The association of physical multimorbidity with suicidal ideation and suicide attempts in England: A mediation analysis of influential factors

Abstract

Background:

Suicide is one of the most important causes of deaths in the United Kingdom, and the numbers are currently increasing.

Aim:

There are numerous identified determinants of suicidality, and physical multimorbidity is potentially important but is currently understudied. Thus, this study aims to investigate the association of physical multimorbidity with suicidality.

Methods:

Cross-sectional data from the Adult Psychiatric Morbidity Survey 2007, which was conducted in England between October 2006 and December 2007 by the National Center for Social Research and Leicester University were analyzed. Respondents were asked about 20 physical health conditions, and suicidal ideation and suicide attempts were assessed.

Results:

Out of 7,403 individuals aged 16 years or over, the prevalence of physical multimorbidity, suicidal ideation, and suicide attempts were 35.1%, 4.3%, and 0.7%, respectively. After adjustment for potential confounders, compared to no physical conditions, 1, 2, 3, and ⩾4 conditions were associated with significant 1.79 (95% CI [1.25, 2.57]), 2.39 (95% CI [1.63, 3.51]), 2.88 (95% CI [1.83, 4.55]), and 6.29 (95% CI [4.12, 9.61]) times higher odds for suicidal ideation. Mediation analysis showed that cognitive problems (mediated percentage 39.2%) and disability (37.5%) explained the largest proportion between multimorbidity and suicidal ideation. Pain (38.0%) and cognitive problems (30.7%) explained the largest proportion between multimorbidity and suicide attempts.

Conclusion:

In this large sample of UK adults, physical multimorbidity was associated with significantly higher odds for suicidal ideation and suicide attempts. Moreover, several potential mediators were identified, and these may serve as future targets for interventions that aim to prevent suicidality among people with physical multimorbidity.

Keywords

Multimorbidity suicide suicidal ideation suicide attempt United Kingdom adult

Introduction

Considering the high and increasing prevalence of suicides in the UK, it is of utmost importance to identify risk factors of suicidal ideation and suicide attempts to aid in the development of targeted interventions. While there is a multitude of identified determinants of suicidal behaviors, one potentially important but understudied risk factor is that of physical multimorbidity.

In the UK, multimorbidity is highly prevalent with data estimating that it affects one in four adults (Stafford et al., 2017). Physical multimorbidity (i.e. ⩾2 physical conditions) may increase the risk for suicidal ideation and suicide attempts via, for example, social isolation, disability, perceived burdensomeness, and financial difficulties (Xiong et al., 2020). In terms of the association between physical multimorbidity and suicide attempts, there are much fewer studies. Furthermore, there are currently no studies on physical multimorbidity and suicidal thoughts and behavior from the UK, while the mediators of this association are largely unknown. It is important to identify such mediators as they can act as targets for suicide prevention interventions.

Thus, the present study aimed to investigate associations of physical multimorbidity with suicidal ideation and suicide attempt in a large sample of English adults. A further aim was to investigate to what extent a broad range of psychological, physical, and environmental factors mediate the physical multimorbidity-suicidal ideation/suicide attempt relationships.

Methods

Cross-sectional data from the Adult Psychiatric Morbidity Survey (APMS) 2007 (NHS, 2021) were analyzed. The survey was conducted in England between October 2006 and December 2007. To obtain a nationally representative sample of the adult population aged ⩾16 years old residing in private households, multistage stratified probability sampling was used. The small user Postcode Address File (PAF) served as the sampling frame with postcode sectors serving as the primary sampling units (PSUs). Sectors were stratified by both region and socioeconomic status. One person was invited to participate from each household that was randomly selected. Information was obtained through computer-assisted personal interviews (CAPI) and computer-assisted self-interviews (CASI). The survey response rate was 57%. To correct for the probability of selection and survey non-response, sampling weights were created to obtain a representative sample of the intended target population. The Royal Free Hospital and Medical School Research Ethics Committee provided ethical approval for the study with all participants providing written informed consent.

Physical conditions and physical multimorbidity

Respondents were asked about 20 physical health conditions (cancer, diabetes, epilepsy, migraine, cataracts/eyesight problems, ear/hearing problems, stroke, heart attack/angina, high blood pressure, bronchitis/emphysema, asthma, allergies, stomach ulcer, or other digestive problems, liver problems, bowel/colon problems, bladder problems/incontinence, arthritis, bone/back/joint/muscle problems, infectious disease, and skin problems). To be counted, conditions had to have been diagnosed by a doctor or other health professional and have been present in the previous 12 months. The number of physical conditions was summed and categorized as 0, 1, 2, 3, 4, and ⩾5. Multimorbidity was defined as two or more physical conditions (NICE, 2021).

