Indirect effects of health-related quality of life on suicidal ideation through psychological distress among cancer patients

The interrelationships of suicidal ideation, psychological distress, and impaired health-related quality of life (HRQoL) in cancer patients are complex and multifaceted. Limited empirical evidence exists on the indirect effects of impaired HRQoL on suicidal ideation through psychological distress among cancer patients. To fill this research gap, 250 cancer patients were recruited through a cross-sectional hospital-based research design. Structural equation model (SEM) results indicated that impaired HRQoL is a predictor of psychological distress (β = 0.153; p < 0.05), and psychological distress positively predicts suicidal ideation (β = 0.647; p < 0.000). The study found no direct effects of impaired HRQoL on suicidal ideation (β = −0.05; p = 0.223). Indirect effects of HRQoL on suicidal ideation was confirmed, showing a full-mediation effect β = 0.099 (SE = 0.048, CI = [0.030, 0.189], p < 0.05) (i.e. the pathway impaired HRQoL predict suicidal ideation is through psychological distress). Cognitive-behavioral therapy and other emotional support programs should be considered for cancer patients to mitigate psychological vulnerabilities linking impaired HRQoL to suicidal ideation.


Introduction
More than 800,000 suicides occur yearly, mak ing it the second leading cause of death among those aged 15-29 years old (WHO, 2019).In recent years, extensive research has been con ducted on suicide among cancer patients as a serious public health concern (Anguiano et al., 2012;Du et al., 2020;Musa et al., 2021Musa et al., , 2022)).As a debilitating health condition, cancer dis rupts physical, mental, and social functioning, leaving survivors with unmet needs (Fang et al., 2018;Silvaggi et al., 2023;Yang et al., 2023).

Indirect effects of health-related quality of life on suicidal ideation through psychological distress among cancer patients
Adverse effects of cancer are typically accom panied by multimorbidity, such as psychological distress and suicidal ideation.Suicidal ideation is the process of mulling over killing oneself (Klonsky et al., 2016).It may range from fleet ing thoughts to detailed planning, depending on the severity of the mental illness (Bryan et al., 2019).The odds of suicidal ideation become higher with increasing psychological pro blems among cancer patients (Peltzer and Pengpid, 2022).Currently, suicidal ideation poses significant policy and intervention chal lenges (Desalegn et al., 2020;WHO, 2019).
The assessment of suicidal ideation is not a simple process since no single factor determines the likelihood of a completed suicide.Several risk factors for suicide have been identified among the general population, including a his tory of selfharm or attempts, mental health diagnosis, psychological distress, and impaired healthrelated quality of life (HRQoL) (CDC, 2018;LopezCastroman et al., 2020;Twenge et al., 2018).Notably, suicidal ideation is a sig nificant precursor to suicide among cancer patients, with prevalence ranging from 1.1% to 19.8% (Casey et al., 2008).Despite reports that cancer patients are at risk of suicide (Anguiano et al., 2012;Du et al., 2020), not all cancer patients report suicidal ideation.Demoralization, financial burden, living conditions, and cancer types are found to contribute to suicidal ideation among cancer patients (Lai et al., 2022).Due to a lack of data on these psychosocial indicators, it is increasingly imperative to examine risk fac tors for suicidal ideation in cancer patients, especially those in resourcelimited populations (Amiri and Behnezhad, 2020;Tang et al., 2022).Having such data is essential to educating clini cians and researchers regarding the risk factors and indirect mechanisms associated with sui cidal ideation in cancer patients.
HRQoL is a multidimensional construct encompassing an individual's physical, psycho logical, and social condition (Megari, 2013).Several studies have assessed the HRQoL of patients with various typologies of cancer (e.g., Ho et al., 2021;Lin et al., 2017;Wu et al., 2023).Suicidal ideation is known to be exacerbated by impaired HRQoL in cancer patients (van Oers and Schlebusch, 2020).Moreover, psychological distress can also lead to suicidal ideation (Lai et al., 2022).Psychological distress is character ized by depression, anxiety, and stress (Keles et al., 2020), existing as psychopathological comorbidities among cancer patients (Koné and Scharf, 2021;Naser et al., 2021).Psychological distress is more likely to occur in cancer patients with impaired HRQoL (Chan et al., 2014;Halvorsen et al., 2018).Suicidal ideation in can cer patients is further exacerbated by the interre lationships between HRQoL and psychological distress (Phoosuwan and Lundberg, 2022).Until now, the indirect effects of impaired HRQoL on suicidal ideation through psychological distress are generally less discussed among cancer patients, especially in resourcelimited popula tions.Cancer patients from lowincome coun tries are faced with socioeconomic disadvantages, poor healthcare infrastructure, and limited psy chological interventions that could buffer sui cidal ideation risks or even promote HRQoL (Adejoh et al., 2013).This study extends empiri cal evidence of HRQoL's indirect effect on sui cidal ideation through psychological distress using data from cancer patients.

