The effect of having a care home physician on the care home residents’ rate of contacts with primary care physicians and hospitals: a retrospective register-based study

Abstract

Purpose

Care home residents often suffer from several morbidities, including dementia, and are prone to acute hospital admissions. Attempts to reduce avoidable hospitalisations from care homes are diverse, but they can include a strengthened collaboration between care home facilities and primary care physicians (PCPs). We aim to investigate whether residents of care homes with and without a care home physician have different contact patterns with PCPs, out-of-hours service, and hospitals.

Methods

Retrospective register-based study combined data from the Danish national health registries with information from a telephone interview, obtaining information on the care homes’ collaboration with local PCPs. Care homes were divided into two groups: Facilities with (n = 26) and without (n = 10) a care home physician in 2019. We used Quasi-Poisson regression to analyse whether residents of facilities with and without a care home physician had a similar pattern for contacts with PCPs, the out-of-hours service, and hospitals.

Results

Residents of facilities with a care home physician (n = 1,368) had more than twice as many home visits by the PCP and 37% fewer e-mail consultations when compared to residents of facilities with no care home physician (n = 500). They tended to have fewer contacts with the out-of-hours service, but this result was not significant. We found no differences in the use of hospital services.

Conclusion

We saw an increased availability of PCPs in care homes with a care home physician and lower use of the out-of-hours service, but no impact on hospital admissions.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12913-025-13189-4.

Background

Acute hospital admissions are frequent among care home residents due to their multimorbidity, vulnerability, susceptibility to falls and fractures, deterioration of chronic diseases, and acute infections [1]. When hospitalised, these residents face a high risk of adverse events, including delirium, loss of functional capacity, and hospital-acquired infections, all of which contribute to high in-hospital mortality [2]. Estimates suggest that between 17 and 67% of admissions from care homes could potentially have been avoided [3–6]. Therefore, many studies have focused on strategies to reduce these avoidable admissions, and interventions vary greatly [1, 4]. One effective approach to lowering the rate of admissions is to strengthen collaboration and communication between care home staff and primary care physicians (PCP) [7, 8].

The introduction of PCPs working as care home physicians in Danish care homes (described in detail in the method section) has positively affected inter-professional collaboration between care home staff and physicians [9]. However, the impact of this arrangement on the care home residents’ utilisation of primary care and hospital services is only sparsely investigated [10–12]. This retrospective, register-based study aims to assess the effects of having a PCP collaborate with the care home as a care home physician with regularly scheduled visits. Specifically, we will.

1. Assess the number of care homes with a care home physician and the number of residents registered with the care home physician.

2. Investigate whether care homes with and without a care home physician differ in terms of nursing staff, size of care home (number of beds), and distance to hospital.

3. Investigate whether residents in care homes with and without a care home physician differ regarding sex, age and morbidities.

4. Examine whether residents in care homes with a care home physician exhibit different patterns of contact with the healthcare system, hypothesising that they will have:

   • More contacts with the PCP,
   • Fewer contacts with the out-of-hours service,
   • Fewer acute hospital admissions.

compared to residents in care homes without a care home physician.

Method

Study design, setting and population

Study design

This is a register-based population study. Data from national health registers were supplemented with information obtained through a telephone interview with all care homes in Southern Jutland regarding their collaboration with local PCPs.

Setting

Southern Jutland comprises four municipalities with approximately 225,000 inhabitants living in rural and urban areas, hereof 23.4% aged 65 + years, and 0.67% of citizens living permanently in care home facilities in 2019 [13, 14]. Population demographics in Southern Jutland are comparable to those of the Danish population.

Danish citizens can apply for a care home residency when they suffer from permanent and substantial impairment of physical or mental function. The municipalities appoint residencies based on needs, regardless of financial means, and all care homes are subject to the Danish Law on Social Services [15]. During 2019, 38 care home facilities were operating in Southern Jutland, with around 1,600 long-term beds in total [16]. The care homes are staffed 24 h/day with healthcare assistants or healthcare helpers with 1.6-3 years of education [17], supplemented by unskilled workers. Most Danish care homes are staffed with a nurse during the daytime on weekdays to handle tasks requiring nursing skills. Care homes without a nurse receive weekly planned visits by the community nurses. During evenings, nights and weekends, the community nurses are on call in all the care homes [17]. Danish care homes employ no physicians.

