When Treatment Hasn't Worked: What Clinical Research Can and Cannot Offer

For many patients, treatment begins with reasonable hope. Yet some people living with depression, anxiety, PTSD, or schizophrenia continue to suffer after multiple medication trials, dose changes, psychotherapies, or other appropriate treatments. Each incomplete response can make the next appointment harder and the health-care system feel less reliable. After years of trying, a patient may stop asking, “Which treatment comes next?” and begin asking, “Is anyone learning from what is happening to me?”

Disclosure, before you read further

• I practice at Inland Psychiatric Medical Group, and IPMG runs a psychiatric research department that enrolls patients in clinical studies. This article tells you to ask who pays the investigator and what interest they hold in the sponsor. So here is mine. I work for a group that enrolls research participants, and a reader who acts on this article could end up in one of its studies.

• I have handled it the only way I know how. This article names what research cannot promise before it names anything it can. Every protection described here is cited to the federal rule or agency source that creates it, and every limit on that protection sits in the same sentence. Nothing here is specific to any study at my group, and nothing here is an invitation to join one.

Key takeaways

• Not responding to treatment is not the same as failing at treatment, and it does not mean every path has been tried.
• Clinical research is one possible path. It is not a rescue, and no one should present it as one.
• Asking about a study commits you to nothing. Screening is not enrollment, and enrollment can be stopped.
• Patients can shape research without ever enrolling — as advisors who help choose questions, outcomes, and consent language.
• The protections are real, and they are not guarantees. Knowing the difference is what makes a “yes” informed and a “no” respected.

That question matters. When a treatment does not help enough, the result is not merely a number in a chart. It may mean another missed birthday, another semester postponed, another relationship strained, or another morning when getting out of bed feels impossibly far away. In a qualitative meta-synthesis of sixteen studies, people living with treatment-resistant depression described a sense of futility — that the return of depression was inevitable — alongside a desperate desire to get better, and experiences of isolation, hopelessness, and self-blame; some also described disappointment in the mental health care system and a loss of faith that their providers would listen.1 Those findings illuminate experiences rather than tell us how every patient feels. Nonresponse does not mean that the patient failed, that the clinician necessarily failed, or that every available path has been exhausted.

Clinical research can be one of those paths. It is not a guaranteed rescue, and it should never be presented as one. Research exists to produce reliable knowledge about health and illness — knowledge that can eventually improve decisions for many people.2 A participant may or may not receive the investigational intervention, may or may not improve, and may encounter inconvenience, disappointment, or harm. But for a person who freely chooses it after understanding those uncertainties, research can offer something ordinary care cannot: a structured opportunity to help answer the question that current care has not yet answered.

The first step is not enrollment. It is learning. A patient or family member can begin by searching ClinicalTrials.gov and bringing questions to the patient’s treating clinician. Psychiatric research also happens locally. Inland Psychiatric Medical Group, where I practice, runs a research department that enrolls Inland Empire patients in psychiatric studies, and a clinician there can say what is open. Asking about a study creates no obligation to join.

Zoom

Research begins with a human question

Research is sometimes pictured as a laboratory, a molecule, or a spreadsheet. Those are tools. The real beginning is a question: What is still not working, for whom, under what conditions, and what outcome would make a meaningful difference?

Patients are uniquely able to describe the burden of an illness and the tradeoffs they are willing — or unwilling — to accept. The U.S. Food and Drug Administration’s patient-focused drug development program describes a systematic approach to ensuring that patients’ experiences, perspectives, needs, and priorities are captured and meaningfully incorporated into drug development and evaluation, including the information most important to patients about treatment benefit, risk, and burden.3 The Patient-Centered Outcomes Research Institute similarly describes meaningful engagement as “an active partnership between researchers, patients, caregivers, clinicians and others throughout the entire research process — from planning the study to conducting the study and disseminating study results.”4

This means patients can advance research in more than one role. A research participant volunteers to follow a protocol and contribute data. A patient partner or advisor may help choose the research question, identify meaningful outcomes, improve consent language, reduce the burden of visits, review recruitment materials, interpret findings, or help communicate results. Some people do both; many do only one. Enrollment is not the price of having a voice.

Evidence about patient engagement is encouraging but not magical. Reviews have reported benefits such as better-designed protocols and more relevant chosen outcomes, greater understanding and rapport between researchers and communities, and service users who felt valued and empowered. The same reviews describe real burdens: lack of preparation and training, an overarching worry of tokenism, and the extra time and funding engagement requires. More recent work names harms the earlier reviews did not — negative attitudes toward public contributors, the emotional burden of involvement, frustration and disappointment, and further marginalization.5 Meaningful partnership therefore requires clear authority, preparation, accessibility, compensation when appropriate, and evidence that patient input actually changed something.

A brief American history: progress built from discovery — and from wrongdoing

The modern U.S. research system was not created in a single act, and its safeguards were not gifts handed down by an always-trustworthy system. Many arose after preventable harm, deception, exclusion, or public exposure of unethical conduct.

