What a medicine does, and what it does to this person
Nurses give almost every medicine a patient receives, which makes them the last point at which a wrong one can be stopped. This subject is what turns that from a checking task into a clinical one: knowing what a drug is for lets you notice when it is not working, and knowing what it does to the body lets you recognise harm before the patient can describe it.
A nurse who can describe exactly how a medicine works and who draws up the wrong strength has not made a pharmacology error; they have made an administration error, and the patient is harmed by the second failure however complete the first body of knowledge was. The two are taught apart because they fail apart. What this subject adds is the ability to notice: to see that a medicine is not working, or that something happening to the patient is being caused by what you gave them.
Medicines are prescribed, dispensed and administered, and a nurse stands at the administration stage — the last point at which an error from any earlier stage can still be caught. A prescribing error that a pharmacist misses and a nurse also misses reaches the patient; the same error caught by the nurse reaches nobody. Understanding what a drug is for is what makes that catch possible, because a nurse who does not know why a medicine is being given cannot know when it should not be.
The World Health Organization treats medication-related harm as a major and largely avoidable source of injury worldwide, serious enough to be the subject of a global patient safety initiative. The harms concentrate in a few predictable places: transitions between settings, high-risk medicines, and patients taking many things at once. Knowing that the risk is concentrated rather than evenly spread is what lets a nurse pay the most attention where it does the most good.
Not every drug. Classes: what this group of medicines does, what it is used for, what it does to the body that you can observe, what the common and the dangerous adverse effects are, what it interacts with, and what a nurse monitors. A student who learns fifteen classes properly can reason about a new agent inside one of them; a student who memorises agents cannot reason about anything.
How a medicine gets into the bloodstream, and how much of it arrives. The oral route is convenient and unreliable — food, gut contents, vomiting, diarrhoea and the liver's first pass all change what actually reaches the circulation. Injected routes bypass that and are therefore faster and less forgiving. For a nurse this explains why a medicine is prescribed with or without food, why a vomiting patient may need a different route, and why the route is part of the prescription rather than a convenience.
Where the drug goes once it is in. Some stay mostly in the blood, some spread into tissue, some into fat, and some cross into the brain or the placenta or breast milk. This is why body composition matters: an older person with less water and more fat, a dehydrated patient, a very thin patient and a pregnant woman all handle the same medicine differently, and why a dose right for one adult is not automatically right for another.
Mostly the liver, changing a drug into something that can be removed — sometimes deactivating it, sometimes activating it. A patient whose liver is failing clears more slowly and accumulates. So does a patient taking something that slows the same pathway, which is where a great many interactions come from. A nurse noticing that a patient has become drowsy or unsteady since something was started has spotted accumulation without needing to name the enzyme.
Mostly the kidneys. A patient with failing kidneys accumulates what a healthy patient clears, and this is the commonest reason a dose that was correct last year is not correct now. Reduced urine output is therefore a pharmacological observation as well as a circulatory one, and flagging it in a patient on medicines that the kidneys remove is a nursing contribution with real consequences.
None of it is examined for its own sake. It is examined because it explains the four questions that actually arise at a bedside: why this route, why this patient is different, why these two together are a problem, and why the patient has changed since yesterday. A student who can answer those from principles will out-perform one who has memorised warnings.
Most medicines work by fitting somewhere the body already uses — switching something on, or occupying the place so the body's own signal cannot. That single idea explains most drug names and most effects: a blocker of a system that speeds the heart will slow it, and stopping that blocker suddenly can let the system rebound. Knowing whether a drug pushes or blocks tells you what to expect and what withdrawal will look like.
For most medicines there is a range that helps and a range that harms, and for a few the gap between them is very small. Those few — anticoagulants, insulin, some heart medicines, some anticonvulsants, some antibiotics that damage kidneys and hearing — are the ones where checking matters most and where monitoring is built into the prescription. Knowing which medicines have a narrow window is more useful than knowing any individual figure.
