It is estimated that 765,000 suicide attempts occur in the United States each year, and there may be 5 million Americans alive today who have attempted suicide.1 In 2003, general medical emergency departments (EDs) treated 411,128 people following suicide attempts.2 Of those who attempt suicide, up to 20% will attempt again, and 1%-2% will die by suicide within 12 months.3 As the eleventh leading cause of death in the United States,1 suicide is clearly a significant public health issue.
Although suicide targets the individual, for every person who attempts or dies by suicide, a much wider circle of family and friends is affected. A growing body of literature has noted that, while bereavement after suicide is not necessarily more severe than other types of bereavement,4,5 certain aspects or features may make it much more difficult, such as length of bereavement, prevalence of anger about the death, and questioning why the death occurred.6 Those bereaved by suicide also seem to experience increased rates of complicated grief,6 a syndrome that shares features with both depression and posttraumatic stress disorder, including intrusive symptoms of yearning for the deceased as well as persistent symptoms of trauma as a result of the death.7 The ED is often the place where consumers first seek help, providing a window of opportunity for medical personnel to assist consumers and families to access the help and services they need. However, little is known about consumers' and family members' experiences in the ED.
In a series of forums and surveys involving 59 consumers who had experienced a psychiatric emergency, Allen et al.8 found that the majority of consumers reported specific negative experiences, such as feeling staff did not treat them with respect, see them in a timely manner, or listen to them or their version of events. More than half also felt that staff members were not equipped to deal with specific needs related to their cultural, racial, or religious backgrounds. Despite these negative experiences, 59% of respondents reported acts of kindness by the staff. Because this study examined a convenience sample of consumers, only a quarter of whom had been seen for suicidal behavior, it may not be representative of experiences that follow suicide attempts.
No research has been published describing family members' or friends' presence or usual role in the ED following suicide attempts. However, family and friends can play very useful roles in supporting the consumer, monitoring for signs of problems after discharge, and encouraging the consumer to seek follow-up care. A better understanding of what family members and friends experience in the ED following a suicide attempt, as well as their perceived needs, can suggest ways in which the satisfaction of family members and friends might be improved. This change, in turn, may lead to improved clinical outcomes, for example, by yielding more complete and accurate clinical assessments and by increasing compliance with follow-up care.
We conducted a web-based survey to specifically explore the experiences of psychiatric consumers and family members in the ED following a suicide attempt. Surveys were designed and made available on the website of the National Alliance on Mental Illness (NAMI), a grassroots organization that provides self-help, support, and advocacy for consumers and families of people with mental illnesses. NAMI's website (www.nami.org) receives over 10,000 hits each month, 80% from unique Internet Protocol (IP) addresses.
Description of Study Procedures
Two surveys were made available on NAMI's website from June 20, 2004 until August 20, 2004. Separate surveys were created for adult consumers who had made a suicide attempt and been to the ED and for family members and friends of people who had been to the ED due to suicidal behavior. These surveys were designed with the input of National Consumer and Family Advisory panels convened by NAMI for this project and were advertised through NAMI's listserve network, which includes several thousand consumers and family members. The questions were based in part on those developed by Allen et al. concerning emergency experiences.8 However, answers were limited to yes or no to simplify the process for the Internet. The survey contained 49 questions, which asked about demographics, the suicide attempt itself (e.g., method, arrival at ED), specific treatments received and if those services were desirable to the consumer or family member, attitudes and potential problems encountered in the ED, and one open-ended question that allowed respondents to elaborate on experiences.
To assure that responses could not be traced to individuals, all identifiers were removed from the data set by NAMI staff before it was sent to the investigators for analysis. The University of Rochester Institutional Review Board (IRB) determined that this study was exempt from a requirement for informed consent.
Statistical analysis software (SAS) was used to examine frequency of survey responses and compare differences between groups. Open-ended responses were categorized into themes by the first author and a research assistant using an iterative process and consensus meetings to resolve any differences that were identified between these two individuals' lists of themes.
