Evidence-Based Public Health

Maximizing the health of a population is largely achieved through three core functions of public health; Assessment, policy development, and assurance. Within each core function lies a subset of the 10 essential functions of public health:

As mentioned in the the previous module on ethics, it is essential to consider alternative courses of action to maximize benefit, minimize adverse consequences and make the most efficient use of scarce resources. As a result decisions on how to address public health must be evidence-based, meaning they are based on accurate and complete data which have been rigorously analyzed.

Two broad types of evidence are essential to the practice of public health:

The Evidence regarding the efficacy and adverse consequences of interventions
Evidence regarding the determinants of health and disease

In addition, it is useful to have evidence regarding how a disease spreads across a population, including social and political factors. Ex. are there members of the population at higher risk due to behavior, age, preexisting conditions or geography?

Even though most research depends on quantitative data, a fine understanding usually requires qualitative data.

Learning Objectives:

Describe the general steps involved in identifying and addressing public health problems
Describe and give examples of the role of surveillance systems in public health
Explain what is meant by the term "notifiable/reportable disease"
Describe the breadth of modern public health surveillance systems and give specific examples
Define "syndromic surveillance" and explain how it can be advantageous over regular surveillance
Explain how the characteristics of person, place, and time are used to formulate hypotheses in acute disease outbreaks and in studies of chronic diseases
Distinguish among case reports, case series, cross-sectional surveys, and ecological studies and explain their importance
Explain the concept of ecologic fallacy
Describe the difference between descriptive and analytic epidemiologic studies
Define and explain the distinguish between retrospective and retrospective cohort studies
Define and explain features of a case-control study
Explain and distinguish features of of an intervention study (clinical trial)
List and define 3 major threats to validity in analytical studies
Describe the contribution of quantitative and qualitative research to public health

Four Steps

Regalman summarized the four steps to public health approach as followed:

1. Identifying the Problem

In the US there are number of ways in which public health problems are identified:

Surveillance systems monitor and record occurrences of specific infectious diseases, and information is communicated from municipalities or counties to state health departments and then federal agencies
States maintain cancer registries
Computerized registries are increasingly used for diseases like diabetes, kidney failure, and a variety of other problems
Vital statistics regarding births, deaths, marriages, and divorces are collected by each state and other jurisdictions and through co-op agreements that data is shared with the US National Vital Statistics System
Many federal agencies periodically conduct surveys to monitor health status of the population
Health care providers report new or unusual problems to health departments and disseminate findings in the form of case reports or case series
Local and federal organizations and systems are in place to inspect food and monitor water quality

2. Identifying Contributory Causes

Epidemiology is a discipline that studies the distribution of health outcomes and its possible determinants in populations in order to identify casual associations. Epidemiology us described as the basic science of public health, because it is based in scientific method and identifies factors associated with health states by systematically collecting, analyzing and interpreting quantitative data.

3. Identifying What Works

This is also largely achieved through quantitative methods that evaluate the efficacy of interventions through clinical trials and program evaluation.

4. Identifying the Means

This may rely on quantitative methods, but increasingly relies on qualitative research which aims at understanding people's perceptions and developing an understanding of why they believe what they do. Ex interviewing a key informant to see why a particular intervention did or did not work in order to improve future programs.

Surveillance Systems

Surveillance for public health is based on a variety of data sources that track health status and health-related behaviors in a population over time in order to identify new problems, such as infectious disease outbreaks. In addition to monitoring trends these data help identify new health problems and provide clues about the causes of new issues in a population. Surveillance data enables us to:

Monitor and report on the health status and health related behaviors in a population
Identify emerging health problems
Alert us to potential threats
Establish public health priorities for a population
Evaluate the effectiveness of intervention programs
Explore potential associations between risk factors and health outcomes in order to generate hypotheses about the determinants of disease

Passive Surveillance

Routine reporting of health data. Can create a full picture when combined but is limited by under-reporting.

Hospital Data
Disease registries
Notifiable diseases - Some countries have laws which list diseases which are required to report

Active Surveillance

Actively sought out information:

Active case finding
Testing blood markers
Community Surveys

Sentinel Surveillance

A group of selected institutions or groups pool health data together on a certain disease. This can be used to monitor disease outbreaks and trends, but is unable to detect cases outside sites/groups making it ineffective for rare diseases.

Rumor Surveillance

Relies on unofficial sources of information (blogs, social media, hearsay, etc.). It might alert authorities to investigate further for signs of an outbreak.

Syndromic Surveillance

Monitors for non-specific illness syndromes recorded at medical clinics, along with tells like medicine purchases or absenteeism from school or work. Allows for early identification and can rely on automatic reporting methods. A "syndrome" is a cluster of signs and symptoms.

