Abstract
Background
Hospital family physicians are recognized for their excellence in managing complex issues. This study aimed to reveal the level of complexities of patients referred to hospital family physicians by community centers, and the degree of change in these complexities following care provided by a health care team that includes hospital family physicians.
Methods
A retrospective cohort analysis. Patients introduced by community centers between 2020 and 2023 were identified. The patients received team-based comprehensive care. Complexity was calculated before and after the care, using the patient-centered assessment method (PCAM), which evaluates 12 items across four domains. Each item is rated from 1 to 4, yielding a total score range of 12 to 48. Pre- and post-intervention scores were compared using paired-sample t-tests, with standardized mean difference calculated using Hedges’ g.
Results
Of 41 referred patients, three died shortly after the initial consultation. Among the 38 remaining patients, 24 were admitted, and 14 were treated as outpatients. The mean PCAM score significantly decreased from 36.9 to 23.7 after interventions (P < .001, Hedges’ g = 2.54). Scores improved significantly across all domains: health and well-being (2.96 vs 1.95; P < .001, g = 2.00), social environment (3.09 vs 1.96; P < .001, g = 2.38), health literacy and communication (2.78 vs 2.46; P < .001, g = 0.67), and service coordination (3.61 vs 1.57; P < .001, g = 4.68).
Conclusion
Hospital family physicians in Japan often manage patients with exceptionally complex problems and improve patient outcomes across multiple domains.
Keywords: delivery of health care, family medicine, geriatrics, patient-centered care
Key messages.
Hospital family physicians can manage patients with exceptionally complex needs.
Hospital-based family practice can effectively enhance service coordination.
Most patients with complex social needs have undifferentiated health issues.
Family physicians can use hospital wards to establish a safer living environment.
Introduction
One of the foundational principles of family medicine is the delivery of comprehensive and continuous care [1, 2]. Family physicians are highly trained to manage a broad range of illnesses and provide holistic, patient-centered care to individuals across all age groups [3]. Their versatility enables them to practice in various settings, including clinics, home-visit care, hospitals, and emergency departments [1]. Hospital care constitutes a substantial share of healthcare expenditures, and improving the quality of chronic care and minimizing emergency department visits and hospital admissions has become a pivotal goal for family medicine [4]. To address these challenges, a progressive model of comprehensive care has been introduced: family physicians with special interests and expertise in hospital medicine [5]. A prominent example of this model is the hospital generalist, or “extensivist,” who delivers integrated support encompassing mental health, social welfare, and lifestyle interventions, particularly for frail older adults [6]. Furthermore, hospital care is a vital aspect of family medicine education, highlighting the importance of equitable resource allocation and efficient healthcare management [7].
Although there are reports indicating a shift away from hospital-based practice among family physicians in the United States [8] and Canada [9], family physicians in Japan commonly remain active in hospital settings. Remarkably, nearly half of the family physicians certified by the Japanese Primary Care Association are employed in hospitals [10]. These hospital family physicians provide a wide range of services, including inpatient ward care, outpatient consultations, emergency medicine, home-visit care, and community health initiatives, representing a unique and comprehensive model of family practice in Japan [11]. Japan’s universal health insurance system and relatively high levels of health equity coexist with significant challenges stemming from an aging population and increasing health disparities. These factors present family physicians with numerous opportunities to address the complex social and health needs of their patients. Within this context, hospital family physicians in Japan often regard their ability to manage complex and multifaceted cases as a defining standard of professional excellence [12, 13]. Thus, clarifying how hospital family physicians in Japan deliver care across diverse settings for patients with complex needs is crucial for understanding their expertise and the broader role of family physicians. Furthermore, such insights can inform how primary care systems worldwide adapt to the increasing demand for managing older patients with multifaceted healthcare needs. However, despite the distinctiveness of their practices, research on the characteristics and scope of hospital family physicians’ practice remains limited. Especially, there is a paucity of research investigating the complexity of cases handled by hospital family physicians and the outcomes of their care on patients’ levels of complexity.
In response to Japan’s increasing aging population, the government has implemented a comprehensive community care system designed to integrate housing, medical care, nursing care, preventive services, and lifestyle support tailored to the specific needs of each community. This system enables individuals requiring nursing care to maintain their daily lives within familiar neighborhoods until the end of their lives [14]. Within this system, community comprehensive support centers play a central role, providing holistic assistance to improve healthcare access and promote welfare by offering the necessary support for maintaining health and life stability [15]. Additionally, community comprehensive support centers collaborate with medical institutions to support older patients with complex and difficult health and social challenges that require medical care. Other community resource centers, such as municipal offices, public health centers, and social welfare councils, also coordinate with medical institutions to refer patients with such complexities. However, only a limited number of medical institutions are equipped to accommodate patients with such complexities. In this regard, hospital family physicians, acting as extensivists, are particularly well-positioned to manage these cases. Therefore, the level of complexity of patients referred to hospital family physicians by community centers may serve as a potential metric to evaluate the expertise and effectiveness of hospital family physicians.
We hypothesized that the complexity of cases referred from community centers to hospital family physicians is exceptionally high, and the team-based practice can effectively mitigate this complexity. This study aims to address the following research questions: (i) What is the level of complexity of older patients referred to hospital family physicians by community centers? (ii) How does the level of complexity of referred patients change following care provided by a health care team that includes hospital family physicians?
