Introduction
Cancer is presently viewed as a modern epidemic (Simões, Reference Simões2014) and one of the leading causes of death worldwide (World Health Organization, 2015); it is the second leading cause of death in both Portugal (Instituto Nacional de Estatística, 2015) and the United States (Siegel et al., Reference Siegel, Miller and Jemal2016).
Cancer affects not only the patient but also encompasses the entire family system and its elements (Rolland, Reference Rolland2005). The patient suffering from cancer and his or her family members are interdependent, with each individual having a significant effect upon another (Northouse, Reference Northouse2012). Healthcare professionals should afford special consideration to the manner in which a family responds to cancer and, when necessary, families should benefit from formal support. Rait (Reference Rait, Holland, Breitbart and Butow2015) suggests that traditional patient-centered care would benefit if replaced with family-centered care.
The terminal phase of cancer is particularly demanding on the family system. The imminent death of a family member is the most difficult crisis a family must confront and adjust to; as such, it is considered a “powerful nodal event that shakes the foundation of family life” (Walsh & McGoldrick, Reference Walsh, McGoldrick, Walsh and McGoldrick2004, p 8). Families are forced to begin a major reorganization of the system (e.g., realignment of relationships, redistribution of roles and functions), simultaneously situational and anticipatory (Relvas, Reference Relvas1989), while grieving for the dying and assessing parallel losses (Walsh & McGoldrick, Reference Walsh, McGoldrick, Walsh and McGoldrick2004).
Facing such a crisis within the system, family members, and particularly family caregivers, may experience difficulties in functionally adjusting to terminal cancer and its demands. Despite the positive outcomes (e.g., personal growth) related to caring for a terminally ill person (Kang et al., Reference Kang, Shin and Choi2013), family caregivers may show increased levels of psychological morbidity, such as depression (Fasse et al., Reference Fasse, Flahault and Brédart2015; Govina et al., Reference Govina, Kotronoulas and Mystakidou2015), anxiety (Perez-Ordonez et al., Reference Perez-Ordóñez, Frías-Osuna and Romero-Rodríguez2016; Rumpold et al., Reference Rumpold, Schur and Amering2016), heightened burden (Govina et al., Reference Govina, Kotronoulas and Mystakidou2015; Rha et al., Reference Rha, Park and Song2015), and complicated anticipatory grief responses (Tomarken et al., Reference Tomarken, Holland and Schachter2008). Family caregivers tend to report lower levels of quality of life (Morishita & Kamibeppu, Reference Morishita and Kamibeppu2014) and are at higher risk for developing or exacerbating physical health disorders (e.g., constipation, hypertension) (Haley et al., Reference Haley, LaMonde and Han2001), followed by an increased risk of mortality (Schulz & Sherwood, Reference Schulz and Sherwood2008).
Some predictors are commonly identified in the literature as negatively influencing outcomes of caring for a terminally ill family member. Female (Given et al., Reference Given, Wyatt and Given2004; Morgan et al., Reference Morgan, Williams and Trussardi2016; Valeberg & Grov, Reference Valeberg and Grov2013) and older individuals (Marwit & Meuser, Reference Marwit and Meuser2002), spouses (Fasse et al., Reference Fasse, Flahault and Brédart2015; Götze et al., Reference Götze, Brähler and Gansera2014; Rumpold et al., Reference Rumpold, Schur and Amering2016), adult children (Given et al., Reference Given, Wyatt and Given2004), and individuals with a history of psychiatric disorders (Tomarken et al., Reference Tomarken, Holland and Schachter2008) are more likely to develop some kind of psychological morbidity. Furthermore, concurrent stressful life events (Tomarken et al., Reference Tomarken, Holland and Schachter2008), financial burden (Götze et al., Reference Götze, Brähler and Gansera2014; Tomarken et al., Reference Tomarken, Holland and Schachter2008), and perceived low social support (Götze et al., Reference Götze, Brähler and Gansera2014; Tomarken et al., Reference Tomarken, Holland and Schachter2008) may predict family caregivers’ maladjustment. Finally, with respect to the patient's characteristics, younger patients (Tang et al., Reference Tang, Chang and Chen2013), the patient's physical decline (Valeberg & Grov, Reference Valeberg and Grov2013), and patient's psychological maladjustment (Götze et al., Reference Götze, Brähler and Gansera2014) seem to increase family caregivers’ psychological morbidity.
