Abstract
Context
Many family members of patients enrolled in hospice for less than seven days state that the hospice referral was made “at the right time.”
Objectives
To examine bereaved family members’ perceptions of the timing of hospice referral to identify aspects of the referral process that can be improved.
Methods
Open-ended interviews were conducted in seven hospice programs, interviewing bereaved family members of hospice patients who died within the first week of hospice enrollment.
Results
Of the 100 narrative interviews, 99 respondents stated that their family member was either referred “too late” (n=41) or “at the right time” (n=58) to hospice services. When families stated that referral was “at the right time,” their perceptions were based on the patient having refused earlier referral (n=8), a rapid decline in the patient’s condition resulting in the late referral (n=20), or a belief in all things coming together as they were meant to (n=11). In contrast, when families stated that referral was “too late,” their reasons were centered on concerns with the health care providers’ role in decision making (n=24), with the leading concerns being inadequate physician communication (n=7), not recognizing the patient as dying (n=11), or problematic hospice delays in referral from the nursing home or home health agency (n=4). Despite the patient refusing an earlier hospice referral, five family members believed the referral was “too late.”
Conclusion
Whereas family members identified expected concerns with communication, more than one in three stated an earlier hospice referral was not possible.
Key Words
Introduction
Over 1.4 million people receive end-of-life services through hospice each year. Yet, 34.4% of hospice patients are enrolled for only seven days or less,
1
defined as a “short” hospice stay. Over half of persons enrolled in hospice die within the first 30 days.National Hospice and Palliative Care Organization. NHPCO facts and figures: hospice care in America. Available from http://www.nhpco.org/files/public/Statistics_Research/Hospice_Facts_Figures_Oct-2010.pdf. Accessed December 28, 2010.
2
Often, late-stage admissions to hospice are “crisis, chaotic, [and] emotional.”3
Late-stage admissions mean different things for providers and families. From a provider’s perspective, some concerns with later referrals are that the dying patient and family do not receive the full benefits of hospice services4
and that family members may be at risk for depression.5
Experts believe that the appropriate referral to hospice comes three months before death.6
However, there is evidence to suggest the importance and value of short-stay admissions to hospice for the patient and family.7
Speer et al.8
determined that a patient’s predeath length of stay in hospice did not make statistically significant differences in caregivers’ grief adjustment. Our previous work found that hospice length of stay was not associated with bereaved family member perceptions of the quality of hospice care. Rather, the perception of the referral being “too late” correlated with higher rates of families reporting unmet needs, concerns, and greater overall dissatisfaction with the quality of end-of-life care.9
An analysis of the Family Evaluation of Hospice Care (FEHC) data repository, maintained by the National Hospice and Palliative Care Organization, showed that there is a substantial variation in the rates of patients’ family members stating that they were referred “too late,” from 0% to 28.1%, with the 25th percentile of 9.2% and 75th percentile of 14.0%.
9
Using data from the 2006 repository, Farrell et al.10
found that neither the examined characteristics of the hospice (for-profit status, average daily census, whether the hospice was free-standing) nor the market (hospice penetration, intensive care unit utilization at the end of life) predicted the response that the patient was referred to hospice “too late.”Based on these results, we undertook a qualitative study conducted with bereaved family members sampled from seven different hospice programs, with variation in the rate of family members reporting that the referral to hospice services came “too late,” to understand what shapes perceptions of the timeliness of referrals. Our narrative interviews focused on 1) Why are persons referred to hospice in the last seven days of life? and 2) Why do family members believe a referral to hospice within seven days of death was either “at the right time” or “too late”?
Methods
Study Population
We conducted a mortality follow-back telephone survey sampling of family members from seven hospice programs across the U.S. Six of these hospice programs were FEHC participants. Two hospice programs were selected by their ranking in the lowest quartile of their rate of respondents stating the timing of referral was “too late,” whereas four programs were sampled based on their ranking in the highest quartile. One hospice program was a non-FEHC participant, but cared for a large, urban population. Sampled hospice programs represented the following geographic regions of the U.S.: New England, Middle Atlantic, Southwestern, South Atlantic (n=2), and Pacific (n=2). One hospice program focused on an inner-city population and another served a rural population. Four of the hospice programs were for profit.
