priority after initial alarm, inadequate assessment and care
`plans, avoidance of confrontation, little cooperation between
`agencies,
`ineffective interventions, and a lack of policies
`despite general agreement that they were needed and that
`elder abuse was important.“
`Fisk has suggested that physicians and psychiatrists for
`elderly people (as well as social workers, primary health care
`teams, and the police) are especially well placed to detect elder
`abuse. A lower threshold for suspicion, despite the abused
`person’s denials, may be required than has prevailed up till
`now. Once abuse has been confirmed the priorities for action
`are, firstly, the safety of the victim; secondly, the physical and
`psychological health of the victim; thirdly, the physical and
`psychological health of the abuser; and, fourthly, a plan to
`prevent recurrence of the abuse. Preventive measures might
`include information packs on caring for elderly people;
`support groups—self help or supervised; financial support for
`carers; physical, psychological, and financial support for
`elderly people; and specialist teams (from health authorities
`and social services) to detect, intervene in, and prevent elder
`abuse. Legislation may be needed to provide for mandatory
`reporting of abuse and protection for vulnerable elderly
`people.
`Some may be sanguine about the effects of the implementa-
`
`tion of the white paper Caringfor People next April. " More are
`deeply concerned that
`there will be a period of chaotic
`struggling to assess priorities for scant resources—when the
`needs of many old people and their carers will not be met. An
`audit of elder abuse, using the baseline now offered by Ogg
`and Bennett, should be required by potential purchasers. It
`may help to give some political ammunition to those who
`insist that worthy intentions must be seen to work.
`BRICE PITT
`
`Professor of the Psychiatry of Old Age,
`St Mary’s and the Royal Postgraduate Medical Schools,
`London W10 6DZ
`
`l Baker AA. Granny battering. Modem Geriatrics 1975;S(August):20-4.
`2 Eastman M. Old age abuse. Mitcham: Age Concern, 1984.
`3 Lau EH, Kosberg JI. Abuse of the elderly by informal care providers. Aging 1979 Sept/Oct:l0-5.
`4 Block MR, Sinott JD. The battered elder syndrmne: an exploratory study. Baltimore: Center on Aging,
`University of Maryland, 1979.
`S Fisk]. Abuse of the elderly. In: Jacoby R, Oppenheimer C, eds. Psychiatry in the elderly. Oxford:
`Oxford University Press, 1991.
`6 US Congress Select Committee on Aging, Subcommittee on Human Services. Elder abuse: an
`examination ofa hidden problem. Washington, DC: Government Printing Office, 1981.
`7 Wolf RS. Abuse ofthe elderly: ten years later._7Am Gerialr Soc 1988;315:756-62.
`8 Homer AC, Giulleard C. Abuse of elderly people by their carers. BM] 1990;301:359-62.
`9 Pillemer K. The dangers of dependency: new findings on domestic violence against the elderly.
`Social Problems 1985;33:146-523.
`10 Ogg J, Bennett G. Elder abuse in Britain. BM] 1992;305:998-9.
`11 Social Services Inspectorate of the Department ofHealth. Confronlingelderabuse. London: HMSO,
`1992.
`12 Secretaries of State for England and Wales. Caringjbr people. London: HMSO, 1990.
`
`Corticosteroids in advanced cancer
`
`If they are not working stop them
`
`their specific and
`Systemic corticosteroids are used for
`general effects in patients with advanced cancer.‘ For their
`specific anti—inflammatory effects they are used in raised
`intracranial pressure, compression of the spinal cord, and
`obstruction of the superior vena cava or other hollow organ.“
`In addition, in one third of elderly patients with breast cancer
`corticosteroids result in regression or cessation of progression
`of their cancer for as long as one year? Patients with prostatic
`cancer may obtain similar benefit.‘
`The general effects of corticosteroids include improved
`appetite, mood, and strength. In a controlled trial of methyl-
`prednisolone 32 mg a day for two weeks in 40 patients with
`terminal cancer, appetite increased in 77%, mood in 71%, and
`activity in 68%] Consumption of analgesics decreased in
`71%. All patients continued taking methylprednisolone for a
`further 20 days; most measures had worsened by the end of
`this time, although there was still significant benefit compared
`with baseline values. This worsening could reflect either the
`loss of effect of the drug or the progression of disease, or both.
