misleading and could only suggest to the reader that the source of the press leak was a member of one or other of our departments, which are regularly responsible for the care of the vast majority of HIV infected individuals in Birmingham. T his is simply not true. Confidentiality is a fundamental tenet of the service we provide for HIV infected individuals and under no circumstances would we divulge a patient's name to the media-certainly not to provide "a focus around which to challenge... mounting complacency about AIDS," as suggested by Fitzpatrick. The article further implies that we have a "tendency to regard HIV infection as more a moral condition than a disease" and that we judge people with HIV and AIDS "according to the degree of their individual responsibility for their condition." This does not reflect the attitude of the doctors and other staff dedicated to caring for such people in Birmingham.. M J WOOD C I ELLIS J A INNES S DRAKE
Regional HIV/AIt)S Unit, I)irectorate of'Infectious D)iseases, East Birniingham NtIS TIruist, Birminghaial B9 5sFr M SHAHMANESH K RADCLIFFE
Departmnctti of Genuitourittary Medicine, Genieral Hospital, Btrmingham 1 FitzpatrickiM. AIDS and ethics in Birttluitgham: a betrayal of trust. B.MJ 1992;305:259-60.
AUTHOR'S REPLY, - I am happy to accept that Dr Wood and his colleagues in the regional HIN/AIDS unit and the department of,genitourinary medicine were not responsible for the breach of confidentiality that I described in my article. However, somebody with access to detailed clinical infortnation concerning people with HIV/AIDS in Birmingham clearly did leak such information to the press and that remains a matter of concern. M FITZPATRICK
Lotudon N 16 9JT
Litigation over illness associated with tryptophan is possible EDITOR, --The illness associated with tryptophan known as the eosinophilia-myalgia syndrome was newsworthy two years ago. Five Scottish patients were reported on in the correspondence columns of the BMJ` and the Lancet."4 In the United States around 1500 cases were notified to the Centers for Disease Control, but there may have been 500010 000 cases.' Two years on relatively few patients, possibly only 10%, are free of symptoms.6 The commonest persisting symptoms are myalgia, fatigue, paraesthesia, muscle cramps, and scleroderma-like skin changes. A problem for patients with such symptoms is that there is no objective indicator; there is no recurrence of eosinophilia or important increase in the erythrocyte sedimentation rate. Doctors should be alert to the risk of dismissing such symptoms as non-organic. Reports have suggested that the symptoms may slowly improve with time. Steroids frequently provide symptomatic relief, but there is no convincing evidence of an effect on the duration of the disease.7 The legal case is still being argued in the American courts. The cause has not yet been defined, but there is some evidence of an association with some batches of tryptophan made by the Japanese firm Showa Denko. The material was
BMJ VOLUME 305
12 SEPTEMBER 1992
prepared microbiologically,5 and several changes were made in the manufacturing process, including a change in the organism. T here may be more than one contaminant.9 In the United States numerous support groups for patients with the eosinophilia-myalgia syndrome are well organised into a national network. According to a recent newspaper report in the United States, there are about 1400 plaintiffs with cases against Showa Denko. 10 British sufferers who are generally less litigious, are probably unaware that they could join such action, and the opportunity to do so will soon be time barred. The American patient support groups are willing to guide British sufferers. If any further information is required I am prepared to respond to any doctor writing to this address on issues affecting individual patients. A S DOUGLAS
University Department of Medicine and Therapeutics, University Miedical School, Aberdeen Royal infirmary, Aberdeen AB9 2ZD
or to have AIDS could be flagged at the NHS Central Registry as part of the notification system. This would need a level of identifying patients which is available only in some cases. When the flagged patients died the accuracy of their death certificates could be investigated. Secondly, record linkage of surveillance systems for HIV infection and AIDS with death certification data as held at the General Register Office in Scotland is possible. Such record linkages could be performed periodically, and matching cases with the same soundex code and date of birth could be identified and analysed. The advantage of this method would be that an anonymised identification system such as the soundex code could be used and confidentiality preserved. The second method is also a check on the accuracy of the anonymised data typically held in national and local surveillance systems for HIV infection and AIDS. It could form part of a regular check to validate data held in these surveillance systems. ANDREW RILEY
I Douglas AS, Eagles JM, Mowat NAG. Eosinophilia myalgia syndrome associated with trvptophan. BMJ 1990;301:387. 2 Maclennan AC, Stewart DG. Eosinophilia myalgia syndrome associated with tryptophan. BMJ 1990;301:387-8. 3 Adamson DJA, Legge JS. L--tryptophan-induced eosinophilia withouLt myalgia. Lancet 1991;337:1474-5. 4 Walker KG, Eastmond CJ, Best PV, Matthews K. Eosinophiliamyalgia syndrome associated with prescribed L-tryptophan. Lancet 1991;336:695-6. 5 Ahmed SR, Clauw D. USA: EMS and L-tryptophan. Lsancet 1991;338: 1512. 6 Kaufman LD, Gruber BL, Gregersen PK. Clinical follow-up and immunogenetic studies of 32 patients with eosinophiliamyalgia syndrome. Lancet 1991;337:1071-4. 7 Culpepper RC, WCilliams RG, Mease PJ, Koepsell TD, Kobayashi JMI. Natural history of the eosinophilia-myalgia syndrome. Ann Intern Med 1991;115:437-42. 8 Belongia EA, Hedberg CW, Gleich GJ, White KE, Mayeno AN, Loegering DA, et al. An insestigation of the cause of the eosinophilia-mvalgia syndrome associated with tryptophan use. N Englj Med 1990;323:357-65 9 Swinbanks D, Anderson C. Search for contaminant in EMS outbreak goes slowly. Nature 1992;358:96. 10 Stieber T. L-tryptophan trial starts in US court. Albuquerquie journal 1992 June 4.
