| Frequent attenders in primary health care: A cross-sectional study of frequent attenders’ psychosocial and family factors, chronic diseases and reasons for encounter in a Finnish health centre | ||
|---|---|---|
| Prev | Chapter 2. Review of the literature | Next |
Why, whenever there is an outbreak of diarrhoea or an influenza epidemic, do the same damn patients seem to get it every time?" | |
| Manton (1985) |
Backett et al. reported in 1954 that 16% of patients in a general practice had ten or more consultations per year and were responsible for 52% of the doctor’s workload (Backett et al. 1954). Since then, numerous studies have examined a variety of associated characteristics of frequent attenders, indicating that frequent attenders comprise a highly heterogeneous group of patients with a wide variety of needs (Neal et al. 1996). They have high rates of somatic diseases, mental disorders and social difficulties. These patients are likely to have multiple complex problems, often including chronic somatic diseases with or without psychological and social problems. (Schrire 1986, Gill & Sharpe 1999.)
Neal et al. (1996) note that, despite extensive research, it has not been established whether frequent attendance is simply behavior at one end of the normal distribution of consulting frequency or something more special. There are two hypotheses about frequent attender patients. Firstly, they may be individuals behaving appropriately in response to real need, who happen to be at the top end of the consulting spectrum. Secondly, they may be deviant individuals, who cause an unnecessary and unwelcome workload and with whom “something must be done”. (Neal et al. 1996.)
Frequent attenders are often considered “heartsink patients” (O’Dowd 1988) or hateful patients (Groves 1978). These names reflect the frustration GPs often feel with these patients, who are never cured in spite of the GP’s persevering efforts at care. More synonyms of frequent attenders are listed in Table 1.
The studies of frequent attendance have been reviewed by Schrire (1986), Karlsson et al. (1994), Karlsson (1996) and, more recently, by Gill & Sharpe (1999).
It is difficult to define a frequent attender. The first difficulty is how to define consultation and the second difficulty is how to define the cut-off point in the numbers of consultations. Thirdly, one must define the observation period for which the consultation rate is calculated. (Neal et al. 1996.)
In the previous studies of frequent attenders, face-to-face contacts with GPs have been defined as consultations, and the observation period has mostly been one year (Table 2). Some researchers have included home visits (Heywood et al. 1998) and telephone contacts (Dowrick et al. 2000) as consultations.
Courtenay et al. (1974) based their definition on the upper quartile of visits stratified by age and sex. The top decile of consultations stratified by age and sex has been used in some studies as the criterion of frequent attendance (Westhead 1985, Von Korff et al. 1992, vab & Zaletel-Krakelj 1993). Most of the studies have used more or less arbitrary numerical definitions of frequent attendance, varying from five to 20 consultations per year (Table 2). Goodridge (1982) used fat folders (100 g or more) to define frequent attenders. Dowrick et al. (2000) defined frequent attenders as patients having an annual rate of consultation over twice as high as the practice’s sex- and age-related mean.
The use of health care services is increasing in most countries, and the concept of frequent attender may be viewed as a phenomenon changing over time.
Table 1. Synonyms for frequent attender.
| Synonym | References |
|---|---|
| Black hole | Lin et al. 1991 |
| Chronic complainer | Rittelmeyer, Jr. 1985 |
| Chronic doctorer | Kemp 1963 |
| Chronic neurotic | Schrire 1986 |
| Clinger | Groves 1978 |
| Constant attender | McCormick 1972 |
| Crock | Drossman 1978 |
| Demander | Groves 1978 |
| Denier | Groves 1978 |
| Doctor promiscuity | Manton 1985 |
| Doctor shopper | Demers 1995 |
| Familiar face | Kemp 1963, McCormick 1972 |
| Frequent attender | McArdle et al. 1974, Hood & Farmer 1974, Browne et al. 1982 |
| Frequent clinic attender | Toomey et al. 1982 |
| Frequent user | Fenstermacher 1984 |
| Frustrating patient | Lin et al. 1991 |
| Habitual frequent attender | Westhead 1985 |
| Hard core patient | Backett et al. 1954 |
| Hateful patient | Groves 1978 |
| Heartsink patient | O’Dowd 1988, Neal et al. 1996, |
| Help rejector | Groves 1978 |
| High attender | Freer et al. 1985 |
| High consulter | Morris et al. 1992 |
| High consumer | Borgquist et al. 1993 |
| High user | Semmence 1969, Schrire 1986 |
| High utilizer | Weimer et al. 1983, Katon et al. 1990 |
| Inappropriate utilizer | Olbrisch 1977 |
| Misuser | Wagner & Hendrich 1993 |
| Multi-user | Borgquist et al. 1993 |
| Obnoxious patient | Groves 1978 |
| Overutilizer | McKinlay 1972 |
| Patient with fat folder | Goodridge 1982 |
| Problem patient | Drossman 1978, Larivaara 1988, Lin et al. 1991 |
| Repeater | Andren & Rosenquist 1985 |
| Returning patient | Wamoscher 1966 |
| Self-pitier | Groves 1978 |
| Thick-file case | Franklin 1971 |
| Turkey | Drossman 1978 |
Table 2. Review of the previous studies of frequent attenders. Study populations, definitions of FA, key questions and main results.