Suicidal ideation and suicide attempts

Suicidal ideation was assessed by the question ‘Have you ever thought of taking your life, even if you would not really do it?’ and suicide attempt by the question ‘Have you ever made an attempt to take your life, by taking an overdose of tablets or in some other way?’ For those who answered ‘yes’, a follow-up question was asked about whether this had occurred in the previous 12 months. Answering ‘yes’ to the first questions and the follow-up question was considered past 12-month suicidal ideation and suicide attempts, respectively.

Mediators

The mediators (i.e. depression, anxiety, insomnia, pain, cognitive problems, perceived stress, loneliness, disability, and social support) were selected based on previous literature (Hajek et al., 2020; HHS.gov, 2021; Pu et al., 2017; Sindi et al., 2020; St John et al., 2014; Stravynski & Boyer, 2001; Wei et al., 2020). The interviewers administered the Clinical Interview Schedule Revised (CIS-R). This can be administered by lay interviewers and was used to generate ICD-10 diagnoses of depressive episode and anxiety disorder (generalized anxiety disorder, panic disorder, phobia, and obsessive-compulsive disorder) in the prior week (Lewis et al., 1992). The reliability and validity of the CIS-R have been reported in previous publications (Jordanova et al., 2004; Lewis et al., 1992). Insomnia was defined as fulfilling all the following three criteria: (i) problems trying to get to sleep or getting back to sleep (if had woken up) in the past month; and in the past 7 days, (ii) had problems with sleep on at least four nights in addition to (iii) taking at least 1 hour trying to get to sleep on the night with least sleep (Freeman et al., 2010). The question on pain was ‘During the past 4 weeks, how much did pain interfere with your normal work (including both work outside the home and housework)?’. The answer options were ‘not at all’, ‘a little bit’, ‘moderately’, ‘quite a bit’, and ‘extremely’. Cognitive function was assessed by two questions with ‘yes’ and ‘no’ answer options: ‘In the past month, have you had any problems in concentrating on what you were doing?’ (concentration); and ‘Have you noticed any problems with forgetting things in the past month?’ (memory). Two separate dichotomous variables were created for concentration and memory based on these answer options. Cognitive problems were defined as answering ‘yes’ to at least one of these two questions. For perceived stress, participants were asked whether their tasks at work and at home were very stressful over the past 2 weeks. Answer options were ‘most of the time’, ‘usually’, ‘occasionally’, and ‘not at all’. Disability was defined as having one or more difficulties with seven types of activities of daily living (ADL): personal care, medical care, preparing meals, mobility, shopping, housework, practical tasks such as decorating, dealing with paperwork and managing money (Brewin & Wing, 1989). Respondents were asked to assess to what extent they had felt ‘lonely and isolated from other people’ in the previous 2 weeks. Answer options were ‘very much’, ‘sometimes’, ‘not often’, and ‘not at all’. The level of social support was assessed by seven questions on whether: family and friends did things to make them happy, made them feel loved, could be relied on no matter what, would see that they were taken care of no matter what, accepted them just the way they are, made them feel an important part of their lives, and gave them support and encouragement. The answer options to these questions consisted of a 3-point scale: ‘not true (coded = 0)’, ‘partly true (coded = 1)’, or ‘certainly true (coded = 2)’. Following a scoring method used in a previous publication (Wickham et al., 2014), the answers to these items were summed to create a scale score that ranged from 0 to 14 with higher scores indicating greater social support (Cronbach’s α = .88).

Control variables

The selection of control variables was based on past literature (Huh et al., 2019) and included age, sex, equivalized income tertiles (high ⩾£29,826, middle £14,057<£29,826, and low <£14,057), education [qualification (degree, non-degree, A-level, GCSE, and other): yes or no)], ethnicity (white British, or other), smoking (never, past, and current), and alcohol dependence. The Alcohol Use Disorders Identification Test (AUDIT) was used to assess alcohol consumption (Saunders et al., 1993). Those with an AUDIT score of 10 and above were assessed for alcohol dependence with the Severity of Alcohol Dependence Questionnaire (SADQ-C; Stockwell et al., 1994), with scores of 4 and above (out of 60) on this measure being used to determine past 6-month presence of alcohol dependence.

Statistical analysis

The statistical analysis was done with Stata 14.2 (Stata Corp LP, College station, Texas). The difference in sample characteristics was tested by Chi-square tests and Student’s t-tests for categorical and continuous variables, respectively. Multivariable logistic regression analysis was done to assess the association between number of physical conditions (exposure) and suicidal ideation or suicide attempts (outcomes). Test of trend was conducted by including the number of physical conditions in the model as a continuous variable rather than a categorical variable.