A side effect is an expected consequence of how the drug works — drowsiness from something that sedates. An adverse effect is harm. An allergy is an immune reaction and is different in kind: rash, swelling of the face or mouth, wheeze, collapse. Patients routinely describe nausea as an allergy, and a record that says allergic to a whole class because somebody felt sick once can deny them an important treatment for life. Asking what actually happened is a clinical act.
Some medicines produce a diminishing effect over time, some produce physical dependence, and stopping several classes abruptly is dangerous — sedatives, some heart medicines, steroids and antiepileptics among them. A patient admitted for something unrelated who quietly misses days of a long-term medicine can deteriorate from the omission rather than the illness, and omitted doses are one of the commonest and least investigated medication errors there is.
The patient identified unambiguously, the drug, the strength, the dose, the route, the frequency, the start date, any stop date, the prescriber's identity, and the indication where the service requires it. Anything missing is a prescription to query rather than interpret. A nurse who fills in a gap from assumption has quietly become the prescriber without the authority or the information.
A trailing zero after a decimal point reads as ten times the amount if the point is missed. A decimal with no leading zero is read as a whole number. Abbreviations for units and routes are misread. Two drugs whose names look alike sit next to each other on a shelf. These are not carelessness; they are predictable failures of reading under pressure, which is why services ban the notations rather than asking people to be careful.
If it looks wrong, it is queried before it is given, and the query goes to the prescriber. This is a professional obligation everywhere and is the behaviour students find hardest to perform in front of a senior. Having a neutral form of words ready helps: I want to check this before I give it. Giving something you do not understand is never defensible, however confident the person who wrote it appeared.
A dose not given is recorded as not given, with the reason. Patients are harmed by medicines that were prescribed and never reached them — because they were off the ward, asleep, nil by mouth, vomiting, or the supply had run out — and because a blank on a chart looks like an oversight rather than a decision, nobody investigates. Writing the reason turns an omission into information.
The UK's competence test devotes an entire paper to numeracy, passed independently at a high threshold, because a nurse who is excellent clinically and unreliable arithmetically is dangerous in a specific and repeatable way. Calculation errors do not produce small mistakes; they produce ten-fold ones, and they concentrate in exactly the high-risk medicines where a ten-fold error is fatal.
Every calculation in practice reduces to the same shape: what has been prescribed, what is available, and therefore how much of what is available to use. Setting it out that way — in writing, with the units carried through every line — catches most errors, because an answer whose units are wrong is visibly wrong before the arithmetic is checked.
Before trusting any answer, ask whether it is plausible. An answer requiring a very large volume of an injection, or many whole tablets at once, or a fraction of a tablet that cannot be divided, is almost always an error rather than an unusual prescription. Nurses who calculate well are not faster at arithmetic; they are more suspicious of their own answers.
Two people doing the calculation separately and then comparing finds errors; two people watching one person do it does not, because the second person anchors on the first person's answer. Where your service requires a check for high-risk medicines, doing it independently is what makes it worth the time. A check performed as a signature is a check that has not happened.
This is a perishable skill. A nurse who last did a calculation under pressure two years ago is slower and less accurate and will not feel either. Short, frequent practice holds it far better than a revision session before an examination, and asking for it in a workplace is a reasonable professional request.
Antibacterials, antivirals, antifungals and antiparasitics act on something the organism has and we do not — a cell wall, an enzyme, a replication step. That is why they can be given at all, and why the ones with the smallest difference between the organism and us are the ones with the most toxicity. Knowing this predicts which agents need monitoring of kidneys or hearing.
Stopping early is one of the commonest patient behaviours and one of the most consequential. Explaining why the length is what it is, rather than instructing, changes adherence more than repetition does. Where a shorter course is now recommended for a condition, that is a change in evidence rather than a concession, and a nurse should be teaching what their own current guidance says.