Throughout this article, the term "consumer" is used to refer to the patient who presented to the ED with a suicide attempt. The term "family member" is used to refer to those individuals whose loved one or friend, a consumer, was seen in the ED following a suicide attempt (i.e., individuals included as family members were not necessarily family relations). It was not possible to determine if a consumer and his or her family member both responded.
Surveys were completed by 465 consumers and 254 family members. As the web-based survey allowed participants to stop at any time and demographic questions were included at the end of the study, full demographic information was available only for 355 consumers and 188 family members and the total number of participants who responded varied from item to item, depending on when individuals stopped the survey.
The majority of consumer respondents were female (86%), white, non-Hispanic (88%), and in the range of 25-44 years of age (56%), with at least some college education (79%); 43% were married or in a committed relationship. The majority of family members who responded to the survey were female (84%), white, non-Hispanic (94%), over 45 years of age (63%), had at least some college education (88%), and were in a married/ committed relationship (72%). There was an even geographic distribution of consumers and family members from the west (consumers 21%, family 23%), midwest (29%, 25%), south (27%, 28%), and northeast (23%, 24%) (Table 1).
Suicide Attempt Specific Variables
Consumers. The majority of consumers reported their suicide attempt was by overdose (63%). Consumers' suicide attempts occurred equally in the last year, 1-5 years ago, and over 5 years ago (Table 2).
Slightly fewer than half of the consumers (48%) arrived at the ED by ambulance. while almost a quarter reported a family member brought them to the ED (Table 2). Approximately half of consumers reported that at least one family member was with them in the ED (n = 220), about a third (n = 145) reported they were alone in the ED, while the remainder (n = 71) reported that they were accompanied by a non-family member such as a coworker or friend. Many consumers (n = 63) were accompanied by more than one individual. Of the 220 consumers who were accompanied by family members, most were accompanied by a parent (47%, n = 104) or a spouse or partner (44%, n = 97) while it was less common to be accompanied by an adult child (5%, n = 11) a minor child (5%, n = 12), or another family member (10%, n = 21).
Family members. Almost half of family members (47%) related that the consumer's suicide attempt and ED visit was precipitated by an overdose, the great majority of which had occurred in the last 5 years (84%). Fewer than half of family members reported that an ambulance had taken the consumer to the ED, while a third reported that a family member had been the method of transport (Table 2).
Of the 222 family members who indicated how they learned about the suicide attempt, over a third (34%, n = 76) were with their loved ones at the time of the suicide attempt, 27% (n = 60) were called by another family member or friend, 12% (n = 27) were called by the consumer who asked to be taken to the hospital or to be met there, and 5% (n = 10) were called by ED staff. The remaining 22% (n = 49) found out about the attempt by other means.
Of the 242 family members who identified the person who attempted suicide, over half indicated that the attempt was by either an adult (n = 85, 35%) or minor child (n = 51, 21%). Other individuals accompanied to the ED included a spouse or partner (n = 35, 15%), a parent (n = 15, 6%), a friend (n = 24, 10%), or other (n = 32, 13%). Family members reported experiences involving approximately equal percentages of females (n = 121, 51%) and males (n = 115, 49%). Of the 232 family members who reported the age of the consumer involved, a quarter indicated that the person was under 18 years of age (n = 59, 25%), 31% (n = 72) reported on the suicide attempt of someone 18-24 years of age, 33% (n = 76) reported on someone 25-44 years of age, and the reports of 11% (n = 25) of family members involved someone over 45 years of age.
Eleven percent of respondents (n = 27/235) reported that their family member or friend subsequently died by suicide, either as a result of the attempt reported in this survey or at a later time. Sixty-two percent (n = 145/235) of respondents reported that their family member had attempted suicide more than once.
Forced-Choice Responses About the Experience
Family respondents were more likely than consumer respondents to report that ED staff treated them with respect (73.9% vs. 55.2%; χ2 = 19.89, p < 0.001); listened to them and their version of events (57.1% vs. 41.3%, χ2 = 13.44, p < 0.001); and described the nature of treatments to them (53.8% vs. 40.5%, χ2 = 9.39, p < 0.01). Family members were less likely than consumers to feel that staff made them feel punished or stigmatized due to the suicide attempt (28.8% vs. 54.5%, χ2 = 35.26, p < 0.001) (Figure 1).
