Epidemiology

Question 1

Transmission Basic reproductive number

Answer 1

R0 = BxCxD B = probability of transmission, C= exposure rate of susceptible partners, D= duration of infectious period

Question 2

Transmission Basic reproductive number

Answer 2

R0>1 -> epidemic, goal control R0<1

Question 3

Transmission STI transmission

Answer 3

B Probability of transmission: Infectivity (viral or bacterial load), biological interactions with other organisms, behaviours which increase/decrease risk (condom, anal vs vaginal intercourse. C Behavioural parameters: Rate of partner change, patterns of sexual mixing (gender, age, assortative (serosorting), disassortative (age gap). D Duration: Infectious period of organism, treatment seeking behaviour

Question 4

Epidemiology Young and Solomon 10 questions

Answer 4

1 Is the study question relevant? 2 Does the study add anything new? 3 What type of research question is being asked? 4 Was the study design appropriate for the research question? 5 Did the study methods address the most important potential sources of bias? 6 Was the study performed according to the original protocol? 7 Does the study test a stated hypothesis? 8 Were the statistical analyses performed correctly? 9 Do the data justify the conclusions? 10 Are there any conflicts of interest?

Question 5

Outbreak Definition

Answer 5

An outbreak is defined as the occurrence of more cases of an adverse health event than expected in a given geographic area over a particular period of time

Question 6

Outbreak Case definition is key

Answer 6

Proven (confirmed), Probable (suspect) or Possible an example is: Possible = symptoms and likely epi risk, Probable = possible who died (ie cannot be confirmed) and Proven is Possible + micro confirmation (serology or culture depending on organism)

Question 7

Outbreak Point source curve

Answer 7

An outbreak that results from exposure to the same harmful influence (eg infectious agent or toxin0 from the same source (eg contaminated water). Same source, exposed period is brief - all cases occur within one incubation period, eg Legionnaire’s disease, does not spread, eg food-borne disease

Question 8

Outbreak Extended common- source

Answer 8

An outbreak that results from exposure over multiple incubation periods to the same harmful influence (eg infectious agent or toxin) from the same source (eg contaminated water) eg Cholera - does not spread, eg foodborne disease outbreak

Question 9

Outbreak Propagated source

Answer 9

Increasing depth of wave transmission occurs from person-to-person rather than a common source and can last longer than common source outbreaks. May have multiple waves. Cases in one peak may be sources for cases in a subsequent peak. May have progressively taller peaks, an incubation period apart. if the incubation period and the infectious period are similar. (eg SARS-CoV-2)

Question 10

Outbreak Intermittent source

Answer 10

Intermittent exposures: multiple peaks - length: no relation to the incubation period (reflects intermittent times of exposure) eg contaminated food product sold over period of time

Question 11

Outbreak What would stop VHF happening again?

Answer 11

After the investigation was complete, the following recommendations were made: 1 Maintain active national surveillance for acute haemorrhagic disease. 2 Distribute pertinent information to medical and other personnel participating in surveillance. 3 Organise a national campaign to inform health personnel of the proper methods for sterilising syringes and needles. 4 Maintain a list of experienced Zairian personnel so that appropriate action can be taken without delay in the event of a new epidemic. 5 Maintain a stock of basic medical supplies and protective clothing for use in suspected outbreaks. 6 Keep plasma from immune donors in readiness and obtain further information concerning the effectiveness of this treatment

Question 12

Outbreak 12 Steps to Outbreak Response

Answer 12

1 Prepare for field work - identify an investigation team, mobilise appropriate resources. 2 Establish the existence of an outbreak/epidemic. 3 Verify diagnoses of cases. 4 Establish a working case definition. 5 Systematic case finding to identify additional cases. 6 Conduct descriptive epidemiological studies (time, place, person). 7 Develop hypotheses. 8 Evaluate hypotheses (often with a case-control study). 9 If required, reconsider/refine hypotheses and do additional studies to test them. 10 Implement control and prevention measures (as early as possible) 11. Communicate findings. 12 Institute surveillance (Team/resources, identify, verify diagnoses, working case definition, case finding, descriptive epi (time place person), develop hypothesis, evaluate hypothesis, refine hypothesis, control measures, communicate, surveillance)

Question 13

Outbreak Case control study

Answer 13

A case-control study is designed to help determine if an exposure is associated with an outcome (ie disease or condition of interest). Good for outbreaks, quick, cheap, simple.