| Author(s), year | Study population | Definition of FA | Key questions, hypothesis | Main results |
|---|---|---|---|---|
| Wamoscher 1966 | Population of two settlements in the south of Israel. 169 FAs were compared with other patients. | FA: 10 or more consultations per year (including home visits). | A one-year survey of attendance rate and related problems. | FAs (14.4%) were responsible for 48.8% of all doctor / patient contacts. Number of consultations (mean) 16.2 versus (vs.) 2.9 (FAs vs. other patients). 61% of FAs were women. 43% of FAs came with purely organic complaints. Functional complaints were more common among female FAs. FAs with functional complaints visited more frequently. There were ”FA families” in this study, including one family of four generations. |
| Semmence 1969 | 1136 patients of all of the two GP practice patients (2468) who consulted the GP during one quarter of a year in Abingdon, UK. FAs: 165 males, 265 females, altogether 430. No COs. | FAs: 3 consultations per quarter of a year (12 times a year). | Aim: to identify FAs and to compare the results of treatment between FAs with psychiatric conditions and FAs with other conditions. A two-year follow-up of FAs was included. | 165 male FAs (13.9%) made 64% of male consultations. 265 female FAs (21% made 1310 (69%) of female consultations. No social class associations with FA was found. 13% of FAs had a mental diagnosis and they accounted for 14% of visits. In follow-up, 32% of FAs remained FAs over the next two years. Of the ”mental high users”, 80% remained FAs. The results of treatment of psychoneurotic illnesses were poor. Patients with psychoneurotic illness are most likely to remain FAs. |
| Courtenay et al. 1974 | A random sample of 382 patients in a two-doctor practice of 3400 patients in south London, UK. 98 FAs, 284 (rest of sample) COs. | FAs: contacts equal to or greater than the upper quartile for age and sex. | The purpose of the study was to determine whether FAs had any special characteristics compared with the rest of the sample. | 98 FAs (26% of the sample) made 61% of contacts, FAs were more common in smaller than larger families. Slight evidence of clustering of FAs within families. Marital status did not differ between groups. Distance from surgery was not related to FA. FAs did not associate with the time on GP’s list. 43% of FAs had at least one major somatic diagnosis. 27% of FAs had al least one major psychological diagnosis. 6% had both a somatic and a psychological diagnosis. Three-year follow-up revealed a slight decrease of FA in the followed sample of 40 families (35% to 24%). |
| Hood & Farmer 1974 | Six-doctor practice in Birmingham, UK. Patients 25–40 year old. 23 FAs, 23 COs. | FA: 5 visits or more per 3 months (20 per year). CO: age and sex-matched, one visit or under per 3 months (= < 4 per year). | The study attempted to identify differences between FAs and infrequent attenders, including attitudes towards health, doctors, themselves and reasons for consulting. | Social class distribution similar. Fewer FAs married (13 vs. 21) (p<0.005). Mean number of children of married subjects equal. GHQ score higher than CO (p<0.001). Eyschenk Personality Inventory points NS. Osgood Semantic Differential difference (p<0.05): low self-confidence, negative evaluation for help, present situation of life rated worse. Less than half labeled as psychiatric cases by GP. |
| McArdle et al. 1974 | A sample of 33 FAs among 11400 patients of four GPs in HC in Glasgow, UK (children and pregnant women excluded). Semistructured interview. 33 FAs, no COs. | FA: 12 or more visits per year, no CO group. | An interview study of a sample of FAs to find out the principal complaints, medical (physical or nervous/social or multiple) problems and sociodemographic backgrounds of FAs. | 33 FAs were interviewed. FAs consume a substantial portion of HC resources. Problems of unemployment, loneliness, housing and alcohol were found. 58% of FAs had psychological problems. Of FAs, 42% had been referred to a psychiatrist during the 20 preceding years, 48% had physical-social complaints, 36% had nervous-social complaints and 15% had multiple complaints. |
| Smedby 1974 | A sample of Swedish population over 15 years old 1963. Numbers of FAs and COs N.A. | FA: 7 or more visits per year. CO: all interviewed persons. | Combined interview and register-based study among a sample of adult Swedish population concerning their illnesses and use of health care services and related factors. | FAs (10% , men 8%, women 11%) made 43% of visits. Age distribution: highest proportion in the 55–64 year group (15%). Place of residence: FAs were more frequent in large cities. Difference between social groups: FAs were more frequent among the lowest social group. Self-perceived health, symptoms and number of illnesses related to FA. FAs had more hospital care days and health insurance days. One third of FAs had ”nervous disorders”. |
| Videman et al. 1976 | A random sample of 1927 patients in a HC in Kuusamo, Finland, during 1969–1970. Emergency and on-call visits were excluded. | FA: 13 or more visits per two years. CO: other patients. | The aim of the study was to find out the accumulation of the use of health care services and related factors in GPs’ surgeries. | FAs (3.9% of patients, 2.7% of population) made 19.8% of visits. Neoplasms, mental illnesses and diseases of the cardiovascular and locomotive systems were common among FAs. |
| Browne et al. 1982 | A random sample of 200 FA (47% of all FAs) in a Canadian Family Practice (9313 patients) compared to 200 modal users (7% of all modal users) and 200 zero users (7% of all zero users). | FA: 9 or more visits per year. Modal user: 1–2 visits per year. | The purpose was to explore the prevalence and characteristics of FAs in a Canadian health service compared to modal and zero users. | FA (4.5%) used 21% of visits. FAs were more physically, socially and emotionally distressed and more of them were single. FAs had more problems of self-esteem, and they were more externally controlled, had greater degree of family dysfunction, tended to be unemployed, retired or mothers of infants, had low incomes. FAs tended to be high users of other services and used more pills. FAs presented more emotional and digestive system complaints. |
| Goodridge 1982 | A sample of one doctor’s patients in an urban five-doctor group practice of 12850 patients, UK. | Fat-folder patients: patient folder weight 100 g or more. Female fat-folder patients had age- and sex-matched controls. | The null hypothesis was that patients with fat folders did not have special characteristics. | 4% of female patients and 0.5% of male patients had fat folders. 54.9% of female fat folder patients were FAs ( > 10 visits per year). There were more divorced women, changing of the doctor, use of psychotropic medication among fat-folder patients than among COs. Evidence of marital disharmony was found among fat-folder patients. One third of fat-folder patients were patients with mainly organic illnesses and another one third were patients with equally organic and emotional illnesses. |
| Westhead 1985 | A sample 1491 patients (every second) of practice patients in a two-man GP practice, Whitehaven, Cumbria, UK. | FA: the top decile of most frequent attenders of each decade age group for each sex during four years.CO: same number of age- and sex-matched non-frequent attenders. | The aim was to compare a range of social, medical and psychological characteristics of habitual FAs with those of COs matched for age and sex. | 109 FA patients, 86 controls. FAs (10%) used 30% of consultations. FAs’ visits: male 7.0–11.7, female 8.6–11.0, COs’ visits: male 0.9–1.7, female 1.0–1.7. Marital breakdown more common among FAs. FAs’ neuroticism scores (Eysenk) higher than among COs. 45% of FAs had a minor neurotic illness in GHQ vs. 15% of COs. 48% of FAs assessed their physical health as poor vs. 9% of COs. The most common category of physical illnesses was cardiovascular diseases. |
| Larivaara 1987 | A rural Finnish health centre in Kolari.Population 4913 (all ages).370 FAs. | FA: 8 or more visits per year. CO: whole population. | The purpose was to determine the number of FAs in a HC and the features characterizing FAs and the doctors’ possibilities for treating FAs and the effectiveness of treatment measures of FAs. | FAs 7.5% of whole population of municipality.Of FAs 56% were women. FAs used 31.8% of visits. Mean number of visits was 10.3 per year. More married and widowed or divorced persons and persons from lower socioeconomic class among FAs. 48% visited general care units, 6% mental care units. 48% of FAs had psychosomatic illness or symptoms. 75% of FA had need for a treatment plan. |
| Savonius 1988 | An urban Finnish health centre in Espoo. Patients seen during one month constituted the study population (all ages, n = 4051). 349 FAs. | FA: 10 or more visits per year. No CO population. | The aim was to determine the prevalence of FAs, the reasons for encounter by FAs and the continuity of care among FAs. | FAs accounted for 8.6% of patients during one month. 65% were women. Continuity of care was low: continuity index 0.47 among FAs. Frequent attendance and related problems were not usually mentioned in the patient records. |