`Another controlled trial also found that
`the effects
`
`diminished with time.*‘ In this trial dexamethasone 3 mg and
`6 mg daily were compared with placebo——the higher dose
`being comparable with methylprednisolone 32 mg. Subjec-
`tive improvement in appetite and strength was noted after two
`weeks but had disappeared by four weeks.
`The benefits seen in time limited trials are much better than
`
`those reported in this issue of the journal by Needham et al
`(p 999)? These authors surveyed corticosteroid use by 100
`patients admitted to a hospice for terminal care. On admission
`33 patients were taking corticosteroids, and seven had done so
`in the past. Of the 28 patients who completed the question-
`naire, only eight said that they had benefited; nine were
`undecided and 11 said that they had not benefited. Five of the
`11 who said that they had not benefited had started treatment
`more than one month before; among those who were
`
`undecided was a woman who had been taking prednisolone
`30 mg daily for two years. Patients who had taken corti-
`costeroids were more likely to complain of anorexia, weight
`loss, or weakness than those who had not.
`Needham et al initiated their survey after three patients had
`been admitted within a month with severe adverse effects
`from corticosteroids (proximal myopathy, excessive weight
`gain, and skin changes). Other reports have also highlighted
`proximal myopathy and, less commonly, avascular necrosis of
`bone.” " Furthermore,
`in a prospective survey of several
`hundred patients with advanced cancer who received corti-
`costeroids nearly one third developed oral candidiasis,
`accounting for four fifths of all such cases in that unit.‘ One in
`10 experienced hypomania, agitation, hyperkinesia, or
`insomnia, and in one in 20 treatment with corticosteroids was
`stopped because of unacceptable adverse effects.‘
`Peptic ulceration may occur," although the concurrent
`use of non—steroidal anti—inflammatory drugs may be respon-
`sible.” Necropsy studies in patients with cancer have shown
`that death may be precipitated by complications of peptic
`ulceration (such as bleeding or perforation) in 5% of patients
`receiving corticosteroids compared with 1% of others.”
`Although a risk of this order is acceptable in patients with a
`specific need for corticosteroids, it cannot be ignored in other
`circumstances.
`
`It is disturbing, therefore, that Needham ez al found that
`more than half of the patients receiving corticosteroids did not
`know why they were taking the drug or how long they were
`meant to continue taking it. More than two thirds did not have
`a steroid card, and a similar proportion did not know that long
`term corticosteroid treatment should not be stopped suddenly.
`If this sample is representative it seems that, once started,
`corticosteroids are stopped only rarely and that the impact of
`the treatment is not adequately monitored. Needham et al
`conclude that many doctors do not exercise the same care with
`
`BM] VOLUME 305
`
`24 OCTOBER 1992
`
`Amerigen Exhibit 1108
`Ameri gen Exhibit 1108
`Amerigen v. Janssen IPR2016-00286
`Amerigen v. Janssen IPR20 l 6-00286
`
`969
`
`

`
`corticosteroids in patients with advanced cancer as they do in
`patients with other conditions.
`As an essential safeguard, therefore, doctors should state
`clearly in their notes why a corticosteroid is being prescribed
`and tell their patients why. Except where the aim is to control
`the tumour, the corticosteroid should be prescribed initially
`on a trial basis for no more than a week: the chances of
`
`obtaining a better response after this time are poor? Treat-
`ment should be continued only if subjective or objective
`benefit occurs. Using corticosteroids for their general effects
`(those on appetite, mood, and strength) should be avoided
`as far as possible in anxious patients and in patients with
`diabetes because of the risk of worsening the associated
`condition.
