 |
|
A Questionnaire Survey of Family Practice Physicians' Perceptions of Bereavement Care
Jeanne P. Lemkau, PhD;
Barbara Mann, PhD;
David Little, MD;
Philip Whitecar, MD;
Paul Hershberger, PhD;
Jeremiah A. Schumm, BS
Arch Fam Med. 2000;9:822-829.
ABSTRACT
Survey responses were obtained from 113 family physicians in an exploratory investigation of bereavement care by family physicians. Respondents generally indicated that they believed that bereavement presented significant health risks to their patients and that the identification and treatment of bereaved patients was an important part of their role. However, physicians were highly variable in how they reported identifying and responding to bereaved patients in terms of counseling, addressing spiritual concerns, and medically treating symptoms. The "grief-responsive" physician is described based on these data. We document the interest and need for training in bereavement care.
INTRODUCTION
Bereavement is a common major stress in the lives of patients and has been shown to have a negative impact on health for many individuals.1 As initially demonstrated in the classic "broken-heart" study of Parkes and colleagues,2 heart disease takes an especially heavy toll in the early months of bereavement. A recent study of 1.5 million adults, aged 35 to 84 years, demonstrated a 20% to 35% excess mortality from ischemic heart disease within 5 years of the death of a spouse.3 Bereavement has also been associated with the development of high blood pressure,4 increases in high-risk health behaviors,4-5 and a variety of depressive and anxiety disorders, including posttraumatic stress disorder.6-7
The adverse and long-term effects of bereavement are especially evident in parents after the death of a child,7-8 and in individuals who have experienced the sudden or traumatic death of a loved one.9-10 For example, a major study of the long-term effects of sudden and unexpected loss found that 4 to 7 years after bereavement, bereaved parents and spouses of innocent victims of automobile accidents had higher mortality and psychiatric morbidity than nonbereaved controls.10 Death need not, however, be either sudden or unexpected to require significant time to grieve. Studies of the elderly have demonstrated that, although significant resolution of distress occurs within the first year of bereavement, high levels of grief symptoms are common 2 to 4 years after the death of a spouse.11-12 Review of empirical literature on grief resolution suggest that full resolution may never be achieved, that multiple losses complicate bereavement recovery, and that "a substantial minority of individuals exhibit distress for a much longer period of time than would commonly be assumed."13(p353)
If interventions to mitigate the suffering and negative consequences of bereavement were of no avail, the role of the physician in the bereavement process would be of little significance. However, empirical studies and case reports suggest that both psychologic and psychopharmacologic interventions can be of significant benefit.14-17 Although controlled trials of interventions to improve health outcomes and diminish suffering among the bereaved have been limited in number and are rife with methodological flaws,14 they lend support to the probable benefit of some forms of bereavement care. This lends importance to the question of the appropriate role for primary care physicians in providing such care.
Primary care physicians have been identified as important providers of grief counseling and of referral for complicated grief reactions.18-20 By virtue of their longitudinal contact with patients and their families, family practice physicians are in a good position to identify bereaved individuals, monitor negative sequelae, and support patients through the grieving process. Depending on the physician's training, practice style, and time constraints, such "following" may best be accomplished either through referral or in coordination with a treatment team that includes mental health professionals and clergy. Attention to patient bereavement is a prime example of the practical application of the biopsychosocial model that is at the core of the specialty of family medicine.
The Institute of Medicine Committee on Health Consequences of the Stress of Bereavement recommends that physicians engage in specific approaches in identifying and treating bereaved patients.19 These include reviewing details of the illness and death with the bereaved, fostering realistic expectations about the course of bereavement, and providing reassurance and support for the normal grieving process, identifying individuals at risk for complications, monitoring progress, and referring bereaved patients to appropriate professionals and community resources.
Given the negative health effects of bereavement, the potential benefits of supportive intervention, the recommendations that physicians become more involved in bereavement care, and the fact that physicians often see bereaved patients, it is remarkable how little is known about what physicians actually think or do in the area of bereavement care. To our knowledge, no empirical work has been published on bereavement care provided by family practice physicians in the United States. However, 2 studies conducted in Great Britain bear on the topic. One found that practices that kept death registers and were interested in palliative care were more likely to routinely reach out to bereaved relatives.20 Another study, based on qualitative interviews, found that many physicians felt guilty and conflicted about their role with bereaved patients and were in need of support and learning to treat bereavement.21 Unfortunately, it is not clear how applicable these studies are to understanding physician behavior in other health systems. A descriptive and exploratory study of primary care physicians in the United States is both appropriate and overdue.
