The question hits most doctors somewhere between final-year medical school and FY2. Some people know from day one — the surgeon who was always going to be a surgeon. Most don't. Most arrive at a specialty through a combination of positive experiences, eliminated alternatives, and deadline pressure.
There's nothing wrong with that. But there is a better way to think about it than "I liked my cardiology rotation" — because liking a four-week rotation as a student is a poor predictor of liking a 30-year career.
Here's a framework that forces honest thinking.
The four dimensions that actually matter
1. Intellectual pattern
Every specialty has a cognitive signature — a dominant type of thinking it rewards.
Pattern recognition and rapid decision-making: Emergency medicine, acute medicine, anaesthetics. You see a presentation, match it to a pattern, and act. The satisfaction comes from speed and decisiveness.
Diagnostic puzzle-solving: Radiology, pathology, clinical genetics, neurology. You assemble data from multiple sources and synthesise a conclusion. The satisfaction comes from being right.
Longitudinal relationship management: General practice, psychiatry, palliative care, geriatrics. You manage people over time, navigating complexity and uncertainty. The satisfaction comes from continuity and trust.
Procedural mastery: Surgery (all subspecialties), interventional cardiology/radiology, ophthalmology. You develop a physical skill to a high level and deploy it under pressure. The satisfaction comes from craft.
Systems thinking: Public health, occupational medicine, medical management. You work at population level, designing interventions for groups rather than individuals. The satisfaction comes from scale.
Most doctors have a natural affinity for one or two of these patterns. The most common career dissatisfaction comes from choosing a specialty whose cognitive pattern doesn't match your wiring — the relationship-builder stuck in radiology, the puzzle-solver stuck in GP.
2. Lifestyle architecture
Be honest about what you want your life to look like. Not "work-life balance" in the abstract — the specific, concrete details.
On-call intensity: Surgery, emergency medicine, obstetrics, and acute medicine have the most demanding on-call commitments. GP, dermatology, radiology, pathology, and public health have the least.
Predictability of hours: Some specialties let you plan your week; others don't. Elective surgery is more predictable than emergency surgery. Outpatient-heavy specialties (rheumatology, endocrinology, dermatology) are more predictable than acute ones.
Part-time feasibility: GP, psychiatry, dermatology, and radiology accommodate less-than-full-time (LTFT) training and practice relatively well. Surgery and emergency medicine are harder to do part-time, though not impossible.
Geographic flexibility: GP, psychiatry, and acute medicine jobs exist everywhere. Highly subspecialised roles (paediatric cardiothoracic surgery, for instance) exist in a handful of centres. If you care about where you live, this matters.
3. Income trajectory
Doctors rarely talk about money openly, which leads to uninformed decisions. The broad picture in the UK:
Highest earning potential: GP partner, private-heavy surgical specialties (orthopaedics, plastics, ophthalmology), radiology (private reporting). These can exceed £150,000–£250,000 with a private practice component.
Solid middle: Most hospital consultant posts: £93,000–£126,000 NHS base salary (2024 scales) + on-call supplements + clinical excellence awards. Limited private practice depending on specialty and geography.
Lower end (but still comfortable): Public health, academic medicine, medical education, and pure NHS specialties without a private market.
The important nuance: training length varies dramatically. A GP completes training 5–7 years before an orthopaedic surgeon. That's 5–7 years of consultant-level earning the surgeon doesn't have. By the time lifetime earnings converge, the GP may have earned more in total despite a lower peak salary.
4. Competition and probability
The hardest conversation: can you realistically get into your chosen specialty?
Highly competitive (competition ratios 5:1+): Dermatology, plastics, ophthalmology, cardiothoracic surgery, neurosurgery, clinical radiology.
Competitive (3–5:1): Cardiology, gastroenterology, orthopaedics, ENT, general surgery.
Moderately competitive (1.5–3:1): Emergency medicine, anaesthetics, obstetrics, paediatrics, most medical subspecialties.
Less competitive (<1.5:1): General practice, psychiatry, geriatrics, public health, chemical pathology.
Competition ratios change year to year and vary by region. But the structural pattern is stable: specialties perceived as having good lifestyle, good income, and good prestige are oversubscribed. Specialties perceived as difficult, underfunded, or low-status are undersubscribed — and often offer the fastest career progression and greatest professional autonomy precisely because of the demand-supply imbalance.
The decision process
Step 1: Eliminate. Remove any specialty whose cognitive pattern fundamentally doesn't suit you. If you hate uncertainty, don't go into GP or emergency medicine. If you hate procedures, don't go into surgery. This usually removes half the options.
Step 2: Rank on lifestyle. Of the remaining options, which ones fit the life you want? Not the life you think you should want — the life you actually want. This usually reduces the list to 3–5 options.
Step 3: Reality-check on competition. Of your shortlist, which ones can you realistically access given your portfolio, exam scores, and geographic flexibility? If your top choice is dermatology but you have no research publications and no dermatology experience, you need either a plan to build that portfolio or a realistic alternative.
Step 4: Test it. Taster weeks, clinical attachments, and talking to registrars (not consultants — registrars are living the training reality right now) are the best way to validate your thinking before committing.
Step 5: Accept uncertainty. You will not feel 100% sure. Nobody does. The doctors who are happiest in their careers are not the ones who chose perfectly — they're the ones who committed to a reasonable choice and invested in making it work.
The most common mistakes
Choosing based on a single rotation. A great consultant makes any specialty look appealing. A terrible rotation makes any specialty look miserable. Neither is representative.
Choosing based on prestige. Prestige fades quickly when you're five years into a training programme you don't enjoy. The opinion of people at medical school dinners matters far less than how you feel on a Tuesday morning.
Not choosing. Delaying the decision by doing extra foundation years, research fellowships, or "keeping options open" indefinitely is itself a choice — and it has an opportunity cost. Perfect information will never arrive. Decide with the best information you have.
iatroX is built by a GP who chose the specialty deliberately — and built a clinical AI platform and exam qbanks to help the next generation make informed choices.