`
`Stopping corticosteroids abruptly after a week is safe if no
`more than prednisolone 40 mg a day or its equivalent
`(methylprednisolone 32 mg or dexamethasone 6 mg a day),
`has been taken." Short courses of larger doses and longer
`courses of lower doses will suppress the hypothalamic-
`pituitary-adrenal axis for prolonged periods, and doses must
`be tapered off over several days or weeks according to
`circumstances.
`
`Needham et al also point out that advanced cancer and
`polypharmacy tend to go hand in hand. Stopping drugs that
`are not yielding benefit will therefore help to ease the patients’
`burden of tablet taking and may improve compliance with
`other drugs. Furthermore, because the biological half lives of
`corticosteroids are relatively long (for example, 18-36 hours
`for prednisolone and 36-54 hours for methylprednisolone)”
`they should be taken once a day unless the number of tablets
`precludes this.
`An important unresolved question is the choice of dose; in
`controlled trials to treat anorexia the dose has varied between
`
`the equivalent of 15 mg and 40 mg of prednisolone a day.7 " '“ '7
`It may be better to start with a relatively high dose in order not
`to miss an effect of treatment and then to reduce to a lower
`
`maintenance dose if treatment is to continue beyond seven
`days. In patients receiving anticonvulsants such as phenytoin
`
`and phenobarbitone, starting with an even higher dose may be
`advisable because these drugs enhance the metabolism of
`corticosteroids. "‘
`
`Finally, well documented alternatives for treating anorexia
`exist. For example, many patients benefit from megestrol
`acetate, and the effect is still detectable after two months.’”"
`Megestrol is, however, considerably more expensive. Given
`the 50% response to placebo,” the best initial step may well be
`dietary advice with or without multivitamin tablets.
`ROBERT TWYCROSS
`
`Macmillan Clinical Reader in Palliative Medicine,
`Oxford University,
`Oxford OX1 2]D
`
`Hanks GW, Trueman T, Twycross RG. Corticosteroids in terminal cancer: a prospective analysis
`ofcurrent practice. l’o.ttgn1d Med] 1983;59:702-6.
`Gilbert RW, Kim JH, Posner JB. Epidural spinal cord compression from metastatic tumours:
`diagnosis and treatment. Amt Neural 1978;3:4-0-51.
`Carter RL,1’ittam MR, Tanner NSB. Pain and dysphagia in patients with squamous carcinomas of
`the head and neck: the role of perineural spread.] R Soc Med 1982;75:598-606.
`Flombaum CD, Schroy 1’, Watson R, Vanamee 1’. Treatment of acute obstructive renal failure with
`high-dose methylprednisolone. Arch Intern Med 1986;146:581-61.
`_
`Minton MJ, Knight RK, Rubens RD, Hayward JL. Corticosteroids for elderly patients with breast
`cancer. Cancer 1981;48:883--7.
`Tannock J, Gospodarawicz M, Meakin W, Panzarella T, Stewart L, Rider W. Treatment of
`metastatic prostatie cancer with low-dose prednisolone: evaluation of pain and qua1it_\' of life as
`pragmatic indices of response. ]' Clin Oncol 1989;7:590-7.
`Bruera E, Roca E, Cedaro L, Carraro S, Chacon R. Action of oral methylprednisolone in terminal
`cancer patients: a prospective randomized double-blind stttdy. (Iurircr Treat Rep l985',69:7S 1 -4.
`Moertel CG, Schutt AJ, Reitemeier RJ. Hahn RG. Cortieosteroid therapy of prcterminal
`gastrointestinal cancer. (lancer 1974,33: 1607-9.
`Needham PR, Daley AG, Lennard RF. Steroids in advanced cancer: a survey of current practice.
`BM] 1992 3051999.
`Weissman DE, Dufer D, Vogel V, Abelofl'MD. Corticosteroid toxicity in netiro-oncology patients.
`_7Neurourim1 l987‘,5: 125-8.