| Von Korff et al. 1992 | HMO primary care clinics, State of Washington, US. Patients 18–75 year s old. 145 depressive high utilizers selected. No COs. | FA: the top decile of the numbers of ambulatory care visits stratified by age and sex. No CO group. | Hypothesis: Depressive FAs whose depressive symptoms improve, show a more favourable course of disability. | FAs’ depression and disability show synchrony over time. Depression and disability are controlled by some other factor (chronic disease or personality disorder). The causal relationship between depression and disability warrants more research. |
| vab & Zaletel-Krakelj 1993 | A rural health centre in Slovenia. A random sample of the population (623 persons, 304 men and 319 women). | FA: the top decile of most frequent attenders in each age group CO: the rest of the sample. 188 FA patients, 320 COs. | The aim was to examine morbidity, symptoms, prevalence of superficial contacts with GP, referrals and prescriptions among FAs compared with COs. | The mean contact frequency of FAs was 11.8. FAs made 67% of visits in the study population. The three main ICD categories as the reason for encounter were: 1) respiratory, 2) cardiovascular, 3) traumas.When all reasons were considered: 1) respiratory, 2) cardiovascular, 3) musculoskeletal. FAs had more mental, malignant and gastrointestinal diagnoses as reasons for encounter than COs. There were more superficial contacts and more referrals to specialists among FAs. The differences in prescribing drugs were not significant. |
| Ward et al. 1994 | Three general practices in Western Australia. All patients during a 6-month period and again after 11 months during a 6-month period were studied (n = 7199). | FA: 7 or more visits per 6 months (14 per year), medium attenders 4–6 visits, low attenders 1–3 visits. | The aim was to examine the stability of attendance patterns in terms of the number of visits and diagnoses over an 18-month period. | 8% of patients were FAs during both 6-month periods. 22% of FAs during the first period remained FAs during the second period. Long-term FAs were older, there were more women among them and they suffered from chronic diagnoses, such as circulatory, musculoskeletal and mental disorders. Short-term FAs suffered from more self-limiting conditions, such as depression and pregnancy. |
| Andersson et al. 1995 | An urban Swedish health centre in Umeå in 1991. 179 FAs and 179 age- and sex-matched COs of all ages were studied. | FA: 5 or more consultations during a year. CO: 1–4 consultations during a year. | The aim was to describe the sociodemographic patterns, consultations and nature of problems for FAs in a HC. | 1.7% of inhabitants were FAs and they accounted for 15% of consultations and the average consultation rate was 6.3 per year. GPs used more time during the consultations of FAs. Problems of musculoskeletal system, psychological and social problems were the most common reasons for encounter among FAs. More of female FAs than male FAs and COs were divorced. |
| Karlsson 1996 | An urban Finnish health centre in Turku. FAs (96) and COs (466) (age between 18 and 65 y.) were selected of 1000 consecutive patients’ sample of HC patients. | FA: 11 or more visits per year. CO: other patients. | The aim was to investigate FAs multidimensionally (sociodemographic factors, physical and psychiatric illnesses, psychological distress, alexithymia) and to create a clinically useful way of grouping FAs. The study included a follow-up of 18 months. | FAs had lower vocational training and lower socioeconomic status. More disability pensioners, more physical illnesses, more mixed (physical-psychiatric) complaints among FAs. FAs more distressed ( 44% vs. 26%). 54% of FAs had a psychiatric diagnosis, mostly depression (24%) and anxiety (24%). Alexithymia associated with FA mediated by distress. Five distinct clinical groups of FAs were identified:’physical’,’psychiatric’, ‘crisis’, ‘chronically somatizing’ and ‘multiproblem’ groups of FAs. In follow-up, clear differences emerged in attendance rates, psychological distress and self-perceived satisfaction between the defined clinical groups. |
| Báez et al. 1998 | Nine general practices in Spain, a case-control study. Cases: patients with the highest decile of patient-initiated visits. Controls: patients with single patient-initiated visits.102 cases and 100 controls were selected by stratified sampling. | FA: the highest decile of patient-initiated visits. CO: patients with single patient-initiated visits. | The aim was to assess the association of chronic physical illnesses, mental disorders, life stress and sociodemographic factors with patient-initiated frequent attendance in primary health care. | Medium high life stress, chronic physical illness, mental disorder, and age associated with FA. FAs (10% of patients) accounted for 27.5% of all patient-initiated visits. Mean number of visits was 9 per year. Age, chronic physical illness, mental disorder and life stress together explained 82% of FA. FAs were older, more often females, less educated and more often widowed or divorced. There were more pensioners among FAs. Family life cycle: FAs were more often in the situation of break-up or contraction. There were no differences in family dysfunction or social support between FAs and COs. Authors emphasize biopsychosocial approach and need for qualitative research |
| Heywood et al. 1998 | A teaching general practice of 12400 patients in UK.204 FAs identified, of whom 132 were interviewed (65%). 204 age- and sex-matched controls selected, of whom 102 attended the interview (50%). | FA: 12 or more consultations per year CO: frequency 11 or fewer. Consultation: face-to-face contact between a fully registered patient and a GP, including home visits and out-of-hours contacts. | To define and identify frequent attenders and to characterize their attributes and use of services. | FAs’ (3.1% of patients) consultation frequency was 15 times a year, five times that of COs. FAs accounted for 15.4% of all consultations. Sex: FAs 86% of FAs were females vs. 52% among general population. Married: FAs 60% vs. COs 78% (p<0.01). More divorced, widowed or separated and living alone among FAs. More FAs were in lower social classes. 94% of FAs had chronic health problem vs. 39% of COs (p<0.01). FAs had more prescriptions and referrals to a hospital consultant (p<0.001). FAs were equally satisfied with services and doctor-patient relationship compared to COs. 52% of FAs were depressed on GHQ-28 compared to 29% of COs. FAs showed more distress in all modalities of well-being on NHP. |
| Neal et al. 1998 | Four general practices in Leeds, UK comprising 61000 patients. The data included all consultations (592000 consultations) during 41 months. | FA: The most frequently consulting 3%, CO: the rest of the patient population. Consultation: face-to-face doctor contacts, including home visits and planned health promotion. | The aim was to examine the distribution of the number of consultations per patient and to estimate the workload generated by FAs and to demonstrate the contribution of age, sex and practice to the likelihood of FA. | The most frequent 3% of all attenders accounted for 15% of all consultations. Females and older people were more likely to be FAs. The cut-off point of 3% equals to about 12 consultations per year. The overall appropriateness of FAs needs to be addressed. |
| Dowrick et al. 2000 | Two general practices, one of 10500 patients in Liverpool, UK and one of 24000 population in Granada, Spain. 127 FAs and 175 COs take part in the survey. No mention of those who did not want to participate. | FA: consultation rate over twice as high as the practice’s sex- and age-related mean. CO: consultation rate below the practice’s sex- and age-related mean. Consultations included telephone contacts and home visits, antenatal visits were excluded. | The aim was to assess whether FAs are more likely to be associated with depressive symptoms than with physical health problems. | Consultation frequency: FAs 12.7 vs. COs 3.7. FAs were more likely to be female (OR 2.7), widowed or divorced (OR 2.1) and from social class V (OR 3.0). No differences in education. Depressive symptoms (BDI score) higher and proportion of depressive persons higher among FAs (OR 26.6 (p<0.001)). Self-reported health lower (OR 7.6) and more physical health problems. More psychological, social, respiratory problems in ICHPPC-2. Depression, self-reported health and respiratory problems were associated with FA in logistic regression analysis. |
In the frequent attender studies, the prevalence of frequent attendance depends on the criteria used (Table 2). Information of population-based prevalence is given in some studies. Andersson et al. (1995) found 2% of inhabitants to be frequent attenders, while in Finland Larivaara considered 8% of the population as frequent attenders (Larivaara 1987, Larivaara et al. 1996). Usually, prevalence is determined as per patient population. The patient population prevalence of frequent attendance varies from 4% to 18% (Table 2).
Frequent attenders consume a substantial portion of the health care service resources in both primary and secondary health care (Table 2), (Gill et al. 1998). The mean consultation rates per year depend on the criteria used to define frequent attender. Westhead (1985) found the mean consultation rate to range from seven to 11, depending on the age group. In the study of vab & Zaletel-Krakelj (1993), the mean attendance rate was 12 per year. In the cut-off based studies, the mean consultation rates have varied from six to 16 (Table 2).