Methods
Setting
This retrospective cohort study was conducted through a medical chart review at a small hospital located in a suburban area of Japan, where the first and second authors were employed. The hospital, with 88 beds, served as a primary care facility, as is typical for small hospitals in Japan [16]. During the research period, the hospital employed approximately five to six family physicians who provided care across various settings, including outpatient clinics, inpatient wards, the emergency room, community (e.g. school physician, industrial physician), and patients’ homes (through home-visit care). Other medical departments in the hospital included orthopedics, pediatrics, psychiatry, and surgery. The hospital handled approximately 230 outpatients daily and maintained an average of 80 inpatients per day. An annual inpatient count was approximately 1,700, and about a half was admitted in the Family Practice ward. The hospital provides non-discriminatory and equitable medical care and welfare services, with extensive experience in managing patients facing socially complex and challenging circumstances. No other hospital in the city employed hospital family physicians. There are three community comprehensive support centers located near the hospital, at distances of 1.0 km, 1.3 km, and 3.6 km, respectively.
Participants
The study analyzed medical records of patients referred to by community resource centers to the hospital between 2020 and 2023. The centers referred patients in socially challenging situations and with complex social needs, including health care needs, who were unlikely to seek medical care independently. Patients who missed follow-up within six months after the initial consultation or before referral to another institution, were excluded from the analysis. In addition, we excluded patients who died shortly after the initial consultation because of inability to compare their complexity between before and after the care. Figure 1 shows the flow diagram. The second author, who managed referral coordination, possessed in-depth knowledge of the medical and social backgrounds of all referred patients through her routine work. She compiled a comprehensive list of eligible patients for the study.
Figure 1.
A flow diagram.
Care for referred patients, whether in ambulatory settings or hospital wards, generally encompasses the following three key components, each tailored to the patient’s unique circumstances through a team-based approach:
Comprehensive Assessment: Physicians, nurses, and medical social workers conduct thorough interviews to evaluate patients’ clinical and social conditions, including daily living activities, social security status, financial situation, available formal and informal support systems, as well as their personal preferences and values.
Medical Management: Identification and treatment of both acute and chronic conditions to ensure comprehensive and continuous healthcare.
Interdisciplinary Collaboration: Coordination among multiple professionals to establish supportive environments that enable patients to maintain their quality of life and independence.
Measures
We quantified each patients’ complexity using the patient-centered assessment method (PCAM) [17]. The PCAM evaluates 12 items across four domains: health and well-being (4 items); social environment (4 items); health literacy and communication (2 items); and service coordination (2 items). Each item is rated from 1 to 4, yielding a total score range of 12 to 48 (health and well-being: 4 to 16; social environment: 4 to 16; health literacy and communication: 2 to 8; service coordination: 2 to 8). Table 1 summarizes the PCAM components and scoring systems. A higher score indicates more complexity. The PCAM is a practical instrument for assessing bio-psycho-social dimensions of patients with chronic conditions, and healthcare professionals can utilize PCAM to systematically assess patients’ needs, taking into account their severity and urgency [17]. The Japanese adaptation of the PCAM has undergone validation in several primary care settings in Japan [18, 19]. The high score of PCAM is related to increased length of hospital stay [20] and alcohol use disorders [21]. Previous studies reported that an average PCAM score was of 25.0 (standard deviation (SD): 7.3) among inpatients in a small hospital in Tokyo, Japan [20], 21.4 (SD: 5.7) among outpatients at a clinic on a remote island in Japan [19, 21], and 16.5 (SD: 5.1) among outpatients at family medicine clinics in Japan [18]. A previous study in hospital settings in Japan shows that PCAM has a two-factor structure: medicine-oriented complexity (external factors such as social health service and physical health needs) and patient-oriented complexity (internal patient factors such as physical and mental well-being and literacy) [20].
Table 1.
Summary of the PCAM components and scoring system
| Health and well-being (4 items; Max 16, Min 4) |
|---|
| 1. Physical health needs From “All problems are already scrutinized” (score: 1) To “Problems influence daily living severely” (socre:4) |
| 2. Physical health impacting mental well-being From “No problem is found” (score:1) To “Severe mental disturbance impedes daily living” (socre:4) |
| 3. Lifestyle impacting well-being (e.g. alcohol, drug, diet, exercise) From “No problem is found” (score:1) To “Severe problems in lifestyle influence well-being, sometimes other people’s one” (socre:4) |
| 4. Other mental well-being concerns From “No problem is found” (score:1) To “Severe problems impede most of daily function” (socre:4) |
| Social environment (4 items; Max 16, Min 4) |
| 1. Home environment From “Consistently safe” (score: 1) To “Neither safe nor stable” (score: 4) |
| 2. Daily activities From “Benefit or no problem” (score: 1) To “Severely damaged mental well-being” (score: 4) |
| 3. Social network From “Adequately participating” (score: 1) To “Lonely and isolated” (score: 4) |
| 4. Financial resources From “Stable, enough income” (score: 1) To “Unstable, little income” (score: 4) |
| Literacy (2 items; Max 8, Min 2) |
| 1. Health literacy From “Good understanding and practice” (score: 1) To “Not understanding important things for health management” (score: 4) |
| 2. Engagement in discussion From “Open communication” (score: 1) To “Severe difficulty in communication” (score: 4) |
| Service coordination (2 items; Max 8, Min 2) |
| 1. Other services From “No other care/service needed” (score: 1) To “Needing but not receiving care/service at all” (score: 4) |
| 2. Service coordination From “Consistently safe” (score: 1) To “Neither safe nor stable” (score: 4) |
Data collection
The PCAM scores were assessed at two distinct time points: (i) pre-intervention, or the initial consultation, and (ii) post-intervention, defined as the discharge date following the initial admission for inpatients, or the conclusion of follow-up care (or 6 months after the initial consultation) for outpatients. The first and second authors reviewed each patient’s medical chart and assigned PCAM scores retrospectively through recurrent discussions to ensure consistency and accuracy. First, the two authors independently assigned PCAM scores to the three participants, after reading all related documents and information in the medical record. The authors shared the scores with each other, discussed the differences, and reached a consensus. The PCAM scores were then assigned to the remaining participants, discussed, and a definitive consensus was reached.