Our study used survey research methods to describe and explore how family practice physicians view their roles with bereaved patients, how they identify and treat bereaved patients, and what they believe about the effects of bereavement. We explored physician beliefs, attitudes, and behaviors in relation to demographic, training, practice, and personal characteristics.
MATERIALS AND METHODS
DEVELOPMENT OF QUESTIONNAIRE
A questionnaire, entitled "Family Physicians and Grieving Patients," was specifically designed for this study. It was developed and revised on the basis of multiple discussions among us and pilot-tested with faculty in our academic department not directly involved in the study. Considerations of comprehensiveness were balanced by concerns for brevity and response rate. The final survey consisted of 35 close-ended items and could be completed in 10 to 15 minutes.
The survey covered demographic, background, attitudinal, behavioral, and knowledge variables. Response formats included 5-point Likert scales and multiple-choice answers. Likert scales were used to assess how important each respondent considered the physician's role to be in identifying and treating grieving patients, and the extent to which they believed grief contributed to health problems. Similarly, respondents rated their level of satisfaction working with grieving patients and how comfortable they were interacting with a crying patient. They were also asked to estimate how interested bereaved patients were in discussing their grief with their family physician and how helpful they thought it was for patients to do so. Finally, physicians were asked to indicate how helpful they found it to talk to others about losses they had experienced.
Multiple choice formats were used for the remaining items, with either forced choice or multiple-endorsement options as appropriate. Questions tapped such areas as demographic variables, practice characteristics, degree of respondent involvement in bereavement care, circumstances under which physicians inquired about bereavement, and factors they considered in deciding how to respond to bereaved patients. Respondents were asked to check barriers they perceived as preventing optimal treatment of grieving patients and items that accurately described their behavior with such patients. Questions were asked about how they approached the spiritual concerns of patients and about their prescription practices with the recently bereaved. Several items queried sources and quality of training in bereavement care and interest in continuing medical education (CME) on this topic. Two questions asked about the personal experiences of respondents with bereavement and with sudden and traumatic loss. Finally, respondents were asked to estimate time typically required "to return to their previous state of social and psychological functioning" for 2 scenarios: a parent dealing with sudden and traumatic death of a child and an older adult dealing with death of a seriously ill spouse. A copy of the questionnaire is available from us.
SAMPLE AND PROCEDURES
The questionnaire was sent to 400 family physicians randomly selected from a mailing list provided by the Ohio Academy of Family Physicians, consisting of members active in clinical practice. The initial mailing took place in June 1998. Each questionnaire packet included a cover letter explaining the survey and its voluntary and confidential nature, a prestamped return envelope, and a token incentive (a bumper sticker from the Ohio Academy of Family Physicians). Respondents were asked to separately return a prestamped postcard that was included with the mailing to confirm that they had completed and returned the questionnaire.
Within 1 month of the initial mailing, attempts were made to contact nonrespondents to confirm their identity and location. All nonrespondents whose location could be verified were sent a second survey packet 3 to 5 weeks after the initial mailing.
DATA ANALYSIS
Descriptive statistics were calculated for all questions. The  2 tests were used to compare the sample to demographic statistics provided by the Ohio Academy of Family Physicians for all active members at the time of the survey. The 5 questions concerning physician beliefs and attitudes toward bereaved patients were factor analyzed and reduced to a single factor. We then used this grief-responsiveness measure and multiple logistic regression to study physician characteristics associated with grief responsiveness. Multiple regression was used to relate this factor to demographic, practice characteristics, and other potential predictor variables. Spearman correlation was used to examine relationships between the factor and other dimensions of physician responses.
