Ali so pogosti obiskovalci za zdravnika
tudi težavni bolniki? Are frequent attenders also heartsink patients?
Avtor / Author	Janko Kersnik1, Marija Kuralt2
Ustanova / Institute	1	Medicinska fakulteta Univerze v Mariboru, Katedra za družinsko medicino 2	Zdravstveni dom Kranj
Izvleček
Abstract
Ključne besede:
osnovno zdravstvo, družinska medicina, odnos zdravnik-bolnik, obiski, stres, konzultacija, težaven bolnik, pogosti obiskovaleci Key words:
primary care, family practice, physician-patient relations, office visits, stress, consultation, difficult patients, frequent attenders
Clanek prispel / Received
05.07.2008
Članek sprejet / Accepted
19.10.2008
Naslov za dopisovanje / Correspondence
Prof. Janko KERSNIK, PhD, MSc, medical doctor
Medicinska Fakulteta Maribor Katedra za družinsko medicino Koroška ul. 2 SI-4280 Kranjska Gora. Tel.: ++ 386 4 58 84 603. Fax: ++ 386 4 58 84 610. E-mail: janko.kersnik@ozg-kranj.si or janko.kersnik@uni-mb.si
Namen: Pogoste obiskovalce ambulant imamo pogosto za težaven bolnike, ki za zdravnike predstavljajo velik stres. Naš namen je bil ugotoviti, če pogosti obiskovalci za zdravnike resnično predstavljajo tak stres.
Metode: Izvedli smo presečno opazovalno raziskavo v zdravstvenem domu v Kranju. S seznama registriranih bolnikov smo izbrali naključni vzorec 400 bolnikov, starih od 18 do 95 let. Izračunali smo odstotek pogostih obiskovalce v skupini težavnih bolnikov. Izdelali smo multivariatni model, ki napoveduje lastnosti težavnih bolnikov.
Rezultati: Pogosti obiskovalci so bili pogosteje starejši bolniki, ki živijo z družinami, so upokojeni, imajo kronično bolezen, rakasto bolezen, v opazovanem letu poškodbo ali akutno stanje, so potrebovali administrativno storitev (bolniški list, ponovno predpisovanje recepta itn.) in presenetljivo, pogosteje bolniki, ki jih je zdravnik označil kot prijetne. Regresijska analiza je pokazala, da večja starost, alkoholizem, veliko število obiskov pred
Purpose: Frequent attenders were often labelled as "heartsink patients", who produce feelings of stress in their general practitioners (GPs). Our aim was to determine how often GPs label frequent attenders as difficult patients and to outline the characteristics of frequent attenders.
Methods: We performed a cross sectional survey in general practice in Kranj, Slovenia. We draw a random sample of 100 frequent attenders and 300 non-frequent at-tenders aged 18 to 95 years from the GP list. Their GP have them a heartsink rating: very nice, nice, difficult, very difficult. We developed a regression prediction model for heartsink patients.
Results: Frequent attenders were more likely patients labelled by a GP as nice or very nice patients. They were more likely to be older; to live with their families; to be retired; to have chronic disease, cancer, an injury or acute disease during the study year; and to need administrative services such as sickness leave
opazovanim letom in manj pogosti obiski v opazovanem letu neodvisno napovedujejo, da jih bo zdravnik označil kot težavne bolnike.
Zaključek: Težavni bolniki niso zgolj podskupina pogostih obiskovalcev, ali celo lastnost pogostih obiskovalcev, zato bi morali njihove lastnosti podrobneje proučiti, da bi zdravnikom omogočili prenesti stres zaradi stikov s takimi bolniki. •
certificates or repeat prescriptions. Regression analysis showed that older age, alcohol abuse, a higher number of visits in the year before the study year, and not being a frequent attender in the study year predicted assignment to the difficult patient group made by the GP.
Conclusion: Heartsink patients are not a simple subgroup of frequent attenders. Nor is the ability to induce heartsink a characteristic of frequent attenders. The characteristics of heartsink patients should be further examined to help GPs surmount the job stress caused by them. •
Introduction
Frequent attenders are an interesting group of patients because of their disproportionate health care consumption and subsequent costs to providers, insurers and society (1-4). Surveys have shown that frequent attenders are predominantly older, female and more frequently have a lower socio-economic status (5). Frequent attenders have been studied for their psychological, family and social characteristics and for the effect of physician practice style on a number of visits made by them (6, 7). They are more likely to come from distressed families and to have chronic conditions (5, 8-10); to be less educated, to report satisfaction with their GP and to have higher scores of anxiety and depression and lower perceived quality of life (6, 10, 11); and to be less likely to try self-care and more likely to use other health services. Some frequent attenders might use GPs in lieu of a social network (7).