Next, in order to gain an understanding of the extent to which various factors may explain the relationship between physical multimorbidity (i.e. ⩾2 physical conditions) and suicidal ideation or suicide attempts, we conducted a mediation analysis. We used the khb (Karlson Holm Breen) command in Stata (Breen et al., 2013) for the mediation analysis. This method can be applied in logistic regression models and decomposes the total effect (i.e. unadjusted for the mediator) of a variable into direct and indirect effects. Using this method, the percentage of the main association explained by the mediator can also be calculated (mediated percentage). Each potential mediator was included in the model individually.

All regression analyses including the mediation analysis were adjusted for age, sex, income, education, ethnicity, smoking, and alcohol consumption. As approximately one-fifth of the participants did not provide information on income, a missing category was included in the models only for this variable to avoid the exclusion of a large number of participants from the analysis. The sample weighting and the complex study design were taken into account in all analyses. Results from the regression analyses are presented as odds ratios (ORs) with 95% confidence intervals (CIs). The level of statistical significance was set at p < .05.

Results

The final sample consisted of 7,403 individuals aged ⩾16 years [M (SD) age 46.3 (18.6) years; 48.6% males]. The prevalence of physical multimorbidity (i.e. ⩾2 physical conditions), suicidal ideation, and suicide attempts were 35.1%, 4.3%, and 0.7%, respectively. The most common individual physical conditions were bone, back, joint/muscle problems (21.9%), and high blood pressure (16.9%), while the most common pairs of physical conditions were bone, back, joint/muscle problems co-occurring with arthritis (5.6%) or high blood pressure (5.1%; Supplemental Appendix Figure S1). The sample characteristics are provided in Table 1. Physical multimorbidity and suicidal ideation were both associated with higher prevalence of female sex, lower levels of wealth, past or current smoking, depression, anxiety, insomnia, pain, cognitive problems, disability, and higher levels of perceived stress, and loneliness. The prevalence of suicidal ideation and suicide attempts were significantly higher in people with physical multimorbidity, asthma, bladder problems/incontinence, bowel/colon problems, bronchitis/emphysema, epilepsy, and migraine or frequent headaches (Table 2). People with arthritis, bone, back, joint or muscle problems, liver problems, skin problems, and stomach ulcers/other digestive problems had a significantly higher prevalence of suicidal ideation but not suicide attempts. The prevalence of suicidal ideation and suicide attempts increased linearly with an increasing number of physical conditions (Figure 1). For example, the prevalence of suicidal ideation was 3.0% among people without physical conditions but this increased to 9.2% among those with ⩾4 physical conditions. After adjustment for potential confounders, compared to no physical conditions, 1, 2, 3, and ⩾4 conditions were associated with significant 1.79 (95% CI [1.25, 2.57]), 2.39 (95% CI [1.63, 3.51]), 2.88 (95% CI [1.83, 4.55]), and 6.29 (95% CI [4.12, 9.61]) times higher odds for suicidal ideation (Table 3). A 3 (OR = 5.64; 95% CI [1.50, 21.19]) and ⩾4 (OR = 13.25; 95% CI [4.58, 38.37]) physical conditions were associated with significantly increased odds for suicide attempts. The test for trend showed that there is a significant dose-dependent association for both suicidal ideation and suicide attempts. Mediation analysis showed that cognitive problems (mediated percentage 39.2%) mediated the largest proportion of the association between physical multimorbidity and suicidal ideation, followed by disability (37.5%), pain (35.6%), loneliness (30.5%), anxiety (25.5%), insomnia (18.6%), perceived stress (11.7%), and depression (11.5%; Table 4). For suicide attempts, the significant mediators were pain (38.0%), cognitive problems (30.7%), disability (26.3%), loneliness (22.4%), anxiety (17.2%), and depression (6.9%).

Table 1.

Sample characteristics (overall and by physical multimorbidity or suicidal ideation).

