COMMITTEE FOR
HEALTH, SOCIAL SERVICES AND PUBLIC SAFETY

OFFICIAL REPORT
(Hansard)

Inquiry into the Prevention of Suicide and Self Harm

28 February 2008

Members present for all or part of the proceedings:
Mrs Michelle O’Neill (Deputy Chairperson)
Mr Thomas Buchanan
Dr Kieran Deeny
Mr Alex Easton
Mr Tommy Gallagher
Mr John McCallister
Ms Carál Ní Chuilín
Ms Sue Ramsey

Witnesses:
Dr Dermot O’Reilly ) Queen’s University Belfast

The Deputy Chairperson (Mrs O’Neill):
I welcome Dr Dermot O’Reilly from Queen’s University Belfast, who has recently co-authored an article on suicide risk. Members will have a copy of Dr O’Reilly’s article.

Dr O’Reilly, I invite you to make a short presentation, after which members will ask questions.

Dr Dermot O’Reilly (Queen’s University Belfast):
Thank you; I shall make a short presentation.

We set out to answer one simple question: are suicide rates affected by where someone lives? Throughout Northern Ireland, and, indeed, in most European countries, there are variations in suicide rates. The question that has been asked is whether those variations have something to do with the areas in which people live, or whether they simply reflect the types of people who live in those areas.

It could be said that people in rural areas are isolated. However, there is a good case for saying that people in certain parts of the cities are also isolated. Cities can be considered to be hostile and antisocial places, which are factors that may increase the risk of suicide. Furthermore, it could be suggested that living in deprived areas is itself depressing, and that may also increase the risk of suicide. Therefore, a case could be made that the character of an area in which people live adds to their risk of suicide.

The question of whether the areas or the type of people who live in those areas increase the risk of suicide could not have been answered previously for Northern Ireland. However, we now have a new data set, called the Northern Ireland Longitudinal Study, which allows us to answer that question and which provides a wonderful tool that enables us to learn more about the health of the population, and, in this instance, about suicide rates. Committee members may want to ask me about that study.

The characteristics of individuals and of the areas in which they live need to be considered. Those two aspects then need to be separated. The longitudinal study starts with the census, which is a rich data source. The mortality statistics for the five years after the census was taken are then examined. Therefore, five years after the census is conducted, we can find out who has died and who is still alive. The factors that are associated with an increased risk of suicide can then be gleaned from that information.

We considered individual factors — the sort of information that the census provides, such as age, gender, martial status, and indicators of the extent of deprivation or disadvantage of the household in which they live. We also considered economic activity, as it is known that unemployment and economic inactivity increase the risk of suicide.

We then added three indicators about the character of the area in which people live. The first concerned deprivation, which, rather than concentrating on the family, asks whether the area itself is disadvantaged. The second was about urban and rural residence; therefore, we used population density, which is a good proxy for that indicator. The third was about social fragmentation, which could be considered as the opposite of social capital, about which the Committee is probably aware. Social fragmentation is evident in a community that is not close-knit and in which people feel isolated and have no sense of cohesion or trust.

More recent research that was published in the British Medical Journal (BMJ) suggested that social fragmentation was even more of a risk factor for suicide than deprivation. That research found that areas with high levels of social fragmentation were more likely to have higher suicide rates. Furthermore, it showed that the more fragmented that areas became, the more likely it was that suicide rates would increase.

By examining individual and household characteristics, we confirmed that the factors that have been shown to increase the risk of suicide in other countries were also relevant to Northern Ireland. Our research found that young people are more at risk of committing suicide than older people; males are three times more likely to commit suicide than females; and single people, including those who are divorced or separated, are at higher risk than those who are in a relationship. I have provided the Committee with more detailed data on that.

Our research also showed that people who live in deprived households are five to six times more likely to commit suicide than those who live in more affluent households. Unemployment also increases the risk significantly. Finally, there is a suggestion that people who live alone are at greater risk of suicide than those who do not, and that could be linked to a sense of social isolation.

We then considered area characteristics. We replicated the analysis that was conducted some years ago in Northern Ireland that showed that people in deprived areas and socially fragmented areas are at a higher risk of committing suicide. We could not, though, identify an urban-rural gradient. However, there was a suggestion — but only a suggestion — that suicides rates tended to be a little higher in both the most urban and the most rural areas, but they did not reach the levels of statistical significance that may have been expected.

When the individual and area factors were added in to the analysis, the area factors lost their significance. Therefore, the conclusion was that individual and household characteristics are important, not where an individual lives, or whether that area is deprived or affluent, urban or rural, or socially fragmented. Let me put the conclusions another way: if a person is young, male, and unemployed, irrespective of where they live — whether their area is deprived or affluent, socially fragmented, urban or rural — the risk of suicide for that individual is high. If a person is older, female, and in a relationship — regardless of where they live — their risk is lower.

The Deputy Chairperson:
OK, thank you very much. May I ask whether your findings are at variance with other research, or does it support your conclusions?

Dr O’Reilly:
There is no equivalent research in the UK. We could have conducted the research in England, Wales, or Scotland, but because of the Northern Ireland Longitudinal Study, we decided to attempt to find a Northern Ireland solution to a Northern Ireland problem. Our research is in keeping with the wonderful record-linkage studies of some Scandinavian countries. Researchers in Denmark and Finland, where the populations are larger, have examined the issue for several years. Their findings are similar to ours: it is who you are rather than where you live that affects your suicide risk.

