Age and Ageing (1978), 7, 74
DRUG PRESCRIBING PATTERNS IN THE ELDERLY: A CROSS-SECTIONAL STUDY OF IN-PATIENTS L. J. CHRISTOPHER1, B. R. BALLINGER2, A. M. M. SHEPHERD3, A. RAMSAY4 AND G. CROOKS8 1
Aberdeen-Dundee Medicines Evaluation and Monitoring Group, Nineioells Hospital, Dundee, Royal Dundee Liff Hospital, Dundee, 8 University of Dundee, Dundee, *Drug Monitoring Unit Royal Dundee Liff Hospital, Dundee, ''University of Aberdeen, Aberdeen
i
Summary A cross-sectional one-day survey of drug prescribing and use in elderly in-patients was carried out in Dundee hospitals. Altogether 873 patients were studied and analyses of prescribing patterns were carried out by patient categories—medical, surgical, geriatric, psychiatric and mentally subnormal—and by major drug groups. While the average number of drugs prescribed per patient on the study day was 3.3, which was not excessive, other results highlighted some possible problems of therapy, viz. the wide range of drugs used within drug groups, the common use of psychotropic drugs, dose regimes which although used in a geriatric population differed little from general adult dosages, and which sometimes involved many administrations per day. Examples were described relating to the use of nitrazepam, phenothiazines and tricyclic antidepressants. INTRODUCTION
The elderly population is increasing and the over-65s now comprise over 13% of the total population in Great Britain compared with 5% early this century (Office of Population Censuses and Surveys 1971). Because of a higher incidence of multiple disabilities than in younger age groups, the elderly require relatively more hospital in-patient care and are frequently subject to multiple drug prescribing (U.S. Department H.E.W. 1968). Since this may be accompanied by altered drug handling (Crooks, O'Malley & Stevenson 1976) and often by the added factor of poor drug compliance (Parkin et al. 1976) it would not be surprising to find the elderly more at risk of adverse drug effects than their younger counterparts. Several authors have published evidence that this is indeed the case, for example Hurwitz (1969) found that only 3% of in-patients aged 20-29 years experienced adverse drug reactions compared with over 21% of patients aged 70-79. At present there is little available information on prescribing patterns in the elderly and this study describes the findings in an in-patient population in Dundee hospitals, examines potential problems of prescribing that emerged and suggests ways in which they may be minimized. Patients and Methods The study covered one 24-hour period in July 1975. All patients of 65 years and over in Dundee hospitals were included in the study. Patient data were: name, case reference number, ward, age, sex, all drugs currently prescribed, their dose, route and frequency of administration. Data
Drug Prescribing Patterns in the Elderly
75
extraction was facilitated by the use of standard forms for drug prescription and administration recording in all hospitals (Crooks et al. 1967). Patients were grouped into five categories, medical, surgical, geriatric psychiatric and mentally subnormal, according to the wards they occupied. The drugs prescribed for 873 patients (625 females and 248 males) were studied. Two hundred and sixty-six (43 %) of the females were over 80 years old and the majority of them occupied geriatric (128) and psychiatric (78) beds. While the female population increased with age from 81 in the 65-69 age band to 266 in the 80 + group, males were fairly evenly distributed, with a range of 52 to 69 patients in each group. Psychiatric (301) and geriatric (279) were the largest patient categories. The others, in decreasing order of size were medical (158), surgical (100) and mentally subnormal (35).