The Danish healthcare system is tax-funded and free of charge for all Danish citizens. PCPs are self-employed and are reimbursed by the Regions on a fee-for-service basis combined with a capitation fee based on the list of patients [18]. The PCPs act as the first point of contact when a patient needs medical care and manage the vast majority of cases without hospital referral [19]. They also manage all access to hospital care, except in medical emergencies. PCPs staff the out-of-hours service in Southern Denmark and are available for consultations and home visits in cases of acute illness emerging outside the PCPs’ opening hours [20].

Care home physicians

Before 2016, Danish citizens would retain their usual PCP after moving to a care home [21], unless they moved to a care home located far away from the PCP. As a result, care home staff collaborated with many PCPs, who attended to patients in several different care homes, which could challenge the collaboration between care home staff and PCPs. Therefore, since 2016, the Danish government, the Local Government Denmark (KL), the five Danish Regions and the Danish Organisation of General Practitioners (PLO) have encouraged care homes and PCPs to strengthen their collaboration by having PCPs take on the role of care home physicians [22, 23]. It is up to the care homes and PCPs to initiate this collaboration, and there are no formal requirements for its content. However, the partnership is encouraged to consist of scheduled visits by the physician and guidance or teaching of the care home staff on general topics related to health promotion and disease prevention among care home residents [22]. Care home physicians visit the care homes regularly while still maintaining their private practice, and the Regions pay the fee for services provided by the care home physician. Upon moving to the care home, residents are encouraged but not required to select the care home physician, and some may choose to keep their usual PCP.

Study population

The in- and exclusion criteria are illustrated in Fig. 1. All individuals living in one of the 38 care homes in Southern Jutland in 2019 were eligible for the present study. We excluded residents aged < 65 years when moving to a care home. Residents of care homes that gained a care home physician or terminated their collaboration with the care home physician during 2019 were excluded from the study. If a care home resident moved out of Southern Jutland during the study period, the individual was excluded from the move date. If a care home resident moved from one care home in Southern Jutland to another during the study period, only the last care home residency was included. The exclusion of care homes and residents that potentially changed their care home physician status during the study period was to avoid biasing the results, as the applied statistical method could not adjust for dependent observations.

Fig. 1.

Showing the in- and exclusion criteria of participants

Data sources

The study was based on register data obtained from the highly valid Danish National Health Registers and information obtained through telephone interviews.

Register data

The Danish Health Data Authority provided all register-based data for the present study, and the registries included are described in Table 1.

Table 1.

Description of registers included in the study

Name of register	Description
Care Home Data (Plejehjemsdata)	The register was introduced in 2020 and can identify all Danish citizens registered with a care home address by combining information from the Civil Registration System with a list of every individual care home apartment in Denmark [24]. The register provides the date of moving to a care home, the date of moving out (either to another care home or the date of death) and an encrypted care home name.
The Civil Registration System	This registry was founded in 1968 and contains general personal information on all Danish citizens [25]. From this register, we obtain information on date of birth, sex, and date of death (if relevant). A unique 10-digit identifier assigned to all individuals links to other national registers.
The Danish National Health Service Register	Since 1990, data on services provided by health contractors in the primary health care system is recorded in this register. As the data are collected for administrative purposes and are linked to the reimbursement of PCPs, the coverage is assumed to be high [26]. From this register, we collect individual information on registration with PCPs, dates of contact with PCPs and the out-of-hours service, and the types of services provided.
The Danish National Patient Register	Information on somatic in-patient activities has been collected since the establishment of the register in 1977. Since 1995, the register has included information on all in- and outpatient activities in both the somatic and psychiatric secondary health care systems [27]. From this register, we obtain information on the dates of hospital activities, the type of activity (e.g., admission or outpatient contact), and the urgency (planned or acute). For every contact, one primary and optional secondary diagnosis is recorded, and the diagnoses are registered according to the international classification of diseases (ICD-10).
The Danish National Prescription Registry	Individual-level data on all prescription medicine sold in Denmark has been recorded since 1995 [28]. Medicines are categorised according to the Anatomical Therapeutic Chemical Index (ATC codes) [29].

Telephone interview

The telephone interview aimed to collect information on the care homes’ present and historic collaboration with local PCPs and whether the care home employed a nurse in the daytime on weekdays. The care homes were asked whether they had a partnership with a PCP working as a care home physician. Some care homes answered that they had signed a contract and initiated a collaboration with a PCP recently, as encouraged by the Danish Regions. Other care home facilities described an agelong collaboration with the local PCP, with weekly preventive visits, like the arrangement proposed by the government, but never with a written contract. These care home facilities were considered to have a care home physician on equal terms as the care home facilities with a signed agreement with a PCP. The remaining part of the care homes answered that they had no care home physician.