Antimicrobial resistance is among the major global health threats, and it is driven by how much and how carelessly these medicines are used. The nursing contributions are concrete: taking cultures before treatment starts where that is the plan, giving doses on time because intervals are part of the effect, noticing when a patient is still on a broad agent days after a result came back, asking whether the intravenous route is still needed, and explaining to families why an antibiotic is not indicated for this illness.
Whether the patient is actually improving, and whether the treatment is harming them: rash and other allergic features, diarrhoea including the severe kind that follows disruption of gut organisms, kidney function, hearing where relevant, and interactions with what else they take. A patient who is not improving by the time expected is a finding to raise rather than to wait out.
Pain relief is generally stepped: simple agents first, stronger ones added rather than substituted, and adjuncts used for pain that does not respond to ordinary analgesia — nerve pain being the usual example. This is why a patient on a strong opioid is often still prescribed a simple analgesic; they work differently and the combination allows less of the stronger one. Students frequently assume the stronger agent replaces the weaker.
They relieve pain and they also sedate, slow breathing, cause constipation, cause nausea and, over time, tolerance and dependence. The observation that matters most is the level of sedation together with the respiratory rate, because sedation increases before breathing slows — which means a rousable but increasingly drowsy patient is the warning, not the falling rate. Constipation should be anticipated and prevented rather than treated after it becomes a problem.
Analgesia given after a patient has failed to get out of bed has been given at the wrong time. Pain relief timed so that it is working when movement, dressing changes or physiotherapy happen achieves far more from the same prescription, and arranging that is nursing judgement rather than a prescribing decision.
Infants, people with dementia, people who cannot describe pain in the staff's language, people with a history of substance use, and people whose reports are disbelieved. Under-treatment of pain in these groups is documented and consistent. Behavioural signs — grimacing, guarding, restlessness, resisting care, or a patient who has gone quiet and still — are assessed where a report is not available, and withdrawal is a recognised pain response that is routinely mistaken for settling.
Several classes do it by different routes — relaxing vessels, removing fluid, slowing the heart, blocking hormone systems that constrict vessels. What they share is the nursing consequence: the patient may become dizzy on standing, may fall, and may feel worse before they feel better. Measuring blood pressure lying and standing in anybody newly started or recently changed is the observation that finds this.
Medicines that slow the heart are checked against the pulse before they are given, and a service will usually specify a rate below which the dose is withheld and the prescriber informed. This is one of the clearest examples of a nurse's assessment gating an administration, and it is also one of the most frequently skipped when a round is running late.
Diuretics relieve breathlessness and swelling and remove electrolytes along with the water, which is why potassium and kidney function are monitored. They also make a patient need the toilet urgently, which in an older person who cannot get there quickly is a direct cause of falls and of the indignity that makes people stop taking them. Timing them so they do not act overnight is a small nursing decision with a large effect on adherence.
These prevent strokes and clots and they cause bleeding, and the gap between the two is narrow. A patient on one of them who falls, who bleeds from anywhere, who bruises easily or who is having a procedure needs that flagged. Monitoring requirements differ between agents, and missed or doubled doses matter more here than almost anywhere else.
Several cardiovascular medicines cause a rebound worse than the original problem when stopped abruptly, which makes an admitted patient's missed doses a real risk rather than an administrative gap. A patient who says they have not had their tablets for two days is reporting a clinical event.
It has a narrow margin, it is easy to mis-measure, its name and its strengths look alike, and the consequence of an error is immediate. Every service has specific rules about who may give it, what device is used, and how it is checked, and those rules exist because of real deaths. A nurse who treats insulin like any other prescription has misunderstood the single most important fact about it.
Insulin and several oral agents lower blood sugar whether or not the patient has eaten. A patient who is fasted for a procedure, who is vomiting, who has refused a meal or whose meal did not arrive is at risk from a treatment that was correct an hour earlier. Noticing and flagging that mismatch is a nursing action that prevents hypoglycaemia, and it is missed constantly on busy wards.