Literature and theoretical review
Studies have examined the relationship between HRQoL and suicidal ideation in cancer patients (Anguiano et al., 2012;Du et al., 2020).However, empirical investigations have not comprehensively examined the theoretical basis for this relationship.The psychological theory of learned helplessness (Maier and Seligman, 2016), provides a premise for understanding the process of these associations.Consistent focus on uncontrollable adverse circumstances (i.e.impaired HRQoL) can result in loss of control, leading to unmanageable and errant thought patterns (Lennerlöf, 1991).Cancer patients may face the reality of their deteriorating health with trepidation until they lose control of their emo tions, leading to unstable thought patterns such as suicidal ideation.Psychological reactance models also reflect these psychosocial conflicts (Rosenberg and Siegel, 2018).Psychological reactance may occur when cancer patients experience unpleasant motivation as a result of their disease condition, which arises from the loss of autonomy over their health.If their health does not improve, they may become frustrated and develop adverse psychological reactions (De Las Cuevas et al., 2014).Cancer patients' deteriorating HRQoL may increase their risk of suicidal ideation as their health autonomy declines.There is evidence that poor HRQoL is linked to suicidal ideation (van Oers and Schlebusch, 2020).
The link between HRQoL and psychological distress has also been verified in several empiri cal studies (Chan et al., 2014;Halvorsen et al., 2018).Based on the transactional model of stress and coping, individuals utilize cognitive and behavioral responses to deal with internal or external situations that exceed their ability to cope and adjust (Folkman and Lazarus, 1980).Psychological distress and HRQoL are deter mined by a transactional process, which includes stressors associated with cancer diagnosis, side effects of treatment, and frailty.The effects of these stressors can result in a subdued emotional adjustment and a further suppression of cogni tive functioning (Baik et al., 2020).Cognitive behavioral theory clarifies the transactional processes between impaired HRQoL and psy chological distress in cancer patients (Lai et al., 2019).Thoughts and thinking patterns can con tribute to psychological distress (González Prendes and Resko, 2012).Negative thinking patterns associated with impaired HRQoL may lead to adverse expectations that elevate distress levels.Cancer patients' psychological distress is significantly influenced by their compromised health status (Phoosuwan and Lundberg, 2022).
Meanwhile, through the process of emo tional regulation (Saarni et al., 1998), cancer patients may experience suicidal ideation aris ing from the difficulty in controlling their exposure to psychological vulnerabilities.In a normal, nonmorbid individual, emotions may be controlled based on limited adverse exter nal or internal influences that could promote deficits in mental health and behavior (Thompson, 1991).The presence of unstable and problematic emotions, such as depression, anxiety, or stress, could indicate a dysfunc tional emotional state in cancer patients.Emotional dysfunction, such as excessive bur den and worry, can result in emotional distress and suicidal ideation (Thullen et al., 2016).Suicidal ideation is further aggravated by emo tional distortions, such as losing meaning and purpose in life (Hirsch et al., 2017).
These assertions discussed above may explain the direct association among HRQoL, psychological distress, and suicidal ideation.However, the understanding of the indirect mechanism warrants further theoretical explo ration.The interrelationships of impaired HRQoL, psychological distress, and suicidal ideation are based on psychopathology sche mas arising from adversity, such as cancer (Ingram and Luxton, 2005).According to stressdiathesis models of suicidal behavior, adversity components, such as impaired HRQoL could motivate selfharm behavior (van Heeringen, 2012).The inability to cope with or control stresses or adversity contributes to psychological distress (Mann and Haghighi, 2010).When HRQoL is compro mised in cancer patients, distress may be aggravated, resulting in suicidal ideation.Postulations from the dualprocess model of illness suggest that individual susceptibili ties (e.g.psychological distress) and stress ors (e.g.impaired HRQoL) could contribute to suicidal ideation (Berrenberg, 1991).Two psychosocial factors are combined and inter act to result in an adverse outcome in the dualprocess model of illness (Pyszczynski et al., 1999).For instance, deteriorating HRQoL and psychological distress could work in tandem through two distinct pro cesses.Among the factors that influence sui cidal ideation, psychological distress may be the most immediate and proximal (Ensel et al., 1996), whereas impaired HRQoL is considered a more distal factor (Murdaugh, 1997).Thus, impaired HRQoL may indirectly contribute to suicidal ideation by increasing psychological distress among cancer patients.Health-related quality of life (HRQoL).HRQoL was measured using the Functional Assessment of Cancer TherapyGeneral (FACTG) devel oped by Cella et al. (1993).The FACTG is a selfreport measure with a 5point Likert response format with a rating scale (0 = Not at all; 1 = A little bit; 2 = Somewhat; 3 = Quite a bit; and 4 = Very much) (Cella et al., 1993).The scale is divided into four subscales: Physical wellbeing (PWB; seven items, score range from 0 to 28), social/family wellbeing (SFWB; 7items, score range from 0 to 28), emotional wellbeing (EWB; 6items, score range from 0 to 24), and functional wellbeing (FWB; 7items, score range from 0 to 28).Two subscales, that is, SFWB and FWB, were reversecoded.Cella et al. (1993) indicated that the Cronbach's alpha of the total scale is 0.89.

