• Login
• Admin

Alothman, Danah (2024) Clinical, behavioural and social indicators of suicide risk: a population-based nationally representative longitudinal study from England. PhD thesis, University of Nottingham.

PDF (Thesis - as examined) - Repository staff only until 11 December 2026. Subsequently available to Repository staff only - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader Available under Licence Creative Commons Attribution Non-commercial No Derivatives. Download (5MB)

Abstract

Background

Suicide is an extremely important public health issue, and its prevention depends on understanding factors that contribute to it. However, the currently recognised risk factors are unlikely to be comprehensive, and most are weak predictors of suicide mortality. Given that most individuals who die of suicide have consulted primary care in the year prior to death, primary care can be an appropriate place for recognising those at risk of suicide death, and hence early intervention. We hence assessed primary care datasets in combination with other electronic heath records looking in depth at under-reported social (ethnicity as well as living alone or loneliness), clinical (dementia diagnosis; Huntington’s disease; analgesics prescribing; and self-harm), and behavioural (patterns of and reasons for primary care consultation) indicators of suicide risk.

Aim and objectives

The overall purpose of this work was to identify opportunities for suicide prevention by examining various indicators of suicide risk using electronic primary care records interlinked with other electronic health records. Our objective was to quantify associations between selected and novel factors and suicide risk, examining effect modification by important subgroups as well.

Method

This was a population-based nationally representative case-control study in England conducted between January 2001 and December 2019. Data were obtained from multiple linked electronic records: Clinical Practice Research Database (CPRD) for primary care records in the UK, Hospital Episode Statistics for secondary care records in England and the Office for National Statistics (ONS) for national mortality records in the UK. As suicide is rare before the age of 15, we included all eligible patients aged 15 years or older registered in the ONS in England with a death coded as suicide or open verdict between 2001 to 2019 (n=14515). For every suicide case, up to 40 live eligible controls were randomly matched on the suicide date (risk-set sampling) and the primary care practice of the corresponding case (N = 594674). Odds ratios (ORs) were estimated using conditional logistic regression.

Results

From the total sample of 594674 patients, 580159 (97.56%) were controls (median [IQR] age at index date, 49.71 [35.94-64.60] years; 289769 male patients [49.95%]), and 14515 (2.44%) died by suicide (median [IQR] age at suicide date, 47.43 [36.03-59.72] years; 10850 male patients [74.75%]). The following were our main findings:

1) Ethnicity: Asian, Black and Other ethnic groups had a significantly lower suicide risk compared to White individuals (reference category), with those of Asian ethnicity having the lowest risk (OR 0.53, 95% Confidence Interval (CI) 0.47-0.60). This ethnicity related suicide risk was significantly modified by age, sex and socioeconomic status.

2) Living alone or loneliness: After adjustment for age and sex, the risk of suicide in individuals who had previously been reported to be either living alone or suffering loneliness was increased (Odds ratio OR 4.89; 95% CI: 4.36 to 5.47). Age substantially modified the level of this risk, with individuals aged 15 to 34 years who were lonely or lived alone having a much higher risk of suicide (OR 16.43; 95% CI: 8.66 to 31.15).

3) Dementia: Compared to patients without dementia, suicide risk was significantly increased in patients diagnosed with dementia before age 65 (age and sex adjusted OR 2.82; 95% C: 1.84-4.33), and within three months of receiving a dementia diagnosis (adjusted OR 2.47; 95% CI: 1.49-4.09). In patients younger than 65 years and within 3 months of diagnosis, suicide risk was 6.69 times (95% CI, 1.49-30.12) higher than in patients without dementia.

4) Huntington’s disease: The age and sex adjusted OR for suicide in patients with Huntington’s Disease was 9.2 (95% confidence intervals, CI 4.9–17.4) compared to those without Huntington’s disease. The increase in risk was higher amongst the younger age group who were ≤45.8 years (median age at suicide for patients with Huntington’s disease) (OR 54.5, 95% CI 10.8–276.1).

5) Prescribed analgesics: In the final year, suicide risks were highest in those prescribed adjuvant analgesics (pregabalin, gabapentin and carbamazepine) (adjusted OR 4.07; 95% confidence intervals CI: 3.62–4.57), followed by those prescribed opioids (adjusted OR 2.01; 95% CI: 1.88–2.15) and those prescribed non-opioid analgesics (adjusted OR 1.48; 95% CI: 1.39–1.58) compared to those not prescribed these medications. Across all three main analgesic categories, suicide risk increased linearly with the number of analgesic prescriptions in the final year (p < 0.01 based on the likelihood ratio test).

6) Self-harm: In both the primary care and secondary care datasets, the one-year relative suicide risk following self-harm was higher in this population-based study than previously reported in the literature (age and sex adjusted OR 310; 95% CI: 244-393 in primary care records; adjusted OR 124; 95% CI: 111-138 in secondary care records). The risk was highest in the first month after self-harm in primary care data (OR 193; 95% CI: 169-221) and secondary care data (adjusted OR 458; 95% CI: 331-633). Suicide risk increased with the number of self-harm events requiring hospitalisation (p<0.0001 for trend); those with ≥3 self-harm attempts in the previous year were at 505-fold increased risk (95 % CI: 227-1123) compared to those with no self-harm records. While suicide risk following self-harm was increased across all age-groups and sexes, it was higher in females and older individuals compared to their counterparts with no self-harm records (p<0.05 for age and sex interaction).

7) Patterns of and reasons for consultation with primary care: Frequent consultations (more than once per month in the final year) were associated with increased suicide risk (age and sex adjusted odds ratio 5.88; 95% CI: 5.47-6.32). The associated rise in suicide risk was seen across all sociodemographic groups as well as in those with and without psychiatric comorbidities. Medication review, depression and pain were the commonest reasons for which patients who died by suicide consulted in the year before death.

Conclusion

This large scale population-based study from England has assessed relatively novel indicators of higher suicide risk. This should inform suicide risk assessment in both primary care and secondary care settings. While we demonstrated associations between selected exposures and suicide risk, it is important to note these may not be causal effects. We recommend future research to understand mechanisms that may link those exposures to suicide risk which may help in the development of interventions to reduce suicide mortality.

Item Type:	Thesis (University of Nottingham only) (PhD)
Supervisors:	Lewis, Sarah Fogarty, Andrew Card, Timothy Tyrrell, Edward
Keywords:	suicide; electronic health record; database; primary care; general practice; risk factor; ethnicity; lonely; dementia; Huntington's disease; analgesic; self-harm; consultation patterns
Subjects:	W Medicine and related subjects (NLM Classification) > WM Psychiatry
Faculties/Schools:	UK Campuses > Faculty of Medicine and Health Sciences > School of Medicine
Item ID:	79863
Depositing User:	Alothman, Danah
Date Deposited:	11 Dec 2024 04:40
Last Modified:	11 Dec 2024 04:40
URI:	https://eprints.nottingham.ac.uk/id/eprint/79863

Actions (Archive Staff Only)

Edit View