Family Practice Vol. 16, No. 3, 283-288
© Oxford University Press 1999
Frequent attenders in out-of-hours general practice care: attendance prognosis
The Research Unit and Department for General Practice and
a The Research Unit for General Practice, University of Aarhus, Vennelyst Boulevarde 6, DK-8000 Aarhus C, Denmark.
Vedsted P and Olesen F. Frequent attenders in out-of-hours general practice care: attendance prognosis. Family Practice 1999; 16: 283–288.
Received 26 January 1998; Revised 5 October 1998; Accepted 28 January 1999.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and methodsResultsDiscussionAppendixReferences |
|---|
Objective. We aimed to describe the use of out-of-hours service and analyse attendance prognosis for frequent attenders and other groups of attenders, and to present a concept describing frequent attendance over time.
Methods. All adult attenders in 1990 were included in a 4-year follow-up study. Frequent attenders (FAs) were defined as those 10% among the attenders who most frequently used the out-of-hours service during a calendar year (12 months). This gave an intersection point of four or more contacts for frequent attenders. Three more groups were defined according to whether they had one, two or three contacts per year. The setting was out-of-hours general practice in Aarhus County, Denmark. Data were collected from the database of the Public Health Insurance, Aarhus County. The county had approximately 600 000 inhabitants, of whom 465 000 were aged 18 years and over. The subjects were 101 321 individuals aged 18 years and over who contacted the out-of-hours service in 1990. Outcome measures were attendance per year, age and sex.
Results. FAs made 42% of the out-of-hours contacts in 1990, and 33% of those who were FAs in 1990 were also FAs in 1991. Among the 1990 FAs, 67% contacted the out-of-hours service at least once in 1991, 25% contacted the service at least once in each of the following 4 years and 7% remained FAs in the following 4 years. The probability of being an FA rose with the duration of previous frequent attendance. Age above 50 years significantly predicted future status as an FA. Females made up two-thirds of FAs, but sex did not predict future frequent attendance.
Conclusion. Frequent attendance could be regarded as a short-lived phenomenon. On the other hand, FAs were the most stable attenders of all groups of attenders over the years. Older FAs had the highest probability of remaining FAs.
Keywords. Denmark, family practice, follow-up studies, frequent attenders, health service use.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsDiscussionAppendixReferences |
|---|
The individuals who most often consult GPs are called frequent attenders (FAs). We know very little about FAs. A letter1 and an editorial2 stress the absence of thorough analysis of fundamental questions concerning FAs such as definitional and conceptual issues, FA identification strategies and long-term attendance patterns.
Most research in this area focuses on daytime FAs, but GPs' workloads are also very heavy out of hours. The resultant strains on GP resources, human as well as financial, warrants research into the underlying mechanisms, patterns and motivations of such out-of-hours patient behaviour.
The aim of this paper is to present a 4-year follow-up study of adult FAs in out-of-hours general practice care sampled in 1990. Attendance patterns are examined and compared with the patterns for other groups of attenders. The attendance prognosis is analysed on the basis of age, sex and attendance.
|
Material and methods |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsDiscussionAppendixReferences |
|---|
In Denmark, more than 97% of the population has free access to medical care through a tax-financed public health insurance. GPs work as independent contractors to the public health insurance and are paid on a fee-for-service basis during out-of-hours. The health care system is based on the GPs working as gatekeepers.3 Outof-hours care is arranged by GPs in rota systems.4 The out-of-hours period runs from 4.00 p.m. to 8.00 a.m., Mondays to Fridays, and throughout Saturdays, Sundays and holidays.
Data
Aarhus County is a region in Denmark (population 5.2 million) with approximately 600 000 inhabitants in 1990, of whom 465 000 were aged 18 and over. The county roughly mirrors the rest of the country in terms of age and sex, urbanization, social groupings, etc. For reasons of accounting, the Public Health Insurance in Aarhus County receives electronic information on all contacts with the out-of-hours service. The Public Health Insurance runs a highly reliable database of all contacts to the out-of-hours service which can be traced to the individual user via his or her personal identification number (civil registration number).3 A contact with the out-of-hours service is registered as one of three main types: telephone consultation given by a GP, a consultation at a GP's clinic or a home visit. Out-of-hours contacts are registered as telephone consultations only if the GP gives the necessary advice by telephone and does not subsequently see the patient either at his clinic or in the patient's home.
