MPH Master pieces

Listeriosis epi

Listeriosis is, apparently, the "leading cause of death due to a food pathogen" in the UK. Famously associated with soft cheese, here is an update on its epidemiology.

A gastrointestinal infection

Here is a peek into communicable disease from a different viewpoint: that of the physician trying to treat the patient. I won't spoil the story, but enjoy the discussion section, which covers many of the common infectious causes of diarrhoea.

E coli outbreak report

Here is the formal report on the outbreak of E coli in Germany earlier this year.

I guess you might ask what purpose is served by publishing so long after it is all over, but it's an interesting read nonetheless.

 

Economics of TB control

Here is an interesting evaluation of a tuberculosis control programme in London. The title says it all - it's the Find and Treat service. Table 1 in the paper shows that very few patients are lost to follow up if referred to the service early (though confusingly only 60 - 70% seem to complete treatment: or maybe I've misunderstood that row in the Table.)

Tuberculosis is particularly common among groups particularly likely to default from treatment: the paper cites prevalences of 

"788 per 100 000 in homeless people, 354 per 100 000 in people with problematic drug use, and 208 per 100 000 in prisoners. By comparison, the overall prevalence of tuberculosis in London was 27 per 100 000 people"

And outside London prevalence is likely to be even lower - for example the incidence cited in a recent report for Surrey was 7 per 100 000 (albeit with some worries about underreporting).  (Note that some of these figures are for incidence and some for prevalence. Incidence you get from reporting or notification of new cases, which should be available everywhere; prevalence estimates require a special survey, or some assumptions about the relationship between incidence and prevalence.)

But back to the economics. The paper uses a model, which hinges on transition probabilities between different compartments, for example between being treated and being lost to follow up , or between being treated and being cured. This concept of transition probabilities between different states is very common in economic modelling. Sometimes costings in economic papers are bold guesses but here we seem to have accurate budget figures. The effect of the programme is estimated in years of life and gain in quality, using the standard metric of an EQ-5D score converted into a utility.

And there is a happy outcome after all this modelling - the service comes in under NICE's cutoff of £30k per QALY.   

Peanuts

I seem to be posting a series of communicable disease items. Here is a readable account of a large outbreak in the USA. The outbreak strain was isolated from the suspect foodstuff - that's quite unusual (and maybe it's what made this story publishable). Often, by the time a particular foodstuff is suspected it has all been eaten or destroyed.

Does public health work?

As public health staff in England move from the NHS to local authorities, we expect councillors and their chief executives to question the value of all those highly trained, expensive specialists. Never easy to show that you make a difference so I was interested in this brave attempt. Unfortunately the data turn out to be a lot more messy than I had hoped, and I think some of the authors' assertions are too strong. But we should applaud the research concept.

Part of the problem is that they used two different sources of data for information on food poisoning. Note that one includes viral outbreaks and the other does not.

 

 

TB in London (and elsewhere)

An interesting piece about TB in London - shows how epidemiology informs the planning of health services.

Epidemiology of HIV

Here's a pretty important question: is it ok to have sex with many women provided that you end one affair before starting the next one? Is serial monogamy ok? - in which case the message is "Be faithful (to one woman at a time)".

Mathematical modelling suggested that HIV will spread particularly fast in communities where men typically have a wife and a mistress. This is said to be common in sub-Saharan Africa and so explain the very high rates of infection there.

But this paper, based in the real world, contradicts the model prediction: what matters is how many women a man has had sex with. Serial monogamy is not ok.

The actual method is complicated because the authors wanted to answer a geographical question: rates of concurrent and lifetime partnership vary from community to community and they tried to extrapolate from survey data, which only provides particular data points, rates across an entire area. This entails modelling, and geographical modelling gets complex very quickly. But the end result was that women are more likely to be infected with HIV in communities where the men have a lot of lifetime partners; communities where concurrent partnership is common does not raise the risk (except insofar as it implies a minimum of two lifetime partners).

Molecular epidemiology

Identifying precisely the organism which is causing an outbreak is essential - and nowadays that means molecular typing of the strain. Here is a good account of this process. Like all the best outbreak investigations, it reads like a detective story.

Case case studies

The standard technique for outbreak investigation is the case-control study (though I wish it was called a case-comparison study). We look at the people who have fallen ill, compare them with people who have not fallen ill, and try to spot the difference. If the problem stems from , say, a wedding reception then the study group is everyone who attended (a cohort); those who fell ill are the cases and everyone else at the reception are the controls (a nested case control study).

But how to select the controls in a community outbreak? Standard methods include asking the cases to nominate a friend as a control (perhaps matched for age), or using neighbours in a house-to-house investigation, or if doing the investigation by phone using the number one above and one below the case's phone number. These methods are becoming problematical in today's society and so later this month the Health Protection Agency is running a workshop on new techniques. One such is the case-case method - a name even more unhelpful than 'case-control'. The idea is that you compare the cases with other people who are infected with a different subtype of the same organism. Here is a worked example.

Meanwhile here is an excellent account of an outbreak investigation using classical methods.