RESULTS
DESCRIPTION OF SAMPLE
One hundred thirteen questionnaires were returned after 2 mailings. When attempts were made to verify addresses of nonrespondents by telephone, 88 of the original 400 physicians were identified as having relocated, retired, or died. For many nonrespondents, we were unable to either verify the address or reach the physician by telephone. Deleting from the original list only those whom we were able to clearly identify as having relocated, retired, or died, the response rate was 36% (113/312). To examine possible response bias, this sample of respondents was compared with the Ohio Academy of Family Physicians database as a whole on the available characteristics of sex, age, board certification status, residency completion status, and US medical school graduation status. The representativeness of our sample was supported by the finding that there were no significant differences on any of these variables.
Of those physicians who responded, 73.5% were men. Median age was 43 years, with a range of 29 to 80 years. Most of the physicians (92.9%) had medical degrees; the rest were osteopathic physicians. Most had graduated from a medical school in the United States (92.0%), completed a residency in family medicine (83.2%), and achieved board certification in family medicine (92.0%). The median number of years in practice was 15, with a range of 3 to 54.
Most respondents (85.8%) reported that their main practice setting was an outpatient clinic. More than half of the sample (55.4%) was in, or very close to, a city of 25,000 or more people. Of the remainder, 26.8% practiced in a large town or small city (population, 5000-25,000) and 17.9% were in a rural area or village of fewer than 5000 people. A substantial minority (40.7%) held faculty appointments in a medical school and/or a family practice residency.
KNOWLEDGE AND BELIEFS
Five questions were used to quantify the physician's beliefs about grief and grieving patients. The responses to these questions are summarized in Table 1. The physicians in our sample generally reported strong beliefs that grief contributes significantly to health problems and that the primary care physician can play an important role in identifying and treating grieving patients. They also reported that their bereaved patients were interested in discussing their grief and that these patients were helped by these discussions. Clearly, these family physicians see themselves as having an important role with bereaved patients; one that is integrally related to their patients' overall health status. They also report finding it at least moderately satisfying to work with grieving patients.
|
|
|
|
Table 1. Physician Beliefs and Attitudes*
|
|
|
Two questions addressed the physicians' knowledge of recovery time for bereaved individuals. Most respondents (58.1%) estimated that recovery after the sudden death of a child would be complete in 2 years or less. Even more (75.9%) estimated that recovery after the death of a spouse would take 2 years or less.
IDENTIFICATION AND TREATMENT OF BEREAVED PATIENTS
Physicians were asked about the situations in which they routinely inquired about deaths within a patient's family or social network (Table 2). Some physicians reported doing so during initial visits, annual physical examinations, or routine follow-up appointments, but the most relied on outside information about the bereavement or medical findings to prompt inquiry.
|
|
|
|
Table 2. Identification of Bereaved Patients
|
|
|
These physicians reported considering many factors in deciding how to respond to a bereaved patient. The most important considerations fell into 3 groups: the patient's context (relationship to deceased and patient support system), the nature of the death (sudden death or suicide), and the physician's relationship with the patient and/or the deceased.
Most respondents (83.2%) reported that they evaluated most grieving patients themselves and referred only complicated cases to other professionals. A small group (8.8%) said that they generally refer grieving patients to mental health professionals and/or clergy. Few reported that they either evaluated all grieving patients themselves (4.4%) or had minimal involvement in identifying or evaluating grieving patients (3.5%).
When asked about their own behavior toward grieving patients, the physicians reported a wide variety of responses (Table 3). Most reported that they expressed their condolences, and high percentages reported other supportive responses. A majority reported some response to patients on a religious or spiritual level.
|
|
|
|
Table 3. Psychosocial Issues in the Treatment of Grieving Patients
|
|
|
Most respondents reported prescribing psychotropic medications for grieving patients under at least some circumstances (Table 4). The most common indicators for such medication were sleep difficulties and clinical depression. Physicians typically recommended full doses of antidepressants. Benzodiazepines were prescribed for sedation or anxiety treatment by a significant minority of physicians; a few reported prescribing antihistimines or low-dose antidepressants for sedation only.
|
|
|
|
Table 4. Prescribing Behavior
|
|
|
Physicians were asked to check what they perceived to be important barriers to the treatment of grief in their patients (Table 5). By far, the most frequently endorsed barrier reported was physician lack of time.
|
|
|
|
Table 5. Barriers and Learning Experiences in Treatment of Grieving Patients
|
|
|
SOURCES OF TRAINING
Respondents reported having learned about grief and bereavement through many sources (Table 5). Much of the learning was experiential, including experiences with patients and within the physician's own family or social network. Other sources of education included residency training and personal reading. Although most respondents reported that their professional training was good or excellent in this area, most also expressed interest in CME on grief identification and treatment.