Confusion, aggression, sweating, tremor, slurred speech or reduced consciousness, frequently mistaken for intoxication or a behavioural problem. It is rapidly reversible and rapidly harmful. A conscious patient able to swallow safely is given fast-acting carbohydrate; an unconscious one is given nothing by mouth and managed as an emergency, following the protocol where you work.
They suppress inflammation and immunity and they do a great deal else: raise blood sugar, thin skin, weaken bone, alter mood and sleep, mask the signs of infection, and suppress the body's own production so that stopping abruptly or becoming acutely unwell on long-term treatment is dangerous. A patient on long-term steroids who becomes unwell needs urgent escalation, and this is a pattern every nurse should recognise.
A very large proportion of patients use inhalers incorrectly, which means the medicine never reaches the lung and the treatment is judged ineffective. Watching a patient use their own device, rather than asking whether they can, is the assessment. Teaching technique and spacer use is one of the highest-value nursing interventions in respiratory care and prevents more admissions than anything done during one.
They work by keeping a steady concentration, which makes timing and never missing doses central rather than incidental. Missed doses are a common cause of seizures in hospital, in patients admitted for something else entirely. Several interact widely with other medicines, and several are unsafe in pregnancy, which makes a woman of childbearing age on one of them a specific conversation rather than a routine prescription.
They have substantial physical effects — weight, blood sugar, movement disorders, effects on the heart's rhythm, and sedation — which is why physical health monitoring is core mental health nursing rather than an addition to it. People with severe mental illness die substantially younger, mostly of physical disease, and part of that gap is treatment effects that nobody monitored.
Sedatives, some antihistamines, some bladder medicines and several others cause confusion, falls and worsening function in older people, and the risk rises sharply when several are taken together. A nurse who notices that a patient has become confused or unsteady since a medicine was started has made a clinically important observation, and saying so may be what prompts a review that nobody else was going to request.
Each medicine may be justified and the combination may still be harmful. Interactions multiply, side effects are mistaken for new conditions and treated with further medicines, and the practical burden of taking a dozen things at different times of day is one most people cannot sustain. This is a recognised clinical problem rather than an inevitability, and review is the treatment.
A medicine causes a side effect, the side effect is treated as a new problem with a second medicine, and that one causes a third. Recognising it requires somebody to ask when the symptom started relative to when the medicine started — a question nurses are well placed to ask because they are the ones taking the history at admission and watching day to day.
An accurate list of what the patient actually takes, which is frequently different from what is prescribed. Noticing changes after a start or a change. Asking what gets in the way of taking them. Flagging medicines that may no longer be needed. Documenting side effects clearly enough that a prescriber can act on the record rather than on a conversation nobody wrote down.
Admission, transfer and discharge are where medicines get lost, duplicated and changed without anybody intending it, and medication-related harm concentrates at exactly these points. Comparing what the patient was taking with what is now prescribed, and resolving every difference deliberately, is the single most effective safety activity in this whole subject.
Many medicines cross to the fetus or into milk, and decisions are made by weighing the risk of the medicine against the risk of leaving the condition untreated — stopping a necessary treatment is not automatically the safe option and frequently is not. What belongs to nursing is asking every woman of childbearing age about the possibility of pregnancy before certain medicines, and flagging it rather than assuming somebody else has.
Almost everything is calculated by weight, which means an accurate current weight is a clinical measurement rather than an administrative one, and an estimate introduces error into every calculation that follows. Formulations, concentrations and routes differ, and several medicines safe in adults are not used in children at all. This is the area where checking culture is strictest for good reason.
Slower clearance, less water and more fat, more medicines at once, and greater sensitivity to anything sedating. The practical rule taught everywhere is that treatment is started low and increased slowly, and the nursing consequence is that an older person who becomes drowsy, confused or unsteady after a change is having a drug effect until proven otherwise.
These patients accumulate what others clear, and they are also the patients most likely to be on several medicines. A rising creatinine or a falling urine output in somebody on medicines the kidneys remove is a pharmacological warning as much as a renal one, and raising it is within any nurse's scope.