Psychological distress. The Depression Anxiety
Stress Scales (DASS21) was designed to meas ure the core symptoms of depression, anxiety, and stress and has demonstrated excellent psy chometric properties across studies mostly con ducted in Western societies (Lovibond and Lovibond, 1995).DASS21 is a 21item stand ardized instrument developed as a shorter ver sion of DASS42.It is an instrument that assesses depression, anxiety, and stress levels of individuals.Each item was rated on a 4point Likert scale from 0, meaning "did not apply to me at all," to 3 "applied to me very much or most of the time."The convergent validity of DASS21 has been found to have a positive relationship with the Beck Depression Inven tory between the three subscales (Asghari et al., 2018).Among agespecific populations, the DASS21 acceptable factor loading ranges from 0.76 to 0.92 across different times and subscales (Cao et al., 2023a(Cao et al., , 2023b;;Chen et al., 2023).For the total score of DASS21, the Cronbach alpha = 0.94.The Cronbach alpha for Depression, Anxiety, and Stress scales were 0.85, 0.85, and 0.87, respectively.Prevalence of psychological distress among the participant as reported in Supplemental Figure 1  Analysis.Descriptive analysis of the study was done through bivariate analysis (corre lations), mean and standard deviations were computed through the Statistical Package for Social Sciences (SPSS Version 25.0).In addition, AMOS Version 20 was used to ana lyze the latent and observed variables through maximum likelihood analysis based on the structural equation model (SEM).As part of the test of the SEM model in the dataset, we examined the model's goodness of fit through confirmatory factor analysis (Bentler, 1990).A bootstrapping method was used to investigate the indirect effect of psychological distress in the theoretical model (n = 5000 bootstrap ping randomly selected samples) (Preacher and Hayes, 2008).To verify the SEM model fit, we used Comparative Fit Index (CFI), Incremental Fit Index (IFI), and Normed Fit Index (NFI) benchmarked at 0.95, and the root means square error of approxima tion (RMSEA) acceptable when below 0.06 (Bentler et al., 1987;Hu and Bentler, 1999).Impaired HRQoL was measured by scoring the four subscales as an observed variable.Two subscales (SFWB and FWB) of the HRQoL were recoded to allow the observed variable to indicate impaired HRQoL across the sub scales.Psychological distress was measured as a latent variable with factor loading anxi ety = 0.52, stress = 0.42, and depression = 0.88.The factor loading of the four items retained of suicidal ideation ranges from 0.53 to 1.00.In addition, to resolve the multicollinearity issues concerning measurements of HRQoL and psy chological distress, a varianceinflated factor (VIF) analysis was conducted (Jung and Park, 2018).The VIF assumes accepting that the pre dictors have no multicollinearity if VIF values are less than 5 as detailed in Table 1 (Mar coulides and Raykov, 2019).The preanalysis results indicate excellent model fit for the data (χ 2 (18, N = 250) = 25.985,p > 0.100; CMIN/ DF = 1.444, and with NFI = 0.979, TLI = 0.990, CFI = 0.993, and RMSEA = 0.040).