GRIEF RESPONSIVENESS
When the questions listed in Table 1 were factor-analyzed, a single factor emerged that accounted for 52.7% of the total variance of these questions. A summary of the factor weights is included in Table 1. Because the factor weights were fairly similar, we added the responses to these questions to obtain an overall measure of grief responsiveness. Respondents who scored high on this scale were more likely to believe that bereaved patients were interested in discussing their grief and that such discussions were helpful to their patients. They were more likely to report high satisfaction from working with grieving patients and to believe that the role of primary care physicians in identifying and treating these patients is very important. They were also more likely to believe that grief significantly contributes to health problems.
The results of the multiple regression relating this factor to physician characteristics (personal and practice) are summarized in Table 6. Physicians who were more comfortable interacting with crying patients and those who personally found it helpful to talk about their losses were significantly more likely to score high on the grief-responsiveness dimension. Physicians whose patients routinely see the same physician also tended to score higher on grief responsiveness.
|
|
|
Table 6. Predictor Variables of Grief Responsiveness With  Coefficients
|
|
|
On the other hand, personal experience with the death of loved ones was not related significantly to grief responsiveness, nor were hours of patient care worked per week and patient volume. Finally, neither demographic characteristics, years of practice experience, faculty status, or perceived quality of grief training were significant predictors of grief responsiveness.
Table 7 presents Spearman correlations between grief responsiveness and several other dimensions of physician responses measured on ordinal scales. A significant relationship was found between grief responsiveness and the degree of active involvement of the physician in the spiritual concerns of patients. Grief responsiveness was not related to the degree to which physicians reported evaluating grieving patients, the desire for CME on bereavement, or the reported number of conditions under which the respondent usually prescribed psychotropic medications to such patients.
|
|
|
|
Table 7. Spearman Correlations Between Grief-Responsiveness and Physician Behaviors
|
|
|
COMMENT
Our survey results indicate that family practice physicians strongly believe that the identification and treatment of grieving patients is an important element of their clinical responsibility. This can be attributed, in part, to their recognition of the fact that bereavement represents an important risk to the health and well-being of their patients. The family physicians responding to our survey reported that they inquire about deaths in their patients' family or social network in a variety of clinical situations. Many physicians reported taking an active role in treating bereaved patients by providing counseling, inquiring about spiritual needs and resources, and prescribing medications when necessary.
While there was general agreement about the importance of grief recognition and treatment in the family practice setting among our physician respondents, there was much variability reported in how grieving patients are treated. Many factors evidently affect the clinical decision-making process. Patient characteristics, features of the death itself, practice characteristics, and physician lack of time influence the degree to which physicians involve themselves with grieving patients. One of our strongest findings was that the physician's comfort level with crying patients seems to be an important factor in grief responsiveness. An increased emphasis on continuity of care, and a higher degree of involvement with the spiritual concerns of the patients, were also associated with grief-responsiveness.
Although respondents were positive in their assessment of their training related to bereavement, they nevertheless expressed a high level of interest in CME on the identification and treatment of grieving patients. The need for further education on the topic of bereavement is also evident in the noteworthy underestimation of the period the respondents believe to be required for grieving the death of a spouse or the sudden and traumatic death of a child.
Our findings support the importance of addressing both cognitive and affective issues in educating physicians in bereavement care. On the cognitive front, residency training, CME presentations, and written materials could be designed to include information on differentiating normal and complicated grief, realistic time frames for grief resolution, and treatment and referral strategies for grieving and bereaved patients. Given the time constraints reported by our physician respondents, education should emphasize treatment options appropriate for the primary care setting, the usefulness of interdisciplinary teams, and indications for referral. Special attention may need to be drawn to cases where loss is sudden, violent, and unexpected, since these may require treatment for trauma as well as support for the grieving process.22
On the affective front, physicians need to be supported to examine their styles of dealing with loss and the implications of their personal experience on their treatment of bereaved patients. Our findings regarding grief responsiveness suggest that attention should be given to enhancing physician comfort with crying patients and helping physicians differentiate between appropriate and inappropriate self-disclosure. Mentorship and modeling by experienced family physicians, precepting by mental health professionals, and Balint group methods may all be helpful in addressing the goals of affective education. Finally, we need to recognize the value of supporting medical students, residents, and colleagues through the grieving process, knowing that most of us remember and pass on sensitive treatment that we have received.