Frequent attenders cause a remarkable workload in primary health care, which can be estimated as the proportion of services used by frequent attenders. In previous studies, frequent attenders made 15%–69% of all contacts to GPs, the percentage depending on the definition of frequent attendance and the populations studied (Table 2). In the study of Neal et al. (1998) in Leeds, UK, the most frequent 1% of attenders accounted for 6% of all consultations, and the most frequent 3% for 15% of all consultations. Gill et al. (1998) reported a progressive increase in GPs’ total workload, and frequent attenders were responsible for a large part of the increase.
The frequent attenders in primary health care are frequent attenders for all health care services (Smedby 1974) and also social agencies (Corney & Murray 1988). In the study of Heywood et al. (1998), the referral rate of frequent attenders to hospital was more than five times greater than that of controls. Larivaara found that 48% of frequent attenders used general hospital services, but only 6% mental care services (Larivaara 1987, Larivaara et al. 1996). vab & Zaletel-Krakelj (1993) found more referrals to specialists in their frequent attender group. Consistently high users (26%) among an elderly population accounted for 55% of hospital admissions (Freeborn et al. 1990). In a study from Alaska, high utilizers had a higher risk of hospitalization during a six-year follow-up period than low utilizers (Nighswander 1984).
At the one end of the continuum of frequent attendance, doctor-shopping frequent attenders can be defined as patients with many visits to health care in a wide variety of health care sources, e.g. primary health care, private doctors and specialized care (Demers 1995). In Canada, 1% of patients appeared to be doctor shoppers, who received ambulatory care from more than 20 physicians. They received 10 times more medical services than the overall patient population, and their mean costs per patient was also 10 times higher. The most frequent diagnoses among doctor-shopping frequent attenders were anxiety, abdominal pain, drug and alcohol dependence and depression. 82% of doctor shoppers had at least one diagnosis related to mental disorders or ill-defined symptoms. (Demers 1995.) Doctor-shopping behavior associates with chronic illnesses, mental disorders, somatization, hypochondriasis and disturbances in the doctor-patient relationship (Kasteler et al. 1976, Sato et al. 1995).
Out-of-hours frequent attenders (4 or more visits to out-of-hours family practice service) made up 9.5% of all attenders in an out-of-hours service in Denmark and accounted for more than 40% of the contacts and the aggregate costs (Vedsted & Olesen 1999a). Age and previous frequent attendance in out-of-hours service were risk factors for long-term frequent attendance (Vedsted & Olesen 1999b).
The frequent consulting behavior seems to persist in many cases (Gill et al. 1998, Gill & Sharpe 1999), and the persistence may be associated with emotional distress (Semmence 1969). According to Freer et al. (1985), frequent attenders’ attendance rates remained stable over a three-year period. Of the frequent attenders studied, 25% had no change, while 43% increased or decreased their use by one or two visits. Ward et al. (1994) found that 22% of frequent attenders remained frequent attenders during 18 months’ follow-up. Frequent attenders’ consultation frequency persisted at a higher level than controls’ frequency during successive years (Andersson et al. 1995, Gill et al. 1998). Kokko found that only 3% of patients were frequent attenders during a nine-year period of observation (Kokko 1988, Kokko 1990).
In various models that have explained the use of health care services, subjective perceived health, various symptoms and subjective experiences of illness are included as important predisposing factors of health care use.
Illness may be defined as the perception, evaluation, explanation, and labelling of symptoms by the patient and his family and the social network, whereas disease may be defined as the malfunctioning of biological and / or psychological processes (Rosen et al. 1982, Kokko 1988). Illness is the subjective experience of the sufferer, while disease is the pathological process (Fry 1986). Health status and the perceived severity of illness seem to be most important predictors of health care utilization (Mechanic 1979, Krakau 1991).
According to the WHO definition of health, health is more than absence of disease: it is complete mental and physical well-being (Fry 1986). René Dubos describes health as an ability to adapt one’s environment or, if necessary, to adapt to it (Dubos 1965).
Population studies have shown that women report symptoms more often than men (Mechanic 1976, Heistaro et al. 1997). Higher education associates with better perceived health (Lahelma et al. 1995). A personal perception of oneself as healthy seems to be the major factor distinguishing between non-attenders and attenders (Schrire 1986). Health need, as measured by the perceived health status and the number of health problems, was found to be consistently associated with increased utilization of medical services, both primary health care and specialist visits (Dunlop et al. 2000).
Frequent attenders have been found to show a clear correlation between self-perceived health and visits to GPs (Smedby 1974, Westhead 1985, McFarland et al. 1985). According to the study of Borgquist et al. (1993), frequent attenders had poor perceived health, and Freeborn et al. (1990) reported a similar finding among elderly high users. Very frequent attenders showed a poor quality of life on all modalities of well-being on the Nottingham Health Profile (NHP) (Heywood et al. 1998). Self-reported ill health was associated with frequent attendance in the study of Dowrick et al. (2000).
Sociodemographic background affects the person’s health behavior (Mechanic 1992, Viinamäki et al. 1997). It is well known that men’s health behavior is more risky than women’s (Kivelä & Lammi 1985).
Hood & Farmer (1974) found that insufficient positive motivation and low self-confidence make frequent attenders more likely to adopt a sick role when confronted by problems in life. The problem concerning frequent attenders in primary health care is whether the uneven distribution of care matches an uneven distribution of needs (Courtenay et al. 1974).
Various symptoms belong to the everyday life of human beings, although the experienced discomfort or threat of symptoms and the person’s coping factors finally contribute to the health behavior carried out, such as consulting a GP. Robinson & Granfield (1986) describe the differences between frequent attenders and non-frequent attenders in meeting the discomfort of symptoms and affect of mood or attention to experienced and consciously felt symptoms. The threshold of tolerance and the degree of perceived threat or anxiety generated by pain influence the decision to consult a doctor (Schrire 1986). Women seem to report more subjective symptoms than men. It may be, however, that women do not perceive any more symptoms than men, but are be more willing to report them (Mechanic 1976, Kroenke & Spitzer 1998). A strong association between somatic symptoms and psychiatric morbidity has been found (Kisely et al. 1997).
Frequent attenders report a higher number of physical symptoms than average users (McFarland et al. 1985) and cannot generate normalizing explanations for common bodily sensations (Sensky et al. 1996). A correlation between patients’ symptoms and the use of health care has been found (Smedby 1974, Robinson & Granfield 1986, Corney & Murray 1988). Frequent attenders are more likely to seek care for minor symptoms (Wagner et al. 1995).
The reason for encounter has been defined as the agreed statement of the reason(s) why a person enters the health care system, representing the demand for care by that person (ICPC Working Party 1987).
In Finland, illness was the reason for encounter in 74% of the visits to health centres in 1983 (Kekki 1983). In the study of Hagman (1981), where the International Classification of Diseases 9th revision (ICD-9) was used, the three main reasons for encounter were respiratory, cardiovascular and musculoskeletal illnesses. Respiratory and musculoskeletal illnesses and accidents were the three main reasons for encounter in eastern Finland (Kokko 1988).