Frequent attenders are often not aware of their high consultation rates and have complex expectations of the consultation (12). It is difficult to explain the variation in the visiting rates by patient characteristics (13), but one-fifth of the variation can be explained by practice factors (14). Stewart and O'Dowd showed that over half of frequent attenders attend with clinically inexplicable problems (4).
This clinical complexity and/or the heavier workload to the practice may be responsible for the label "heartsink patients". One study found frequent attenders mistrusted GPs and reported dissatisfaction about GPs (5); another study found that frequent attenders were more satisfied with their GP than other groups (6).
O'Dowd pioneered the research on the concept of heartsink patients (15), a concept that goes hand in that with that of "heartlift" patients who involve themselves in negotiating the doctor-patient relationship, are interesting or virtuous, and have a positive effect (16). Heartsinking reflects a difficult relationship between GP and patient and might become a source of stress, burnout and low job satisfaction in GPs. Frequent attenders who are also heartsink patients might be at higher risk of missed diagnoses, but such a link has not been proven (4, 15). Although there are many anecdotal examples about heartsink patients causing stress in GPs, the body of research is small. Relatively little is known about doctors' perceptions of heartsink patients and frequent attenders.
We wanted to examine the relationship between frequent attenders and heartsink patients in general practice. We hypothesized that higher visit
rates and particular patient characteristics would lead to higher rates of being labelled by a GP as a heartsink patient.
Material and methods
The survey was conducted in an urban general practice in Kranj, Slovenia, which had 2443 patients on its list in the year 2001. Patient ages from 19 to 95 years, mean 48.7 years (S.D. = 16.4 years); 1047 (42.9%) were female, 754 (30.9%) were of non-Slovenian origin, and 19 died in the study year. The patients made between 0 and 64 visits in the year 2001: median 3 visits, lower quartile 0 visits, and upper quartile >7 visits. We counted each contact with the practice except phone consultations as a visit. Frequent attenders were identified by the method described by Westhead, namely, the upper quartile of patients who visited the practice most often in the year (17).
We draw a random sample of 100 patients from the group of frequent attenders and a random sample of 300 patients from the rest of the practice list. Deceased patients were excluded. We collected data on gender, age, marital status, ethnic group, education, employment status, type(s) of health problem(s) seen by the GP in 2001 and number of visits in 2000 and 2001 from the patients records for each sample. Each patient was subjectively evaluated by the practice GP (MK)
for the presence or absence of psychological distress and assigned a "heartsink score" from 1 (very nice patient) to 4 (very difficult patient) by the GP.
We used SPSS 13.0 package (SPSS Inc, Chicago, IL, USA) for the statistical analysis. Descriptive statistics were calculated, t-test or chi-square tests were used as appropriate; and we performed multi-variate logistic modelling to extract independent predictors for being labelled a difficult or very difficult patient. For that purpose we dichotomized the heartsink score into two groups, "nice patients" (very nice and nice patients) and "difficult patients" (difficult and very difficult patients).
Results
We extracted data from the medical records for all of the 400 patients in the two samples. Overall, the mean age was 49.1 years (S.D. = 17.2 years) but the group of frequent attenders were older (54.3 years vs. 47.3 years; p<0.001). There were 175 (43.8%) female patients, 232 (58.0%) patients with a secondary school or higher education level, 288 (72.0%) living in their families and 117 (29.3%) being non-Slovenian patients. The two samples were not significantly different in gender, education level and ethnic origin but frequent attenders were more likely to live with their families (p=0.02).
Table 1: Heartsink scores in frequent attenders and non-frequent attenders (Pearson chi-square = 6.105, df = 3, p=0,107)
Patient label	Veiy nice	Nice	Difficult	Veiy difficult
(average no. of visits in 2001)	(7.3)	(4.8)	(5.7)	(5.1)
Non-frequent attenders	20 (58.8%)	198 (76.4%)	65 (74.7%)	17 (85.0%)
Frequent attenders	14 (41.2%)	61 (23.6%)	22 (25.3%)	3 (15.0%)
Total	34 (8.5%)	259 (64.7%)	87 (21.8%)	20 (5%)
There were 201 (50.3%) patients working or having a job (of whom 112 (54.9%) were manual workers), 48 (12.0%) were currently without a job and 151 (37.8%) were retired. In the frequent attenders group there were 54 (35.8%) retired patients, 42 (20.9%) patients with a job and 4 (8.3%) jobless patients (p<0.001).
