The epidemiology of crime

Today we will get back to the epidemiology of crime (see last instalment here).

If the hypotheses that crime is spread by memes and that the causation of crime is multi-factorial are valid, it should be possible to use epidemiological methods to understand the causation of crime – but with great caution.

Let me give a health-related comparison to explain why caution is needed. It was long thought that stomach (or peptic) ulcers in humans were caused by a reaction to environmental stressors. Treatment was primarily palliative, with the aim of reducing symptoms through a controlled diet and by reducing levels of stress.

A study in 1994 produced these conclusions:

“Age, inheritance, and cigarette smoking are all important risk factors for peptic ulcer. The increased risk associated with low educational background indicate that social strains, comprising lifestyle and diet habits, are part of the multifactorial aetiology of peptic ulcer. No support was found for the assumption that peptic ulcer disease is a psychosomatic disorder. This study did not support the view that duodenal and gastric ulcers have different aetiologies-rather it showed a similarity in risk pattern.”

This study was typical of the time and was considered credible, valid and reliable.Now we know (or we think we do, anyway) that most if not all peptic ulcers are caused by a bacterium, namely H. Pylori, with these consequences:
  • 80% of peptic ulcers are cured with dual or triple therapy
  • Two weeks’ treatment with amoxycillin, metronidazole and omeprazole are sufficient to eradicate the bacterium
  • Short term recurrence rates are low
  • Long term recurrence rates are unknown, as yet
  • Drugs have changed the need for ulcer surgery over last 20 years
  • Admissions for elective surgery have significantly reduced
  • The incidence of complications remains unchanged
  • May be increasing due to increased NSAID use in elderly
  • Bleeding and perforation still have mortality of >10%

So, it is important in structuring any epidemiological study not to start with the wrong premise. The 1994 study was valid because it was predicated on the assumption that peptic ulcers were caused by environmental factors – that is then what the study looked for. What you measure is what you get.

Next time we will consider how to avoid falling into this trap when studying the causation of crime.

(Cross-posted at Temporanea)

No comments:

Post a Comment

All comments containing Chinese characters will not be published as I do not understand them