16 Feb 2017
by Lesley McCollum
A new study published in the February issue of the American Journal of Psychiatry reports that the rate of new psychosis cases in rural England is much higher than previously estimated, indicating substantial demand for early intervention services. In the largest first-episode epidemiological study since early intervention services were introduced in England, James Kirkbride of University College London and colleagues found that within rural communities, people of ethnic minorities or from the most densely populated or deprived neighborhoods have an increased risk of developing psychosis.
“This kind of information is really vital,” said Kirkbride. Although early intervention services have been established in the UK for some time, the planning of those services has been challenging. Kirkbride added that the findings in the new study should allow services planners to allocate resources better and therefore improve the efficiency with which services work, which will hopefully translate to better care for patients.
Research has indicated that treating psychosis early can improve the functional and clinical outcomes for patients. However, the need for early intervention services has mainly been informed by epidemiological data collected before their introduction. Additionally, the estimation of services is often based on the narrow definition of schizophrenia. The new study included a broader spectrum of psychosis, including affective psychoses, such as bipolar disorder or psychotic depression, which accounted for 12.3 percent of the cases.
“It’s very compelling,” said Susan Essock of Columbia University and New York State Psychiatric Institute, who wrote an editorial accompanying the article. “It’s a wonderful use of available data,” she said, noting that the public health system in the UK and the widespread specialized clinics in the area studied provide confidence that the people entering those services represent the underlying population.
Improving estimates of psychosis incidence
Previous epidemiological research has focused heavily on urban areas (see SRF related story), limiting incidence estimates to just one end of the rural-urban spectrum. In this study, Kirkbride and colleagues broadened these estimates to more rural areas by examining a mixed rural and urban setting in the east of England. The study included 687 participants, aged 16 to 35 years old, who presented to early intervention services and went on to develop a psychotic disorder. The researchers estimated an incidence of 34 new cases per 100,000 people per year, compared with previous estimates of about 15 per 100,000 people per year.
“We can now combine information from rural areas with information we already have from more urban areas to make more accurate predictions about how many people with psychosis you might see in different parts of the country,” said Kirkbride.
Consistent with previous studies (McGrath et al., 2004; Kirkbride et al., 2012), patients were more likely to be men, younger, from an ethnic minority background, of lower socioeconomic status, and from more deprived and densely populated neighborhoods. In particular, ethnic minorities had a 1.5-fold higher risk than the baseline white British population, and the most densely populated neighborhoods were associated with a 1.4- increased risk. The largest increase in incidence occurred in neighborhoods with the highest proportion of households disadvantaged for at least two of four factors—employment, education, health, or living environment—where the risk for psychosis more than doubled.
“The most surprising part of these findings was the fact that population density and deprivation didn’t show a strictly linear association with psychosis risk,” said Kirkbride, noting that risk seems to be concentrated in the most populated and disadvantaged environments. The findings suggest there may be a threshold of exposure necessary to have an effect.
The analysis was unable to identify a causal link for the associations, which remains an important etiological problem. For example, the study cannot rule out the reverse explanation—that people with increased risk may tend to move to more populated areas. However, the findings are consistent with research investigating environmental risk factors, such as ethnicity and urbanicity (see related SRF stories here and here; see also SRF online discussion), and the consistency in findings hints that there may be grounds for causation.
“Evidence continues to pile up suggesting that environmental risk factors increase the likelihood of schizophrenia emerging in the presence of genetic risk,” said Essock, which she highlights in her editorial.
Informing implementation of early intervention services
Detailed epidemiological data have been difficult to acquire in the US, where the nature of the healthcare system provides just a narrow slice of the larger population. But Kirkbride notes that in the broader literature around psychosis risk, evidence for increased risk in urban areas and more deprived communities is fairly consistent across a variety of regions. So although they can’t say for sure that the findings will generalize to the US, the authors hope that their findings can help in an American context.
Essock is also hopeful that the study will impact implementation of early intervention services in the US, which are just starting to take hold. Initiatives such as the Recovery After an Initial Schizophrenia Episode (RAISE) program (see related SRF related story) have been important for establishing the efficacy of early intervention care in a US context, and Essock points out that the findings of this study have huge practical implications for where to locate teams to serve individuals with first-episode psychosis.
“This work lets us know that when we want to get new teams up and running, if we want to cover a population, we need to put more such teams in these areas of very high population density and high adversity,” said Essock, adding that this will be important for ensuring equal access in areas with increased incidence of the disorder.