Medication, physical dependence, and addiction: questions without shame

“Some things are in our control and others not.”

Epictetus, Enchiridion, §1, Elizabeth Carter translation; verified text. A philosophical frame, not an account of how medication works.

What this article is about

A missed tablet can prompt an anxious question: “If I feel unwell without it, does that mean I'm addicted?” Sometimes the question comes from the person taking it, sometimes from someone close who wants to help them give up “all those chemicals.” But the same label brings together different situations: a condition returning, withdrawal symptoms, the body's adaptation, and use involving loss of control.

This article helps you understand what to discuss with a doctor when medication is needed for a long time, no longer works as it did, or is difficult to stop. It concerns adults; it does not determine individual prescriptions, doses, or tapering schedules. Length of use and a single unpleasant feeling after a missed dose do not establish a diagnosis.

If you take medication regularly, do not deliberately skip doses or stop abruptly to test your independence. Abrupt benzodiazepine withdrawal can be particularly dangerous. Seizures, a marked disturbance of consciousness, or difficulty breathing require urgent medical care. A planned treatment review and care for an acute condition are different tasks. FDA, benzodiazepine information.

I. Four questions rather than one label

Start by distinguishing the concepts. Physical dependence means the body has adapted to a medicine and a change in use may cause withdrawal symptoms. Tolerance means less of the previous effect from the same amount, or needing more to obtain that effect. Addiction involves problematic use and difficulties with control. Craving may be a strong desire to take the medicine, but it needs to be understood in context, rather than from one word the person uses. The MHRA explicitly distinguishes physical dependence from addiction: the former does not necessarily mean the latter. MHRA, patient advice and definitions.

These phenomena can coexist, but they do not have to occur together. It is useful to expand “am I addicted?” into four questions: is treatment beneficial; is there physical adaptation; what is happening to control over medication use; and what are the risks of continuing or changing it? One answer does not replace the others.

“I'm afraid of being left without my medication” is also ambiguous. Someone may fear severe pain returning, withdrawal symptoms, losing the ability to function, or may crave a particular effect. Meaning emerges from the history: what they fear, what has already happened, and how they actually use the medicine. Fear should not be treated as proof of deception; understanding the fear does not remove the need to assess risk.

II. A question about dependence does not erase treatment benefits

Considering medication solely in terms of whether it can be stopped loses sight of treatment's original purpose. It matters why it was prescribed and what changed: symptoms and the ability to work, sleep, care for yourself, and maintain relationships. Consider a hypothetical example: someone can manage ordinary tasks again after treatment for depression but also notices unpleasant side effects. Benefit and difficulty deserve discussion together; the choice is not limited to “everything is perfect” or “it was all a mistake.”

The Royal College of Psychiatrists recommends regularly reviewing antidepressant treatment and discussing both benefits and difficulties associated with long-term use. Some conditions need prolonged treatment. That does not automatically establish addiction, and it does not mean treatment cannot change. Royal College of Psychiatrists, Stopping antidepressants.

For opioid pain treatment, the CDC likewise recommends jointly assessing the benefits and risks of continuation and, separately, the benefits and risks of reduction. Either decision needs a basis. This pain recommendation applies within the guideline's scope; it does not replace specific decisions in cancer-related pain, palliative care, or treatment of opioid use disorder. CDC, Continuing Opioid Therapy, scope of the CDC 2022 guideline.

III. Feeling worse after a missed dose: withdrawal or a returning condition?

The cause of deterioration after a change in medication is not always clear. Antidepressant withdrawal can produce symptoms resembling anxiety or depression. The Royal College of Psychiatrists recommends considering when symptoms began, what they feel like, and the history of medication changes. Withdrawal symptoms often start sooner than returning depression, but delays are possible; timing alone is insufficient. Royal College of Psychiatrists, distinguishing withdrawal from returning symptoms.

Report precisely when you last took the usual dose, what changed, when symptoms appeared, whether you had experienced them before, and what else was happening then. This helps examine possible explanations. Do not alternate stopping and restarting on your own to test the response: observation should not create additional treatment fluctuations.

If symptoms ease after restarting, that is not ready-made proof of addiction either. Discuss the response with the prescribing professional alongside the rest of the history. New thoughts of suicide or a sudden, dangerous deterioration require prompt professional help; do not wait for a routine appointment while trying to determine the cause first. Royal College of Psychiatrists.

