BMJ No 7115 Volume 315
Education and debate Saturday 25 October 1997
Optimising drug treatment for elderly people: the prescribing cascade
Paula A Rochon, Jerry H Gurwitz
| Summary points |
| The "prescribing cascade" cascade begins when an adverse drug reaction is misinterpreted as a new medical condition |
| Another drug is then prescribed, and the patient is placed at risk of developing additional adverse effects relating to this potentially unnecessary treatment |
| To prevent the prescribing cascade, doctors should always consider any new signs and symptoms as a possible consequence of current drug treatment |
| Before any new drug treatment is started, the need for the drug should be re-evaluated and a non-drug treatment should be considered |
| If drug treatment is necessary the lowest feasible dose of the drug should be used and alternative drugs with fewer adverse effects considered |
The most frequent medical intervention performed by a doctor is the writing of a prescription. Because chronic illness increases with advancing age, older people are more likely to have conditions that require drug treatment. Advanced age, frailty, and increased use of drugs are all factors that contribute to a patient's risk of developing a drug related problem. As many as 28% of hospital admissions in the United States of older people are as a result of drug related problems,(1) up to 70% of which are attributed to adverse reactions to drugs.(1) Creating optimal drug regimens that meet the complex needs of elderly people requires thought and careful planning.
Inappropriate prescribing is expensive. In a recent study the
costs of preventable adverse drug events - namely, injury resulting from
a drug related medical intervention - occurring during a stay in
hospital were estimated to be $2.8m (£1.75) annually in two large
American teaching hospitals.(2) The national cost of
managing the consequences of inappropriate prescribing remains
uncertain. One estimate has put the annual cost of drug related
morbidity and mortality in outpatient clinics at $76.6bn.(3)
Drug related morbidity and mortality is an important area to target
both to improve the quality of medical care for elderly p A prescriber can do little to modify age related physiological changes
in trying to minimise the likelihood that an older person will develop
an adverse drug reaction. However, when assessing a patient who is
already taking drugs, a doctor should always consider the development
of any new signs and symptoms as a possible consequence of the
patient's drug treatment. This article will focus on an
under-recognised, and largely preventable drug related problem that we
have termed the "prescribing cascade."
(4)
The prescribing cascade begins when an adverse drug
reaction is misinterpreted as a new medical condition. A drug is
prescribed and an adverse drug effect occurs that is mistakenly
diagnosed as a new medical condition. A new drug is prescribed, and the
patient is placed at risk of developing additional adverse effects
relating to this potentially unnecessary treatment (fig 1). Drawing
prescribers' attention to this disturbing sequence of events may be an
important step in minimising the occurrence of preventable adverse drug
events associated with suboptimal prescribing
decisions.
Non-steroidal anti-inflammatory drugs are among the most
frequently prescribed drugs to elderly patients. An estimated 10-15%
of people aged 65 years or older are prescribed such
drugs.(5) Their anti-inflammatory properties seem to result
from their ability to inhibit cyclo-oxygenase, a critical enzyme in the
biosynthesis of prostaglandins.(6) Good evidence exists to
suggest that prostaglandins have an important role in the modulation of
two major determinants of blood pressure: vasoconstriction of
arteriolar smooth muscle and control of extracellular fluid volume. The
effects of non-steroidal anti-inflammatory drugs are most prominent in
patients with existing hypertension.(7)
The high prevalence of use of non-steroidal anti-inflammatory drugs
among older people emphasises the importance of studying the clinical
impact of these drugs on blood pressure in elderly people. To determine
whether there is an increased risk associated with starting
antihypertensive treatment in older people prescribed non-steroidal
anti-inflammatory drugs (table), a case-control study was performed
involving patients enrolled in the New Jersey Medicaid programme who
were aged 65 years or older.(8) Over 9,000 patients who were
newly started on an antihypertensive drug were compared with a similar
number of randomly selected control patients. The adjusted odds ratio
for starting antihypertensive treatment for recent users of
non-steroidal anti-inflammatory drugs compared with non-users was 1.66
(95% confidence interval, 1.54 to 1.80). The odds ratio increased with
increasing daily dose of the anti-inflammatory drug: compared with
non-users, the adjusted odds ratio for users of low average daily doses
was 1.55 (1.38 to 1.74), for medium dose users was 1.64 (1.44 to 1.87),
and for high dose users was 1.82 (1.62 to 2.05). The conclusion was
that the use of non-steroidal anti-inflammatory drugs may increase the
risk associated with starting antihypertensive treatment in older
people. Given the high prevalence of use of non-steroidal
anti-inflammatory drugs by elderly people, this association could have
important public health implications for the care of older
patients.