We also collected various factors influencing patient complexity. First, a review of medical records was conducted to identify participants’ comorbidities and determine whether they had any of the top 20 chronic conditions associated with multimorbidity, as defined by Prados-Torres et al. in their systematic review [22]. These conditions include chronic obstructive pulmonary disease, diabetes, hypertension, malignancy, stroke, dementia, depression, joint disease, anxiety, chronic heart failure, coronary heart disease, asthma, cardiac arrhythmia, thyroid disease, anemia, hearing impairment, dyslipidemia, obesity, prostatic hypertrophy, and osteoporosis. Second, participants’ cognitive decline level was assessed using the mini-mental state examination (MMSE). The MMSE scores range from 0 to 30, with lower scores indicating greater cognitive decline. A cutoff score of 24 allows for the detection of dementia with a sensitivity of 0.85 and a specificity of 0.90 [23]. As part of routine care, patients with clinically suspected cognitive decline were assessed on the MMSE by a psychiatrist or clinical psychologist. The MMSE score first measured after referral was collected by chart review. MMSE was not calculated in participants with no clinically suspected cognitive decline. Third, participants’ levels of long-term care (LTC) certification were collected. In Japan, individuals aged 65 and older can apply for LTC certification if they currently require or are anticipated to require LTC. Applicants are categorized based on their condition into three groups: not applicable, in need of support, or in need of care. The LTC certification consists of seven levels: Support Levels 1 and 2 (for individuals requiring daily assistance) and Care Need Levels 1 to 5 (for those needing support with fundamental activities such as eating and bathing). Higher levels indicate a greater need for nursing care [24]. This classification system is applied nationwide in Japan and has been clinically validated [25]. Generally, individuals classified at Care Need Level 2 experience significant limitations in instrumental activities of daily living [26]. The LTC status, whether certified at the time of referral or first approved after referral, was collected through chart review. This data was not assessed for patients ineligible for LTC certification. Patients who applied for LTC but were transferred before receiving certification results, with no subsequent data available, were classified as missing data. These data were collected and abstracted by the first author. Missing data was deleted.
Data analysis
The PCAM scores (both each item, each domain, and total) were summarized as an average with SD and compared between pre-intervention and post-intervention using paired-sample t-tests. To quantify the magnitude of changes, the standardized mean difference was calculated using Hedges’ g, a widely recognized measure [27]. Based on previous research, Hedges’g values of 0.16, 0.38, and 0.76 were interpreted as small, medium, and large effects, respectively [28]. These analyses were performed by the first author and comprehensively reviewed by the third author, who is not a staff of the hospital.
Ethical considerations
Patient consent was obtained on an opt-out basis. A notice about the study was posted in the hospital waiting room and on the website. Patients who did not want their data to be used for the study were given the opportunity to refuse. The study was approved by the Ethical Committee of Ehime Seikyo Hospital (No. 51-2-2023-003; approval date: Feb 15, 2024). This study adhered to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement [29] and the Consensus Reporting Items for Studies in Primary Care (CRISP) statement [30].
Results
A total of 41 cases were enrolled in the study. Of these, 3 patients died shortly after the initial consultation. There was no patient with missing follow-up. Among the remaining 38 patients, 24 were admitted as inpatients, while 14 were managed on an outpatient basis. Admission days of inpatients was 33.5 on median (IQR: 17–46.3). Psychiatric liaison was offered to 25 patients. The median age was 73.5 years (interquartile range [IQR]: 68.25–82). The median number of comorbidities among the top 20 chronic conditions was 4 (IQR: 2.25–4). Other than the 20 conditions, 5 participants (13%) developed alcohol use disorder. A total of 3 participants (7.9%) were certificated as Support Level, and 21 participants (55%) as Care Need Level. A total of 16 participants (42%) had clinically intact cognitive function. MMSE was assessed among 20 participants and the median score was 17 (IQR: 13–20.75).
The number of patients referred from three community comprehensive support centers located near the hospital was 10, 4, and 10, respectively. Additionally, three other centers situated farther from the hospital within the city referred a total of four patients, while a center outside the city referred two patients.
The municipal office of the city referred two patients, while two offices in other cities collectively referred three patients. The city’s public health center referred one patient, whereas a center in another city referred two patients. Additionally, the city’s social welfare council referred two patients, and a council in another city referred one patient. Table 2 shows the details of participants’ demographics.
Table 2.