Characteristic	Overall	Physical multimorbidity^a			Suicidal ideation
Characteristic	Overall	No	Yes	p-Value^b	No	Yes	p-Value^b
Age (years)
M (SD)	46.3 (18.6)	41.6 (16.6)	55.2 (19.0)	<.001	46.6 (18.6)	40.3 (17.4)	<.001
Sex
Male	48.6	51.7	42.9	<.001	49.1	38.2	.002
Female	51.4	48.3	57.1		50.9	61.8
Qualification
No	23.9	19.7	31.8	<.001	23.8	27.1	.209
Yes	76.1	80.3	68.2		76.2	72.9
Ethnicity
British White	85.1	83.0	89.1	<.001	85.1	85.7	.808
Other	14.9	17.0	10.9		14.9	14.3
Income
High	35.8	39.2	29.8	<.001	36.5	22.3	<.001
Middle	32.6	31.9	33.8		32.7	30.4
Low	31.6	28.8	36.4		30.9	47.3
Smoking
Never	34.8	36.7	31.4	<.001	35.2	25.6	<.001
Past	42.1	38.8	48.1		42.5	32.6
Current	23.1	24.5	20.5		22.2	41.8
Alcohol dependence
No	94.1	93.8	94.7	.180	94.6	83.2	<.001
Yes	5.9	6.2	5.3		5.4	16.8
Depression
No	97.0	98.0	95.1	<.001	98.2	71.1	<.001
Yes	3.0	2.0	4.9		1.8	28.9
Anxiety
No	93.2	95.3	89.4	<.001	95.0	54.5	<.001
Yes	6.8	4.7	10.6		5.0	45.5
Insomnia
No	86.3	90.6	78.2	<.001	87.5	59.2	<.001
Yes	13.7	9.4	21.8		12.5	40.8
Pain
Not at all	58.7	70.8	36.3	<.001	59.7	36.4	<.001
A little bit	21.9	19.6	26.2		21.7	26.8
Moderately	7.8	4.5	13.9		7.6	13.0
Quite a bit	8.8	4.0	17.7		8.5	16.4
Extremely	2.8	1.1	5.8		2.6	7.4
Cognitive problems
No	63.9	70.9	51.0	<.001	66.0	19.9	<.001
Yes	36.1	29.1	49.0		34.0	80.1
Perceived stress
Most of the time	5.7	4.9	7.1	.002	5.0	21.0	<.001
Usually	7.0	7.0	7.0		6.5	17.2
Occasionally	47.9	49.0	45.8		48.2	43.1
Not at all	39.4	39.0	40.1		40.3	18.7
Loneliness
Very much	3.1	2.4	4.2	<.001	2.0	25.6	<.001
Sometimes	17.4	15.6	20.7		16.0	47.4
Not often	16.5	16.5	16.7		16.8	11.0
Not at all	63.0	65.5	58.4		65.2	16.1
Disability
No	67.1	78.6	46.1	<.001	68.7	35.7	<.001
Yes	32.9	21.4	53.9		31.3	64.3
Social support^c
M (SD)	13.2 (1.9)	13.2 (1.8)	13.1 (2.2)	.160	13.2 (1.8)	11.9 (3.2)	<.001

Note. Data are % unless otherwise stated. SD = standard deviation.

Physical multimorbidity referred to ⩾2 physical conditions.

p-Value was calculated by Chi-squared tests and Student’s t-tests for categorical and continuous variables, respectively

Based on a scale ranging from 0 to 14 with higher scores representing higher levels of social support.

Table 2.

Prevalence of suicidal ideation and suicide attempts by absence or presence of physical diseases and physical multimorbidity.

Condition	Suicidal ideation (%)			Suicide attempts (%)
	Physical condition			Physical condition
	Absent	Present	p-Value^a	Absent	Present	p-Value^a
Physical multimorbidity^b	3.42	5.97	<.001	0.46	1.05	.013
Allergies	4.23	5.24	.250	0.63	1.14	.133
Arthritis	4.14	5.58	.037	0.69	0.52	.515
Asthma	3.96	8.00	<.001	0.48	2.65	<.001
Bladder problems/incontinence	4.14	9.44	.001	0.63	1.91	.039
Bone, back, joint, or muscle problems	3.84	6.01	.001	0.63	0.79	.549
Bowel/colon problems	4.09	8.59	<.001	0.61	1.81	.003
Bronchitis/emphysema	4.19	9.84	<.001	0.62	2.83	.001
Cancer	4.30	5.72	.442	0.66	1.06	.517
Cataracts/eyesight problems	4.32	4.26	.923	0.73	0.30	.073
Diabetes	4.29	4.82	.593	0.67	0.71	.927
Ear/hearing problems	4.19	5.98	.068	0.66	0.73	.884
Epilepsy/fits	4.27	12.33	.009	0.64	6.22	<.001
Heart attack/angina^c	4.32	3.98	.807	NA	NA	NA
High blood pressure	4.33	4.24	.879	0.67	0.66	.948
Infectious disease	4.32	3.21	.676	0.66	1.37	.471
Liver problems	4.28	9.41	.034	0.67	1.00	.688
Migraine or frequent headaches	3.71	11.21	<.001	0.53	2.20	<.001
Skin problems	4.12	6.03	.021	0.65	0.79	.681
Stomach ulcer/other digestive problems	4.13	7.43	.001	0.65	0.97	.384
Stroke^c	4.30	8.26	.206	NA	NA	NA

p-Value was based on Chi-squared tests.

