Our methods

  • MethodsBanner

Our research is focussed on identifying the social and environmental factors which increase a person’s risk of developing schizophrenia and other psychiatric disorders in their lifetime. To address this problem we design large studies of the population and examine whether differences exist between various groups of people according to certain risk factors we think might be related to particular disorders (such as whether men experience schizophrenia more than women).

The broad name for this type of scientific research is called epidemiology. Epidemiology is concerned with identifying the causes of diseases which affect populations. One central goal of epidemiology is to identify these causes (such as the link between tobacco smoking and lung cancer) in order to prevent or lower the occurrence of disease. The branch of scientific research which aims to improve the health of populations (by targeting large groups of people with prevention strategies, for example by launching campaigns to quit smoking) is called public health.

A sub-branch of epidemiology, which deals with the particular challenges of psychiatric disorders is known as psychiatric epidemiology. The PsyLife lab uses the principles of epidemiology and psychiatric epidemiology to conduct our research. By carefully evaluating the size, timing, strength, validity and reliability of differences in risk between different groups we can identify potential clues as to which social or environmental factors might be related to different psychiatric disorders. The Lab’s research is mostly based on observational research studies (you can find out more about these on our epidemiology page). We use a variety of observational study designs to answer our questions, depending on the project or question, and these have different strengths and limitations:

  • PsyLife projects use observational data from longitudinal (or “cohort”) studies. These studies follow a large group of people over time (who initially don’t have the disorder of interest, for example, schizophrenia). Data on different risk factors is collected on these people before they get the disorder and we then observe who goes on to develop the disorder. We use statistical approaches to determine how important these differences might be. The major advantage of these types of study is that they collect information on people before they have the disorder, so the causal direction of the association between exposure and disorder is clearer than in studies where people might be asked retrospectively about  their exposure to different risks in their past. The major disadvantage is that they require long follow-up times (often decades) and large samples.
  • SEPEA uses a cross-sectional study design to examine the incidence of psychotic disorder at a particular point (or short period) in time. Data on the exposure and the outcome are collected at the same time. This method is useful for establishing the incidence rate in a population but it can be difficult to establish whether the exposure occurred before the disorder. Differences are compared using suitable statistical methods for incidence data.
  • The EU-GEI study has two designs: there is a cross-sectional incidence design, similar to the SEPEA study, and a case-control design. Case-control studies sample a group of people with a disorder (like schizophrenia) and a representative sample of people without the disorder (i.e. “controls”). Both cases and controls are asked about potential risk factors they may have experienced and differences are compared using statistical methods suitable for case-control studies. The advantage is that these studies are easier and cheaper to conduct than cohort studies, but questions about social or environmental risks are often asked retrospectively, which may introduce various issues.

You can read more about epidemiology, statistics and public health on the relevant pages.