- Email: [email protected]
Family physicians and the risk of suicide in the depressed elderly
Journal of Affective Disorders 54 (1999) 193–198
Brief report
Family physicians and the risk of suicide in the depressed elderly a, b a a a Gabriela Stoppe *, Hagen Sandholzer , Claudia Huppertz , Hauke Duwe , Juergen Staedt a
b
Department of Psychiatry, Georg-August University, Von-Siebold Str. 5, 37075 Goettingen, Germany Department of General Practice, Hannover Medical School, Carl-Neuberg Str. 1, 30625 Hannover, Germany Received 26 May 1998; received in revised form 25 July 1998; accepted 3 August 1998
Abstract Background: Depression is the most frequent psychiatric disorder in the elderly. It is the reason for most suicides in this age group. Method: We performed a representative survey in primary care. Two written case vignettes were presented to 170 family physicians in face-to-face interviews which took place in their practices. The case vignettes described either (Case 1) a mildly depressed otherwise healthy old patient or a severely depressed patient (Case 2) with somatic comorbidity. Afterwards the interviewers asked standardized open questions. The physicians were not let into the mental health focus of the study. Results: The response rate was 77.6%. Depression was considered for primary or differential diagnosis by 91.2% of the physicians in Case 1 and by 70% in Case 2 (X 2 -test; p , 0.01). For further anamnesis, only 2.4% of the physicians were interested in suicidal ideation of the patient. When directly asked at the end of the interview, 76.9% of the physicians said they would talk about suicide. Those who would not, thought that the patient would communicate suicidal intent himself / herself, or they feared to induce suicide by asking directly. Conclusion: Thinking of suicidality and its prevention is not uppermost in the physicians’ mind. Therefore, and also with regard to the relatively high rate of depression recognition, we conclude that educational means should not only focus on the recognition and screening of depression, but also on the management—‘how to talk about . . . ’—of complex problems like suicide in the elderly, in order to change suicide rates. 1999 Elsevier Science B.V. All rights reserved. Keywords: Suicide; Depression; Old age; Primary care; Family physicians; Cued and uncued answers; Case vignettes
1. Introduction Recent studies in the US revealed increasing suicide rates in the elderly in recent years (Diekstra *Corresponding author. Tel.: 1 49-551-39-95-63; fax: 1 49551-39-66-92. E-mail address: [email protected] (G. Stoppe)
and Gulbinat, 1993; Centers for Disease Control and Prevention, 1996). Psychiatric illness accounts for about 90% of all suicides, with depression making up much more than 50% especially in the elderly (Henriksson et al., 1993, 1995; Conwell et al., 1996). Recent evidence reveals that early recognition and treatment of the underlying psychiatric disorder is the most important method of suicide prevention
0165-0327 / 99 / $ – see front matter 1999 Elsevier Science B.V. All rights reserved. PII: S0165-0327( 98 )00149-9
194
G. Stoppe et al. / Journal of Affective Disorders 54 (1999) 193 – 198
(Isacsson et al., 1996; Rutz et al., 1989, 1992; Rihmer et al., 1995; Ahrens et al., 1995). Different guidelines recommend routine screening and evaluation of suicide risk in high risk groups (American Medical Association, 1994; Canadian Task Force on the Periodic Health Examination, 1994). In the elderly, the recognition and treatment of depression is of major importance because it is a frequent disorder (Beekman et al., 1995) and can be treated as successful as in younger age groups (Katona, 1995; Lebowitz et al., 1997). However, more than for younger ages, it goes under-recognized and undertreated (Lebowitz et al., 1997; NIH Consensus Development Panel on Depression in Late Life, 1992). Since most of the elderly exclusively visit their family physicians, these hold the key position to better recognition and management. Two thirds of all suicides visited their family physician within the month preceding the suicide (Lloyd and Jenkins, 1995; Cattell and Jolley, 1995). However, only one fifth of those who would commit suicide within the following 4 weeks had discussed suicide during their last appointment, according to a Finnish study (Isometsa et al., 1995). In old age, some patient factors make detection and treatment of depression and suicidality even more difficult (NIH Consensus Development Panel on Depression in Late Life, 1992; Gallo et al., 1994; Lyness et al., 1995). In addition different attitudes of the physicians and the community with regard to suicide in old compared to young patients must be taken into account (Duberstein et al., 1995). There is some evidence that gender differences on the part of the patients as well as on the part of the physicians may be influencing this situation. Duberstein et al. (1995) found that female physicians ‘‘hold attitudes toward suicide that are different from male physicians and, perhaps, from other women’’. In the Gotland study the rate of female depressive suicides decreased dramatically after the training program, while the proportion of male depressive suicides was almost unchanged (Rihmer et al., 1995). Finally, Isometsa et al. (1994) found significant sex differences in current and previous treatment of major depressed patients who later died of suicide. With a two-step design the following study reveals information about the willingness of family physi-
cians to elicit a history of suicidal ideation in their elderly depressed patients.
2. Materials and methods
2.1. The case vignettes We designed written sample case histories. The severity of the depression and the gender of the patient were the main factors which were experimentally varied. Case 1 describes a patient free of other diseases with a mild depression. The symptoms have been described to be typical of old age depression (Ernst and Angst, 1995; Brodaty et al., 1991). Case 2 describes a patient with moderate to severe depression with the same social characteristics as the patient in Case 1. The severity is underlined by some typical depressive delusions. The patient is brought to the doctor by his / her daughter and he / she suffers from physical diseases and is treated with medication on a regular basis. The case history is given in two versions, each describing either stroke or thyroid disorder. These have a high rate of depressive comorbidity in common (Brodaty et al., 1991; Starkstein and Robinson, 1989), but differ with regard to the origin (‘primary cerebral’ vs. ‘primary non-cerebral’). For each case different versions with regard to patient’s gender were used: in Case 1 only the gender of patients varied (Case 1a: male; Case 1b: female), in Case 2 both the gender (female / male) and the anamnesis (stroke / hypothyreosis) varied (Case 2a: female 1 stroke; Case 2b: female 1 thyroid; Case 2c: male and stroke; Case 2d: male and thyroid).
