Introduction
Patients with complex care needs are often required to be cared for by different health professionals in multiple settings (Coleman, Reference Coleman2003). Changes in the setting and/or the clinicians involved in the healthcare process are classified as transitions (Schumacher & Meleis, Reference Schumacher and Meleis1994). The transition of patients from acute care to palliative care is one of the most challenging aspects for health services because transition can lead to the discontinuity of care, service duplication, medical errors, and inappropriate care (Coleman & Boult, Reference Coleman and Boult2003).
In particular, the transition from curative to palliative care merits greater attention because of the extreme vulnerability of the patients involved (Ryndes & Emanuel, Reference Ryndes and Emanuel2003). The World Health Organization defines palliative care as “a multidisciplinary approach that improves the quality of life of patients and their families facing the challenges associated with a life-threatening illness” (WHO, 2017). Palliative care should begin at the earliest stages of the trajectory of a life-threatening illness simultaneously with treatments that are intended to extend life. Palliative care can be delivered at primary and secondary care levels, or in specialist palliative care services that include integrated services such as medical consultation, home care, and residential care (Payne & Radbruch, Reference Payne and Radbruch2009).
Despite the imperative to provide palliative care throughout the course of an illness, it certainly has a more prominent role when a patient's disease advances and there is less to be achieved in terms of life prolongation (Robinson et al., Reference Robinson, Gott and Gardiner2016). The transition to a specialist palliative care service is necessary to ensure the maximum focus on meticulous symptom control, comfort, psychosocial and spiritual support, and bereavement care (Sutherland, Reference Sutherland2009).
During the transition from acute to palliative care, patients and families usually experience a shift of focus from the hope for a cure, to a hope for the elimination of suffering, as the patient suddenly faces a progressive and fatal illness. This change in expectation can be associated with feelings of a loss of control for both patients and families (Back et al., Reference Back, Young and McCown2009).
Although clinical evidence has demonstrated the benefit of a well-managed and smooth transition to palliative care in terms of quality of life and psychosocial well-being (Kirby et al., Reference Kirby, Broom and Good2014), the transition to palliative care is a process that has not been studied extensively (Marsella, Reference Marsella2009; Rocio et al., Reference Rocio, Rojas and Gonzalez2017) and the transition of terminally ill patients is a challenging aspect of clinical work. However, current evidence suggests that continuity of care is the best way to ensure effective palliative care transitions, while also enhancing the quality of the time left to live for patients and their caregivers (Gardiner et al., Reference Gardiner, Ingleton and Gott2015; Hamano et al., Reference Hamano, Morita and Fukui2017).
Several studies have focused on the continuity of care at life end (Bakitas et al., Reference Bakitas, Bishop and Caron2010; Corrales-Nevado et al., Reference Corrales-Nevado, Alonso-Babarro and Rodriguez-Lozano2012). In particular, a conceptual model (D'Angelo et al., Reference D'Angelo, Mastroianni and Hammer2015) exists that captures the elements of continuity of care during end of life (Haggerty et al., Reference Haggerty, Reid and Freeman2003). This model explains how the quality of a transition can be enhanced through careful information transfer, a trusting relationship, and by guaranteeing that flexible care is able to meet patients’ and caregivers’ changing needs throughout the progression of a life-threatening illness.
To implement an appropriate plan of care, there is a need for a better understanding of the transition phenomenon through its accurate appraisal. However, to date, no instrument has been available to evaluate the transition from curative treatment to palliative care (Hanson et al., Reference Hanson, Scheunemann and Zimmerman2010; Lofmark et al., Reference Lofmark, Nilstun and Bolmsjo2007; Schenck et al., Reference Schenck, Rokoske and Durham2014).
