Epidemiology & Biostatistics Review

Question-and-Answer flashcards covering key epidemiology and biostatistics concepts from lecture notes.

What are the three main methods for collecting epidemiologic information?

Epidemiology maps, graphics (charts), and tables.

List three shared features every table, graphic, or map should have.

A clear numbered title, a brief summary of the content, and limits that keep the content easy to understand.

What is the primary benefit of a table in epidemiology?

It displays numbers of events (illness, death, injury, disability) for each category of a variable so comparisons are easy.

How many rows does a one-variable table contain?

Two rows – one header row and one data row.

Which three factors control the width of a confidence interval?

Sample size, confidence level, and variability in the population.

Define a point estimate.

A single numerical value used to approximate an unknown population parameter.

Define a confidence interval for a population parameter.

A range (lower and upper limits) calculated from sample data that is expected to contain the true population parameter with a stated level of confidence.

State the midrange formula and the midrange of 0, 7, 9, 2, 21, 13, 21, 5, 6, 14, 11, 10.

Midrange = (minimum + maximum) / 2; here (0 + 21)/2 = 10.5.

Which value is the mode of 29, 31, 24, 28, 30, 25, 24, 20, 33, 26, 27?

24 (it occurs most often).

What is the median of the data set 5, 6, 9, 9, 9, 9, 3, 3, 3, 3, 6, 11, 8, 14, 18, 1, 4?

8 (the middle value when ordered).

Give the formula for relative risk from a 2×2 table.

RR = [a/(a+b)] ÷ [c/(c+d)] where a = exposed diseased, b = exposed non-diseased, c = non-exposed diseased, d = non-exposed non-diseased.

Calculate the relative risk for: Exposed 200 disease / 9800 no disease; Non-exposed 100 disease / 9900 no disease.

RR = (200/10 000) ÷ (100/10 000) = 2.

In a diagnostic test table, which cell represents true negatives?

The cell where Disease- (standard test) and Test- intersect.

How many true negatives are in the table: T+ 4/7; T- 86/3?

86 true negatives.

Define conditional probability in diagnostic testing terms.

The probability of disease given a positive test result; P(D+ | T+).

What is the unconditional probability of disease in a population?

Total diseased / total population (regardless of test results).

What structure must every 2×2 table have?

Exactly two rows and two columns of data (plus totals).

What does a probability value of 0 mean?

The event cannot occur.

Perinatal mortality rate refers to deaths occurring at what infant age?

Under 7 completed days of life (often combined with late fetal deaths).

State two main purposes of tables, graphics, and maps in epidemiology.

To summarise data and to display/communicate data for analysis.

Write the population variance formula for a sample of n observations.

Variance = Σ(x − mean)² / n.

If one data point is wrongly recorded as 80 instead of 8, which measure changes most?

The mean increases dramatically (median and mode change little or not at all).

Name the two broad classes of frequency-distribution data.

Categorical data and numerical data.

Define the statistical term ‘mode’.

The value that appears most frequently in a data set.

Interpret a maternal mortality ratio of 60 per 1000 live births.

Out of 1 000 live births, about 60 mothers die during pregnancy, delivery, or within 42 days postpartum in that year.

Define the arithmetic mean.

The sum of all values divided by the number of values.

Using only ‘classic’ highly specific diagnostic cases tends to bias estimates in what direction?

It decreases measured incidence and prevalence.

High false-positive (poor specificity) diagnostic tests bias incidence and prevalence in what way?

They inflate both incidence and prevalence estimates.

List two situations that increase prevalence in a community.

Increase in new cases (incidence) or prolonged duration of existing cases.

What factor immediately increases incidence?

An increase in the number of new cases entering the population.

Give the neonatal mortality rate (per 1 000 live births) formula.

(Deaths of infants <28 days / live births) × 1 000.

State two main uses of an epidemiology map.

To show geographic distribution of disease and to mark patterns/clusters across areas.

Give two advantages of graphics over raw tables.

They highlight distribution patterns and demonstrate relationships (size versus frequency) at a glance.

List two defining features of a bar chart.

Bars are equal width and do not touch; bar length is proportional to the quantity represented.

Name three key elements every graph should display clearly.

A concise title, labelled axes with units, and darker data lines than axis lines.

What are the X and Y axes in a scatter plot used for?

X shows the independent variable; Y shows the dependent variable; points display paired values.

How many columns does a one-variable table normally need?

One column of categories and one column of numbers (two columns total).

Cite two characteristics of a good table.

Clear concise title and each column/row labelled with units; totals provided where appropriate.

Where should a data source be cited for a figure or table?

Immediately below the table, graph, or map.

List the three sequential steps recommended for presenting numerical data.

Organise figures, put them in a table, then illustrate with a graphic or map.

Provide the odds ratio formula for a 2×2 table.

OR = (a × d) / (b × c).

Calculate the odds ratio for: a=200, b=9 800, c=100, d=9 900.

OR = (200 × 9 900) / (9 800 × 100) ≈ 2.02.

Convert an 80 % probability of winning into odds.

Odds = 0.8 / 0.2 = 4 (4 to 1).

Write the standard chi-square formula.

χ² = Σ (O − E)² / E, where O = observed frequency and E = expected frequency.

State two main applications of the chi-square test.

Goodness-of-fit, test of independence/homogeneity for categorical data.

Is the chi-square test parametric or non-parametric?

Non-parametric (distribution-free) test for categorical data.

In which year was the chi-square test introduced?

1890s; commonly credited to Karl Pearson in 1900 (closest to “around 1890–1900”).

How many basic logical methods exist to frame an alternative hypothesis (H₁)?

Two methods – one-sided or two-sided alternatives.

Give the incidence rate formula.

(Number of new cases during a specified time / population at risk) × k (e.g., 1 000 or 100 000).

Define attack rate in an outbreak investigation.

(Number of new cases / number of persons exposed) × 100 %.

Name the four measurement scales used for variables.

Nominal, ordinal, interval, and ratio scales.

Distinguish point prevalence from period prevalence.

Point prevalence measures existing cases at a single point in time; period prevalence measures all cases present during a specified period.