HIV infection and certification of death EDITOR,-John Aldous and colleagues' paper discussing the impact of HIV infection on mortality in young men in a London health authority highlights many of the problems regarding death certification, HIV infection, AIDS, and confidentiality.' The authors found that HIV infection was the leading cause of death in 15-44 year old men and emphasise the importance of accuracy in death certification. The fact that HIV infection is the underlying cause of death, however, is not as important as ascertaining the HIV infection status of those who have died. The actual cause of death may have no immediate or obvious link with HIV infection, but the knowledge that the person was infected could be highly relevant, for example, in cases of suicide. It may also be wrong to assume that HIV infection was the underlying cause of death in people known to be infected but certified as having died of other causes. The most that may be assumed is that HIV infection might have been a contributory cause of death in those who are found to be infected after death. While the information on death certificates remains non-confidential inaccuracy in certification will occur. The principle must be to maximise the accuracy of the initial certification. Apart from the current facility to offer further information on cause of death, methods should be established for checking the accuracy of death certification and linking with surveillance systems for HIV infection and AIDS. I suggest two possible methods of linkage. Firstly, people known to be infected with HIV
Department of Public Health Medicine, Lothian Health Board, Edinburgh EH8 9RS 1 Aldous J, Hickman M, Ellam A, Gazzard B, Hargreaves S. Impact of HIN' infection on mortality in young men in a lIondort hcalth authority. B.J 1992;305:219-21. (25 July.)
EDITOR,-John Aldous and colleagues, in their study of male residents of Riverside, concluded that HIV infection was the leading cause of death in those aged 15-44 and the third commonest cause of death in those aged 15-64 years.' The operational criteria for HIV or AIDS related death in this study is misleading as the diagnosis of HIV infection or AIDS does not necessarily imply a cause of death. On this basis, it is unclear what proportion of deaths was due to common bacterial pathogens or non-HIV related causes in those infected with the virus. McCormick, to whom Aldous and colleagues referred quite extensively, was careful in claiming a cause-effect relation and stated only that the observed increase in mortality is "unattributed to but probably associated with HIV infection."2 Mientjes et al reported a high and rising incidence of bacterial pneumonia among HIV infected intravenous drug users in Amsterdam without a consequential rise in non-AIDS mortality.' By contrast, an earlier study in New York showed that deaths caused by bacterial pathogens have risen greatly during the AIDS epidemic.4 The greater mortality among HIV infected intravenous drug users in New York than in Amsterdam would seem to reflect differences in these two cities in the early detection of bacterial pneumonia among intravenous drug users and accessibility to inpatient and outpatient medical services, both of which are important factors in preventing deaths due to common bacterial pathogens. Aldous and colleagues did not consider the health consequences of homelessness in their study or the social and psychological deprivations which invariably follow the diagnosis of HIV. Social isolation, a sense of hopelessness and alienation, suicide or accidental drug overdose, and failure of early detection and treatment of bacterial pneumonia in people infected with HIV deserve specific comment in the analysis of data and interpretation of research findings. Bacterial pneumonia, endocarditis, thromboembolism, and deliberate or accidental self poisoning were major causes of death among intravenous drug users before the advent of HIV and remain largely so. Among single men aged 15-54, HIV arguably has contributed in some degree to increased mortality in the decade 1981-91; that this picture will persist through to the next decade is unlikely, particularly in view of AIDS related self organisation and behaviour change among homosexual groups. We remain vigilant about
647
mortality within the heterosexual population, but there is a serious danger in interpreting every morbid phenomenon in our lives today in the context of HIV. That we have reached the stage where it is fashionable to blame most deaths in the population on HIV or AIDS may distort the perception of what the real problem is. JAMES EDEH Division of Addictive Behaviour, St George's Hospital Medical School, London SW17 ORE 1 Aldous J, Hickman M, Ellam A, Gazzard B, Hargreaves S. Impact of HIV infection on mortality in young men in a London health authority. BMJ 1992;305:219-21. (25 July.) 2 McCormick A. Excess morbidity associated with the HIV epidemic in England and Wales. BMJ7 1991;302:1375-6. 3 Mienties GH, Van Amei;den EJ, Van den Hoek AJAR, Continho RA. Increasing morbidity without rise in non-AIDS mortality among HIV infected intravenous drug users in Amsterdam. AIDS 1992;6:207-12. 4 Selwyn PA, Hartel D, Wasserman W, Drucker E. Impact of AIDS epidemic on morbidity and mortality among intravenous drug users in a New York city methadone maintenance programme; AmJ7 Public Health 1989;79:1358-62.