IV. Different medication classes need different conversations

Antidepressants

These treat specified conditions, including depression and some anxiety disorders. Difficulty stopping may involve physical adaptation and withdrawal. The RCP distinguishes this from addiction. Serious symptoms deserve attention even when addiction has not been established. “It isn't addiction” should not be taken to mean “nothing is happening.” Royal College of Psychiatrists.

Benzodiazepines and Z-drug sleeping medicines

Benzodiazepines carry distinct risks of addiction, physical dependence, and withdrawal. A prescription does not remove them. The FDA warns about severe reactions to abrupt discontinuation, including seizures. That is a reason to discuss changes with a doctor, rather than endure dangerous symptoms alone. FDA.

Z-drugs are a separate class of sleeping medicines. In January 2026, the MHRA strengthened its communication about dependence, addiction, withdrawal, and tolerance for this class, benzodiazepines, and gabapentinoids. The UK update supports distinguishing the risks; it does not establish the same probability for every medicine and person. MHRA.

Gabapentinoids

Pregabalin and gabapentin cannot be assessed using the rule “from a pharmacy means safe” either. The MHRA update addresses their risks in the relevant prescribing contexts. However, its assessment did not include specialist use for epilepsy. Advice about stopping symptomatic treatment should not be transferred to an antiseizure prescription: the purpose and continuing need for treatment require separate discussion. MHRA, review scope and patient advice.

Opioid pain medicines

Physical adaptation also needs attention here. However, in the CDC's description of DSM-5 criteria for opioid use disorder, tolerance and withdrawal are not counted when opioids are taken solely under appropriate medical supervision. This specific qualification does not rule out assessing other signs or make opioids harmless. CDC, Opioid Use Disorder: Diagnosis.

For a planned pain-treatment review, the CDC cautions against abrupt discontinuation unless there is a life-threatening issue, such as warning signs of an impending overdose. That exception needs a medical response; it is not an instruction to diagnose an overdose at home and manage doses yourself. CDC, Continuing Opioid Therapy.

Prescription stimulants

These medicines can be used for conditions such as ADHD. The FDA recognises their therapeutic benefits when used appropriately while warning about misuse, addiction, and overdose. Treatment for an indicated condition differs from taking someone else's medicine or changing your own use without agreement. Do not share your medication with others. FDA, prescription stimulant warnings.

V. Misuse and addiction are also different

Misuse can mean taking medication differently from how it was prescribed or taking someone else's prescription medicine. This is important safety information. But a single departure does not replace a full assessment of a disorder. The FDA distinguishes prescribed use from misuse; the CDC considers a pattern and combination of signs when assessing opioid use disorder. FDA, stimulants, CDC, diagnosis.

Consider a hypothetical example: after pain worsens, someone begins taking medicine more frequently than agreed. Report that directly. Pain, the current prescription, safety, and possible impaired control need assessment. You do not have to declare yourself addicted before receiving help, and you should not hide your actual use to protect the image of a “good patient.”

Discussing craving also requires more than asking whether the medicine produces pleasure. Someone may continue problematic use in an effort to relieve an unpleasant state. Control, consequences, and reasons for taking it matter. Medical monitoring should not become a moral examination, and openness does not imply that every form of use is acceptable.

VI. Medication for addiction can support recovery

Opioid use disorder is a distinct situation. Buprenorphine and methadone are used to treat it; other medication options with different mechanisms also exist. The CDC reports that medication treatment for OUD is associated with lower risks of overdose and overall mortality. Maintenance medication use does not, by itself, mean recovery has failed. CDC, Opioid Use Disorder: Treating.

“Does the person take a substance?” does not describe the whole outcome of treatment. Safety, control, health, and everyday life matter. This does not mean the medication has no risks or can be self-prescribed. Choice, support, and access must be assessed individually and in the particular country.

The CDC also advises against treating OUD with detoxification alone without appropriate medication because of increased risks of returning to use and overdose. Getting off every medication as quickly as possible is not a universal treatment goal. Duration of care is decided according to the condition and circumstances, rather than to provide outward proof of “real sobriety.” CDC, Opioid Use Disorder: Treating.