This relation also shows a clear sequence of events where the use of
one treatment leads to the start of a second that might have been
avoided. Based on the findings of numerous epidemiological and clinical
studies that have characterised the adverse consequences of use of
non-steroidal anti-inflammatory drugs in older people, recommendations
have been made to avoid using these agents when clinically
feasible.(5) As with other drugs prescribed to elderly
patients, the most prudent approach is to limit prescribing
non-steroidal anti-inflammatory drugs to situations in which benefits
clearly outweigh risks and to use them only after potentially safer
alternatives have been tried.(11) Because of the multiple
adverse effects attributable to these drugs, for some indications (such
as osteoarthritis) treatments such as acetaminophen, gentle
exercise, and weight reduction may be effective
alternatives.(12-14) When treatment with non-steroidal
anti-inflammatory drugs is necessary, the lowest feasible dose should
be used for the shortest time required to achieve the desired effect.
Furthermore, if patients require extended treatment with non-steroidal
anti-inflammatory drugs, periodic monitoring of blood pressure is
warranted, as such treatment may contribute to newly detected rises in
pressure. With recognition of this association between non-steroidal
anti-inflammatory drugs and rises in blood pressure, the starting or
intensifying of antihypertensive treatment may be avoided.
The development of some degree of hyperuricaemia is a well
documented side effect of treatment with thiazide
diuretics.(15-18) Population based studies have shown an
association between hyperuricaemia and the development of gout. For
example, data from the Framingham study document a cumulative incidence
of gout of 36% over 12 years in patients with serum uric acid
concentrations g476 |gmmol/l, compared with less than 3% in those
with lower concentrations.(19) The occurrence of
hyperuricaemia that has been induced by thiazide diuretics raises some
important issues about the use of these diuretics in elderly people.
Ample data show the efficacy of these agents in treating hypertension
in elderly patients and in preventing major sequelae such as
stroke - data that are absent for many other commonly used
antihypertensive drugs.(20-22) The impact of thiazide
diuretics on serum uric acid concentrations, however, raises questions
about whether this treatment may precipitate the use of additional
drugs.
This question was recently examined in a retrospective cohort study of
9,249 patients enrolled in the New Jersey Medicaid programme aged 65 or
older who had been started on a variety of antihypertensive
agents.(9) None of the patients in the cohort had previously
used treatment for gout (allopurinol, colchicine, or uricosuric agent).
Follow up extended for up to two years, and exposure to
antihypertensive drugs was characterised over this period according to
the following categories: thiazide diuretics alone; non-thiazide
antihypertensive drugs alone; thiazide diuretics combined with any
non-thiazide antihypertensive drug; and no use of antihypertensive
drugs. The relative risk for starting treatment for gout was 1.00 (0.65
to 1.53) for non-thiazide antihypertensive drugs alone, 1.99 (1.21 to
3.26) for thiazide diuretics alone, and 2.29 (1.55 to 3.37) for
thiazide diuretics combined with any non-thiazide drug. Risk for
starting treatment for gout was significantly increased for thiazide
doses of 25 mg/day (in hydrochlorothiazide equivalents) or more; no
significant increase in risk was seen for lower doses. It was concluded
that the use of thiazide diuretics in doses of 25 mg/day and over was
associated with a significantly increased risk for starting treatment
for gout, relative to antihypertensive regimens that did not include
the use of a thiazide diuretic.
Considerable evidence supports the efficacy of low doses of thiazide
diuretics in the treatment of hypertension in elderly
people.(13-15) The dose-response relations found in this
study support the use of lower doses of thiazide diuretics when
treatment is indicated. Although the recommendations of the United
States's joint national committee on detection, evaluation and
treatment of high blood pressure suggest starting antihypertensive
treatment at low doses in all patients,(17) thiazide
diuretics are commonly started at doses that extend well beyond the low
dose range. Low doses of thiazide diuretics - for example, 12.5 mg of
hydrochlorothiazide - often produce as large an antihypertensive effect
as larger doses, with a reduced risk of metabolic abnormalities. In
fact, evidence exists that a dose of hydrochlorothiazide as low as
6.25 mg can be as efficacious in treating hypertension in many older
patients, when combined with a low dose of another antihypertensive
drug.(23-25) When hyperuricaemia does occur during treatment
with a thiazide diuretic, clinicians should bear in mind that
asymptomatic hyperuricaemia alone does not warrant treatment.
