Perceived Burdensomeness in Older Adults and Perceptions of Burden on Spouses and Children

Abstract

Older adults are at high risk of suicide, and perceived burdensomeness is an important suicide risk factor in this population. Additionally, previous research indicates that older adults feel like a burden on their spouses most often, though perceptions of burden on younger generations were associated with greater overall perceived burdensomeness. The current study sought to clarify this complex pattern of results, and found that perceptions of burden on a spouse were linked to the most severe form of perceived burdensomeness—belief that others would be better off if one were gone—though perceptions of burden on a child were not significantly associated with this most severe form of perceived burdensomeness in the current sample. These results indicate that older adults’ perceptions of burdensomeness on a spouse may be especially harmful. Clinical implications, including the need to assess not only whether older adults perceive themselves to be a burden on others, but also who is perceived to be burdened, and the importance of targeting perceived burdensomeness through cognitive restructuring, are discussed.

Older adults have elevated rates of suicide in the United States (Centers for Disease Control and Prevention, 2012), and it is not currently known how to prevent suicide in later life. It is known, however, that suicide ideation is an important risk factor for death by suicide in older adults (Waern, Beskow, Runeson, & Skoog, 1999). Further, the interpersonal theory of suicide (Joiner, 2005; Van Orden et al., 2010) has identified two factors that contribute to suicide ideation, low belongingness and perceived burdensomeness.

Perceived burdensomeness is a belief that one is so incapable that he or she is a burden or liability to other people (Joiner, 2005). This belief leads some individuals to consider their deaths to be more valuable than their lives to people around them, and may result in shame, low self-esteem, and self-hatred (Van Orden et al., 2010). Perceived burdensomeness may be especially relevant to older adults, as they may require more assistance with activities of daily living, finances, and other areas of life as they age, leading to potential feelings of burdensomeness on other people. Research indicates that perceived burdensomeness is found more often in older adults than younger adults (Foster, 2003), and is linked to suicide risk (e.g., Cukrowicz, Cheavens, Van Orden, Ragain, & Cook, 2011; Jahn & Cukrowicz, 2011; Jahn, Cukrowicz, Linton, & Prabhu, 2011).

Older adults with elevated functional impairment, such as those with terminal illness, or those residing in assisted living facilities often experience perceptions of burden on others (e.g., Akechi et al., 2004; de Catanzaro, 1995). This may be the result of needing help with activities of daily living, requiring financial assistance, or feeling like an emotional burden on loved ones. Perceptions of burdensomeness are a source of concern among older adults with terminal cancer (Akechi et al., 2004); such concerns have been linked to the will to live in this population (Chochinov et al., 2005) and are associated with death by suicide (Filiberti et al., 2001). Additionally, among older adults living in senior housing, perceptions of burden on others were common and a major correlate of suicide ideation (de Catanzaro, 1995).

Recent research has begun to examine various issues that may influence perceived burdensomeness (Jahn & Cukrowicz, 2011). For example, research has examined the type of relationship (e.g., spouse, child, sibling, parent) perceived as burdened. Findings indicated that older adults report greater scores on a self-report measure of overall perceived burdensomeness (e.g., feeling like a liability to others, like one makes things worse for others) when they endorse feeling like a burden on a younger generation (e.g., child, grandchild), as opposed to when they endorse feeling like a burden on their own (e.g., spouse, sibling) or older (e.g., mother, father) generations. Yet, when older adults were asked which person in their lives was most strongly burdened by them, a spouse was identified by over 57%, whereas children were only identified by only 22%. This suggests that older adults believe that they are a burden on their children, but feel that they are a greater burden on their spouses. Thus, a complex pattern emerged: perceived burden on children was associated with greater overall perceived burdensomeness, yet spouses were perceived to be the most burdened (Jahn & Cukrowicz, 2011). This may be because older adult parents are not comfortable receiving certain types of help from their adult children, such as financial help (Hamon & Blieszner, 1990). Therefore, receiving this type of help may lead older adults to commonly feel like a burden on children, though they may not feel like they severely burden their children as a result of this help. However, same-generation familial relationships, such as a spouse, are considered the closest social relationships, and spousal relationships in particular are very intimate (Antonucci & Akiyama, 1995). Spouses, then, may spend the most time with an older adult and be relied on the most frequently for various forms of assistance, leading older adults to feel like their spouses are most severely burdened.

