 |
 |
|
|
|
 Complicated grief commonly refers to extremes in the intensity and duration of grief symptoms and their clinical complications (1,2). Studies investigating the prevalence of CG among psychiatric outpatients report levels ranging from 20% to 33% (3–5). Despite these high levels, CG is often not detected in initial assessments. This may be due to lack of focus on death losses by both patients and clinicians. Clinicians are expected to inquire about presenting problems, precipitating events, mental status, potential for self-harm, previous disorders and treatments, and current disorders. Little time is available to explore such important events as deaths. Consequently, information relevant to treatment decisions is not obtained. It would be useful to have a few screening questions that possess high sensitivity (that is, identification of patients with CG) and high specificity (that is, identification of patients without CG). Our recent prevalence study (5) provided responses to 2 sets of potential screening items. Table 1 lists 5 items from the TRIG (6) that cover events that clinical reports (7) claim are risk factors or indicators of CG. Affirmative responses to items 3 and 5, and negative answers to items 1, 2, and 4, are thought to indicate increased likelihood of CG. However, supportive research evidence is lacking.
Another set of items came from questionnaires used to determine the presence of CG: the PGQ (8) and the IES (9). Subsequent to our prevalence study, we conducted a factor analysis of items from the 2 questionnaires and the BDI-II (10). Separate grief and depression factors emerged (11), which indicated unique symptoms for CG. For the current study, we chose a few items that were highly loaded on the grief factors. We hypothesized that one or more combinations of the items would emerge as CG identifiers that could be used to screen patients. MethodSetting and Procedure A detailed description of the methodology for the previous prevalence study is found in Piper and others (5). The study received approval from the university and hospital ethical review boards for research involving human subjects. We collected data for 1 year at 2 psychiatric outpatient clinics. Written informed consent was obtained from 235 patients. During intake, the assessor inquired about significant lifetime death losses. The assessor asked the 5 TRIG questions and also asked patients to complete questionnaires that included definitional criteria for CG. Patients The 235 patients received diagnoses according to DSM-IV criteria (12). Most ( 92%) received Axis I diagnoses, the most frequent being major depression (41%) and dysthymia (11%). About 53% received Axis II diagnoses, the most frequent being OCD (16%) and dependent personality disorder (14%). The patients’ mean age was 42.2 years, and 57% were women. About 86% had at least a high school education, and 53% were employed. The patients had experienced an average of 2.9 significant losses, as follows: loss of parent (47%), friend (14%), grandparent (10%), sibling (6%), partner (4%), child (3%), and other (14%). The mean time since the loss was 10.4 years, which indicates long-term but not necessarily continuous CG. Criteria for CG Three questionnaires determined whether the patient met criteria for CG. One was the PGQ (8). The other was the IES (9) with subscales for Intrusion and Avoidance. The patient indicated the frequency with which events related to the losses occurred during the past week. The third questionnaire was the SAS-SR (13), which covers social dysfunction in 6 areas. To meet criteria for CG, the patient had to have a score of 10 or higher on the PGQ, the Intrusion subscale, or the Avoidance subscale for at least one loss, and a score of 2.0 or higher on at least one SAS-SR subscale. The loss must have occurred at least 3 months before the patient answered the questionnaires. ResultsScreening Item Selection In addition to the TRIG items, we selected the 2 items from the factor analysis with the highest loadings from each of 2 grief factors (specifically, Grief Symptoms and Grief Avoidance). We dichotomized the responses to the grief factor items into low frequency (that is, “not at all” or “rarely”) and high frequency (that is, “sometimes” and “often”) categories. TRIG responses were already dichotomized (“yes” or “no”). Relation Between Each Item and CG Table 1 shows the percentages of test-positive and test-negative responses. We investigated the relation between each of these 9 items and the presence or absence of CG with 2-by-2 chi-square analyses. The following 3 TRIG items were significantly related to CG: “Do you feel that you have really grieved?” (response n = 224; c2 = 11.87, df 1; P < 0.001), “Do you feel you are now functioning about as well as you were before the death?” (response n = 226; c2 = 25.27,df 1; P < 0.001), and “Do you get upset each year at about the same time that he/she died?” (response n = 199; c2= 9.43, df 1; P = 0.002). All 4 grief factor items were significantly related to CG: “I tried to remove it from memory” (response n = 230; c2 = 45.98, df 1; P < 0.001), “I tried not to think about it” (response n = 234; c2 = 75.77, df 1; P < 0.001), “I had waves of strong feelings about it” (response n = 232; c2 = 75.19, df 1; P < 0.001), and “Pictures about it popped into my mind” (response n = 234; c2 = 80.87, df 1; P < 0.001). Sensitivity and Specificity Table 2 shows results for significant TRIG items. With regard to total correct percentage, at 69.6%, which is not particularly high, the best performer was the pair of items “not having really grieved” or “not functioning as well.” For combined items such as this, patients had to test positive to any of the items but not to all items. Table 3 shows results for the grief factor items. Because the item “I tried to remove it from memory” had a low total correct percentage (65.2%), it was dropped. On total correct percentage, the best performer was the pair of items “Pictures about it popped into my mind” and “I tried not to think about it.” Sensitivity (88.6%) and specificity (89.2%) were high. Nearly 90% of the patients with and without CG were correctly identified.
