Family Practice Vol. 17, No. 3, 248-251
© Oxford University Press 2000
Frequent attenders in general practice: an attempt to reduce attendance
Institute of General Practice and Primary Care, Northern General Hospital, Herries Rd, Sheffield S5 7AU, UK.
Jiwa M. Frequent attenders in general practice: an attempt to reduce attendance. Family Practice 2000; 17: 248–251.
Received 6 August 1999; Accepted 21 December 1999.
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
Background. ‘Frequent attenders' in general practice are known to include patients with a variety of problems. Most studies of frequent attenders have not assessed the impact of providing GPs with detailed summaries of the clinical records of these patients on consultation rates. Good medical records are associated with good care. If it is not relatively easy or quick for GPs to ascertain which chronic illness or psychosocial problems the patient has from the records, it will be difficult to manage the patient proactively.
Objective. The aim of this study was to investigate the impact on the consultation rate of providing a detailed and accessible summary of patients' problems including physical, social and psychological data based on information already recorded in the patients' records.
Method. A prospective controlled study was made of frequent attenders at one UK general practice comprising four full-time GPs. A total of 104 frequent attenders were identified by examining the lists of patients attending the surgery and by including the names of patients identified by GPs from memory. The final study groups were patients who consulted a GP 11 times or more in a year. The sample was divided into two groups. In both groups, the clinical notes were summarized for data relating to physical, social and psychological problems. In the intervention group, summaries were displayed prominently in the notes which were marked with a distinctive label. GPs were asked to read and initial the available summaries in these notes. In the control group, summaries were not included in the notes and the notes were not highlighted in any way. The consultation rate of these 104 patients was recorded for 5 months after the availability of the summaries in the notes.
Results. Summarizing the notes in the style described in this study failed to reduce the consultation rate of the identified frequent attenders (Kruskal–Wallis H = 2.75, P = 0.1) Furthermore, in the intervention group, patients for whom there was evidence that the summaries had been consulted by their GP (24; 46%) continued to attend as frequently as those whose summaries were not endorsed as consulted (mean attendance 4.8 consultations compared with 4.2 in 5 months).
Conclusion. Frequent attenders often have multiple problems, but prominently displayed summaries of their history for use by GPs during consultations do not reduce the frequency of consultations.
Keywords. Clinical notes, frequent attenders, general practice, summaries.
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Patients who attend their GP more frequently than most have been given a variety of labels: O'Dowd (1988) called them ‘heartsink’ patients,1 Groves (1986), a psychiatrist, referred to his group as ‘hateful’,2 Ellis (1986), a GP, uses the term ‘dysphoria’,3 Stokewell (1984), a nurse, calls them ‘unpopular’,4 Westhead5 (1985) and Courtenay6 (1974) both refer to ‘frequent attenders' and finally Bass (1990), a psychiatrist, introduced the term ‘chronic somatization'.7 Other groups8 have dealt with such patients under the heading ‘problem patients' and ‘difficult patients', defining them as those who regularly complain of a number of symptoms often in the absence of verifiable physical pathology.
There is no single explanation for the phenomenon of ‘frequent attenders’. They may comprise a "heterogeneous group who may have a genuine need for care, or a troublesome group who simply create an unnecessary workload".9 Whatever the roots of the problem, several approaches have been recommended. A plan, which sets out as its first step an information gathering exercise, including a review of the notes, was published by Corney et al. in 1988.8 Good medical records have been associated with good care.10 If it is not possible to ascertain from the records or from memory what chronic illness or psychosocial problems the patient has, it will be difficult to manage the patient well.11
Further recommendations, from a hospital-based study, suggest that for some types of frequent attender, where psychosomatic symptoms predominate, the doctor should ‘broaden the agenda’ to include discussion about the patient's emotional problems, and minimize contact with specialists if the risk of iatrogenesis is to be diminished.12 Most guidelines on the management of this group of patients recommend sharing information about the patient with partners in the practice. In a study of the GPs of so-called ‘heartsink’ patients, one team suggests: "...avoid blaming the patient for the problem and encourage the development of alternative management strategies".13 The authors took the view that GPs cope better with such patients if they adopt a strategy that includes information gathering as central to the process.