Each program sent us the names of family members of hospice patients who died while enrolled in hospice with a length of stay of seven days or less. A letter was sent to the family member requesting participation in the study. A total of 367 letters were mailed, of which 121 persons were successfully contacted by phone on follow-up. Fifty-nine persons refused, 47 phone numbers were incorrect, and 144 persons could not be contacted despite repeated phone call attempts.
Data Collection
For those who agreed to participate, a nurse interviewer completed the entire FEHC survey, which includes the question, “In your opinion, was the patient referred to hospice too early, at the right time, or too late during the course of his/her final illness?” Following the survey, respondents were audiotaped during a brief interview regarding the process of referral to hospice, focusing on why they believe that their family member was referred either “too late” or “at the right time.”
Approach to Qualitative Data Analyses
All narratives were transcribed. Two of the authors independently reviewed the narratives. One conducted her review after examining the FEHC survey that provided information regarding bereaved family member perceptions of the quality of care. A content analysis was conducted to answer why the bereaved family member believed the hospice patient was referred either “at the right time” or “too late.” Separate analyses were performed with the narratives from the “too late” respondents and the “at the right time” respondents. Weekly conference calls were conducted to discuss and arrive at a consensus on the approach to the coding and labeling of categories for the first narrative. These calls were held until agreement was reached on how to sort and formulate the codes into categories. Then one reviewer independently coded the remaining narratives. Each week, one reviewer created a worksheet that summarized relevant information from the interviews, highlighted narrative quotations that were used in arriving at classification of the family members’ responses, and noted differences between the narrative and survey data. As categories emerged, two other reviewers independently reviewed a working document with the proposed categories and examples of the narratives. Results were presented at two research seminars to allow multiple disciplines to review and comment on our approach to the content analyses.
Results
A total of 121 persons agreed to complete the FEHC survey. Of these, four were not knowledgeable about the referral process, 10 refused to be taped, and there was a recording failure for five respondents. We further eliminated two narratives because the patients were readmissions to hospice services, leaving 100 narratives for analysis. Of these 100 narratives, 58 persons stated the referral was “at the right time,” 41 stated it was “too late,” and one person stated that hospice referral was “too early.” Characteristics of the decedents and respondents are reported in Table 1.
Table 1Sample Description
| Characteristic | %; n=118 |
|---|---|
| Age (mean, SD, range) | 80.3 (13.1, 33–100) |
| Male | 44.1 |
| Race | |
| White | 82.2 |
| Black | 5.9 |
| Hispanic | 4.2 |
| Diagnosis | |
| Cancer | 39.8 |
| Cardiac | 14.4 |
| Dementia | 8.5 |
| Perception of hospice referral | |
| Too late | 39.8 |
| At the right time | 58.5 |
| Respondent | |
| Spouse | 32.5 |
| Child | 51.7 |
| Rating of the hospice quality of care | |
| Excellent | 67.8 |
| Very good | 17.8 |
Family Member Perceptions of Referrals That Came “At the Right Time”
Separate content analyses were performed based on the respondent’s perception of the timing of hospice referral. We identified four predominant explanations of family member perception that hospice referral came “at the right time.” A sudden, acute change in the medical condition resulted in a hospice referral shortly prior to death in 20 cases (Table 2). For nine of these cases, the patient had a stroke or intracranial hemorrhage that resulted in death in a short time period. Six frail, older persons developed an acute infection (usually pneumonia) or another sudden illness that resulted in their death. An example of a narrative from the family member of a 90-year-old woman who developed a sudden illness is as follows:
…she walked in (to the hospital) on Monday, and Wednesday was the day we found out that she only had a couple of days to live so we were, we were all kind of blindsided by that and, ahh, if they had said Monday, I would have said, “What does she need hospice for?”