Research on caregivers has focused more on the role of the family's primary caregivers (Given, Reference Given, Wyatt and Given2004), leaving the impact of terminal cancer on other family members who may not be exclusively responsible for the care of the person with cancer inadequately represented in the literature. Therefore, describing the impact of terminal cancer on family caregivers, both primary and nonprimary, together with identifying which factors may be associated with how they experience psychological morbidity, are essential steps in designing interventions that might prevent or reduce psychopathological symptoms in family caregivers of people with terminal cancer. The purpose of this study was thus twofold: (1) to determine the prevalence of psychological morbidity with regard to psychological distress, depression, anxiety, somatization, and complicated anticipatory grief (high to severe grief levels) in family caregivers of people living with terminal cancer and (2) to describe which factors may be related to an unadjusted emotional response.
Method
Data collection and procedures
Data were collected from January 2014 to March 2016 at two hospice palliative care units from an oncology hospital and a general hospital and through a specialist home-based palliative care team from a general hospital. For inclusion in the study, participants fulfilled the following prerequisites: (1) a family member of a person with terminal cancer providing some level of care to his/her relative; (2) aged 18 years or older; and (3) able to understand the study, read, and complete the scales. Permission to conduct this study was received from the institutional review boards and ethic committees of each medical institution; all participants provided written informed consent.
The sample collection procedures differed depending on the context of care: hospice palliative care units or home-based palliative care. In hospice palliative care units, family caregivers, either closely accompanying their relatives in the unit or regularly visiting them, were invited by a healthcare professional (e.g., nurse) to participate in this study. If the family caregiver agreed to participate, he or she was then taken to a private room where the interviewer clarified the goals of the study. Upon obtaining written consent, participants completed a self-administered protocol, with the interviewer available nearby for any queries.
In home-based palliative care, family caregivers receiving support from the home professional care team were contacted by telephone by a healthcare professional (e.g., nurse) to participate in the study. If the family caregiver agreed to participate, the interviewer administered the protocol in similar fashion to the hospice context at the person's home.
In both cases, the assessment protocol lasted approximately 30 minutes.
Measurements
Family caregivers’ demographics and patients’ medical information
A sociodemographic and complementary data questionnaire was used to collect general information related to: (1) demographics, (2) family, (3) patient care, (4) social support, (5) psychological/spiritual support, and (6) patient medical data.
Psychological distress, depression, anxiety, and somatization
Participants responded to anxiety, depression, and somatization subscales of the Portuguese version of the Brief Symptom Inventory (BSI) (Canavarro, Reference Canavarro, Simões, Gonçalves and Almeida1999). To achieve the goals of this study, we also computed one global severity index (GSI), summing up the items proposed in a short version of the BSI (Derogatis, Reference Derogatis2001) to obtain a measure of individual's global distress. T scores were used as indicative of caseness: subjects were considered has clinical cases or noncases according to the general rule for caseness suggested by Derogatis (Reference Derogatis2001): T ≥ 63 in GSI or in at least two subscales. The Portuguese version of BSI shows good psychometric properties (Canavarro, Reference Canavarro, Simões, Gonçalves and Almeida1999). For this study, the following Cronbach's α results were obtained: “somatization,” α = 0.81; “depression,” α = 0.81; “anxiety,” α = 0.84; and GSI (sum of the selected items), α = 0.91.