The setting for the present study was a UK NHS general practice. The practice is based in a small market town in North Nottinghamshire (Jarman score = –4.26). It serves 7200 patients and consists of four full-time equal partners working on a combined list basis. The great majority (95%) of the practice population live within 6 miles of the practice, and an appointment system is in operation. The practice average consultation rate was calculated as 3.5 per patient per year.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Identifying frequent attenders
Frequent attenders were identified in two ways: (i) by analysis of attendance registers at the surgery; and (ii) by asking GPs to name patients who they considered to attend more frequently than most.
By definition, frequent attenders consult at least once or more every month. Thus 2 months in a year were selected by simple random sampling. The names of all patients attending the surgery in October 1995 (2241 names) and April 1996 (1876 names) were selected. Only consultations with a doctor were counted. Each visit to the surgery was treated as a single event regardless of the number of problems presented during that consultation. Home visits, consultations with a nurse, consultations for antenatal care and minor operations were disregarded, as were all consultations out-of-hours.
The names of all patients seen in either month (October 95 or April 96), or during both months and seen more than three times were selected. The latter frequency was chosen arbitrarily in order to focus on the most frequently occurring names on surgery lists. Thus 194 names were selected. In addition, all GPs at the surgery were asked independently to name patients who they considered to attend the most frequently. The GPs named 51 patients. From this list, the names of patients not already identified by the review of attendance records were added to the list of potential frequent attenders, giving a total of 200 patients.
Sample size calculations
It was decided by practice consensus that a reduction in the number of attendances by 50% would be regarded as clinically significant. The clinical notes of the 200 patients identified were examined and the number of times each individual had attended in the year May 1995 to April 1996 was noted. In keeping with a definition adopted in a previous study,14 patients attending 11 times or more in the year (104) were selected (mean attendance 25, range 11–46, SD 14). Clearly these patients consult significantly more often than the practice average of 3.5 per patient per year. These 104 patients, when divided into two groups, give an 80% power to detect a 50% difference (Wilcoxon test) at 5% significance (i.e. 52 patients in each group).
The intervention
A summary was made of the data appearing in the clinical notes of these patients. The summary was produced on an A4 sheet of paper and discussed among the partners of the practice. A consensus was reached on the format of the summary.
The following sections were clearly delineated in the summaries:
- Section 1: age, sex, marital status, occupation, cigarette and alcohol consumption.
- Section 2: drug history and allergies, past medical history and summary of main medical problems.
- Section 3: index of commonest symptoms presented during consultations. After each of these symptom headings, a date was given in each summary indicating where in the notes the symptom was presented. The GP could then refer back to see how the symptom was managed in the past.
- Section 4: clinical signs noted. These included vital signs and any major clinical signs recorded, e.g. cardiac murmurs.
- Section 5: results of any major investigations conducted in the past.
- Section 6: other agencies involved in the care of the patient, including hospital consultants.
- Section 7: symptoms of depression recorded in previous consultations. The symptoms selected were those highlighted by Paykel and Priest.15
- Section 8: social problems discussed with a GP and recorded in the notes. For example, problems with a relative, neighbour or friend.
These summaries were much more detailed and wide ranging than the standard summaries available in the notes already. The latter were essentially chronological lists of the major events in the patient's medical history, focusing mainly on physical problems and data collected for administrative purposes. Only the information already available in the notes was summarized. The intervention group had their summaries entered in the notes. These patients had their notes marked with a distinctive yellow sticker indicating that a detailed summary was available. The control group did not have the summaries entered in the notes and their notes were not highlighted in any way. The doctors were asked to initial the summary on each occasion that one was consulted. They were instructed to consult the summaries at their discretion. Furthermore, they could add information if they felt data were missing or there had been a significant change in the patient's circumstances.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
At selection, there were 52 patients in each arm of the study. As the study progressed, there were losses in each group. In the intervention group, one patient moved out of the practice area and one patient died, whereas, in the control group, two patients moved out of the practice area and one patient died.