This study has several limitations related to the restricted sample, the low return rate, and the self-report nature of survey data. Studies of physician beliefs and practices in other parts of the country and in other primary care disciplines will be important to establish or refute the generalizability of our findings. Our low return rate also raises questions about external validity. In general, mail surveys of physician respondents tend to have a slightly higher return rate (average of 54%) in contrast to ours (38%).23 The fact that the survey was conducted during the summer, when many physicians take vacations, probably decreased our response rate. Although response bias cannot be totally ruled out, the fact that respondents did not differ from nonrespondents in sex, age, or training is reassuring. Finally, our study suffers from the biases inherent in self-report data. Physicians, like other people, tend to overreport desirable behaviors on survey instruments relative to behavioral measures of the same behaviors.24-25
A fuller understanding of the role of family physicians in bereavement care will require the application of multiple methods, including qualitative studies of both physician and patient experience. Further research is needed to more fully describe physician behavior, and to understand the relationship between physician interventions and patient expectations, satisfaction, and health outcomes.
CONCLUSIONS
Family physicians play an active role in identifying and treating bereaved patients, although practice patterns seem to be highly variable. Meaningful differences exist among physicians with respect to their knowledge about bereavement and their grief responsiveness. Many physicians expressed an interest in CME to improve the treatment of the bereaved patient. Further studies that relate physician behavior to patient outcomes are needed to improve our understanding of the ideal role of the family physician working with bereaved individuals.
AUTHOR INFORMATION
Accepted for publication May 18, 2000.
Supported by a grant from the Ohio Academy of Family Physicians Foundation and a summer research fellowship, Wright State University School of Medicine, Dayton, Ohio (Mr Schumm).
We thank Gretchen Zimmerman, PsyD, for her contributions to the development of this project.
Corresponding author: Jeanne P. Lemkau, PhD, Department of Family Medicine, Wright State University School of Medicine, 140 E Monument Ave, Dayton, OH 45402.
From the Department of Family Medicine, Wright State University School of Medicine (Drs Lemkau, Mann, Little, and Whitecar) and Primary Care Line, Dayton Veterans Affairs Administration Medical Center (Dr Hershberger), Dayton, Ohio; and the Department of Psychology, Kent State University, Kent, Ohio (Mr Schumm).
REFERENCES
| |
1. Rogers MP, Reich P. On the health consequences of bereavement. N Engl J Med. 1988;319:510-511.
ISI
| PUBMED
2. Parkes CM, Benjamin B, Fitzgerald RG. Broken heart: a statistical study of increased mortality among widowers. BMJ. 1969;1:740-743.
3. Martikainen P, Valkonen T. Mortality after the death of a spouse: rates and causes of death in a large Finnish cohort. Am J Public Health. 1996;86(8, pt 1):1087-1093.
4. Prigerson HG, Bierhals AJ, Kasl SV, et al. Traumatic grief as a risk factor for mental and physical morbidity. Am J Psychiatry. 1997;154:616-623.
ABSTRACT
5. Woof WR, Carter YH. The grieving adult and the grieving general practitioner: a literature review in two parts (part 1). Br J Gen Pract. 1997;47:443-448.
ISI
| PUBMED
6. Chen JH, Bierhals AJ, Prigerson HG, Kasl SV, Mazure CM, Jacobs S. Gender differences in the effects of bereavement-related psychological distress in health outcomes. Psychol Med. 1999;29:367-380.
FULL TEXT
|
ISI
| PUBMED
7. De Vries B, Davis CG, Wortman CB, Lehman DR. Long-term psychological and somatic consequences of later life parental bereavement. OMEGA J Death Dying. 1997;35:97-117.
8. Birenbaum LK, Stewart BJ, Phillips DS. Health status of bereaved parents. Nurs Res. 1996;45:105-109.
PUBMED
9. Rynearson EK, McCreery JM. Bereavement after homicide: a synergism of trauma and loss. Am J Psychiatry. 1993;150:258-261.