International Classification of Primary Care (ICPC) offers a simple, logical and easy way to classify the reasons for encounter in primary health care (ICPC Working Party 1987, Bentsen & Hjortdahl 1991, Lamberts et al. 1992). The three most common ICPC-defined reasons for encounter by primary health care patients in Norway were musculoskeletal (20%), cardiovascular (14%) and respiratory (11%) symptoms (Claussen et al. 1994). Rokstad et al. (1997) obtained similar results. In a large survey in Australia, respiratory problems were the most important reason for consulting a GP (34%), followed by skin symptoms (21%) and musculoskeletal symptoms (16%) (Britt et al. 1998).
In Finland, Mäntyselkä (1998) studied the ICPC-coded reasons for encounter in 30 Finnish health centres and found the three main reasons for encounter to be musculoskeletal (18%) and respiratory symptoms (18%) and unspecific reasons (14%). GPs estimated psychiatric factors to be the main or background reason for encounter in 40% of the cases in Finnish health centres (Winblad et al. 1994). Mental disorders were ranked as fifth in importance among the reasons for encounter in Kokko’s study (Kokko 1988), but in the study of Mäntyselkä (1998), GPs assessed only 2% of the reasons to be psychiatric reasons for encounter. In Norway, psychiatric symptoms were the fourth most important reason for encounter (12%) (Rokstad et al. 1997). Patients seldom report psychiatric reasons for their encounters, as only about 1–4% of patients reported a psychiatric reason as their main reason to visit a GP (Lehtinen et al. 1984, Joukamaa et al. 1991).
The main reasons for frequent attenders to consult a GP are physical complaints (Karlsson et al. 1994). Frequent attenders complain more commonly about respiratory symptoms and gastrointestinal or back pain than controls (Robinson & Granfield 1986). According to Wagner & Hendrich (1993), health care misusers, referred to as frequent attenders, who received such attributes of misuse as inappropriate or psychosomatic users, had more musculoskeletal and mental disorders than controls.
Based on the ICD-9 classification, respiratory, cardiovascular and musculoskeletal diseases were the three main diagnoses found as the reasons for encounter among frequent attenders. Frequent attenders had significantly more commonly diagnoses of malignant diseases, mental disorders and gastrointestinal diseases and less commonly diagnoses of endocrine, respiratory or skin diseases than infrequent attenders. (vab & Zaletel-Krakelj 1993.) According to International Classification of Health Problems in Primary Care (ICHPPC-2), various ill-defined symptoms and signs were the main reasons for encounter among constant frequent attender patients. Musculoskeletal, circulatory and respiratory diagnoses came next on the list. (Ward et al. 1994.) In their ICPC-based study, Andersson et al. (1995) found problems arising from the musculoskeletal system and psychological and social problems to be most common among frequent attenders, often in combination, while chronic diseases, such as cardiovascular diseases and diabetes, were not connected with frequent attendance. Báez et al. (1998) conclude that over 40% of frequent attendance can be attributed to exposure to chronic physical illness, about one third to mental disorders and 15% to life stress.
In Finland, Larivaara (1987) found frequent attenders to complain mostly of ear, nose, throat or eye symptoms or musculoskeletal and general symptoms and less often of emotional problems. Also, Karlsson et al. (1994) found somatic complaints the main reason for consulting, and only 1% of frequent attenders came due to psychiatric symptoms, although both psychiatric and somatic complaints were found to be the reason for encounter in 11% of cases compared to 5% of controls.
Frequent attenders have high rates of somatic diseases, mental disorders and social difficulties. These patients’ complex problems often include chronic somatic diseases with or without psychological and social problems (Báez et al. 1998, Heywood et al. 1998, Gill & Sharpe 1999).
“Frequent attenders are sick people when considered both objectively and subjectively.” | |
| Smedby (1974) |
The presence of chronic disease is a potent risk factor for increased attendance (Schrire 1986, Báez et al. 1998, Heywood et al. 1998). Smedby (1974) found that the number of self-reported illnesses associated positively with frequent attendance. Using ICHPPC-2, Dowrick et al. (2000) found more psychological, social and respiratory problems among frequent attenders than controls. Morris et al. (1992) found high consulters to have a lower cardio-respiratory health and a higher risk factor status than average consulters.
Previously in Finland, Videman et al. (1976) found neoplasms, mental disorders, and diseases of the cardiovascular and musculoskeletal systems common among frequent attenders. Larivaara found every fifth frequent attender in a health centre to have a chronic somatic disease as their main problem (Larivaara 1987, Larivaara et al. 1996). Karlsson et al. found that frequent attenders had more physical morbidity than controls. As far as ICD-9 based diagnoses were concerned, respiratory and musculoskeletal diagnoses tended to be more frequent. Frequent attenders had more multiple diagnoses than controls. (Karlsson et al. 1994.)
According to a study in Alaska, high utilizers have a higher risk for early death than low utilizers, and half of the deaths are associated with alcohol (Nighswander 1984). The mortality of frequent attenders has also been studied among frequent emergency department visitors. Frequent emergency department visits predicted mortality within nine years. Heavy emergency department users (over four times a year) had twofold excess mortality. The excess mortality was most significantly due to violent death (suicide, probable suicide, alcohol / drug abuse), which accounted for one third of the excess mortality among frequent emergency department visitors. (Hansagi et al. 1990.)
"There seems little doubt that a considerable proportion of the complaints about which patients consult their doctors are not primarily or solely related to physical causes but originate from emotional problems." | |
| Balint (1957) |
The prevalence of mental disorders in the Finnish population varies from 16% to 29% (Väisänen 1975, Lehtinen 1988, Lehtinen 1991, Lehtinen et al. 1991, Viinamäki et al. 2000). Of GPs’ patients in the Helsinki area, 8% had a mental disorder and 14% both a mental disorder and a somatic disease (Lönnqvist & Niskanen 1972). The prevalence of mental disorders was 34% among primary health care patients in Turku in a Nordic multicentre study. The overall prevalence in the Nordic countries was lower (26%). (Fink et al. 1995.) Of Finnish health centre patients, 28% had mental disorders when the General Health Questionnaire (GHQ) was used as a measure of mental disorder (Lehtinen et al. 1984). According to Johnstone & Goldberg (1976), 32% of GPs’ patients had conspicuous psychiatric disorders when GHQ was used in the UK. The rate of mental disorders measured with Diagnostic Interview Schedule (DIS) was 22% among medical users in the Epidemiologic Catchment Area (ECA) Program study in the US, and there were high rates of affective disorders among women and notable substance abuse among men (Kessler et al. 1987).
Previously, frequent attendance has been connected with neurotic personality and neuroses (Table 2) (Polliack 1971, Schrire 1986). Kessel (1960) was the first to suggest that neurosis, social problems and intrafamily conflicts are associated with high consultation rates. Smedby (1974) found that one third of frequent attenders reported “nervous disorders”, women more commonly than men. McArdle et al. (1974) found that 58% of frequent attenders had psychological problems and 42% of them had been referred to a psychiatrist during the 20 preceding years. One fourth of frequent attenders had a major psychological diagnosis in the study of Courtenay et al. (1974). According to Báez et al. (1998), mental disorders associated with patient-initiated frequent attendance. Psychiatric morbidity increased the risk to consult a GP among female high attenders (Corney & Murray 1988). Mental disorders are also common among children who are frequently presented to the GP with physical symptoms (Bowman & Garralda 1993). Patients in contact with psychiatric services are frequent attenders of other medical services (Hansson et al. 1997), and the higher use by them can be explained by the concomitant physical symptoms and discomfort (Mechanic et al. 1982).