Table 2Major Reason for Late Referral Stratified by Respondent Perception of the Timing of Hospice Referral
| “At the Right Time” (n=58 Respondents) | |
|---|---|
| Major reason for late referral | n |
| Acute change in medical condition | 20 |
| Patient refused earlier hospice referral | 8 |
| Futility model of decision making | 5 |
| All things came together as meant to be | 11 |
| “Too Late” (n=41 Respondents) | |
| Concern with health care provider | |
| Physician | 17 |
| Nurse in a nursing home | 5 |
| Late transition because of the hospice or home health agency | 5 |
| Patient refused early admission | 6 |
Three persons had a diagnosis of cancer. One person’s cancer was diagnosed on presentation to the hospital with intestinal blockage, and the perception of the family member was that the patient was doing well until this precipitous decline. This was also true for the other two cancer patients’ families. As noted by the wife of this 62-year-old patient with lung cancer:
No, there was no, there would have been no need for an earlier referral. It, it just happened like really, you know, really fast.
Finally, in two cases, the patient either suffered a medical error that resulted in death, or an acute decline of a chronic condition led to a late hospice referral.
In eight of the cases in which the family member stated that hospice services came “at the right time,” the patient refused hospice at an earlier time point. The category of hospice refusal was straightforward, with narrative text reflecting a patient choice to undergo treatments to prolong life. For example, a 52-year-old man with sarcoma understood his poor chance of survival, but he wanted to pursue aggressive care at all costs. His children struggled with this decision, but honored his choice, with a hospice referral occurring only in his last days of life.
In five narratives, we found a futility model of decision making. In this category, all aggressive treatment was pursued until toxicity or decline in the patient’s condition developed. At this point, the patient and/or family elected hospice services. None of these cases had had discussion of the potential of a hospice referral at an early time point.
The final category comprised a group of 11 respondents who believed that all things came together as they were meant to be. Often, these persons were following the guidance of physicians, their belief in God, or they simply had come to the acceptance that this was the way things were meant to be. This was best illustrated by a quotation from the son of an 89-year-old woman who died at home from chronic obstructive pulmonary disease. In response to a question regarding whether any additional services from hospice would have allowed an earlier referral, the son said:
Nothing that I’m aware of. And in her particular case, it, you know, as I mentioned, she had passed in a peaceful way and it was, umm, just appropriate, I think all of the parts came together in an appropriate manner.
For the remaining cases, there were an array of categories including persons who were only referred to hospice when there was a need for additional help in care (n=3), lack of insurance (n=1), and a patient with a feeding tube who was not accepted onto hospice at an earlier time period (n=1), as well as nine persons who stated they did not know.
Family Member Perceptions of Referrals That Were “Too Late”
Concerns With Health Care Providers
Nearly two-thirds of bereaved family members who believed that their family member was referred “too late” identified concerns with a physician (n=17), a nurse in a nursing home (n=5), or poor transition to hospice either because of a home health agency (n=2) or because hospice itself did not see the patient in a timely fashion (n=3). The majority of concerns with physicians stemmed from lapses in communication and/or recognition that the patient was dying. On retrospection, family members were frustrated with the failure of the physician in taking the lead to recognize and communicate the patient’s prognosis. This attitude is exemplified by the wife of a 71-year-old man with Stage IV prostate cancer whose neighbor helped to make the hospice referral:
I mean the doctor obviously had to know…He was a fighter, mentally, but he had to look at his condition and, and realize this is a terminal illness and it’s the ending soon. It was almost like he didn’t want to recognize that.
The daughter of an 89-year-old woman with congestive heart failure voiced her frustration at her mother not being given adequate time to prepare for her death:
…it was ABSOLUTELY clear beyond a doubt that she was dying and she needed hospice. It is definite. She is dying. Even with that, the cardiologist didn’t want to give up, I’m afraid they see patients more as machines to tinker with, to keep it going, than a human being with a soul and emotions that need to get prepared for death.
In addition, there were two cases in which a delay in the physician seeing the patient resulted in a delay in hospice referral.