RESULTS
The average number of prescriptions per patient was 3.3 during the 24-hour period of observation. Patients in geriatric wards were prescribed most drugs with an average of 4.0. The number of drugs most commonly prescribed was three. While 19% received this number, 52% of all patients had from 1 to 6 drug doses administered, 33% had 7 or more and 15% 10 or more. On the other hand, 15% (128) patients did not receive any drug. It is known that drugs prescribed on an 'as required* basis comprise about 20% of all hospital prescriptions in Dundee hospitals (Aberdeen-Dundee Medicines Evaluation and Monitoring Group: unpublished); in this survey, approximately 50% of patients were prescribed one 'as required' drug and most of these received regular medication as well. Some 67% of patients in geriatric wards were prescribed one or more 'as required' medicine compared to 31% of those in the medical category. On the study day, only 21% of 'as required' medication was given, the proportion of given to not given being similar between patient categories. Use of drugs by groups
Table I shows patterns of prescribing by major drug groups within different patient categories. Phenothiazines (see also Figs. 1 and 2), chloral derivatives, diuretics and laxatives were the most frequently prescribed, each being used in more than 200 patients. While popular drugs tended to be widely prescribed between patient categories, certain features stand out. Diuretics and mineral supplements were most used by medical patients, probably reflecting a high incidence of cardiac disease (see Table II). Analgesics were most frequently prescribed for surgical cases, chloral derivatives for geriatric patients and, as expected, phenothiazines for psychiatric patients. The mentally subnormal represented only a small sector of the study (35 patients) and no drug group was particularly favoured. The use of hypnotics was high in the four main patient categories and this becomes more obvious when prescriptions for chloral derivatives, benzodiazepines and 'other hypnotics and sedatives' are pooled together; it is then evident that hypnotic prescribing ranges from over 40% in medical to more than 70% in geriatric wards. The most commonly prescribed hypnotic was the chloral derivative triclofos, particularly in geriatric wards. It was remarkable that only one of the 35 patients in the mentally subnormal group was prescribed a hypnotic.
76
L. J. Christopher et al.
Table I. Prescribing patterns of the most commonly used drug groups (shown as percentage of patients within each category)
Drug group Neuroleptics Chloral derivatives Diuretics Laxatives Mineral supplements Analgesics Benzodiazepines Iron preparations Cardiac drugs Antibiotics and anti-infectives Anticholinergics Antidepressants Vitamins Anti-Parkinson's Other sedatives and hypnotics (eic. barbiturates)
Medical N = 158
Surgical N = 100
%
/o
12 16 40 10 35 22 17 15 27
Geriatric Psychiatric N-=279 N = 301
Mentally tub-normal N = 35
All
N = 873
0/
0/
0/
/o
/o
/o
/o
19 17 24 7 21 61 23 14 14
36 61 35 42 23 22 11 17 13
48 23 22 30 15 9 18 9 7
6 3 17 17 14 17
286 (33) 280 (32) 257 (29) 235 (27) 192 (22) 189 (22) 135 (16) 114(13) 111 (13)
24 10 5 5 2
22 7 1 7
11 13 9 10 4
5 3 10 4 10
3
11
7
1
4
6 3
11 3
106 (12) 67 (8) 66 (8) 58 (7) 45 (5) 37 (4)
If a patient is prescribed one or more drugs in a given group this is counted once. 60-,
Ptyehtatrle CD All agn 145 Non-Pivdiitrtc FS\ Pi\»tm 141
50-
40-
30-
20-
10-
64-69
70-74
75-79
80+
AGE
Fig. 1. Neuroleptic drug prescribing by age groups: psychiatric versus non-psychiatric.
77
Drug Prescribing Patterns in the Elderly Up to 50 mg/dsy
K\N
Mora than 50 me/d*y I
1
1tt Column • Pjychiatric Patients 2nd Column • Non-Ptychiatrtc
50-
40-
1 20 -
10 -
65-69
70-74
F
75-79
I 80+
AGE
Fig. 2. Phenothiazine daily dosage by age groups (257 patients).
Table II. Diuretic prescribing/potassium supplementation Short-stay medical N = 158
Short-stay surgical N = 100
Long-stay geriatric N = 279
Psychiatric N = 301
Mentally subnormal N = 35
Total
Bendrofluazide Frusemide Other drugs Total
12 37 19 68
6 12 9 27
43 23 32 98
43 11 13 67
1 4 2 7
105 (39%) 87(33%) 75 (28%) 267 (100%)
No. of different diuretics
6
7
9
8
3
13
51
21
57
46
5
180
8 59
4 25
15 72
4 50
0 5
31 211
Prescription
Separately prescribed potassium supplement Diuretic and potassium supplement Total