EDITOR,-John Aldous and colleagues report the mortality in men with HIV infection in Riverside Health Authority; eight of the 213 men who died committed suicide.' Suicidal behaviour in people with HIV infection has been the subject of considerable interest, and Marzuk et al argued that suicide is considerably increased in this group,2 although there are complex methodological problems in establishing its prevalence accurately. ' We have identified six completed suicides in people with HIV infection over two years in Riverside Health Authority. Four of the patients had a history of psychiatric illness: in two this preceded the diagnosis of HIV infection and included a history of deliberate self harm. The two patients without a history of psychiatric illness were in an advanced stage of disease, and one had planned his death for over a year. The mode of suicide was predominantly by jumping-usually a rare form of suicide-but suicides by overdose may be underreported in patients with AIDS. We also looked at deliberate self harm and found 23 cases in people positive for HIV over two years; HIV infection played a part in most cases.4 In recent years a significant increase in suicides in young men has been noted, and HIV infection may well be a contributing factor.5 Staff who care for patients with HIV infection should be aware of the risk of suicidal behaviour and should refer to psychiatric services patients with suicidal ideas, particularly those with a history of psychiatric illness or deliberate self harm, especially as the disease becomes more advanced. KATHRYN PUGH JOSE CATALAN Academic and Clinical Department of Psychological Medicine, Charing Cross and Westminster Medical School, London SWIP 2NS 1 Aldous J, Hickman M, Ellam A, Gazzard B, Hargreaves S. Impact of HIV infection in young men in a London health
authority. BMJ 1992;305:219-21. (25 July.) 2 Marzuk P, Tierney H, Tardiff K, Gross EM, Morgan EB, Hsu M-A, et al. Increased risk of suicide in patients with AIDS.
JAMA 1988;259:1333-7. 3 Marzuk P. Suicidal behaviour and HIV illnesses. International Review ofPsychiatry 1991;3:367. 4 Catalan J. Deliberate self-harm in HIV disease. Research into clinical and ethical aspects. In: Beskow JE, Bellini M, Sampaio Faria JG, Kerkhof AD, eds. Report on the WHO consultation, Bologna, September 1990, HIV and AIDS related suicidal behaviour. Bologna: Monduzzi Editore, 1991:61-8. S Hawton K. Suicide in young men. BMJ 1992;304:100O.
Treatment of natal cleft sinus EDITOR,-Patients who have had recurrent treatment for persistent pilonidal sinus' 2 are sometimes referred, in the last resort, to plastic surgeons. This is done in the hope that the plastic surgeon can either create skin flaps to obliterate the natal cleft 648
or perform expert skin grafting which will remove the hair follicles in and around the natal cleft. In a few patients who have tightly opposed, muscular buttocks friction between the buttocks may play an important part in the persistence of pilonidal disease. I carried out a combination of liposuction and limited skin grafting in such a patient. Liposuction of the medial aspect of the buttocks reduced the sharp angle of the cleft -and eliminated the tendency of one buttock to rub against the other. Excision of the scarred area in the base of the cleft and use of a quilted meshed skin graft completed the procedure. This was followed by primary healing, and the patient has been free of any pilonidal problems for the past six months. He had suffered recurrent disease every month or so for the previous four years despite innumerable operations. I hesitate to make any claims for this procedure based merely on one case, but I thought it worth reporting because other surgeons might be interested in trying it. I am unlikely to be able to build up a series because these difficult recurrent cases are relatively rare. H P HENDERSON
Department of Plastic Surgery, Leicester Royal Infirmary, Leicester LEI 5WW 1 Correspondence. Treatment of natal cleft sinus. BMJ 1992;305: 311-2. (1 August.) 2 Khawaja HT, Bryan S, Weaver PC. Treatment of natal cleft sinus: a prospective clinical and economic evaluation. BMJ 1992;304:1282-3. (16 May.)