Among the tools developed to assess transition, the Care Transition Measure (CTM) was psychometrically tested and extensively used with older patients who had experienced care transitions (Coleman et al., Reference Coleman, Smith and Frank2002, Reference Coleman, Mahoney and Parry2005). It was used to evaluate the quality of “transitional care” in a large number of American centers, although it has not been specifically tested for use in the transition to palliative care (Hanson et al., Reference Hanson, Scheunemann and Zimmerman2010). The CTM was developed as a patient self-report measure. It underlines that discussion between healthcare professionals and patients is essential to foster an anticipatory preparation based on clarifications on what to expect during the transition (Broom et al., Reference Broom, Kirby and Good2013; Schofield et al., Reference Schofield, Carey and Love2006).
Because the communication between terminally ill patients and healthcare providers is often less than optimal (Coleman et al., Reference Coleman, Ground and Maul2015), clinicians find it difficult to communicate bad news (Horlait et al., Reference Horlait, Chambaere and Pardon2016); therefore, family caregivers, who are closely involved in patient care, represent a reference point for healthcare providers, to plan advanced care. Moreover, often when direct patient information is hard to collect, it is the caregivers that are the most reliable source of information (Coleman & Roman, Reference Coleman and Roman2015; Peruselli et al., Reference Peruselli, Paci and Franceschi1997). When this situation arises, assessing the quality of a palliative care transition from the caregiver's perspective is of paramount importance. Furthermore, in Italy, although specialist palliative care services are progressively increasing the application of the “continuum of care” model during illness progression, further improvement is required. Hence, it is not surprising that for those patients who require palliative care, the transition to a specialist palliative care setting represents the predominant way to be cared for (Italian Ministry of Health, 2015).
The purpose of this study was to test the psychometric properties of the Palliative Care Transition Measure for Caregivers (PCTM-C), as modified from the Care Transition Measure as used for patients.
Methods
To establish the psychometric properties of the CTM as modified for palliative care, two main phases were followed. Phase I focused on the construction of the modified Italian version of the instrument, whereas Phase II focused on testing its psychometric properties.
Phase I
Development of the PCTM-C
In the first phase, the original CTM was translated into Italian. Content validity of this Italian version was evaluated through focus groups and by calculating the content validity index. A first focus group was conducted with two caregivers who had recently experienced palliative care transitions, two oncology nurses, two palliative nurses, and two palliative physicians. The participants evaluated whether, in their opinion, the items of the CTM appropriately addressed the family caregivers’ experiences of the transition from acute care to palliative care. After feedback from the panel of experts the CTM was modified.
In a second focus group, the same participants were asked to judge whether the items of the modified CTM were understandable and clearly defined.
Standard focus group techniques were adopted and the discussion was moderated by the principal researcher (DD), who encouraged participants to share their experiences. The scale was modified by rewording the items to assess the caregiver's perception rather than that of the patient (e.g., “before I left the hospital” was changed into “before my relative left the hospital”). Whenever possible, the sentences that focused on the self-care aspects that did not fit with the caregivers perspectives, were reformulated (e.g., “I had all the information I needed to be able to take care of myself” was changed into “I had all the necessary information about palliative care”; “I clearly understood the purpose for taking each of my medications” was changed into “I understood the general purpose of medications that would be prescribed in palliative care to my relative”).
The items that detailed information such as how to take medications, the side effects of medications, the written list of appointments, and the written care plan were removed.
When patients are referred to a palliative care center, the information gathered should not be provided by the discharging team, but by the team that enrolls them. Because a paternalistic approach in which the clinical team usually takes health-related decisions by itself is still broadly used in Italy, we removed items that stressed the sharing of decision-making, and this reality may hinder the applicability of those items in our context.
In an effort to better address transitions from acute to palliative care, caregivers were asked about the timeliness of transitions, their feelings after a patient's discharge, and the negotiation process in regard to the goals of the care provided. More emphasis was given to feelings of trust towards the palliative care service with specific items focused on the patient's symptom management, and the caregiver's relief and support.