In Finland, Videman et al. (1976) found mental disorders to be one of the main groups of diseases among frequent attenders. Larivaara (1987) found that psychosomatic illness or symptoms were the main problem of frequent attender patient in 48% of cases. According to Karlsson et al. (1995b), 54% of frequent attenders had a psychiatric diagnosis.
In his review article, Hinkle (1973) quotes the definition of stress by Selye and Wolff. According to them, stress in biology indicates a state within a living creature, which results from the interaction of the organism with noxious stimuli or circumstances. Later on, stress and coping have conceptualized by Coelho et al. (1974) and Cassel (1976). According to Antonovsky’s theory of coping, the essential factor in coping with stress is the sense of coherence (Antonovsky 1979). Coping with stress is a multilevel process, including effects of psychological, hormonal, metabolic and immunological functions and also of autonomic and central nervous system functions (Chrousos & Gold 1992, Lindholm & Gockel 2000). The buffer effect of social support on stress has been presented by Kaplan et al. (1977) and Payne & Jones (1987).
Recently, the interactions between stress, coping and social support were reviewed by Thoits (1995). It has been claimed that life stress has increased in the industrial countries (Huttunen 1981). In Finland, nearly 15% of the population have feelings of stress (Berg et al. 1990). Daily stress associates positively with the use of health care services (Gortmaker et al. 1982).
A relationship between the accumulation of stressful life events and the onset of physical illness has been demonstrated (Rahe et al. 1964). Stressful life events also predict higher utilization. The more stressful life events test the person’s coping ability, and the more the distress the person feels, the more likely he/she is to attend his doctor (Tessler et al. 1976, Gortmaker et al. 1982, Schrire 1986, Báez et al. 1998). In the study of Robinson & Granfield (1986), frequent consulters had fewer stressful life events but they coped less well with them because they had less satisfactory family and social support. The death of the spouse is the most stressful life event one can face (Holmes & Rahe 1967), and it affects injuriously the health of the widowed person (Wan 1982).
The word distress means pain, anxiety, sorrow, acute physical or mental suffering, affliction or trouble (Webster’s Encyclopedic Unabridged Dictionary 1996). Psychological distress associates with subjective ill health (Mechanic 1979, Cockerham et al. 1988), illness behavior (Mechanic 1978), health need (Preville et al. 1998), propensity to reporting physical symptoms (Cockerham et al. 1988) and use of health services (Tessler et al. 1976, Manning & Wells 1992). Balint (1957) stated that, underlying a physical ticket of entry, there is often psychological distress. Musculoskeletal illness and poor outcome of musculoskeletal illness were connected with psychological distress (Jörgensen et al. 2000a, Jörgensen et al. 2000b).
Frequent attenders are more likely to feel psychological distress than non-frequent attenders (Browne et al. 1982, McFarland et al. 1985, Katon et al. 1990, Karlsson et al. 1995a), even among the elderly population (Freeborn et al. 1990). In the study of Katon et al. (1990), 51% of frequent attenders were considered psychologically distressed. In Finland, Karlsson et al. (1995a) found 44% of frequent attenders distressed, and the mean score of psychological distress on the Symptom Checklist – 25 (SCL-25) scale was significantly higher among frequent attenders than among controls. The psychologically distressed frequent attenders are perceived as frustrating by their physicians (Lin et al. 1991).
Studies concerning the prevalence of depression in the general population have been recently reviewed by Angst (1992), Poutanen (1996) and Rajala (1997). Altogether 10–20% of men and 15–30% of women have depressive symptoms, and the prevalence of clinical major depression varies within 2–6% among men and 1–12% among women. The ten-year prevalence of depression is estimated to be about 15% (Angst 1992). In Finland, the prevalence of depression is comparable to the international figures (Väisänen 1975, Lehtinen et al. 1990), although elderly populations have higher prevalence rates (Kivelä et al. 1988, Rajala et al. 1995, Rajala 1997). It has been said that one fifth of people have depressive episodes during their life (Lehtinen 1995).
The prevalence of depression among primary care patients varies within 12–25% (Katon 1987), and the one-month prevalence of depression among Finnish working age patients is 11% (Poutanen 1996, Salokangas et al. 1996). GPs’ ability to recognize depression is not good (Paykel & Priest 1992, Joukamaa et al. 1994, Joukamaa et al. 1995, Coyne et al. 1995, Poutanen 1996).
Depression is connected with frequent attendance (McFarland et al. 1985). Dowrick et al. found that 59% of frequent attenders were depressive compared to 5% of controls. Depressive symptoms were the major predictor of frequent attendance in this study. (Dowrick et al. 2000.) According to Heywood et al. (1998), 52% of very frequent attenders were depressive compared to 29% of controls. Of distressed frequent attenders, 24% had major depression and 17% dysthymic disorder. Two thirds of these patients had a lifetime history of major depression. (Katon et al. 1990.) In Finland, Karlsson et al. (1995b) found 24% of frequent attenders to have depression.
“It appears as if these patients believe it is better to be sick than crazy.” | |
| Wagner & Hendrich (1993) |
Lipowski defined somatization as a tendency to experience and communicate somatic distress in response to psychosocial stress and to seek medical help for it (Lipowski 1988). Due to differences in the criteria used to define somatization and the differences in the study populations, the prevalence of somatization varies from 1% to 12% (Quill 1985, Escobar et al. 1987, Lipowski 1988, Noyes et al. 1995). According to Väisänen (1975), 7% of the Finnish population have serious psychosomatic disorders and 53% have milder symptoms of somatization. Among primary care patients, the prevalence of somatization varies from 8% to 25% (Bridges & Goldberg 1985, Kirmayer & Robbins 1991, Noyes et al. 1995, Kroenke et al. 1997, Kisely et al. 1997, Escobar et al. 1998). In Finland, one fourth of health centre patients were considered to have a psychosomatic reason for their encounters with GPs (Winblad et al. 1994).
Somatizing patients have a negative perception of their health (Katon et al. 1991, Gureje et al. 1997). Somatizing patients often have psychosocial difficulties (Mechanic 1992), experience substantial distress (Noyes et al. 1995) and show enhanced sensitivity to normal physical sensations (Robinson & Granfield 1986, Barsky & Wyshak 1990, Blackwell & DeMorgan 1996). Somatizing patients tend to use bodily symptoms to communicate because they have difficulties to express their feelings in words. Alexithymia (Nemiah et al. 1976) associates positively with somatization (Bach & Bach 1995, Bach & Bach 1996, Taylor et al. 1997). Somatizing patients are characterized by abnormal illness behavior following the interpretation and attribution of bodily perceptions (Lipowski 1988, Mechanic 1992, Noyes et al. 1995).
Somatizing patients prefer general medical services to mental health services (Simon 1992) and use an excess of hospital care (Zoccolillo & Cloninger 1986, Fink 1992). Somatization frequently goes unrecognized (Quill 1985, Fink et al. 1999), and physicians tend to exclude organic etiologies through multiple tests and procedures (Quill 1985, Margo & Margo 1994), which causes the health care system a great deal of expense (Smith et al. 1986, Lipowski 1988, Ford 1992, Blackwell & DeMorgan 1996). Somatizing patients are challenging in the doctor-patient relationship, and physicians are commonly frustrated by such patients (Katon et al. 1991, Mechanic 1992, Blackwell & DeMorgan 1996).