Patients’ demographics
| Age (median: IQR) [years old] | 73.5 (68.25–82) |
|---|---|
| Sex (Female) [n, %] | 19 (50.0%) |
| Inpatient [n, %] | 24 (63.2%) |
| Psychiatric Liaison [n, %] | 25 (65.8%) |
| Encounters in… | |
| 2020 | 5 |
| 2021 | 9 |
| 2022 | 12 |
| 2023 | 12 |
| Referring from… | |
| Community comprehensive support centers (near in the city) | 24 (from 3 centers) |
| Community comprehensive support centers (distant in the city) | 4 (from 3 centers) |
| Community comprehensive support centers (outside the city) | 2 (from 1 center) |
| Municipal office (of the city) | 2 |
| Municipal offices (outside the city) | 3 (from 2 offices) |
| Public health center (of the city) | 1 |
| Public health centers (of the city) | 2 (from 1 center) |
| Social welfare council (of the city) | 2 |
| Social welfare council (outside the city) | 1 (from 1 council) |
| The number of comorbidities [n, %] | |
| 1 condition | 4 (11%) |
| 2 conditions | 6(16%) |
| 3 conditions | 7 (18%) |
| 4 conditions | 7 (18%) |
| 5 conditions | 6 (16%) |
| 6 conditions | 6 (16%) |
| 7 conditions | 1 (2.6%) |
| 8 conditions | 1 (2.6%) |
| The number of participants with each chronic condition [n, %] | |
| Chronic obstructive pulmonary disease | 6 (16%) |
| Diabetes | 12 (32%) |
| Hypertension | 18 (47%) |
| Malignancy | 4 (11%) |
| Stroke | 11 (29%) |
| Dementia | 22 (58%) |
| Depression | 6 (16%) |
| Joint disease | 13 (34%) |
| Anxiety | 4 (11%) |
| Chronic heart failure | 6 (16%) |
| Coronary heart disease | 2 (5.3%) |
| Asthma | 0 (0%) |
| Arrhythmia | 3 (7.9%) |
| Thyroid disease | 3 (7.9%) |
| Anemia | 11 (29%) |
| Hearing impairment | 3 (7.9%) |
| Dyslipidemia | 12 (32%) |
| Obesity | 3 (7.9%) |
| Prostatic hypertrophy | 3 (7.9%) |
| Osteoporosis | 4 (11%) |
| Long-term care certification [n, %] | |
| No functional decline | 8 (21%) |
| Support Level 1 | 2 (5.2%) |
| Support Level 2 | 1 (2.6%) |
| Care Need Level 1 | 9 (24%) |
| Care Need Level 2 | 4 (11%) |
| Care Need Level 3 | 5 (13%) |
| Care Need Level 4 | 3 (7.9%) |
| Care Need Level 5 | 0 (0%) |
| No available data | 6 (16%) |
| Mini-mental State Examination [n, %] | |
| No cognitive decline (clinically intact) | 16 (42%) |
| 21–25 | 5 (13%) |
| 16–20 | 8 (21%) |
| 11–15 | 4 (10%) |
| < 10 | 3 (7.9%) |
| Unable to test for severe cognitive decline | 1 (2.6%) |
| No available data | 1 (2.6%) |
The mean initial PCAM score was 36.9 (SD: 6.1), with a range of 21 to 45. The domain-specific averages were as follows: health and well-being (2.96 [SD: 0.71]), social environment (3.09 [SD: 0.66]), health literacy and communication (2.78 [SD: 0.66]), and service coordination (3.61 [SD: 0.62]). Following interventions, the overall PCAM score was 23.7 (SD: 4.1), with domain-specific averages of 1.95 (SD: 0.40), 1.96 (SD: 0.46), 2.46 (SD: 0.72), and 1.57 (SD: 0.55), respectively. Hedge’s g was largest in the domain of service coordination (4.68), and thus social environment (2.39), health and wellbeing (2.00), and smallest in the domain of literacy (0.67). Table 3 shows the details. There was no missing data about PCAM scores.
Table 3.
Transition of PCAM scores
| Pre [average (SD)] | Post [average (SD)] | P-value (t test) | Hedge’s g | |
|---|---|---|---|---|
| Total | 36.9 (6.1) | 23.7 (4.1) | <.001 | 2.54 |
| Health and well-being | 2.96 (0.71) | 1.95 (0.40) | <.001 | 2.00 |
| -Physical health needs | 3.11 (0.83) | 1.34 (0.53) | <.001 | 2.52 |
| -Physical health impacting mental well-being | 2.94 (0.77) | 2.18 (0.61) | <.001 | 1.10 |
| -Lifestyle impacting mental well-being | 2.76 (1.1) | 2.03 (0.64) | <.001 | 0.85 |
| -Other mental well-being concerns | 3.03 (0.85) | 2.26 (0.55) | <.001 | 1.06 |
| Social environment | 3.09 (0.66) | 1.96 (0.46) | <.001 | 2.39 |
| -Home environment | 3.13 (0.99) | 1.47 (0.56) | <.001 | 2.06 |
| -Daily activities | 3.16 (0.82) | 2.11 (0.61) | <.001 | 1.46 |
| -Social network | 3.47 (0.56) | 2.37 (0.59) | <.001 | 1.93 |
| -Financial resources | 2.58 (1.08) | 1.89 (0.69) | <.001 | 0.75 |
| Literacy | 2.78 (0.66) | 2.46 (0.72) | <.001 | 0.67 |
| -Health literacy | 2.97 (0.72) | 2.53 (0.83) | <0.001 | 0.58 |
| -Engagement in discussion | 2.58 (0.79) | 2.39 (0.72) | .009 | 0.24 |
| Service coordination | 3.61 (0.62) | 1.57 (0.55) | <.001 | 4.68 |
| Other services | 3.63 (0.59) | 1.58 (0.55) | <.001 | 3.60 |
| Service coordination | 3.58 (0.68) | 1.56 (0.60) | <.001 | 3.15 |
The amount of change in PCAM scores among the inpatient group was larger than that among the outpatient group (Hedge’s g: 3.62 vs 0.72). The same trend was observed in each category. Table 4 shows the details.