Question 14

Person, place and time Epidemiology

Answer 14

The distribution and determinants of disease (or health): how much [condition] is there, who gets it? We’re also interested in natural history: what happens to people with [condition]. We might be interested in trying to establish causation: what factors influence who gets [disease], what is the cause of this?. We might want to evaluate interventions to treat individuals or control disease: is treatment A better than treatment B? does this intervention prevent this disease?

Question 15

Person, place and time Key points

Answer 15

When thinking about disease in populations, and when considering whether you have a real outbreak on your hands: 1 the proportion of the population with the disease is more useful than just the absolute numbers (though both are useful and absolute numbers are obviously important for planning healthcare delivery) 2 ‘what is the denominator’ (what is the size of the population which produced the cases?)

Question 16

Person, place and time Key concept

Answer 16

Disease seen in hospital may not be representative of disease in the community

Question 17

Person, place and time Basic descriptive epidemiology

Answer 17

Describing a disease in a population in terms of person (who), place (where), time (when)

Question 18

Person, place and time Mapping/survey

Answer 18

Survey of disease in community must be a representative sample. Random means each individual (or household) has an equal chance of being included - it does NOT mean haphazard sampling

Question 19

Person, place and time Summary

Answer 19

The purpose of epidemiology is to understand the distribution and determinants of disease in populations in order to improve disease control. Use of epidemiological techniques (description of health and disease in a population, natural history of disease, risk factors/causation of disease, evaluating interventions). Basic descriptive epidemiology (time, place, person. need to know denominator. Percentage population with disease usually more useful than absolute number). Interpreting data (many possible explanations for apparent differences (or lack of) in routinely-collected data). Disease in hospital vs disease in community (the ears of the hippopotamus, influenced by disease severity but also (access to hospital, quality of care in community)). Surveys for disease in a community (must be a representative sample, random not haphazard).

Question 20

Quantifying disease frequency Prevalence & Incidence definitions

Answer 20

Prevalence: The number of cases of a disease in a defined population. Incidence: the number of new cases of disease in a defined population over a specified PERIOD of time (this is usually measured in cohort studies)

Question 21

Quantifying disease frequency Relationship of prevalence and incidence

Answer 21

The number of cases of disease at a given point in time (prevalence, P) will depend upon: the number of new cases that arise (incidence, I) AND how long each case lasts (duration, D) Example: Measles (high incidence, short duration, population prevalence low) HIV (low incidence, lifelong duration, population prevalence can be very high) HIV prevalence among gold miners in South Africa = 29% (out of every 100 gold miners, 29 are HIV+ at the time measured). HIV incidence among gold miners = 2% per year (2 out of 100 previously HIV-negative gold miners acquire new HIV infection every year)

Question 22

Quantifying disease frequency Incidence risk and incidence rate

Answer 22

Both count the number of new cases - the difference is in the denominator. Incidence risk counts number of new cases that arise in a population and divides them by the number of people who started off at risk (eg annual incidence of Salmonellosis in Liverpool supporters attributed to chicken pies is 15/1000). Incidence rate counts number of new cases that arise in a population and divides them by the person-time at risk. eg annual incidence rate of salmonellosis in Liverpool supporters attributed to chicken pies is 25/1000 (because not all Liverpool supporters identified at start of season attended every match, so denominator is smaller)

Question 23

Quantifying disease frequency Importance of denominator when calculating incidence: concept of population at risk

