Evolving practice of general internal medicine in the current medical landscape

Abstract

General internal medicine is a specialty that is heterogeneously implemented across the world. In some places, it includes physicians who have undergone basic internal medicine training with no subspecialisation, whereas in other places, these physicians undergo further training to become ‘advanced internists’ (with the corollary name of advanced internal medicine). General internists run inpatient medical wards and ambulatory care services, provide consultation to general practitioners and/or have their own primary care practice. Within the hospital, the role of the general internist had traditionally been ambiguous – often serving as a ‘catch‐all’ medical service for patients deemed unsuitable to be managed under subspecialty teams. Over the years, different models of care within general medicine have been trialled to varying effects, ranging from the ‘generalist‐specialist’ model (also known as the hospitalist model, where the general internist coordinates care for multimorbid patients with liaison with relevant subspecialties) and ‘acute medicine’ model (general medicine care only for the first 72 h of admission in an acute medical unit, after which patients who require continued admission go to a subspecialty) to the ‘specialist‐generalist’ model (where all specialists are to care holistically for general medical issues beyond the specific condition that warrants subspecialty review). In this article, we advocate for a renewed framework of general internal medicine that focusses on managing undifferentiated, complex and multimorbid patients, providing clinical care that addresses prevailing gaps in hospital‐based practice, developing robust hospital‐at‐home services and championing quality improvement, education and administrative excellence within the medical ecosystem.

Introduction

General internal medicine (GIM) is an internal medicine subspecialty that was formerly recognised in the 1970s. ¹ Since then, it has largely remained fairly ambiguous and heterogeneously implemented across the world. ¹ In the United States, basic training (3 years) in internal medicine rotations would suffice to become a general internist/hospitalist, while training can be even shorter (2 years) in places like Russia and Kazakhstan. ¹ On the other hand, in European countries, and in Singapore, GIM, also known as advanced internal medicine (AIM), requires an additional 2–3 years of further training after basic internal medicine training. ¹ , ² In the United Kingdom, GIM is typically part of dual accreditation in concert with subspecialty training. ³ Unlike most countries where general internists are only involved in hospital‐based clinical services, some places such as the United States and Japan also require them to provide primary care services in the community. ⁴

In Australasian countries like Australia and New Zealand, general and acute care medicine is a well‐established, stand‐alone specialty, with a dedicated training pathway, fixed curriculum and clearly defined clinical duties in both inpatient services and ambulatory care. ⁵ Unfortunately, this is not the case in many other parts of the world, where GIM often languishes in an ambiguous zone, with heterogeneous implementation and acceptance of its model of care.

In the past, GIM used to be seen as running a ‘catch‐all’ inpatient service, where their role was ‘pragmatically defined … as caring for all non‐elective admissions not positively claimed by any specialty’. ⁶ Over time, other models have emerged. For instance, the ‘acute medicine’ model in the United Kingdom National Health Service limits generalist care to the first 72 h of admission to the acute medical unit (AMU), after which patients who require continued admission end up in a subspecialty ward team assigned by the AMU team. ³ Elsewhere, the role of the general internist may be seen as providing holistic care for complex, multimorbid patients ⁷ or serving as the ‘bridge’ between subspecialties, ⁴ while executing eventual treatment decisions as the overall physician‐in‐charge with a helicopter view of the patient's overall health situation. More recently, radical models such as the specialist‐generalist approach have been proposed, where specialists are trained and required to go ‘broader’ in their care of patients beyond their domain expertise. ⁸

In this article, we propose a renewed framework for conceptualising the modern GIM service, which comprises four key domains: (i) provision of general medicine services for undifferentiated, complex and/or multimorbid cases, (ii) creation of adjunctive subspecialty consultation services to meet pre‐existing needs/gaps in hospital‐based care, (iii) development of robust hospital‐at‐home services and (iv) championing quality improvement and educational and administrative efforts within the medical ecosystem.