2.1.1. Text of Case 1 A 70-year-old man / woman whom you have known for several years presents himself / herself at your clinic. You do not know him / her to be severely ill. He / She takes no regular medication. Following the death of his wife / her husband 5 years ago, he / she has been living alone in a three-bedroom
G. Stoppe et al. / Journal of Affective Disorders 54 (1999) 193 – 198
apartment and caring for the household on his / her own. His / Her only daughter works full-time and lives with her husband and children in the same community. The patient complains of progressive tiredness and a loss of drive. He / She says he / she is hardly able to concentrate and becomes increasingly irritated. After further inquiry he / she says that he / she has less appetite and is worried about his / her health. The patient appears to be in low spirits but he / she seems to cheer up during the consultation.
2.1.2. Case 2 a,c A 70-year-old male / female patient is brought to your clinic by his / her daughter, who lives nearby. ♠He / She has suffered from hypertension for years. One year ago he / she had a stroke. An initial hemiparesis has disappeared with the exception of a slight disability of one arm. ♠He / She is on aspirine, an ACE-antagonist and a diuretic on a regular basis. With a moderate diet, serum levels of glucose and lipids are within normal range. The patient has not shown up at the practice recently and is now brought on his / her daughter’s instigation. The patient himself / herself is sitting downheartedly in his / her chair as the daughter reports that his / her personality and behaviour has considerably changed recently. He / She does nothing on her own, and even has to be persuaded to eat. Therefore he / she has had a considerable weight loss over the preceding 2 weeks. Talking to him / her makes him / her irritable. He / She is convinced that he / she had spent too much on his / her living and that his / her complaints were a punishment for his / her previous lifestyle. As the patient himself / herself is asked he / she says that being incurably ill he / she certainly will die soon. Case 2b,d differ only with regard to the information given between the ♠ signs: Some time ago he / she was diagnosed to have a hypothyroidism, which is treated with oral l-thyroxine 125 mg daily. Since then thyroid function has been normal on controls. Because of arterial hypertension, he / she is on an ACE-antagonist and a diuretic on a regular basis.
195
2.2. Study sample, interview procedure and analysis The study was carried out from July to September 1995. On the basis of previous experience and pilot studies we developed the following design: We obtained a list of all doctors who run a private practice in the city of Kassel (approximately 200 000 inhabitants) and the surrounding regions from the ¨ regional association of physicians (Kassenarztliche Vereinigung). We selected all family physicians (FP), who—in the German health care system—are general practitioners and many internal-medicine specialists. Two trained investigators (C.H. and H.D.) phoned all eligible FP and requested a face-to-face interview after explaining that the purpose of the study would be the management of old patients in primary care. All versions of Case 1 and Case 2 were randomly assigned in advance to the doctors consenting to participate in the study. The interviewers visited their practices and asked the physicians to fill out a short questionnaire concerning statistical data (age, sex, medical education, practice size). Then the interviewers presented a pair of Case 1 and Case 2 vignettes to the physician. Following a short period of perusal, after each vignette, the interviewers asked them standardized questions and categorized the answers to a classification scheme developed in the pilot studies. Thus suggestive effects of given answer categories were avoided. Finally we asked them whether they regarded the case history as somewhat typical for their practice. The whole procedure had a mean duration of 30 minutes. With regard to the management of suicidality we developed a two-step procedure in our questionnaire. The initial two questions were on (differential) diagnostic considerations. The third question— ‘Which anamnestic details would you like to know?’ (multiple answers possible)—was the first, upon which consideration of suicide risk could have been mentioned. As no cues were given at this stage of the interview, we expected that the answers to this question reflected the actual frequency in daily practice. The next questions focused on diagnostics, consultations, treatments and caregiver information. At the end (because we would not induce a bias in favour of the mental health focus), the interviewers
196
G. Stoppe et al. / Journal of Affective Disorders 54 (1999) 193 – 198
asked explicitly whether the physicians would ask the patient for suicidal ideation. Those who answered that they would not were asked for their reasons. Statistical analysis was performed with a computer software (SPSS). We analysed independent research questions by use of the X 2 -test. Results were regarded as significant at a level of p , 0.05.
3. Results In the investigation area 239 physicians work in private practices according to the list of the chamber of physicians. A total of 20 doctors had to be excluded as they did not practise any more (n 5 6) or they were not primary care physicians (11). From the remaining physicians (n 5 219), 77.6% (125 general practitioners and 45 primary care internists) took part in the study. In detail, the group consisted of 131 males and 39 females, with a mean age of 47.04 years, who had been running their practices for 10.7 years. They estimated the percentage of old patients to be 39% on average. Eighty-eight percent of the respondents regarded the case vignettes as typical for their practice. The responder group and the nonresponder group did not differ significantly on the basis of any of the available data. Thus, we met with the criteria for a representative study. Depression was considered for primary or differential diagnosis by 91.2% of the physicians in Case 1 and by 70% in Case 2 (X 2 -test; p , 0.01). About half of the physicians, with regard to both cases, wanted further information regarding, for example lifestyle and gastrointestinal disturbances. However, only 4 (2.4%) physicians considered suicidality in Case 1 and 10 (5.9%) in Case 2. To the direct question 130 (76.9%) physicians in Case 1 and
126 (74.6%) doctors in Case 2 answered that they would ask the patient for suicidal ideation during the initial consultation or later. The answer category ‘later’ was not further specified during the interview. The differences between cued and uncued answers were highly significant (Case 1: X 2 5 126.01, df 5 1, p , 0.01; Case 2: X 2 5 116.01, df 5 1, p , 0.01). There were no significant differences between the answers given to Case 1 compared to Case 2. About 50% of those who would not check for suicidality, argued that ‘there was no reason’ to do so. Other reasons and details are given in Table 1. Again there were no significant differences between the answers given to Case 1 and Case 2. Considering the gender focus of our study, we found no significant differences in the cued and uncued answers nor with regard to the reasons given not to speak about suicidal ideation. This holds true with regard to the patients’ and the physicians’ gender.