History should not be used to frighten patients, but neither should it be softened into a victory story in which abuse happened long ago and every weakness has been solved. Safeguards reduce risk; they do not eliminate it. Rules apply differently depending on the funder, institution, intervention, and type of study. The Common Rule does not automatically govern every U.S. research project, and FDA’s human-subject rules remain similar to — but not identical with — the Common Rule; proposed rules to harmonize them have not been finalized. Institutions may also impose protections beyond the federal floor. Oversight is carried out by people and institutions that can make mistakes. Trust is healthiest when it is informed, specific, and continually earned.9

What protects a participant today

A legitimate study should be able to explain its protections in plain language. For many drug trials, a sponsor must submit an Investigational New Drug application containing animal pharmacology and toxicology data, manufacturing information, data from any prior human research, information about the investigator, and proposed clinical protocols before human testing proceeds. FDA can delay or stop an investigation when, for example, participants are exposed to unreasonable or significant risk or the materials given to volunteers are misleading.11 This federal review does not prove that an investigational product is safe or effective; discovering those facts is part of the reason for doing the trial.

An Institutional Review Board, or IRB, reviews covered research before enrollment. Under the Common Rule, an IRB must determine, among other things, that risks are minimized, remaining risks are reasonable in relation to anticipated benefits and the importance of the knowledge expected, participant selection is equitable, consent will be sought and documented as required, safety monitoring is adequate when appropriate, and privacy and confidentiality provisions are adequate when appropriate. An IRB can require changes and disapprove a study, and it can suspend or terminate approval of research that is not being conducted in accordance with its requirements or that has been associated with unexpected serious harm to participants.12 It is a risk-governance layer — not a warranty that no participant will be harmed.

Informed consent is another essential protection, but a signature alone is not informed consent. When consent is required — and federal rules contain limited waivers and narrow emergency exceptions — the applicable rules require an understandable explanation of the research purpose, procedures, expected duration, foreseeable risks or discomforts, reasonably expected benefits, alternatives, confidentiality, contacts, and the voluntary nature of participation. The process should give the person sufficient opportunity to discuss and consider the decision and should minimize the possibility of coercion or undue influence. Under the Common Rule, consent must also begin with a concise and focused presentation of the key information most likely to help a person understand why they might or might not want to take part. Consent may not include language through which a participant waives, or appears to waive, any legal right, or releases researchers, sponsors, or institutions from liability for negligence.13

Some populations receive additional protections when the applicable rules cover the study. HHS regulations contain added requirements for research involving pregnant women, human fetuses and neonates, prisoners, and children; FDA rules add protections for children in FDA-regulated clinical investigations. IRBs must also ensure additional safeguards when participants are likely to be vulnerable to coercion or undue influence — for example children, prisoners, individuals with impaired decision-making capacity, or economically or educationally disadvantaged persons.14 These protections should not become automatic exclusion: fair access to potentially valuable research is also a matter of justice.

The distinction between care and research belongs at the center of that conversation. Clinical care is organized around the well-being of an individual patient. Research follows a protocol designed to generate generalizable knowledge. A protocol may use random assignment, blinding, fixed visit schedules, an active comparison, or a placebo because those methods help answer the scientific question. They can also limit how much the study team can tailor treatment to one participant. Confusing those purposes is known as therapeutic misconception. A systematic review of 31 studies in psychiatry research found variable and often high rates — roughly 13% to 86%, depending on the population studied and which aspect of misunderstanding was measured.15

Safety monitoring continues after enrollment. Investigators assess symptoms, side effects, laboratory findings, and other protocol-defined measures. Sponsors of investigational-drug studies must report qualifying serious risks — including suspected adverse reactions that are both serious and unexpected — to FDA and to participating investigators, generally within 15 calendar days, and within 7 calendar days for unexpected fatal or life-threatening reactions.16 Monitoring should match the study’s size, complexity, and risk. NIH policy requires an independent Data and Safety Monitoring Board for multi-site clinical trials involving interventions that entail potential risk, but not every study has one; in lower-risk or smaller studies, monitoring may take another form.17

Conflict safeguards operate at several levels. Under federal IRB rules, no IRB may have a member participate in the board’s review of any project in which the member has a conflicting interest, except to provide information requested by the IRB. FDA also requires specified financial information about clinical investigators when covered studies are submitted in marketing applications, so the agency can assess the extent to which potential bias has been minimized.18 These rules do not mean every professional or financial interest will appear in every consent form. A prospective participant can still ask whether the investigator, institution, or treating clinician is paid by or has another interest in the sponsor.

Privacy protections also deserve precise questions. The HIPAA Privacy Rule sets conditions under which covered entities may use or disclose protected health information for research — sometimes with the patient’s authorization, and sometimes under a waiver or alteration approved by an Institutional Review Board or a Privacy Board.19 Research funded by NIH that collects identifiable, sensitive information is automatically deemed to hold a Certificate of Confidentiality, which limits compelled disclosure in many circumstances but has exceptions — including disclosures required by other federal law, disclosures necessary for the person’s own medical treatment with their consent, and disclosures the person consents to. Certificates are also available by request for some research that is not federally funded.20 Neither HIPAA nor a Certificate makes data invisible, and neither applies identically to every project. Participants should ask who can see their records, whether data or samples may be shared for future research, how identifiers are handled, how long information is retained, and what could happen if confidentiality is breached.