78
L. J. Christopher et al.
Of the cardiac drugs (111 patients), digoxin was prescribed most frequently (89 patients), the main users being medical (34) and geriatric (27) patients; 44% of all patients on digoxin were 80 years old or over (see Fig. 3). Of 66 patients who received antidepressants, 29 were psychiatric and 24 were geriatric. The most popular sub-group, the tricyclics, accounted for 80% of all antidepressant prescriptions (see also Table III). 40I
und*r 0.25 mg/dty I 0.25mg+/d»y
30
20o
i
10-
1 6&-69
70-74
80+
75-79 AGE
8%
9%
10%
12% of all ptttenu in survty
Fig. 3. Digoxin daily dosage by age groups (89 patients).
Table III. Tricyclic antidepressants : tunes per day prescribed Drug Amitriptyline Imipramine Others (3) Total
1 Daily
2 Daily
3 Daily
7 6 1 14
5 4 1 10
11 16 1 28
As required 1
1
Total 24 26 3 53
Drug Prescribing Patterns in the Elderly
79
DISCUSSION
The study method was simple in that drug prescription sheets in each ward are normally held in one place—the drug Kardex folder—and all items needed are contained therein (Crooks et al. 1967). Two of the study group collected all data required over a period of a fortnight. In a 'one-day1 cross-sectional study it is not possible to comment on an individual patient's course of treatment which unfolds over the whole period of hospitalization. It follows that information such as the number of drugs prescribed per patient in this survey is not comparable to the number of drugs prescribed per patient when this figure is derived from hospital discharge records. Since about half of the patients were prescribed one or more drugs to be given 'as required' this probably produced some exaggeration of the true picture of drug use, particularly with analgesics, hypnotics, sedatives and phenothiazines. This mode of prescribing apparently occurs more frequently in the elderly since in the overall hospital population it accounts for approximately 20% of all prescriptions. Apart from continuing intermittent need for certain prescribed drugs, factors such as low doctor or nurse to patient ratios in geriatric wards could contribute to this picture (Ballinger & Ramsay 1974). It is notable that 120 patients were prescribed six drugs or more and 132 patients received 10 or more drug doses on the study day. Whether such drug regimes, particularly those requiring many administrations per day can be effectively complied with by elderly patients in their homes is debatable and their simplification prior to discharge would be advantageous (Parkin et al. 1976). Laxatives were high on the list of prescribed drugs. Evidence is accumulating that the addition of roughage to the diet could achieve the same objective (Burkitt 1972). With encouragement from the nursing staff, one geriatric ward has managed practically to eliminate the use of laxatives by adding bran to the patients' morning cereal. While the cost of laxatives is relatively small, the gain in nurses' and pharmacists' time should be considerable. The use of diuretics and potassium supplements was high in all patient categories in this survey. Figures showed that frusemide was used to a remarkable extent (87 patients, 33%) (Table II). Since the therapeutic result for the chronic sick would probably be no worse if a less potent (and often cheaper) diuretic was used, it is not easy to explain the common use of a powerful diuretic like frusemide which may produce over-flow incontinence or aggravate urinary retention in elderly males. The number of different diuretic preparations prescribed in this survey was 13, and while different ones may be selected for individual needs, a common approach to diuretic prescribing could lead to a reduction in the range of diuretics employed, without prejudice to patient benefit. Without joining in the current debate, the need for potassium supplementation should at least be further investigated especially in the light of recent evidence which questions its usefulness as a routine measure (Lawson et al. 1976). Neuroleptics (mainly phenothiazines) were prescribed for 286 (33%) patients. As expected, psychiatric patients accounted for the major share and 48% of patients in this category were prescribed a neuroleptic, compared to 25% of non-psychiatric patients. While patient age was independent of frequency of use in psychiatric patients, in non8