4. Discussion The method we used is well known in epidemiological research as a tool to measure physicians’ competence, attitudes and also actual behaviour (Duberstein et al., 1995; Jones et al., 1990; Holt and Mazzucca, 1992; Yoder et al., 1990). We have discussed the advantages and problems of this method previously in publications of an earlier study of our group (Stoppe et al., 1994). An advantage with regard to the questions of this study is that a standardised patient, which was regarded by 88% of the physicians as typical for their practice, could be described in two versions, which only differed with regard to patient’s gender, depression severity or
Table 1 The reasons mentioned by those physicians who would not ask the patient for suicidal thoughts
There is no reason The patient will mention this topic himself / herself Fear to induce suicide All other reasons
Case 1 (n 5 37) (%)
Case 2 (n 5 40) (%)
45.9 27.1 18.9 8.1
50 17.5 12.5 20.0
Data are given as a percentage of the ‘no’-answers (n). There are no significant differences between the answers given for Case 1 and those given for Case 2.
G. Stoppe et al. / Journal of Affective Disorders 54 (1999) 193 – 198
somatic comorbidity. By this means we can see the extent to which these variables account for differences in the answers. Another advantage is that the physicians were not informed about the mental health focus nor about the gender focus of the study. Thus the design fulfilled the demand that decision-making studies should embed the psychosocial questions within a list of other possible issues (Robbins et al., 1994). The face-to-face interview allowed standardised questions without the risk of potential suggestions by given answer categories and allowed clarifying questions if necessary. The technique was—according to our experience—also the reason for the high response rate and the representativity of the study. One might argue that relevant factors like the physicians’ interview skills and behaviour or their sensitivity, for example to non-verbal communications, make no difference in the interview situation. However, their knowledge, their psychological awareness and their interest for psycho-social problems could influence their decisions in this hypothetical situation. In some way the case vignette serves as a ‘matrix’, upon which the condensed experience of the physicians could be projected. This is a major advantage when studying complex topics like gender differences or suicidality. The two-step design to evaluate the physicians’ willingness to consider suicidality in their patients led to the most striking results of this study: Overall 2.4% of the physicians thought of a suicide risk spontaneously and 76.5% when being asked explicitly. The latter value could be influenced by social desirability, whereas the former could reflect the actual behaviour. Taking into account that most physicians considered depression for primary or differential diagnosis in both cases, we can assume that depression was on their mind. The answers to the direct question may reflect the knowledge that suicide is a feature of depression and that one should talk about it. However, the low—compared to other information— number of uncued answers suggests that thinking of suicidality is not uppermost in the physicians’ mind. There may be two reasons for this. First, family physicians may not know how frequent suicide is and how important they are in the prevention of suicide. In Germany, until recently,
197
family physicians did not have an obligatory training in psychiatry or geriatrics in their curriculum although psycho-social problems make up about 30– 50% of their consultations and elderly patients about 40%. Secondly, physicians may be insecure about how to manage suicidality. This is underlined by the result that about 20% of those who would not ask fear to induce suicidality by talking about it. One must also discuss that some fear the consequences if the patient says that she / he suffers from suicidal ideation, although no physician mentioned this in this study. The patient should be referred to hospital or to a primary care neuropsychiatrist, both of which elderly patients will resist (NIH Consensus Development Panel on Depression in Late Life, 1992). The answers with regard to suicidality do not vary with the patient’s or physician’s gender or with the severity of depression. This also indicates that the answers may be explained mostly by knowledge and skill factors on the part of the physicians. With regard to physicians’ gender our results differ from those of other studies investigating the attitudes towards assisted suicide (Duberstein et al., 1995; Cohen et al., 1994). However, we did not focus on attitudes to (assisted) suicide in our study, and this may explain the differences. Educational training in primary care seems to be necessary to reduce suicide rates. This has been proven preliminary effective, although the effect seems to fade after some time and repetition of training is necessary (Rutz et al., 1989, 1992). In agreement with recent studies (Tiemens et al., 1996; Callahan et al., 1996), the results of this study indicate that educational means should not only focus on the recognition and treatment, for example of depression, but also on the management—‘how to talk about . . . ’—of complex problems like suicide.