Many applicable FDA-regulated trials and NIH-funded trials must be registered and have summary results reported on ClinicalTrials.gov — generally within 21 days of first enrollment for registration and within one year of the primary completion date for results — but reporting requirements do not cover every study.21 A public record can help a patient check the sponsor, protocol summary, eligibility criteria, locations, outcome measures, recruitment status, and sometimes results. Registration is valuable transparency; it is not an endorsement of scientific quality or a guarantee that the information is complete or current.

Finally, participation is voluntary. Covered consent rules require patients to be told that participation is voluntary, that refusal will involve no penalty or loss of benefits to which they are otherwise entitled, and that they may discontinue participation at any time on the same basis.13 Withdrawal can have limits that should be explained before enrollment. In many regulated studies, data already collected remain in the research record and may still be analyzed; stopping the study intervention, stopping new data collection, and allowing limited safety follow-up can be separate decisions.22

Two protections are often misunderstood because federal rules require explanation rather than a universal outcome. For research involving more than minimal risk, consent must explain whether any compensation and whether any medical treatments are available if injury occurs; federal consent rules do not themselves guarantee that either will be provided.13 Continued access to an investigational intervention after a trial also is not assured — expanded-access programs are created by the manufacturer, not by FDA. Patients should ask about injury coverage, an open-label extension, product supply, cost, and the plan for returning to ordinary care before they enroll.23

When treatment has not helped enough

Mental health treatment has changed many lives. Antidepressants, antipsychotics, and evidence-based psychotherapies are the usual foundations; brain-stimulation treatments have an established role when those have not worked well enough; and rehabilitative services and community supports matter throughout. Which of these fits, and in what order, depends on the diagnosis and the person. At the same time, average efficacy does not tell an individual patient whether the next treatment will restore sleep, concentration, safety, work, relationships, or hope.

In depression, NIMH uses the term treatment-resistant depression when a person “doesn’t get better after trying at least two antidepressants.” NIMH also notes that finding the best treatment may take trial and error, that brain stimulation therapy is an option when other depression treatments have not worked, and that the right plan depends on a person’s needs, preferences, and medical situation.24 Note what that public-facing definition leaves out: most research definitions also require that each trial was given at an adequate dose for an adequate duration. Two abandoned trials at subtherapeutic doses are not the same thing as treatment resistance.

In the landmark STARD study, 36.8% of participants reached remission during the first medication step — just over one in three, by the study’s self-report measure. Remission became less common at each later step (30.6%, 13.7%, and 13.0%), and relapse during follow-up was more frequent among those who had required more steps (40.1% after step one, rising to 71.1% after step four).25 STARD is an important 2006 study, not a prophecy for any one person and not a complete picture of today’s options — and a 2023 reanalysis of its patient-level data, disputed by the original investigators, put the cumulative remission figure closer to 35% than the widely cited 67%.26

For anxiety disorders, the phrase treatment resistant has been less standardized. A systematic review found wide variation in definitions, and a 2024 international Delphi panel proposed separate criteria for pharmacologic and psychotherapeutic nonresponse — broadly, failure of at least two adequate medication trials from different classes, or at least one adequate course of a standardized psychotherapy — chiefly to improve research and regulatory consistency. The panel stated that its criteria still require empirical validation, and they do not turn different anxiety disorders into one condition.27 The uncertainty is itself a research problem: if studies define “resistance” differently, it becomes harder to compare results or identify who might benefit from a new approach.

For PTSD, recommended trauma-focused psychotherapies help many people, yet a 2024 meta-analysis of 86 studies found that a substantial subgroup continued to have clinically important symptoms after first-line psychological treatment. The authors also reported substantial to considerable heterogeneity and a high risk of bias in half of the included studies — important reasons to avoid treating a single percentage as a universal truth. A 2026 systematic review of 55 studies likewise found substantial variation in how treatment-resistant PTSD is defined.28 Research is needed not only for new interventions, but also for better definitions, matching, sequencing, delivery, retention, and understanding of why effective therapies do not reach or work for everyone.

In schizophrenia, an international consensus group defined treatment resistance for research purposes around persistent symptoms of at least moderate severity, together with at least moderate functional impairment, despite at least two adequate antipsychotic treatment episodes — while emphasizing careful verification of dose, duration, and adherence.29 NIMH notes that if a person’s symptoms do not improve with usual antipsychotic medication they may be prescribed clozapine, which requires regular blood tests, and that psychosocial treatments remain important for day-to-day functioning and personal life goals.30 Even then, major questions remain about cognition, negative symptoms, side effects, recovery, prevention, and individualized treatment.

Before anyone is labeled “resistant,” clinicians should examine whether the diagnosis fits, treatments were delivered at adequate doses and durations, side effects limited use, medications interacted, access or cost interrupted treatment, substance use or medical conditions contributed, or the outcomes being measured failed to capture what the patient values. Across psychiatric diagnoses, the shared logic is a correct diagnosis, adequate treatment, and inadequate response — but the specific criteria vary considerably between disorders and guidelines.31 This is not an exercise in blaming the patient. It is a way of distinguishing biological nonresponse from a treatment that never had a fair or feasible test.

And a crucial language change helps: the person is not medication-resistant. A condition or set of symptoms has not improved enough with particular treatments under particular circumstances. The patient remains a person — not a failed case, not an eligibility category, and not raw material for science.