Ms Ní Chuilín:
Thank you very much for your presentation. I am surprised at your findings because I would have thought that where an individual lived had a bigger influence on suicide risk than you have claimed. I suppose that your conclusions were reached after an assessment was carried out of the various factors that contribute to people ending their lives. However, I am not entirely clear: are you saying that the social, rather than the physical, factors have the real influence? Do you believe that a person who lives in a high-rise block of flats is at no more risk of committing suicide than someone who lives in a house in the same area?

Dr O’Reilly:
Areas can be characterised in different ways. We have looked at three such characteristics: deprivation, urban-rural, and fragmentation. I am sure that you could come up with other traits that we have not measured that influence suicide risk. However, we cannot be sure about those traits.

Ms Ní Chuilín:
In my constituency of North Belfast, which is a very deprived area that has a massive housing problem and lots of other social problems, there is a lot of self-harming and mental ill-health. The suicide rate is high, particularly among young people.

When we were in Scotland, examining evidence and sharing experiences, we found that older people living in rural communities, particularly the Highlands, were at the greatest risk of suicide. Therefore, we automatically thought that where a person lived was a massive contributory factor. I am not contradicting what you are saying; I am not in a position to do that. However, having read your paper and heard your testimony, I am surprised that you do not consider that where a person lives has a greater impact on their risk of suicide.

Dr O’Reilly:
I concur with your feeling. Understandably, deprived areas have the highest rates of suicide in Northern Ireland. People who are deprived and live in materially disadvantaged households are, by and large, based in impoverished areas. Therefore, the rates of suicide in deprived areas are higher, not because the area itself is destitute, but because there is a high concentration of deprived people in that locality. We are merely suggesting that if those people who are deprived happened to live somewhere more affluent, they would be at the same risk of suicide.

Ms Ní Chuilín:
Do you believe that it is the individual rather than the environment that affects the risk of suicide?

Dr O’Reilly:
If you walk or drive around deprived areas, they are depressing to look at. However, that does not seem to add anything to an individual’s risk. I should also point out that the risk factors for self-harm and suicide are not quite the same. The epidemiology and risk factors overlap to a certain extent, but there is a distinct epidemiology behind both self-harm and suicide. Therefore, we cannot make a direct comparison.

The findings for Northern Ireland might not apply in other parts of the world. Our findings equate to those from some Scandinavian countries. However, in certain parts of the United States, suicide rates tend to be higher in rural than in urban areas, but in other parts of the world that is reversed. To some extent, the results depend on the country.

Dr Deeny:
Thank you, Dermot. I am interested in this topic, as a GP and as a member of the Committee. I am a little surprised, but not completely shocked, at what you say about social fragmentation. We tend to think that rural communities are strong and that people help each other, but some suicides still shock us completely, so what you say does not entirely surprise me.

Given that you attribute the risk factors to individual circumstances, does that suggest that we should examine families that have a history of suicide? Your research looked specifically at areas, but it does not explain why so many more men than women — especially young men —commit suicide and why many more young women self-harm.

Dr O’Reilly:
No; that was not the question that the research set out to answer. There is a tendency to examine individual suicides from which to extrapolate theories. That cannot be done. The circumstances surrounding each suicide are unique. We have tried to draw out the general patterns from the 566 suicides in Northern Ireland that we examined. One or two cases may not fall into the pattern that we have established, given that those people were influenced by what happened locally. Our results show what happens in general. We cannot say what the risk factors are for individuals or why men, especially young men, are more at risk of suicide.

Another study has been commissioned by the Northern Ireland research and development office, of which the Committee will know. That is a psychological autopsy study that will try to understand, as far as possible, the reasons behind suicide in individual cases.

Dr Deeny:
Is it possible that family history plays a part?

Dr O’Reilly:
I really cannot answer that from the data, and I do not want to speculate. I must be careful, because some people might think that there is a genetic propensity towards suicide. I suggest that, if suicides happen to occur in certain families, the reasons are much more likely to be environmental. Families share much more than genes; they share an environment as well as levels of material disadvantage, isolation and so on.

Ms S Ramsey:
On that point, I know that Dr Deeny was not implying anything, but we must be careful not to scaremonger. We are dealing with several families who are suffering because of suicide or self-harm, and we do not want to add to the pressure that they are under by assuming that other family members could take their own lives. We must conscious of that.

The Deputy Chairperson:
How do your research findings correlate with the suicide-prevention strategy? Do you feel that, based on those findings, the strategy must go in a different direction?

Dr O’Reilly:
No. About five or six years ago, there was a suggestion from the research field that we needed to take the policy initiative in a different direction and that we needed to examine specifically certain factors at an area level. This is where it gets tricky. When I say area-level factors, I do not mean that we should not concentrate on areas. It is evident that we should do that, because, for example, there are likely to be many people at risk in deprived areas. However, we should not aim to change the characteristics of areas and expect that to have an impact on suicide risk. The evidence from my study, and from the Scandinavian studies, suggests that that will not have a significant impact on suicide risk.

The Deputy Chairperson:
As members have no more questions, I thank you for coming today to help us with our inquiry.