Table 4.
PCAM scores among the inpatient group and among the outpatient group.
| Inpatient (N = 24) | Outpatient (N = 14) | |||||||
|---|---|---|---|---|---|---|---|---|
| Sex (male) [N, %] | 12 (50%) | 7 (50%) | ||||||
| Age (Average) [years old] | 70.7 | 72.7 | ||||||
| Psychiatric Liaison [N, %] | 16 (66.7%) | 9 (64.3%) | ||||||
| PCAM scores | Pre [average (95% CI)] | Post [average (95% CI)] | P-value (t test) | Hedge’s g | Pre [average (95% CI)] | Post [average (95% CI)] | P-value (t test) | Hedge’s g |
|---|---|---|---|---|---|---|---|---|
| Total | 39.8 (3.97) | 22.7 (3.45) | <.001 | 3.62 | 32.1 (6.22) | 25.4 (4.77) | <.001 | 0.72 |
| Health and wellbeing | 3.20 (0.62) | 1.88 (0.27) | <.001 | 2.39 | 2.55 (0.70) | 2.09 (0.55) | <.001 | 0.55 |
| Social environment | 3.35 (0.56) | 1.88 (0.47) | <.001 | 2.97 | 2.63 (0.59) | 2.11 (0.42) | <.001 | 0.73 |
| Literacy | 2.94 (0.60) | 2.49 (0.67) | <.001 | 0.86 | 2.50 (0.71) | 2.43 (0.83) | .17 | 0.08 |
| Service coordination | 3.83 (0.35) | 1.38 (0.45) | <.001 | 7.81 | 3.21 (0.78) | 1.89 (0.56) | <.001 | 1.42 |
Discussion
This study revealed the high complexity of patients referred from community centers to hospital family physicians and the significant impact of their interventions in reducing this complexity. Comparing with previous studies [18, 19, 21, 22], the pre-intervention average score of 36.9 indicates the extreme complexity of cases managed by hospital family physicians. Although no consensus exists on the clinically significant change in PCAM scores, the average post-intervention score of 23.7 is comparable to that of inpatients in a small hospital (25.0) [20], suggesting a level of complexity that aligns with common clinical presentations. Such a notable reduction in patient complexity is particularly evident in the domain of service coordination (the average PCAM score: from 3.61 to 1.57). This finding aligns with previous research emphasizing that structured care coordination improves outcomes for high-need, high-cost patients by addressing fragmented care and social determinants of health [31]. For patients in complex and challenging situations, establishing a connection to medical care serves as a pivotal turning point, potentially increasing their openness to a range of care options. Successfully linking patients with complex needs to appropriate care necessitates intensive care coordination, team-based approaches, integration of medical and behavioral health services, and mitigation of social barriers [32]. These findings underscore that an integrated care approach may be also highly effective in addressing the needs of patients with extremely complex and multifaceted challenges.
In the domain of health and well-being, notable improvements were also observed (from 2.96 to 1.95), particularly in addressing undiagnosed conditions and symptoms requiring further evaluation. Patients in complex and challenging situations often present with a higher prevalence of ongoing medical issues, inadequately treated chronic conditions, and ill-defined symptoms and functional impairments [33]. Furthermore, these patients frequently experience comorbid psychiatric and cognitive disorders [33], especially in cases where significant barriers exist between them and medical institutions. Our findings indicate that hospital family physicians play a critical role in diagnosing and simplifying these multifaceted problems, facilitating appropriate referrals, and optimizing treatment plans. Importantly, this underscores the diagnostic expertise of hospital-based family physicians, who integrate biomedical, psychological, and social factors to ensure a patient-centered, equitable, and safe diagnostic process [34]. Such efforts highlight the unique contributions of hospital family physicians in navigating complex healthcare needs and fostering integrated care solutions.
In the domain of social environment, significant improvements were observed (from 3.09 to 1.96), particularly in housing stability and social networks. This suggests that hospitalization itself may provide an opportunity to identify and intervene in social determinants of health. Hospital wards can serve as temporary safe environments, particularly for patients experiencing unsafe living conditions or social isolation. A comparable model is the acute frailty unit (AFU) [35], which focuses on mitigating functional decline among older individuals during acute hospitalization. AFUs operate under the principles of tailored environments, patient-centered care, comprehensive discharge planning, and thorough medical reviews to minimize polypharmacy and unnecessary interventions. Evidence suggests that compared to standard hospitalization, AFUs are associated with fewer adverse outcomes, such as falls and delirium [36], and are effective in reducing emergency department visits and hospital readmissions [37]. These findings reinforce the need for hospital-based family physicians to adopt a holistic approach that extends beyond medical stabilization to social reintegration.