The telephone interview was conducted in the spring of 2021. A questionnaire was set up in REDCap [30, 31] electronic data capture tool hosted at Open Patient data Explorative Network (OPEN) [32]. GSK collected all data by calling the 39 care home facilities in Southern Denmark, and all care homes responded. Of the 39 care home facilities interviewed, 38 were operating in 2019, which represents the study period. The person interviewed at the care home facility was either the care home manager or the nurse. Raw data from the telephone interview collected in REDCap was transferred to the Danish Health Data Authority and combined with register data.

Data variables

Care home residents’ general characteristics

The general characteristics of the care home residents are described in terms of age, sex, and morbidities at baseline, defined as the date of moving to the care home. Morbidities were assessed by collecting all primary and secondary ICD-10 diagnoses assigned to all in- and outpatient hospital contacts during the past ten years from baseline. To reduce the risk of underestimating disease prevalences, the ICD-10 diagnoses were combined with ATC codes on all dispensed prescriptions within one year from baseline. ATC codes were excluded from the assessment of morbidities if the medicine had several different indications. The assessment of morbidities was developed for a previous study [33], and a comprehensive description of ICD-10 codes and ATC codes is included in Additional File 1.

Days in care home and person-years

We calculated the average time in care home during the one-year study period by dividing the number of days spent in care home by the number of individuals in the study. The days spent in care home were equal to the days between inclusion in the study, either January 1, 2019, or the date of moving to the care home if later, and the end of follow-up, either December 31, 2019 or the date of death/moving out of Southern Jutland, if earlier. Person-years is equivalent to the total time in years that all participants contributed to the study.

Primary outcome: contacts with primary care physicians and hospital

Contacts with PCPs were categorised as either consultations, home visits, telephone consultations, e-mail consultations, or preventive contacts (e.g. follow-up home visits, discharge conferences, planning meetings, and annual consultations regarding chronic diseases). Contacts with out-of-hours PCPs were categorised as either consultations, home visits or telephone consultations. The distinction between different service types was based on the service numbers presented in Additional File 2.

Hospital contacts were categorised as acute treat-and-release visits (duration < 12 h), short acute admissions (duration 12–48 h), more prolonged acute admissions (duration > 48 h), and planned admissions.

Telephone interview

The information collected through the telephone interview contained the following variables:

• Name of care home facility.

• Size of care home (number of beds).

• Distance to the nearest acute care hospital (in kilometres (km)).

• Whether the care home facility has a care home physician (yes/no).

  • ◦ Name of the PCP.
  • ◦ Date of implementing the care home physician.

• Whether the care home facility employs a nurse in daytime on weekdays (yes/no).

Before the transfer of data obtained from the telephone interview to the Danish Health Data Authority, the name of the PCP working as a care home physician was changed to a specific caregiver number, which is used for the identification of healthcare providers in the primary healthcare system. Names of care home facilities and caregiver numbers of PCPs were encrypted upon transfer to the Danish Health Data Authority.

Data analysis and statistical methods

The prevalence of care homes with a care home physician and the proportion of care home residents with and without a care home physician is presented in numbers and proportions as of January 1, 2018, 2019, 2020, and 2021, respectively.

Care homes were grouped as care homes with and without a care home physician. Likewise, the care home residents were divided into two groups: Residents of facilities with and without a care home physician. Descriptive statistics were presented as frequency counts and proportions for categorical variables. Means and standard deviations (SD) were presented for normally distributed numeric variables.

We analysed whether there was a significant difference between the characteristics of care home facilities and residents of care home facilities with and without a care home physician. For numerical variables with a normal distribution, we used the t-test. We used the Wilcoxon rank-sum test for numerical non-normal distributed variables. For dichotomous categorical variables, we used the Chi-squared test. According to Cochran’s rule, Fischer’s exact test was used for categorical variables with five observations or less.

The annual rates of contact with PCPs, out-of-hours service and hospitals were calculated by dividing the number of contacts by the number of person-years.

We used Quasi-Poisson regression to analyse whether residents of facilities with and without a care home physician had different rates of contact with PCPs, the out-of-hours service, and hospital. The model allowed for extra variance, as mean and variance were not equal as Poisson regression required. We adjusted for clustering at a care home level because residents of the same care home live in the same environment, share the same nurse and often the same PCP, which can affect the rates of contact with PCPs and hospitals.