Physical multimorbidity referred to ⩾2 physical conditions.

Estimates for suicide attempts could not be obtained as there were no suicide attempts among people with these conditions.

Figure 1.

Prevalence of suicidal ideation and suicide attempts by number of physical conditions.

Table 3.

Association between number of physical conditions and suicidal ideation or suicide attempts (outcomes).

		Suicidal ideation			Suicide attempts
		OR	95% CI	p-Value	OR	95% CI	p-Value
No. of physical conditions	0	1.00			1.00
	1	1.79	[1.25, 2.57]	.002	2.54	[0.91, 7.11]	.076
	2	2.39	[1.63, 3.51]	<.001	4.77	[1.52, 14.95]	.008
	3	2.88	[1.83, 4.55]	<.001	5.64	[1.50, 21.19]	.011
	⩾4	6.29	[4.12, 9.61]	<.001	13.25	[4.58, 38.37]	<.001

Note. Models are adjusted for age, sex, income, education, ethnicity, smoking, and alcohol dependence. Significant test for trend for both suicidal ideation and suicide attempts (p < .05). OR = odds ratio; CI = confidence interval.

Table 4.

Mediators in the association between physical multimorbidity (i.e. ⩾2 physical conditions) and suicidal ideation or suicide attempts.

Mediator	Effect	Suicidal ideation			Suicide attempts
Mediator	Effect	OR [95% CI]	p-Value	% Mediated^a	OR [95% CI]	p-Value	% Mediated^a
Depression	Total	2.15 [1.58, 2.92]	<.001	11.5	3.30 [1.44, 7.56]	.005	6.9
	Direct	1.97 [1.44, 2.68]	<.001		3.04 [1.31, 7.04]	.009
	Indirect	1.09 [1.06, 1.12]	<.001		1.09 [1.05, 1.13]	<.001
Anxiety	Total	1.96 [1.47, 2.61]	<.001	25.5	2.79 [1.29, 6.03]	.009	17.2
	Direct	1.65 [1.23, 2.21]	.001		2.34 [1.07, 5.13]	.034
	Indirect	1.19 [1.14, 1.24]	<.001		1.19 [1.11, 1.28]	<.001
Insomnia	Total	2.27 [1.73, 2.99]	<.001	18.6	3.85 [1.83, 8.07]	<.001	NA
	Direct	1.95 [1.47, 2.58]	<.001		3.61 [1.68, 7.73]	.001
	Indirect	1.16 [1.12, 1.21]	<.001		1.07 [0.99, 1.15]	.091
Pain	Total	2.34 [1.77, 3.08]	<.001	35.6	3.92 [1.85, 8.29]	<.001	38.0
	Direct	1.73 [1.30, 2.30]	<.001		2.33 [1.05, 5.20]	.038
	Indirect	1.35 [1.23, 1.48]	<.001		1.68 [1.27, 2.22]	<.001
Cognitive problems	Total	2.46 [1.86, 3.27]	<.001	39.2	4.15 [1.98, 8.70]	<.001	30.7
	Direct	1.73 [1.32, 2.28]	<.001		2.68 [1.31, 5.51]	.007
	Indirect	1.42 [1.32, 1.54]	<.001		1.55 [1.28, 1.88]	<.001
Perceived stress	Total	2.41 [1.83,3 .18]	<.001	11.7	3.90 [1.91, 7.98]	<.001	NA
	Direct	2.17 [1.65, 2.86]	<.001		3.66 [1.81, 7.41]	<.001
	Indirect	1.11 [1.06, 1.15]	<.001		1.07 [0.99, 1.15]	.089
Loneliness	Total	2.27 [1.70, 3.04]	<.001	30.5	3.32 [1.57, 7.00]	.002	22.4
	Direct	1.77 [1.32, 2.37]	<.001		2.54 [1.24, 5.22]	.011
	Indirect	1.28 [1.20, 1.38]	<.001		1.31 [1.14, 1.50]	<.001
Disability	Total	2.48 [1.87, 3.28]	<.001	37.5	4.10 [1.97, 8.53]	<.001	26.3
	Direct	1.76 [1.33, 2.34]	<.001		2.83 [1.30, 6.18]	.009
	Indirect	1.41 [1.30, 1.52]	<.001		1.45 [1.18, 1.78]	<.001
Social support	Total	2.39 [1.82, 3.14]	<.001	NA	3.90 [1.89, 8.03]	<.001	NA
	Direct	2.36 [1.80, 3.10]	<.001		3.87 [1.88, 7.97]	<.001
	Indirect	1.01 [0.99, 1.03]	.272		1.01 [0.99, 1.02]	.359

Note. Models are adjusted for age, sex, income, education, ethnicity, smoking, and alcohol dependence. OR = odds ratio; CI = confidence interval.

