Veterinary Epidemiology: Causality & Disease Determinants

• Introduction to Epidemiology for a Master's Degree in Veterinary Medicine at Egas Moniz School of Health & Science, held on March 17th, 2025, by Professor Dr Rita Santos (PhD).

Program & Evaluations

• The course uses problem-based learning.
• Course includes topics of causality, frequency measures, field epidemiology, building tools survey, association measures, preparing research proposals, sampling, diagnostic tests, and research proposal presentation.
• There is practical assessment and mini tests scheduled, alongside an integrative test.

Topics to be Discussed

• Recommended reading: "Veterinary Epidemiology" 3rd edition.
• Definition of Epidemiology (p22-28).
• Concepts and criteria of causality (p34-38).
• Descriptive Epidemiology (p28-32).
• Determinants of disease and health in populations (p75-95).
• Ecological concepts of disease (p116-120).
• Agent-host interdependence and the environment.
• Introduction to confounding factors, types of bias (p 40-44).
• Temporal and spatial patterns of disease (p137-145).

Frequency Measures

• Frequency measures in disease and mortality indicates how often diseases and deaths occur.

Epidemiologic Indicators

• Prevalence and Incidence are key indicators.
• Indicators of death and mortality help understand how diseases lead to deaths in a population.
• Direct and indirect Standardization are methods used.
• Prevalence measures the total number of existing and new cases of a disease in a population at a specific time.
• Incidence measures the number of new cases of a disease occurring over a specific period.
• Standardization is used to compare mortality or disease rates between populations with different age distributions
• Direct Standardization uses a standard population to compare two different populations, e.g., comparing cancer death rates in two countries by adjusting for age.
• Indirect Standardization uses expected deaths based on a reference population, e.g., comparing workplace accidents in a factory to national statistics.

Application of Epidemiological Indicators

• The confusion matrix is used to measure how well a diagnostic test detects disease.
• It helps in quantifying disease events and is useful for disease surveillance, observational investigations, and outbreak investigations.

Quantifying Disease Events

• Disease surveillance is used in routine monitoring of diseases in a population.
• Observational investigation is used to study disease risk factors and spread.
• The investigation of disease outbreaks/clusters is used when there's a sudden increase in disease cases.
• A True Positive (TP) correctly identifies the disease.
• A False Positive (FP) test says the disease is present, but it is wrong.
• A False Negative (FN) test indicates no disease, but the person is actually sick.
• A True Negative (TN) correctly identifies that there is no disease.

Types of Epidemiological Indicators

• These indicators help measure disease frequency in a population.
• Count measures the number of disease cases, e.g., "There were 50 cases of flu in the city last week." It has limited use due to the absence of a denominator.
• Proportion is the ratio between the number of cases (numerator) and the total population at risk (denominator). For example, if 50 out of 1,000 people have the flu, the proportion is 50/1000 = 0.05 (or 5%). It is limited as it does not indicate the size of the population, so different areas cannot be compared.
• Odds are the ratio between the number of disease cases (numerator) and the number of individuals without the disease (denominator), e.g., if 50 people have the flu and 950 people don't, the odds are 50/950=0.0526.
• Rate means the denominator includes the variable representing the total amount of time at risk. It measures the number of cases within a specific time period, e.g., if there were 50 new flu cases in a population of 1,000 over a year, the flu incidence rate might be 50 per 1,000 per year.

Type of Epidemiological Indicators Table

Exercise: Epidemiological Indicators in a Veterinary Clinic

• A clinic monitors canine parvovirus spread in a dog kennel with 100 dogs, collecting data over one month.
• 20 dogs tested positive.
• 80 dogs remained uninfected.
• The dogs accumulated a total of 500 dog-days in the kennel over the month.

Exercise Example Table

Types of Incidence

• Incidence in epidemiology refers to the rate at which new cases of a disease occur in a population over a specific period.
• Incidence-Time measures the time until an event occurs.
• Incidence-Count is the number of new disease cases in a population.
• Incidence-Risk is the probability that an individual will contract a disease within a specific time period.
• Incidence-Rate is the number of new disease cases in a population divided by animal-time units over a defined time period.

Incidence Type Table

Exercise: Incidence Measures in a Dog Kennel

• A veterinary clinic monitors a canine parvovirus outbreak in a dog kennel.
• The kennel hosts 100 dogs at any given time.
• 20 new cases were reported this month.
• The median time until the first case appeared was 10 days.
• The dogs spent 500 dog-days at the kennel over the month.
• The probability of a dog contracting parvovirus within the month was 15%.

Exercise Example Table

Proportion - Risk

• Population: Incidence-proportion is the proportion of individuals in a population at risk who are affected by a disease within a specific time period.
• Individual - Probability is the probability that an individual in a population at risk will be affected by a disease.
• R = Number of new cases in a defined time period / Population at risk during the same period.

Population at Risk

• In a Closed population, new animals do not enter, and there are few or no exits during the evaluation period.
• In an Open population, animals enter and leave during the study.
• Stable population describes when the number of entries equals the number of exits.