80
L. J. Christopher et al.
psychiatric patients drug use increased with age, and of the 65-69 age group 16% were prescribed a neuroleptic in contrast to 28% of the over-80s (Fig. 1). This probably reflects the increasing frequency of psychiatric problems in the ageing in-patient population occupying general and geriatric beds (xia for linear trend=4.7; P
DRUG PRESCRIBING PATTERNS IN THE ELDERLY: A CROSS-SECTIONAL STUDY OF IN-PATIENTS L. J. CHRISTOPHER1, B. R. BALLINGER2, A. M. M. SHEPHERD3, A. RAMSAY4 AND G. CROOKS8 1
Aberdeen-Dundee Medicines Evaluation and Monitoring Group, Nineioells Hospital, Dundee, Royal Dundee Liff Hospital, Dundee, 8 University of Dundee, Dundee, *Drug Monitoring Unit Royal Dundee Liff Hospital, Dundee, ''University of Aberdeen, Aberdeen
i
Summary A cross-sectional one-day survey of drug prescribing and use in elderly in-patients was carried out in Dundee hospitals. Altogether 873 patients were studied and analyses of prescribing patterns were carried out by patient categories—medical, surgical, geriatric, psychiatric and mentally subnormal—and by major drug groups. While the average number of drugs prescribed per patient on the study day was 3.3, which was not excessive, other results highlighted some possible problems of therapy, viz. the wide range of drugs used within drug groups, the common use of psychotropic drugs, dose regimes which although used in a geriatric population differed little from general adult dosages, and which sometimes involved many administrations per day. Examples were described relating to the use of nitrazepam, phenothiazines and tricyclic antidepressants. INTRODUCTION
The elderly population is increasing and the over-65s now comprise over 13% of the total population in Great Britain compared with 5% early this century (Office of Population Censuses and Surveys 1971). Because of a higher incidence of multiple disabilities than in younger age groups, the elderly require relatively more hospital in-patient care and are frequently subject to multiple drug prescribing (U.S. Department H.E.W. 1968). Since this may be accompanied by altered drug handling (Crooks, O'Malley & Stevenson 1976) and often by the added factor of poor drug compliance (Parkin et al. 1976) it would not be surprising to find the elderly more at risk of adverse drug effects than their younger counterparts. Several authors have published evidence that this is indeed the case, for example Hurwitz (1969) found that only 3% of in-patients aged 20-29 years experienced adverse drug reactions compared with over 21% of patients aged 70-79. At present there is little available information on prescribing patterns in the elderly and this study describes the findings in an in-patient population in Dundee hospitals, examines potential problems of prescribing that emerged and suggests ways in which they may be minimized. Patients and Methods The study covered one 24-hour period in July 1975. All patients of 65 years and over in Dundee hospitals were included in the study. Patient data were: name, case reference number, ward, age, sex, all drugs currently prescribed, their dose, route and frequency of administration. Data
Drug Prescribing Patterns in the Elderly
75
extraction was facilitated by the use of standard forms for drug prescription and administration recording in all hospitals (Crooks et al. 1967). Patients were grouped into five categories, medical, surgical, geriatric psychiatric and mentally subnormal, according to the wards they occupied. The drugs prescribed for 873 patients (625 females and 248 males) were studied. Two hundred and sixty-six (43 %) of the females were over 80 years old and the majority of them occupied geriatric (128) and psychiatric (78) beds. While the female population increased with age from 81 in the 65-69 age band to 266 in the 80 + group, males were fairly evenly distributed, with a range of 52 to 69 patients in each group. Psychiatric (301) and geriatric (279) were the largest patient categories. The others, in decreasing order of size were medical (158), surgical (100) and mentally subnormal (35).