There was no difference between the two groups in terms of age, sex, marital status, smoking habit or problem drinking. However, this study was not designed to detect the impact of any differences between the groups, and further analysis was not possible.
There was no statistically significant difference in the attendance rate in the intervention group compared with the controls after the intervention period. The results of attendance are given in Table 1
. In the intervention group, there was no significant reduction in attendance during the study period for patients whose summaries were read and initialled by the GP (24 cases; 46%) compared with cases where this did not occur. For those whose summary was read and initialled, the mean attendance was 4.8, whereas for others in the intervention group it was 4.2
|
Therefore, patients whose summaries were read and initialled by the GPs remained frequent attenders under the definition used in this study (i.e. >11 consultations per year).
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
The patients studied and labelled as ‘frequent attenders' are a heterogeneous group of patients with a variety of needs. Patients included had a number of problems including chronic illness, terminal illness, recurrent minor illness, chronic unexplained symptoms, chronic infections, chronic pain and a plethora of social and psychological problems. Many had multiple problems. When asked to identify the frequent attenders, the GPs could only name 28 out of the final group of 104 patients. This finding highlights the difficulty with identifying all frequent attenders in a practice unless a systematic approach is adopted.
Some patients who were identified as frequent attenders in the period used to assess this feature appear not to have been ‘frequent attenders' in the past. Furthermore, in the follow-up period, some demonstrated a marked decline in attendance. Clearly there are patients who may attend exceptionally frequently but do not continue to do so indefinitely. This reflects the many reasons why patients consult GPs and the interplay of social,16 economic17 and psychological18 factors. It is also recognized that some patients appear to consult their GP without the reason for such attendance being recognized. Such patients receive no diagnosis or effective treatment other than contact with their doctor, and have been labelled ‘temporarily dependent’.19
It is clear from the results that even in cases where summaries were endorsed and therefore shown to have been read by the consulting GP, patients continued to attend frequently. One cannot exclude the possibility that a greater knowledge of the patient's background may not have had any significant impact on the doctor's decisions during consultations. It has been suggested in at least one study that despite having extensive knowledge of their patients, specifically in relation to housing and occupation, this played no part in the majority of decisions taken.20 Alternatively, the decisions taken did not have the desired effect of reducing the consultation rate.
This project aimed to introduce an element of continuity of care merely by providing written summaries in complex cases. Patient satisfaction with the consultations was not compared between the summary group and the controls. This was deemed beyond the scope of this study. However, it may have been a critical failing of this approach. Pereira Gray (1979) noted that by adopting a personal list system, the gold standard for continuity of care, he reduced the consultation rate of his patients.21 The provision of summaries did not introduce other aspects that are evident in a personal list system, including what Balint (1957) called the ‘mutual Investment Company’ which symbolizes the emotional investment by both parties in the doctor– patient relationship.22 Many academics including McCormick (1996) have continued to emphasize the importance of this aspect in general practice consultations.23 Some of the patients in this study may have suffered from ‘multiple doctor syndrome’ which was not remedied by the intervention adopted. The potential for such a dilution of personal care has been highlighted as a feature of group practices organized with combined lists.24
|
Acknowledgments |
|---|
I wish to thank Dr Judy Jones and Dr Roland Petchy for their help with the design of the study, Pat Ward from the Department of Public Health, North Nottinghamshire Health, Staff and Partners at Bridgegate Surgery, Retford, Professors Nigel Mathers and Mike Campbell of the Institute of General Practice, University of Sheffield, for cogent and constructive criticism, and Trent Focus and Trent health for financial assistance in attending the Master of Medical Science course at Nottingham University. This report is a dissertation submitted in partial fulfilment of the requirements for the degree of Masters of Medical Science in Primary Health Care, Department of General Practice, University of Nottingham, UK.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