A case-control study, again involving patients enrolled in the New
Jersey Medicaid programme aged 65 years or older, showed that patients
taking metoclopramide were three times more likely to begin using a
drug containing levodopa than patients not taking metoclopramide (odds
ratio 3.09 (2.25 to 4.26)).(10) The risk increased with
increasing daily metoclopramide dose: the odds ratio was 1.19 (0.50 to
2.81) for 10 mg/day and over, 3.33 (1.98 to 5.58) for >10 mg/day to
20 mg/day, and 5.25 (1.16 to 8.50) for >20 mg/day (fig 2). In
summary, metoclopramide confers an increased risk of starting treatment
generally reserved for managing idiopathic Parkinson's disease. Such
multiple prescribing may represent the misdiagnosis of Parkinson's
disease in patients with drug induced parkinsonian symptoms.
The prescribing cascade, whereby additional drug treatment is
started after a patient develops an adverse reaction to a drug, is
largely preventable by carefully considering whether any new medical
condition might be the result of an existing drug treatment. The
prescription of a new drug specifically to treat an adverse drug effect
should be considered the choice of last resort in the care of older
patients. More prudent strategies include:
Division of
Geriatric Medicine, Meyers Primary Care Institute,
Correspondence to: Dr Rochon
References
1 Col N, Fanale J E, Kronholm P. The role of medication
non-compliance and adverse drug reactions in hospitalizations of the
elderly. Arch Int Med 1990;150:841-5.
2 Bates D , Spell N, Cullen D J, Burdick E, Laird N, Petersen L A,
et al. The cost of adverse drug events in hospitalized patients.
JAMA 1997;277:307-11.
3 Johnson J A, Bootman J L. Drug-related morbidity and mortality:
A cost-of-illness model. Arch Intern Med
1995;155:1949-56.
4 Rochon P A, Gurwitz J H. Drug therapy. Lancet
1995;346:32-6.
5 Ray W A, Griffin M R, Avorn J. Evaluating drugs after their
approval for clinical use. N Engl J Med
1993;329:2029-32.
6 Meconi M, Taylor L, Martin B, Polgar P. A review:
prostaglandins, aging, and blood vessels. J Am Geriatr
Soc 1987;35:239-47.
7 Pope J E, Anderson J J, Felson D T. A meta-analysis of the
effects of nonsteroidal anti-inflammatory drugs on blood pressure.
Arch Intern Med 1993;153:477-84.
8 Gurwitz J H, Avorn J, Bohn R L, Glynn R J, Monane M, Mogun H.
Initiation of antihypertensive treatment during nonsteroidal
anti-inflammatory drug therapy. JAMA 1994;272:781-6.
9 Gurwitz J H, Kalish S C, Bohn R L, Glynn R J, Monane M, Mogun H,
et al. Thiazide diuretics and the initiation of anti-gout therapy.
J Clin Epidemiol 1997;50:953-9.
10 Avorn J, Gurwitz J H, Bohn R L, Mogun H, Monane M, Walker A.
Increased incidence of levodopa therapy following metoclopramide use.
JAMA 1995;274:1780-2.
11 Solomon D H, Gurwitz J H. Toxicity of nonsteroidal
anti-inflammatory drugs in the elderly: Is advanced age a risk factor?
Am J Med 1997;102:208-15.
12 Bradley J D, Brandt K D, Katz B P, Kalasinski L A, Ryan S I.
Comparison of an anti-inflammatory dose of ibuprofen, an analgesic does
of ibuprofen, and acetaminophen in the treatment of patients
with osteoarthritis of the knee. N Engl J Med
1991;325:87-91.
13 Kovar P A, Allegrante J P, MacKenzie C R, Peterson M G E, Gutin B,
Charlson M E. Supervised fitness walking in patients with osteoarthritis
of the knee: a randomized, controlled trial. Ann Intern
Med 1992;116:529-34.
14 Felson D T, Zhang Y, Anthony J M, Naimark A, Anderson J J. Weight
loss reduces the risk for symptomatic knee osteoarthritis in women: the
Framingham study. Ann Intern Med 1992;116:535-9.
15 Kelly W N, Schumacher H R Jr, Gout. In: Kelley W N, Harris E D Jr,
Ruddy S, Sledge CB, eds. Textbook of rheumatology. 4th
ed. Philadelphia: W B Saunders, 1993:1291-336.
16 Fletcher A E. Adverse treatment effects in the trial of the
European working party on high blood pressure in the elderly. Am
J Med 1991;90:42-3S.
17 The Fifth Report of the Joint National Committee on Detection,
Evaluation, and Treatment of High Blood Pressure (JNCV). Arch
Intern Med 1993;153:154-83.