Anticipatory grief
The Marwit-Meuser Caregiver Grief Inventory – Short Version (MM-CGI-SF) was used to measure the anticipatory grief experience. Its 18 items are based on a 5-point Likert scale. The MM-CGI-SF allows for a “total grief level” and three subscale scores: “personal sacrifice burden,” “heartfelt sadness and longing,” and “worry and felt isolation” (Marwit & Meuser, Reference Marwit and Meuser2005). For the purpose of this study, only the total grief level was considered. Total scores above the 75th percentile (high levels of grief) and 90th percentile (severe levels of grief) were considered to represent the at-risk level for a complicated anticipatory grief response. The MM-CGI-SF shows good psychometric properties for both the original (Marwit & Meuser, Reference Marwit and Meuser2005) and Portuguese (Areia et al., Reference Areia, Major, Relvas, Relvas and Major2016 b) versions. For this study, a Cronbach's α = 0.86 was obtained for the entire scale.
Unmet needs
The Family Inventory of Needs (FIN) is a questionnaire designed to measure the importance of caregiving-related needs of families with advanced cancer patients and the extent to which these needs are met by healthcare professionals (Kristjanson et al., Reference Kristjanson, Atwood and Degner1995). It includes 20 items rated on two subscales: FIN-importance and FIN-fulfillment. FIN-importance measures the importance of each need on a scale that ranges from 1 (not important) to 5 (extremely important). On the subscale FIN-fulfillment, respondents indicate whether those needs rated as being important (2 or more) were met, partly met, or not met (Fridriksdottir et al., Reference Fridriksdottir, Sigursdardottir and Gunnarsdottir2006). FIN proved to be a reliable tool in both the original (Kristjanson et al., Reference Kristjanson, Atwood and Degner1995; Fridriksdottir et al., Reference Fridriksdottir, Sigursdardottir and Gunnarsdottir2006) and Portuguese (Areia et al., Reference Areia, Major, Relvas, Relvas and Major2016 a) versions. For the purpose of this study, only the total number of unmet needs, obtained from the FIN-fulfillment, was used. The FIN-fulfillment items showed a Cronbach's α = 0.92 for this study.
Family functioning
The Systemic Clinical Outcome Routine Evaluation-15 (SCORE-15) is a measure of family functioning comprising 15 items and based on a 5-point Likert scale. Items are organized into three dimensions—“family strengths,” “family difficulties,” and “family communication”—and a total score. Both the original (Stratton et al., Reference Stratton, Bland and Janes2010) and Portuguese versions of SCORE-15 (Vilaça et al., Reference Vilaça, Silva, Relvas, Relvas and Major2015) have been shown to be valid and reliable tools. For the purpose of this study, only the total score was used. In this case, the higher the total score, the poorer the family functioning. For this study, a Cronbach's α = 0.86 for the SCORE-15 total score was obtained.
Statistical analyses
Statistical analyses were performed using IBM SPSS Statistics and AMOS, version 22. Descriptive and frequency statistics were calculated to provide information about participants’ characteristics and the patient's clinical information and to determine the prevalence of severe psychological morbidity. With regard to the BSI-18, a comparison with the Portuguese community norms (Canavarro et al., Reference Canavarro, Nazaré, Pereira, Gonçalves, Simões and Almeida2017) was also conducted through the calculation of the difference effect size, Glass's delta (∆).
To determine the predictors of psychological morbidity, the structural equation modelling method was applied. Simple linear regression and uni- and multivariate multiple regression models were tested through the maximum likelihood estimation. More specifically, analyses were conducted to understand the influence of demographic characteristics (age, sex, education), relationship to the family member suffering from cancer, type of patient care (caregiving role, time spent in caregiving per day, care setting), unmet needs by healthcare professionals (i.e., number of important unmet needs), and family functioning with respect to the experience of psychological morbidity. These predictors were chosen for inclusion in the study following consultation of previous international literature.
For each tested model, violations of assumptions were assessed beforehand. Mahalanobis distance was used to detect the presence of outliers. The normality of variables was evaluated through uni- and multivariate skewness and kurtosis. Variance inflation factors were used to detect multicollinearity. Any violation of assumptions was identified in the tested models.