FREE FULL TEXT
10. Lehman DR, Wortman CB, Williams AF. Long-term effects of losing a spouse or child in a motor vehicle crash. J Pers Soc Psychol. 1987;52:218-231.
FULL TEXT
|
ISI
| PUBMED
11. Zisook S, Schuchter SR. Time course of spousal bereavement. Gen Hosp Psychiatry. 1985;7:95-100.
FULL TEXT
| PUBMED
12. Thompson LW, Gallagher-Thompson D, Futterman A, Gilewski MJ, Peterson J. The effects of late-life spousal bereavement over a 30-month interval. Psychol Aging. 1991;6:434-441.
FULL TEXT
|
ISI
| PUBMED
13. Wortman CB, Silver RC. The myths of coping with loss. J Consult Clin Psychol. 1989;57:349-357.
FULL TEXT
|
ISI
| PUBMED
14. Kato PM, Mann T. A synthesis of psychological interventions for the bereaved. Clin Psychol Rev. 1999;19:275-296.
FULL TEXT
| PUBMED
15. Parkes CM. Research in thanatology: a critical appraisal. OMEGA J Death Dying. 1987;18:365-377.
16. Potocky M. Effective services for bereaved spouses: a content analysis of the empirical literature. Health Soc Work. 1993;18:288-301.
PUBMED
17. Zygmont M, Prigerson HG, Houck PR, et al. A post hoc comparison of paroxetine and nortriptyline for symptoms of traumatic grief. J Clin Psychiatry. 1998;59:241-245.
ISI
| PUBMED
18. Charlton R. Bereavement: a protocol for primary care. Br J Gen Pract. 1995;45:427-430.
ISI
| PUBMED
19. Osterweis M, ed, Solomon F, ed, Green M, ed. Bereavement: Reactions, Consequences, and Care. Washington, DC: National Academy Press; 1984.
20. Harris T, Kendrick T. Bereavement care in general practice: a survey in South Thames Heath Region. Br J Gen Pract. 1998;48:1560-1564.
PUBMED
21. Saunderson EM, Ridsdale L. General practitioners' beliefs and attitudes about how to respond to death and bereavement: qualitative study. BMJ. 1999;319:293-296.
FREE FULL TEXT
22. Rando TA. Treatment of Complicated Mourning. Champaign, Ill: Research Press; 1993.
23. Asch DA, Jedrziewski MK, Christakis NA. Response rates to mail surveys published in medical literature. J Clin Epidemiol. 1997;50:1129-1136.
FULL TEXT
|
ISI
| PUBMED
24. Leaf DA, Neighbor WE, Schaad D, Scott CS. A comparison of self-report and chart audit in studying resident physician assessment of cardiac risk factors. J Gen Intern Med. 1995;10:194-198.
ISI
| PUBMED
25. Montano DE, Phillips WR. Cancer screening by primary care physicians: a comparison of rates obtained from physician self-report, patient survey, and chart audit. Am J Public Health. 1995;85:795-800.
FREE FULL TEXT
RELATED ARTICLES
Good Grief . . .
Cynthia G. Olsen
Arch Fam Med. 2000;9(9):833-834.
FULL TEXT
The Archives of Family Medicine Continuing Medical Education Program
Arch Fam Med. 2000;9(9):887-891.
FULL TEXT
THIS ARTICLE HAS BEEN CITED BY OTHER ARTICLES
Bereavement Practices of Physicians in Oncology and Palliative Care
Chau et al.
Arch Intern Med 2009;169:963-971.
ABSTRACT
| FULL TEXT
A Broken Heart--The Physician's Role: Bereaved Parents' Perceptions of Interactions with Physicians
Bright et al.
CLIN PEDIATR 2009;48:376-382.
ABSTRACT
Reactions and Needs of Tristate-Area Pediatricians After the Events of September 11th: Implications for Children's Mental Health Services
Laraque et al.
Pediatrics 2004;113:1357-1366.
ABSTRACT
| FULL TEXT
How well do general practitioners deliver palliative care? A systematic review
Mitchell
Palliat Med 2002;16:457-464.
ABSTRACT
Good Grief . . .
Olsen
Arch Fam Med 2000;9:833-834.
FULL TEXT
|