This complex pattern of findings may also be understood by conceptualizing perceived burdensomeness as a continuum in which some feelings of burdensomeness are relatively less severe (e.g., “I feel like a burden on others”) and others are more severe (e.g., “The people in my life would be better off if I were gone”). Recent work has examined specific items that measure perceived burdensomeness to identify items that are most closely associated with suicide ideation and represent the extreme end of the perceived burdensomeness continuum (Van Orden, Smith, O'Riley, & Conwell, 2011). One item (i.e., “These days the people in my life would be better off if I were gone”) has particularly good specificity and sensitivity in predicting suicide ideation, and appears to be a measure of severe perceived burdensomeness, whereas other items measure less acute or critical components of perceived burdensomeness that are less closely associated with suicide ideation. Other components of perceived burdensomeness, such as the belief that one makes things worse for others, may create distress, but are not strongly linked to elevated suicide risk. However, the “better off if I were gone” item represents the severe end of the perceived burdensomeness continuum that is closely linked to suicide ideation. This item may serve to help shed light on the complex pattern of findings presented previously. For example, because spouses were perceived to be the most severely burdened in previous research (Jahn & Cukrowicz, 2011), this severe burden may be associated with the most severe form of perceived burdensomeness (i.e., the belief that others would be better off if one were gone). However, perceptions of burden on children may be associated with less severe perceptions of burdensomeness as they are perceived to be less severely burdened. Given this rationale, we hypothesized that perceived burdensomeness on a spouse would be associated with endorsing the belief that others would be better off if the respondent was gone, whereas perceived burdensomeness on a child would not be associated with endorsing this belief.

Methods

Participants

Older adults (i.e., those over age 65) who participated in this study were recruited from family and internal medicine clinics at a large university medical center in the south central United States. A total of 70 participants agreed to participate, were not disqualified due to exclusionary criteria (i.e., cognitive impairment, current substance abuse, or current/past psychotic symptoms), and provided complete information on the necessary questionnaires. All participants included reported being a burden on at least one person in their lives. Participants’ mean age was 72.79 (SD = 6.56, Range = 65-93). A majority of the sample was female (56%), married (69%), and Caucasian (91%). More details about the sample are provided in previous publications (Jahn & Cukrowicz, 2011).

Measures

Perceived Burdensomeness Questionnaire (PBQ; Jahn & Cukrowicz, 2011)

The PBQ is a five-item questionnaire that measures perceptions of burden on different relationships, among other topics. The item used in the current study was, “The last time I thought I had a negative impact on the people in my life, I was concerned with affecting the following people: (Mark all that apply).” There are 41 different specific categories listed for this item (e.g., husband, granddaughter, sister, brother-in-law). The PBQ measures feelings of burden on specific relationships, making internal consistency reliability estimates inappropriate and, as feelings of burden may change over time, test-retest reliability is also not likely beneficial in examining the psychometric properties of this questionnaire. Previous research has identified significant relations between the PBQ and the only other perceived burdensomeness measure available to date, providing evidence of construct validity. In addition, the face validity of the measure is strong. More details about this questionnaire are available in previous publications (Jahn & Cukrowicz, 2011). For this study, relationships that were considered spousal relationships were husband and wife (cohabitating partners, ex-spouses, non-married romantic partners were not considered spouses); relationships that were considered children included son, daughter, son-in-law, daughter-in-law, stepson, and stepdaughter.

Interpersonal Needs Questionnaire (INQ; Van Orden, Cukrowicz, Witte, & Joiner, 2012)

The INQ has two subscales, one that assesses perceived burdensomeness and one that assesses low belongingness. Participants are asked to complete the items based on how they have been feeling recently in the instructions, and each item starts with, “These days,” to remind participants of this instruction. In the current study, one item that assesses perceived burdensomeness was used (i.e., “These days the people in my life would be better off if I were gone”). Participants rated the item on a Likert scale from one (Not at all true for me) to seven (Very true for me). These scores were then dichotomized, with participants indicating that the item was not at all true for them in one group (n = 58) and those indicating it was at least slightly true for them (i.e., scores of two or above) in the second group (n = 12). Because the goal of this study was to examine differences between those who reported at least some belief that others would be better off if they were gone and those who did not endorse this belief at all, dichotomization of this item was most appropriate for our analyses. These groups were used in all analyses for the current study.

Procedures

This study was part of a larger study funded by the American Foundation for Suicide Prevention (Principal Investigator: Kelly C. Cukrowicz, Ph.D.) and all procedures were approved by the university's Institutional Review Board. Detail about procedures are included in a previous publications (Cukrowicz et al., 2011; Jahn & Cukrowicz, 2011; Van Orden et al., 2012). In sum, participants were recruited by research assistants at appointments with primary care physicians and scheduled an appointment to participate in the study. All interviews and questionnaires, including those included in the current study, were completed in one session. Participants were screened for suicide risk during participation in the study, and any elevated risk was addressed as appropriate (e.g., referrals to outpatient mental health facilities, crisis response plans, hospitalization).

Data Analysis

To test the hypotheses, we conducted two logistic regressions. For both regressions, the dependent variable was belief that others would be better off if the respondent were gone (yes/no). For the first regression, the predictor was whether the respondent endorsed burdensomeness on a spouse (yes/no). For the second regression, the predictor was whether the respondent endorsed burdensomeness on a child (yes/no).