Answer 23

Strictly speaking, incidence is a measure of risk (or rate). Fairly meaningless to describe incidence of a disease in a population if many people are not at risk (eg risk of testicular cancer in the general population vs incidence in males in general population). You can’t get a disease if you are not at risk (so ideally the denominator of an incidence estimate should only include the ‘population at risk’). Example: calculation of HIV incidence: in a district of 30,000 adults there are 200 new cases in one year. But if 1000 people already had HIV, these should not be included int he denominator for the calculation of HIV incidence. Incidence = NEW cases arising in a defined period among a population FREE OF DISEASE at the start of the period of observation

Question 24

Quantifying disease frequency Incidence risk

Answer 24

Numerator (top number) = number of people in a population who acquire disease in a specified time period. Denominator (bottom number) = total number of people in the population who were free of disease at the START of the period of observation. Result usually expressed as a percentage or per 1000. Example HIV: Cohort 10,000 adults, all have HIV status measured on 1 January and again on 31 December. on 1/1, 1000 are HIV positive, by 31/12 200 have acquired new HIV. Population at risk is 10,000-1000=9,000, Cumulative incidence = 200/9000 = 2.2%. Incidence risk is easier to calculate than incidence rate (but problematic) - simply count how many people you start with and count how many have developed your outcome disease at the end. This works fine if all the people you start with can be found and checked to see if they have your outcome disease of interest. If they can’t be found - they are lost to follow up, then you can’t verify them as diseased or disease free, so your numerator lacks accuracy

Question 25

Quantifying disease frequency Incidence rate

Answer 25

If the population changes during period of observation, we need to take account of this. For incidence rate, denominator is person-time at risk. Person-time at risk = total time that each healthy individual contributes during the period of observation. Logically, the number of new cases must depend on how many people you follow, and how long you follow them for. Example: Consider PrEP for sex workers - you plan to evaluate the effect of PrEP on the HIV incidence rate so you measure HIV status at the start, and quarterly thereafter. You count the number of new HIV infections over 1 year. Denominator takes into account not just the number of HIV negative women at the start, but how long each woman remains HIV negative, until she 1 becomes HIV positive 2 leaves the programme (moves away or dies) or 3 gets to the end of the period of observation.

Question 26

Quantifying disease frequency Incidence risk versus rate

Answer 26

With larger numbers and more losses, the differences can become substantial. This sis particularly important if comparing between populations (eg where there were more losses from one than from the other). This calculation of incidence rate assumes that each woman became HIV positive on the day that she was tested positive - this is clearly not correct. To improve precision, need to ‘estimate’ when each woman actually became HIV positive - by convention, fi this is not known, it is estimated as halfway between the last negative test and the first positive test - ie, if testing is done every 3m, assume HIV was acquired 1.5m after last negative test

Question 27

Quantifying disease frequency Summary

Answer 27

Prevalence (number of new cases of a disease in a defined population at a given point in time). Incidence (number of new cases of disease rising in a defined population during a specified period of time). Numerator for incidence is the number of NEW cases. Denominator can be calculated in different ways: Risk (number free from disease at start). Rate (person TIME at risk). Prevalence depends on both incidence and duration.

Question 28

Study design Observational vs intervention studies

Answer 28

Observe and count. Descriptive (describe a disease in a population - what percentage of people in village have [hepatitis B]. What percentage of people with [hepatitis B] develop [liver cancer]. Analytic: explain the pattern of disease in a population by identifying risk factors (what are the risk factors for cerebral malaria in children? Does smoking cigarettes cause lung cancer?. Intervention studies - change something - INTERVENE- and see what happens

Question 29

Study design Exposure and outcome

Answer 29

Exposure = something to which an individual is exposed which may affect their risk of disease (often called ‘risk factor’ but may also be protective ie bednets). Outcome = the disease or event we are interested in. Examples E vs O - cigarette smoking and lung cancer, antismoking campaigns and cigarette smoking, lung cancer and anxiety

Question 30

Study design Case reports and case series

Answer 30

Principle: Describe experience of single patient (case report) or group of similar patients (case series). Can provide early information about new diseases. Multiple case reports may suggest a new disease or epidemic. Individuals in a case series can lead to the formulation of a hypothesis of causation. Advantages: enable rapid sharing of information about rare occurrences, can be useful in hypothesis formation. Disadvantages: often represent experience of one researcher, any association reported may be coincidental, there is a lack of a comparison group.