Who benefits most from research?

The honest answer is: it depends on the study and on what kind of benefit we mean.

The justice question is not only who bears the burden of research, but who receives its benefits. The Belmont Report observed that “during the 19th and early 20th centuries the burdens of serving as research subjects fell largely upon poor ward patients, while the benefits of improved medical care flowed primarily to private patients,” and warned that groups sought for their ready availability must be protected from involvement in research “solely for administrative convenience.”8 The opposite problem also matters: excluding older adults, racial and ethnic minorities, people with disabilities, rural communities, people with comorbidities, or those facing language and transportation barriers can leave the evidence poorly matched to the population that will eventually use a treatment. FDA states that “participants in clinical trials should represent the patients that will use the medical products,” and NIH inclusion policy seeks broader representation in NIH-supported clinical research.32

Fair participation therefore requires more than diverse recruitment photographs. It includes accessible information, qualified interpretation, practical visit schedules, transportation or remote options when scientifically appropriate, fair reimbursement, community relationships, and returning understandable results. A study should not ask communities to contribute data while keeping the knowledge, credit, and eventual benefits out of reach.

What a patient can leave behind

There is a profound temptation to tell a suffering person, “Join research and your pain will have meaning.” That sentence is emotionally powerful — and ethically dangerous. No patient owes science a body, a story, or another exhausting appointment. A person who declines research has not wasted an experience or failed future patients. Their suffering already deserves care, respect, and attention.

The more honest possibility is this: if a patient freely chooses to participate, an experience that once existed only between symptoms and a medical chart may become part of a disciplined question. One story is not proof. Across participants, however, carefully collected experiences can reveal patterns: a benefit that deserves confirmation, a side effect that changes monitoring, an outcome researchers forgot to measure, a subgroup that responds differently, or a promising idea that does not work and should be left behind.

That contribution can begin before a trial opens. A patient advisor might say, “A six-point symptom change means little to me if I still cannot return to work,” prompting the team to measure function. A caregiver might explain that a morning appointment excludes families who cannot safely travel then. A former participant might rewrite a confusing consent paragraph or ask the team to explain what happens after the last visit. These are not decorative stories; they can change the question, the protocol, and the usefulness of the answer.34

For some people, that is a form of legacy — not immortality and not guaranteed impact, but a contribution others may be able to build upon. Research cannot promise that suffering was “for something.” It can offer a voluntary route by which lived experience becomes protocol-defined evidence — and, when a study deliberately includes patient-reported outcomes, interviews, or patient advisers, a route for priorities and feelings to be heard more directly. Even a null result can narrow the field and protect future patients from misplaced confidence.

Questions to ask before saying yes

The Office for Human Research Protections publishes a fuller question list for prospective volunteers.33 For a mental health study, these are especially useful:

• Purpose and design: What exactly is the study trying to learn? What phase is it? Is there an NCT number? Who is the sponsor and funder? Who reviewed it?

• Research versus care: Which decisions follow the protocol rather than my individual clinician’s judgment? Is my treating clinician also an investigator? Who is responsible for my ordinary care?

• Assignment: Is the study randomized or blinded? What are the study arms? Could I receive placebo or a comparison treatment, and what is the chance of each assignment?

• Evidence and uncertainty: What is already known from laboratory work or earlier human studies? What benefits are reasonably possible? What is unknown?

• Risks and worsening symptoms: What physical, emotional, cognitive, social, privacy, or medication-withdrawal risks are foreseeable? Whom do I contact after hours? What happens if my symptoms worsen or I need urgent clinical care?

• Current treatment: Would I need a medication change or washout? How would that be supervised? How will the study team coordinate with my psychiatrist, therapist, primary-care clinician, or pharmacy? Do not start, stop, taper, or change medication to qualify for a study without guidance from the responsible clinicians.

• Daily burden: How many visits, calls, questionnaires, blood draws, scans, or digital measurements are required? How long are visits? What travel, work, caregiving, or childcare burdens should I expect?

• Costs and payment: What does the study pay, reimburse, or not cover? Could my insurance be billed? Who pays for medical care if I am injured by the research?

• Privacy and future use: Who can see my identifiable records? Will data or samples be shared, retained, sold, or used in future research? Is there a Certificate of Confidentiality? What happens to my data if I withdraw?

• Alternatives and aftercare: What evidence-based options are available if I decline? What happens at the end of the trial? If the investigational intervention helps, can I continue it, and who would pay? If it harms me or does not help, how will care transition?

• Voice and support: May I take the consent form home, ask for simpler language, use an interpreter, and include a trusted person? How is decision-making capacity assessed if there is a concern? Can I speak with someone independent of the recruitment team?

• Results and accountability: Will I receive overall results or clinically relevant individual findings? Does the study measure quality of life, daily functioning, and outcomes patients helped choose? Where will results be posted? Who can I contact about participant rights or a complaint?

Family support without taking away the patient’s voice

Family members can help by organizing questions, noticing changes, supporting transportation, reviewing practical burdens, and participating in safety planning when the patient wants that help. For an adult who has decision-making capacity, however, family involvement ordinarily follows the patient’s permission. Love does not replace consent.

