What chronic disease are prevalent in your country / community? • Prioritize them from the most urgent to the least urgent. ?

The science of public health

Greek epi (upon), demos (people), logos (study)

Now includes injury, disability, risk factors, quality of life, and, traditionally, disease

Center for Disease Control (CDC) defines epidemiology as “the study of the distribution and determinants of health-related states in specified populations, and the application of this study to the control of health problems.”

Definition of Epidemiology

Based on the CDC’s definition, a breakdown of the terms:

Study – quantitative, statistic- and research-based

Distribution – characterized through people and how health events happen in populations

Determinants – causes, factors, implemented in analytical epidemiology

Health-related states – not just diseases, a spectrum of ailments

Populations – groups of people, not just individuals

Control – implement decision-making processes to prevent health issues

Definition of Epidemiology, contd.

Diseases of known microorganisms (AIDS, flu, etc.) focus the study of epidemiologists

Likewise, injuries are acute, with often immediate health consequences (car accidents, falling, etc.)

However, many chronic diseases originate early in life through unhealthy behaviors, which, in turn, increase the risk in later life to develop these diseases

This (detailed below) is called the chronic disease continuum

The Chronic Disease Continuum

Descriptive Epidemiology, Analytic Epidemiology, Intervention/Evaluative Research

Developmental, social, environmental, genetic determinants

Health risk behaviors and other factors

Chronic conditions

Chronic diseases

Adapted from Remington et al. (2010)

An important question is at what point are epidemiologist trying to “control” a chronic disease? At which stage of the continuum?

Controlling has become not only achieving lower mortality rates, but also lowering detrimental effects of the disease

Both of these factors are used to describe the natural history of a given disease

The Chronic Disease Continuum, contd.

Uses readily available data to examine a disease and who it affects in populations by time, place, or person

Assesses burden, at risk groups, trends

Public tends to focus on incurable diseases or environmental diseases, instead of chronic diseases that fall on the continuum.

Descriptive Epidemiology

Frequently, the count of affected individuals is seen as the fundamental measure

Implemented when analyzing the need for healthcare to aide in lessening the burden

Count is a simple way to present this burden to the general public, but it is dependent on the population size

Other measures can be used by epidemiologists to highlight different elements of the disease

The Burden of Disease

Rates are used to compare populations – incidence and prevalence

Incidence rate – new cases over period ÷ person-years

Person-years = persons × period monitored

Cumulative incidence – probability of developing a disease, ranging from 0 to 1

Prevalence – another proportion, this one measured at a point in time instead of a period

Less valuable than incidence in identifying causes

Calculating Rates

Compared using risk ratios/relative risks

Rate ratio – incidence rate of one group compared to another

Rate difference – subtracting the rates from one another

Calculates how much risk is different given certain variables in the study

Rates also used to aim program targeting goals

At-risk subgroups are found within populations

Can assist healthcare professionals in joint interventions

Comparing Rates

Randomized controlled trials (RCTs) – the most thorough and scientific study

Subjects assigned to prevention or non-prevention

Disease and death-rates monitored and observed over time

Often impractical/impossible

Need comparison groups for analysis

Comparisons between (least effective) national data and (most effective) concurrent collection vary in appropriateness

Study Designs

Study Designs, contd.

Adapted from Remington et al. (2010)

Historical or retroactive studies can also be a viable option, if the correct records are available

Downsides include that no incidence rates or cumulative incidence can be gathered

Upsides include that, in certain cases, information was found that affected risks which could be modified and benefit a population in the future

Since RCTs can be unethical, epidemiologists need to observe without intervention

Based on doctor recommendations

Prospective cohort – exposures not randomly assigned, but first identified

Active and intense measurements can be made

Observational Studies

Strengths/Limitations of Study Designs

Adapted from Remington et al. (2010)

Most chronic disease studies are from observational studying

Therefore, errors are present in a number of different ways:

Measurement, selection of subjects, bias, etc.

Experts are needed to quantify uncertainty in research methods and decide quality studies

Confounding and bias are important types of error in epidemiological studies

Assessing Valid Study Results

“The influence of an exposure of interest is mixed with the effect of another” (Remington et al. 2010)

In RCT, confounders can be equally distributed among each study group

In observational studies, confounders have to be measured and adjusted

As long as confounders are measurable and adjusted, then no need for concern

However, some confounders are difficult to measure or unknown

Confounding

Faulty sampling methods or refusal to participate leads to samples with higher or lower measurement risk

Poorly arranged control groups are major threats to accuracy

Phone samples introduce bias when families without landlines are not taken into account

Classification errors also present a potential bias

Also known as information bias

Selection/Information Bias

Attributable Risk

How much burden is lessened is exposure is eliminated

Relative risk - 1 ÷ relative risk

Population Attributable Risk

How common the exposure is

Rate (total pop.) - rate (unexposed) ÷ rate (total pop.)

Also Pe (relative risk - 1) ÷ 1 + Pe (relative risk - 1)

Where Pe proportion of population exposure

Prevention through Intervention

Often, critical analyses will lead to “important” or “groundbreaking” new information, but this must be taken lightly

One study is never sufficient to answer all or even one question about a chronic disease.

Validity must be measured through repetition and variety or intervention

Single vs. Series

Systematic reviews

Researchers consolidate information using comprehensive methods to assess relevant science behind a single health issue

Increasingly, meta-analysis is used in these systematic reviews

Meta-analysis

Identify relevant studies, inclusion/exclusion criteria, data abstraction, heterogeneity across statistical analysis

Systematic Reviews, Meta-analysis, Expert Panels

An approach to looking at risks posed by adverse exposures like pollutants

A “bridge” between the scientific community and those who create policies and protocol

Four steps to proper risk assessment:

Hazard identification, risk characterization, exposure assessment and risk estimation

Expert panels used to assess validity of epidemiological studies through peer review

Risk Assessment/Expert panels

Incidence rate – New events in a specified period / persons exposed to risk within period

Relative risk – Risk of death or disease in population exposed to risk / risk of death or diseased in unexposed population

Population attributable risk – Rate of disease in population attributed to risk factor / total rate of disease

Key Concepts