Content validity
The second step consisted of establishing the content validity of the pre-final version of the PCTM-C, using a rigorous judgment quantification process (Lynn, Reference Lynn1986), and the pre-final version of the scale was sent for content validity to an independent panel of eight experts (two palliative physicians, two oncologists, two nurses, a social worker, and a psychologist). They were asked to evaluate the relevance of each item of the modified CTM and of the whole CTM, using a Likert Scale from 1 = not relevant to 4 = very relevant, and to give comments and/or suggestions for modifications when appropriate. The response options were dichotomized according to the procedure by Lynn to compute the Content Validity Index (CVI) (Lynn, Reference Lynn1986).
Phase II
Psychometric testing
In the second phase of the study, the PCTM-C was tested for its psychometric properties. To evaluate the construct validity of the scale, confirmatory factor analysis (CFA) was used with MPlus (Muthén & Muthén, Reference Muthén and Muthén1998-2012). The assumption underlying the transition model is that the quality of the transitional process is strongly connected with the continuity of the care experience (Gardiner et al., Reference Gardiner, Ingleton and Gott2015). Thus, the components of the continuity of care model (D'Angelo et al., Reference D'Angelo, Mastroianni and Hammer2015), specifically communication, a trusting relationship, and integrated care, were hypothesized as the main dimensions of the scale to be confirmed through CFA. In addition, a second order factor was hypothesized to account for the holistic nature of the concept of continuity.
CFA was performed using the robust maximum likelihood estimator method, which produces standard maximum likelihood parameter estimates with corrected standard errors and chi-square test statistic robust to non-normality and to non-independence of observations. To evaluate the adequacy of the measurement model, the following fit indices and indicators of good fit were considered: Comparative Fit Index >0.95; Tucker and Lewis Index >0.95; root mean square error of approximation 0.05 to 0.08, p < 0.05; and standardized root mean square residual: < 0.08 (Barbaranelli et al., Reference Barbaranelli, Lee and Vellone2015). The internal consistency was examined through the Cronbach's alpha coefficient.
Design
This was a cross-sectional validation study.
Setting and participants
The second phase of the study was conducted between July 2015 and April 2016 in three Italian palliative care centers in the north and center of Italy, respectively named “Madonna dell'Uliveto,” “Antea Network,” and “San Marco.”
The participants were family caregivers of the patients that had transferred from an acute care to either of the palliative care centers involved in the study. Family caregivers were recruited if the patient was admitted to a palliative service for at least five days, and were identified by the healthcare providers who cared for the patient. In particular, given the specificity of the instrument to investigate factors related to the transitions from acute care to palliative care, the healthcare providers were instructed to pay careful attention to identify the primary caregiver involved in the transition process.
Other inclusion criteria included the requirement to be aged 18 years and older, a willingness to participate in the study, and the provision of a signed informed consent. Each caregiver completed the 15 items of the PCTM-C on a scoring system ranged from 1 (completely disagree) to 4 (completely agree). Lower scores indicate a poorer quality transition, and higher scores indicate a better transition experience. To obtain a user-friendly 0 to 100 score, we used the same formula outlined by Coleman et al. (Reference Coleman, Smith and Frank2002).
Results
Phase I
Content validity of the PCTM-C
The pre-final version of the PCTM-C encompassed 15 items. The CVI for each item was higher than 0.80, whereas the average CVI for the whole scale was 0.95. These results indicated a satisfactory agreement among the participating experts.
There was, however, a general concern voiced by the expert panel who evaluated the content validity regarding “not applicable” as a response choice combined with “do not know/do not remember.” The researchers discussed these comments in depth and agreed that the presence of too many response choices might hinder the clarity and comprehensiveness of the items. It was felt that the term “not applicable” could cause confusion, and that it would be preferable to remove it.
Based on these comments, the researchers decided to withhold the pre-final version of the PCTM-C and to remove only the term “not applicable” from the response choices. The PCTM-C scale was finally drafted based on these results.