A psychiatric diagnosis alone does not establish incapacity. Research on schizophrenia finds that decisional capacity varies widely within the group, correlates most strongly with neuropsychological impairment, and is often responsive to remediation — and it cautions against treating the diagnosis itself as proof that a person cannot consent.34 Capacity is specific to the decision and may change over time. When it is impaired, applicable law, institutional policy, IRB requirements, and the study’s risk level determine whether a legally authorized representative may decide and what additional protections are required. Even when someone else provides permission, the person’s preferences and objections should be respected to the greatest extent possible.

The best family posture is neither “You must try this” nor “Research is too dangerous.” It is: “Let’s learn what this study is, what it is not, and what matters to you. Then the decision is yours.”

The next step can be small

For someone who has spent years repeating treatments that did not help enough, even exploring research can feel like one more demand. Make the first step smaller: read one page, write down three questions, and discuss them with a clinician and a trusted person.

Patients can search the federal ClinicalTrials.gov database and ask a treating clinician to help interpret a listing. Study availability and eligibility can change, and an inquiry or screening conversation is not consent to enroll.

Local studies work the same way. Inland Psychiatric Medical Group runs a psychiatric research department in the Inland Empire. Asking it what is enrolling costs nothing and commits nothing. A clinician there can tell you which studies are open, what each one would ask of you, and what it would not change about your regular care. If any conversation starts to sound like an offer of treatment rather than a description of research, that is the moment to slow down and work through the questions above.

Asking what is enrolling is a question, not a commitment. Eligibility varies. Participation is voluntary. Personal benefit is not guaranteed.

Research may offer access to a new intervention. It may offer closer measurement, a language for experiences that have gone unheard, or the knowledge that one’s participation contributes to a larger answer. It may also offer no personal improvement. All of those possibilities belong in the same honest conversation.

Hope does not require certainty. It requires a next question worth asking, people willing to answer truthfully, safeguards strong enough to protect freedom, and a patient whose “yes” is informed — or whose “no” is respected.

Frequently asked questions

Does asking about a study mean I have signed up for it? No. An inquiry is a question, and a screening conversation is a separate step from consent. You can ask what a study involves, take the consent form home, and decide later or not at all. Federal consent rules require that you be told participation is voluntary and that refusing will not cost you benefits you are otherwise entitled to.

If I join, will I definitely get the new treatment? No. Many studies assign participants at random, and some compare an investigational treatment against a placebo or an existing treatment. You may not be told which group you are in. Ask directly what the study arms are and what your chance is of each assignment.

Can I keep seeing my own psychiatrist or therapist? Usually, but do not assume it. Ask who is responsible for your ordinary care during the study, whether your treating clinician is also an investigator, and how the study team will coordinate with your prescriber, therapist, and pharmacy. Never start, stop, or change a medication to qualify for a study without the clinician who manages it.

What happens if I want to stop? You may discontinue at any time. But withdrawal has limits worth understanding first: in many regulated studies, data already collected stay in the research record and may still be analyzed. Stopping the study intervention, stopping new data collection, and agreeing to limited safety follow-up can be three separate decisions.

Who pays if the research injures me? For studies involving more than minimal risk, the consent form must explain whether compensation and whether medical treatment are available if injury occurs. The federal rules require that explanation — they do not require that either be provided. Read that paragraph closely and ask who pays before you sign.

Can I help shape research without being in a study? Yes. Patient advisors help choose research questions, identify outcomes that matter, simplify consent language, reduce visit burden, review recruitment materials, and interpret findings. Enrollment is not the price of having a voice.

This article is for general education. It is not medical advice, not a recommendation to enter any study, and not a substitute for study-specific informed consent. Reading this does not create a clinician-patient relationship. Do not start, stop, taper, or change prescribed treatment to pursue research without the clinicians responsible for your care. Study availability, eligibility, risks, benefits, costs, payment, privacy protections, and continuation of ordinary care vary by protocol and site. Every regulatory, statistical, and citation claim here was verified against primary federal, journal, and agency sources on September 15, 2026; regulations, guidelines, and study listings change, so confirm current status before acting on anything here.

Related reading on NP FADY

• When the first two don’t work: a map of what comes next in treatment-resistant depression
• IV ketamine for depression
• Spravato, or esketamine, for depression
• How TMS works
• Modern ECT for depression: procedure, benefits, memory, and safety
• What psychiatric genetic testing can and cannot tell you

References

1. Cernat A, Pahwa M, Hadid D, et al. Patient experiences of treatment-resistant depression (TRD): A systematic review and qualitative meta-synthesis. PLOS Mental Health. 2024;1(6):e0000128. doi:10.1371/journal.pmen.0000128. PMID 41661783. Sixteen qualitative studies were included from 18,568 screened records; fourteen focused on experiences of specific treatments, so the findings should not be read as prevalence estimates.