We obtained data about individuals aged 18 years and over who had contacted the out-of-hours service of the Aarhus County in 1990. These individuals were followed from 1991 to 1994.
Definition of FAs
FAs were defined as those 10% among the attenders who most frequently used the out-of-hours service during a calendar year (12 months). All attenders were ranked according to the number of contacts per year. Starting with the highest rank, individuals were included consecutively until the group consisted of 10% of all attenders. In order to avoid attenders with the same number of contacts being placed in two different groups, the intersection point between FAs and other attenders had to be between two integers. Therefore, the intersection point was moved to the integer number of contacts that defined the FA group as close to the 10% as possible. This gave an intersection point of four or more contacts per year for each of the years 1990–1994. The other attenders were placed pragmatically in three other groups with one, two and three contacts per year, respectively.
Analysis
In order to describe the prognosis for attendance, we used two outcome measures (see also the appendix). We calculated the proportion of FAs who were FAs within a period (frequent attendance) and the proportion of FAs who attended the out-of-hours service at least once within a period (attendance).
Frequent attendance and attendance were analysed in two ways. When an FA in the follow-up period did not attend one year and re-attended in a later year, we refer to this attendance as ‘irregular’. When the FA could be found every year from inclusion until the year of interest, we refer to this attendance as ‘regular’.
Data were encrypted and did not contain any personal identifiers. The investigation did not influence the treatment of the patients.
The chi-square and Kruskal–Wallis non-parametric tests were used to test differences between the groups. Probabilities of 
0.05 were chosen as significance levels. Logistic regression was used to calculate the predictability of future frequent attendance of the independent variables: age in 1990, sex and being an FA in 1990. Age was divided into four design variables (18–30, 31–50, 51–65 and 
66) with the age group 18–30 as a reference. Frequent attendance in 1990 was dichotomized into FA or non-FA. Ninety-five per cent confidence intervals were calculated for the odds ratios (OR). All variables were entered in the model. Data were processed and analysed in SAS ver. 6.11 and SPSS ver. 8.
|
Results |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsDiscussionAppendixReferences |
|---|
In 1990, 101 321 individuals aged 18 and over contacted the out-of-hours general practice at least once, corresponding to 22% of the adult population. In all, 284 746 different individuals contacted the out-of-hours service at least once within the 5-year period (approximately 60% of the adult population).
Table 1
presents descriptive data about attender groups and contacts with the out-of-hours service in 1990. The FAs accounted for 42% of all contacts. The mean age and the female/male ratio increased proportionally with the number of contacts per year.
|
Table 2
shows the 1-year attendance and frequent attendance of the attenders in 1990: 66.5% of the FAs contacted the service again in 1991 as compared with 34.8% of the other attenders.
|
Figure 1
shows the 4-year follow-up on the regular and irregular attendance and frequent attendance of the FAs defined in 1990. Among the 1990 FAs, 25% showed regular attendance by having contacted the out-of-hours service at least once during each of the years from 1990 to 1994, whereas 42% showed irregular attendance. The irregular frequent attendance shows that 16% of FAs in 1990 were also FAs in 1994, and the regular frequent attendance shows that 7% were FAs for all 5 years.
|
In 1991, with 2 years as FAs, 64% were females and 36% males. In 1994, after 5 years of regular frequent attendance, 70% were females and 30% were males.
An FA in 1990 had a 0.33 probability of being an FA in 1991. With a regular FA duration of 2, 3 and 4 years, the corresponding probabilities of becoming an FA the next year were 0.49, 0.62 and 0.70, respectively.