Incidence rates

• Incidence rate I = Number of new cases in a defined time period/Number of animal – time units at risk during the defined time period
• Cow-Month: Measure of time at risk for cows. One cow-month totals 1 cow for 1 month.
• Dog-Day: Measurement of time at risk for dogs totals 1 dog for 1 day.
• Average time to disease occurrence in a closed or stable population = 1/I.

Incidence Rate Example

• Defined Time Period: 30 days. Population (N): 4 animals.
• 1 animal did not get sick, so that animal is at risk for 30 days.
• 1 animal became sick on day 10, so that animal is at risk for 10 days.
• 1 animal became sick on day 20, so that animal is at risk for 20 days.
• 1 animal was sold on day 15, so that animal is at risk for 15 days.
• Total Animal-Time Units = 30+10+20+15 = 75 animal-days.
• Conversion of Animal-Days to Animal-Months: 1 month=30 days
• Total Animal-Time in Animal-Months = 75/30 = 2.5
• I = 2 / (11+10.33+10.67+10.5) = 2/2.5 = 0.80 cases/animal-month.
• Average time to disease occurrence in a closed or stable population = 1/I = 1/0.8 = 1.25 months.

Incidence Rate in a Cattery

• A veterinary clinic is monitoring a feline upper respiratory infection (URI) outbreak in a cattery.
• The cattery has a population of 6 cats over a 2-month period.
• Cats were observed for 2 months.
• 1 cat remained healthy for the entire 2 months.
• 2 cats got sick on day 15.
• 2 cats got sick on day 45.
• 1 cat was adopted and left the cattery on day 30.

Relationship between Risk and Rate

• Incidence is the number of new cases/population at risk R = number of new cases/population at risk.
• Incidence Rate relates to R: I = number of new cases/(N * ∆t) is equivalent to R = I * ∆t
• N = Population at Risk.
• △t = timeframe.

Prevalence

• Prevalence = the number of disease cases that exist at a specific point in time or over a defined period = number of disease cases in a defined period / Total number of animals at risk in the same period
• In a stable of 75 horses that are tested for equine infectious anaemia 3 test positive: P = 3/75 = 0.04 = 4%.

Mortality Indicators

• Mortality rate (MR) = Number of deaths in a defined time period / Number of diseased animal / time units at risk during the defined period.
• Case Fatality rate (FR) = Number of deaths in a defined time period / Number of diseased animal / time units during the defined period.
• Rate is not the same as Risk.
• Attack rate (Risk) AR = Number of disease cases in a disease outbreak / Number of animal – time during the defined period
• Secondary Attack Rate (Risk) = Number of disease cases in an outbreak – Number of initial cases /Number of animal-time units of disease animals during the defined period.
• Case Mortality Risk = Number of deaths from a specific during a time period/Number of animals with the disease.
• Mortality Ratio = Number of deaths from a specific disease during a time period/Number of deaths from all diseases during the same time period.

Venn Diagram

• Population correlates to mortality rate
• Overlapping population and animals with the disease, represents the Case Fatality Rate.

Standard Deviation and Prevalence

• Standard Deviation for Prevalence equation is as follows: sqrt(P(1 - P)/N)
• The 95% Confidence Interval equates to P + 1.96 * DP where P = Prevalence
• P = Number of disease cases in a defined period/Total number of animals at risk in the same period
• Standard Deviation for Incidence = sqrt(A/t^2)
• 95% Confidence Interval = I + 1.96 * DP where A = Number of new cases (incidence) and T = time period

Crude and Standardized Rates

• Crude rates are a general measure for the entire population without adjustments, for example, the mortality rate in Portugal across all age groups.
• Adjusted and Standardized rates are adjusted to demographic characteristics to allow fair comparisons, for example, the age-adjusted mortality rate in Portugal for individuals aged over 50 years.

Standardized Rates

• Direct Standardization: Rates (or risks) are calculated for a subpopulation and then applied to a reference population, useful for comparing disease rates across different populations by adjusting for demographic factors such as age, gender, or socioeconomic status.
• Indirect Standardisation: Rates (or risks) are calculated based on parameters from the reference population. Allows the calculation of standardised mortality ratios (SMR) or standardised morbidity ratios, which are useful when the age-specific rates of the study population are not available, or the population size is small.

Direct Standardisation (Risk) Population A

• Assumes various farms with dairy and/or beef numbers with cases exhibiting specific rates, leading to the percentage of the total population in the Standardised risk calculation.

Direct Standardisation (Risk) Population B

• Assumes various farms with dairy and/or beef numbers with cases exhibiting specific rates, leading to the percentage of the total population in the Standardised risk calculation.

Indirect Standardisation (Risk) Population A

• Assumes various farms with dairy and/or beef numbers with cases exhibiting specific rates, leading to the percentage of the total population in the Standardised risk calculation.

Indirect Standardisation (Risk) Population B

• Assumes various farms with dairy and/or beef numbers with cases exhibiting specific rates, leading to the percentage of the total population in the Standardised risk calculation.

Take Aways of the Day

• Epidemiological indicators.
• Prevalence and Incidence.
• Disease and mortality indicators.
• Direct and Indirect Standardisation.