References ¨ ¨ Ahrens, B., Berghofer, A., Wolf, T., Muller-Oerlinghausen, B., 1995. Suicide attempts, age and duration of illness in recurrent affective disorders. J. Affect. Disord. 36, 43–49. American Medical Association, 1994. Guidelines for Adolescent Preventive Services (GAPS), Recommendations and Rationale, American Medical Association, Chicago, pp. 131–143. Beekman, A.T.F., Deeg, D.J.H., van Tilburg, T., Smit, J.H.,
198
G. Stoppe et al. / Journal of Affective Disorders 54 (1999) 193 – 198
Hooijer, C., van Tilburg, W., 1995. Major and minor depression in later life: A study of prevalence and risk factors. J. Affect. Disord. 36, 65–75. Brodaty, H., Peters, K., Boyce, P., Hickie, I., Parker, G., Mitchell, P., Wilhelm, K., 1991. Age and depression. J. Affect. Disord. 23, 137–149. Callahan, C.M., Dittus, R.S., Tierney, W., 1996. Primary care physicians’ medical decision making for late life depression. J. Gen. Intern. Med. 11, 218–225. Canadian Task Force on the Periodic Health Examination, 1994. Canadian Guide to Clinical Preventive Health Care, Canada Communication Group, Ottawa, Canada, pp. 456–470. Cattell, H., Jolley, D.J., 1995. One hundred cases of suicide in elderly people. Br. J. Psychiatry. 166, 451–457. Centers for Disease Control and Prevention, 1996. Suicides among Older Persons, United States, 1980–1992, MMWR (Morb. Mortal. Weekly Rep.) 45, 3–6. Cohen, J.S., Fihn, S.D., Boyko, E.J., Jonsen, A.R., Wood, R.W., 1994. Attitudes towards assisted suicide and euthanasia among physicians in Washington State. N. Engl. J. Med. 331, 89–94. Conwell, Y., Duberstein, P.R., Cox, C., Herrmann, J.H., Forbes, N.T., Caine, E.D., 1996. Relationship of age and axis I diagnoses in victims of completed suicide: A psychological autopsy study. Am. J. Psychiatry 153, 1001–1008. Diekstra, R.F.W., Gulbinat, W., 1993. The epidemiology of suicidal behaviour: A review of three continents. Wld. Hlth. Statis. Quart. 46, 52–68. Duberstein, P.R., Conwell, Y., Cox, C., Podgorski, C.A., Glazer, R.S., Caine, E.D., 1995. Attitudes toward self-determined death: A survey of primary care physicians. J. Am. Geriatr. Soc. 43, 395–400. Ernst, C., Angst, J., 1995. Depression in old age. Is there a real decrease in prevalence? A review. Eur. Arch. Psychiatry Clin. Neurosci. 245, 272–287. Gallo, J.J., Anthony, J.C., Muthen, B.O., 1994. Age differences in the symptoms of depression: A latent trait analysis. J. Gerontol. 49, P251–264. Henriksson, M., Aro, H.A., Marttunen, M.J., Heikkinen, M.E., ¨ Isometsa, E.T., Kuoppasalmi, K.I., Lonnqvist, J.K., 1993. Mental disorders and comorbidity of suicide. Am. J. Psychiatry 150, 935–940. Henriksson, M.M., Marttunen, M.J., Isometsa, E.T., Heikkinen, ¨ M.E., Aro, H.M., Kuoppasalmi, K.I., Lonnqvist, J.K., 1995. Mental disorders in elderly suicide. Int. Psychogeriatr. 7, 275– 286. Holt, W.S., Mazzucca, S.A., 1992. A written case simulation of osteoarthritis as a predictor of prescribing behaviour among family practitioners. Acad. Med. 67, 414–415. Isacsson, G., Bergman, U., Rich, C.L., 1996. Epidemiological data suggest antidepressants reduce suicide risk among depressives. J. Affect. Disord. 41, 1–8. Isometsa, E.T., Heikkinen, M.E., Marttunen, M.J., Henriksson, M.M., Aro, H.M., Lonnqvist, J.K., 1995. The last appointment
before suicide: Is suicide intent communicated?. Am. J. Psychiatry 152, 919–922. Isometsa, E.T., Henriksson, M.M., Aro, H.M., Heikkinen, M.E., Kuoppasalmi, K.I., Lonnqvist, J.K., 1994. Suicide in major depression. Am. J. Psychiatry 151, 530–536. Jones, T.V., Gerrity, M.S., Earp, J., 1990. Written case simulations: Do they predict physician’s behavior?. J. Clin. Epidemiol. 43, 805–815. Katona, C.L.E., 1995. Detecting and managing depression in older people. Hum. Psychopharmacol. 10, S229–S234. Lebowitz, B.D., Pearson, J.L., Schneider, L.S., Reynolds, C.F., Alexopoulos, G.S., Bruce, M.L., Conwell, Y., Katz, I.R., Meyers, B.S., Morrison, M.F., Mossey, J., Niederehe, G., Parmelee, P., 1997. Diagnosis and treatment of depression in late life. Consensus statement update. JAMA 278, 1186–1190. Lloyd, K., Jenkins, R., 1995. The economics of depression in primary care. Department of Health Initiatives. Br. J. Psychiatry 166 (Suppl. 27), 60–62. Lyness, J.M., Cox, C., Curry, J., Conwell, Y., King, D.A., Caine, E.D., 1995. Older age and the underreporting of depressive symptoms. J. Am. Geriatr. Soc. 43, 216–221. NIH Consensus Development Panel on Depression in Late Life, 1992. Diagnosis and treatment of depression in late life, JAMA 168, 1018–1024. Rihmer, Z., Rutz, W., Pihlgren, H., 1995. Depression and suicide in Gotland: An intensive study of all suicides before and after a depression training programme for general practitioners. J. Affect. Disord. 35, 147–152. ´ Robbins, J.M., Kirmayer, L.J., Cathebras, P., Yaffe, M.J., Dworkind, M., 1994. Physician characteristics and the recognition of depression and anxiety in primary care. Med. Care 32, 795– 812. Rutz, W., von Knorring, L., Walinder, J., 1989. Frequency of suicide on Gotland after systematic postgraduate education of general practitioners. Acta Psychiatr. Scand. 80, 151–154. Rutz, W., von Knorring, L., Walinder, J., 1992. Long-term effects of an educational program for general practitioners given by the Swedish Committee for the Prevention and Treatment of Depression. Acta Psychiatr. Scand. 85, 83–88. Starkstein, S.E., Robinson, R.G., 1989. Affective disorders and cerebral vascular disease. Br. J. Psychiatry 154, 170–182. Stoppe, G., Sandholzer, H., Staedt, J., Winter, S., Kiefer, J., ¨ Kochen, M., Ruther, E., 1994. Diagnosis of dementia in primary care: Results of a representative survey in lower Saxony. Eur. Arch. Psychiatry Clin. Neurosci. (Germany) 244, 278–283. Tiemens, B.G., Ormel, J., Simon, G., 1996. Occurrence, recognition, and outcome of psychological disorders in primary care. Am. J. Psychiatry 153, 636–644. Yoder, C.Y., Shute, G.E., Tryban, G.M., 1990. Community recognition of objective and subjective characteristics of depression. Am. J. Commun. Psychol. 18, 547–566.