In the nursing home setting, bereaved family members identified concerns with the nurse who did not recognize the patient as dying (n=3) or the staff who did not listen to the families’ requests for hospice referral (n=2). Problematic transitions to hospice services were encountered, such as a home health agency not initiating a prompt referral (n=2) and a hospice program taking too long to enroll a patient (n=3).
Patient Refusal, But Family Believe Referral Was “Too Late.”
In six narratives, the family member noted that although the patient refused hospice services at an earlier time period, they still believed that the patient was referred “too late.” In these cases, the family members acknowledged the prior refusal by the patient, but voiced their own wish for an earlier referral, noting that either the patient and/or they would have benefited from hospice services. For example, the daughter of this 89-year-old patient who died at home noted:
…when I said that she was referred too late, because she was in very, very bad pain by the time, we talked, my husband and talked her into going to hospice. So I don’t think being referred earlier would have happened. I wished that she would have accepted it earlier.
There were two cases in which the family member stated that they did not know about hospice services. In one case, the family believed the nursing home was inattentive to the patient’s condition. In three cases, the families’ reasons for the late referral were not clear. In the final case, the referral was made only when there was a need for additional help with the patient.
Discussion
Short hospice length of stay raises concerns about whether the dying patient and family receive the full range and benefits of hospice services. Although our results confirmed previous noted concerns with physician communication and prognostication, slightly more than one in three hospice referrals that occurred within seven days of death would not have been able to be referred to hospice earlier because either the patient refused hospice services or suffered a severe acute illness (e.g., a stroke) that swiftly resulted in death. Our study supports the conclusion that a portion of late hospice referrals cannot occur at an earlier time, which suggests the need for systems to be in place to deliver high-quality care for persons who will receive hospice services for only a short period of time.
Our previous research has found that, even when the hospice patient dies shortly after enrollment, only a small number of persons state that they were referred “too late,” with the majority of bereaved family members stating that they were referred “at the right time.” Ferrell et al.
10
did not find that characteristics of the hospice, the market, or sociodemographic characteristics were strongly associated with a bereaved family member perception of being referred “too late.” Using cases drawn from seven diverse hospice programs from geographically diverse regions, we examined 100 interviews with family members of hospice patients who died within one week of enrollment. Among those 58 persons who believed their family member was referred “at the right time,” 48% refused earlier hospice services or the trajectory of their terminal illness resulted in late referral. Waldrop,11
in her qualitative study of 59 family members with a short hospice length of stay, found that 44% of cancer patients in her sample were diagnosed in the month prior to death. When the cancer is diagnosed at an advanced stage, patients typically die shortly after hospice enrollment.Our study examined reasons for late referral based on patients with both cancer and noncancer diagnoses. A cancer diagnosis only accounted for three of 20 cases that we categorized as rapid disease trajectory. Nearly one-half of these were stroke cases that resulted shortly in the patient’s death, and six cases were frail, older persons who the family member perceived as doing well prior to the development of an infection or acute illness that resulted in the patient’s death.
Family members believed that physician communication and/or difficulty with prognostication were the reasons for a late hospice referral. Given the retrospective nature of the interviews, our findings should be interpreted with some caution. Prognostication can be difficult for noncancer diagnoses and the interviews with family members occurred after the death of the patient. However, prior studies
11
, 12
support their concerns. Physicians tend to overestimate survival, and communication with the patient and family regarding prognosis is often lacking.12
Additionally, lapses in communication arise because physicians can often be uncomfortable discussing death with patients and their families, or refuse to accept that they cannot administer a cure.13
, 14
For persons dying in a nursing home, family members identified concerns with the nurse or social worker that resulted in late hospice referral in five cases. In three narratives, nurses did not recognize that the patient was dying. Nursing home staff in two other cases did not follow through on family requests for a hospice referral. Nursing home staff’s recognition of decline and personal beliefs about hospice, and uncertainty of diagnosis and terminality, affect hospice referral.15
Family members also noted problems with the transition from home health agency to hospice.In six narratives, patients refused an earlier referral to hospice, yet their family members expressed their wish that hospice enrollment had come sooner. Perceptions of the optimal time for hospice enrollment may be earlier for families than for patients. Retrospectively, families gain a better appreciation of how earlier enrollment might have been helpful for the patient.