The statistical significance level was set at 5%. Most analyses were conducted in Stata Statistical Software: Release 18 [34]. However, Quasi-Poisson regression was performed in R Statistical Software, version 4.3.3 [35], as Stata could not adjust for clustering in the Quasi-Poisson model.

Results

A total of 38 care homes were operating in Southern Jutland 2018–2021. The number of care homes with a care home physician rose from 18 (47%) in January 2018 to 26 in 2019 and 29 (76%) in January 2020 and 2021, respectively. Figure 2 illustrates the increase in care home residents registered with a care home physician during 2018–2021. Focusing only on care homes with a care home physician, the proportion of care home residents registered with the care home physician rose from 60% on January 1, 2018, to 67% on January 1, 2021. During 2019, which represents the study period, the proportion of care home residents registered with the care home physician rose from 63 to 66%.

Fig. 2.

The attendance among care home residents of the 38 care homes in Southern Jutland to the arrangement with care home physicians as of January 1, 2018, 2019, 2020, and 2021, respectively

Of the 38 care home facilities operating in Southern Jutland in 2019, 26 had a care home physician throughout 2019, while 10 did not have a care home physician. Two care homes signed a contract with a PCP in 2019. Residents of these two care homes were excluded from the study (n = 51). Residents of the remaining 36 care homes were identified through the Care Home Data Register, and all individuals aged 65 + years at the time of moving to care home were included in the study (n = 1,868), as illustrated in Fig. 1.

Characteristics of care home facilities with and without a care home physician

Care home facilities with and without a care home physician were similar regarding the number of beds and distance to the nearest hospital with an ED, as seen in Table 2. There was a tendency that facilities without a care home physician more often did not employ a nurse, but the results were not significant.

Table 2.

Characteristics of care homes with and without a care home physician in Southern Jutland in 2019

Characteristics of care homes	With a care home physician	Without a care home physician	p-value
Care home facilities, n	26	10
Number of beds, mean (SD)	44.8 (± 16.6)	45.1 (± 25.0)	p = 0.96 *
Distance to acute care hospital (km), mean (SD)	22.6 (± 12.4)	25.1 (± 17.4)	p = 0.63 *
No nurse employed at care home, n (%)	< 5 (< 19.2%) Ⴕ	5 (50%)	Ⴕ

*p-values calculated using the t-test

Ⴕ Less than five care homes with a care home physician did not employ a nurse. Due to the Danish Data Protection Law and Danish Health Data Authority guidelines, we cannot provide an exact number or p-value

Characteristics of residents of care home facilities with and without a care home physician

The study included 1,868 residents living in care homes in Southern Jutland at any time during 2019, with 1,368 residing in the 26 facilities with a care home physician and 500 residents of the 10 facilities without a care home physician. Table 3 shows the baseline characteristics of the care home residents, stratified by type of care home facility. There were no statistically significant differences between residents of facilities with and without a care home physician.

Table 3.

Baseline characteristics of residents of care homes with and without a care home physician

Characteristics of residents	With a care home physician	Without a care home physician	p-value
Care home residents, n (%)	1,368 (73.2%)	500 (26.8%)
Days in care home during the study period, mean (SD)	276.9 (± 119.2)	267.7 (± 124.9)	n.s. *
Age in years at moving to care home, mean (SD)	82.9 (± 7.5)	83.1 (± 7.8)	n.s. Ⴕ
Females, n (%)	915 (66.9%)	345 (69.0%)	n.s. ǂ
Morbidities
Cancer	266 (19.4%)	100 (20.0%)	n.s.ǂ
Diabetes	248 (18.1%)	83 (16.6%)	n.s. ǂ
Dementia	717 (52.4%)	249 (49.8%)	n.s. ǂ
Parkinson’s disease	79 (5.8%)	24 (4.8%)	n.s. ǂ
Alcohol abuse	82 (6.0%)	32 (6.4%)	n.s. ǂ
Hypertension	765 (55.9%)	258 (51.6%)	n.s. ǂ
Ischemic heart disease	221 (16.2%)	88 (17.6%)	n.s. ǂ
Heart failure	130 (9.5%)	56 (11.2%)	n.s. ǂ
Atrial fibrillation	288 (21.1%)	105 (21.0%)	n.s. ǂ
Stroke	339 (24.8%)	121 (24.2%)	n.s. ǂ
COPD*/asthma	229 (16.7%)	86 (17.2%)	n.s. ǂ
Osteoporosis	311 (22.7%)	126 (25.2%)	n.s. ǂ
Number of selected morbidities			n.s. ǂ
0	81 (5.9%)	33 (6.6%)
1–2	569 (41.6%)	210 (42.0%)
3–4	556 (40.6%)	204 (40.8%)
5+	162 (11.8%)	53 (10.6%)