The total correct percentage for the best-performing pair of grief factor items was 19% greater than the best-performing pair of TRIG items. This difference was statistically significant (response n = 458; c2 = 24.76, df 1; P < 0.001). DiscussionStudies indicate that CG is highly prevalent in outpatient psychiatry clinics. Because assessment time is limited, many cases probably go undetected. This study attempted to discover a few screening questions that could identify out- patients likely experiencing CG. Items that reflected clinical beliefs about indicators and risk factors associated with CG did not perform well. By contrast, a set of grief factor items performed very well. Two items correctly identified nearly 90% of patients with and without CG. These items can be transformed into questions that can be used in initial assessments. If a patient tests positive, the interviewer can acquire further information about the losses and (or) refer the patient to someone who specializes in the assessment and treatment of patients with CG. Cost-efficient treatments such as short-term group therapy can effectively treat psychiatric outpatients with CG (14,15). Before it can be treated, however, CG needs to be identified. We suspect that, in busy psychiatric outpatient clinics, a few pertinent questions will identify more CG patients than existing requests to conduct more thorough assessments of CG during intake. In other words, less may provide more. Although the current study involved a large sample from 2 outpatient clinics, the findings would benefit from cross-validation. They may not apply to the general population. Different findings might emerge with different criteria for CG. Despite these limitations, we believe that the current study makes a reasonable case for the use of screening questions to identify psychiatric outpatients who are probably experiencing CG. Funding and SupportThis research project was supported by Grant MT-13481 from the Medical Research Council of Canada. References1. Jacobs S, Mazure C, Prigerson H. Diagnostic criteria for traumatic grief. Death Studies 2000;24:185–99. 2. Prigerson HG, Bierhals AJ, Kasl SV, Reynolds CF, Shear MZ, Day N, and others. Traumatic grief as a risk factor for mental and physical morbidity. Am J Psychiatry 1997;154:616–23. 3. Zisook S, Shuchter SR, Schuckit M. Factors in the persistence of unresolved grief among psychiatric outpatients. Psychosomatics 1985;26:497–503. 4. Zisook S, Lyons L. Bereavement and unresolved grief in psychiatric outpatients. Omega 1989/90;20:307–22. 5. Piper WE, Ogrodniczuk JS, Azim HF, Weideman R. Prevalence of losses and levels of complicated grief in psychiatric outpatient clinics. Psychiatr Serv 2001;52:1069–74. 6. Faschingbauer TR, Zisook S, DeVaul RA. The Texas Revised Inventory of Grief. In: Zisook S, editor. Biopsychosocial aspects of bereavement. Washington (DC): American Psychiatric Press; 1987. p 109–23. 7. Rando TA. Treatment of complicated mourning. Champaign (IL): Research Press; 1993. 8. Prigerson HG, Frank E, Kasl SV, Reynolds CF, Anderson B, Zubenko GS, and others. Complicated grief and bereavement-related depression as distinct disorders: preliminary empirical validation in elderly bereaved spouses. Am J Psychiatry 1995;152:22–30. 9. Horowitz MJ, Wilner N, Alvarez W. Impact of Event Scale: a measure of subjective stress. Psychosom Med 1979;41:209–21. 10. Beck AT, Steer RA, Brown GK. Manual for Beck Depression Inventory-II. San Antonio (TX): Psychological Corporation; 1996. 11. Ogrodniczuk JS, Piper WE, Joyce AS, Weideman R, McCallum M, Azim HF and others. Differentiating symptoms of complicated grief and depression among psychiatric patients. Can J Psychiatry 2003;48:87–93. 12. American Psychiatric Association. Diagnostic and statistical manual of mental disorders. 4th ed. Washington (DC): American Psychiatric Association; 1994. 13. Weissman MM, Bothwell S. Assessment of social adjustment by patient self-report. Arch Gen Psychiatry 1976;33:1111–5. 14. Piper WE, McCallum M, Azim HFA. Adaptation to loss through short-term group psychotherapy. New York: Guilford Press; 1992. 15. Piper WE, McCallum M, Joyce AS, Rosie JS, Ogrodniczuk JS. Patient personality and time-limited group psychotherapy for complicated grief. Int J Group Psychother 2001;51:525–52. Author(s)Manuscript received April 2004, revised, and accepted March 2005. 1. Professor, Department of Psychiatry, University of British Columbia, Vancouver, British Columbia. 2. Assistant Professor, Department of Psychiatry, University of British Columbia, Vancouver, British Columbia. 3. Clinical Associate Professor, Department of Psychiatry, University of British Columbia, Vancouver, British Columbia. Address for correspondence: Dr WE Piper, Department of Psychiatry, University of British Columbia, 2255 Wesbrook Mall, Vancouver, BC V6T 2A1 e-mail: piper@interchange.ubc.ca
1 | 2
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||