The domain of literacy showed relatively minimal improvement compared to other domains, potentially reflecting the higher prevalence of irreversible cognitive decline in older populations. Assessing patients with complex problems requires an evaluation of their health literacy, including factors such as mistrust of healthcare professionals and entrenched health beliefs [31]. Establishing strong therapeutic relationships with such patients, coupled with an emphasis on patient autonomy and shared decision-making, may prove instrumental in achieving further quality improvements [38, 39]. This approach not only enhances communication but also fosters greater engagement and alignment with the patient’s values and preferences. Future studies should explore targeted interventions to enhance health literacy among patients with complex needs.
In this study, inpatients exhibited higher levels of complexity before the intervention and lower levels after the intervention. This may be attributed to the concentrated care provided in the ward and the tendency of medical professionals to recommend hospitalization for patients experiencing significant disruptions in their health and social situations. Such interventions, however, cannot be effectively achieved by medical professionals working in isolation. Collaborative care models, in which healthcare professionals work closely with community health workers, have been associated with improved outcomes in managing patients with complex needs [32]. Even when care is delivered in a hospital ward, it is essential to develop and implement these interventions in partnership with the community to ensure continuity and sustainability of care.
Clinicians can derive several key insights from the findings of this study. First, hospital-based family medicine plays a crucial role in enhancing service coordination. Collaborative care involving multiple professionals both within and outside the hospital can significantly improve patients’ experiences of care. Second, family physicians must recognize that most patients facing socially complex and challenging situations also present with multiple undifferentiated health problems. Clarifying and streamlining these issues is a fundamental role of physicians, facilitating effective collaboration among multidisciplinary professionals, patients, families, and other stakeholders. Third, hospital family physicians can strategically utilize hospital wards—environments distinct from patients’ everyday living conditions. By leveraging the ward setting, family physicians can create the necessary time and space to stabilize patients’ health concerns while concurrently working to establish a safer living environment.
This study has several limitations. First, many individuals in complex and challenging situations addressed by community centers decline hospital referrals. According to feedback from community center staff, only a small fraction of individuals with complex needs agree to be referred, and those with the most severe needs are often the ones who refuse hospital care. Future research should focus on understanding the experiences and challenges of patients with complex healthcare-related needs who do not seek or accept hospital care. In addition, neither the community centers nor the hospital used standardized criteria for referrals. The referral process was influenced by multiple factors, potentially complicating the interpretation of results. To mitigate selection bias, we included all referred participants during the study period. Second, this study did not reveal the specifics of what the patients experienced, what interventions were implemented, and which aspects of care were most effective in reducing complexity. Understanding the nature and impact of interventions is crucial for refining and optimizing care strategies. We outlined our principles of patient-tailored care to aid in the interpretation of findings. Third, PCAM scores were assessed retrospectively, introducing the potential for recall bias. Although PCAM scores were determined through consensus between two authors and statistical analysis was reviewed by a researcher outside the hospital, the first author was involved in all aspects of study design, data collection, and statistical analysis, posing a risk of bias. Previous studies have highlighted concerns regarding the reliability of medical record reviews [40]. In this study, interobserver reliability was not tested, and blind analysis was not conducted. Fourth, we did not adjust for background factors in statistical analyses. Despite our efforts to summarize participants’ backgrounds, some key variables (e.g., socioeconomic status) were not collected. Lastly, this study was conducted at a single institution, which limits the generalizability of its findings. We provided detailed descriptions of the hospital’s context, its surrounding healthcare infrastructure, and the characteristics of the community centers making referrals. Additionally, we noted that most participants were older adults, many with cognitive decline and some with alcohol use disorders. Contextualized care is a defining feature of family medicine, and the specific challenges patients face are shaped by the practice setting. The hospital studied operates within a distinct community healthcare network developed over years, and readers should consider differences from their own practice environments when interpreting these findings.
To address these limitations, future research should focus on several key areas. First, there remains a significant knowledge gap regarding the experiences and challenges of patients with complex healthcare needs who do not seek or accept hospital care. Second, detailed analyses of the care provided by hospital family physicians and other healthcare professionals, along with patient-reported outcomes, are needed. Third, to minimize recall bias and enhance the reliability of data sources, prospective and blinded studies should be conducted. More comprehensive data collection on participants’ backgrounds would also improve the robustness of statistical analyses. Finally, multicenter studies are essential to validate these findings and ensure their applicability across diverse settings and populations.
The strength of this study lies in demonstrating that hospital family physicians actively accept patients with highly complex healthcare needs and deliver collaborative, community-based care that effectively reduces their complexity. This finding highlights the crucial role of hospital family physicians in addressing the challenges of an aging population with multifaceted medical and social needs.
Conclusion
Hospital family physicians in Japan frequently manage patients with exceptionally complex health and social problems referred from community centers. The findings of this study imply that hospital family practice care may mitigate patient complexity across multiple domains. This study underscores the unique and vital role of hospital family physicians in providing integrated, patient-centered care. Their ability to address complexity with diagnostic excellence, coordination of resources, and collaboration with community-based systems highlights their professional competence and importance in meeting the needs of patients in challenging circumstances. Further research is needed to expand upon these findings and explore strategies for optimizing care for complex patients in diverse settings.
Contributor Information
Junki Mizumoto, Center for General Medicine Education, School of Medicine, Keio University, Tokyo, Japan; Department of Family Practice, Ehime Seikyo Hospital, Matsuyama, Japan.