There were no significant differences in the percentage of family members versus consumers who felt that the injury was not taken seriously by ED staff (28.6% vs. 31.2%), that they were not seen in a timely fashion (31.4% vs. 24.4%), that ED staff did not address cultural considerations appropriately (28.9% vs. 39%), or that ED staff used jargon or words they did not understand (14.5% vs. 15.3%) (Figure 1)
Over a third of family members (37.6%) reported ED staff did not want to communicate with them about their loved one.
Consumer experiences. The survey contained one open-ended question that asked respondents to "please describe anything that was particularly helpful or hurtful about your ED experience." Of the 465 consumers, 257 provided a total of 552 separate comments: 47 were about positive experiences, 15 were neutral, and 490 were about negative experiences.
Of the 47 positive comments, 43 consumers reported that staff had been particularly helpful. Representative comments included the following:
The nurse who assisted me was one of the most thoughtful and emotionally comforting people I have come across.
The lady who did my assessment was really nice, and comforted me when I cried.
Negative experiences in the ED were reported by 197 separate consumers. The most common category of negative experiences involved a perception of unprofessional behavior by staff, which was involved in 83 comments, for example:
The nurse asked me what kind of mother could I be, doing this to my children? Didn't I know I was committing a sin?
The social worker in the ER treated me like a criminal. She told me how selfish what I'd done was
I was told I was going to hell if I died.
Other common negative responses were categorized as
Feeling the staff did not value them (44 comments):
I was just taking their time away from real patients who needed their help.
The staff don't take time to adequately listen to what went wrong. I feel that their goal is more to determine where the next placement should be.
Feeling lonely or ignored (27 comments):
I felt very lonely and desperate at the time and I needed support.
Feeling the suicide attempt was not taken seriously (25 comments):
They told me I just did it for attention.
I was treated like I had the flu. They did not care I was depressed and did not want to address that.
After I had taken pills, all they did was send me home with a sheet that said I needed to drink plenty of fluids and rest.
Feeling that the waiting time was too long (22 comments):
For two days I lay in the emergency room waiting for a psych bed.
I didn't see anyone [on staff] for 5 hours. I froze on the gurney.
Family experiences. Of the 254 family members who responded to the survey, 139 responded to the openended question. Of the 270 separate comments written by family members, 43 were about positive experiences, 2 were neutral, and 225 were about negative experiences. As was the case for the consumers' responses, the majority of positive comments concerned positive experiences with staff (25 comments) such as:
The ED my son was taken to saved his life. They were very kind to me
Being provided with information about a family member's care was also highly valued (13 comments) such as:
It was very helpful that they gave me truthful and accurate information about his physical condition-it gave me confidence in their treatment and allowed me to make realistic decisions.
We were assigned a social worker within moments of arriving... she kept checking on us approximately every 15 minutes with updates.
Like the consumers, the family members' most frequent negative comments concerned a perception of unprofessional staff behavior (30 comments):
Nurses scolded [my mother] in front of students and other patients-telling her she will end up in hell and asking how she could do that to her family.
Other categories of frequent negative responses included
A perception that the suicide attempt was not taken seriously (26 comments):
They acted like she just had a bad day and had too much to drink.
They sent her home because she lied to them and told them it was an accident... We were so scared every minute for several months that she would do it again.
A feeling that the waiting time was too long (18 comments):
We had to wait hours on end while those with no obvious traumas were treated and released.
We were in the ER for 20 hours before he was transferred to an inpatient facility.
A need for better communication of discharge plans or instructions (15 comments):
It would have been very helpful to receive some written materials/handouts.
I would have liked information on family and friend support groups.
No one in the ER explained what was going to happen, what we should expect, what to do and not do in relation to the admission of a child to a psychiatric unit.
This study represents the first attempt to examine consumers' and family members' experiences in the ED following a consumer's suicide attempt. The majority of consumers and family members felt that staff addressed their ethnic and cultural issues appropriately, saw them in a timely manner, and did not use jargon or words they did not understand. At the same time, fewer than 40% of consumers felt that staff listened to them, their story or their version of events, described the nature of treatments, or took their injury seriously. While family members were more likely than consumers to feel heard or to receive information about treatment, less than two-thirds of family members reported these experiences. More than half of consumers and almost a third of family members felt directly punished or stigmatized by staff.