The percentages do not add up due to rounding

n.s. not significant

* p-values calculated using the Wilcoxon rank-sum test

Ⴕ p-values calculated using the t-test

ǂ p-values calculated using the Chi-squared test

Primary outcome: contacts with primary care physicians and hospital

Table 4 shows the annual rates of contacts between care home residents and PCPs, out-of-hours service and hospitals. We found that residents of facilities with a care home physician had more than twice as many home visits by their PCP (p < 0.001) and 36.5% fewer e-mail consultations (p < 0.001).

Table 4.

Annual rates of contact with primary care physicians, out-of-hours service, and hospitals among residents of care home facilities with and without a care home physician, respectively

	With a care home physician	Without a care home physician	p-value *
Person-years	1,037.87	366.73
Primary care physician
Rate of contacts per person-year
Consultations	6.50	5.70	p = 0.37
Home visits	6.21	2.71	p < 0.001
Telephone consultations	5.03	5.67	p = 0.44
E-mail consultations	11.06	17.46	p < 0.001
Preventive contacts	0.89	0.58	p = 0.09
Out-of-hours primary care physician
Rate of contacts per person-year
Face-to-face contacts	0.66	0.82	p = 0.11
Telephone consultations	1.64	1.96	p = 0.12
Hospital
Rate of contacts per person-year
Acute treat-and-release visits (< 12 h)	0.32	0.32	p = 0.87
Short acute admissions (12–48 h)	0.17	0.15	p = 0.54
Long acute admissions (48 + hours)	0.30	0.26	p = 0.56
Planned admissions	0.02	0.04	p = 0.14

* p-values calculated using the Quasi-Poisson model

Residents of facilities with a care home physician tended to have fewer consultations or home visits by the out-of-hours service, but the results were not significant (p = 0.11). Consultations with the out-of-hours service accounted for only 1.6% of all face-to-face contacts, while the remaining 98.4% represented home visits. We found no differences in the rates of hospital contacts regarding acute treat-and-release visits, short or longer acute hospital admissions, or planned admissions.

Discussion

Key findings

The study showed an increase in the proportion of care homes with a care home physician during 2018–2021 from 47 to 76%. Focusing only on care homes with a care home physician, the proportion of residents registered with the care home physician rose from 60% on January 1, 2018, to 67% on January 1, 2021. Care homes and residents of care homes with and without a care home physician showed similar characteristics.

We found that residents of facilities with a care home physician received more than twice as many home visits by their PCP and had 36.5% fewer e-mail consultations when compared to residents of facilities without a care home physician. There was a tendency for residents of facilities with a care home physician to have fewer contacts with the out-of-hours service; however, these results were not significant. We detected no differences in the rates of acute treat-and-release visits, or acute or planned admissions.

The implementation of care home physicians

The increase in the number of care homes in Southern Jutland with a care home physician corresponds to the results from a report on implementing care home physicians in Denmark in general [36]. Correspondingly, we showed an increase in the proportion of care home residents registered with the care home physician in 2018–2021. Still, in 2021, one-third of care home residents living in facilities with a care home physician were registered with another PCP, similar to findings from a Danish study evaluating the pilot project introducing care home physicians in 2012–2013. Some of the residents registered with another PCP may have moved to care homes before the implementation of the care home physician, and a continuation with their regular PCP was, therefore, the natural choice. Others may have actively chosen to continue with their regular PCP when moving to a care home that offers registration with the care home physician. Continuing with the regular PCP after moving to a care home means preserving the long-term relationship between PCP and patient based on trust and familiarity. Studies show that PCP continuity among older adults is associated with fewer acute hospitalisations, lower use of out-of-hours service and lower mortality [37–39], which may impact the results on rates of contacts with PCPs and hospitals in the present study. On the other hand, one might argue that the continuity of care is improved by choosing the care home physician who provides scheduled visits at the care home, thereby providing easier access to PCP services. Furthermore, the care home physician has a long-term collaboration with the care home staff, who manages most of the contacts between the care home resident and physician, which is beneficial for the resident who chooses the care home physician.

