General practitioners ' office consultations with elderly patients – diagnoses and prescriptions : A report from the Møre & Romsdal Prescription Study

Study objectives: To analyse office consultations with elderly people in general practice: reason for encounter, diagnostic indications for prescribing drugs, and what drugs are prescribed for the different diagnoses. Design: Cross-sectional survey. Setting: The Norwegian county of Møre and Romsdal. Participants: In 1988 and 1989 the general practitioners recorded all encounters with patients over a period of two months. The participation rate was close to 100%, and we report data from 10,850 office consultations with elderly patients (65 years and older). Main results: 60% of the consultations involved female patients, mean age was 74 years. One of three consultations were for new diagnoses, two thirds were follow-ups. The most common diagnostic groups were cardiovascular (28%), musculoskeletal (13%), psychiatric (8%) and respiratory diseases (8%). Almost 10% of all consultations were for hypertension. Drugs were prescribed during 56% of all consultations. 27% of all prescriptions were for cardiovascular drugs, and 25% were for the central nervous system. The 20 most common diagnoses constituted more than half of all the diagnoses, and the ten most common drugs (therapeutic groups) constituted almost 70% of all prescriptions. Conclusion: In the context of quality assurance large benefits may be gained even if focus is put on rather few diagnoses and drugs.

General practitioners (GPs) prescribe most of the drugs taken by elderly people.Drugs are issued in almost six out of ten contacts with the GP, more often for women than for men, and more often with increasing age of the patient (7).
Most of the pharmacoepidemiological surveys in Norway have been performed without diagnostic data, making it difficult to evaluate the quality of the prescriptions.We have performed a survey where it is possible to relate the diagnosis (indication for prescribing) to the drug actually prescribed for the individual patient.
Most contacts in general practice take place in the doctor's surgery (8,9).In this paper we have analysed such contacts with elderly people: main reason for contact, indications for prescribing drugs, and what drugs are prescribed for the different diagnoses.Two separate periods were chosen for the survey because between the two periods an intervention study was conducted on improving the GPs' prescription patterns (for insomnia and acute cystitis) by prescribing feedback and therapeutic recommendations sent them by mail (10).

METHODS
During the survey the GPs used a specially designed prescription form, one part of which was handed to the patients as their prescription sheet.A carbon copy was retained with a questionnaire.On this the GPs filled in information about: • The patients' age and sex • Type of GP-patient contact (office consultation, home visit, telephone consultations, or indirect contact through the receptionist) • Whether the contact was initial or follow-up for the diagnosis • Reason for encounter, and diagnostic indication(s) for the prescriptions • Prescription information, and whether the prescription was a first time or repeat one Several drugs could be issued on the same prescription sheet during the same ordination.The questionnaires were completed during all contacts irrespective of drugs were issued or not.The twelve pharmacies in the county kept a record every time a GP used his or her own personal prescription form instead of the form designed for the study.This showed that other forms were used in less than 0.5% of the prescriptions (10).
The GPs were provided with a list of about 250 of the most common diagnoses used in general practice in terms of The International Classification of Primary Care (ICPC).The GPs were not obliged to use specific diagnostic criteria.During the coding we modified the ICPC-list: in some cases, different codes that covered similar clinical conditions (e.g., hypertension and elevated blood pressure) were grouped together as can be read from the tables.
The drugs were classified according to the Anatomical Therapeutic Chemical (ATC) classification.In 1988, 752 different drugs were listed in the Pharmaceutical Industry's Catalogue in Norway.

MATERIAL
Of the 156 GPs in the county, 149 (95.5%) participated in the MRPS during the first period.The participation rate rose to 98% during the second period.During the MRPS the GPs altogether recorded 90,458 contacts with patients during which 74,079 drug prescriptions were issued.Of all the contacts, 21,512 were with patients aged 65 years or more, and about half of them (10,850) were office consultations.During the 10,850 consultations, 9,521 drug prescriptions were issued.
The 10,850 office consultations and the 9,521 prescriptions constitute the material for this article.

RESULTS
Of the 10,850 office consultations, 60% were with female patients.The mean age of the patients was 74 years, 58% of all were younger olds (Figure 1).Two of three of the consultations were for diagnoses established previously, while every third consultation was for a new problem.Most of the consultations took place on the five weekdays with most consultations on Tuesdays (23%) and the least on Fridays (16%).Only 1.2% of the office consultations were performed during weekends.The most common reasons for visiting the GPs' surgeries were diagnoses in the cardiovascular system (ICPC-group K; 28%), followed by the musculoskeletal group (ICPC-group L; 13%), psychiatric problems (ICPC-group P; 8%), and respiratory tract illnesses (ICPC-group R; 8%).The 20 most commonly recorded diagnoses for seeing the GP comprised more than half of all diagnoses (Table 1).A little less than one in ten consultations were for hypertension, more commonly for females and for the younger olds (Table 1).
The GPs prescribed one or more drugs during 56% of all office consultations.The proportion of the consultations during which drug treatment was issued, varied with the diagnoses: from 28% of the consultations for cerebrovascular stroke, to 99% of the consultations for insomnia (Table 1).Of the consultations during which drug prescriptions not were issued, 47% of the patients were already taking drugs issued previously, and another 11% were referred to hospital.In Table 2 the 9,521 drug prescriptions are shown distributed according to the main anatomical ATC groups.Twenty-seven percent of the prescriptions were for drugs acting on the cardiovascular system (ATC class C), 25% were drugs for the central nervous system (ATC class N).Altogether 358 (of 752 possible) different drugs were issued.