Yumiko Hironaka, Regional medical coordination office, Ehime Seikyo Hospital, Matsuyama, Japan.
Hirohisa Fujikawa, Center for General Medicine Education, School of Medicine, Keio University, Tokyo, Japan.
Conflict of interest
None declared.
Funding
This work was supported by JSPS KAKENHI Grant Number 24K23757.
Ethical approval
The study was approved by the Ethical Committee of Ehime Seikyo Hospital (No. 51-2-2023-003).
Data availability
The data underlying this article will be shared on reasonable request to the corresponding author.
References
- 1. College of Family Physicians of Canada. Four principles of family medicine. Mississauga, ON: College of Family Physicians of Canada, 2006. www.cfpc.ca/principles [Google Scholar]
- 2. Starfield B, Shi L, Macinko J. Contribution of primary care to health systems and health. Milbank Q 2005;83:457–502. https://doi.org/ 10.1111/j.1468-0009.2005.00409.x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. American Academy of Family Physicians. Family medicine specialty. https://www.aafp.org/about/the-aafp/family-medicine-specialty.html (17 April 2025, date last accessed). [Google Scholar]
- 4. Newton W, Sutton M, Asher GN. The elephant in the room. Fam Med 2014;46:85–7. [PubMed] [Google Scholar]
- 5. Starfield B, Shi L, Macinko J. Contribution of primary care to health systems and health. Milbank Q. 2005;83:457–502. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Mears P. New models of care for a local district general hospital. Future Hosp J 2015;2:107–10. https://doi.org/ 10.7861/futurehosp.2-2-107 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. The SHM/STFM Joint Task Force for Hospital Medicine Training of Family Medicine Residents. Hospital Medicine Curricular Guidelines or Track Recommendations for Family Medicine Residencies. https://resourcelibrary.stfm.org/viewdocument/hospital-medicine-curricular-guidel?CommunityKey=2751b51d-483f-45e2-81de-4faced0a290a&tab=librarydocuments (17 April 2025, date last accessed). [Google Scholar]
- 8. Jetty A, Jabbarpour Y, Petterson S, et al. The declining presence of family physicians in hospital-based care. J Am Board Fam Med 2019;32:771–2. https://doi.org/ 10.3122/jabfm.2019.06.190152 [DOI] [PubMed] [Google Scholar]
- 9. Neimanis I, Woods A, Zizzo A, et al. Role of family physicians in an urban hospital: Tracking changes between 1977, 1997, and 2014. Can Fam Physician 2017;63:221–7. [PMC free article] [PubMed] [Google Scholar]
- 10. Tooi T, Murata A, Ota H, et al. Survey on the activities of certificated family physicians. Journal of Japan Primary Care Association 2016;39:243–249. [in Japanese]. https://doi.org/ 10.14442/generalist.39.243 [DOI] [Google Scholar]
- 11. Yokota Y, Watari T. Various perspectives of “General Medicine” in Japan—Respect for and cooperation with each other as the same “General Medicine Physicians.”. J Gen Fam Med 2021;22:314–5. https://doi.org/ 10.1002/jgf2.500 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Kaneko M, Oishi A, Sawa N, et al. Perspectives on expert generalist practice among Japanese family doctor educators: a qualitative study. BJGP Open 2021;5:BJGPO.2021.0011. https://doi.org/ 10.3399/BJGPO.2021.0011 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Mizumoto J, Mitsuyama T, Eto M, et al. Primary care physicians’ perceptions of social determinants of health recommendations: a qualitative study. BJGP Open 2023;7:BJGPO.2022.0129. https://doi.org/ 10.3399/BJGPO.2022.0129 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Ministry of Health, Labour and Welfare. Comprehensive community care system.https://www.mhlw.go.jp/stf/seisakunitsuite/bunya/hukushi_kaigo/kaigo_koureisha/chiiki-houkatsu/ (17 April 2025, date last accessed). [Google Scholar]
- 15. Ministry of Health, Labour and Welfare. Guidance on community comprehensive support center. 2007. https://www.mhlw.go.jp/topics/2007/03/tp0313-1.html (17 April 2025, date last accessed). [Google Scholar]
- 16. Kato D, Ryu H, Matsumoto T, et al. Building primary care in Japan: literature review. J Gen Fam Med 2019;20:170–9. https://doi.org/ 10.1002/jgf2.252 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Pratt R, Hibberd C, Cameron IM, et al. The Patient Centered Assessment Method (PCAM): integrating the social dimensions of health into primary care. J Comorb 2015;5:110–9. https://doi.org/ 10.15256/joc.2015.5.35 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Mutai R, Sugiyama Y, Yoshida S, et al. Development and validation of a Japanese version of the Patient Centred Assessment Method and its user guide: a cross-sectional study. BMJ Open 2020;10:e037282. https://doi.org/ 10.1136/bmjopen-2020-037282 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Sugiyama Y, Mutai R, Yoshimoto H, et al. Structural validity and internal consistency of the Patient Centred Assessment Method in a primary care setting in a Japanese island area: a cross-sectional study. BMJ Open 2022;12:e050566. https://doi.org/ 10.1136/bmjopen-2021-050566 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Yoshida S, Matsushima M, Wakabayashi H, et al. Validity and reliability of the Patient Centred Assessment Method for patient complexity and relationship with hospital length of stay: a prospective cohort study. BMJ Open 2017;7:e016175. https://doi.org/ 10.1136/bmjopen-2017-016175 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Sugiyama Y, Matsushima M, Yoshimoto H. Association between alcohol consumption/alcohol use disorders and patient complexity: a cross-sectional study. BMJ Open 2020;10:e034665. https://doi.org/ 10.1136/bmjopen-2019-034665 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Prados-Torres A, Calderón-Larrañaga A, Hancco-Saavedra J, et al. Multimorbidity patterns: a systematic review. J Clin Epidemiol 2014;67:254–66. https://doi.org/ 10.1016/j.jclinepi.2013.09.021 [DOI] [PubMed] [Google Scholar]
