Measuring and Describing Disease part 2 Flashcards by Sydney Carrera-Brown

List the 6 types studies in the hierarchy of evidence from strongest to weakest.

Systematic reviews and meta-analysis

Randomised controlled trials

Cohort studies

Case-control studies

Case series, Case report

Editorials, expert opinion

How well did you know this?

Not at all

Perfectly

What are the two main research methods?

Qualitative

Quantitative

How well did you know this?

Not at all

Perfectly

What are the two main study designs?

Observational research

Interventional research

How well did you know this?

Not at all

Perfectly

What are the two main types of epidemiological approaches?

Descriptive epidemiology

Analytical epidemiology

How well did you know this?

Not at all

Perfectly

What is the main purpose of qualitative research?

Explores underlying ideas and themes to inform possible research questions and help in establishing a future hypotheses

How well did you know this?

Not at all

Perfectly

How are findings and outputs expressed in qualitative research?

Findings described in words

How well did you know this?

Not at all

Perfectly

How does the sample size differ in qualitative research versus quantitative?

Sample size of participants is smaller but goes into substantial detail

How well did you know this?

Not at all

Perfectly

What are the 3 main causes of morbidity/mortality? List them from most common to least common in the UK population

Non-communicable diseases (e.g. CVD, Cancers)

Communicable diseases (e.g. malaria, infections)

Injuries (e.g. traffic accidents, violence)

How well did you know this?

Not at all

Perfectly

What are the 4 measures of frequency?

Odds

Prevelance

Cumulative incidence

Incidence rate

How well did you know this?

Not at all

Perfectly

Define odds

The ratio between the number of people who have the disease to the number of people who dont have the disease

N.B. The ratio of the probability (P) of an event to the probability of its complement (1-P).

How well did you know this?

Not at all

Perfectly

In the context of odds, what is the output number above which it becomes more likely an event takes place than does not take place?

1.0

Because if half the population has the outcome the calculation will be 1 so any number higher than half the number will produce a value higher than 1

(e.g. 6 people sleep 6 people awake = 6/6 =1. 7 people sleep 5 people awake = 7/5 =>1)

How well did you know this?

Not at all

Perfectly

Define prevalence

Proportion of individuals in a population who have the disease or attribute of interest at a specific point in time

N.B. proportion not ratio

How well did you know this?

Not at all

Perfectly

How is prevelance calculated?

Number of people with disease/ Total population

N.B. The measure can be a percentage or in decimals and you have to specify the timepoint in which this value was calculated (e.g. 20% of students have sleeping disease at 10am)

How well did you know this?

Not at all

Perfectly

What is a flaw of prevalence?

Prevalence provides no information on new cases of a disease (is only a snapshot).

Therefore not a good measure when trying to establish frequency of short duration diseases

How well did you know this?

Not at all

Perfectly

Define cumulative incidence

Proportion of the population with a new event during a given period of time.

N.B. Useful when wanting to see how the frequency of cases change over given periods of time

How well did you know this?

Not at all

Perfectly

How is cumulative frequency calculated?

Number of new cases in period of interest/ Number of disease free individuals at start of time period

N.B. Remember to state this period of time (e.g. 50% cumulative incidence of sleeping disease over a 5 hour time period)

N.B. Do not include people that are already ill at start of time period as part of the bottom denomiator

How well did you know this?

Not at all

Perfectly

What does a cumulative interest of 0% indicate?

No new incidences have occured within this time period

What does a cumulative incidence of 100% indicate?

All individuals developed the disease during this time period

What are 2 alternative names for cumulative incidence?

Risk

Incidence proportion

What is a weakness of cumulative incidence?

Cumulative incidence can look at the proportion of disease but it cannot tell us when a disease occurs within time period (cannot calculate the rate of new cases of disease)

Can cumulative incidence be calculated in studies without a follow up?

No as they would not be able to see how many people in the observed sample developed the disease if they arent seen again

N.B. another weakness

What are the 2 main implications of follow up studies?

Some people in study population may drop out/die before follow up

Some people may enter the study population before follow up

What measures can address the problems associated with cumulative incidence?

Incidence rate

Define person-time

Measures time participants spend in the study.

(person hours stop at moment they develop disease or leave study)

(e.g. person enters a 9am lecture at 10am and develops sleeping disease at 1pm means they contributed 3 person hours)

N.B. bottom person in grey doesnt contribute to anything as they already had disease during start of study period

Define incidence rate

The count of new cases during the follow-up period, divided by the total person-time. e.g. 2.5 new cases per month

How is incidence rate calculated?

Number of new cases during the follow up period/Total person time by **disease free individuals** **N.B. THIS INCLUDES THE HOURS UP UNTIL THEYD DEVELOP DISEASE** e.g. 2 new cases/10 people hours = 0.2 new cases per hour

Is person time always in hours?

No it can be in minute, days ect which is why its always important to state the measure of person-time

What is the most appropriate measure of frequency when investigating chronic diseases?

Prevelance

What is the most appropriate measure of frequency when investigating short lasting diseases?

Incidence rate (prevalence would be difficult as patients may have already recovered)

How does incidence and preference differ from eachother?

Prevalence differs from incidence proportion as prevalence includes all cases (new and pre-existing cases) in the population at the specified time whereas incidence is limited to new cases only

Give an example how standardisation could be used

Helps to determine whether the difference in incidence is related to differences in demographic or not. (e.g. town may have higher incidence of strokes but could be due to the fact that it has a higher demopgraphic of older people, standardisation can help to determine whether this is true or not)

What are the two types of standardisation

Direct standardisation Indirect standardisation

Define direct standardisation

Type of adjustment that allows us to compare a like-for-like between populations by giving a comparable incidence

What is the difference between crude death rates and standardized death rates?

Crude death rates looks at the incidence of deaths occuring within the population as a whole Standardized death rates are based on particular characteristics within that population (e.g. age or gender). N.B. Crude rates arent good to compare against different populations as particular differences in demographics (e.g. age or sex) may influence differences in rates.

Define indirect standardisation

Type of standardisation that gives out a ratio

How is crude rate calculated?

Total number of cases in time period / Total population x k K = constant usually 1000 (per 1000) or 100,000 (per 100,000)

What is a standard population?

Arbitrary population based off a general average from multiple different populations

How is the expected count (expected number of deaths) calculated?

rate (within specific age range/or gender) x standard population / k (e.g. per 100k) N.B. You then add all the age specific expected counts to create a total expected count within that population

Give 3 possible explanations for findings that may be presumed to be abnormal (e.g. more deaths in this hospital than previously)

1) Unwaranted variation (variation that cannot be explained by illness, medical need, or the dictates of evidence-based medicine) **[E.g. Hospital is dangerous]** 2) Explained variation 3) Statistical artefact (e.g. hospital is better at recording deaths especially outside of hospital so may appear to have higher death rates)

How do you calculate the standardised mortality ratio?

Observed count (of deaths)/ Expected count (of deaths) N.B. A SMR of 1.65 would mean there were 65% more deaths than expected based on national estimates adjusting for the mix of procedures

What is an implication of crude rates?

They are standardised. Crude rates do not take into account factors such as age, gender