Results
Subjects
One hundred and twelve family caregivers for persons living with terminal cancer completed the assessment protocol. The majority of participants were women (n = 92; 82.1%) with a mean age of 44.45 years (SD = 15.32) and educational level at less than a university degree (n = 71; 63.4%). Concerning the familial relationship to the person with terminal cancer, 42 participants (37.5%) were adult children and 22 (19.6%) were spouses. As for the caregiver role, 52 participants (46.4%) perceived themselves as primary caregivers and 60 (53.6%) as nonprimary caregivers. The majority of individuals with terminal cancer were receiving outpatient care (n = 62; 55.4%) and were diagnosed with either gastrointestinal cancer (n = 27; 24.5%), respiratory cancer (n = 23; 20.9%), or other solid tumors (n = 23; 20.9%). Last, one-half of the patients with cancer had received their diagnosis less than a year before the study (n = 56; 50.0%). Table 1 presents a summary of the sociodemographic data and clinical information.
Table 1. Sample characteristics, N = 112
* Mean (SD).
Prevalence of psychological morbidity
According to the rule for caseness (Derogatis, Reference Derogatis2001), a considerable proportion of participants was found to present clinically significant psychological morbidity. Specifically, (1) 66.1% of family caregivers reported high levels of distress (mean = 6.34; SD = 14.38), (2) 68.8% showed high risk of depression disorder (mean = 9.67; SD = 5.58), (3) 72.3% showed high risk of anxiety disorder (mean = 10.10; SD = 5.82), and (4) 50.9% of the sample reported high levels of somatization (mean = 6.57; SD = 5.18).
Observed results were higher than the community national average (Canavarro et al., Reference Canavarro, Nazaré, Pereira, Gonçalves, Simões and Almeida2017). Indeed, when comparing with the national average, large size effects were obtained for distress (Glass’ ∆ = 0.96), depression (Glass’ ∆ = 1.01), and anxiety (Glass’ ∆ = 0.89). The effect size for somatization was medium (Glass’ ∆ = 0.56).
Regarding the anticipatory grief level measured by the MM-CGI-SF, the obtained mean score was 56.16 (SD = 12.61) and 25.9% of participants reported high (16.1%) to severe level of grief (9.8%).
Predictors of psychological morbidity
Multivariate multiple regression models were established with the endogenous variables depression, anxiety, and somatization, considering that the dependent variables errors are correlated (re1 depression).re2(anxiety) = 0.76–0.79; re1(depression).re3(somatization) = 0.47–0.54; re2(anxiety).re3(somatization)= 0.62–0.74). Linear and univariate multiple regression models were established considering the endogenous variables distress and anticipatory grief independently.
Models were defined with regard to the following exogenous variables: age, sex and education (models 1, 2, and 3); being a spouse and being an adult child (models 4, 5, and 6); caregiving role, hours of care per day, care setting (models 7, 8, and 9); number of unmet needs by healthcare professionals (models 10, 11, and 12); and family functioning (models 13, 14, and 15).
Results of linear and multivariate multiple regression analysis and simple regression analysis are detailed in Table 2.
Table 2. Standardized regression coefficients (β) and determination (adjusted R 2) for each tested model
* Multivariate multiple regression model.
† Univariate multiple regression model.
‡ p < 0.05.
§ p < 0.01.
‖ Simple linear regression model.
¶ p < 0.001.
** A higher score represents a poorer family functioning.
Sex revealed to be a statistically significant predictor of psychological morbidity. In particular, women show higher levels of depression (β depression.sex = −0.25, p < 0.01), anxiety (β anxiety.sex = −0.25, p < 0.01), somatization (β somatization.sex = −0.27, p < 0.01), and distress (β distress.sex = −0.29, p < 0.01).
Older individuals are at higher risk of developing complicated anticipatory grief responses (β grief.age = 0.24, p = 0.01). This is also true when the family member is a spouse (β grief.spouse = 0.30, p < 0.01) or a primary caregiver (β grief.caregiving role = −0.34, p < 0.01).