Question 31

Study design Cross-sectional studies

Answer 31

Principle: A representative sample of a population is studied to quantify a disease, an exposure, or both. A ‘snapshot’ of the population at a single point in time. Exposure and disease assessed simultaneously. Surveys are usually cross-sectional. Trends in disease over time can be examined using repeated cross-sectional studies. Examples: What is the prevalence of [condition] in [certain population]? ie Is the prevalence of A lumbricoides more common in boys than girls, how does this new rapid diagnostic test perform for the diagnosis of malaria, what is the spectrum of disease among HIV-positive people admitted to this hospital? Advantages: quick and (relatively) easy - often the first step in investigation of a problem, can measure prevalence of disease, can measure prevalence of a risk factor, can measure demand for a service (for health service planning), hypothesis generation - can make a rapid assessment of an association, then go on to investigate more closely with another design. Disadvantages: measure prevalence at a single point in time - prevalence (eg epilepsy in pig farmers) may be subject to effect upon survival. Cannot show that exposure came before disease.

Question 32

Study design Case-control studies

Answer 32

Principle: People with diseases (cases) and people without disease (controls) are compared by looking at the past exposure of each group to potential causes. “Retrospective” (look backwards from disease to cause) but much better to call them case-control studies (cohort studies can also be retrospective…) Can be useful for investigating causation, particularly if disease is rare, or long ‘incubation’ period after exposure (when cohort studies impractical). Examples: Are people with lung cancer more likely to have been cigarette smokers than people without lung cancer? Does BCG vaccine prevent leprosy? Does living near a canal increase the risk of malaria? What food caused this outbreak of Salmonella enteritis?. Advantages: relatively quick and inexpensive, useful for rare diseases, useful for disease with long ‘incubation’ periods, can examine multiple exposures. Disadvantages: impractical if the exposure is rare, vulnerable to bias (selection/information), not always certain that exposure preceded disease, cannot measure disease incidence

Question 33

Study design Cohort studies

Answer 33

Principle: simplest cohort is a ‘natural history’ (descriptive cohort) study - people with a risk factor are followed to see if they develop disease, but more usually and more usefully we do analytic cohort studies. Recruit people free from the disease, classify participants at baseline as exposed or unexposed, follow exposed and unexposed groups to determine outcome. Compare the incidence of the outcome between exposed and unexposed groups. These are usually prospective ie look forward in time from exposure to disease. Usually measure incidence, they are useful for studies of prognosis and causation. Example research questions: do people who have coma due to severe malaria subsequently have cognitive impairment? What proportion of people with antibodies to HIV develop AIDS? What is the effect of HIV infection on the outcome of TB treatment? Do smokers have a higher risk of lung cancer compared with non-smokers? Advantages: useful for rare exposures, can examine multiple outcomes, can show that exposure comes before disease, exposure and outcome data are usually measured directly - so more accurate and less risk of bias. Disadvantages: impractical/inefficient if the outcome is rare, vulnerable to losses to follow up, can be expensive and time-consuming

Question 34

Study design Prospective vs retrospective cohorts

Answer 34

Prospective: recruit people without the outcome of interest, follow up in real time to determine when the outcome happens, better data on order of events, can look at time to event, exposure and outcome data measured directly. Retrospective: construct the cohort based on past records (eg people starting ART to determine 6m outcomes), efficient if accurate outcome data (eg vital status if good records), problematic if poor data on outcome (eg adverse events, or if deaths cannot be determined among those lost to follow up)

Question 35

Study design Nested case-control studies

Answer 35

A case-control study may be ‘nested’ into a cohort or trial. At the end of the cohort study when outcomes have occurred, investigators select cases (participants who developed the outcome of interest during follow up), controls (sample of participants who did not develop outcome during follow up), and perform additional analyses on cases and controls. Useful where additional analyses are expensive/complex so can’t be done on everyone. Example: prospective cohort study with active TB as an outcome. At study end, select those who developed TB (cases) plus sample of those with no TB (controls) and compare transcriptomic signature from samples taken at baseline between cases and controls.

