01Purpose and principlesWhat the treatment does and how it fits into care.
Antimicrobial prescribing combines a diagnostic hypothesis with a plan to test it. The initial hypothesis identifies the likely site, pathogens and severity; the prescription supplies effective exposure while balancing adverse effects and resistance. The hypothesis may be wrong or become more precise as cultures, imaging and clinical response arrive. Good stewardship therefore includes both appropriate early treatment and willingness to stop treatment when infection is unsupported. Broad coverage can be necessary in severe illness, but breadth is not a substitute for assessing the source or obtaining information that allows a more focused decision.
The presence of an organism does not itself establish a treatable infection. A sample may reflect contamination, colonisation or disease, depending on where it came from and the patient’s symptoms. Conversely, a negative culture does not exclude infection after antibiotics or with a poorly collected specimen. Regimens must also fit the anatomical site: a medicine that achieves useful urinary exposure may not treat renal tissue infection. Duration belongs to the particular syndrome and response, not to a habit of prescribing a standard week for every illness. National recommendations, local resistance patterns and product-specific administration information have complementary roles.
Key points
- Establish the infection site and severity before choosing an antimicrobial or interpreting a positive sample.
- Check allergy phenotype, pregnancy possibility, recent antibiotics, previous susceptibility results, renal function and interactions.
- Use an infection-specific regimen with route, dose, interval and total course documented from the start.
- For suitable non-pregnant adult women with lower UTI and eGFR at least 45, NICE recommends nitrofurantoin modified-release 100 mg orally twice daily for three days.
- Arrange an antimicrobial review using clinical response and microbiology; intravenous prescriptions should be considered for review at 48–72 hours.
- Stop, narrow, switch route or define further treatment at review; continuing by default is an avoidable prescribing decision.
02Indications, selection and cautionsWho may benefit, who needs urgent treatment and important alternatives.
Distinguish a probable bacterial syndrome from a self-limiting illness and assess how safely uncertainty can be observed. A stable patient may benefit from advice and reassessment, whereas physiological instability requires urgent action. Record the evidence supporting infection rather than only the symptom that led the person to seek care.
Consider whether the medicine reaches the relevant tissue and whether drainage, removal of an infected device or another source-control intervention is needed. Repeatedly broadening antibiotics will not reliably correct an undrained collection. Persistent illness should trigger re-examination of anatomy and diagnosis as well as susceptibility.
Pregnancy, immunosuppression, impaired kidney function, previous resistant organisms and recent healthcare exposure may change the choice or route. Ask about genuine allergy features and important interactions. A label such as penicillin allergy should prompt a focused history; do not erase a serious reaction because the preferred regimen contains a beta-lactam.
Diarrhoea, rash, new breathlessness, jaundice or tendon pain after starting an antimicrobial may represent harm from treatment. Separate adverse effects from progression of infection while assessing severity. Adding another antibiotic for every new symptom can obscure the original problem and increase exposure without benefit.
03Assessment before treatmentTests and checks that guide safe selection.
Read from the initial assessment onwards. Tests may run in parallel in urgent care; first-line, preferred, confirmatory, definitive and gold-standard labels appear only when the chapter explicitly states them.
- 01
Focused clinical assessmentFirst step - Why
- Define the suspected syndrome and identify immediate physiological risk.
- Interpretation and limitations
- Use history, examination and observations to decide the site and severity. Include symptom duration, hydration and ability to take oral medicines. An inflammatory marker alone cannot choose an antibiotic; the clinical syndrome supplies the question that laboratory and imaging results help answer.
- 02
Microbiological sampling - Why
- Obtain information that could change antimicrobial selection or duration.
- Interpretation and limitations
- Collect the appropriate specimen before treatment when indicated and feasible without harmful delay. Document recent antibiotics and the clinical question. Interpret growth alongside collection quality and symptoms, and identify who will review the result after the patient leaves the setting.
- 03
Renal function and exposure assessment - Why
- Adjust selection and dose to drug handling and prior treatment.
- Interpretation and limitations
- Check current kidney function and use the measure specified for the selected medicine. Review previous antibiotic courses, organism susceptibilities and hospital exposure. A positive historical result is relevant but does not prove the current illness has the same cause or susceptibility.
- 04
Response and source-control assessment - Why
- Determine why apparent treatment failure is occurring.
- Interpretation and limitations
- Ask whether doses were taken and retained, whether the chosen drug covers the plausible organism, whether a collection or obstruction persists and whether the diagnosis is correct. Repeat tests selectively to answer these questions. Failure to improve is not automatically evidence that a broader drug is needed.
04Treatment approachPreparation, options, escalation and aftercare.
01Worked caseSelect a lower-UTI regimenFirst stepA 32-year-old non-pregnant woman has dysuria and frequency without vaginal symptoms, fever or flank pain. She is stable, can take oral medicine, has eGFR 82 mL/min/1.73 m² and no relevant allergy, G6PD deficiency or recent resistant isolate.+
- 1EscalationConfirm that the findings fit lower urinary infection and do not suggest a different diagnosis or upper-tract disease. Discuss immediate versus back-up treatment according to symptom burden, complication risk and preference; make the observation and escalation plan explicit.
- 2If immediate treatment is chosen, nitrofurantoin is a NICE first-choice option for this setting. Select the modified-release preparation and prescribe the complete three-day regimen shown in the prescribing section; the formulation determines the dosing interval.
- 3Explain administration and what to do if symptoms worsen or fail to begin improving within 48 hours. Fever, flank pain or systemic illness changes the clinical question and requires reassessment rather than automatic extension of the same bladder-directed antibiotic.