Where the prescribed agent is unavailable, substituting within a class is a prescribing decision and not a nursing one, however obvious it seems. What a nurse does is establish what is actually available, tell the prescriber, and document the gap. Quietly giving something similar is how a patient receives a medicine nobody has checked against their allergies or their other treatment.
Heat, light and humidity degrade medicines, and a medicine that has degraded looks exactly like one that has not. Where there is no reliable cold chain or air conditioning, knowing which of your stock is vulnerable, storing it as well as circumstances allow, and reporting the conditions honestly are part of the job. A treatment failure caused by degraded stock is usually recorded as resistance or non-adherence.
These are a genuine and documented problem in many markets, and they cause treatment failure and death. Buying through recognised supply chains, being alert to packaging that looks wrong, and reporting suspicion through the proper route are what a health worker can do. Patients who bought a medicine informally should be asked about it without judgement, because they will not volunteer it.
If somebody will not be seen again, what they leave knowing is the whole of the treatment. How to take it, for how long, what to do about a missed dose, what side effects to expect and which ones mean come back. Asking them to say it back, in their own words, is the only check that takes a minute and works.
Anything new since a medicine started, until proven otherwise. Rash, fever, breathlessness, swelling of the face or mouth, collapse, bleeding, confusion, a change in bowel habit, a change in urine output. The question that finds most of them is simple and rarely asked: what has changed, and what started recently?
Rapid onset after an exposure with airway, breathing or circulatory compromise, usually with skin changes. Stop the trigger, call for emergency help, position appropriately — flat with legs raised where circulation is compromised, sitting up where breathing is the problem, never stood up suddenly — and assist. Treatment is prescribed and appears nowhere in this manual; what a nurse contributes is speed and position.
Both to your own incident system and, where one exists, to the national pharmacovigilance scheme. Adverse reactions are detected after licensing largely because health workers report them, and nurses report far less than they encounter. A reaction that is managed and not reported is one that will happen to somebody else.
Make the patient safe, tell somebody senior immediately, document factually, complete the incident report, and accept support. Waiting to see whether the patient is affected is the temptation to resist, because the window in which harm can still be prevented is short. Services that respond by finding an individual to blame get fewer reports rather than fewer errors, and the harm continues unmeasured.
The World Health Organization maintains a model list of medicines that meet the priority health needs of a population, selected on need, evidence and comparative cost, and most countries build their own national list from it. Understanding that a formulary is a deliberate selection rather than everything that exists explains why a familiar agent may be unavailable and why substitution is a decision rather than a shrug.
Cultures before treatment where planned. Doses on time. Noticing a broad agent still running after a result returned. Asking whether the intravenous route is still needed. Documenting indication and review date. Explaining honestly to families why an antibiotic is not indicated. None of these are prescribing and all of them change outcomes.
What it is for, what to expect, what to watch for, what to do about a missed dose, and what would mean stop and seek help. Patients who understand their medicines take them; patients handed a bag and a leaflet frequently do not. This conversation is the most-skipped and one of the most consequential parts of a discharge.
That knowing what a drug is for is what lets you notice when it is not working. That the risk concentrates in a few places — transitions, high-risk medicines, many medicines at once. That anything new since a medicine started is that medicine's fault until proven otherwise. That omissions are errors. And that querying a prescription you do not understand is an obligation rather than an impertinence.
Oral, sublingual, rectal, topical, inhaled, injected under the skin, into muscle, into a vein, into the spinal space. Each changes how fast the medicine arrives, how much of it arrives, and how forgiving an error will be. A route is never changed for convenience — crushing a tablet that was designed to release slowly delivers the whole of it at once, and giving something by a route it was not prepared for has killed people.
Some tablets are built to release their contents over many hours. Crushing one destroys that and delivers the entire amount immediately, which for several medicines is a lethal overdose. A patient who cannot swallow needs a different preparation or a different route, decided by a prescriber and a pharmacist, not a nurse with a tablet crusher and a time pressure.