Question 36

Study design Randomised controlled trials

Answer 36

Generally regarded as the ‘gold standard’ study design. Principle: a cohort study, but investigator allocates exposure, participants are randomly allocated to an intervention group or a control group. Intervention and control groups are followed up and evaluated for specified outcomes (which should have been decided upon at the outset of the trial. Rationale: since the allocation is random - the groups should be similar with respect to all risk factors, apart from the intervention, therefore any any difference in outcome between intervention and non-intervention arms should be attributable to the intervention. Useful for questions about interventions (treatment or prevention). Example research questions: does bolus IV fluid reduce mortality among critically-ill children? Do impregnated bed nets reduce the risk of malaria? Is an HIV ‘test and treat’ strategy superior to ART delivered based on a CD4 threshold in preventing new HIV infections? Advantages: RCTs provide strong evidence of the effect of the intervention, data is gathered prospectively, more than one outcome can be examined, randomisation potentially eradicates effect of bias. Disadvantages: can be expensive and time consuming, if not properly planned they may be too small or too short, not always practical (if the number of patients needed is prohibitively high, if randomisation is unethical (cannot randomise an exposure thought harmful eg cigarette smoking), trial results do not necessarily reflect what happens when intervention is delivered routinely (in ‘real life’)

Question 37

Study design PICO

Answer 37

Population, Intervention, Comparator, Outcome (PICOST includes Situation and Type of study)

Question 38

Fact Faget’s sign

Answer 38

Fever and bradycardia - Sphygmothermic dissociation. Yellow fever, Salmonella typhi, Tularaemia, brucellosis, Colorado tick fever, Legionella pneumophilia, Mycoplasma pneumoniae

Question 39

Critical appraisal Phases

Answer 39

Phase I - is the treatment safe? Phase II Does the treatment work? Phase III Is the new treatment better than existing treatments? (efficacy) Phase IV is there a better way to use this treatment? Or implement this intervention (effectiveness)

Question 40

Critical appraisal Why do a critical appraisal?

Answer 40

Caution (be cautious, many published studies are poorly done, results are unreliable or wrong), Need (to assess whether the design, analysis and interpretation were adequate and reasonable), Need (to determine if results are important and generalisable)

Question 41

Critical appraisal Validity of a study

Answer 41

Any finding that treatment A is better than B always has four potential explanations: 1 True, 2 result of confounding (there’s an alternative explanation), 3 result of bias (some kind of systematic error in selection) 4 Chance. The question of validity is really about how well the study addressed the roles of chance, bias, and confounding. The roles of chance and bias are usually evaluated/negated by statistical analysis and randomisation (reduce role of bias). Inadequate quality of studies may distort results of metaanalyses and systematic reviews -> The influence of the quality of included studies should be examined: can have HUGE implications -> Key issues for validity are: concealed treatment allocation, blinded outcome assessment, handling of patient attrition in the analysis. Summary points: Valid studies often produce findings that are unlikely to change current practice. They may have been carried in such different settings or on such different populations to your own, that the results cannot be safely applied.

Question 42

Critical appraisal Errors and bias

Answer 42

All studies have the potential for error - in fact, errors are probably impossible to avoid. However some errors are random, others are systematic

Question 43

Critical appraisal Random error

Answer 43

Result is inaccurate by CHANCE or by imprecision. Random error is shown with p values and confidence intervals within studies

Question 44

Critical appraisal Bias

Answer 44

Systematic error - any trend in the collection, analysis, interpretation, publication or review of data that can lead to conclusions that are systematically different from the truth. Bias can occur in different ways.