- 4If a culture has been sent, assign responsibility for reviewing susceptibility and deciding whether a change is required in light of symptoms. A laboratory result should be interpreted with clinical response, not used to trigger an unconsidered additional course.
- 5Verify that the quantity matches the intended course and that the patient understands the stop point. Document the syndrome, relevant exclusions and safety advice so any subsequent reviewer can see why this regimen was appropriate at the time.
02Early treatment decisionChoose no antibiotic, back-up treatment or immediate therapyA patient presents with a possible infection but the diagnosis and likely benefit from antibiotics require assessment.+
- 1Explain the likely course of the illness and the expected effect of antibiotics. For uncomplicated acute cough without systemic illness or higher complication risk, NICE advises against routine antibiotic treatment; provide practical symptom advice and reasons for reassessment.
- 2If a back-up prescription is appropriate, state the specific trigger for using it and the circumstances needing clinical review instead. A delayed prescription without clear instructions simply postpones uncertainty to the patient.
- 3When immediate antibiotics are indicated, use the current syndrome-specific recommendation and relevant local susceptibility information. State route, dose and duration, account for contraindications and interactions, and consider the need for sampling or source control.
- 4Use systemic fluoroquinolones only when other commonly recommended antibiotics are inappropriate, as required by MHRA restrictions. Record why alternatives cannot be used rather than selecting a fluoroquinolone for convenience or broad coverage alone.
03Antimicrobial reviewConvert new evidence into a treatment decisionAn inpatient has received empirical intravenous antibiotics and microbiological results or a clearer clinical picture are now available.+
- 1Reassess whether infection remains the best explanation. If evidence supports a non-infective diagnosis, stop unnecessary antibiotics and address the actual cause rather than finishing a course simply because it was started.
- 2If infection is confirmed, review susceptibility and source control, then narrow treatment where suitable. Broad initial coverage should not remain an unexplained default once the relevant organism and site are known.
- 3Consider intravenous review at 48–72 hours, including clinical response and microbiology. Assess whether oral treatment is reliable and reaches the required site; vomiting, impaired absorption or certain deep infections may prevent a straightforward route switch.
- 4Document the decision to stop, change, continue or switch route, together with the total intended duration. Count treatment already given and communicate the remaining course at discharge so a new prescription does not inadvertently restart the clock.
05Regimens, contraindications and interactionsTreatment details and the circumstances that modify them.
Nitrofurantoin for uncomplicated lower UTI in non-pregnant adult women
NICE NG109 regimen: 100 mg modified-release orally twice daily for 3 days when eGFR is at least 45; if that formulation is unavailable, 50 mg orally four times daily for 3 days.Take with food or milk. Exclude upper-tract infection; avoid G6PD deficiency, previous nitrofurantoin liver injury and relevant hypersensitivity. Pregnancy and eGFR below 45 require a different individual assessment; stop and investigate new respiratory or hepatic symptoms.
Doxycycline when antibiotics are indicated for adult acute cough
NICE NG120: 200 mg orally on day 1, then 100 mg orally once daily for 4 further days, giving a 5-day course in total.Avoid in pregnancy; review interactions and administration instructions, including swallowing with water while upright and separating relevant mineral-containing products according to product advice.
Amoxicillin as an acute-cough alternative
NICE NG120 alternative first-choice regimen for a suitable adult: 500 mg orally three times daily for 5 days, with review if deterioration or non-response occurs.Do not use with a relevant serious penicillin allergy; review kidney function and interactions. This regimen should not be transferred to a different infection without checking its specific guidance.
06Complications, monitoring and follow-upAdverse effects, response and longer-term review.
- Review clinical response using symptoms, function and observations appropriate to the site. Persistent fever, new instability or worsening focal signs should prompt earlier reassessment than a scheduled routine review.
- Act on microbiology through a named process. Confirm that a result requiring a treatment change reaches the patient and prescriber, particularly after discharge or a back-up prescription.
- Watch for treatment harms and give clear stop-and-seek-advice instructions for serious symptoms. Nitrofurantoin can cause pulmonary and hepatic reactions with short or prolonged exposure; new breathlessness should not be dismissed as unrelated.
- At completion, confirm whether further infection follow-up is required and remove unintended repeat authorisation. Recurrent symptoms need a new assessment rather than automatic reuse of leftover antibiotics.
07Special situationsVariants, exceptions and circumstances that change the usual approach.
A result needs a question
The same organism can be clinically meaningful in one specimen and a contaminant in another. Interpret the collection method, site and symptom pattern together. Laboratory susceptibility cannot by itself establish that treatment is needed.
Course length is cumulative
A patient discharged after several inpatient treatment days may need only the remainder of the planned course. Document start date and intended final dose to prevent an unnecessary additional full course at each transfer.
Oral can be effective
Intravenous treatment is a route, not a guarantee of greater clinical effectiveness. Once an appropriate oral drug can deliver reliable exposure for the syndrome, switching can reduce line-related burden and facilitate recovery outside hospital.
Stewardship includes urgent treatment
Avoiding unnecessary exposure and treating serious infection promptly are compatible duties. The skill is matching the speed and breadth of treatment to the patient’s risk while arranging early reassessment as uncertainty decreases.
08Common pitfallsFrequent interpretation and management errors.
- 01
Treating asymptomatic bacteriuria as symptomatic UTI in a non-pregnant adult can expose the person to harm without addressing the reason for their presentation.
- 02
Copying a duration from a different infection ignores differences in tissue penetration, source control and the evidence supporting course length.
- 03
Using a positive culture to explain every symptom may miss an alternative diagnosis or medicine-related complication.
- 04
Continuing broad intravenous therapy without a documented review leaves diagnosis, route and stop point unresolved.