Over the counter candidiasis
trlatment for
EDITOR,-Susan Mitchell and Caroline Bradbeer provide an interesting interpretation of the presentation patterns for vaginal candidiasis that may follow the change in legal status for intravaginal imidazoles.' The recommendation of the Committee on Safety of Medicines to exempt such antifungals from prescription only control recognises the efficacy of the products and the role of community pharmacy in primary health care and resolves the distinction between infection in men for whom over the counter preparations have been available for some time, and that in women. Because of the complexity of the differential diagnosis in women presenting with "thrush," over the counter imidazoles are available only to clients experiencing a suspected recurrence of candidiasis. The pharmaceutical press has made strenuous efforts to ensure that pharmacists are aware of the circumstances for medical referral, describing 11 scenarios in which over the counter treatment would be inappropriate.2 Accurate diagnosis of vaginal candidiasis will therefore necessitate assimilation of potentially embarrassing information from a client who may expect to receive her over the counter preparation automatically on request. This requires a private area for counselling and competency in communication skills on the part of the pharmacist. The importance of communication skills forms an integral part of undergraduate education for pharmacists.3 The most problematic issue, however, is obtaining a history of sexually transmitted disease or exposure to a partner with sexually transmitted disease. A client may presume that she has a recurrence of candidiasis when the aetiological agent is in fact a sexually transmitted pathogen, such as herpes simplex virus or Trichomonas vaginalis. The danger is that a woman may not disclose, or even be aware of, exposure to a sexually transmitted disease. Exempting vaginal imidazoles from prescription only control may be of considerable benefit to women who can recognise their recurrent symptoms and can afford to pay for their treatment over the counter. The caveat is clearly that wider
availability may lead to more cases of treatment failure and undiagnosed sexually transmitted diseases. Ultimately, medical practitioners will detect whether a change in the pattern of presentation of vaginal candidiasis has indeed occurred. GRETA FORSTER DAVID G WEBB
Royal London Trust, Royal London Hospital, London El 1BB I Mitchell S, Bradbeer C. Over the counter treatment for candidiasis. BMJ 1992;304:1648. (27 June.) 2 Li Wan Po A. Treatment of vaginal candidiasis. Pharmaceutical 3rournal 1992;249: i-is. 3 Committee of Inquiry. Pharmacy: a report to the Nuffield Foundatnm. London: Nuffield Foundation, 1986:85-104.
Screening for depression in elderly patients EDITOR,-In describing use of a new screening instrument for depression in elderly patients' Fiona Adshead and colleagues address a problem with self rating scales2": many elderly patients are unable to use them and require help. This difficulty is illustrated by the use of a mixed mode of presentation of the geriatric depression scale in the study (self administered or staff assisted). This is clinically relevant as a staff assisted scale is associated with a lower score than the self administered form.2 Two points are worthy of comment. The brief assessment schedule depression cards (BASDEC) are printed in a large typeface. If all patients could read the BASDEC but some could not read the geriatric depression scale this suggests that the typefaces were of different sizes: could this partly explain the patients' modest preference for the BASDEC? Although the BASDEC took less time to administer in Adshead and colleagues' study, a shortened form of the geriatric depression scale is under evaluation4 and may reduce the difference in time required to perform the measures. The geriatric depression scale has an advantage over the BASDEC in that it has been shown to be useful in people with mild to moderate cognitive impairment.' If the BASDEC can be shown to share this quality its utility to clinicians will be further enhanced. DESMOND O'NEILL
Elderly Services, Selly Oak Hospital, Birmingham B29 6JD I Adshead F, Day Cody D, Pitt B. BASDEC: a novel screening instrument for depression in elderly medical inpatients. BMJ 1992;305:397. (15 August.) 2 O'Neill D, Rice I, Blake P, Walsh JB, Coakley D. The geriatric depression scale: rater-administered or self-administered? International Journal of Geriatric Psychiatry 1992;7:511-5. 3 Toner J, Gurland B, Teresi J. Comparison of self-administered and rater-administered methods of assessing levels of severity of depression in elderly patients. J Gerontol 1988;43:136-40. 4 Yesavage JA. Geriatric depression scale. Psychopharmacol Bull 1988;24:709-1 1. 5 O'Riordan T, Hayes J, O'Neill D, Shelley R, Walsh JB, Coakley D. The effect of mild to moderate dementia on the geriatric depression scale and on the general health questionnaire in the hospitalized elderly. Age Ageing 1990;19:57-61.