Result
Data from 250 cancer patients were analyzed.The data distribution includes 73.6% females and 26.4% males.Other information on the socioeconomic and demographic characteris tics of the study population is shown in Supplemental Table 1.
The mean, standard deviation, and correla tion for the study variables are shown in Table 1.The VIF result indicates that the variables are not multicollinear.Correlations among all the measured variables indicate that suicidal idea tion is positively correlated with psychological distress indicators (i.e.stress (r = 0.244, p < 0.01), depression (r = 0.635, p < 0.01), and anxiety (r = 0.421, p < 0.01)).In addition, sui cidal ideation is positively correlated with EWB (r = 0.206, p < 0.01) and SFWB (r = 0.231, p < 0.01).However, suicidal ideation is nega tively associated with FWB (r = −0.249,p < 0.01).The outcome based on the bivariate exploration of socioeconomic and demographic variables indicated that age is negatively corre lated with suicidal ideation (r = −0.134,p < 0.05), while cancer types were positively associated with suicidal ideation (r = 0.529, p < 0.01).
Figure 1 shows the path outcome in the model such that impaired HRQoL was positively asso ciated with psychological distress (β = 0.153; p < 0.05), indicating that when HRQoL worsens by 1 standard deviation, psychological distress goes up by 0.153 standard deviation.Psychological distress is positively associated with suicidal ideation (β = 0.647; p < 0.001), indicating that with 1 unit standard deviation increase in psychological distress, suicideal idea tion goes up by 0.647 standard deviation.
Overall, analysis shows that the model explained 21.2% of the variance in psychological distress.
In addition, a 66% variance in suicidal ideation in the cancer patients was explained by the model.
To verify the indirect relationships in the theoretical model, we generated 5000 samples (through a bootstrapping sampling strategy) from the primary dataset by random sampling.The result in Table 2 shows that the indirect effects of impaired HRQoL on suicidal idea tion through psychological distress was β = 0.099 (SE = 0.048, CI = [0.030,0.189], p < 0.05).This result indicated that the stand ardized indirect effect of impaired HRQoL on suicidal ideation is 0.099 (i.e. when HRQoL worsens by 1 standard deviation, suicidal idea tion goes up by 0.099 standard deviations).These results consolidate the understanding that the indirect pathway through which impaired HRQoL could result in suicidal idea tion among cancer patient is through psycho logical distress.