VII. How to prepare a treatment review

Gather information about medicines and actual use, including other medication, alcohol, and substances. Combining benzodiazepines with alcohol or opioids is particularly important to discuss because of the risk of serious harm. Do not try to solve an interaction problem by abruptly stopping one medicine on your own. FDA, benzodiazepines.

Five questions can help at an appointment:

What is treatment intended to achieve now, and what benefit are we actually seeing?
What might explain new symptoms: the condition, a side effect, withdrawal, another medicine, or a combination of causes?
Are there signs of tolerance or difficulties with control, and what further assessment is needed?
What options exist for continuation, change, and support, and who is responsible for the agreed plan?
What should happen if things get worse, who can I contact between visits, and how can continuity of prescribed treatment be maintained?

If your doctor is unavailable or your medicine is running out, tell the treating service or an available medical professional in advance. A pharmacist can help check the name, formulation, instructions, and interactions, but should not replace necessary medical assessment. Do not borrow another person's schedule: the same medication name does not mean the same situation.

Someone close to you can ask “what do you need to discuss with your doctor, and how can I help you prepare?” rather than demand that you throw away the tablets immediately. They can help record observations with your consent. They should not independently confiscate prescribed medicine or decide that someone else's treatment must stop.

VIII. Regaining a say in the decision

Participating in treatment does not require unquestioning trust in every prescription or proving freedom by refusing any pill. It begins with being able to name the benefit, difficulty, uncertainty, and goal. You can ask for explanations, discuss alternatives, and seek another professional opinion.

Physical dependence creates a medical task requiring support. If addiction is identified, that is a reason to select appropriate help, rather than deny it. In both cases, your life extends beyond a medication list: assessment should help you live more safely and freely, taking your actual condition into account.

IX. Freedom does not begin with a test for the body

A question about dependence can quickly become an argument about worth. Someone is told: if you are truly free, manage without the pill. But a family vote cannot decide how a body will respond. Feeling worse after a missed dose tells us there is a situation to understand; it does not establish a person's moral value. The opposite extreme also removes participation: because the issue is medical, the patient supposedly has nothing to ask. You can lack control over a bodily reaction and still take part in decisions, within the limits of your condition, information, and available support.

At the beginning of the Enchiridion, Epictetus distinguishes what depends on us from what does not. A modest application here is to stop demanding immediate control over every bodily reaction and look for an action within reach: report a change, ask a question, request help. Yet this ancient account of authority over judgment has limits. Severe anxiety, confusion, exhaustion, and illness may make those very actions difficult. Being unable to reason calmly does not prove weak will. Epictetus, §§1–2 offers an ethical conversation, not a clinical test of independence.

In the preface to Part III of the Ethics, Spinoza contrasts understanding human emotions with ridicule and condemnation. This can redirect a question from blame toward what is happening and what help is needed. His metaphysics, however, does not establish a symptom's cause or a medicine's effects. Part II, Proposition 7 links ideas and things within that system; it is not modern neuropharmacology. Spinoza, Ethics, II.7; III, preface.

In John's Gospel, the disciples encounter a man blind from birth and ask who sinned. Jesus rejects the proposed blame of the man or his parents. The comparison relevant here is limited: suffering does not authorize an automatic assignment of guilt. The account also contains a theological explanation and a healing; it does not describe medication dependence. It cannot require readers to call illness a gift, wait for a miracle instead of seeking help, or expect recovery if they have enough faith. John 9:1–7, KJV.

These texts do not say the same thing. Stoic choice, Spinoza's account of causes, and the Gospel encounter have different foundations. We can place them alongside one another around a narrower question: how to respect a person without denying bodily vulnerability or the consequences of actions. That is where philosophical comparison ends and medical work begins: identifying the medicine, circumstances, and appropriate help. The attractive word freedom cannot do that work for us.

X. When others debate the decision and you live with the consequences

Consider a fictional situation. Lena takes prescribed medication and recently felt worse after missing a dose. Her sister says this proves addiction, while an acquaintance advises her to simply endure a few days. Lena is afraid both to believe them and to disagree; she does not understand the reaction herself. Her first step need not be winning an argument about terminology. It is more useful to tell an accessible qualified medical professional what she takes, what actually changed, and which symptoms appeared. The explanation remains open.

