Next of Kin’s Notification of Patient’s Death: A Qualitative and Quantitative Preliminary Analysis

Introduction

Most reports in the literature relate to the notification of death by physicians and other professionals, such as policemen, victim advocates, social workers, and chaplains. ^{1 -8} The treating physician is usually the one who notifies the patient’s next of kin. This is one of the most challenging and difficult situations that the medical staff can experience. Many publications have assessed the training, experience, and preferences of those who notify families of death (notifiers). ^{1 -4,7 -11}

Home palliative care units have been established in many countries and patients are cared for in these units throughout the entire day. As a result, patients treated in these units die at home with the family at hand.

To our knowledge, the circumstance of a next of kin notifying the medical staff of the occurrence of a family member’s death has not been investigated or reported before. Our home care palliative unit takes care of patients at home, in most cases, patients with terminal cancer. A personal physician and nurse are assigned to each patient. They call on the patient at home, as needed, during working hours and are also available after working hours. When a patient dies at home, the next of kin calls a medical staff member and notifies him about the death. A physician then comes to the patient’s home, confirms death, and provides a death certificate.

Evaluating the nonmedical next of kin’s notification about patient’s death at home can help the medical staff to understand the notifier better and to respond better to the notification. Because there is a lack of research on next of kin death notifications, the aim of the current study was to characterize death notifications by next of kin to the medical staff in a home palliative care setting. We aimed to conduct a qualitative analysis of the content of the death notifications to discover the overt and hidden meanings. In addition, we aimed to document and conduct a quantitative analysis of the circumstances of the death notification in relation to the notifier’s identity, the timing of the notification, the identity of the staff member who received the death notification, the notifier’s requests, and demographic characteristics of the deceased and his or her family.

Methods

Results

Participants

During the 2½-year study period, 371 patients were treated in the home palliative care unit. Two hundred sixty (70%) patients died at home. One hundred fifty-three (41% of all patients and 58.8% of those who died at home) were included in the final qualitative analysis. Some patients who died at home during the study period were not included in the study because the death notification could not be documented in real time and/or there were communication barriers between the medical staff members and family members.

Of the 153 patients, 147 (96%) had cancer, exactly the same as the percentage of patients with cancer among all the patients treated in the home palliative care unit during the study period. All but one of the death notifications were made by telephone to a medical staff member. The one exception was an e-mail message by the next of kin (who was also a physician) to a medical staff member. Table 1 depicts the sociodemographic characteristics of the patients whose next of kin notified the medical staff of death.

OPEN IN VIEWER

Table 1.

Sociodemographic Characteristics of the Patients.^a,b

Variable	n	%
Gender
Male	68	47.6%
Female	75	52.4%
Age (years)
≤50	8	5.4%
51-70	56	39.2%
71+	79	55.2%
	143
Mean (SD)	70.9 (15.1)
Range	4-94
Marital status
Married	88	61.5%
Widow	37	25.9%
Divorced	15	10.5%
Single	3	2.1%
	143
Country of birth
Former Soviet
Union	51	35.7%
North Africa	28	19.6%
Israel	27	18.9%
Asia	16	11.2%
East Europe	14	9.8%
West Europe and North America	4	2.8%
South America	3	2.1%
	143
Primary next of kin (more than one possible)
Spouse	73	51.0%
Child	52	30.9%
Foreign caregiver	17	11.9%
Other family member	9	6.3%
Parent	2	1.4%
	143
Duration of stay in home care unit (days)
Mean (SD)	49.9 (57.8)
Median	32.5
Range	0-307

Abbreviation: SD, standard deviation.

^aN = 143.

^bDue to missing data in 10 cases, only 143 of the 153 notifications were included in the statistical analysis.

Characteristics of Death Notifications

Table 2 shows the characteristics of death notifications. The gender of the notifying next of kin was similar to the male/female distribution in the general population. Most of the notifiers were children of the deceased (54.9%). In only 25.4% of the cases, the notifiers were spouses, though they comprised 51% of the next of kin.

OPEN IN VIEWER

Table 2.

Characteristics of the Notification of Death.^a,b

Variable	n	%
Gender of notifying next of kin
Male	65	47.1%
Female	73	52.9%
	138	Missing = 5
Relationship of notifying person to patient
Children	78	54.9%
Spouse	36	25.4%
Other family member	22	15.5%
Friend	5	3.5%
Foreign worker	1	0.7%
	142	Missing = 1
Time of death notification
8 am to 4 pm	48	34.0%
4 pm to midnight	42	29.8%
Midnight to 8 am	51	36.2%
	141	Missing = 2
Staff member who received notification
Nurse	71	50.0%
Senior physician	42	29.6%
Resident physician	26	18.3%
Social worker	2	1.4%
Clinic clerk	1	0.7%
	142	Missing = 1
Did the receiver of the notification perceive the notification as meaning the patient had actually died?
Yes	134	97.1%
No	4	2.9%
	138	Missing = 5
Was the patient actually dead when the medical staff arrived at the home?
Yes	128	90.1%
No	14	9.9%
	142	Missing = 1

^aN = 143.

^bDue to missing data in 10 cases, only 143 of the 153 notifications were included in the statistical analysis.

Qualitative Analysis

Figures 1 and 2 present the thematic map that has emerged from the qualitative analysis. Two themes emerged from the analysis, direct notification and indirect notification. The map is divided into 2 parts, each representing one of the 2 themes. The different codes on the left side of the map were in part equal and in part unequal between the 2 themes. Selected quotations that captured the essence of the codes are presented and then the subthemes and the 2 themes.