RESULTS
The average number of prescriptions per patient was 3.3 during the 24-hour period of observation. Patients in geriatric wards were prescribed most drugs with an average of 4.0. The number of drugs most commonly prescribed was three. While 19% received this number, 52% of all patients had from 1 to 6 drug doses administered, 33% had 7 or more and 15% 10 or more. On the other hand, 15% (128) patients did not receive any drug. It is known that drugs prescribed on an 'as required* basis comprise about 20% of all hospital prescriptions in Dundee hospitals (Aberdeen-Dundee Medicines Evaluation and Monitoring Group: unpublished); in this survey, approximately 50% of patients were prescribed one 'as required' drug and most of these received regular medication as well. Some 67% of patients in geriatric wards were prescribed one or more 'as required' medicine compared to 31% of those in the medical category. On the study day, only 21% of 'as required' medication was given, the proportion of given to not given being similar between patient categories. Use of drugs by groups
Table I shows patterns of prescribing by major drug groups within different patient categories. Phenothiazines (see also Figs. 1 and 2), chloral derivatives, diuretics and laxatives were the most frequently prescribed, each being used in more than 200 patients. While popular drugs tended to be widely prescribed between patient categories, certain features stand out. Diuretics and mineral supplements were most used by medical patients, probably reflecting a high incidence of cardiac disease (see Table II). Analgesics were most frequently prescribed for surgical cases, chloral derivatives for geriatric patients and, as expected, phenothiazines for psychiatric patients. The mentally subnormal represented only a small sector of the study (35 patients) and no drug group was particularly favoured. The use of hypnotics was high in the four main patient categories and this becomes more obvious when prescriptions for chloral derivatives, benzodiazepines and 'other hypnotics and sedatives' are pooled together; it is then evident that hypnotic prescribing ranges from over 40% in medical to more than 70% in geriatric wards. The most commonly prescribed hypnotic was the chloral derivative triclofos, particularly in geriatric wards. It was remarkable that only one of the 35 patients in the mentally subnormal group was prescribed a hypnotic.
76
L. J. Christopher et al.
Table I. Prescribing patterns of the most commonly used drug groups (shown as percentage of patients within each category)
Drug group Neuroleptics Chloral derivatives Diuretics Laxatives Mineral supplements Analgesics Benzodiazepines Iron preparations Cardiac drugs Antibiotics and anti-infectives Anticholinergics Antidepressants Vitamins Anti-Parkinson's Other sedatives and hypnotics (eic. barbiturates)
Medical N = 158
Surgical N = 100
%
/o
12 16 40 10 35 22 17 15 27
Geriatric Psychiatric N-=279 N = 301
Mentally tub-normal N = 35
All
N = 873
0/
0/
0/
/o
/o
/o
/o
19 17 24 7 21 61 23 14 14
36 61 35 42 23 22 11 17 13
48 23 22 30 15 9 18 9 7
6 3 17 17 14 17
286 (33) 280 (32) 257 (29) 235 (27) 192 (22) 189 (22) 135 (16) 114(13) 111 (13)
24 10 5 5 2
22 7 1 7
11 13 9 10 4
5 3 10 4 10
3
11
7
1
4
6 3
11 3
106 (12) 67 (8) 66 (8) 58 (7) 45 (5) 37 (4)
If a patient is prescribed one or more drugs in a given group this is counted once. 60-,
Ptyehtatrle CD All agn 145 Non-Pivdiitrtc FS\ Pi\»tm 141
50-
40-
30-
20-
10-
64-69
70-74
75-79
80+
AGE
Fig. 1. Neuroleptic drug prescribing by age groups: psychiatric versus non-psychiatric.
77
Drug Prescribing Patterns in the Elderly Up to 50 mg/dsy
K\N
Mora than 50 me/d*y I
1
1tt Column • Pjychiatric Patients 2nd Column • Non-Ptychiatrtc
50-
40-
1 20 -
10 -
65-69
70-74
F
75-79
I 80+
AGE
Fig. 2. Phenothiazine daily dosage by age groups (257 patients).
Table II. Diuretic prescribing/potassium supplementation Short-stay medical N = 158
Short-stay surgical N = 100
Long-stay geriatric N = 279
Psychiatric N = 301
Mentally subnormal N = 35
Total
Bendrofluazide Frusemide Other drugs Total
12 37 19 68
6 12 9 27
43 23 32 98
43 11 13 67
1 4 2 7
105 (39%) 87(33%) 75 (28%) 267 (100%)
No. of different diuretics
6
7
9
8
3
13
51
21
57
46
5
180
8 59
4 25
15 72
4 50
0 5
31 211
Prescription
Separately prescribed potassium supplement Diuretic and potassium supplement Total