16
When interpreting these results, certain limitations of this study should be kept in mind. First, the family members’ narratives are retrospective recollections of the events. It is always possible that family members’ perceptions are influenced by their own experience of the patient’s illness and death, and that an earlier hospice referral could have been possible.
Secondly, this study included only seven hospice centers, although they were from diverse geographic regions and types of hospice programs. Additionally, we were able to locate and interview only 121 of the 367 persons identified as family members of hospice patients with a length of stay of seven days or less, which raises concerns about the generalizability of our results.
Short hospice length of stay raises concerns whether the dying patient and family receive the full range of benefits of hospice services. Our results yielded the expected findings of concerns with physician communication and prognostication. Additionally, we found slightly more than one in three hospice referrals would not have been able to be referred to hospice at an earlier point in time. These patients did not enroll mainly because the patient refused or an acute illness (e.g., a stroke) resulted in death. These results suggest the need for systems to be in place to deliver high quality of care for persons who will receive hospice services for only a short period of time.
Disclosures and Acknowledgments
This research was supported by American Cancer Society Grant PEP-07-265-01-PEP1, Hospices Organized to Promote Excellence. The funding source had no role design and conduct of the study; collection, management, analysis, and interpretation of the data; and preparation, review, or approval of the article. Drs. Joan Teno and David Casarett had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. The authors report no financial conflicts of interest.
References
National Hospice and Palliative Care Organization. NHPCO facts and figures: hospice care in America. Available from http://www.nhpco.org/files/public/Statistics_Research/Hospice_Facts_Figures_Oct-2010.pdf. Accessed December 28, 2010.
- Making the transition to hospice: exploring hospice professionals’ perspectives.Death Stud. 2009; 33: 557-580
- Can short hospice enrollment be long enough? Comparing the perspectives of hospice professionals and family caregivers.Palliat Support Care. 2009; 7: 37-47
- Length of hospice enrollment and subsequent depression in family caregivers: 13-month follow-up study.Am J Geriatr Psychiatry. 2006; 14: 264-269
- Timing of hospice referral and families’ perceptions of services: are earlier hospice referrals better?.J Am Geriatr Soc. 2005; 53: 819-823
- Extent and determinants of error in doctors’ prognoses in terminally ill patients: prospective cohort study.BMJ. 2000; 320: 469-472
- Short terminal admissions to a hospice.Palliat Med. 1993; 7: 289-294
- The relationship between hospice length of stay and caregiver adjustment.Hosp J. 1995; 10: 45-58
- Timing of referral to hospice and quality of care: length of stay and bereaved family members’ perceptions of the timing of hospice referral.J Pain Symptom Manage. 2007; 34: 120-125
- It’s too late: examining the role of hospice and market factors in late hospice referrals.J Am Geriatr Soc. 2008; 56: S1-S2
- At the eleventh hour: psychosocial dynamics in short hospice stays.Gerontologist. 2006; 46: 106-114
- “I’m not ready for hospice”: strategies for timely and effective hospice discussions.Ann Intern Med. 2007; 146: 443-449
- Why don’t patients enroll in hospice? Can we do anything about it?.J Gen Intern Med. 2010; 25: 1009-1019
- Overcoming obstacles to hospice care: an ethical examination of inertia and inaction.J Clin Oncol. 2002; 20: 2752-2755
- Referral and timing of referral to hospice care in nursing homes: the significant role of staff members.Gerontologist. 2008; 48: 477-484
- Are we referring patients to hospice too late? Patients’ and families’ opinions.J Palliat Med. 2005; 8: 521-527
Article info
Publication history
Published online: January 30, 2012
Accepted:
May 12,
2011
Identification
Copyright
© 2012 U.S. Cancer Pain Relief Committee. Published by Elsevier Inc.