OPEN IN VIEWER

Figure 1.

A thematic map of the qualitative analysis of death notifications: direct notifications.

OPEN IN VIEWER

Figure 2.

A thematic map of the qualitative analysis of death notifications: indirect notifications.

A notification was considered direct if the notifier stated clearly and in specific words that the patient was dead or not alive. A notification was considered indirect if the notifier spoke about death in other words, such as “it’s over” or “the patient finished” or “is not with us,” or used physical descriptions relating to death such as cessation of breath or pulse or no movement, and so on, without mentioning the words death or not being alive.

Direct as opposed to indirect death notifications

Although in both groups the notifiers used the pronouns he or she or mentioned their relation to the deceased (my husband, my mother, etc), the deceased’s first name was mentioned only in direct notifications. Alongside the breath pattern description, both groups used a wide range of symptoms (“her hand turned blue, “…his leg is cold…, “Foam is coming from his mouth…”), but only in the case of indirect notifications, there was confusion between death and sleeping (“She is sleeping and we can’t wake her up,” “…he sleeps very deep…”). The direct notifications were characterized with specific requests from the staff members, while in indirect notifications, there were more general requests for nonspecific help. Another characteristic of indirect notifications is related to expressed doubt and/or uncertainty regarding the circumstances. The affect in the direct notifications was usually calm and accepting, while in indirect notifications was more confused or distressed.

Discussion

Guidance papers on how to deliver news of a patient’s death in a skillful manner have been published to educate medical personnel, including paramedics, general and palliative care practitioners, medical students and interns, pediatricians, emergency medicine staff, as well as policemen, victim advocates, chaplains, and social workers. ^{1,2,9,10,13,14}

In the present mixed quantitative and qualitative study, we explored the situation in which the patient’s next of kin notified the treating medical staff, in a home-based palliative care unit, of a patient’s death. The results of this study show that the gender of the notifier is distributed as in the general population. The notifiers were usually children or spouses of the deceased. Although spouses were more likely to act as primary caregivers, their children and other family members were more likely to deliver the notification of death when the patient died at home. This finding may indicate that children and other family members are more involved in the final hours or days before patients die at home from terminal illness but are less involved as caregivers during the earlier phase of the terminal illness. Another explanation can be the feeling of grief that renders the primary caregiver unable to make the call.

The notifications of death, which were direct in 42% of the cases, were more likely to be direct when delivered over the late hours of the night. An explanation for this finding might be that people are more tired at that time so they tend to communicate in a short and straightforward manner.

The patient was actually dead in only 90% of the cases when the medical staff reached the home after the death notification. This situation of near death is difficult to deal with emotionally for relatives, so it is possible that a layman can misinterpret the patient’s condition.

Only 20% of the notifications contained emotional expressions including pain, suffering, and longing, and at the other extreme closure, acceptance, and peace of mind. This finding can be explained by the fact that 20% of the notifiers were not close relatives (more distant family members, foreign workers). In some cases, in Israel, sick people who require nursing care and close supervision are cared for by foreign workers who are brought to Israel specifically for this purpose. They work under the supervision of the patient’s family. Furthermore, most of the patients had a terminal disease and their death was expected and inevitable. Support for this explanation can be found in the results of a survey of professionals about their past experience and training in delivering death notifications, which was conducted between 1995 and 1998 in geographically diverse regions of the United States. The participants were policemen, advocates, chaplains, social workers, psychologists, among others. They were asked to identify the circumstances of death that would be the most emotionally demanding for them to provide death notifications. Terminal illness and chronic condition posed the least emotionally demanding situation in comparison to other cases of deaths such as the death of a child, violent crime, drunken driving, accidents, suicide, a sudden-onset medical problem, or short-term illness or disease. ³

In another study, urgent care physicians were interviewed about communication with families following the death of a loved one. They concluded that communication with the family of a child was more difficult than communication with the family of an adult. ¹⁰

In 30% of the cases, a degree of uncertainty on the patient’s current condition was expressed in the notification of death. There was a greater degree of uncertainty regarding death by next of kin since they were not medical professionals. Despite this uncertainty, in 90% of the cases, the patient was pronounced dead when a medical professional arrived at the scene.

The emotions and uncertainty expressed in the death notifications highlight the need for empathy, patience, and reassurance on the part of the medical staff member who receives the telephone notification of death. The following recommendations for handling death notifications by telephone to the next of kin have been published in the medical literature:

Inquire whether the notifying person is alone and what his location is.

Other recommendations on death notifications include:

Speak slowly and in single sentences.

In 37% of the cases, the notifier mentioned that the patient had stopped breathing. This sign is easily noticed by laymen and is a cardinal sign of death, which explains its broad use.

A specific innovation of the qualitative analysis evolves from the distinction between direct to indirect notification and the characteristics of each. Awareness of the differences enables the medical staff members to respond respectively to each accurately. In both types of notification, the medical staff member should respond in an empathic and supportive attitude, but in an indirect notification, it seems that the medical staff member should be more patient and ask more questions to clarify the situation and promise his or her early arrival to calm the patient and family.

A limitation of this study is that only about 60% of the patients treated by the home palliative care unit and who died at home during the study period were included in it. Another limitation is that the telephone calls were not audio-recorded, so an interpretation bias could occur. Interpersonal differences between the notifiers could also be a source of bias.

Conclusion

This descriptive qualitative and quantitative survey is unique since there are no similar studies of death notifications delivered to the medical staff by next of kin. The results afford a better understanding of the process of death notifications by a next of kin to the medical staff in cases where death is expected and occurs at home.