The increase in the number of care homes in Southern Jutland with a care home physician stagnated in 2020–2021, reflecting that not all care homes can find a care home physician. The reason for this is unknown, but it could reflect that the local PCP is close to retirement or has too many patients and does not wish to take on the role of care home physician. Or the local PCP is already a care home physician at another care home facility, or perhaps there are no PCPs near the care home, e.g., in rural areas. One might hypothesise that some care homes can find a care home physician more easily than others, e.g., care homes in larger cities or with fewer residents. If true, this can bias the results. We analysed differences between care home facilities by comparing the general characteristics of the facilities with and without a care home physician. The care homes were similar in size and distance to the nearest hospital with an ED. We observed a tendency for care homes without an employed nurse to lack a care home physician more often; however, the results were not statistically significant. The tendency may reflect that having a permanent collaborator in the care home makes it more attractive for the local PCP to take on the role of care home physician. A qualitative Swedish study supports this, where PCPs described the importance of relying on the care home nurse’s observations and their medical knowledge about the residents, emphasising that having a skilled nurse familiar with the residents is essential to the quality of medical care [40].

We furthermore analysed for differences between care homes with and without a care home physician by comparing the care home residents’ characteristics, and we found no differences between the two groups of residents regarding sex, age, duration of stay or morbidities. The assessment of morbidities included ICD-10 diagnoses from hospital contacts and ATC codes on dispensed prescriptions. All characteristics of care home residents were assessed at baseline, which was defined as the date of moving to care home. Therefore, a possible different approach to prescribing/de-prescribing and diagnostic assessments or hospital admissions among care home physicians and regular PCPs could not impact the estimated prevalence of morbidities among residents.

Contacts between residents and primary care physicians, out-of-hours service and hospital

We saw that residents of facilities with a care home physician received markedly more home visits by the PCP, presumably resulting in the lower rate of e-mail contacts between PCP and care home staff. A Danish study on one of the biggest municipalities in Denmark has shown similar results [10]. The rate of contacts with PCPs per person-year was slightly higher than the rates from two older Danish studies reporting 22–24 contacts per resident per year [17, 41]. That residents of care home facilities with a care home physician received more home visits is the natural effect of having a care home physician who performs planned visits at the care home. This makes it easier for the care home physician and staff to schedule medical appointments for a larger number of residents and handle both acute and chronic conditions. Presumably, there is a lower threshold for scheduling appointments with the care home physician, meaning that the care home physician sometimes sees patients for conditions that would otherwise have waited longer before contact to the physician. The regular visits by the care home physician may facilitate the handling of less urgent matters, such as medication reviews, or end-of-life conversations between residents, physician and staff. Although not a result of the present study, other studies have shown that an increased availability of the PCP effectively reduces acute hospital admissions from care homes [4, 8, 42].

In the present study, we saw that residents in facilities with a care home physician tended to have fewer face-to-face contacts with the out-of-hours service (98.4% home visits). However, this result was not statistically significant, presumably due to few observations. A systematic review found that a more timely access to PCPs allowed early medical assessment of ill care home residents with early diagnosis and onset of treatment, reducing the number of ED transfers [8]. It might be possible that emerging acute illnesses were diagnosed and treated earlier in residents of care home facilities who received frequent scheduled visits by the care home physician, thereby lowering the need for the out-of-hours service. However, larger studies are needed to obtain significant results. Contacts with the out-of-hours service are problematic for care home residents due to the unfamiliarity between the on-call PCP and the patient, and the care home residents’ complex care needs, often leading to acute admission [40, 43]. Furthermore, home visits by the out-of-hours service are more cost-intensive for the healthcare system than regular PCP home visits. If more emerging acute illnesses are detected and treated by the care home physician, it is beneficial to both the care home resident and the healthcare system in general.

We hypothesised that residents of facilities with a care home physician had fewer acute hospitalisations than those without a care home physician. However, we found no difference in the rates of hospital contact between the two groups. This contrasts with results from the pilot study of care home physicians in Denmark, where a reduction in preventable admissions was seen [12]. Preventable admissions were defined as admissions that could have been avoided if symptoms were precisely identified and treated effectively. However, the care homes and PCPs in this pilot project volunteered to participate, which could affect their motivation to reduce admissions during the study period. Christensen et al. evaluated the effect of introducing care home physicians in one large Danish municipality. They found a reduction in acute admissions and readmissions only when the implementation of the care home physician was successful, defined as having more than 60% of care home residents listed with the care home physician [10]. This indicates that it can take time before the collaboration between care home staff and physician is fully implemented and effectively reduces acute admissions of residents.