A higher number of unmet needs was related to higher scores in terms of depression (β depression.unmet needs = 0.09, p = 0.01), anxiety (β anxiety.unmet needs = 0.13, p < 0.001), distress (β distress.unmet needs = 0.24, p = 0.01), and anticipatory grief (β grief.unmet needs = 0.21, p = 0.05). Finally, family functioning revealed to be a strong predictor of psychological morbidity. In particular, lower family functioning was related to higher levels of depression (β depression.family functioning = 0.35, p < 0.001), anxiety (β anxiety.family functioning = 0.29, p = 0.002), distress (β distress.family functioning = 0.31, p < 0.001), and anticipatory grief (β grief.family functioning = 0.34, p < 0.001). Figure 1 depicts these findings.
Fig. 1. Summary of the findings on the predictors of psychological morbidity of family caregivers of people with terminal cancer.
Discussion
The aim of the present study was to examine the prevalence and predictors of psychological morbidity of family caregivers of individuals living with terminal cancer. Specifically, the prevalence of high levels of depression, anxiety, somatization, psychological distress, and complicated anticipatory grief was analyzed. Demographic data, care-related factors, unmet needs, and family functioning were also investigated as potential predictors of psychological morbidity.
Prevalence of psychological morbidity
Findings from this study showed a high prevalence of psychological morbidity in family caregivers of individuals with terminal cancer receiving palliative care. It is noteworthy that more than half of the participants reported a positive risk for psychological distress, depression, anxiety, and somatization, showing significantly higher levels than the community national average (Canavarro et al., Reference Canavarro, Nazaré, Pereira, Gonçalves, Simões and Almeida2017). Additionally, more than one-quarter of the participants reported a positive risk for complicated anticipatory grief. These prevalence rates of psychological morbidity are somewhat alarming and intensify the sense of urgency, as stated by Rait (Reference Rait, Holland, Breitbart and Butow2015), for expanding patient-centered care more in the direction of family-centered care. Undoubtedly, these results corroborate the idea that cancer is a “we-disease” (Kayser et al., Reference Kayser, Watson and Andrade2007) and family members are cosufferers alongside cancer patients (Northouse, Reference Northouse2012), showing heightened levels of psychological morbidity.
Factors associated with psychological morbidity
As for the predictors, it was found that older individuals, women, spouses, primary caregivers, important unmet needs by healthcare professionals, and poor family functioning may influence the presence of some form of psychological morbidity.
Some predictors were expected, such as age and sex. Indeed, the literature frequently points out that older individuals (Marwit & Meuser, Reference Marwit and Meuser2002) and women (Burridge et al., Reference Burridge, Barnett and Clavarino2009; Morgan et al., Reference Morgan, Williams and Trussardi2016) are more likely to develop mood-related complications when facing an imminent death of a family member.
Notwithstanding, results regarding the caregiving role and the relationship to the person with terminal cancer were slightly surprising. It was expected that being a primary caregiver and/or a partner would predict greater psychological morbidity. Indeed, it is well established in the literature that primary caregivers and spouses are at higher risk of developing severe mood complications, such as depression and anxiety (Fasse et al., Reference Fasse, Flahault and Brédart2015; Given, Reference Given, Wyatt and Given2004; Gotze, Reference Götze, Brähler and Gansera2014). The current results showed that primary caregivers and nonprimary caregivers share a similar risk of developing psychological morbidity, in terms of distress, depression, anxiety, and somatization. The caregiving role and relationship to the person with terminal cancer significantly influenced the levels of grief only. This indicates that primary caregivers and partners (20 of 22 spouses in this study considered themselves primary caregivers) are more likely to develop complicated anticipatory grief responses.