Most of the somatizing patients have psychological symptoms and half of them meet the criteria for a current psychiatric diagnosis, typically a somatization disorder, hypochondriasis, depression or anxiety (Bridges & Goldberg 1985, Lipowski 1988, Katon & Russo 1989, Simon & VonKorff 1991, Noyes et al. 1995, Rogers et al. 1996, Fink et al. 1999).
Somatization has been connected with frequent use of health services (O’Reilly 1988, Katon et al. 1991, Ford 1992, Portegijs et al. 1996, Karlsson et al. 1997). Portegijs et al. (1996) found the prevalence of somatization to be 45% in a group of frequent attender patients in general practice. Of distressed frequent attenders, 20% were somatizers (Katon et al. 1990).
In Finland, previous studies have revealed an association between somatization and frequent attendance in primary health care. Larivaara found psychosomatic disease or symptoms the chief problem of frequent attender patients in 48% of cases in a rural health centre in Kolari (Larivaara 1987, Larivaara et al. 1996). About one fifth of frequent attenders were classified as chronically somatizing patients in an urban health centre in Turku (Karlsson et al. 1997).
Alexithymia refers to a cognitive-affective disturbance in psychic functioning characterized by difficulties in the capacity to verbalize affects and to elaborate fantasies (Taylor 1984). It comes from a Greek word meaning “no word for emotions”. The term alexithymia was first coined by Sifneos (1973). The concept of alexithymia has since been refined and developed (Taylor et al. 1991, Taylor et al. 1997). Numerous studies have shown that alexithymia is associated with many different somatic diseases and mental disorders (Taylor et al. 1997). The construct of alexithymia was recently reviewed by Taylor et al. (1991), Kauhanen (1993), Taylor et al. (1997) and Salminen et al. (1999a).
The following four dimensions are considered to constitute alexithymia: difficulty in identifying and describing feelings verbally, difficulty in distinguishing between feelings and bodily sensations of emotional arousal, paucity of fantasies and an externally oriented, concrete cognitive style. There has been a debate in the literature as to whether alexithymia is a stable personality trait or a transient state (Warnes 1986, Kauhanen 1993, Salminen et al. 1994). Primary and secondary alexithymia have been tentatively distinguished: primary alexithymia is a life-long disposition, whereas secondary alexithymia could be result from a somatic illness or some other kind of stress (Freyberger 1977).
The prevalence of alexithymia depends on the measures and cut-off points of alexithymia used and the population studied. In a normal population, the prevalence is about 10%–13%, being higher among men than among women (Salminen et al. 1999b, Honkalampi et al. 2000) and higher among the elderly (Joukamaa et al. 1996). Alexithymia is more common among less educated persons and among persons from lower socioeconomic classes (Kauhanen et al. 1993, Saarijärvi et al. 1993, Salminen et al. 1999b). Among the different groups of psychiatric patients, the prevalence varies within 38–40% (Taylor et al. 1992, Saarijärvi et al. 1993). Alexithymia correlates with increased illness behavior (Lumley et al. 1996, Lumley et al. 1997), frequently reported physical symptoms (Cohen et al. 1994) and psychological distress (Saarijärvi et al. 1993). Alexithymia also associates with somatization (Bach & Bach 1995, Bach & Bach 1996, Taylor et al. 1997), hypochondriasis (Rodrigo et al. 1989, Kauhanen et al. 1991), panic disorder (Joukamaa & Lepola 1994), depression (Honkalampi et al. 2000) and heavy alcohol use (Kauhanen et al. 1992, Kauhanen 1993).
Among frequent attenders, the prevalence of alexithymia was 25% in the study of Joukamaa et al. (1996). Among distressed frequent attenders, the prevalence was 30% (Joukamaa et al. 1996).
Hypochondriasis comes originally from a Greek word hypochondrios, meaning the upper abdomen, the presumed seat of melancholy (Webster’s Encyclopedic Unabridged Dictionary 1996). Hypochondriasis means, according to the DSM-IV classification, preoccupation of at least six months’ duration with fears of having or the idea that one has a serious disease based on a misinterpretation of bodily symptoms, despite appropriate medical evaluation and reassurance. This preoccupation causes clinically significant distress or impairment in some important areas of functioning. (American Psychiatric Association 1994.)
According to a review by Kellner (1985), constitutional factors, disease in family or in childhood and previous disease predispose to hypochondriasis, and stress may be a precipitating factor. Hypochondriacal subjects show enhanced perceptual sensitivity to illness cues, which may further enhance their concern with bodily symptoms (Hitchcock & Mathews 1992). Barsky and Klerman write about the perceptual amplification of bodily sensations and the cognitive misinterpretation by hypochondriacal patients, and they propose a new general concept of “amplifying somatic style”. They suggest this term to be used instead of hypochondriasis, which has a stigmatizing connotation. (Barsky & Klerman 1983.) Barsky & Wyshak (1990) reported enhanced somatosensory amplification among hypochondriacs.
The prevalence of hypochondriasis as a disorder varies from 0.4% to 14%, depending on the population surveyed and the methods used (Beaber & Rodney 1984, Kellner 1985, Barsky et al. 1990). The prevalences among males and females vary in different studies (Pilowsky 1970, Barsky et al. 1990). Hypochondriasis is associated with various somatic and psychological problems, especially depression (Beaber & Rodney 1984, Barsky et al. 1986b).
Hypochondriacal attitudes are associated with frequent use of medical services (Barsky et al. 1986a, Pålsson 1988) and doctor-shopping behavior (Kasteler et al. 1976). Hypochondriacal beliefs often go unrecognized in primary health care (Beaber & Rodney 1984, Pålsson 1988), although GPs are aware of their patients’ concerns and fears of disease and bodily preoccupation. Hypochondriacal patients are often described as frustrating patients (Barsky & Klerman 1983).
Franklin (1971) claimed that frequent attenders have two components that interact with each other, one being an underlying chronic but unrecognized psychiatric illness and the other a massive hypochondriacal superstructure.
The occurrence of multiple diagnoses (comorbidity) is an issue of major importance in health care. Psychiatric comorbidity has been found in chronic somatic diseases (Mayou et al. 1988, Ruoff 1996) and vice versa: mental disorders contribute to the risk of somatic symptoms and diseases and their outcome (Wells et al. 1991, Aromaa et al. 1994, Sherbourne et al. 1996). There is considerable comorbidity between various mental disorders. Comorbidity of anxiety with depression (Coyne et al. 1994, Roy-Byrne 1996, Honkalampi et al. 2000) and vice versa (Montgomery 1990) are well known. Comorbidity of somatization with other mental disorders (Escobar et al. 1998, Fink et al. 1999), e.g. depression (Rogers et al. 1996), has been reported. Posttraumatic stress disorder (PTSD) is particularly likely to be comorbid with affective disorders, other anxiety disorders, somatization, substance abuse and dissociative disorders (Brady 1997) as well as physical health problems (Beckham et al. 1998). On the other hand, psychiatric comorbidity relates to better recognition and outcome of other psychological disorders (Ormel et al. 1990). Comorbidity studies have been recently reviewed by Angold et al. (1999).