Fifty-four (13.5%) patients visited the practice because of an injury, of whom 32 (59.3%) were frequent attenders (p<0.001). 164 (41.0%) patients had an acute problem, of whom 63 (38.4%) were frequent attenders (p<0.001). 206 (51.5%) had a chronic problem, of whom 72 (35.0%) were frequent attenders (p<0.001). 174 (43.5%) were current smokers, 43 (10.8%) were problem drinkers and 3 (0.8%) had a substance abuse problem other than nicotine or alcohol. The differences in visit rates between the two samples was not statistically significant. There were 12 (3.0%) cancer patients, with 6 in the frequent attenders group (p=0.04). There were 11 (2.8%) patients with a psychiatric diagnosis. The differences in visit rates between the two samples was not statistical-
ly significant. 116 (29.0%) came for administrative reasons, 80 (69.0%) of whom were frequent attenders (p<0.001). The doctor evaluated 34 (8.5%) of patients as very nice, 259 (64.7%) as nice, 87 (21.8%) as difficult and 20 (5%) as very difficult. Surprisingly, 41.2% of very nice patients and only 15.0% of very difficult patients were frequent attenders (table 1).
Table 1. Heartsink scores in frequent attenders and non-frequent attenders (Pearson chi-square = 6.105, df = 3, p=0,107)
Assignment to the "nice patient" or "difficult patient" group can be predicted for 74.4% of patients with the multivariate model (table 2). Older age, alcohol abuse, a higher number of visits in the year before the study year and not being a frequent attender in the study year predicted assignment to the difficult patient group.
Table 2: Independent predictors for being labelled by a GP as a difficult patient (model chi-square = 52.804, goodness of fit = 387.140, df=18, p< 0.001).
Variable	B	Wald	Sig.	R	Exp. (B)
Constant	-2.6723	13.6396			
Patient chaiacteiistic					
Age	.0280	4.8556	.03	.0785	1.0284
Health pioblemc					
Alcoholism	.8152	4.0618	.04	.0667	2.2595
Patient behaviour					
No. of visits in previous year	.0821	9.7826	.002	.1296	1.0856
Frequent attendance group	-.9183	4.1623	.04	-.0683	.3992
Discussion
This study answers the question of whether frequent attendance is the prime patient characteristic that causes heartsink in GPs and offers insights in some patients characteristics that may contribute to high visit rates. The frequent attenders in our study were more likely to be older, live with their families, be retired, need administrative services (such as sickness leave certificates or repeated prescription). They had a diagnosis of either a chronic disease, cancer, injury or acute disease in the study year indicating that visiting rate did not depend solely on a patient medical condition. This is in line with the findings of previous research (5-10).
However, patients labelled very nice patients were more likely to be frequent attenders. This, too, accords with findings of previous research (15). Patients who use services more frequent are usually more satisfied with them. Perhaps we can assume that the doctor-patient relationship evolves through the repeat contacts, so that doctors themselves become more satisfied and more likely to consider frequent attenders to be nice patients. On the other hand, we have to take into account that one-quarter of the patients in the study had not visited the doctor in the year preceding the study year. Presumably, due to low rate of contacts they could not be given an extreme heartsink score by their doctor, either as very difficult or very nice. Also, in our study the number of the visits in the previous year independently predicted a heartsink label in the current years. However, a Croatian study found that only 8% of frequent attenders maintained high consultation rates throughout the three-year study period (20), which could indicate that giving a label of a heartsink patient due to high visiting rates should be continuously revisited by a doctor. This finding deserves attention in future surveys.
Regression analysis showed that frequent attendance did not predict assignment to the heartsink
group. Older age, alcohol abuse, a higher number of visits in the year before the study year and not being a frequent attender in the study year were independent predictors for being labelled a difficult patient. There are a number of potential explanations for this. Alcohol abuse, for example, could be explained by the divergent views of doctor and the patient about the core health issues. Higher numbers of visits in the preceding year may indicate the futility of some health-care interventions or the presence of clinically inexplicable health problems (4, 20). Higher patient age was only weakly associated with a heartsink label.
The study has several limitations. First, although the visit rates in our study were similar to those found in a national sample (6), the extent to which the findings of a single general practice can be generalized to other practices is questionable. Next, the assignment of heartsink scores was arbitrarily done by the practice GP and is difficult to replicate. Research in this area is hampered by the lack of a generally accepted definition for heartsink patients; the inclusion of visits where doctors do not know the patients very well and therefore cannot label them; or in case of external observers who also do not know the patients. However, we succeeded in quantifying visit rates and determining demographic and utilization characteristics of the two samples, and the high volume of patients from multiethnic backgrounds in the practice in question allowed analysis of this variable. Finally, we could not ascertain the personal psychological profile of the patients - which might play an independent role in triggering heartsink feelings in doctors - or test the impact of the GP's practice style on visit rates (7, 18, 19). As the heartsink phenomenon is predominantly a doctors' problem, future studies should focus on refining the concept of the difficult doctor-patient relationship and on understanding the psychological features of heart-sink patients. The effects of doctors' practice styles on their patients' visit rates should also be studied in more details.