2. National Institute of Mental Health. Clinical Trials — Information for Participants. Reviewed April 2024. NIMH describes clinical trials as research studies that look at ways to prevent, detect, or treat diseases and conditions, states that people take part in part “to help doctors and researchers learn more about a disease and improve health care,” and states that participation is entirely voluntary and may be withdrawn at any time. https://www.nimh.nih.gov/health/trials

3. U.S. Food and Drug Administration, Center for Drug Evaluation and Research. CDER Patient-Focused Drug Development. Accessed September 15, 2026. https://www.fda.gov/drugs/development-approval-process-drugs/cder-patient-focused-drug-development

4. Patient-Centered Outcomes Research Institute. Why Patient Engagement in Research Matters. Posted February 2, 2026. Accessed September 15, 2026. https://www.pcori.org/engagement-research/why-patient-engagement-research-matters

5. Domecq JP, Prutsky G, Elraiyah T, et al. Patient engagement in research: a systematic review. BMC Health Services Research. 2014;14:89. doi:10.1186/1472-6963-14-89. PMID 24568690; Brett J, Staniszewska S, Mockford C, et al. A systematic review of the impact of patient and public involvement on service users, researchers and communities. The Patient. 2014;7(4):387–395. doi:10.1007/s40271-014-0065-0. PMID 25034612; Wearn A, Brennan-Tovey K, Adams EA, et al. Evaluating process and outcomes of public involvement in applied health and social care research: a rapid systematic review. Health Expectations. 2025;28(1):e70160. doi:10.1111/hex.70160. PMID 39840654. The Wearn review is UK-focused, covers January 2006–July 2024, includes 19 studies, and identified no specific outcome evaluations.

6. U.S. Food and Drug Administration. Milestones in U.S. Food and Drug Law. Accessed September 15, 2026. https://www.fda.gov/about-fda/fda-history/milestones-us-food-and-drug-law; U.S. Food and Drug Administration. Part III: Drugs and Foods Under the 1938 Act and Its Amendments (Kefauver–Harris informed-consent requirement). https://www.fda.gov/about-fda/changes-science-law-and-regulatory-authorities/part-iii-drugs-and-foods-under-1938-act-and-its-amendments

7. Centers for Disease Control and Prevention. The Untreated Syphilis Study at Tuskegee Timeline. Last reviewed September 4, 2024. https://www.cdc.gov/tuskegee/about/timeline.html

8. U.S. Department of Health and Human Services, Office for Human Research Protections. The Belmont Report: Ethical Principles and Guidelines for the Protection of Human Subjects of Research. April 18, 1979 (44 FR 23192). https://www.hhs.gov/ohrp/regulations-and-policy/belmont-report/read-the-belmont-report/index.html; Centers for Disease Control and Prevention. Effects on Research. Last reviewed September 4, 2024. https://www.cdc.gov/tuskegee/about/effects-research.html

9. HHS Office for Human Research Protections. Federal Policy for the Protection of Human Subjects (“Common Rule”). https://www.hhs.gov/ohrp/regulations-and-policy/regulations/common-rule/index.html; HHS Office for Human Research Protections. Revised Common Rule (compliance dates January 21, 2019 and January 20, 2020). https://www.hhs.gov/ohrp/regulations-and-policy/regulations/finalized-revisions-common-rule/index.html; 45 C.F.R. Part 46, current text. https://www.ecfr.gov/current/title-45/subtitle-A/subchapter-A/part-46. FDA’s proposed harmonization rules (87 FR 58733 and 87 FR 58752, September 28, 2022) had not been finalized as of September 15, 2026; 21 C.F.R. 56.114 and 50.25 retain their pre-harmonization text.

10. National Institutes of Health. NIH Policy and Guidelines on the Inclusion of Women and Members of Racial and/or Ethnic Minority Groups in Clinical Research (current hub; policy revised by NOT-OD-25-131, July 17, 2025). https://grants.nih.gov/policy-and-compliance/policy-topics/inclusion/women-and-minorities; Clinical Trials Registration and Results Information Submission, Final Rule, 81 FR 64982 (September 21, 2016), effective January 18, 2017, codified at 42 C.F.R. Part 11. https://www.ecfr.gov/current/title-42/chapter-I/subchapter-A/part-11

11. U.S. Food and Drug Administration. Step 3: Clinical Research. https://www.fda.gov/patients/drug-development-process/step-3-clinical-research

12. 45 C.F.R. §§ 46.109, 46.111, 46.113. https://www.ecfr.gov/current/title-45/subtitle-A/subchapter-A/part-46

13. 45 C.F.R. § 46.116, especially (a)(2), (a)(4), (a)(5)(i), (a)(6), (b)(6), (b)(8), and (f). https://www.ecfr.gov/current/title-45/subtitle-A/subchapter-A/part-46; 21 C.F.R. §§ 50.20, 50.24, 50.25. https://www.ecfr.gov/current/title-21/chapter-I/subchapter-A/part-50. The “key information” and “reasonable person” requirements appear in the Common Rule at 45 C.F.R. § 46.116(a)(4)–(a)(5)(i); FDA’s parallel rule at 21 C.F.R. § 50.25 does not contain them.

14. 45 C.F.R. Part 46, Subpart B (pregnant women, human fetuses and neonates), Subpart C (prisoners), Subpart D (children), and § 46.111(b). https://www.ecfr.gov/current/title-45/subtitle-A/subchapter-A/part-46; 21 C.F.R. Part 50, Subpart D (children in FDA-regulated clinical investigations). https://www.ecfr.gov/current/title-21/chapter-I/subchapter-A/part-50

15. Thong ISK, Foo MY, Sum MY, et al. Therapeutic misconception in psychiatry research: a systematic review. Clinical Psychopharmacology and Neuroscience. 2016;14(1):17–25. doi:10.9758/cpn.2016.14.1.17. PMID 26792036. Thirty-one studies were included; reported rates ranged from 12.5% to 86%, and the authors caution that methods and definitions were not standardized across studies.