Her sister can help, if Lena wants: find an earlier treatment record, check the service's opening hours, or join a conversation with an agreed role. Help stops being support when someone hides the medication, reads private records without permission, or makes stopping treatment a condition of care. Respecting choice also does not mean relatives must stay silent about dangerous observations. They can describe a fact, such as unusual drowsiness after substances were combined, and seek necessary help without announcing an accusatory diagnosis. Problems with breathing or consciousness require urgent medical help, not a family debate about the degree of dependence.

Another important case is a medicine that helps and is taken as prescribed, while the next appointment is expensive or far away. Asking why someone still takes it misses the immediate question: who will support safe continuity and a review? An unavailable clinician does not turn an unsupervised interruption into a justified experiment. Contact the treating service or an accessible qualified medical professional in advance and state the problem directly: the medicine is running out and the plan is unclear. A pharmacist may help identify the product and local routes to assistance; their powers differ between countries. There is no universal promise that they can renew a prescription.

If there is no energy for a diary, an app, or a long list, three spoken facts can be enough to begin: the medication's name, what changed in its use, and what is happening now. If you cannot remember the name, a package or earlier prescription may help when available. Missing records do not remove your right to ask for care. A safe relative is not assumed, and family involvement is not required. If no answer is accessible, that remains an access barrier; it does not show that the person lacks commitment to recovery.

A harder answer is possible: an assessment may identify problems with control or a substance use disorder. Distinguishing the terms should not become a way of defending against all unwelcome information. Its purpose is to make help more precise. You can acknowledge a risk, reject humiliation, and ask what specific support is being offered. A useful conversation is measured by whether the next safe step becomes clearer, not by whether a comfortable label survives.

A quieter outcome is possible too: treatment is helpful, no urgent change is needed, and tolerability can be discussed at a planned review. Continuing agreed care can also follow a considered decision. This article does not prescribe continuation or discontinuation to every reader. It gives each question its own basis: observed changes for benefit, assessment for risk, understandable options and conditions for choice, and respectful treatment for dignity.

XI. Ending a conversation when certainty is still incomplete

Not every appointment ends with a confident explanation. A clinician may need to reconstruct prescribing history, check combinations, or assess several possible causes. That differs both from refusing to listen and from reaching a settled conclusion. Ask what remains unknown and what action could clarify it. A vague promise to see how things go becomes more useful when someone is responsible for the assessment, its question is clear, and there is a contact route if the situation changes. The service has a responsibility to explain; this is not an exam in perfectly completed patient homework.

Consider another fictional example. Oleg wants one answer: addiction or no addiction. During the conversation, however, it becomes clear that he remembers the name of only one of several products, and a recent change of packaging may matter to the history. He can leave without a final label but with a specific next step: ask his service to clarify the list and provide it to the professional assessing him. If he cannot collect the information himself, available assistance needs discussion. Missing documentation does not mean symptoms are invented, and a need for clarification does not prove any proposed mechanism.

Another person may understand a calm explanation but feel unready for the proposed change. A previous experience was very difficult, and contact between appointments is uncertain. That concern belongs in the decision; it is not an obstacle to hide in order to appear agreeable. They can say: I understand the aim, but I do not understand what help will be available if I feel worse. This asks for conditions to be clarified; it does not require the clinician to endorse every preferred option. A real disagreement may remain. Naming it and discussing a second opinion, where accessible, is more useful than pretending agreement has been reached.

An agreed plan must be understandable to the person who will live with it. Asking for another explanation does not make them less capable. If the consultation takes place in an unfamiliar language, request interpreting or simpler language when possible; professional interpreting depends on the service. A child or an unsafe relative should not be treated as a required intermediary. Where private information is involved, it helps to establish whom the person wants included and what may be discussed in their presence. Receiving support does not require sharing an entire treatment history with everyone around you.

The day after an appointment, relief and doubt may coexist. That does not automatically show either that the decision was poor or that enough information was provided. Check a narrower result: is the current question understandable, is the responsible contact known, and can an observation be distinguished from a proposed cause? Emotional certainty does not always arrive with practical clarity. Conversely, a confident voice does not establish that an explanation has been verified. What matters is being able to return to a specific uncertainty and receive an answer within the professional's competence.

Families may need to change their aim too. Instead of promising to ensure that someone never takes anything again, they can agree on practical help with seeking care. If a problem with substance use is identified, support need not become a reward withheld until all difficulties disappear immediately. Relatives may still set their own boundaries, such as not sharing prescriptions and refusing dangerous requests. Compassion and boundaries can coexist; neither requires prescribing treatment for someone else.