Table 3 shows the result of the logistic regression analysis predicting frequent attendance in the follow-up period 1991–1994 when the model included age, sex and status as an FA in 1990. Frequent attendance in 1990 was the best predictor of frequent attendance in one of the following years. Age above 50 years in 1990 also predicted frequent attendance in one of the following years. Sex did not predict future frequent attendance.
|
Table 4
shows the results of the logistic regression analysis predicting regular frequent attendance for 2, 3 and 4 years: the FAs sampled in 1990 were FAs again in 2–4 consecutive years after 1990. The model included frequent attendance in 1990, age in 1990 and sex. Frequent attendance in 1990 was a powerful predictor of regular frequent attendance. All age groups had increased ORs for regular frequent attendance, especially so because the age group 51–65 had the highest ORs. Sex did not predict regular frequent attendance.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsDiscussionAppendixReferences |
|---|
One-fifth of the adult population contacted the out-of-hours service in 1990. In the five-year period 1990–1994, 60% of the adult population contacted the out-of-hours service at least once. The FAs accounted for 42% of all contacts with the service in 1990. Other general practice studies have found that FAs (4.5–25% of the patient population) accounted for 21–67% of all consultations within a period of 3–12 months.5–9
The FAs were characterized by a high rate of use of the out-of-hours services: two-thirds of 1990 FAs also called upon the service in 1991, which is more than in any of the other groups; 42% contacted the service again in 1994; and 25% were regular attenders making at least one annual contact during the following 4 years.
The 1-year regular frequent attendance of FAs defined in 1990 was only 33%, but only 7% fell within the category of regular frequent attendance throughout the entire 4-year period. For most individuals, frequent attendance therefore could be regarded as a short-lived phenomenon. In 1965, Semmence found that 32% of 430 high utilizers defined for a 4-month period remained high utilizers for the next 2 years.5 Ward et al. reported that among Australians defined as FAs on a 6-month basis, 30% were still FAs 11 months later.10 Andersson et al. found that on average 20% of FAs defined in 1991 had been FAs during at least one of the three previous years.11 In out-of-hours clinical practice working with FAs will therefore include a large group of ‘temporary’ FAs. On the other hand, in the logistic regression model including age, sex and FAs in 1990, the latter parameter was the best predictor of frequent attendance in one of the following four years.
The propensity of being an FA increased with the previous duration of one's status as an FA. A study of out-patient medical care utilization in the USA found that the probability of being a high utilizer was 0.54 after 1 year, 0.65 after 2, 0.70 after 3, and 0.76 after 4 consecutive years.12 Another US study of all physician contacts (office, home, hospitals) found a probability of 0.37 after 1 year and 0.57 after 2 years (own calculation).13
Females made up the majority of attenders in the groups as a whole and were particularly frequent among FAs. This has also been found in other studies.6,11,14 Among regular frequent attenders, the proportion of females increased during the study period. Surprisingly, when controlling for age and FA in 1990, sex did not predict regular frequent attendance throughout the period. The women's probability of regular frequent attendance was therefore not higher than that of the men. This is in conflict with the results of another follow-up study.10
The mean age of the FAs in 1990 was significantly higher than that of the other attenders. Both logistic regressions showed age to predict future frequent attendance: age above 50 years predicted frequent attendance in one of the following years, although only to a small degree. As far as regular frequent attendance throughout the study period is concerned, those in the age group 51–65 years gave the highest ORs. We may therefore conclude that 51–65-year-old patients who have been FAs for 1 year will have the highest propensity to remain FAs irrespective of their sex. This conclusion must, of course, be seen in the light of the lack of data on mortality and migration; the former of which may have contributed to the low OR among patients aged above 65 years
A definition of FAs as those 10% among the attenders who most frequently used the out-of-hours service during 12 months gave an intersection point of four or more contacts per year for all 5 years. This observation is in agreement with previous findings.5,10 Our definition was adjusted to be an integer number of contacts to avoid the possibility of misclassification.
