Mental Health and Behavioural Medicine
Assessment
Use a structured approach: risk (suicide, self-harm, harm to others), symptoms, function, substance use, social context. Know PHQ-9/GAD-7 as screening aids, not diagnoses. Understand mental state examination domains.
Common disorders
- Depression: core symptoms, duration ≥2 weeks, severity, stepped care (psychological therapies, SSRIs).
- Anxiety disorders: GAD, panic, social anxiety, OCD, PTSD.
- Bipolar disorder: mania criteria, mood stabilisers, avoid antidepressants alone.
- Psychosis: first-episode referral, antipsychotics, metabolic monitoring.
- Substance misuse: brief interventions, withdrawal management.
- Eating disorders: red flags (bradycardia, hypokalaemia, rapid weight loss).
Suicide risk
Ask directly. Risk factors: male, older, previous attempt, access to means, isolation, substance use, hopelessness. Protective factors: social support, children, religious beliefs. Document risk assessment and safety plan.
Traps
- Prescribing an SSRI alone in bipolar disorder (risk of mania).
- Missing organic causes of psychiatric symptoms (hypothyroidism, delirium, drugs).
- Failing to ask about suicidal ideation for fear of 'planting' the idea.
- Ignoring postpartum psychosis - a psychiatric emergency.
Worked rule
A young adult with low mood and no risk: offer psychological therapy first for mild–moderate depression; SSRIs for moderate–severe or if therapy unavailable, with review at 2 weeks for suicidality.
Musculoskeletal, Skin and Eye Disorders
Musculoskeletal
Distinguish inflammatory (morning stiffness >1 hour, improvement with use) from mechanical pain. Know osteoarthritis (clinical diagnosis, exercise, weight loss, analgesia), rheumatoid arthritis (symmetrical small joints, anti-CCP, early DMARDs), gout (monosodium urate crystals, acute treatment, urate-lowering), polymyalgia rheumatica (age >50, shoulder/hip girdle, ESR, steroids), and back pain (red flags: age, trauma, cancer history, neurological deficit, systemic symptoms).
Skin
Common conditions: eczema (emollients, topical steroids), psoriasis (plaque, systemic associations), acne, fungal infections, cellulitis (marking the border, systemic signs), and skin cancer (ABCDE, urgent referral).
Eye
- Red eye: conjunctivitis vs iritis vs acute angle-closure glaucoma (painful, fixed pupil, halos - emergency) vs keratitis (contact lens).
- Vision loss: sudden painless (retinal artery occlusion, vitreous haemorrhage) vs painful (angle closure, uveitis).
- Diabetic retinopathy screening.
Traps
- Treating acute angle-closure glaucoma as conjunctivitis - a sight-threatening error.
- Missing cauda equina syndrome (saddle anaesthesia, urinary retention).
- Delaying DMARDs in RA.
- Using topical steroids on suspected fungal or herpetic lesions.
Worked rule
Painful red eye with halos and a mid-dilated fixed pupil: refer immediately for acute angle-closure glaucoma; do not prescribe drops and send home.
Geriatrics
The geriatric approach
Use the comprehensive geriatric assessment (CGA): medical, functional, cognitive, psychological, social. Know the geriatric giants: falls, immobility, incontinence, intellectual impairment (delirium/dementia), and iatrogenic harm.
Key syndromes
- Delirium: acute onset, fluctuating, inattention; causes (infection, drugs, pain, constipation, retention, environment); manage the cause and environment; avoid antipsychotics unless necessary.
- Dementia: progressive cognitive decline; assess with MMSE/MoCA; differentiate types (Alzheimer, vascular, Lewy body - fluctuating cognition, visual hallucinations, parkinsonism).
- Falls: multifactorial assessment; medication review; vitamin D; exercise.
- Polypharmacy: STOPP/START criteria; deprescribing.
- Incontinence: stress vs urge vs overflow; conservative management first.
Medication safety
Physiological changes increase drug sensitivity. Avoid anticholinergics, benzodiazepines, and NSAIDs where possible. Use the Beers criteria conceptually.
Traps
- Diagnosing dementia when the patient has delirium (reversible, acute).
- Missing constipation or urinary retention as a cause of delirium.
- Ignoring drug interactions and renal dosing.
- Attributing weight loss or falls to 'old age' without assessment.