Mediated percentage was only calculated in the presence of a significant indirect effect (p < .05).

Discussion

Findings from the present study support and add to previous literature. They support previous literature (Xiong et al., 2020) by further confirming that physical multimorbidity is associated with higher odds for suicidal ideation and suicide attempt among adults in England, and add to this literature by identifying potential key mediating variables.

There are several plausible pathways that likely explain the association between physical multimorbidity and suicidal ideation/suicide attempt. First, several physical conditions were individually associated with suicidal ideation and suicide attempt in our study. For example, epilepsy may increase risk for suicidal behavior as limbic epileptogenic areas in temporal lobe epilepsy, such as the amygdala, are involved in social behaviors, including impulse control, anxiety, and emotional memory. Moreover, altered serotonergic neurotransmission has been described in epilepsy, as well as impulsive and suicidal behavior (de Oliveira et al., 2011). In the case of migraine, literature has shown that the levels of cortisol and functioning of the hypothalamic-pituitary-adrenocortical (HPA) axis are affected by stressful events such as headaches. Specifically, HPA activity has been found to be correlated with low grade cognitive stress in those who experience migraines, and this may subsequently lead to suicidal behaviors (Berhane et al., 2018). Other individual physical conditions identified as being associated with suicidal ideation or suicide attempts may also increase suicidal thoughts and behaviors through suffering caused by the symptoms, hopelessness, and perceived burden, etc.

Interestingly, in our study, cognitive impairment explained the largest proportion of the association between physical multimorbidity and suicidal ideation (% mediated 39.2%) or suicide attempts (30.7%). Multiple individual physical conditions identified in this study as being associated with suicidal thoughts and/or behaviors have also been found to be associated with cognitive impairment [e.g. migraines (de Araújo et al., 2012), asthma (Caldera-Alvarado et al., 2013), arthritis (Wallin et al., 2012), epilepsy (Miller et al., 2016), and emphysema (Torres-Sánchez et al., 2015)]. For example, emphysema may lead to cognitive decline via low peripheral oxygen saturation (⩽ 88%) which has been strongly associated with a risk of cognitive impairment in patients with chronic obstructive pulmonary disease, and the use of home oxygen therapy has been associated with a reduction in that risk (Torres-Sánchez et al., 2015). Migraines may lead to cognitive impairment via central nervous system dysfunction underlying migraine pathophysiology (de Araújo et al., 2012). Furthermore, people with physical diseases are more likely to have sleep problems (Koyanagi et al., 2014), and sleep problems, in turn, may give rise to cognitive impairment (Vanek et al., 2020).

In addition to this, some pairs of physical conditions may mutually influence each other to accelerate cognitive decline (e.g. heart disease and cerebrovascular disease; Vassilaki et al., 2015), while drug interactions in people with multimorbidity may also increase the risk for cognitive decline (Koyanagi et al., 2018). In turn, cognitive impairment may increase suicidality owing to neurocognitive deficits that may lead to an incorrect appraisal of one’s life situation and consequently poor decision-making (Pu et al., 2017).

Disability (% mediated 26.3%–37.5%) and pain (35.6%–38.0%) were also identified as key mediating variables. Physical multimorbidity can give rise to disability and pain owing to specific symptoms of individual physical conditions (e.g. stroke and arthritis), and these may increase the risk for suicidal thoughts/behavior not only via the specific suffering due to the symptoms but other factors such as social exclusion (Milner et al., 2019) and hopelessness (Hooley et al., 2014). Moreover, the desire to escape physical pain by death is often seen as the only solution, and it has been suggested that chronic exposure to pain and pain tolerance can reduce fear about death and increase the capability to carry out suicide (Jacob et al., 2018).

The large sample of UK adults and the identification of potential mediating variables in the physical multimorbidity/suicidal thought and behavior relationship are clear strengths of the present study. However, findings must be interpreted in light of the study limitations. First, the information used in this study was based on self-report, and therefore, social desirability and recall bias may exist. Second, as cross-sectional data were used, the temporal ordering of physical multimorbidity and suicidality cannot be unequivocally determined. Third, data on the duration of physical illness, disease severity, and other specific information (e.g. cancer type) were not available, despite the fact that these could be important determinants of risk for suicidality. Future studies on this topic should consider incorporating this information. Finally, the variable on cognition was a crude measure consisting of only two answer options. Thus, it is possible for the mediated percentage to have differed if cognition was based on a different measure (e.g. Likert scale).