Brief report
Family physicians and the risk of suicide in the depressed elderly a, b a a a Gabriela Stoppe *, Hagen Sandholzer , Claudia Huppertz , Hauke Duwe , Juergen Staedt a
b
Department of Psychiatry, Georg-August University, Von-Siebold Str. 5, 37075 Goettingen, Germany Department of General Practice, Hannover Medical School, Carl-Neuberg Str. 1, 30625 Hannover, Germany Received 26 May 1998; received in revised form 25 July 1998; accepted 3 August 1998
Abstract Background: Depression is the most frequent psychiatric disorder in the elderly. It is the reason for most suicides in this age group. Method: We performed a representative survey in primary care. Two written case vignettes were presented to 170 family physicians in face-to-face interviews which took place in their practices. The case vignettes described either (Case 1) a mildly depressed otherwise healthy old patient or a severely depressed patient (Case 2) with somatic comorbidity. Afterwards the interviewers asked standardized open questions. The physicians were not let into the mental health focus of the study. Results: The response rate was 77.6%. Depression was considered for primary or differential diagnosis by 91.2% of the physicians in Case 1 and by 70% in Case 2 (X 2 -test; p , 0.01). For further anamnesis, only 2.4% of the physicians were interested in suicidal ideation of the patient. When directly asked at the end of the interview, 76.9% of the physicians said they would talk about suicide. Those who would not, thought that the patient would communicate suicidal intent himself / herself, or they feared to induce suicide by asking directly. Conclusion: Thinking of suicidality and its prevention is not uppermost in the physicians’ mind. Therefore, and also with regard to the relatively high rate of depression recognition, we conclude that educational means should not only focus on the recognition and screening of depression, but also on the management—‘how to talk about . . . ’—of complex problems like suicide in the elderly, in order to change suicide rates. 1999 Elsevier Science B.V. All rights reserved. Keywords: Suicide; Depression; Old age; Primary care; Family physicians; Cued and uncued answers; Case vignettes
1. Introduction Recent studies in the US revealed increasing suicide rates in the elderly in recent years (Diekstra *Corresponding author. Tel.: 1 49-551-39-95-63; fax: 1 49551-39-66-92. E-mail address: [email protected] (G. Stoppe)
and Gulbinat, 1993; Centers for Disease Control and Prevention, 1996). Psychiatric illness accounts for about 90% of all suicides, with depression making up much more than 50% especially in the elderly (Henriksson et al., 1993, 1995; Conwell et al., 1996). Recent evidence reveals that early recognition and treatment of the underlying psychiatric disorder is the most important method of suicide prevention
0165-0327 / 99 / $ – see front matter 1999 Elsevier Science B.V. All rights reserved. PII: S0165-0327( 98 )00149-9
194
G. Stoppe et al. / Journal of Affective Disorders 54 (1999) 193 – 198
(Isacsson et al., 1996; Rutz et al., 1989, 1992; Rihmer et al., 1995; Ahrens et al., 1995). Different guidelines recommend routine screening and evaluation of suicide risk in high risk groups (American Medical Association, 1994; Canadian Task Force on the Periodic Health Examination, 1994). In the elderly, the recognition and treatment of depression is of major importance because it is a frequent disorder (Beekman et al., 1995) and can be treated as successful as in younger age groups (Katona, 1995; Lebowitz et al., 1997). However, more than for younger ages, it goes under-recognized and undertreated (Lebowitz et al., 1997; NIH Consensus Development Panel on Depression in Late Life, 1992). Since most of the elderly exclusively visit their family physicians, these hold the key position to better recognition and management. Two thirds of all suicides visited their family physician within the month preceding the suicide (Lloyd and Jenkins, 1995; Cattell and Jolley, 1995). However, only one fifth of those who would commit suicide within the following 4 weeks had discussed suicide during their last appointment, according to a Finnish study (Isometsa et al., 1995). In old age, some patient factors make detection and treatment of depression and suicidality even more difficult (NIH Consensus Development Panel on Depression in Late Life, 1992; Gallo et al., 1994; Lyness et al., 1995). In addition different attitudes of the physicians and the community with regard to suicide in old compared to young patients must be taken into account (Duberstein et al., 1995). There is some evidence that gender differences on the part of the patients as well as on the part of the physicians may be influencing this situation. Duberstein et al. (1995) found that female physicians ‘‘hold attitudes toward suicide that are different from male physicians and, perhaps, from other women’’. In the Gotland study the rate of female depressive suicides decreased dramatically after the training program, while the proportion of male depressive suicides was almost unchanged (Rihmer et al., 1995). Finally, Isometsa et al. (1994) found significant sex differences in current and previous treatment of major depressed patients who later died of suicide. With a two-step design the following study reveals information about the willingness of family physi-
cians to elicit a history of suicidal ideation in their elderly depressed patients.
2. Materials and methods
2.1. The case vignettes We designed written sample case histories. The severity of the depression and the gender of the patient were the main factors which were experimentally varied. Case 1 describes a patient free of other diseases with a mild depression. The symptoms have been described to be typical of old age depression (Ernst and Angst, 1995; Brodaty et al., 1991). Case 2 describes a patient with moderate to severe depression with the same social characteristics as the patient in Case 1. The severity is underlined by some typical depressive delusions. The patient is brought to the doctor by his / her daughter and he / she suffers from physical diseases and is treated with medication on a regular basis. The case history is given in two versions, each describing either stroke or thyroid disorder. These have a high rate of depressive comorbidity in common (Brodaty et al., 1991; Starkstein and Robinson, 1989), but differ with regard to the origin (‘primary cerebral’ vs. ‘primary non-cerebral’). For each case different versions with regard to patient’s gender were used: in Case 1 only the gender of patients varied (Case 1a: male; Case 1b: female), in Case 2 both the gender (female / male) and the anamnesis (stroke / hypothyreosis) varied (Case 2a: female 1 stroke; Case 2b: female 1 thyroid; Case 2c: male and stroke; Case 2d: male and thyroid).