We excluded individuals under 18 years because children's attendance is often strongly influenced by their parents.
To our knowledge, this study is the first to present a comprehensive follow-up of out-of-hours FAs. The figures of regularity illustrate the need for a method with which to identify FAs who will continue to be FAs (high regular frequent attendance), and not to treat all FAs in one year as a homogeneous group.13,15 Future research must focus on how FAs are characterized according to other variables and on the influence of these variables on attendance.
|
Appendix |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsDiscussionAppendixReferences |
|---|
The following table shows the fundamental concept of describing future attendance of frequent attenders (FA) defined at t0. The FAs are then followed by means of measurements of attendance at t1, t2 ... until ti.
|
|
Acknowledgments |
|---|
We thank Mr Jørgen Nørskov Nielsen, Head Clerk at The Public Health Insurance, Aarhus County, for help with data collection.
|
References |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsDiscussionAppendixReferences |
|---|
1 Neal RD. Frequent attenders (Letter). Fam Pract 1995; 12: 370.
2 Neal R, Dowell A, Heywood P, Morley S. Frequent attenders: Who needs treatment? Br J Gen Pract 1996; 46: 131–132.
3 de Fine Olivarius N, Hollnagel H, Krasnik A, Pedersen PA, Thorsen H. The Danish National Health Service Register. Dan Med Bull 1997; 44: 449–453.[Web of Science][Medline]
4
Olesen F, Jolleys JV. Out of hours service: the Danish solution examined. Br Med J 1994; 309: 1624–1626.
5 Semmence A. Chronic high users in a general practice. A preliminary study. J R Coll Gen Pract 1969; 17: 304–310.[Medline]
6 Wamoscher Z. The returning patient. A survey of patients with high attendance rate. J R Coll Gen Pract 1966; 11: 166–173.
7 Courtenay MJ, Curwen MP, Dawe D, Robinson J, Stern MJ. Frequent attendance in a family practice. J R Coll Gen Pract 1975; 24: 251–261.[Medline]
8
Karlsson H, Lehtinen V, Joukamaa M. Frequent attenders of Finnish public primary health care: sociodemographic characteristics and physical morbidity. Fam Pract 1994; 11: 424–430.
9 Browne GB, Humphrey B, Pallister R, Browne JA, Shetzer L. Prevalence and characteristics of frequent attenders in a prepaid Canadian family practice. J Fam Pract 1982; 14: 63–71.[Web of Science][Medline]
10
Ward AM, Underwood P, Fatovich B, Wood A. Stability of attendance in general practice. Fam Pract 1994; 11: 431–437.
11 Andersson SO, Mattsson B, Lynoe N. Patients frequently consulting general practitioners at a primary health care centre in Sweden —a comparative study. Scand J Soc Med 1995; 23: 251–257.[Web of Science][Medline]
12 McFarland BH, Freeborn DK, Mullooly JP, Pope CR. Utilization patterns among long-term enrollees in a prepaid group practice health maintenance organization. Med Care 1985; 23: 1221– 1233.[Web of Science][Medline]
13 Densen PM, Shapiro S, Einhorn M. Concerning high and low utilizers of service in a medical care plan, and the persistence of utilization levels over a three year period. Milbank Mem Fund Q 1959; 37: 217–250.[Web of Science][Medline]
14 Backett EM, Heady JA, Evans JC. Studies of a general practice (II). The doctor's job in an urban area. Br Med J 1954; 228: 109–115.
15
Schrire S. Frequent Attenders—A Review. Fam Pract 1986; 3: 272–275.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  J. A. Savageau, M. McLoughlin, A. Ursan, Y. Bai, M. Collins, and S. B. Cashman Characteristics of frequent attenders at a community health center. J Am Board Fam Med, May 1, 2006; 19(3): 265 - 275. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Vedsted, H. T. Sorensen, J. N. Nielsen, and F. Olesen Variation in proportion of frequent attenders between Danish general practices Scand J Public Health, May 1, 2004; 32(3): 188 - 193. [Abstract] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||