Strengths and limitations

The present study included a large proportion of the care home residents aged 65 + years in the four municipalities of Southern Jutland in 2019, as we only excluded 51 residents of the two care homes who obtained a care home physician in 2019. The high validity of the Danish national health registers ensured that we could identify all care home residents listed with a care home physician. Information on care homes provided by the municipalities and information on general characteristics of the residents obtained from the national health registers allowed us to control for possible differences between care homes and residents.

However, our study has some limitations. The study assessed the effect of the care home physician by comparing rates of contact with PCPs, out-of-hours service, and hospital visits between residents of care home facilities with and without a care home physician. We thereby measure the effect on the entire facility, and not the individual residents, even though one-third of residents of facilities with a care home physician were registered with another PCP. We argue that having a care home physician can affect all the residents of the care home facility, as having a care home physician also implies teaching the care home staff. When the care home physician teaches prevention of urinary tract infection, fall prevention, early signs of disease in older adults, or close communication between care home staff and PCP, this will be beneficial for all residents of the care home facility [7, 8]. However, the presence of individuals registered with another PCP in care homes with a care home physician may lead to an underestimation of the results. Our results, therefore, show the effect of the care home physicians on a larger public health level and not only how the care home physician affects the individual residents.

We measured the effect of the care home physicians in a relatively early stage of implementation. Many of the care homes initiated a collaboration with a PCP in 2018. As it takes time for the care home staff, residents, and PCPs to be acquainted with one another [12], the effect of the collaboration might be more significant if evaluated on newer data. Furthermore, as care home physicians are increasingly popular, more care home residents may choose to select the care home physician today, which could produce different results in a newer study. Although more recent data from 2020 to 2021 were available for the present study, we chose not to include this time period. The COVID-19 pandemic meant a radical change in the pattern of contact with PCPs, out-of-hours service and hospitals for all Danish citizens, but especially for the care home residents [44]. To avoid a distortion of the study results, data from these years were not included in the study.

Implications for future studies

Future studies should include more recent data to detect a possible increase in the number of care home residents listed with a care home physician and a potential reduction in the use of out-of-hours service and hospital admissions.

Conclusion

Residents of care home facilities with a care home physician received more home visits from their PCP, had fewer e-mail consultations, and tended to use the out-of-hours service less. We saw no differences in hospital contacts between residents of facilities with and without a care home physician.

Abbreviations

ATC

Anatomical Therapeutic Chemical Index

ICD-10

International Classification of Diseases

KL

Local Government Denmark [Kommunernes Landsforening

Km

Kilometres

OPEN

Open Patient data Explorative Network

PCP

Primary Care Physician

PLO

Danish Organisation of General Practitioners [Praktiserende Lægers Organisation]

SD

Standard Deviation

Authors’ contributions

GSK conceived and designed the study, supported by all co-authors. GSK performed data collection and analyses, supported by SRP. GSK wrote the first draft of the manuscript, and all co-authors substantially revised the work. All authors have read and improved the final version of the manuscript.

Funding

Open access funding provided by University of Southern Denmark. The research was funded through a grant from the Region of Southern Denmark.

Data availability

The data supporting this study’s findings are available from the Danish Health Data Authority. Restrictions apply to the availability of these data, which were used under license for the current study and are not publicly available. Data are, however, available from the authors upon a reasonable request and with the permission of the Danish Health Data Authority.

Declarations

Ethics approval and consent to participate

The Danish Health Data Authority provided all data for the present study. No data were missing. Data were processed on the Danish Health Data Authority’s secure platform, Forskermaskinen. All methods were carried out in accordance with relevant guidelines and regulations.

The processing of personal data was approved by the Region of Southern Denmark (internal record 19/432,119) cf. Art 30 of The General Data Protection Regulation. According to Danish law, studies based on register data do not require approval from an ethics committee or informed consent from study participants [45].

The telephone interview was developed for the present study. Prior to the telephone interview, we contacted the health directors in the individual municipalities and obtained written permission to contact the care homes. The telephone interview collected no personal information, only overall organisational and publicly available information.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.