Human milk banks EDITOR,-In an editorial on human milk banks D P Davies states, "At 9 months the mean developmental quotient was shown to be significantly lower in infants fed donor breast milk than in those fed a preterm formula,"' citing the work of Lucas et al.I Although this is widely believed, it is a view not entirely borne out by the data. According to Lucas et al's paper, at 9 months of age the mean (SD) developmental quotient of the babies fed banked human milk was 97-2 (8 5) and that of the babies fed preterm formula 98-2 (11.0).2 Thus the difference between the groups was not BMJ
VOLUME 305
12 SEPTEMBER 1992
Regional HIV/AIt)S Unit, I)irectorate of'Infectious D)iseases, East Birniingham NtIS TIruist, Birminghaial B9 5sFr M SHAHMANESH K RADCLIFFE
Departmnctti of Genuitourittary Medicine, Genieral Hospital, Btrmingham 1 FitzpatrickiM. AIDS and ethics in Birttluitgham: a betrayal of trust. B.MJ 1992;305:259-60.
AUTHOR'S REPLY, - I am happy to accept that Dr Wood and his colleagues in the regional HIN/AIDS unit and the department of,genitourinary medicine were not responsible for the breach of confidentiality that I described in my article. However, somebody with access to detailed clinical infortnation concerning people with HIV/AIDS in Birmingham clearly did leak such information to the press and that remains a matter of concern. M FITZPATRICK
Lotudon N 16 9JT
Litigation over illness associated with tryptophan is possible EDITOR, --The illness associated with tryptophan known as the eosinophilia-myalgia syndrome was newsworthy two years ago. Five Scottish patients were reported on in the correspondence columns of the BMJ` and the Lancet."4 In the United States around 1500 cases were notified to the Centers for Disease Control, but there may have been 500010 000 cases.' Two years on relatively few patients, possibly only 10%, are free of symptoms.6 The commonest persisting symptoms are myalgia, fatigue, paraesthesia, muscle cramps, and scleroderma-like skin changes. A problem for patients with such symptoms is that there is no objective indicator; there is no recurrence of eosinophilia or important increase in the erythrocyte sedimentation rate. Doctors should be alert to the risk of dismissing such symptoms as non-organic. Reports have suggested that the symptoms may slowly improve with time. Steroids frequently provide symptomatic relief, but there is no convincing evidence of an effect on the duration of the disease.7 The legal case is still being argued in the American courts. The cause has not yet been defined, but there is some evidence of an association with some batches of tryptophan made by the Japanese firm Showa Denko. The material was
BMJ VOLUME 305
12 SEPTEMBER 1992
prepared microbiologically,5 and several changes were made in the manufacturing process, including a change in the organism. T here may be more than one contaminant.9 In the United States numerous support groups for patients with the eosinophilia-myalgia syndrome are well organised into a national network. According to a recent newspaper report in the United States, there are about 1400 plaintiffs with cases against Showa Denko. 10 British sufferers who are generally less litigious, are probably unaware that they could join such action, and the opportunity to do so will soon be time barred. The American patient support groups are willing to guide British sufferers. If any further information is required I am prepared to respond to any doctor writing to this address on issues affecting individual patients. A S DOUGLAS
University Department of Medicine and Therapeutics, University Miedical School, Aberdeen Royal infirmary, Aberdeen AB9 2ZD
or to have AIDS could be flagged at the NHS Central Registry as part of the notification system. This would need a level of identifying patients which is available only in some cases. When the flagged patients died the accuracy of their death certificates could be investigated. Secondly, record linkage of surveillance systems for HIV infection and AIDS with death certification data as held at the General Register Office in Scotland is possible. Such record linkages could be performed periodically, and matching cases with the same soundex code and date of birth could be identified and analysed. The advantage of this method would be that an anonymised identification system such as the soundex code could be used and confidentiality preserved. The second method is also a check on the accuracy of the anonymised data typically held in national and local surveillance systems for HIV infection and AIDS. It could form part of a regular check to validate data held in these surveillance systems. ANDREW RILEY
I Douglas AS, Eagles JM, Mowat NAG. Eosinophilia myalgia syndrome associated with trvptophan. BMJ 1990;301:387. 2 Maclennan AC, Stewart DG. Eosinophilia myalgia syndrome associated with tryptophan. BMJ 1990;301:387-8. 3 Adamson DJA, Legge JS. L--tryptophan-induced eosinophilia withouLt myalgia. Lancet 1991;337:1474-5. 4 Walker KG, Eastmond CJ, Best PV, Matthews K. Eosinophiliamyalgia syndrome associated with prescribed L-tryptophan. Lancet 1991;336:695-6. 5 Ahmed SR, Clauw D. USA: EMS and L-tryptophan. Lsancet 1991;338: 1512. 6 Kaufman LD, Gruber BL, Gregersen PK. Clinical follow-up and immunogenetic studies of 32 patients with eosinophiliamyalgia syndrome. Lancet 1991;337:1071-4. 7 Culpepper RC, WCilliams RG, Mease PJ, Koepsell TD, Kobayashi JMI. Natural history of the eosinophilia-myalgia syndrome. Ann Intern Med 1991;115:437-42. 8 Belongia EA, Hedberg CW, Gleich GJ, White KE, Mayeno AN, Loegering DA, et al. An insestigation of the cause of the eosinophilia-mvalgia syndrome associated with tryptophan use. N Englj Med 1990;323:357-65 9 Swinbanks D, Anderson C. Search for contaminant in EMS outbreak goes slowly. Nature 1992;358:96. 10 Stieber T. L-tryptophan trial starts in US court. Albuquerquie journal 1992 June 4.