16. U.S. Food and Drug Administration. IND Application Reporting: IND Safety Reports (21 C.F.R. § 312.32). https://www.fda.gov/drugs/investigational-new-drug-ind-application/ind-application-reporting-ind-safety-reports

17. National Institutes of Health. NIH Policy for Data and Safety Monitoring. NOT-98-084, June 10, 1998; still listed as current NIH policy. https://grants.nih.gov/grants/guide/notice-files/not98-084.html

18. 45 C.F.R. § 46.107(d). https://www.ecfr.gov/current/title-45/subtitle-A/subchapter-A/part-46/subpart-A/section-46.107; U.S. Food and Drug Administration. Financial Disclosures by Clinical Investigators (21 C.F.R. Part 54). https://www.fda.gov/science-research/clinical-trials-and-human-subject-protection/financial-disclosures-clinical-investigators

19. HHS Office for Civil Rights. Research. HIPAA Privacy Rule guidance. https://www.hhs.gov/hipaa/for-professionals/privacy/guidance/research/index.html

20. National Institutes of Health. Certificates of Confidentiality. Page last updated August 10, 2024. https://grants.nih.gov/policy-and-compliance/policy-topics/human-subjects/coc; NIH Notice NOT-OD-17-109, implementing 21st Century Cures Act § 2012, effective October 1, 2017. https://grants.nih.gov/grants/guide/notice-files/NOT-OD-17-109.html

21. 42 C.F.R. §§ 11.24, 11.44. https://www.ecfr.gov/current/title-42/chapter-I/subchapter-A/part-11; ClinicalTrials.gov. Clinical Trial Reporting Requirements. https://clinicaltrials.gov/policy/reporting-requirements. The Final Rule implements requirements for “applicable clinical trials” under FDAAA § 801; a separate NIH policy reaches a broader set of NIH-funded clinical trials.

22. HHS Office for Human Research Protections. Guidance on Withdrawal of Subjects from Research. September 21, 2010; content reviewed June 2, 2026. https://www.hhs.gov/ohrp/regulations-and-policy/guidance/guidance-on-withdrawal-of-subject/index.html; U.S. Food and Drug Administration. Data Retention When Subjects Withdraw from FDA-Regulated Clinical Trials. October 2008. https://www.fda.gov/regulatory-information/search-fda-guidance-documents/data-retention-when-subjects-withdraw-fda-regulated-clinical-trials

23. U.S. Food and Drug Administration. How Can I Get Access to a Drug That Is in Testing but Has Not Yet Been Approved? FDA describes clinical trials and expanded access as distinct routes and states that manufacturers, not FDA, create expanded-access programs. https://www.fda.gov/drugs/frequently-asked-questions-popular-topics/how-can-i-get-access-drug-testing-has-not-yet-been-approved

24. National Institute of Mental Health. Depression. https://www.nimh.nih.gov/health/publications/depression

25. Rush AJ, Trivedi MH, Wisniewski SR, et al. Acute and longer-term outcomes in depressed outpatients requiring one or several treatment steps: a STAR*D report. American Journal of Psychiatry. 2006;163(11):1905–1917. doi:10.1176/ajp.2006.163.11.1905. PMID 17074942. Remission was defined as QIDS-SR16 ≤ 5. The paper also reports a theoretical cumulative remission rate of 67%, which assumes no dropouts and that those who exited would have remitted at the same rate as those who stayed.

26. Pigott HE, Kim T, Xu C, Kirsch I, Amsterdam J. What are the treatment remission, response and extent of improvement rates after up to four trials of antidepressant therapies in real-world depressed patients? A reanalysis of the STARD study’s patient-level data with fidelity to the original research protocol. BMJ Open. 2023;13(7):e063095. doi:10.1136/bmjopen-2022-063095. PMID 37491091. The original investigators responded in Rush AJ, et al. The STARD data remain strong: reply to Pigott et al. American Journal of Psychiatry. 2023;180(12):919–920. doi:10.1176/appi.ajp.20230869. The point is contested, not settled.

27. Bokma WA, Wetzer GAAM, Gehrels JB, Penninx BWJH, Batelaan NM, van Balkom ALJM. Aligning the many definitions of treatment resistance in anxiety disorders: a systematic review. Depression and Anxiety. 2019;36(9):801–812. doi:10.1002/da.22895. PMID 31231925; Domschke K, Seuling PD, Schiele MA, et al. The definition of treatment resistance in anxiety disorders: a Delphi method-based consensus guideline. World Psychiatry. 2024;23(1):113–123. doi:10.1002/wps.21177. PMID 38214637. Thirty-six experts and stakeholders reached consensus on 14 recommendations at a ≥75% agreement threshold.