Question 45

Critical appraisal Selection bias

Answer 45

Arises if the groups being compared are not comparable. Eg in a case-control study if the controls are not carefully selected - if they are from different populations etc. Questions you can use to check for evidence of selection bias: Was the study population clearly defined? What were the inclusion and exclusion criteria? What percentage of those eligible were recruited to the study? For studies with follow-up (cohort/intervention), was there a high proportion of loss to follow up? Are the exposed and unexposed groups similar except for the exposure of interest or intervention? Is the follow up period similar for the two groups? For case-control studies, are the controls drawn from the same population as the cases?

Question 46

Critical appraisal Information bias

Answer 46

Any error in the measurement of exposure or outcome resulting in systematic differences in the accuracy of information collected between groups being compared. If information is equally inaccurate in all groups that is measurement error not bias. Questions you can use to check for evidence of information bias: Where the exposures/outcomes clearly defined? Were the measurements as objective as possible? Was the study ‘blinded’ as much as possible? Were the observers/interviewers carefully trained? Was data collection standard?

Question 47

Critical appraisal Precision

Answer 47

If I use this measure many times, do I get nearly the same results? Also called reproducibility, reliability, consistency

Question 48

Critical appraisal Accuracy

Answer 48

Does my variable measure what it is supposed to - how close am I to measuring the actual thing I am trying to measure. Example a dosing height pole for Praziquantel dosing in Schistosomiasis

Question 49

Statistical analysis Sensitivity

Answer 49

The proportion of those who truly have a disease, who have a positive test =A/A+C. Sens/Spec are characteristics of the TEST, will be altered by use of different cut-offs

Question 50

Statistical analysis Specificity

Answer 50

The proportion of those who truly do not have a disease, who have a negative test = D/D+B Sens/Spec are characteristics of the TEST, will be altered by use of different cut-offs

Question 51

Statistical analysis PPV

Answer 51

The proportion of those with a positive test, who truly have a disease = A/A+B PPV/NPV are altered by the PREVALENCE of the disease

Question 52

Statistical analysis NPV

Answer 52

The proportion of those with a negative test, who truly do not have a disease = D/D+C PPV/NPV are altered by the PREVALENCE of the disease

Question 53

Statistical analysis 4x4 table

Answer 53

Gold standard at the top, new test on the side. Represent as a ++ b-+ c+- d– a = true pos b = false pos c= false neg d= true neg Sens = a/a+c Spec = d/b+d PPV=a/a+b NPV=d/c+d

Question 54

Statistical analysis Confounder

Answer 54

The situation where the association between an exposure and an outcome is entirely or partially due to ANOTHER exposure. A confounding variable must satisfy two conditions: must be associated with the exposure of interest, must be a risk factor for the outcome of interest

Question 55

Statistical analysis Odds ratio

Answer 55

Risk ratio = A/A+B C/C+D. Odds ratio = A/B C/D

Question 56

Statistical analysis Ways of dealing with confounding

Answer 56

DESIGN: Restriction (study only individuals who are similar with respect to the confounder) Randomisation (to exposure of interest) Matching (choose controls who are the same sex as cases) Identify and measure confounders at the design stage, in order to be able to assess confounding at the analysis stage. ANALYSIS: Stratification (analyse data separately according to presence or absence of the confounder), Statistical adjustment (logistic regression modelling)

Question 57

Statistical analysis Confounding vs bias

Answer 57

Confounding (arises because exposures of interest in a population are not randomly distributed, but are often associated with each other (eg low income, crowded living conditions, poor sanitation) (eg smoking cigarettes, drinking alcohol, use of other drugs) confounding is a nuisance of which we need to be aware, is not caused by poor study design, can (to some extent) allow for confounding in data analysis providing right data was collected). Bias (arises because of imperfect study design or conduct, a biased study fails to give a true representation of the situation we want to describe or the association we are trying to analyse, data analysis cannot correct for bias - if you study the wrong people or collect the wrong information clever analysis cannot correct this)

