Membership of the Faculty of Public Health (MFPH) Health Economics, Decision Science and Health Information Flashcards

24 sample cards from the Health Economics, Decision Science and Health Information deck

Sampled from the end of the deck, so these are different cards from the ones shown on the syllabus page.

1. What does the decennial Census provide for public health?

   The Census provides a complete population count and demographic/socioeconomic data (age, sex, ethnicity, housing, occupation) every ten years. It is the denominator for many rates and underpins population estimates and resource allocation, though it becomes dated between censuses.

2. Define crude birth rate and crude death rate.

   Crude birth rate = live births in a year per 1,000 mid-year population. Crude death rate = deaths in a year per 1,000 mid-year population. Both are 'crude' because they are not adjusted for population age structure.

3. What is a population pyramid and what does its shape indicate?

   A population pyramid is a back-to-back bar chart of age-sex structure. A wide base/narrow top (expansive) indicates high birth rates and a young, growing population; a more rectangular/top-heavy (constrictive) shape indicates low birth rates and an ageing population.

4. Define the dependency ratio.

   $$\text{Dependency ratio} = \frac{\text{population aged }0\text{--}14 + \text{aged }65+}{\text{population aged }15\text{--}64} \times 100$$ It expresses the number of dependents per 100 people of working age and indicates the economic burden on the productive population.

5. What is the demographic transition?

   The demographic transition is the shift, as countries develop, from high birth and death rates to low birth and death rates. It typically passes through stages of falling mortality first (rapid growth) then falling fertility (stabilising), changing the age structure toward an older population.

6. List the main dimensions of data quality.

   Commonly cited dimensions: completeness, accuracy (validity), timeliness, relevance, consistency, accessibility, and reliability. (Some frameworks add validity, coverage, and granularity.) Each affects how fit-for-purpose a dataset is.

7. Distinguish validity from reliability as data quality concepts.

   Validity (accuracy) is whether the data measure what they are intended to measure — closeness to the true value. Reliability is the consistency/reproducibility of the measurement. Data can be reliable (consistent) yet invalid (consistently wrong).

8. What is information governance and what UK legislation underpins it?

   Information governance is the framework of policies, standards and processes ensuring personal health information is handled legally, securely, ethically and efficiently. In the UK it is underpinned by the Data Protection Act 2018, UK GDPR, the common law duty of confidentiality, and Caldicott principles.

9. What are the Caldicott Principles concerned with?

   The Caldicott Principles govern the use of confidential patient information — e.g. justify the purpose, use only when necessary, use the minimum necessary, access on a strict need-to-know basis, everyone aware of responsibilities, comply with the law, and the duty to share information can be as important as the duty to protect.

10. What is data linkage and a key challenge it poses?

    Data linkage joins records relating to the same individual across different datasets (e.g. linking hospital, cancer registry and mortality data) to create richer information. Key challenges include matching accuracy (false matches/missed matches) and protecting confidentiality/privacy.

11. What is a decision tree and how is an expected value calculated?

    A decision tree is a diagram mapping decisions (square nodes), chance events (circular nodes) and outcomes over a single time frame. The expected value of a branch is the sum of each outcome's value weighted by its probability: $$EV = \sum_{i} p_i \times v_i.$$

12. In decision-tree analysis, what does 'rolling back' (averaging out and folding back) mean?

    Rolling back is evaluating the tree from right to left: at chance nodes compute expected values from outcome probabilities, and at decision nodes choose the option with the best expected value. This identifies the optimal strategy and its expected cost/effect.

13. What is a Markov model and when is it preferred over a decision tree?

    A Markov model represents a disease as a set of mutually exclusive health states, with patients transitioning between them over discrete cycles according to transition probabilities. It is preferred for chronic/recurring conditions over long time horizons, where a decision tree would become unwieldy.

14. What is the Markov assumption (memorylessness) and what is a Markov 'cycle'?

    The Markov assumption is that transition probabilities depend only on the current state, not on history (memorylessness). A cycle is the fixed time interval (e.g. one year) over which transitions between states are modelled and costs/QALYs are accrued.

15. In infectious disease modelling, what do the compartments of an SIR model represent?

    Susceptible (S) — those who can be infected; Infectious (I) — those currently infected and able to transmit; Recovered/Removed (R) — those immune or removed (recovered or dead). Individuals move $S \to I \to R$ over time.

16. Define the basic reproduction number $R_0$ and its epidemic threshold.

    $R_0$ is the average number of secondary cases produced by one infectious individual in a fully susceptible population. If $R_0 > 1$ an epidemic can grow; if $R_0 < 1$ infection dies out. $R_0 = 1$ is the threshold for endemic stability.

17. Distinguish the basic reproduction number $R_0$ from the effective reproduction number $R$ (or $R_t$).

    $R_0$ assumes a wholly susceptible population. The effective reproduction number $R$ (or $R_t$) is the average secondary cases at a given time given existing immunity and interventions: $R = R_0 \times s$, where $s$ is the susceptible fraction. Control aims to keep $R < 1$.

18. State the herd immunity threshold formula and its meaning.

    $$H_c = 1 - \frac{1}{R_0}$$ $H_c$ is the proportion of the population that must be immune to halt sustained transmission. Higher $R_0$ requires a higher immunity threshold (e.g. measles with $R_0 \approx 15$ needs ~93% immunity).

19. Distinguish deterministic from stochastic infectious-disease models.

    Deterministic models use fixed equations (e.g. differential equations) giving the same output for given inputs, suited to large populations. Stochastic models incorporate chance/random variation, important for small populations or rare events (e.g. probability of extinction or outbreak).

20. Distinguish static from dynamic transmission models.

    Static models assume the force of infection is constant (risk to an individual is fixed), like a cohort/Markov model. Dynamic models allow the force of infection to change as the number of infectious people changes, capturing herd immunity and indirect (transmission) effects of interventions.

21. What is the 'force of infection' in transmission modelling?

    The force of infection ($\lambda$) is the per-capita rate at which susceptible individuals acquire infection at a given time. In dynamic models it depends on the prevalence of infection and the contact/transmission rate, so it changes over the course of an epidemic.

22. Give three legitimate uses of mathematical modelling in public health policy.

    (1) Forecasting epidemic spread and healthcare demand; (2) comparing the impact and cost-effectiveness of control strategies (e.g. vaccination vs treatment) including indirect/herd effects; (3) scenario planning and informing resource allocation under uncertainty.

23. State two key limitations of mathematical models that must be communicated to policymakers.

    Models are simplifications dependent on assumptions and input data quality ('garbage in, garbage out'); their projections carry uncertainty and should be presented with ranges/sensitivity analyses rather than precise point predictions. They inform, but do not replace, judgement.

24. What is net monetary benefit (NMB) and its formula?

    Net monetary benefit re-expresses cost-effectiveness on a single monetary scale using the threshold $\lambda$: $$NMB = (\lambda \times \Delta E) - \Delta C.$$ If $NMB > 0$ the intervention is cost-effective. It avoids problems of ratios (e.g. negative ICERs) when ranking options.

What this deck covers

Health Economics, Decision Science and Health Information flashcards FAQ