Phase II
Confirmatory factor analysis of the PCTM-C
Participants
A total of 272 family caregivers completed the questionnaire. The demographic characteristics of the sample are summarized in Table 1. The mean (SD) age of the participants was 54 (14); the majority (62%) were the patients’ children or nephews. More than half of the respondents (64%) were employed and had a high educational level (77%); their relative was diagnosed with cancer (91.8%).
Table 1. Socio-demographic characteristics of caregivers (n = 272)
PC, palliative care; SD, standard deviation.
Construct validity
The CFA confirmed that the data from the transition measure referred to the three dimensions of the continuity of care model with a good fit: chi-square (degrees of freedom 86, N = 272) = 158, p < 0.001; Comparative Fit Index = 0.95; Tucker and Lewis Index = 0.94; root mean square error of approximation = 0.056 (confidence interval90% = 0.042–0.069) p = 0.237; standardized root mean square residual = 0.047. Figure 1 shows the CFA model with the statistically significant factor loadings for each individual item.
Fig. 1. Confirmatory factor analysis of the Palliative Care Transition Measures for Caregivers (PCTM-C).
The correlations between each dimension were positive, high and significant at p < 0.001 (relationship with communication, r = 0.61; relationship with integrated care, r = 0.66; integrated care with communication, r = 0.55). Therefore a second-order factor including these dimensions was specified and named PCTM-C.
Reliability
Cronbach's alpha coefficient was estimated at 0.94 for the total scale and for each factor: 0.88 (communication), 0.87 (integrated care) and 0.93 (trusting-relationship). Table 2 shows in detail the mean scores, skewness, and kurtosis for the PCTM-C items.
Table 2. Item scores of PCTM-C for the total sample
Kurt, kurtosis; PCTM-C, Palliative Care Transition Measure for Caregivers; SD, standard deviation; skew = skewness.
Discussion
Poorly negotiated palliative care transitions are the cause of avoidable medical errors, discontinuity of care, and re-admissions to hospital (Kaur & Mohanti, Reference Kaur and Mohanti2011). Health providers’ efforts to realize the transition to palliative care can lead to significant resistance from patients who are often reluctant or not yet ready to give up hope of a cure. Consequently, caregivers occupy a central position in motivating patients to begin the transition process from acute care to palliative care (Broom et al., Reference Broom, Kirby and Good2015).
Although caregivers’ support has been advocated for quality palliative care, and Coleman (Reference Coleman, Ground and Maul2015) recently developed a tool focused on family caregiver self-efficacy (Coleman et al., Reference Coleman, Ground and Maul2015), instruments that assess their perspectives are lacking (Hudson et al., Reference Hudson, Trauer and Graham2010). To fill this critical gap, we have modified and tested a measure to evaluate the quality of palliative care transition from the family caregivers’ perspectives. During the process of revising the CTM scale (Coleman et al., Reference Coleman, Smith and Frank2002) it was decided to significantly change it, and therefore perfect statistical comparisons among the new and the original version are no longer possible.
The items of the new measure (PCTM-C) were directly built on those factors that may influence the caregivers’ experience of palliative transitions (D'Angelo et al., Reference D'Angelo, Mastroianni and Hammer2015; Marsella, Reference Marsella2009), and careful attention has been paid to emphasize the collaborative efforts between team members (Back et al., Reference Back, Young and McCown2009; Broom et al., Reference Broom, Kirby and Good2013; Kirby et al., Reference Kirby, Broom and Good2014). Because the overall palliative transition experience is more than simply a passage from one care setting to another (Duggleby & Berry, Reference Duggleby and Berry2005), the coordination of interventions as well as smooth referrals may contribute to increase its quality. The new measurement of items such as the flow of information, sharing in decision-making, feelings of confidence, empowerment, and negotiation allows the level of care integration to be fully understood.