2.1.1. Text of Case 1 A 70-year-old man / woman whom you have known for several years presents himself / herself at your clinic. You do not know him / her to be severely ill. He / She takes no regular medication. Following the death of his wife / her husband 5 years ago, he / she has been living alone in a three-bedroom
G. Stoppe et al. / Journal of Affective Disorders 54 (1999) 193 – 198
apartment and caring for the household on his / her own. His / Her only daughter works full-time and lives with her husband and children in the same community. The patient complains of progressive tiredness and a loss of drive. He / She says he / she is hardly able to concentrate and becomes increasingly irritated. After further inquiry he / she says that he / she has less appetite and is worried about his / her health. The patient appears to be in low spirits but he / she seems to cheer up during the consultation.
2.1.2. Case 2 a,c A 70-year-old male / female patient is brought to your clinic by his / her daughter, who lives nearby. ♠He / She has suffered from hypertension for years. One year ago he / she had a stroke. An initial hemiparesis has disappeared with the exception of a slight disability of one arm. ♠He / She is on aspirine, an ACE-antagonist and a diuretic on a regular basis. With a moderate diet, serum levels of glucose and lipids are within normal range. The patient has not shown up at the practice recently and is now brought on his / her daughter’s instigation. The patient himself / herself is sitting downheartedly in his / her chair as the daughter reports that his / her personality and behaviour has considerably changed recently. He / She does nothing on her own, and even has to be persuaded to eat. Therefore he / she has had a considerable weight loss over the preceding 2 weeks. Talking to him / her makes him / her irritable. He / She is convinced that he / she had spent too much on his / her living and that his / her complaints were a punishment for his / her previous lifestyle. As the patient himself / herself is asked he / she says that being incurably ill he / she certainly will die soon. Case 2b,d differ only with regard to the information given between the ♠ signs: Some time ago he / she was diagnosed to have a hypothyroidism, which is treated with oral l-thyroxine 125 mg daily. Since then thyroid function has been normal on controls. Because of arterial hypertension, he / she is on an ACE-antagonist and a diuretic on a regular basis.
195
2.2. Study sample, interview procedure and analysis The study was carried out from July to September 1995. On the basis of previous experience and pilot studies we developed the following design: We obtained a list of all doctors who run a private practice in the city of Kassel (approximately 200 000 inhabitants) and the surrounding regions from the ¨ regional association of physicians (Kassenarztliche Vereinigung). We selected all family physicians (FP), who—in the German health care system—are general practitioners and many internal-medicine specialists. Two trained investigators (C.H. and H.D.) phoned all eligible FP and requested a face-to-face interview after explaining that the purpose of the study would be the management of old patients in primary care. All versions of Case 1 and Case 2 were randomly assigned in advance to the doctors consenting to participate in the study. The interviewers visited their practices and asked the physicians to fill out a short questionnaire concerning statistical data (age, sex, medical education, practice size). Then the interviewers presented a pair of Case 1 and Case 2 vignettes to the physician. Following a short period of perusal, after each vignette, the interviewers asked them standardized questions and categorized the answers to a classification scheme developed in the pilot studies. Thus suggestive effects of given answer categories were avoided. Finally we asked them whether they regarded the case history as somewhat typical for their practice. The whole procedure had a mean duration of 30 minutes. With regard to the management of suicidality we developed a two-step procedure in our questionnaire. The initial two questions were on (differential) diagnostic considerations. The third question— ‘Which anamnestic details would you like to know?’ (multiple answers possible)—was the first, upon which consideration of suicide risk could have been mentioned. As no cues were given at this stage of the interview, we expected that the answers to this question reflected the actual frequency in daily practice. The next questions focused on diagnostics, consultations, treatments and caregiver information. At the end (because we would not induce a bias in favour of the mental health focus), the interviewers
196
G. Stoppe et al. / Journal of Affective Disorders 54 (1999) 193 – 198
asked explicitly whether the physicians would ask the patient for suicidal ideation. Those who answered that they would not were asked for their reasons. Statistical analysis was performed with a computer software (SPSS). We analysed independent research questions by use of the X 2 -test. Results were regarded as significant at a level of p , 0.05.
3. Results In the investigation area 239 physicians work in private practices according to the list of the chamber of physicians. A total of 20 doctors had to be excluded as they did not practise any more (n 5 6) or they were not primary care physicians (11). From the remaining physicians (n 5 219), 77.6% (125 general practitioners and 45 primary care internists) took part in the study. In detail, the group consisted of 131 males and 39 females, with a mean age of 47.04 years, who had been running their practices for 10.7 years. They estimated the percentage of old patients to be 39% on average. Eighty-eight percent of the respondents regarded the case vignettes as typical for their practice. The responder group and the nonresponder group did not differ significantly on the basis of any of the available data. Thus, we met with the criteria for a representative study. Depression was considered for primary or differential diagnosis by 91.2% of the physicians in Case 1 and by 70% in Case 2 (X 2 -test; p , 0.01). About half of the physicians, with regard to both cases, wanted further information regarding, for example lifestyle and gastrointestinal disturbances. However, only 4 (2.4%) physicians considered suicidality in Case 1 and 10 (5.9%) in Case 2. To the direct question 130 (76.9%) physicians in Case 1 and
126 (74.6%) doctors in Case 2 answered that they would ask the patient for suicidal ideation during the initial consultation or later. The answer category ‘later’ was not further specified during the interview. The differences between cued and uncued answers were highly significant (Case 1: X 2 5 126.01, df 5 1, p , 0.01; Case 2: X 2 5 116.01, df 5 1, p , 0.01). There were no significant differences between the answers given to Case 1 compared to Case 2. About 50% of those who would not check for suicidality, argued that ‘there was no reason’ to do so. Other reasons and details are given in Table 1. Again there were no significant differences between the answers given to Case 1 and Case 2. Considering the gender focus of our study, we found no significant differences in the cued and uncued answers nor with regard to the reasons given not to speak about suicidal ideation. This holds true with regard to the patients’ and the physicians’ gender.