Discussion
This study examined impaired HRQoL's direct effects on psychological distress and suicidal ideation.The study extends the theoretical premises of stressdiathesis and the dualpro cess model of illness by investigating the indirect effects of impaired HRQoL on suicidal ideation through psychological distress among cancer patients.
The finding shows no direct and significant association between HRQoL and suicidal idea tion among the cancer patients.This finding is counterintuitive to studies and theories relating to psychological reactance and learned help lessness supporting the effects of impaired HRQoL on suicidal ideation (Lennerlöf, 1991;Rosenberg and Siegel, 2018;van Oers and Schlebusch, 2020).Several subjective and cul tural factors may account for these discrepan cies in the findings and theoretical understanding of the psychopathology.Based on the sociocul tural characteristics of this study population, patients may have access to alternative coping strategies, such as religious and spiritual beliefs, which may prevent suicidal ideation regardless of HRQoL impairment (Kelly et al., 2022;Mabena and Moodley, 2012).These factors could also lead to a reluctance to discuss sui cidal thoughts due to stigmatization or faith based assurance (Bruce et al., 2020;Okyere et al., 2022).Hence, the study cannot confirm that impaired HRQoL and suicidal ideation are associated.This finding confirms that impaired HRQoL is associated with suicidal ideation as a function of the socioenvironment and population.
However, the study found that impairment of HRQoL was a positive predictor of psycho logical distress in cancer patients.Earlier stud ies and theories have suggested that stressors (such as diagnosis and treatmentrelated side effects) may adversely affect the psychologi cal wellbeing of cancer patients (Baik et al., 2020).The study also identified other direct paths, suggesting that psychological distress is positively associated with suicidal ideation.According to previous theories and studies of emotional regulation (Saarni et al., 1998), can cer patients may experience instability and problematic emotions, such as fear, sadness, and anger, leading to thoughts of suicide.Symptoms of these emotional dysfunctions, such as anxiety, stress, and depression, can lead to suicidal thoughts.Consequently, impaired HRQoL could promote psychological distress, and psychological distress is a risk factor for suicidal ideation among cancer patients.
Furthermore, impaired HRQoL indirectly contributes to suicidal ideation through psycho logical distress.This study indicates that two theoretical propositions (the dual process model of illness and the stressdiathesis model) are viable theoretical perspectives that could be used to understand the indirect link between impaired HRQoL and suicidal ideation through psychological distress (Pyszczynski et al., 1999;van Heeringen, 2012).According to the indirect pathway, impaired HRQoL and psy chological distress may interact to exacerbate suicidal ideation in cancer patients (Ensel et al., 1996;Murdaugh, 1997), and together they form psychopathology schemas in which impaired HRQoL leads to psychological distress, and psychological distress to suicidal idea tion.Based on the study findings, cancer patients' suicidal ideation may not be directly caused by impaired HRQoL but rather indrectly through psychological distress, leading to a far reaching adverse outcome.
Age, cancer types, and educational attain ment of the cancer patients were also controlled for in this study's theoretical model, which may have contributed to the results.According to the model, 21% of psychological distress variance is predicted.However, more research is needed to examine psychosocial factors that might con tribute to the unexplained variance in psycho logical distress among cancer patients examined in this study within their unique sociocultural context.Suicidal ideation accounts for 66% of the variance explained by this model.Suicidal ideation may be cushioned by urgent HRQoL and psychological distress interventions for this unique population.
Despite its theoretical and empirical merits, this study has limitations.The relatively small sample size (250 respondents) and the study setting (limited to only University of Nigeria Teaching Hospital, Enugu, Nigeria) seem to limit the generalizability of this finding.Data collection should be expanded to other regions in Nigeria in future studies.A comparative study of the indirect mechanism proposed in this study is also encouraged.Additionally, this  study was crosssectional with selfreport meas ures, so causal inferences cannot be drawn.Future studies should adopt a longitudinal approach, which may help determine causality.
Research replicating the research objectives should control for treatment and lengths of treatments to extend this study.

Theoretical and practical implication
Certain populations, especially those from spe cific sociocultural and environmental back grounds, are less likely to report suicidal ideation when HRQoL is impaired.Suicidal ideation and impaired HRQoL are not associa tive due to numerous coping mechanisms available to cancer patients, including spiritu ality, religion, and social desirability (Bruce et al., 2020;Okyere et al., 2022).Providing emotional support for cancer patients should take precedence on their road to recovery.Psychological reactions and learned helpless ness may not hold in specific psychosocial con texts or among some populations, as critically ill individuals may seek coping mechanisms to arrest suicidal ideation in the face of impaired HRQoL.
In parallel with ongoing debates on the influ ences of psychological distress on the associa tion between HRQoL and suicidal ideation among cancer patients, the study advances empirical evidence.Cancer patients can benefit from psychological distress buffering resources like mindfulness, cognitivebehavioral thera pies, and counseling programs by arresting these negative interactions of HRQoL, psycho logical distress, and suicidal ideation.This approach is necessary because psychological distress is one pathway through which cancer patients with deteriorating HRQoL may experi ence suicidal ideation.Suspending exposure to emotional vulnerabilities becomes a viable intervention option for clinicians and public health policies to cushion the effects of impaired HRQoL on suicidal ideation.Clinicians, the government, and public health stakeholders must work together to provide appropriate policies, cognitive behavioral therapy, and other psychosocial interventions.

Conclusion
Based on the results of this study, impaired HRQoL was associated with psychological dis tress.Suicide ideation is also significantly influenced by psychological distress.There was no direct effect of impaired HRQoL on suicidal ideation among cancer patients.This research evidence suggests that HRQoL affects suicidal ideation indirectly through psycholog ical distress, as indicated by the full mediation effect of the study model.Psychological vul nerabilities of cancer patients should be miti gated through cognitivebehavioral therapies and other programs to cushion the adverse effects.

Informed consent
A signed written informed consent was received from the research participants.All participants were assured of the confidentiality protocol for using the information provided.In addition, questionnaire was anonymized without collecting identifiable data of the participants.
Control variables.In the model, age, cancer types, and level of education were controlled (see Supplemental Table1).

Table 1 .
Mean, SD, and correlation matrix of the study measures.

Table 2 .
Indirect effect of HRQL on suicidal ideation through psychological distress.