78
L. J. Christopher et al.
Of the cardiac drugs (111 patients), digoxin was prescribed most frequently (89 patients), the main users being medical (34) and geriatric (27) patients; 44% of all patients on digoxin were 80 years old or over (see Fig. 3). Of 66 patients who received antidepressants, 29 were psychiatric and 24 were geriatric. The most popular sub-group, the tricyclics, accounted for 80% of all antidepressant prescriptions (see also Table III). 40I
und*r 0.25 mg/dty I 0.25mg+/d»y
30
20o
i
10-
1 6&-69
70-74
80+
75-79 AGE
8%
9%
10%
12% of all ptttenu in survty
Fig. 3. Digoxin daily dosage by age groups (89 patients).
Table III. Tricyclic antidepressants : tunes per day prescribed Drug Amitriptyline Imipramine Others (3) Total
1 Daily
2 Daily
3 Daily
7 6 1 14
5 4 1 10
11 16 1 28
As required 1
1
Total 24 26 3 53
Drug Prescribing Patterns in the Elderly
79
DISCUSSION
The study method was simple in that drug prescription sheets in each ward are normally held in one place—the drug Kardex folder—and all items needed are contained therein (Crooks et al. 1967). Two of the study group collected all data required over a period of a fortnight. In a 'one-day1 cross-sectional study it is not possible to comment on an individual patient's course of treatment which unfolds over the whole period of hospitalization. It follows that information such as the number of drugs prescribed per patient in this survey is not comparable to the number of drugs prescribed per patient when this figure is derived from hospital discharge records. Since about half of the patients were prescribed one or more drugs to be given 'as required' this probably produced some exaggeration of the true picture of drug use, particularly with analgesics, hypnotics, sedatives and phenothiazines. This mode of prescribing apparently occurs more frequently in the elderly since in the overall hospital population it accounts for approximately 20% of all prescriptions. Apart from continuing intermittent need for certain prescribed drugs, factors such as low doctor or nurse to patient ratios in geriatric wards could contribute to this picture (Ballinger & Ramsay 1974). It is notable that 120 patients were prescribed six drugs or more and 132 patients received 10 or more drug doses on the study day. Whether such drug regimes, particularly those requiring many administrations per day can be effectively complied with by elderly patients in their homes is debatable and their simplification prior to discharge would be advantageous (Parkin et al. 1976). Laxatives were high on the list of prescribed drugs. Evidence is accumulating that the addition of roughage to the diet could achieve the same objective (Burkitt 1972). With encouragement from the nursing staff, one geriatric ward has managed practically to eliminate the use of laxatives by adding bran to the patients' morning cereal. While the cost of laxatives is relatively small, the gain in nurses' and pharmacists' time should be considerable. The use of diuretics and potassium supplements was high in all patient categories in this survey. Figures showed that frusemide was used to a remarkable extent (87 patients, 33%) (Table II). Since the therapeutic result for the chronic sick would probably be no worse if a less potent (and often cheaper) diuretic was used, it is not easy to explain the common use of a powerful diuretic like frusemide which may produce over-flow incontinence or aggravate urinary retention in elderly males. The number of different diuretic preparations prescribed in this survey was 13, and while different ones may be selected for individual needs, a common approach to diuretic prescribing could lead to a reduction in the range of diuretics employed, without prejudice to patient benefit. Without joining in the current debate, the need for potassium supplementation should at least be further investigated especially in the light of recent evidence which questions its usefulness as a routine measure (Lawson et al. 1976). Neuroleptics (mainly phenothiazines) were prescribed for 286 (33%) patients. As expected, psychiatric patients accounted for the major share and 48% of patients in this category were prescribed a neuroleptic, compared to 25% of non-psychiatric patients. While patient age was independent of frequency of use in psychiatric patients, in non8
80
L. J. Christopher et al.
psychiatric patients drug use increased with age, and of the 65-69 age group 16% were prescribed a neuroleptic in contrast to 28% of the over-80s (Fig. 1). This probably reflects the increasing frequency of psychiatric problems in the ageing in-patient population occupying general and geriatric beds (xia for linear trend=4.7; P