HIV infection and certification of death EDITOR,-John Aldous and colleagues' paper discussing the impact of HIV infection on mortality in young men in a London health authority highlights many of the problems regarding death certification, HIV infection, AIDS, and confidentiality.' The authors found that HIV infection was the leading cause of death in 15-44 year old men and emphasise the importance of accuracy in death certification. The fact that HIV infection is the underlying cause of death, however, is not as important as ascertaining the HIV infection status of those who have died. The actual cause of death may have no immediate or obvious link with HIV infection, but the knowledge that the person was infected could be highly relevant, for example, in cases of suicide. It may also be wrong to assume that HIV infection was the underlying cause of death in people known to be infected but certified as having died of other causes. The most that may be assumed is that HIV infection might have been a contributory cause of death in those who are found to be infected after death. While the information on death certificates remains non-confidential inaccuracy in certification will occur. The principle must be to maximise the accuracy of the initial certification. Apart from the current facility to offer further information on cause of death, methods should be established for checking the accuracy of death certification and linking with surveillance systems for HIV infection and AIDS. I suggest two possible methods of linkage. Firstly, people known to be infected with HIV
Department of Public Health Medicine, Lothian Health Board, Edinburgh EH8 9RS 1 Aldous J, Hickman M, Ellam A, Gazzard B, Hargreaves S. Impact of HIN' infection on mortality in young men in a lIondort hcalth authority. B.J 1992;305:219-21. (25 July.)
EDITOR,-John Aldous and colleagues, in their study of male residents of Riverside, concluded that HIV infection was the leading cause of death in those aged 15-44 and the third commonest cause of death in those aged 15-64 years.' The operational criteria for HIV or AIDS related death in this study is misleading as the diagnosis of HIV infection or AIDS does not necessarily imply a cause of death. On this basis, it is unclear what proportion of deaths was due to common bacterial pathogens or non-HIV related causes in those infected with the virus. McCormick, to whom Aldous and colleagues referred quite extensively, was careful in claiming a cause-effect relation and stated only that the observed increase in mortality is "unattributed to but probably associated with HIV infection."2 Mientjes et al reported a high and rising incidence of bacterial pneumonia among HIV infected intravenous drug users in Amsterdam without a consequential rise in non-AIDS mortality.' By contrast, an earlier study in New York showed that deaths caused by bacterial pathogens have risen greatly during the AIDS epidemic.4 The greater mortality among HIV infected intravenous drug users in New York than in Amsterdam would seem to reflect differences in these two cities in the early detection of bacterial pneumonia among intravenous drug users and accessibility to inpatient and outpatient medical services, both of which are important factors in preventing deaths due to common bacterial pathogens. Aldous and colleagues did not consider the health consequences of homelessness in their study or the social and psychological deprivations which invariably follow the diagnosis of HIV. Social isolation, a sense of hopelessness and alienation, suicide or accidental drug overdose, and failure of early detection and treatment of bacterial pneumonia in people infected with HIV deserve specific comment in the analysis of data and interpretation of research findings. Bacterial pneumonia, endocarditis, thromboembolism, and deliberate or accidental self poisoning were major causes of death among intravenous drug users before the advent of HIV and remain largely so. Among single men aged 15-54, HIV arguably has contributed in some degree to increased mortality in the decade 1981-91; that this picture will persist through to the next decade is unlikely, particularly in view of AIDS related self organisation and behaviour change among homosexual groups. We remain vigilant about
647
mortality within the heterosexual population, but there is a serious danger in interpreting every morbid phenomenon in our lives today in the context of HIV. That we have reached the stage where it is fashionable to blame most deaths in the population on HIV or AIDS may distort the perception of what the real problem is. JAMES EDEH Division of Addictive Behaviour, St George's Hospital Medical School, London SW17 ORE 1 Aldous J, Hickman M, Ellam A, Gazzard B, Hargreaves S. Impact of HIV infection on mortality in young men in a London health authority. BMJ 1992;305:219-21. (25 July.) 2 McCormick A. Excess morbidity associated with the HIV epidemic in England and Wales. BMJ7 1991;302:1375-6. 3 Mienties GH, Van Amei;den EJ, Van den Hoek AJAR, Continho RA. Increasing morbidity without rise in non-AIDS mortality among HIV infected intravenous drug users in Amsterdam. AIDS 1992;6:207-12. 4 Selwyn PA, Hartel D, Wasserman W, Drucker E. Impact of AIDS epidemic on morbidity and mortality among intravenous drug users in a New York city methadone maintenance programme; AmJ7 Public Health 1989;79:1358-62.