In accordance with these assumptions, the CFA confirmed the possibility of grouping the items into three factors (communication, a trusting relationship, and integrated care), which represent the essential elements for guaranteeing a high-quality transition (D'Angelo et al., Reference D'Angelo, Mastroianni and Hammer2015; Lofmark et al., Reference Lofmark, Nilstun and Bolmsjo2007). This view embodies the three types of continuity of care; informational, relational, and managerial (Haggerty et al., Reference Haggerty, Reid and Freeman2003), and is a good continuity indicator (Breton et al., Reference Breton, Haggerty and Roberge2012) because it enables the capture of all aspects contributing to the overall sense of continuity (Gulliford et al., Reference Gulliford, Naithani and Morgan2006). In line with the findings from other studies (Medigovich et al., Reference Medigovich, Porock and Kristjanson1999; Wong & Chan, Reference Wong and Chan2007), this research included caregivers involved in the transition who were in their midlife and were facing different challenges, from meeting their relative's changing needs to dealing with the demands of their social role (i.e., family, job).
Preventing caregivers’ physical and emotional strain is paramount, and it is also one of the best strategies for managing high-quality improvement in transitional care.
The aspects of care that family members rated as the poorest were those regarding communication. It is worth noting that breakdowns in communication mechanisms significantly predicted whether patients and caregivers experienced feeling abandoned, and may negatively affect their confidence in healthcare providers (Lofmark et al., Reference Lofmark, Nilstun and Bolmsjo2005; Schofield et al., Reference Schofield, Carey and Love2006). Several studies have shown how patients within an official network of organizations were less likely to feel abandoned by the healthcare system (Haggerty et al., Reference Haggerty, Roberge and Freeman2012).
Although the PCTM-C is a useful measure focused on the transition to palliative care, it does not help to overcome the uncertainty regarding when such transition should occur (Krishnasamy et al., Reference Krishnasamy, Wells and Wilkie2007). Despite this, the PCTM-C has the utility to lead healthcare providers to reflect on those elements that facilitate a sensitive transition and to consider the emotional burden and subjectivity underpinning the transition to palliative care. For instance, to ensure tailored communication and trusted relationships, a small number of key health professionals should be identified, whereas multidisciplinary collaboration should be enacted behind the scenes.
Finally, healthcare providers should use the information gathered during the caregivers’ transition experience to guide decision-making in regard to future resource management and to assist caregivers when they face emerging problems. Improving the quality of the transition has the potential to increase an early and smooth referral to palliative care as well as to enhance caregivers’ physical and psychological well-being and satisfaction levels (Casarett et al., Reference Casarett, Pickard and Bailey2008).
Limitations and future directions
The findings from this study need to be viewed in the light of some limitations. First, the study provides data only about one type of transition, from acute care to palliative care. Thus, future research is needed to confirm these findings in larger and more varied samples of family caregivers who are involved in different types of transitions to palliative care (e.g., from a long-term setting to palliative care).
Second, we addressed the caregiver's perspective without having an insight into that of the patient. Further research efforts might include the patients’ perspective with the aim of finding the right communication channels congruent with patients’ care preferences and awareness levels. Finally, our results present only initial evidence for the PCTM-C validation, so future research needs to better establish its psychometric proprieties by using a more rigorous approach, especially with regard to reliability.
Conclusion
PCTM-C allows the evaluation of the quality of care delivered during the transition to palliative care from a caregiver's perspective. It can assist in bridging the gap during the transitional process by offering insights into various interaction elements, such as those between caregivers and the healthcare providers involved in the transition. It has the potential to improve the continuity of care for an early and timely referral to palliative care, while focusing on the coordination of individual elements of the plan of care. This aspect may contribute to other desirable palliative care outcomes, such as the decrement of palliative service abandonment and the reduction of unnecessary transitions.
Acknowledgments
There are no commercial associations neither conflict of interest to disclose from any of the authors. We would like to thank all the healthcare providers and the family caregivers who participated in the study. We are also grateful to Ralph James Holland for the English editing of the manuscript.
