Table 3 Results on frequent attendance among the elderly at the primary care level across reviewed studies

Bergh and Marklund, 2003 [37]

10% most frequent attenders in 12 months/ by sex and age group

face-to-face visits to GP

medical records

Elderly (≥ 65 years):

• Most diagnostic groups and medical prescriptions more frequent among FAs than non-FAs for both sexes

Most common diseases:

• Women: diseases of circulatory and musculoskeletal system, similar for FAs and non-FAs

• Men: circulatory & endocrine diagnoses (FAs), circulatory and musculoskeletal problems (non-FAs)

Gilleard et al., 1998 [38]

Very High Attenders: 10% most frequent attenders in 12 months (> 15 contacts in 12 months)

face-to-face visits to GP, visits to the practice nurse

home visits, out-of-hour visits

computerized records, interviews, questionnaires

Elderly (≥ 65 years):

• 10% FAs responsible for 33% of all visits

• Frequent attendance not associated with psychiatric morbidity, self-reported depression, use of hypnotic or antipsychotic medication

• Use of antidepressants: 9.5% of FAs received prescriptions for antidepressants compared to 2.8% of low average attenders (chi-square = 13.6, df 3, p < 0.01)

Menchetti et al., 2006 [39]

> 1 contact to GP per month in 6 months

n/a

n/a

registered data, questionnaires, clinical judgments of GPs

Elderly (≥ 60 years):

• Frequent attendance associated with moderate or severe physical illness (aOR = 2.89, 95% CI: 1.63–5.11), depression (aOR = 1.92, 95% CI: 1.10–3.35) and unexplained somatic symptoms (aOR = 1.99, 95% CI: 1.05–3.77)

• Depression increased risk of being an FA fivefold and was a risk factor for frequent attendance independent of other clinical predictors

Rennemark et al., 2009 [40]

30% most frequent attenders in 12 months (≥3 contacts in 12 months)

n/a

n/a

questionnaires, cognitive tests, medical records

Elderly (≥ 60 years):

• Number of GP visits positively correlated with age (0.53, p < 0.001), and comorbidity (0.93, p < 0.001), and negatively correlated with functional ability (−0.18, p < 0.001), education level (−0.12, p < 0.01) and internal locus of control (−0.12, p < 0.01)

Results from logistic regression analyses:

• Physical comorbidity as main factor determining frequent attendance (OR = 8.17, 95% CI: 5.54–12.04)

• Sense of coherence (OR = 1.03, 95% CI: 1.00–1.06) and locus of control (OR = 1.14, 95% CI: 1.02–1.27) significantly related to frequent attendance

• Education level and social anchorage not associated with frequent attendance

Scherer et al., 2008 [46]

> 17 contacts in 9 months

n/a

n/a

questionnaires, telephone interviews

Elderly:

• Frequent attendance associated with female sex, living alone, severity of heart failure, psychological distress and quality of life

• In multivariate analysis physical problems (OR = 1.1, 95% CI: 1.0–1.1, p < 0.001) and living alone (OR = 2.4, 95% CI: 1.1–5.1) independently related to frequent attendance

Sheehan et al., 2003 [45]

top third of attenders in 9 months

medical contacts with GP at primary care centre or at home

consultations with practice nurse

patient interview, GP records, GP assessment of patients tendency to somatise

Elderly (≥ 65 years):

• Frequent attendance related to depression (OR = 2.24, 95% CI: 1.11–4.50, p < 0.05), high rates of physical disorder (OR = 1.78, 95% CI:1.16–2.71, p < 0.05), somatic symptom reporting (OR = 1.83, 95% CI:1.13–2.97, p < 0.05), and low social support (OR = 1.73, 95% CI:1.01–2.94, p < 0.05)

• In multivariate regression only low social support and somatic symptoms significantly related to frequent attendance

Svab and Zaletel-Kragelj, 1993 [43]

25% most frequent attenders in 12 months/ by age group

face-to-face visits with

GP, contacts for administrative purposes

telephone contacts

medical records and registered data

Elderly (>65):

• Probability for superficial (administrative) contacts larger for FAs compared to non-FAs (median percentage of superficial contacts among all contacts: FAs = 27.1%/non-FAs = 0.5%, p = 0.05)

• Non-significant trend: larger probability of referral to specialists for FAs compared to non-FAs (median index-value for referral to a specialist: FAs = 8.0/non-FAs = 0.4).

van den Bussche et al., 2016 [44]

A: ≥ 50 contacts with physician practices in 12 months

B: contacts with ≥10 different practices in 12 months

C: contacts with ≥3 different practices of the same medical specialty in 12 months

visits to the practice, home, nursing home visits, telephone contacts, contacts with practice staff

appointments by phone and administrative contacts

insurance claims data/ registered data

Elderly (≥65):

• Type A attendance associated with higher age, dependency on nursing care, multi-morbidity, and high impact somatic diseases

• Types B and C attendance associated with younger age, less dependency on nursing care, and presence of mental diseases

• Number of chronic conditions reduced the risk of being Type C FA

Vedsted et al., 2001 [42]

daytime: 10% most frequent attenders (≥ 12 contacts) in 12 months/ by sex and age group

out-of-hours: 10% most frequent attenders (≥ 4 contacts) in 12 months

daytime: face-to-face visits with GP

out-of-hours: telephone advice, surgery consultations, home visits

telephone contacts during daytime and administrative and routine consultations

electronic records

Elderly (≥ 65 years):

• Frequent attendance during daytime strongly related to the risk of being an out-of-hours FA:

OR and 95% CI of daytime users to be an FA in out-of-hours service compared to non-attenders:

men with 10% most daytime contacts: OR = 72.5 (CI: 48.7–107.9)

women with 10% most daytime contacts: OR = 40.7 (CI: 28.2–58.8)

Vedsted et al., 2004 [41]

10% most frequent attenders (≥ 12 contacts) in 12 months/ by sex and age group

face-to-face visits to GP, home visits during daytime

telephone contacts, administrative and routine consultations (e.g. driver’s licenses)

electronic records

Elderly (≥ 65 years):

• Prevalence ratio for using one or more drugs only slightly higher among FAs compared to the 50%-group with the fewest contacts

• Prevalence for polypharmacy (drugs from 5 or more drug groups) 6.7 times (men) and 4.2 times (women) higher among FAs compared to the 50%-group with the fewest contacts