In conclusion, in this large sample of UK adults, physical multimorbidity was associated with a significantly higher odds for suicidal ideation and suicide attempts. Moreover, several potential mediators were identified, and these may serve as future targets for interventions that aim to prevent suicidality among people with physical multimorbidity.

Supplemental Material

sj-docx-1-isp-10.1177_00207640221137993 – Supplemental material for The association of physical multimorbidity with suicidal ideation and suicide attempts in England: A mediation analysis of influential factors

Supplemental material, sj-docx-1-isp-10.1177_00207640221137993 for The association of physical multimorbidity with suicidal ideation and suicide attempts in England: A mediation analysis of influential factors by Lee Smith, Jae Il Shin, San Lee, Jae Won Oh, Guillermo F López Sánchez, Karel Kostev, Louis Jacob, Mark A. Tully, Felipe Schuch, Daragh T. McDermott, Damiano Pizzol, Nicola Veronese, Junmin Song, Pinar Soysal and Ai Koyanagi in International Journal of Social Psychiatry

Footnotes

Acknowledgements

We would like to thank the National Center for Social Research and the University of Leicester who were the Principal Investigators of this survey. In addition, we would also like to thank the UK Data Archive and the National Center for Social Research as the data collectors and for making these data publically available. They bear no responsibility for this analysis or interpretation of this publically available dataset.

Conflict of interest

All authors confirm to have no actual or potential conflict of interests.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work is supported by the European Union – Next Generation EU to Dr. Guillermo F. López Sánchez.

Submission declaration and verification

The article has not been published previously.

ORCID iDs

Lee Smith

Guillermo F López Sánchez

Karel Kostev

Mark Tully

Data availability statement

The data used in this study can be obtained from the UK data archive ().

Supplemental material

Supplemental material for this article is available online.

References

Berhane

H. Y.

Jamerson-Dowlen

Friedman

L. E.

Berhane

Williams

M. A.

Gelaye

(2018). Association between migraine and suicidal behavior among Ethiopian adults. BMC Psychiatry, 18, 46.

Breen

Karlson

K. B.

Holm

(2013). Total, direct, and indirect effects in logit and probit models. Sociological Methods & Research, 42, 164–191.

Brewin

Wing

(1989). Manual of the MRC needs for care assessment.

Caldera-Alvarado

Khan

D. A.

Defina

L. F.

Pieper

Brown

E. S.

(2013). Relationship between asthma and cognition: The Cooper Center Longitudinal Study. Allergy, 68, 545–548.

de Araújo

C. M.

Barbosa

I. G.

Lemos

S. M. A.

Domingues

R. B.

Teixeira

A. L

. (2012). Cognitive impairment in migraine: A systematic review. Dement Neuropsychol, 6, 74–79.

de Oliveira

G. N.

Kummer

Salgado

J. V.

Filho

G. M.

David

A. S.

Teixeira

A. L

. (2011). Suicidality in temporal lobe epilepsy: Measuring the weight of impulsivity and depression. Epilepsy Behavior, 22, 745–749.

Freeman

Brugha

Meltzer

Jenkins

Stahl

Bebbington

(2010). Persecutory ideation and insomnia: Findings from the second British National Survey of Psychiatric Morbidity. Journal of Psychiatric Research, 44, 1021–1026.

Hajek

Kretzler

König

H. H.

(2020). Multimorbidity, loneliness, and social isolation. A systematic review. International Journal of Environmental Research and Public Health, 17, 1–12.

HHS.gov. (2021). Does depression increase the risk for suicide? 2014. U.S. Department of Health & Human Services.

10.

Hooley

J. M.

Franklin

J. C.

Nock

M. K.

(2014). Chronic pain and suicide: Understanding the association. Current Pain and Headache Reports, 18, 435.

11.

Huh

Nam

G. E.

Kim

Y. H.

Lee

J. H.

(2019). Relationships between multimorbidity and suicidal thoughts and plans among Korean adults. Journal of Clinical Medicine, 8, 1094.

12.

Jacob

Haro

J. M.

Koyanagi

(2018). The association between pain and suicidal behavior in an English national sample: The role of psychopathology.Journal of Psychiatric Research, 98, 39–46.

13.

Jordanova

Wickramesinghe

Gerada

Prince

(2004). Validation of two survey diagnostic interviews among primary care attendees: A comparison of CIS-R and CIDI with SCAN ICD-10 diagnostic categories. Psychological Medicine, 34, 1013–1024.

14.

Koyanagi

Garin

Olaya

Ayuso-Mateos

J. L.

Chatterji

Leonardi

Koskinen

Tobiasz-Adamczyk

Haro

J. M.

(2014). Chronic conditions and sleep problems among adults aged 50 years or over in nine countries: A multi-country study. PLOS ONE, 9, e114742.