Health status is connected with age, sex, socioeconomic status, education, income, language, unemployment and many other sociodemographic factors that characterize the population (Koskinen 1995). Primary care use and costs increase in a linear fashion with declining socioeconomic status (Worrall et al. 1997). It has been found, for example, that cardiovascular risk factor profiles and increased mortality concentrate in the lower social classes but differently among men and women (Pekkanen et al. 1995).
The social problems identified among frequent attenders often connect with chronic illnesses and psychological problems (McArdle et al. 1974, Browne et al. 1982, Schrire 1986, Gill & Sharpe 1999).
Frequent attendance has been connected with older age and female gender (Table 2), although almost all morbidity studies show that females usually attend more frequently than men (Nathanson 1977). Frequent attenders are more likely to be divorced or widowed (Westhead 1985, Larivaara 1987, Larivaara et al. 1996, Báez et al. 1998, Heywood et al. 1998, Dowrick et al. 2000). Lower socioeconomic status has been connected with frequent attendance (Table 2). Lower education has been considered a risk factor for frequent attendance (Karlsson 1996, Báez et al. 1998, Dunlop et al. 2000) but not in all studies (Dowrick et al. 2000). Frequent attenders were more likely to be unemployed (McArdle et al. 1974, Browne et al. 1982) or retired or to be on disability pension (Browne et al. 1982, Karlsson et al. 1994, Báez et al. 1998). Frequent attenders were more common in large cities than in rural areas (Smedby 1974, Vuori et al. 1983). Frequent attenders have been found to live alone (Heywood et al. 1998) and to have problems with housing and alcohol use (McArdle et al. 1974). Cultural differences in illness behavior will influence attendance patterns (Schrire 1986).
Social support is defined as information leading the subject to believe that he is cared for and loved, esteemed, and a member of a network of mutual obligations (Cobb 1976). There is a lot of evidence of the connections between social support and physical and mental health (Kaplan et al. 1977, Holahan & Moos 1981, Payne & Jones 1987, Bowling 1991, Thoits 1995, Stansfeld et al. 1998). Social support has a buffer effect on stressful life events, protecting a subject against psychological distress (Paulsen & Shaver 1991) and the development of mental disorders, especially depression (Paykel 1994, Dalgard et al. 1995).
The concept of social support and its effects on health and health behavior have been reviewed by several researchers (Caplan 1974, Cassel 1976, Cobb 1976, O’Reilly 1988). Israel (1982) reviewed the associations between social network and health.
The availability of social support in the community determines health care attendance (Schrire 1986). People who are lonely and have fewer social resources for coping tend to visit more frequently (McArdle et al. 1974, Browne et al. 1982, Robinson & Granfield 1986). Even owning a pet serves as social support, leading to less use of doctor contacts (Siegel 1990). Lesser social support combined with a high level of psychological distress was a significant risk factor for high use of health care services (Kouzis & Eaton 1998).
Social burden is an opposite to social support, but there is a lack of studies on the effects of social burden on the use of health care services.
"My moral really is that three facets are deeply involved in the thick file case — the patient, the family and the doctor." | |
| Kemp (1963) |
In his review of studies concerning health care service use, McKinlay pays attention to the social network and the family (McKinlay 1972). Family is the major source of social support but also a source of stress, both of which affect health (Cohen & Syme 1985, Campbell 1987, Parkerson et al. 1989, Ell 1996, Pratt 1976). There is ample evidence of impacts of the family on the health and illnesses of the family members (Campbell 1987). The family and its functioning and interactions affect both physical, psychosomatic and mental health (Alanen et al. 1966, Minuchin et al. 1975, Wynne 1981, Wirsching & Stierlin 1982).
The way in which the family copes with and adapts to the illness of one of its members has a strong impact on the physical and psychosocial well-being of all family members and the clinical course and duration of the illness itself (Tansella 1995). Family dysfunction decreases coping behavior. Illness behavior is affected by the family members’ behavior partly via the “family culture of using health care services” and partly through learning. (Schrire 1986.) The reported perception of subjective health seems to be in part learnt in childhood, in part shaped by one’s status in society and in part conditioned by psychological stress (Mechanic 1979). Sudden changes in life and the loss of a significant family member change illness behavior (Dowrick 1992). “Worried” parents tend to perceive common and trivial symptoms, e.g. colds, as threatening and consult physicians (Hansen 1994, Kai 1996).
The relations between family structure, family interaction and marital status and the use of health services have been studied to some extent, but the studies have produced different and even controversial findings (McKinlay 1972). Huygen (1978) reported a significant consistency in the number of consultations per family member within family and across generations. The number of contacts with the family physician as well as subjective symptoms and readiness to seek medical help are rather equal between family members (Picken & Ireland 1969, Huygen 1978). The mother’s psychological distress affects her propensity to seek medical services for her children (Tessler & Mechanic 1978). The maternal use of health services appears to be a more powerful predictor of use by children than other family and maternal variables (Newacheck & Halfon 1986).
The family structure and interactional patterns differ between high and low health service utilizer families. Wamoscher (1966) reported frequently attending families, and Wilson (1977) later pointed out that high utilizer families really exist, and he found high utilization families to have more social and economic problems than families with a lower utilization rate. Frequent use concentrates in large families (Wilson 1977), although opposite results have also been presented (Picken & Ireland 1969, Courtenay et al. 1974, Garcia Lavandera et al. 1996). The families of frequent attenders showed characteristics of dysfunctionality and were less social than low utilizer families (Browne et al. 1982, Weimer et al. 1983). The ”life situation” diagnosis of frequent attender patients reveals family problems and stressful life events (e.g. bereavement) (Larivaara 1987, Larivaara et al. 1996).
Family life cycle as a concept means the series of stages or events that mark a family’s life, offering an organizing schema for viewing the family as a system proceeding through time. The family life cycle theory provides a conceptual framework for understanding the common stresses of marital life. (Medalie 1979, Carter & McGoldrick 1980, McDaniel et al. 1990.) An association between the person’s position in the family life cycle and chronic illness and distress has been confirmed (Rolland 1987, Newby 1996). The literature contains only a few references to the importance of the family life cycle when studying health care use (McKinlay 1972). Báez et al. (1998) found more frequent attenders in the breaking or contraction phases of family life cycle.
Marital status is related to health problems, mortality rate, subjective illness, illness behaviour and use of health services, mainly due to differences in stress, life styles and social support (Morgan 1980, Burman & Margolin 1992). Marital adjustment, satisfaction and happiness associate with physical and psychological health (Renne 1971, Campbell 1987, Saarijärvi et al. 1990, Horwitz et al. 1998). The use of health services is more frequent among divorced, separated, widowed and never-married persons than among those who are married (Evashwick et al. 1984, Westhead 1985).
In some previous studies, marital disharmony or marital stress have been found among frequent attenders (Browne et al. 1982, Goodridge 1982, Weimer et al. 1983, Schrire 1986), but there are also studies without these findings (Courtenay et al. 1974, Karlsson et al. 1995a, Báez et al. 1998).
Frequent attenders may be present personally or via a proxy, frequently using a child in this way (Schrire 1986, Schreier & Libow 1994). The Münchausen by proxy syndrome (MBPS) is a condition in which a mother pretends her infant or child is ill or causes the infant or child to be ill in order to engage in an intensely ambivalent but often destructive relationship with a physician. MBPS was first described by Meadow in 1977. (Schreier & Libow 1994).