28. Semmlinger V, Leithner C, Klöck LM, Ranftl L, Ehring T, Schreckenbach M. Prevalence and predictors of nonresponse to psychological treatment for PTSD: a meta-analysis. Depression and Anxiety. 2024;2024:9899034. doi:10.1155/2024/9899034. PMID 40226730. Eighty-six studies, 117 active treatment conditions, 7,894 participants; weighted average nonresponse 39.23% (95% CI 35.08–43.53), I² = 83.12%, high risk of bias in half the studies. Al-Shamali HF, Meshkat S, Janssen-Aguilar R, et al. Toward a definition of treatment-resistant posttraumatic stress disorder: a systematic review. Journal of Military, Veteran and Family Health. 2026;12(1):17–33. doi:10.3138/jmvfh-2025-0015. Fifty-five of 4,046 screened articles met inclusion criteria. U.S. Department of Veterans Affairs and Department of Defense. VA/DoD Clinical Practice Guideline for the Management of Posttraumatic Stress Disorder and Acute Stress Disorder, Version 4.0 (2023). https://www.healthquality.va.gov/guidelines/mh/ptsd/

29. Howes OD, McCutcheon R, Agid O, et al. Treatment-resistant schizophrenia: Treatment Response and Resistance in Psychosis (TRRIP) Working Group consensus guidelines on diagnosis and terminology. American Journal of Psychiatry. 2017;174(3):216–229. doi:10.1176/appi.ajp.2016.16050503. PMID 27919182. Minimum criteria include at least two antipsychotic trials of ≥6 weeks each at therapeutic dose (≥600 mg/day chlorpromazine equivalents where the target dose is unclear), ≥80% adherence established by at least two sources, at least moderate symptom severity persisting ≥12 weeks, and at least moderate functional impairment.

30. National Institute of Mental Health. Schizophrenia. https://www.nimh.nih.gov/health/publications/schizophrenia

31. Howes OD, Thase ME, Pillinger T. Treatment resistance in psychiatry: state of the art and new directions. Molecular Psychiatry. 2022;27(1):58–72. doi:10.1038/s41380-021-01200-3. PMID 34257409. A systematic review of 23 consensus guidelines found the same three core components across disorders — correct diagnosis, adequate treatment, inadequate response — with considerable variation in the specific criteria used.

32. U.S. Food and Drug Administration. Clinical Trial Participation. https://www.fda.gov/consumers/health-education-resources/clinical-trial-participation; National Institutes of Health. Inclusion of Women and Members of Racial and/or Ethnic Minority Groups in Clinical Research. https://grants.nih.gov/policy-and-compliance/policy-topics/inclusion/women-and-minorities

33. HHS Office for Human Research Protections. Questions to Ask About Volunteering for a Research Study. Reviewed January 28, 2020. https://www.hhs.gov/ohrp/education-and-outreach/about-research-participation/questions-to-ask/index.html

34. Appelbaum PS. Decisional capacity of patients with schizophrenia to consent to research: taking stock. Schizophrenia Bulletin. 2006;32(1):22–25. doi:10.1093/schbul/sbi063. PMID 16177275; Spencer BWJ, Shields G, Gergel T, Hotopf M, Owen GS. Diversity or disarray? A systematic review of decision-making capacity for treatment and research in schizophrenia and other non-affective psychoses. Psychological Medicine. 2017;47(11):1906–1922. doi:10.1017/S0033291717000502. PMID 28441976.

If you or someone you know is in crisis

• Call 911 or go to your nearest emergency room for any life-threatening emergency.
• 988 Suicide & Crisis Lifeline — call or text 988, or chat at 988lifeline.org, available 24/7, free and confidential. En español: marque 988 y oprima 2, o envíe AYUDA al 988. Deaf or hard of hearing: 988 videophone (ASL) or TTY 711 then 988. Veterans: 988 and press 1, or text 838255.
• Crisis Text Line — text HOME to 741741.
• The Trevor Project (crisis support for LGBTQ+ young people) — call 1-866-488-7386, or text START to 678-678.
• National Sexual Assault Hotline (RAINN) — call 1-800-656-HOPE (4673) or text HOPE to 64673; free, confidential, 24/7. Online chat at RAINN.org/hotline.
• National Domestic Violence Hotline — call 1-800-799-SAFE (7233) or text START to 88788; 24/7, help in 200+ languages. Online chat at TheHotline.org. If your phone or computer may be monitored, calling from a safer device is an option.
• Riverside County — Inland SoCal Crisis Helpline 951-686-HELP (4357), 24/7 (Inland SoCal United Way / 211+, in partnership with RUHS-BH); Community Access, Referral, Evaluation and Support (CARES) Line 800-499-3008, 24/7.
• San Bernardino County — Access Unit (Behavioral Health Helpline) 888-743-1478, 24/7; Mobile Crisis/CCRT 800-398-0018 (24/7, all ages) or text 909-420-0560. Arrowhead Regional Medical Center (ARMC) has a dedicated walk-in adolescent psychiatric ER (ages 13–17).
• Children under 13 — call 911 for immediate danger, contact your county's mobile crisis team (they respond to all ages), or go to the nearest pediatric emergency room.
• California Peer-Run Warm Line (non-crisis — someone to talk to) — call or text 1-855-600-WARM (9276); daytime and evening hours, not a 24/7 line.
• NP Fady (non-emergency) — for routine scheduling or questions, call (909) 707-6261. This line is not monitored for emergencies.