4. Discussion The method we used is well known in epidemiological research as a tool to measure physicians’ competence, attitudes and also actual behaviour (Duberstein et al., 1995; Jones et al., 1990; Holt and Mazzucca, 1992; Yoder et al., 1990). We have discussed the advantages and problems of this method previously in publications of an earlier study of our group (Stoppe et al., 1994). An advantage with regard to the questions of this study is that a standardised patient, which was regarded by 88% of the physicians as typical for their practice, could be described in two versions, which only differed with regard to patient’s gender, depression severity or
Table 1 The reasons mentioned by those physicians who would not ask the patient for suicidal thoughts
There is no reason The patient will mention this topic himself / herself Fear to induce suicide All other reasons
Case 1 (n 5 37) (%)
Case 2 (n 5 40) (%)
45.9 27.1 18.9 8.1
50 17.5 12.5 20.0
Data are given as a percentage of the ‘no’-answers (n). There are no significant differences between the answers given for Case 1 and those given for Case 2.
G. Stoppe et al. / Journal of Affective Disorders 54 (1999) 193 – 198
somatic comorbidity. By this means we can see the extent to which these variables account for differences in the answers. Another advantage is that the physicians were not informed about the mental health focus nor about the gender focus of the study. Thus the design fulfilled the demand that decision-making studies should embed the psychosocial questions within a list of other possible issues (Robbins et al., 1994). The face-to-face interview allowed standardised questions without the risk of potential suggestions by given answer categories and allowed clarifying questions if necessary. The technique was—according to our experience—also the reason for the high response rate and the representativity of the study. One might argue that relevant factors like the physicians’ interview skills and behaviour or their sensitivity, for example to non-verbal communications, make no difference in the interview situation. However, their knowledge, their psychological awareness and their interest for psycho-social problems could influence their decisions in this hypothetical situation. In some way the case vignette serves as a ‘matrix’, upon which the condensed experience of the physicians could be projected. This is a major advantage when studying complex topics like gender differences or suicidality. The two-step design to evaluate the physicians’ willingness to consider suicidality in their patients led to the most striking results of this study: Overall 2.4% of the physicians thought of a suicide risk spontaneously and 76.5% when being asked explicitly. The latter value could be influenced by social desirability, whereas the former could reflect the actual behaviour. Taking into account that most physicians considered depression for primary or differential diagnosis in both cases, we can assume that depression was on their mind. The answers to the direct question may reflect the knowledge that suicide is a feature of depression and that one should talk about it. However, the low—compared to other information— number of uncued answers suggests that thinking of suicidality is not uppermost in the physicians’ mind. There may be two reasons for this. First, family physicians may not know how frequent suicide is and how important they are in the prevention of suicide. In Germany, until recently,
197
family physicians did not have an obligatory training in psychiatry or geriatrics in their curriculum although psycho-social problems make up about 30– 50% of their consultations and elderly patients about 40%. Secondly, physicians may be insecure about how to manage suicidality. This is underlined by the result that about 20% of those who would not ask fear to induce suicidality by talking about it. One must also discuss that some fear the consequences if the patient says that she / he suffers from suicidal ideation, although no physician mentioned this in this study. The patient should be referred to hospital or to a primary care neuropsychiatrist, both of which elderly patients will resist (NIH Consensus Development Panel on Depression in Late Life, 1992). The answers with regard to suicidality do not vary with the patient’s or physician’s gender or with the severity of depression. This also indicates that the answers may be explained mostly by knowledge and skill factors on the part of the physicians. With regard to physicians’ gender our results differ from those of other studies investigating the attitudes towards assisted suicide (Duberstein et al., 1995; Cohen et al., 1994). However, we did not focus on attitudes to (assisted) suicide in our study, and this may explain the differences. Educational training in primary care seems to be necessary to reduce suicide rates. This has been proven preliminary effective, although the effect seems to fade after some time and repetition of training is necessary (Rutz et al., 1989, 1992). In agreement with recent studies (Tiemens et al., 1996; Callahan et al., 1996), the results of this study indicate that educational means should not only focus on the recognition and treatment, for example of depression, but also on the management—‘how to talk about . . . ’—of complex problems like suicide.