Question 58

Statistical analysis Summary

Answer 58

Risk ratios and odds ratios: two alternative ways to express the strength of an association. OR are further away from 1 than the corresponding RR. Ors and RRs are similar if the outcome is rare, but diverge more if the outcome is common. Confounding: the situation where the association between an exposure and an outcome is entirely or partially due to another exposure (A confounding variable must satisfy two conditions - it must be associated with the exposure of interest, it must be a risk factor for the outcome of interest)

Question 59

Statistics Error types

Answer 59

Type 1 finding something that is not there, Type 2 not finding something that is there

Question 60

Economics Definitions

Answer 60

Macro: National level decisions, political. Micro: subset of microeconomics within health economics - experimental in its basis and based in individuals. How we can improve the poorest in our community. Economics is essential - need to know the cost of the drugs you are prescribing, as this is important in understanding the drivers of prescribing.

Question 61

Economics Concept

Answer 61

If you do not understand something it is my fault for not explaining it clearly -> how to be!

Question 62

Economics Opportunity cost

Answer 62

If you spend more on one item, then the cost is spending less on another. You want to spend on the optimal BUNDLE of interventions - balance the things you can do with the resources available. If you ‘misallocate’ you will kill just as many as if you use the wrong drugs - you will have less benefit from the resource in front of you (usually highly constrained) - if you do one thing you must do less of another thing

Question 63

Economics Affordability

Answer 63

Is a drug/diagnostic test/procedure affordable? This is very context specific

Question 64

Economics Supply and demand

Answer 64

If supply constrained - and demand goes up, by definition the price will go up (and vice versa). This simple understanding is not understood. If you want to increase demand for condoms, PrEP etc, you need to also increase supply to counterbalance the risk of price rise

Question 65

Economics Monopoly

Answer 65

Competition within reason is good, particularly for medications. Complex monopoly can occur when pharma collude on prices. The only way to know how much a drug actually costs to make is to put a competitor against it, and see how far they can drop the price. The more competitors, the more likely to find the case for a cheaper cost. Companies will try and get into the market when they have a monopoly - there is a big cost to a country to change policy, effective monopoly can be maintained when there are more options because the policy is difficult to change (as is the education for clinicians on the new product)

Question 66

Economics Price

Answer 66

Most of us believe deep in our souls if it is more expensive it is better. Two options - sell lots and sell them cheap, sell almost none and sell them at a high margin - there is psychology around this. Example blue and green bednets in Malawi - same except for shape and colour - expensive ones sold out. Same with shampoo in UK - made by two companies only with different market points

Question 67

Economics Cost-effectiveness

Answer 67

How effective an intervention is (medical question) and how much it costs (economist question). However, important additional concept is of marginal cost effectiveness. Invest in bednet manufacturing - initial outlay is huge, but ever additional bednet you are making is making more and more profit up to the point the limit of the factory is reached, then if need to increase factory - the next bednet is again much more costly - this is not a smooth curve

Question 68

Economics Direct and indirect costs

Answer 68

Important for treating patients. Direct is the cost in money terms of the drug, user fees, consultation fees, fees for diagnostic services, may also be the money you need to pay to enter the facility. You can make treatment free (or very highly subsidised) to reduce the direct cost to the poorest. But other direct costs may not be able to be subsidised (cost of travel ie bus ticket). The much bigger cost for most families is the indirect cost. Indirect is the cost to patients and families in the opportunity cost - money they lost by being away from their work, or the lost day of education, cost of alternative childcare - these are often more important and dominant - you cannot subsidise the indirect costs. Free treatment at one level is fine, but in a rural area, the indirect cost of accessing the free treatment is more than the cost of buying the treatment in the nearest care facility - time really matters during the wet season when the travel will be slow and the opportunity cost will be high. The nearest cost will be in the private sector - do not ignore the impact of the nearby market stall as opposed to the free MSG care that is available but a long way away (and with the loss of opportunity cost)

Question 69

Demography Three drivers

Answer 69

Child mortality (rapidly falling). Rising age of mortality, Reducing fertility - the global population Is unrecognisable to what it was when our grandparents were growing up. If you want populations to expand - stop the children dying and the rest will follow. The only way to persuade mothers to have less children is to stop the children from dying.