EDITOR,-John Aldous and colleagues report the mortality in men with HIV infection in Riverside Health Authority; eight of the 213 men who died committed suicide.' Suicidal behaviour in people with HIV infection has been the subject of considerable interest, and Marzuk et al argued that suicide is considerably increased in this group,2 although there are complex methodological problems in establishing its prevalence accurately. ' We have identified six completed suicides in people with HIV infection over two years in Riverside Health Authority. Four of the patients had a history of psychiatric illness: in two this preceded the diagnosis of HIV infection and included a history of deliberate self harm. The two patients without a history of psychiatric illness were in an advanced stage of disease, and one had planned his death for over a year. The mode of suicide was predominantly by jumping-usually a rare form of suicide-but suicides by overdose may be underreported in patients with AIDS. We also looked at deliberate self harm and found 23 cases in people positive for HIV over two years; HIV infection played a part in most cases.4 In recent years a significant increase in suicides in young men has been noted, and HIV infection may well be a contributing factor.5 Staff who care for patients with HIV infection should be aware of the risk of suicidal behaviour and should refer to psychiatric services patients with suicidal ideas, particularly those with a history of psychiatric illness or deliberate self harm, especially as the disease becomes more advanced. KATHRYN PUGH JOSE CATALAN Academic and Clinical Department of Psychological Medicine, Charing Cross and Westminster Medical School, London SWIP 2NS 1 Aldous J, Hickman M, Ellam A, Gazzard B, Hargreaves S. Impact of HIV infection in young men in a London health
authority. BMJ 1992;305:219-21. (25 July.) 2 Marzuk P, Tierney H, Tardiff K, Gross EM, Morgan EB, Hsu M-A, et al. Increased risk of suicide in patients with AIDS.
JAMA 1988;259:1333-7. 3 Marzuk P. Suicidal behaviour and HIV illnesses. International Review ofPsychiatry 1991;3:367. 4 Catalan J. Deliberate self-harm in HIV disease. Research into clinical and ethical aspects. In: Beskow JE, Bellini M, Sampaio Faria JG, Kerkhof AD, eds. Report on the WHO consultation, Bologna, September 1990, HIV and AIDS related suicidal behaviour. Bologna: Monduzzi Editore, 1991:61-8. S Hawton K. Suicide in young men. BMJ 1992;304:100O.
Treatment of natal cleft sinus EDITOR,-Patients who have had recurrent treatment for persistent pilonidal sinus' 2 are sometimes referred, in the last resort, to plastic surgeons. This is done in the hope that the plastic surgeon can either create skin flaps to obliterate the natal cleft 648
or perform expert skin grafting which will remove the hair follicles in and around the natal cleft. In a few patients who have tightly opposed, muscular buttocks friction between the buttocks may play an important part in the persistence of pilonidal disease. I carried out a combination of liposuction and limited skin grafting in such a patient. Liposuction of the medial aspect of the buttocks reduced the sharp angle of the cleft -and eliminated the tendency of one buttock to rub against the other. Excision of the scarred area in the base of the cleft and use of a quilted meshed skin graft completed the procedure. This was followed by primary healing, and the patient has been free of any pilonidal problems for the past six months. He had suffered recurrent disease every month or so for the previous four years despite innumerable operations. I hesitate to make any claims for this procedure based merely on one case, but I thought it worth reporting because other surgeons might be interested in trying it. I am unlikely to be able to build up a series because these difficult recurrent cases are relatively rare. H P HENDERSON
Department of Plastic Surgery, Leicester Royal Infirmary, Leicester LEI 5WW 1 Correspondence. Treatment of natal cleft sinus. BMJ 1992;305: 311-2. (1 August.) 2 Khawaja HT, Bryan S, Weaver PC. Treatment of natal cleft sinus: a prospective clinical and economic evaluation. BMJ 1992;304:1282-3. (16 May.)