This is a specific competency with real hazards: tube position confirmed before anything is given, the right formulation used, medicines given separately with flushes between them rather than mixed together, and the interaction between feed and certain medicines respected. Blocked tubes are usually caused by crushed tablets, and a blocked tube means a patient who stops receiving both nutrition and treatment.
Medicines applied to skin, eyes, nose or lung are absorbed, and in some cases enough to produce whole-body effects — particularly in infants, in older people with thin skin, and where a preparation is applied over a large area or under a dressing. Treating these as harmless because they are local is a common misunderstanding with real consequences.
They are prescribed, checked, given and recorded like anything else, with the added requirement that they have been stored within their temperature range for their whole life. A vaccine that has been too warm or has been frozen may have lost potency, and the patient walks away believing they are protected when they are not — which is worse than not vaccinating them, because nobody will check again. Cold chain records are clinical records.
It has indications, a target, an endpoint and its own risks, and in a subset of patients with chronic respiratory disease sustained high concentrations are harmful. Treating it as a comfort measure to be turned up because somebody looks anxious, or left running because nobody reviewed it, is exactly the behaviour that distinguishes a medicine from a piece of furniture. Check that the device is actually on the patient every time you pass.
Anything given into a vein acts fast and cannot be taken back, and a small number of routes — into the spinal space in particular — have been the site of catastrophic wrong-route errors that changed practice worldwide. Where your service has specific rules about who may give what by which route, those rules are written in consequence of real deaths.
Some medicines are controlled because of their potential for dependence and diversion, and separately some are designated high-alert because an error with them causes disproportionate harm. The two overlap but are not the same thing, and a nurse needs to know which category a medicine sits in, because the rules that follow are different.
Controlled medicines are stored securely, recorded in a register with running balances, and usually checked and signed by two people at administration and at disposal. These rules feel bureaucratic until a discrepancy appears, at which point the register is the only evidence of what happened. A register signed without the count being done is worse than no register, because it manufactures false assurance.
What is not given is disposed of according to policy, witnessed, and recorded. A discrepancy is reported immediately rather than resolved quietly, however embarrassing, because the alternative is a pattern that grows. Diversion by colleagues happens, it is usually driven by dependence rather than malice, and the professional response is to report it so the person is helped and the patients are protected.
Anticoagulants, insulin, opioids, concentrated electrolytes, sedatives, chemotherapy and certain cardiac medicines recur on high-alert lists worldwide. Knowing your own service's list, and knowing that it exists at all, changes how much checking a nurse does without being told. This is one of the few places where slowing down is unambiguously the safer choice.
Most give a patient on a named class and ask what the nurse assesses, watches for, or does next. They are answerable from class knowledge without knowing the individual agent, which is why this manual is organised by class. Options that adjust a dose, substitute an agent or withhold without informing anybody are outside nursing scope and are there to be rejected.
Check the pulse before a heart-rate-lowering medicine. Check sedation and respiratory rate on opioids, and know that sedation comes first. Match insulin to eating. Watch potassium and kidneys on diuretics. Anything new since a medicine started is a drug effect until excluded. Stopping several classes abruptly is dangerous. Reconcile at every transition.
Practise it under time pressure, in writing, with the units carried through, and sense-check every answer before trusting it. Where your examination has a separate numeracy paper with its own pass mark, treat it as a separate examination, because that is exactly what it is.
Your own formulary, your own scope and your own service's checking rules. This manual is deliberately free of agents and quantities so that it is true everywhere; the specifics that make you safe on a particular ward come from that ward, and you should have read them before your first round rather than during it. The single habit most worth forming from this whole subject is the one that costs nothing: never give a medicine you cannot say the purpose of, and ask when you cannot. Every serious medication error that has ever been reviewed contained a moment where somebody could have asked and did not, and the reason they did not was almost never ignorance — it was that asking felt like admitting something, in front of somebody who seemed certain.