15.

Koyanagi

Lara

Stubbs

Carvalho

A. F.

Stickley

Veronese

Vancampfort

(2018). Chronic physical conditions, multimorbidity, and mild cognitive impairment in low- and middle-income countries. Journal of the American Geriatrics Society, 66, 721–727.

16.

Lewis

Pelosi

A. J.

Araya

Dunn

(1992). Measuring psychiatric disorder in the community: A standardized assessment for use by lay interviewers. Psychological Medicine, 22, 465–486.

17.

Miller

L. A.

Galioto

Tremont

Davis

Bryant

Roth

LaFrance

W. C.

Jr. Blum

A. S.

(2016). Cognitive impairment in older adults with epilepsy: Characterization and risk factor analysis. Epilepsy & Behavior, 56, 113–117.

18.

Milner

Bollier

A. M.

Emerson

Kavanagh

(2019). The relationship between disability and suicide: Prospective evidence from the Ten to Men cohort. Journal of Public Health, 41, 707–713.

19.

NHS. (2021). Adult psychiatric morbidity survey. Author.

20.

NICE. (2021). The National Institute for Health and Care Excellence. Multimorbidity. Author.

21.

Setoyama

Noda

(2017). Association between cognitive deficits and suicidal ideation in patients with major depressive disorder. Scientific Reports, 7, 11637.

22.

Saunders

J. B.

Aasland

O. G.

Babor

T. F.

de la Fuente

J. R.

Grant

(1993). Development of the alcohol use disorders identification test (AUDIT): WHO collaborative project on early detection of persons with harmful alcohol consumption—II. Addiction, 88, 791–804.

23.

Sindi

Pérez

L. M.

Vetrano

D. L.

Triolo

Kåreholt

Sjöberg

Darin-Mattsson

Kivipelto

Inzitari

Calderón-Larrañaga

(2020). Sleep disturbances and the speed of multimorbidity development in old age: Results from a longitudinal population-based study. BMC Medicine, 18, 382.

24.

Stafford

Steventon

Thorlby

Fisher

Turton

Deeny

(2017). Briefing: Understanding the health care needs of people with multiple health conditions. Health Foundation.

25.

St John

P. D.

Tyas

S. L.

Menec

Tate

. (2014). Multimorbidity, disability, and mortality in community-dwelling older adults. Canadian Family Physician, 60, e272–e280.

26.

Stockwell

Sitharthan

McGrath

Lang

(1994). The measurement of alcohol dependence and impaired control in community samples. Addiction, 89, 167–174.

27.

Stravynski

Boyer

(2001). Loneliness in relation to suicide ideation and parasuicide: A population-wide study. Suicide and Life-Threatening Behavior, 31, 32–40.

28.

Torres-Sánchez

Rodríguez-Alzueta

Cabrera-Martos

López-Torres

Moreno-Ramírez

M. P.

Valenza

M. C.

(2015). Cognitive impairment in COPD: A systematic review. Jornal Brasileiro de Pneumologia, 41, 182–190.

29.

Vanek

Prasko

Genzor

Ociskova

Kantor

Holubova

Slepecky

Nesnidal

Kolek

Sova

(2020). Obstructive sleep apnea, depression and cognitive impairment. Sleep Medicine, 72, 50–58.

30.

Vassilaki

Aakre

J. A.

Cha

R. H.

Kremers

W. K.

Sauver

J. L.

Mielke

M. M.

Geda

Y. E.

Machulda

M. M.

Knopman

D. S.

Petersen

R. C.

Roberts

R. O.

(2015). Multimorbidity and risk of mild cognitive impairment. Journal of the American Geriatrics Society, 63, 1783–1790.

31.

Wallin

Solomon

Kåreholt

Tuomilehto

Soininen

Kivipelto

(2012). Midlife rheumatoid arthritis increases the risk of cognitive impairment two decades later: A population-based study. Journal of Alzheimer’s Disease, 31, 669–676.

32.

Wei

M. Y.

Levine

D. A.

Zahodne

L. B.

Kabeto

M. U.

Langa

K. M.

(2020). Multimorbidity and cognitive decline over 14 years in older Americans. The Journals of Gerontology: Series A, 75, 1206–1213.

33.

Wickham

Taylor

Shevlin

Bentall

R. P.

(2014). The impact of social deprivation on paranoia, hallucinations, mania and depression: The role of discrimination social support, stress and trust. PLOS ONE, 9, e105140.

34.

Xiong

Wang

Shen

Guo

Guan

(2020). The relationship between multimorbidity and suicidal ideation: A meta-analysis. Journal of Psychosomatic Research, 138, 110257.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.02 MB