The outcome may be continued monitoring, a clearer history, further assessment, or a separate discussion about changing treatment. Those are different results with different grounds. The reader's useful gain is more modest than a promise of complete independence: a way to discuss a bodily reaction without passing judgment on themselves, and a risk without denying it. An assessment can legitimately remain unfinished. Leaving a person with an unclear plan and a demand to prove their strength is a different matter.

Practice: one note before the conversation

If your condition allows, spend a few minutes on a short note. First record or say a fact without a conclusion: after this change, this happened. Name the concern separately: I am afraid this means addiction, or I am afraid of losing treatment. Then add what you want from the appointment: understand possible explanations, discuss benefits and risks, and establish how to contact the service again.

Choose one question: how can we distinguish physical adaptation, a returning condition, and problems with control in my case? If the reply contains an unfamiliar term, ask for an explanation using observable features and ask what remains unknown. Establish who the next contact is, when it will happen, and what to do if things worsen. Do not change medication use to test the hypothesis in your note.

Writing is optional. You can say these words or ask for language support when available. Everything need not be resolved in one appointment. Acute dangerous symptoms take priority over this sequence. The practice produces a prepared question and a contact route, not a diagnosis made on your own.

Key quotation

“The order and connection of ideas is the same as the order and connection of things.”

Spinoza, Ethics, Part II, Proposition 7, R. H. M. Elwes translation. Text and explanatory note. Here the quotation recalls the obligation to seek grounds for explanations. It belongs to Spinoza's metaphysical system; it proves neither a particular symptom's cause nor the equivalence of all treatments.

Map of connections between traditions

Basis	What it helps us notice	Boundary
Epictetus, Enchiridion, §§1–2	The difference between a bodily outcome and an available action: ask, report, request help	Illness can limit action and judgment too; this is not a test of will
Spinoza, Ethics, II.7 and preface to III	A move from condemnation toward causal explanation	Metaphysics cannot replace medical assessment
John 9:1–7	Refusing to infer personal or family guilt automatically from suffering	A theological narrative is neither a healing promise nor an account of medication effects

These three comparisons keep differences between traditions visible. Medical information has the separate official sources below.

Sources and foundations

Royal College of Psychiatrists. Stopping antidepressants. Patient information: physical adaptation, withdrawal, returning symptoms, and reviewing benefits; UK context.
FDA. Benzodiazepine warnings, 23 September 2020. Addiction, physical dependence, withdrawal, and combination risks; US regulatory communication.
MHRA. Improving Information Supplied with Gabapentinoids, Benzodiazepines and Z-Drugs, 8 January 2026. Updated risk communication, individual support, and review limits; UK.
FDA. Prescription stimulant warnings, 11 May 2023. Therapeutic use and misuse; US.
CDC. Opioid Use Disorder: Diagnosis, 16 April 2024. DSM-5 opioid use disorder and the tolerance/withdrawal qualification under appropriate medical supervision.
CDC. Continuing Opioid Therapy, 7 May 2024, and Clinical Practice Guideline for Prescribing Opioids for Pain — United States, 2022, scope excerpt. Reviewing opioid pain treatment; general information for US professionals, not an individual plan.
CDC. Opioid Use Disorder: Treating, 9 April 2024. Medication treatment for OUD, risks of detoxification alone, and support; availability is determined by the local system.
Epictetus. Enchiridion, §§1–2, Elizabeth Carter translation. Opening sections read in context; an ethical frame for available action, not treatment advice.
Spinoza. Ethics, II, Proposition 7 and note; preface to III, R. H. M. Elwes translation. The specified passages were checked, not the entire work. A metaphysical source for explanation and the limits of condemnation.
Gospel according to John. 9:1–7, King James Version. The complete specified passage was read; a religious comparison concerning blame and suffering.

About the authors

Andrey Zarinsh is the author and founder of United Wisdom. He explores inner transformation, unity across traditions, intercultural dialogue, and contemporary spiritual discipline. Instagram: @andrey.zarinsh.

Maria Evilkut is a co-author of United Wisdom. Alongside her research, she seeks to test these ideas in practice and observe how principles of unity and mutual understanding work in real life between people and cultures. Instagram: @ilkutki.

MHR-088 r01, 25 October 2026.