References ¨ ¨ Ahrens, B., Berghofer, A., Wolf, T., Muller-Oerlinghausen, B., 1995. Suicide attempts, age and duration of illness in recurrent affective disorders. J. Affect. Disord. 36, 43–49. American Medical Association, 1994. Guidelines for Adolescent Preventive Services (GAPS), Recommendations and Rationale, American Medical Association, Chicago, pp. 131–143. Beekman, A.T.F., Deeg, D.J.H., van Tilburg, T., Smit, J.H.,
198
G. Stoppe et al. / Journal of Affective Disorders 54 (1999) 193 – 198
Hooijer, C., van Tilburg, W., 1995. Major and minor depression in later life: A study of prevalence and risk factors. J. Affect. Disord. 36, 65–75. Brodaty, H., Peters, K., Boyce, P., Hickie, I., Parker, G., Mitchell, P., Wilhelm, K., 1991. Age and depression. J. Affect. Disord. 23, 137–149. Callahan, C.M., Dittus, R.S., Tierney, W., 1996. Primary care physicians’ medical decision making for late life depression. J. Gen. Intern. Med. 11, 218–225. Canadian Task Force on the Periodic Health Examination, 1994. Canadian Guide to Clinical Preventive Health Care, Canada Communication Group, Ottawa, Canada, pp. 456–470. Cattell, H., Jolley, D.J., 1995. One hundred cases of suicide in elderly people. Br. J. Psychiatry. 166, 451–457. Centers for Disease Control and Prevention, 1996. Suicides among Older Persons, United States, 1980–1992, MMWR (Morb. Mortal. Weekly Rep.) 45, 3–6. Cohen, J.S., Fihn, S.D., Boyko, E.J., Jonsen, A.R., Wood, R.W., 1994. Attitudes towards assisted suicide and euthanasia among physicians in Washington State. N. Engl. J. Med. 331, 89–94. Conwell, Y., Duberstein, P.R., Cox, C., Herrmann, J.H., Forbes, N.T., Caine, E.D., 1996. Relationship of age and axis I diagnoses in victims of completed suicide: A psychological autopsy study. Am. J. Psychiatry 153, 1001–1008. Diekstra, R.F.W., Gulbinat, W., 1993. The epidemiology of suicidal behaviour: A review of three continents. Wld. Hlth. Statis. Quart. 46, 52–68. Duberstein, P.R., Conwell, Y., Cox, C., Podgorski, C.A., Glazer, R.S., Caine, E.D., 1995. Attitudes toward self-determined death: A survey of primary care physicians. J. Am. Geriatr. Soc. 43, 395–400. Ernst, C., Angst, J., 1995. Depression in old age. Is there a real decrease in prevalence? A review. Eur. Arch. Psychiatry Clin. Neurosci. 245, 272–287. Gallo, J.J., Anthony, J.C., Muthen, B.O., 1994. Age differences in the symptoms of depression: A latent trait analysis. J. Gerontol. 49, P251–264. Henriksson, M., Aro, H.A., Marttunen, M.J., Heikkinen, M.E., ¨ Isometsa, E.T., Kuoppasalmi, K.I., Lonnqvist, J.K., 1993. Mental disorders and comorbidity of suicide. Am. J. Psychiatry 150, 935–940. Henriksson, M.M., Marttunen, M.J., Isometsa, E.T., Heikkinen, ¨ M.E., Aro, H.M., Kuoppasalmi, K.I., Lonnqvist, J.K., 1995. Mental disorders in elderly suicide. Int. Psychogeriatr. 7, 275– 286. Holt, W.S., Mazzucca, S.A., 1992. A written case simulation of osteoarthritis as a predictor of prescribing behaviour among family practitioners. Acad. Med. 67, 414–415. Isacsson, G., Bergman, U., Rich, C.L., 1996. Epidemiological data suggest antidepressants reduce suicide risk among depressives. J. Affect. Disord. 41, 1–8. Isometsa, E.T., Heikkinen, M.E., Marttunen, M.J., Henriksson, M.M., Aro, H.M., Lonnqvist, J.K., 1995. The last appointment
before suicide: Is suicide intent communicated?. Am. J. Psychiatry 152, 919–922. Isometsa, E.T., Henriksson, M.M., Aro, H.M., Heikkinen, M.E., Kuoppasalmi, K.I., Lonnqvist, J.K., 1994. Suicide in major depression. Am. J. Psychiatry 151, 530–536. Jones, T.V., Gerrity, M.S., Earp, J., 1990. Written case simulations: Do they predict physician’s behavior?. J. Clin. Epidemiol. 43, 805–815. Katona, C.L.E., 1995. Detecting and managing depression in older people. Hum. Psychopharmacol. 10, S229–S234. Lebowitz, B.D., Pearson, J.L., Schneider, L.S., Reynolds, C.F., Alexopoulos, G.S., Bruce, M.L., Conwell, Y., Katz, I.R., Meyers, B.S., Morrison, M.F., Mossey, J., Niederehe, G., Parmelee, P., 1997. Diagnosis and treatment of depression in late life. Consensus statement update. JAMA 278, 1186–1190. Lloyd, K., Jenkins, R., 1995. The economics of depression in primary care. Department of Health Initiatives. Br. J. Psychiatry 166 (Suppl. 27), 60–62. Lyness, J.M., Cox, C., Curry, J., Conwell, Y., King, D.A., Caine, E.D., 1995. Older age and the underreporting of depressive symptoms. J. Am. Geriatr. Soc. 43, 216–221. NIH Consensus Development Panel on Depression in Late Life, 1992. Diagnosis and treatment of depression in late life, JAMA 168, 1018–1024. Rihmer, Z., Rutz, W., Pihlgren, H., 1995. Depression and suicide in Gotland: An intensive study of all suicides before and after a depression training programme for general practitioners. J. Affect. Disord. 35, 147–152. ´ Robbins, J.M., Kirmayer, L.J., Cathebras, P., Yaffe, M.J., Dworkind, M., 1994. Physician characteristics and the recognition of depression and anxiety in primary care. Med. Care 32, 795– 812. Rutz, W., von Knorring, L., Walinder, J., 1989. Frequency of suicide on Gotland after systematic postgraduate education of general practitioners. Acta Psychiatr. Scand. 80, 151–154. Rutz, W., von Knorring, L., Walinder, J., 1992. Long-term effects of an educational program for general practitioners given by the Swedish Committee for the Prevention and Treatment of Depression. Acta Psychiatr. Scand. 85, 83–88. Starkstein, S.E., Robinson, R.G., 1989. Affective disorders and cerebral vascular disease. Br. J. Psychiatry 154, 170–182. Stoppe, G., Sandholzer, H., Staedt, J., Winter, S., Kiefer, J., ¨ Kochen, M., Ruther, E., 1994. Diagnosis of dementia in primary care: Results of a representative survey in lower Saxony. Eur. Arch. Psychiatry Clin. Neurosci. (Germany) 244, 278–283. Tiemens, B.G., Ormel, J., Simon, G., 1996. Occurrence, recognition, and outcome of psychological disorders in primary care. Am. J. Psychiatry 153, 636–644. Yoder, C.Y., Shute, G.E., Tryban, G.M., 1990. Community recognition of objective and subjective characteristics of depression. Am. J. Commun. Psychol. 18, 547–566.