18 Langford H G, Blaufox M D, Borhani N O, Curb J D, Molteni A,
Schneider K A, et al. Is thiazide-produced uric acid elevation harmful?
Arch Intern Med 1987;147:645-9.
19 Hall A P, Barry P E, Dawber T R, McNamara P M. Epidemiology of gout
and hyperuricemia: a long-term population study. Am J
Med 1967;42:27-37.
20 SHEP Cooperative Research Group. Prevention of stroke by
antihypertensive drug treatment in older persons with isolated systolic
hypertension. Final results of the systolic hypertension in the elderly
program (SHEP). J Am Med Assoc 1991;265:3255-64.
21 Amery A, Birkenhager W, Brixko P, Bulpitt C, Clement D,
Deruyttere M, et al. Mortality and morbidity results from the European
working party on high blood pressure in the elderly trial.
Lancet 1985;i:1349-54.
22 Dalhof B, Lindholm L H, Hansson L, Schersten B, Ekbom T, Wester
P O. Morbidity and mortality in the Swedish trial in old patients with
hypertension (STOP-hypertension). Lancet
1991;338:1281-5.
23 Frishman W H, Bryzinski B S, Coulson L R, DeQuattro V L, Vlachakis
N D, Mroczek W J, et al. A multifactorial trial design to assess
combination therapy in hypertension. Arch Intern Med
1994;154:1461-8.
24 Ziegler M G, Lernhardt E, Solt-Buzsaki V. Dose response to
hydrochlorothiazide in hypertensives receiving a calcium channel
blocker. Clin Exp Hypertens 1988;A10:791-800.
25 Andren L, Weiner L, Svensson A, Hansson L. Enalapril with
either a "very low" or "low" dose of hydrocholrothiazide is
equally effective in essential hypertension: a double-blind trial in
100 hypertensive patients. J Hypertens 1983;1(suppl
2):384-6.
26 Koller W, O'Hara R, Weiner W, Lang A, Nutt J, Agid Y, et al.
Relationship of aging to Parkinson's disease. Adv
Neurol 1987;45:317-21.
27 Martilla R J, Rinne U K. Epidemiology of Parkinson's disease in
Finland. Acta Neurol Scand
1976;53:81-102.
Fig 1: Prescribing cascade Non-steroidal anti-inflammatory drugs and starting
antihypertensive treatment
Examples of prescribing cascade
Initial treatment
Adverse effect
Subsequent treatment
Non-steroidal anti-inflammatory drugs(8)
Rise in blood pressure
Antihypertensive treatment
Thiazide diuretics(9)
Hyperuricaemia
Treatment for gout
Metoclopramide treatment(10)
Parkinsonian symptoms
Treatment with levodopa
Thiazide diuretics and starting treatment for gout
Use of metoclopramide when starting levodopa
treatment
Metoclopramide hydrochloride is widely used in the treatment of
gastro-oesophageal reflux, in the management of disorders of gastric
emptying including diabetic gastroparesis, and as an antiemetic after
chemotherapy. Its antidopaminergic adverse effects, including unwanted
extrapyramidal signs and symptoms, have long been recognised. Such drug
induced symptoms in older people can be misinterpreted as indicating a
new disease or be attributed to the aging process itself. This
misinterpretation is particularly likely when the symptoms are
indistinguishable from an illness that is seen more often in older
people, such as Parkinson's disease.(26-27)
Fig 2: Adverse effects of drugs that are related to dose, leading to prescribing cascade. Top: Odds ratio for starting treatment with antihypertensive drugs based on dose of non-steroidal anti-inflammatory drug (NSAID)(8) Centre: Relative risk for starting treatment for gout based on dose of thiazide.(9) Bottom: Odds ratios for starting levodopa treatment based on dose of metoclopramide.(10) Conclusion
Carefully re-evaluating the absolute need for the offending
agent; Using non-pharmacological treatment for managing a patient's
medical condition; Reducing the dosage of the implicated drug to the lowest
feasible dose that is effective in treating a patient's medical
condition; and Considering alternative drugs that might be safer in terms of
the risk of adverse effects in older people.
Departments of Medicine and of Preventive Medicine
and Biostatistics,
University of Toronto,
Baycrest Centre for Geriatric
Care,
3560 Bathurst Street,
North York,
Ontario,
Canada M6A 2E1
Paula A Rochon, assistant professor of
medicine
100 Central Street,
Worcester,
MA 01608,
USA
Jerry H
Gurwitz, executive director
Home | Current issue | Past issues |
Classified ads | Career Focus | Feedback
Collections |
About this site | About the BMJ | BMA | Medline