We believe that these results may be explained by an increased awareness of the relative's approaching death and levels of suffering, as well as by the multiple present losses (e.g., loss of freedom) that primary caregivers/spouses may experience as a result of their caregiving tasks. As found by Janze and Henriksson (Reference Janze and Henriksson2014), family primary caregivers/spouses are more likely to become increasingly aware of death because they continually experience first-hand the extent to which their relative's condition is deteriorating. They may also anticipate and imagine the physical deterioration and death of their family member by picturing a variety of scenarios. Although primary caregivers struggle to focus on the present and on responding to practical challenges related to caregiving, it is unavoidable that death is always either in the background or at the forefront (Janze & Henriksson, Reference Janze and Henriksson2014), which may explain an increased anticipatory grief response. Additionally, as stated previously, predeath grief is not only focused on future losses (e.g., the patient's death), but also on present and past losses. Becoming a primary caregiver of a terminally ill person is considered a complex transition process that implies a wide range of changes in identity, roles, relationships, abilities, and patterns of behaviors (Schumacher & Meleis, Reference Schumacher and Meleis1994) over a short period; thus, primary caregivers may be more exposed to several losses. As such, primary caregivers/spouses may not only grieve the imminent death of their family member but also the present changes in the relationship with their relative, roles, social involvement (isolation), personal projects, and pursuits and so on, which explains a higher risk of complicated anticipatory grief responses. Likewise, present caregiving demands are related to higher levels of loss appraisal (Oberst et al., Reference Oberst, Thomas and Gass1989) and, as is well-known, the terminal phase of cancer is particularly demanding on primary caregivers.
Confronting the challenging task before them, primary caregivers tend to postpone their emotional responses by withholding their own feelings and needs. They usually take on the role of being the “strong one” and thus focus on the present by assigning priority to the needs of their ill loved-ones and addressing whatever practical demands and caregiving tasks that may arise (Janze & Henriksson, Reference Janze and Henriksson2014). Although it is unknown whether this response has a positive or negative outcome in the middle and long term (Janze & Henriksson, Reference Janze and Henriksson2014), we believe that this may explain the study results that indicate, contrary to what was previously expected, that being a primary caregiver/spouse does not have a direct effect on distress, depression, anxiety, and somatization. There is also some evidence that being a primary caregiver may bring positive consequences (e.g., Kang et al., Reference Kang, Shin and Choi2013). Indeed, despite the well-known challenges of caring for a terminally ill individual, the caregiving experience may simultaneously serve as a wake-up call about life, heightening and/or altering the sense of priorities, changing one's life patterns of and reorienting life goals and hopes (Walsh, Reference Walsh and Walsh2003). Thus, further research should thoroughly examine whether these positive effects can explain the similar levels of distress, depression, anxiety, and somatization obtained by both primary caregivers and nonprimary caregivers in our study.
The results of this study support the importance of attending not only to family primary caregivers, but also to family nonprimary caregivers because both seem to show a similar probability for developing psychological morbidity.
Additionally, our results show that unmet needs seem to be a strong predictor of psychological morbidity in family caregivers of people with terminal cancer. Similar results were reported by Fridriksdottir et al. (Reference Fridriksdóttir, Saevarsdóttir and Halfdánardóttir2011) and Buzgová et al. (Reference Buzgová, Spatenková and Fukasová-Hajnová2016), who found associations between perceived unmet needs, psychological morbidity (anxiety), and lower quality of life. As briefly mentioned earlier, families commonly prioritize a relative's wishes and needs above their own, deemphasizing their own needs (Milne & Quinn, Reference Milne, Quinn, Hudson and Payne2009). This may partially preclude an adequate fulfilment of these needs by healthcare professionals. In addition, there is still a gap in palliative care settings in adequate care for families of terminally ill people (Milne & Quinn, Reference Milne, Quinn, Hudson and Payne2009). Healthcare professionals still tend to focus on the patient and to dedicate little time to addressing family members’ difficulties, also being reluctant to openly attend to these matters in the cancer patient's presence (Grande et al., Reference Grande, Stajduhar and Aoun2009). Nevertheless, adequate patient-family support must overcome these difficulties, and families should be encouraged to identify their own needs, and in so doing, better fulfill them, as unmet needs seem to have worrying negative outcomes for family members (Milne & Quinn, Reference Milne, Quinn, Hudson and Payne2009).