1 O'Dowd TC. Five years of heartsink patients in general practice. Br Med J 1988; 297: 528–530.
2 Groves JE. Taking care of the hateful patient. New Engl J Med 1986; 298: 883–887.[Abstract]
3 Ellis CG. Making dysphoria a happy experience. Br Med J 1986; 293: 317–318.
4 Stockwell F. The Unpopular Patient. London: Croom Helm, 1984.
5 Westhead JN. Frequent attenders in general practice: medical, psychological and social characteristics. J R Coll Gen Pract 1985; 35: 337–340.[Web of Science][Medline]
6 Courtenay MJF, Curwen MP, Dawe D, Robinson J, Stern MJ. Frequent attendance in family practice. J R Coll Gen Pract 1974; 24: 251–261.
7 Bass C. Somatisation: Physical Symptoms and Psychological Illness. Oxford: Blackwell Scientific Publications, 1990.
8 Corney RH, Strathdee G, Higgs R et al. Managing the difficult patient: practical suggestions from a study day. J R Coll Gen Pract 1988; 38: 349–352.[Web of Science][Medline]
9 Neal R. Frequent attenders: who needs treatment? Br J Gen Pract 1996; 46: 131–132.
10 Lyons TF, Payne BC. The relationship of physicians' medical recording performance to their medical care performance. Med Cure 1974; 12: 714–720.
11 Peppiatt R. Records and prevention for the beginner. J R Coll Gen Pract 1983; 33: 660–661.[Web of Science][Medline]
12 Bass C. Helping the frequent hospital attenders to self help. Psychiatry Pract 1995; July/August: 13–16.
13 Mathers N, Jones N, Hannay D. Heartsink patients: a study of their general practitioners. Br J Gen Pract 1995; 45: 293–296.[Web of Science][Medline]
14
Karlsson H, Lehtinen B, Joukama M. Frequent attenders of Finnish public primary health care: socio demographic characteristics and physical morbidity. Fam Pract 1994; 11: 424–430.
15 Paykel E, Priest R. Recognition and management of depression in general practice: consensus statement. Br Med J 1992; 305: 1198–1202.
16 Robinson J, Granfield A. The frequent consulter in primary medical care. J Psychosom Res 1986; 30: 589–600[Web of Science][Medline]
17
Ben Shlomo Y, White I, McKeigue P. Predictions of general practice workload from census based social deprivation scores. J Epidemiol Commun Health 1992; 46: 532–536.
18 Ellaway A, Wood S, Macintyre S. Someone to talk to? The role of loneliness as a factor in the frequency of GP consultations. Br J Gen Pract 1999; 49: 363–367.[Web of Science][Medline]
19 Thomas KB. Temporarily dependent patients in general practice Br Med J 1974; 1: 625–626.
20
Jones I, Morrell D. General practitioners' background knowledge of their patients. Fam Pract 1995; 12: 49–53.
21 Pereira Gray DJ. The key to personal care. J R Coll Gen Pract 1979; 29: 666–678.[Medline]
22 Balint M. The Doctor, His Patient and the Illness. London: Tavistock Publications, 1957.
23 McCormick J. Death of a personal doctor. Lancet 1996, 348: 667–668.[Web of Science][Medline]
24 Freeman GK, Richards SC. How much personal care in four group practices? Br Med J 1990; 301: 1028–1030.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  J. A. Bellon, A. Delgado-Sanchez, J. de Dios Luna, and P. Lardelli-Claret Patient psychosocial factors and primary care consultation: a cohort study Fam. Pract., December 1, 2007; 24(6): 562 - 569. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Bergh, A. Baigi, and B. Marklund Consultations for injuries by frequent attenders are found to be medically appropriate from general practitioners' perspective Scand J Public Health, May 1, 2005; 33(3): 228 - 232. [Abstract] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||