Worked rule
An older patient with new confusion: screen for infection, drugs, pain, constipation, retention, and environment (the 'PINCH ME' approach) before labelling it dementia.
Palliative Care and Chronic Disease Management
Palliative care principles
Palliative care is person-centred care for serious illness, not only end-of-life care. Know total pain (physical, psychological, social, spiritual), advance care planning, and goals-of-care conversations. Use the surprise question and prognostic tools cautiously.
Symptom control
- Pain: WHO analgesic ladder; regular review; opioids for moderate–severe pain; anticipate adverse effects (constipation, nausea); opioid conversion and breakthrough dosing.
- Dyspnoea: opioids, oxygen if hypoxic, fan, anxiety management.
- Nausea: identify cause; antiemetics by mechanism.
- Terminal phase: anticipatory prescribing, syringe drivers, discontinuation of non-essential drugs.
Chronic disease management
Use the chronic care model: self-management support, delivery system design, decision support, clinical information systems, community resources. Apply motivational interviewing and care plans. Understand multimorbidity and polypharmacy.
Traps
- Withholding opioids for fear of addiction in cancer pain.
- Failing to prescribe laxatives with opioids.
- Continuing futile or burdensome treatments without goals-of-care discussion.
- Missing depression in chronic illness.
Worked rule
A dying patient with agitation and noisy breathing: consider causes (pain, retention, constipation), use anticipatory medication (e.g., midazolam, glycopyrronium), and stop non-essential drugs; comfort is the outcome measure.
Practice Management and Quality
Practice management
Understand appointment systems, access and continuity, team roles (nurse, pharmacist, allied health), record keeping, and medicolegal documentation. Know audit cycle (standard → measure → compare → change → re-audit) versus research (generates new knowledge).
Quality improvement
Use PDSA cycles, root cause analysis, significant event analysis, and clinical governance pillars (education, audit, risk management, patient involvement, information). Know key performance indicators and patient safety concepts (human factors, Swiss cheese model).
Evidence-based medicine
Formulate a PICO question; know levels of evidence (systematic reviews > RCTs > cohort > case-control > case series); understand bias, confounding, randomisation, blinding, and intention-to-treat. Interpret confidence intervals, p-values, NNT, and likelihood ratios.
Traps
- Calling an audit 'research' or vice versa.
- Interpreting a p-value as the probability the null hypothesis is true.
- Ignoring confounders in observational studies.
- Confusing efficacy (trial conditions) with effectiveness (real-world).
Worked rule
A practice notices high no-show rates. Run an audit, identify causes (reminders, transport, scheduling), implement a PDSA change, and re-audit. That is quality improvement, not research.
Research
Know study designs, sampling, validity (internal/external), and ethics (consent, ethics committee).
Evidence-Based Medicine and Research
Asking and answering
Convert a clinical dilemma into a PICO question (Population, Intervention, Comparator, Outcome). Search efficiently: primary sources (PubMed, Cochrane) and secondary sources (guidelines, UpToDate). Appraise for validity, importance, applicability.
Study designs
- RCT: randomisation controls confounding; blinding reduces bias; analyse by intention-to-treat.
- Cohort: incidence, relative risk; susceptible to confounding and loss to follow-up.
- Case-control: rare outcomes; odds ratio; recall bias.
- Cross-sectional: prevalence; cannot establish temporality.
- Diagnostic studies: sensitivity, specificity, likelihood ratios, ROC curves.
Statistics
Know type I error (false positive) and type II error (false negative, related to power), confidence intervals, standard deviation vs standard error, parametric vs non-parametric tests, and correlation vs causation. Understand absolute risk reduction, relative risk reduction, NNT, and number needed to harm.
Bias and critical appraisal
Recall selection, information, recall, publication, and observer bias; lead-time and length-time bias in screening; Hawthorne effect. Use CONSORT, PRISMA, STROBE as reporting standards.
Traps
- Reporting relative risk reduction without absolute numbers.
- Confusing sensitivity (true positive rate) with PPV (post-test probability).
- Assuming a statistically significant result is clinically important.
- Ignoring loss to follow-up in cohort studies.
Worked rule
A drug reduces events from 10% to 8%: ARR = 2%, RRR = 20%, NNT = 50. Quote all three - the RRR alone exaggerates benefit.