Evidence suggests that patients discharged following psychiatric emergency services rarely follow up with treatment recommendations and are at high risk for further suicidal behavior.9 The data presented here provide some explanation for why follow-up does not occur-a significant fraction of consumers and family members feel their needs are not being met in the ED following a suicide attempt. Input from consumers and family is critical to continuous quality assurance in emergency services. Educating family members by providing them with resources such as NAMI's free toolkit for family members,* or referring them to family psychoeducation may assist them to better monitor for the presence of suicidal behavior and limit future psychiatric emergencies.
Over half of the consumers in this sample were accompanied or met by family members in the ED following the suicide attempt. Suicide is often described as a solitary act, but this evidence that consumers are not alone following a suicide attempt suggests that family members and friends are often available and can be valuable resources for assessment and follow-up. As one family member wrote:
Family member(s) must be taken seriously. They need to be listened to. They are the ones who live with the individual on a daily basis and they are more aware of what is going on with the individual than any other person.
The U.S. Health Insurance Portability and Accountability Act of 1996 (HIPAA) allows information to be shared with family members when the information is directly relevant to the family member's involvement and specifically states that family members can provide information about the patient even if ED personnel are reluctant to give them information.10
Using NAMI's website for the survey was helpful in recruiting a broad range of individuals who seek out the website for information. As a consequence, however, the opinions may solely reflect NAMI membership and exposure to NAMI's web-based materials. Because NAMI is an organization specifically for families of individuals with mental illness, the sample described here may over-represent consumers whose families are engaged in their care. In addition, individuals who seek out mental health resources and advocacy organizations may do so specifically because of negative experiences or a personal mission to improve care, further biasing the results. Also, the Internet is not as accessible to individuals with less education and those from diverse ethnic and socioeconomic backgrounds,11 potentially limiting the generalizability of these findings.
The race and gender mix of the sample described here did not appear to match those of a sample of patients described by Doshi et al.12 In that study, Doshi et al. analyzed data for all visits for attempted suicide or selfinflicted injury during 1997 to 2001 from the National Hospital Ambulatory Medical Care Survey, a national probability sample of ED visits, and found that rates were slightly higher among female patients (1.7 visits per 1,000 US citizens) than male patients (1.3 visits/1000 citizens) and among blacks (1.9) than whites (1.5).12 However, the sample in the current study was similar in other respects to that of Doshi et al., who found no difference by U.S. region and that 68% of ED visits for suicidal behavior were due to overdose or poisoning. Finally, since 29% of consumer and 16% of family members reported on experiences that occurred more than 5 years earlier, some caution in extrapolating the findings reported here to current practices is warranted.
Because surveys were de-identified, we were unable to compare the perspectives of family members and consumers regarding their visits to the ED. There is no way to know if these two samples overlap, or if the differences between groups are a result of consumers and family members receiving services at entirely different times and places.
The results of this web-based survey indicate that, although both consumers and family members related positive experiences with ED staff members, most respondents reported negative experiences during their time in the ED. These negative experiences involved a perception of unprofessional staff behavior, feeling lonely or ignored, feeling their suicide attempt was not taken seriously, and long wait times. ED staff may not understand the profound impact that their comments can have on consumers or family members. The ED is an excellent place to provide useful interventions because the vast majority of suicide attempters who receive any medical attention are seen in the ED.13 For example, Rotheram-Borus et al. reported that a reduction in subsequent suicide attempts was seen following simple training sessions aimed at increasing rapport between patients and ED staff at all levels, supporting rather than blaming the family, and increasing respect for the patient and family.14 A shortage of resources, as manifested by the perception (and often reality) of long waits in EDs, is another area that needs to be addressed. Further prospective data need to be systematically collected to serve as a basis for developing ED practices that are sensitive to the needs of consumers, involve family members in care, and decrease future suicidal behavior.