in in older New Zealanders Summer Studentship : : Gis isele - - PowerPoint PPT Presentation

Predictors of repeate ted self-harm in in older New Zealanders

Summer Studentship : : Gis isele Foster

Supervisor: Gary Cheung, Old Age Psychiatrist Department of Psychological Medicine

Shah et al. (2015) Aging & Mental Health

Male Suicide Rates 2001 to 2010

10 20 30 40 50 60 0- 5- 10- 15- 20- 25- 30- 35- 40- 45- 50- 55- 60- 65- 70- 75- 80- 85+ 5-year age group Rates per 100,000

NEW ZEALAND

Late-life Suicide: Risk Factors

• psychiatric illness (particularly depression)
• history of previous suicide attempts
• physical illness
• pain
• functional impairment
• social disconnectedness

Cattell, 2000; Conwell and Thompson, 2008; Conwell et al., 2011; Fassberg et al., 2012

Suicidality: A continuum

Death wish Suicidal ideation Suicide Plan Suicide attempt Completed suicide 28% 11.5% Choi et al., 2015

2-15.9%

9 to 18% older people who had made a suicide attempt would make further attempt within 12 months

Study: Aims

(i) to characterise older suicide attempters in Auckland, Hamilton, Wellington and Christchurch; and (ii) to determine the predictors of repeat suicide attempt in 12 months.

Methods: Inclusion criteria

• Age ≥ 65
• presented to ED following an episode of

self-harm from 1st July 2010 to 30th June 2013

• 7 locations

– Auckland City Hospital, North Shore Hospital, Waitakere Hospital, Middlemore Hospital, Waikato Hospital, Wellington Hospital and Christchurch Hospital

Methods: Databases

(i) the existing electronic coding for “self- harm”, “suicide attempt”, “deliberate self- harm” in the emergency department and medical records department e.g. ICD10 Codes X60 to X84; (ii) the consultation-liaison referral database

Methods: Exclusion criteria

• accidental self injury/overdose e.g. took the

wrong medication purely by mistake

• those who died from a suicide attempt (i.e.

these cases are classed as suicide)

• repeated episodes during the 3 years

period

Methods: Classification of self-harm

• The Columbia Classification Algorithm of

Suicide Assessment (C-CASA) (Posner et al., 2007):

(i) Suicide attempt (ii) Self-injurious behaviour, no suicidal intent (iii) Self-injurious behaviour, suicidal intent unknown

Summer Studentship Gisele Foster

• To combine data from the 7 hospitals and

perform statistical analysis

• Learning included data management, SPSS,

statistics, referencing, academic writing

Results: 1

• 339 older people (age 65-96; mean=74)
• Female 55.2%
• European 92.3%
• Married 44.2%; widow 25.4%
• Lived alone 38.1%

Results: 2

• Non-psychiatric hospital admission in

previous 12 months: 46.9%

• Past mental health history: 64.6%
• Under MH service at time of self harm: 33.6%
• Depression: 55.8%
• Depressive symptoms: 17.7%
• Past attempts: 34.2%
• Positive blood alcohol: 14.2%

Results: 3

C-CASA Classification Methods

Suicide attempt 76.5% Self-injurious behaviour, no suicide intent 13.0% Self-injurious behaviour, suicide intent unknown 10.3% Overdose 68.7% Laceration 9.1%

Acute stressors associated with self harm

20 40 60

death of 1st degree relative percieved physical illness terminal illness in 1st degree … family discord changed relationship/death of … separation financial trouble employment change legal difficulties

Results: prognosis in 12 month

7 (2.1%) committed suicide 50 (14.7%) repeated self-harm

Results: repeat self-harm in 12 months

Repeat Self-harm within 12 months P-value Odds ratio Yes No Under Mental health* 58% 29.4% 0.000 3.1 Past attempts 46% 32.2% 0.039 Taking antidepressant 70% 46.4% 0.001 Positive blood alcohol* 28.0% 11.8% 0.003 3.5

* Remained significant in final logistic regression model; Chi2 =15.4, p=0.032, df=7, R2 = 16.9%

Conclusion

• Predictors of self harm: under mental

health; positive blood alcohol

• Role of physical illness in suicidal behaviour

in older people

• A very risk group

– Suicide 2100 per 100,000 people

• Suicide rate for the general older adult population

= 6 to 19 per 100,000 people