In conclusions the association between frequency of attendance and being labelled a heartsink patient is not a simple linear function but a complex problem. Higher visit rates from previous years is just one of the factors which might be involved in less satisfactory doctor-patient relationships, as evaluated by the GP, but current high visit rates did not have a negative impact in this study. In order to enhance patient satisfaction, and as one of the means of improving patient involvement in their own care, we have to refine the concept of the heartsink patient, the characteristics of such
patients, and the effect they have on job stress levels in their doctors. We should not underestimate the importance of the way doctors perceive their patients on both the quality of care provided and job stress levels.
Acknowledgements
The study was partly supported by the grant "P3-0339" from ARRS. •
References
1.	Demers M. Frequent users of ambulatory health care in Quebec: the case of doctor-shoppers CMAJ 1995; 153: 37-42.
2.	Fries JF, McShane D. Reducing need and demand for medical services in high-risk persons. A health education approach. West J Med 1998; 169: 2017.
3.	Dally DL, Dahar W, Scott A, Roblin D, Khoury AT. The impact of a health education program targeting patients with high visit rates in a managed care organization. Am J Health Promot 2002; 17: 101-11.
4.	Stewart P, O'Dowd T. Clinically inexplicable frequent attenders in general practice. BJGP 2002; 52: 1000-1.
5.	Svab I. Zaletel-Kragelj L. Frequent attenders in general practice: a study from Slovenia. Scand J Prim Health Care 1993; 11: 38-43.
6.	Kersnik J, Svab I, Vegnuti M. Frequent attenders in general practice: quality of life, patient satisfaction, use of medical services and GP characteristics. Scand J Prim Health Care. 2001; 19: 174-7.
7.	Vedsted P, Christensen MB. Frequent attenders in general practice care: a literature review with special reference to methodological considerations. Public Health 2005; 119: 118-37.
8.	Portegijs PJ, van der Horst FG, Proot IM, Kraan HF, Gunther NC, Knottnerus JA. Somatization in frequent attenders of general practice. Social Psychiatry & Psychiatric Epidemiology 1996; 31: 29-37.
9.	Heywood PL, Cameron BG, Cameron IH, Dowell AC. An assessment of the attributes of frequent attenders to general practice. Fam Pract 1998; 15: 198-204.
10.	Smucker DR, Zink T, Susman JL, Crabtree BF. A framework for understanding visits by frequent attenders in family practice. J Fam Pract 2001; 50: 847-52.
11.	O'Dowd T. Heartsink patients - optimising care. Practitioner 1992; 236: 941-2.
12.	Hodgson P, Smith P, Brown T, Dowrick C. Stories from frequent attenders: a qualitative study in primary care. Ann Fam Med. 2005; 3: 318-23.
13.	Andersson SO, Lynöe N, Hallgren CG, Nilsson M. Is frequent attendance a persistent characteristic of a patient? Repeat studies of attendance pattern at the family practitioner. Scand J Prim Health Care. 2004; 22: 91-4.
14.	Vedsted P, S rensen HT, Nielsen J, Olesen F. Variation in proportion of frequent attenders between Danish general practices. Scand J Public Health 2004; 32: 188-93.
15.	O'Dowd T. Five years of heartsink patients in general practice. BMJ 1988; 297: 528-30.
16.	O'Riordan M, Skelton J, de la Croix A. Heartlift patients? An interview-based study of GP trainers and the impact of 'patients they like'. Fam Pract (In press).
17.	Westhead JN. Frequent attenders in general practice: medical, psychological and social characteristics. J R Coll Gen Pract 1985; 35: 337-40.
18.	Roos NP, Carri re KC, Friesen D. Factors influencing the frequency of visits by hypertensive patients to primary care physicians in Winnipeg. CMAJ 1998; 159: 777-83.
19.	Neal R, Dowell A, Heywood P, Morley S. Frequent attenders: Who needs treatment? Br J Gen Pract 1996; 46: 131-2.
20.	Vrca Botica M, Kovacid L, Kujundzid Tiljak M, Katid M, Botica I, Rapid M, Novakovid D, Lovasid S. Frequent attenders in family practice in Croatia: retrospective study. Croat Med J. 2004; 45: 620-4.