Other Types

ProMED - Web-based
GPHIN - Early warning tool to monitor significant health events

Characteristics of a Good Surveillance System (from CDC)

An integrative approach uses all available information and provides the best overall picture of health
It has simple, clearly defined objectives
The data should be complete and accurate
The plan should be flexible without incurring additional costs
Accepted by people and organizations involved in public health surveillance
Sensitive and Positive Predictive Value (PPV), it should pick up most if not all cases of the disease in the population
It accurately measures what it intends to measure
Timeliness to leave room for action
Stable and available when needed

And all surveillance systems should be evaluated on their effectiveness and if they meet public health functions

Data Sources

There are many sources of data that are useful for monitoring health of the populations and exploring disease frequency over time. Ex:

Census and Records - US Census and US National Vital Statistics System and the Massachusetts Registry of Vital Statistics and Records
Disease registries - Massachusetts Cancer Registry, ALS Registry
Hospital Discharge Registries
Community Health Information - In MA the Community Health Information Profile (MassCHIP) provides a dynamic, user-friendly service for obtaining free online access to many health and social indicators. One can obtain community-level data to assess health needs, monitor health status indicators and evaluate health programs.
Infectious Disease Surveillance - Division of Emerging Infections and Surveillance Services (DEISS)
Commercial Data - Sale of tobacco, drugs, etc

Reportable Diseases

Each US state designates a specified list of diseases that are required to report by law. The MA Department of Public Health provides a Reportable Diseases web site as an online resource for all public health organizations. This data is fed into the National Notifiable Disease Surveillance System (NNDSS).

Data for selected nationally notifiable diseases reported by the 50 states and D.C. and the US territories are collated and published weekly in the Morbidity and Mortality Weekly Report (MMWR).

Descriptive Epidemiology

Descriptive epidemiology is defined as epidemiological studies and activities with descriptive components that are much stronger than their analytic components or that fall within the descriptive area of the descriptive-analytic spectrum.

When new health problems arise, health practitioners need to find details on the determinants. The most basic starting point is to collect detailed information about a single case (case-report) and search for additional persons with similar cases and gather information from a series of similar cases (case series). The most fundamental information focuses on three key aspects: Person, place, and time. These fundamental elements enable us to look for suggestive patterns.

Person - all relevant characteristics including age, diet, behaviors, exposures, etc.
Place - Spatial distribution of the problem.
Time - Whether the frequency of disease varies over time.
Differences - Factors may cause the frequency of disease to differ in circumstances
Similarities - A high frequency of disease found in several different circumstances with a common factor
Correlations - If the frequency of disease varies in relation to some factor

Epidemiological Studies

A study is a scientific process of answering a question using data from a population. The first step is having a question to study, then decide a study type, collect and analyze data (ethically), interpret results, and reporting findings.

Types of studies

Ecological Study

Disease rates and information about exposures are made on a group of people. Could be a household or an entire country. Results apply to groups and not individuals, meaning these studies are better at comparing health between populations and generating questions.

Case Series

describes characteristics of people with the same disease or exposure.

Cross-Sectional/Prevalence Study

Health information of a selected population at a given point in time. Usually involves a questionnaire or survey and are relatively inexpensive, but it cannot determine if exposure caused a disease because the data is collected from a single point in time.

Case Control Study

People with a disease, called cased, and people without the disease but with similar characteristics, called control, are asked about past exposures to certain risk factors. For each risk factor the odds of being exposed as a case is compared to the odds of being exposed if they were control, giving us the Odds Ratio (OR).

	Cases (Have Disease)	Controls (No Disease)
Exposed	a	b
Unexposed	c	d

Odds Ratio (OR) = (a/c) / (b/d) = a*d / b*c = Odds that a Case was Exposed / Odds that a Control was Exposed

Relative Risk (RR) = (a / (a + b)) / (c / (c + d)) = Risk of Disease of Exposed / Risk of Disease Unexposed

If the OR is greater than 1 than the risk factor is associated with disease, less than 1 suggests its a protective factor and =1 suggests no association.

This is often used when searching for food-borne illnesses. They are quick and cheap, and can be used for uncommon disease. It falls short when finding rare exposures, and it can be difficult to find a representative control selection who can accurately recall if they were exposed.

Cohort Study

A group of people are followed over a period of time to collect data on risk factors and then compare occurrences of similar outcomes of those who were exposed and those who were not.

The main measurement used in cohort studies is called the Relative Risk (RR), which is the risk of disease for those exposed to a risk factor over the risk of disease in the unexposed group.

RR > 1 is increased risk, RR < 1 is lower risk, RR = 1 is same risk

Advantages are that time sequence can be determine causality and info on various outcomes/risk factors can be collected at the same time. Disadvantages are the fact that many people have to be followed over a long period of time make them costly and not good for rare diseases, also if too many people drop out it will interfere with the results of the study.

Interventional Study

Intervention is done on a group and outcomes are measured. This could be medicine, vaccine, health advise, etc.

The best design for an interventional study is a randomized controlled trial, where people are randomly selected from the population to either receive the intervention or take part in the control group. Ideally the experiment would be double blind, where neither the participant or investigator know who is the control group.

These studies can provide good evidence of causality, and randomization provides equal chance that groups are similar. Disadvantages are the cost of finding participants.

Summaries

Summaries are studies made with other studies. Systematic review finds all relevant studies to a topic, assesses and interprets the results and creates an unbiased, impartial summary of the evidence. A meta-analysis uses data from studies which asked the same question and had a similar design, then uses that data to combine analysis and produce a single summary result.