Results

Of the 70 participants that were included in each analysis, 47 reported feeling like a burden on a spouse (67.1%) and 45 reported feeling like a burden on a child (64.3%). A total of 31 participants reported feeling like a burden on both a spouse and a child, meaning that 16 reported feeling like a burden only on a spouse, and 14 reported feeling like a burden only on a child.

The first logistic regression model indicated that older adults who reported feeling like a burden on a spouse were nearly six times more likely to believe that others would be better off if they were gone compared to those older adults not endorsing perceived burden on a spouse, (Wald statistic = 6.56, odds ratio = 5.73, p = .01). As expected in the second logistic regression model, older adults who reported feeling like a burden on a child were not significantly more likely to believe that others would be better off if they were gone than those older adults not reporting burdensomeness on a child (Wald statistic = 1.26, odds ratio = 2.05, p = .26).

Given that a sizable portion of our sample reported feeling like a burden on a spouse and a child, we examined whether these individuals would be particularly likely to endorse the belief that others would be better off if they were gone. For this examination, we ran two additional logistic regressions. The first compared those who endorsed burden only on a spouse to those who endorsed burden on a spouse and a child; the second compared those who endorsed burden only on a child to those who endorsed burden on a spouse and a child. Neither of these ancillary analyses were significant (first analysis: Wald statistic = 0.48, odds ratio = 2.07, p = .49; second analysis: Wald statistic = 3.49, odds ratio = 5.80, p = .06), suggesting that there was not an additive effect of endorsing burden on both a spouse and a child.

Discussion

The results of this study supported our hypothesis, suggesting that, in our sample, older adults who reported feeling like a burden on a spouse were more likely to believe that others would be better off if they were gone, but older adults who reported feeling like a burden on a child were not. Additionally, there was not an additive effect of endorsing burden on both a spouse and child beyond endorsing burden on either a spouse or a child alone. These results indicate that older adults’ perceptions of burdensomeness on a spouse may be especially harmful in terms of indicators of severe burdensomeness (i.e., feeling like others would be better off if they were gone). This is consistent with the conceptualization of perceived burdensomeness as a continuum, such that there are less severe forms of burdensomeness (e.g., feeling like one makes things worse for others) and more severe forms (e.g., feeling like others would be better off if one were gone). It may also explain the complex pattern of earlier findings. Specifically, results from Jahn and Cukrowicz (2011) suggested that perceived burden on children was associated with greater overall perceived burdensomeness (as measured by the perceived burdensomeness subscale of the INQ, which assesses the full continuum of perceived burdensomeness, including less severe forms). Yet spouses were perceived to be the most severely burdened (Jahn & Cukrowicz, 2011); this perception of severely burdening a spouse is linked to the most severe form of perceived burdensomeness in this study. Therefore, among older adults, severe burden on a spouse may be linked to the belief that others would be better off if one were gone. Additionally, the results of this study in conjunction with previous research (i.e., Jahn & Cukrowicz, 2011) suggest that perceptions of burden on relationships with children may be associated with less severe, but more generalized, perceptions of burdensomeness.

This study has limitations worth noting. First, the data collection was cross-sectional, precluding any conclusions regarding causality, and was based on self-report measures, which may be linked to reporting biases. Our sample was also small, which may have limited our ability to detect effects. For example, the relation between feeling like a burden on a child and the most severe form of perceived burdensomeness was not significant, yet the odds ratio was relatively large (2.05) and may be a significant effect in a larger sample. Additionally, there was substantial heterogeneity among older adults reporting some degree of the most severe form of perceived burdensomeness (e.g., some reported that this belief was only slightly true for them, whereas others reported that it was very true for them) that should be further explored in future studies. For example, a latent class analysis may provide information about the best way to categorize older adults experiencing severe perceived burdensomeness and the outcomes associated with these classes. Our sample was also generally healthy (mean of 4.6 on the first item of the Short Form-8 Health Survey [SD = 1.0], on a scale of 1 to 6) and had relatively low levels of depressive symptoms (mean of 7.1 on the Center for Epidemiologic Studies Depression Scale [SD = 8.9], on a scale of 0 to 48). Therefore, these results should be replicated in samples of older adults with more severe health problems and depressive symptoms to ensure that they generalize to these samples.

Clinicians should assess not only whether older adults perceive themselves to be a burden on others, but also who is perceived to be burdened, as this can influence the severity of perceived burdensomeness. Perceived burdensomeness on spouses should be taken seriously by clinicians when assessing for suicide risk, as it is linked to severe forms of perceived burdensomeness. Additionally, perceptions of burdensomeness may warrant attention as a target of psychotherapeutic treatment, such as cognitive restructuring, to improve quality of life and potentially reduce suicide risk. In conclusion, the results of the current study suggest that there are many important facets of perceived burdensomeness in older adults. The nature of relationships and severity of perceived burden are two important components of perceived burdensomeness that may influence well-being in older adults.