Method
This work is part of the largest survey of patient contacts ever made in Norway.Its strength is that nearly 100% of the physicians participated according to protocol.
The study was performed during November 1988 and November 1989.This may have caused some diagnostic bias concerning seasonal diseases.
Another possible source of error is that diagnoses may be used for justifying the prescription of certain drugs.A request for a hypnotic may be the sole basis for a diagnosis of insomnia, which in turn makes it "legitimate" to prescribe a hypnotic drug (11).This may have caused an overreporting of such diagnoses.It was considered impractical to use standardized diagnostic criteria.Although the diagnostic validity thus may be weakened, the diagnoses reported here are probably representative for diagnoses made during everyday general practice in Norway.
In general practice the distribution of diagnoses seems to be quite stable over time (9,12).As for drugs, minor changes occur all the time since new drugs are introduced, while others are withdrawn from the market.The uses of single compounds therefore may have changed considerably since our data were recorded.However, because we only report main anatomi-cal groups and therapeutical subgroups according to the ATC classification, we believe that our results are still relevant.

Results
In general, approximately 60% of the patients seen in GPs' surgeries are females (12).The present study shows that this holds true also for elderly patients.Even if this is a survey of reasons for contact and not a prevalence study, the results are not very different from prevalence studies reported from Sweden and Denmark (2,3,13,14).Generally, hypertension and heart diseases are the most common problems.Women tend to be given a diagnosis of hypertension more often than men.We found that men more often suffer from coronary heart disease, arrythmias, and heart failure.This finding is in accordance with a systematic health survey among elderly people in Northern Sweden (13).
The data reported here were recorded before the new generation of antidepressant drugs (i.e. the selective serotonin reuptake inhibitors) was launched.Today prescriptions for antidepressants therefore probably constitute a larger share than the 2% reported here.
At the time when the survey was performed GPs had a conservative attitude towards treating hypertension in the elderly.According to later recommendations, more elderly hypertensive men should probably have been treated (15).We can only speculate whether this would have prevented the development of heart diseases in this population.
We found that drugs were prescribed during 56% of all consultations.Another 21% had previously received one or more prescriptions for the same diagnosis.Thus, it seems that altogether 77% of elderly people receive drug treatment for their reason for visiting their GP.Other studies have revealed similar numbers (1,3,16,17).The most commonly prescribed drugs for elderly people are cardiovascular drugs, psychotropic drugs, and analgesics, a finding also reported by others (1)(2)(3)16,17).
The patients do not always know why they are taking their drugs, and this problem increases by increasing age (1).Other studies have revealed that many drugs prescribed to elderly people are given for questionable indications, and that the treatment is inadequately followed up (1,4,5,18).This problem also tends to get worse in old age (4,5).
Since most elderly people live in their own homes and are responsible for their own medication, it is important that the GPs carefully instruct them how to handle their drugs.Key words are better communication (both spoken and written) in what may be called "patient-centered prescribing" (19): The physician and the patient must cooperate in establishing a shared understanding of the problem, the treatment goals, and also regarding the duration of the treatment.
In the context of quality assurance it is an advantage that so few diagnoses and drugs make up a large proportion of the total: "Common diseases commonly occur, rare diseases rarely happen" (20).Even if we focus on a few diagnoses and drugs, a large impact may be made upon the quality of drug prescribing for the elderly in general practice.
This is a report from the Møre & Romsdal Prescription Study (MRPS), in which almost all GPs in the Norwegian county of Møre & Romsdal recorded their contacts with patients and prescriptions during November 1988 and November 1989.

Figure 1 .
Figure 1.Elderly patients (65+) consulting general practitioners in their surgeries, by patients' sex and age groups.

Table 1 .
The twenty most commonly recorded diagnoses for seeing elderly patients in the general practitioners' surgeries, by patients' sex and age groups, and the proportions of the consultations for the various diagnoses where drug prescriptions (Rx) are issued.

Table 3 .
The ten most frequently prescribed drugs (ATC system therapeutic groups) for elderly (65+) patients during office consultations in general practice, and the three most commonly recorded diagnoses (%) for prescribing in each therapeutic group.