Tag: epidemiology

Defining a population

Posted on by Prav

Measuring disease and health outcomes, Part 2 – Population

As mentioned in part 1 understanding frequency alone can be misleading – which is why we need to know our population.

The fact is that health outcomes can differ for a variety of reasons. From different countries, regions, socio-economic backgrounds and even time. Therefore to study an outcome we first have to clearly identify who we want to include – this is our Target population.

Our target population is basically the individuals we want to identify or whose health we want to improve. For example being concerned about the population of Hull City. However if we were to look at the all 200,000 people in Hull City – we would probably be overwhelmed as realistically it would be too difficult to gather and quantify all that data.

To make things easier we identify a Study Population. And this would be a smaller subset of individuals with a particular disease/health outcome we would like to study. For example the study population would be the number of individuals over the age of 30 who are diabetic. If this sample is too large – lets say 50,000 individuals, then an alternative method of collecting information is needed. Which is why we use a Study Sample, which is a smaller number of the Study Population, selected at random to represent the 50,000 individuals.

Target Population = Individuals in Hull City
Study Population = Diabetics > 30 years of age living in Hull ( eg = 50,000 identified)
Study Sample = 2000 individuals chosen at random from the 50,000, to represent the diabetics over 30 years of age.

Why do we need to define our populations?

So we know who are the population at risk. These are individuals who have the risk factor for the condition being studied regardless of whether they are currently healthy or ill. This can be clear cut or vague depending on what is being studied.

For example, if the population at risk for developing lung cancer is smokers, they are clearly identifiable cause we can assess smoking histories. But in some instances such as rare diseases we cannot always identify who is at risk for a variety of reasons. In these instances we use the whole population as an estimate.

So once we identify the population, we can figure the frequency of an incident taking place. Ie if there are 500 cases of lung cancer out of a study population of 4000 smokers(population at risk), then we know the frequency is 1/8.

But let’s say we only have 2 researchers and 4000 is too large a group to manage, this is when we use a study sample. So 1000 smokers are identified at random to represent the 4000. Out of the 1000 smokers, we find 120 cases with lung cancer. The frequency in this instance would still be 1/8 or 120 per 1000 cases. Which when compared with the study population would still give you 500 cases of lung cancer out of the total population at risk of 4000 smokers.

In the next post, we will talk about how we can express frequency – incidence vs prevalence.

Study design – Observational

Posted on by Prav

One of the most important facets of public health is understanding epidemiological studies. I might also add that it is the one thing I repeatedly have to look over and a lot of my peers and colleagues find confusing and struggle with.

A quick recap of what epidemiology is – Put simply, it deals with figuring out the who, what, when, where, why and how certain diseases happen and telling us the way we can overcome/stop it.

Now epidemiology is actually made up of both Observational and Interventional studies, and to start we will talk about observational studies.

They are quite self-explanatory, as they are just that – observational. All we are doing is essentially just having an overall look. However, there are different ways to do this, and that is by being either descriptive or analytical.

So what are the main differences between these two types of research/studies we can carry out and what do we gain from it? I prefer the 5W’s and 1 H method of thinking about it.

Descriptive Analytical
Who?
What?
When?
Where?		 Why?
How?

So descriptive studies, look at an overall picture. It tells us what is going on, what is involved, who is involved, where it is happening and when without telling us Why or How. We basically go in without an idea for cause and effect. It essentially helps us identify them by examining patterns and by giving us an overall idea of the population, the distribution of health based on age, gender, location and time/over a period of time. It is from this, that we might identify a problem leading to ideas for new studies to figure out the why, how and perhaps even move on to an interventional study.

If let’s say I was selling chocolate and I wanted to know more about my customers, my initial descriptive study would tell me about the people who are buying my chocolates, where they live etc. I might learn, that only hipsters in their late 20’s buy my brand of chocolate but I still do not know why (well maybe because it’s unheard of?). My next task is to figure out why and perhaps how I can make my chocolate more appealing to different groups.

Examples of descriptive studies can be further broken down to cross-sectional study=ies like a health survey, ecological studies or even case reports/case series. Remember all it does, is present the facts for what they are and is a starting point for us to make associations and come up with new ideas.

Analytical studies then basically go into how this is happening and why? It is one of the ways to investigate causal relationships. So in these studies, I have a hypothesis/an idea. A health-related example would be that ‘smokers have a higher risk of lung cancer than non-smokers. We then investigate if this is true or not. How we go about this, is either with a case-control study, cohort study, cross-sectional study or an ecological study.

Something visual to help, the rest will be revealed as we move on.

How come cross-sectional and ecological studies are in both descriptive and analytical studies? Well, I will go into that when I talk more about the different types of studies mentioned in the next post.

Disease Detectives?

Posted on by Prav

One of the first activities in my course module – Issues in Public Health was looking at the role of an epidemiologist.

What is an epidemiologist?

The explanation I personally love was by the CDC* who described them quite aptly as ‘Disease Detectives’. These sleuths search for clues looking for the cause of disease, try and identify who are the people at risk and then try to figure out how we can stop the disease from spreading and preventing a future episode.

And while familiar, I definitely did not realise that there were some differeing opinions on what the focus of epidemiologists or public health should be.

The critique has been that the focus has been to much on individual risk factors while excluding all the other factors that contribute to disease.

But should public health professionals be focusing on the social determinants of health solely OR other individual risk factors?

A good example used was in discussing how to combat Malnutrition. How do we solve this and what should the focus be?

Do we take a step back and address the other social issues contributing to poverty leading to malnutrition. Or do we spend time and effort finding out that by supplementing Vitamin A in these individuals we greatly reduce their risk of blindness and death by starting them on Vitamin A supplements? While we haven’t solved the issue of Malnutrition per se, I feel it is better to be doing something that reduces the negative outcomes cause the reality is that pushing for those changes with bureaucracy can be long-drawn out. So, why not assist and intervene where we can?

However, this of course does mean that we are diverting our focus, energy and more importantly our already limited resources. And is it then the best utilisation of our resources? – now that’s a concept in Health Economics that we will discuss in another post.

But what do you think?

*CDC – Centers for Disease Control and Prevention

Public health at the crossroads: which way forward?
https://www.thelancet.com/journals/lancet/article/PIIS0140-6736(97)09494-4/fulltext
–> this article is available for free

Should the mission of epidemiology include the eradication of poverty?
https://www.thelancet.com/journals/lancet/article/PIIS0140-6736(98)01327-0/fulltext
–> this is unfortunately paid unless you have access