Finally, consistent with Melbourne Grief Studies (Kissane & Bloch, 2002), the results of our study confirmed that family functioning is related to individuals’ psychological morbidity. Specifically, poorer family functioning may predict individuals’ maladaptive responses to the approaching death of a family member (e.g., complicated anticipatory grief), whereas it is legitimate to expect that healthier family functioning may act as a buffer against the impact of terminal cancer on family caregivers.
Indeed, in a systemic perspective, family is considered an interactional system and so, the impact of a death, actual or in the near future, would have far-reaching effects for every member and all other relationships within a complex chain of influences that reverberates throughout the family network (Walsh & McGoldrick, Reference Walsh, McGoldrick, Walsh and McGoldrick2004). Maladaptive relational processes within the family system constrain the full coping potential of the family as a whole, the patient, and all other family members (Zaider & Kissane, Reference Zaider, Kissane, Holland, Breitbart and Butow2015), putting them at risk for developing psychological morbidity. Our results, therefore, emphasize the importance of developing and delivering relationship-focused interventions to strengthen the quality of family relationships and reduce the risk for psychological morbidity in people living with terminal cancer and their family caregivers (Zaider & Kissane, Reference Zaider, Kissane, Holland, Breitbart and Butow2015) and also promote not only the family caregivers’ personal growth (Walsh, Reference Walsh and Walsh2003), but also the experience of positive consequences related to the caregiving of the terminally ill (Kang et al., Reference Kang, Shin and Choi2013).
Strengths and Limitations
Within the field of family studies and the contexts of oncology and palliative care, the major strengths of this study lie in how it scrutinized and examined the factors related to family caregivers’ maladjustment, identifying a worrying prevalence of psychological individual morbidity strongly linked to family functioning and to inadequate family support.
The main weaknesses of our study may be found in its cross-sectional nature, the modest sample size, and its homogeneity (e.g., most participants were female). This study was also limited in that it did not consider certain other factors as influencing the presence of psychological morbidity. For example, it would have been interesting to investigate how a patient's characteristics, such as type of cancer, level of physical deterioration, and psychological adjustment, influence his or her relative's adjustment as well as to strictly underscore the interdependence of patient-family adjustment. In addition, other important care-related variables should have been addressed. Fatigue and related perceived burden emerge as crucial variables in palliative care settings (Peters et al., Reference Peters, Goedendorp and Verhagen2015) and, thus, it should have been addressed in depth to determine how much a heightened burden may contribute to family caregivers’ maladjustment to an imminent death of their relatives.
Research and clinical implications
The present study has implications for both research and clinical practice. In the area of research, this study points to the importance of more research focused on factors related to family caregivers’ psychological morbidity. Additionally, a scrutinized study of family caregivers’ responses in palliative care should investigate the mediator role of family functioning in family caregivers’ adaptation to the terminal phase of a cancer.
As for this study's clinical implications, it provides some preliminary evidence that psychosocial interventions should be targeted first to older individuals, women, spouses, and primary caregivers. As is known, clinicians must move forward to a family-centered approach to address and better fulfill family caregivers’ needs.
Given the alarming risk of psychological morbidity for family caregivers of people living with terminal cancer, it is hoped that this work will trigger future research in the field and raise awareness of the importance of better support for families that face the imminent death of a family member.
Conclusions
Family caregivers of individuals with terminal cancer reported worrying levels of psychological morbidity, specifically a high prevalence of global distress, depression, anxiety, somatization, and complicated grief. Factors associated with family caregivers’ psychopathological symptomatology (i.e., global distress, depression, and anxiety) accounted for sex, unmet needs by healthcare professionals and family functioning. Remarkably, all the predictors studied, with the exception of sex, revealed to have an influence on complicated anticipatory grief responses. Palliative care must ensure that whole family support and family-centered interventions adequately replace the traditional patient-, primary caregiver-interventions to consider the entire family as a unit of care.
Acknowledgments
This research was undertaken as part of a doctoral project entitled “Terminal Cancer Patients’ Family Needs in Palliative Care.” Doctoral fellowship financed by FCT (Fundação para a Ciência e Tecnologia) (SFRH/BD/86178/2012).