- 23. Creavin ST, Wisniewski S, Noel-Storr AH, et al. Mini-Mental State Examination (MMSE) for the detection of dementia in clinically unevaluated people aged 65 and over in community and primary care populations. Cochrane Database Syst Rev 2016;2016:CD011145. https://doi.org/ 10.1002/14651858.CD011145.pub2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Yamada M, Arai H. Long-term care system in Japan. Ann Geriatr Med Res 2020;24:174–80. https://doi.org/ 10.4235/agmr.20.0037 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Tsutsui T, Muramatsu N. Care-needs certification in the long-term care insurance system of Japan. J Am Geriatr Soc 2005;53:522–7. https://doi.org/ 10.1111/j.1532-5415.2005.53175.x [DOI] [PubMed] [Google Scholar]
- 26. Takahashi S, Yonekura Y, Takanashi N, et al. Risk factors of long-term care insurance certification in Japan: a scoping review. Int J Environ Res Public Health 2022;19:2162. https://doi.org/ 10.3390/ijerph19042162 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Higgins JPT, Li T, Deeks JJ. Chapter 6: Choosing effect measures and computing estimates of effect. In: Higgins JPT, Thomas J (eds.), Cochrane Handbook for Systematic Reviews of Interventions. Chichester (UK): Ver 6.5 The Cochrane Collaboration, 2024. https://training.cochrane.org/handbook/current/chapter-06 (17 April 2025, date last accessed). [Google Scholar]
- 28. Brydges CR. Effect size guidelines, sample size calculations, and statistical power in gerontology. Innov. Aging 2019;3:igz036. https://doi.org/ 10.1093/geroni/igz036 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. von EE, Altman DG, Egger M, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. J Clin Epidemiol 2008;61:344–9. [DOI] [PubMed] [Google Scholar]
- 30. Phillips WR, Sturgiss E, Glasziou P, et al. Improving the reporting of primary care research: consensus reporting items for studies in primary care-the CRISP statement. Ann Fam Med 2023;21:549–55. https://doi.org/ 10.1370/afm.3029 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Khalili J, Jeong J, Tibbe TD, et al. A novel curriculum for internal medicine residents to care for high-need, high-cost patients. J Med Educ Curric Dev 2024;11:23821205241246889. https://doi.org/ 10.1177/23821205241246889 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Komaromy M, Bartlett J, Zurawski A, et al. ECHO care: providing multidisciplinary specialty expertise to support the care of complex patients. J Gen Intern Med 2020;35:326–30. https://doi.org/ 10.1007/s11606-019-05205-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Quek JS, Lee ES, Low LL, et al. How family physicians in Singapore recognise complexity during consultations: a qualitative study. BMC Prim Care 2024;25:134. https://doi.org/ 10.1186/s12875-024-02368-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Watari T, Schiff GD. Diagnostic excellence in primary care. J Gen Fam Med 2023;24:143–5. https://doi.org/ 10.1002/jgf2.617 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35. Soong JTY, Bell D, Ong MEH. Meeting today’s healthcare needs: medicine at the interface. Ann Acad Med Singap 2022;51:787–92. https://doi.org/ 10.47102/annals-acadmedsg.2022196 [DOI] [PubMed] [Google Scholar]
- 36. Fox MT, Persaud M, Maimets I, et al. Effectiveness of acute geriatric unit care using acute care for elders components: a systematic review and meta-analysis. J Am Geriatr Soc 2012;60:2237–45. https://doi.org/ 10.1111/jgs.12028 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37. Chong E, Zhu B, Tan H, et al. Emergency Department Interventions for Frailty (EDIFY): front-door geriatric care can reduce acute admissions. J Am Med Dir Assoc 2021;22:923–8.e5. https://doi.org/ 10.1016/j.jamda.2021.01.083 [DOI] [PubMed] [Google Scholar]
- 38. Chang E, Ali R, Berkman ND. Unpacking complex interventions that manage care for high-need, high-cost patients: a realist review. BMJ Open 2022;12:e058539. https://doi.org/ 10.1136/bmjopen-2021-058539 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39. Wyman MF, Liebzeit D, Voils CI, et al. “Hopes and wishes”: Goals of high-need, high-cost older patients and their caregivers. Patient Educ Couns 2020;103:1428–34. https://doi.org/ 10.1016/j.pec.2020.02.022 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40. Worster A, Bledsoe RD, Cleve P, et al. Reassessing the methods of medical record review studies in emergency medicine research. Ann Emerg Med 2005;45:448–51. https://doi.org/ 10.1016/j.annemergmed.2004.11.021 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The data underlying this article will be shared on reasonable request to the corresponding author.