Over the counter candidiasis
trlatment for
EDITOR,-Susan Mitchell and Caroline Bradbeer provide an interesting interpretation of the presentation patterns for vaginal candidiasis that may follow the change in legal status for intravaginal imidazoles.' The recommendation of the Committee on Safety of Medicines to exempt such antifungals from prescription only control recognises the efficacy of the products and the role of community pharmacy in primary health care and resolves the distinction between infection in men for whom over the counter preparations have been available for some time, and that in women. Because of the complexity of the differential diagnosis in women presenting with "thrush," over the counter imidazoles are available only to clients experiencing a suspected recurrence of candidiasis. The pharmaceutical press has made strenuous efforts to ensure that pharmacists are aware of the circumstances for medical referral, describing 11 scenarios in which over the counter treatment would be inappropriate.2 Accurate diagnosis of vaginal candidiasis will therefore necessitate assimilation of potentially embarrassing information from a client who may expect to receive her over the counter preparation automatically on request. This requires a private area for counselling and competency in communication skills on the part of the pharmacist. The importance of communication skills forms an integral part of undergraduate education for pharmacists.3 The most problematic issue, however, is obtaining a history of sexually transmitted disease or exposure to a partner with sexually transmitted disease. A client may presume that she has a recurrence of candidiasis when the aetiological agent is in fact a sexually transmitted pathogen, such as herpes simplex virus or Trichomonas vaginalis. The danger is that a woman may not disclose, or even be aware of, exposure to a sexually transmitted disease. Exempting vaginal imidazoles from prescription only control may be of considerable benefit to women who can recognise their recurrent symptoms and can afford to pay for their treatment over the counter. The caveat is clearly that wider
availability may lead to more cases of treatment failure and undiagnosed sexually transmitted diseases. Ultimately, medical practitioners will detect whether a change in the pattern of presentation of vaginal candidiasis has indeed occurred. GRETA FORSTER DAVID G WEBB
Royal London Trust, Royal London Hospital, London El 1BB I Mitchell S, Bradbeer C. Over the counter treatment for candidiasis. BMJ 1992;304:1648. (27 June.) 2 Li Wan Po A. Treatment of vaginal candidiasis. Pharmaceutical 3rournal 1992;249: i-is. 3 Committee of Inquiry. Pharmacy: a report to the Nuffield Foundatnm. London: Nuffield Foundation, 1986:85-104.
Screening for depression in elderly patients EDITOR,-In describing use of a new screening instrument for depression in elderly patients' Fiona Adshead and colleagues address a problem with self rating scales2": many elderly patients are unable to use them and require help. This difficulty is illustrated by the use of a mixed mode of presentation of the geriatric depression scale in the study (self administered or staff assisted). This is clinically relevant as a staff assisted scale is associated with a lower score than the self administered form.2 Two points are worthy of comment. The brief assessment schedule depression cards (BASDEC) are printed in a large typeface. If all patients could read the BASDEC but some could not read the geriatric depression scale this suggests that the typefaces were of different sizes: could this partly explain the patients' modest preference for the BASDEC? Although the BASDEC took less time to administer in Adshead and colleagues' study, a shortened form of the geriatric depression scale is under evaluation4 and may reduce the difference in time required to perform the measures. The geriatric depression scale has an advantage over the BASDEC in that it has been shown to be useful in people with mild to moderate cognitive impairment.' If the BASDEC can be shown to share this quality its utility to clinicians will be further enhanced. DESMOND O'NEILL
Elderly Services, Selly Oak Hospital, Birmingham B29 6JD I Adshead F, Day Cody D, Pitt B. BASDEC: a novel screening instrument for depression in elderly medical inpatients. BMJ 1992;305:397. (15 August.) 2 O'Neill D, Rice I, Blake P, Walsh JB, Coakley D. The geriatric depression scale: rater-administered or self-administered? International Journal of Geriatric Psychiatry 1992;7:511-5. 3 Toner J, Gurland B, Teresi J. Comparison of self-administered and rater-administered methods of assessing levels of severity of depression in elderly patients. J Gerontol 1988;43:136-40. 4 Yesavage JA. Geriatric depression scale. Psychopharmacol Bull 1988;24:709-1 1. 5 O'Riordan T, Hayes J, O'Neill D, Shelley R, Walsh JB, Coakley D. The effect of mild to moderate dementia on the geriatric depression scale and on the general health questionnaire in the hospitalized elderly. Age Ageing 1990;19:57-61.
Human milk banks EDITOR,-In an editorial on human milk banks D P Davies states, "At 9 months the mean developmental quotient was shown to be significantly lower in infants fed donor breast milk than in those fed a preterm formula,"' citing the work of Lucas et al.I Although this is widely believed, it is a view not entirely borne out by the data. According to Lucas et al's paper, at 9 months of age the mean (SD) developmental quotient of the babies fed banked human milk was 97-2 (8 5) and that of the babies fed preterm formula 98-2 (11.0).2 Thus the difference between the groups was not BMJ
VOLUME 305
12 SEPTEMBER 1992
