal historical control group for a phase 1 trial of mesenchymal stromal cell delivery through cardiopulmonary bypass in neonates and infants.. Cardiology in the young, 2023.
• Children and adults differ in how primary and secondary incentives modulate valuation, effort, and cognitive control.. 2026.
• Single-cell and spatial analyses reveal endothelial-macrophage inflammatory crosstalk in dry age-related macular degeneration.. 2026.
• Therapeutic targets for diabetic nephropathy identified by druggable genome mendelian randomization: the role of the gut microbiota-metabolite axis.. 2026.
• Comparative Analysis of Transcription Factor Binding Sites in the Long Control Region Across Human Papillomavirus Types.. 2026.
• Evolocumab Alters Transcriptomic Signatures and Identifies Inflammatory Biomarkers in Brain-Heart Syndrome with Coronary Heart Disease History.. 2026.
• Mitochondrial Metabolic Biomarkers in Periodontitis: Discovery and Clinical Validation.. 2026.
• Placebo control group in COVID-19 vaccine trials: context and timing matters. European Journal of Clinical Pharmacology, 2021.
• The difference between ‘placebo group’ and ‘placebo control’: a case study in psychedelic microdosing. Scientific Reports, 2023.
• Clinical trials: specific problems associated with the use of a placebo-control group. Journal of molecular medicine, 1995.
• Phase III trials: specific problems associated with the use of a placebo control group.. International journal of clinical pharmacology and therapeutics, 1997.
• Efficacy of mepolizumab add-on therapy on health-related quality of life and markers of asthma control in severe eosinophilic asthma (MUSCA): a randomised, double-blind, placebo-controlled, parallel-group, multicentre, phase 3b trial.. The Lancet Respiratory Medicine, 2017.
• Cognitive behavioral group treatment for social phobia: Comparison with a credible placebo control. Cognitive Therapy and Research, 1990.
• Does chess instruction improve mathematical problem-solving ability? Two experimental studies with an active control group. Learning and Behavior, 2017.
• Comparing the Serum Level of Vascular Endothelial Growth Factor (VEGF) in Patients with Active Pulmonary Tuberculosis and the Control Group: A Case Control Study. Iranian South Medical Journal, 2023.
• Complex Intervention Programs Integrating Multiple Intervention Strategies Were Not More Effective than Active Control Groups: Evidence from Randomized Controlled Trials. Behavioral Sciences, 2025.
• Matching Exercise Volume in Active Control Groups for Yoga Interventions. Alternative Therapies in Health and Medicine, 2023.

This article is educational and does not replace institutional policy, professional advice, or applicable safety and regulatory requirements.