`(lapell H. Selected side-effects: steroid therapy and osteonecrosis. l’rcsrnbcr.=']'oumuI 1993;32:32-
`4.
`Schell HW. This risk of adrenal corticosteroid therapy in far-advanced cancer. Ant _7 Med Sci
`1960;252:641-9.
`Gusltindo M, Titohello A. Steroid ulcers: a myth revisited. RM] 1992;304:655-6.
`Byyny RL. Withdrawal from glucocorticoid therapy. N I:'ng[_7Med 1976;295:311-2.
`Speight'1‘M, ed. Azwry’s drug treatment. 3rd ed. Edinburgh: Churchill Livingstone. 1987:5624.
`Wi|1oxJC, Corr J, Shaw J, Richardson M, Calman KC. 1)rennan M. Prednisoloiie as an appetite
`stimulant in patients with cancer. BM] 198-1:288:27.
`17 Twyeross RG, Guppy D. Prednisolone in terminal breast and bronchogenic cancer. [’mi'tttiwwr
`1985;229:574).
`18 Gatnbertoglio JG. Corticosteroids and anticonvulsants. I)ru;,t Interactions Newsletter 1983;12:5S~8.
`19 Tchekmedyian NS, Tait N, Moody M, Greco FA, Aisner J. Appetite stimulation with megestrol
`acetate in cachectic cancer patients. Serrtin Ortwl 1986,1337-43s.
`20 Loprinzi CL, Ellison NM, Schaid 1)J, Krook JE, Altmann LM, Duse AM, Controlled trial of
`megestrol acetate for
`the treatment of cancer anorexia and caehexia. ] Natl Cartrer Inst
`l990;82:1l27-32.
`
`Pet birds and lung cancer
`
`Smoking is still a confounder
`
`Cigarette smoking accounts for about 80% of Britain’s 40 000
`deaths from lung cancer each year.‘ The contribution of other
`causes of deaths from lung cancer in the general population is
`thus small. It may, however, be increasing} 3 and natural
`radiation, occupational exposures, dietary intake of vitamin
`A, and familial predisposition have all been implicated.”
`A more recent hypothesis, advanced and tested by Holst et al
`in 1988,“ is that some cases of lung cancer may be caused by
`exposure to pet birds. This hypothesis is independently tested
`in two studies published in this issue (pp 986-9, 989-92)." '0
`The original study by Holst et al compared 49 patients with
`lung cancer with 98 randomly selected community controls?
`With adjustment for smoking the relative risk of lung cancer
`from exposure to any pet bird five to 14 years before diagnosis
`was estimated at 6'7 (95% confidence interval 2-2 to 20-0).
`The two studies published in this issue are both larger but
`arrive at smaller estimates of risk: Kohlmeier ez al report an
`adjusted odds ratio of 2-12, which was significant,” and
`Gardiner et al an unadjusted value of 1-58, which was
`not.” Gardiner et (11, however, also analysed the effects of
`exposure to individual bird species and,
`though cautious
`
`about the validity of this subgroup analysis, found a signifi-
`cant fourfold increase in risk associated with exposure to
`pigeons. Thus there are now at
`least
`three independent
`reports describing an increased risk of cancer associated with
`exposure to pet birds. How likely is it that these findings are
`valid?
`
`This will depend on the extent to which the investigators
`have eliminated bias and controlled for confounding in their
`study design and analysis. In case-control studies bias arises
`principally from the methods by which cases and controls are
`selected and exposure is measured, and once present it is
`difficult to remove. Confounding by factors that are related to
`both the exposure and the disease can be dealt with in the
`analysis so long as the confounding exposure is recognised and
`measured. The main potential source of confounding in
`studies of the aetiology of lung cancer is smoking, and,
`because both smoking and the keeping of pet birds tend to
`occur in lower socioeconomic groups, confounding of these
`effects is inherently likely.
`Controlling successfully for confounding by smoking
`requires either that cases and controls are closely matched for
`
`970
`
`BM] VOLUME 305
`
`24 OCTOBER 1992