Spectrum of GIM clinical services

First, general internists are best placed to care for patients with undifferentiated medical complaints to institute initial management and subsequently decide on their disposition. The AMU is a prime example of a generalist‐run inpatient service that provides rapid access to care for generic medical admits (e.g. functional decline, falls, giddiness) and have admitting rights for subsequent right siting of patients who require continued admission. ³ In addition, general medicine physicians can provide holistic consultation services for a wide range of clinical complaints suspected to have an organic medical cause. For example, a recent observational study in Japan found that GIM consultations were often sought from other departments for non‐specific complaints such as fever, abnormal laboratory results, fatigue, pain and numbness, for which >70% of consultations led to new established diagnoses, with >60% of patients having favourable outcomes. ⁹ In ambulatory outpatient care, general internists can play a meaningful intermediary role in triaging referrals, such that those who truly require subspecialty input can be given an appointment in a timely fashion while the rest can be managed by generalists. A recent Australian study showed that general medicine clinics were able to triage effectively and manage patients who were initially meant for referrals to rheumatology, achieving dual objectives of minimising clinic wait times and unnecessary subspecialty consultations. ¹⁰

Second, there has recently been an emergence of enhanced care units situated within generalist‐run AMUs as a bridging service between general ward and intensive care, targeting patients who may not (yet) require or desire intensive care admission. ¹¹ Eligible patients are usually those who require closer monitoring (continuous vitals monitoring, lower nursing‐to‐patient ratio), peripheral inotropic support or non‐invasive ventilatory support (e.g. bilevel or continuous positive airway pressure, high‐flow nasal cannula). ¹¹ On the other end of the spectrum, for patients who can be discharged from the emergency department (ED) with early outpatient follow‐up, generalist‐run ambulatory ‘hot’ clinics facilitate timely discharge of suitable patients. ¹²

Third, in the present healthcare landscape featuring a chronic disease epidemic and increasingly ‘complex’ patients with multimorbidities, there is a greater need for a generalist‐led integrated care model to provide consolidated, person‐centred care. By having broad‐based expertise and attunement to social determinants of health, generalists are well‐placed to engage in shared decision‐making with patients based on considerations of ‘whole‐person’ health rather than focussing solely on single‐organ disease. Naturally, this promotes care centricity of clinical decisions – where tests performed are more likely to change patients’ management/disease trajectory, and interventions administered are meaningful and goals‐directed. ¹³ Otherwise, many patients with multiple comorbidities end up with recurrent admissions and excessive clinic follow‐ups, which lead to fragmented care, polypharmacy and poor quality of life.

Under an integrated generalist‐led hospital model, ward patients are managed by a multidisciplinary care team led by a general internist or a specialist‐generalist, and they are admitted at one site, by one team, for the entire care duration. ¹⁴ Even on discharge, shared decision‐making is held with the patient to consolidate outpatient care under a ‘primary’ generalist physician. ¹⁵ Importantly, generalists are expected to manage a broad range of internal medicine conditions, minimise unnecessary referrals and avoid fragmentation of care that happens when single‐organ dysfunction is outsourced to individual subspecialties. ⁸ Unlike traditional models that mandate morning ward rounds and exit rounds with team seniors, this newer care system stratifies patients into acuity levels, where higher acuity patients get more frequent reviews by senior physicians, and lower acuity patients may be predominantly managed by junior staff or nurse clinicians on a day‐to‐day basis. ¹⁵

Addressing holistic care gaps in hospital care

General internists play a unique role in providing services that meet pre‐existing needs and gaps in hospital care. The role of GIM physicians in perioperative medicine is one of the best established, where they provide pre‐operative counselling and assessment of risks, optimise the patient perioperatively and help in multidisciplinary‐based management of post‐operative medical events. ¹⁶ In peripartum care, the involvement of general internists in managing metabolic disease and associated complications during pregnancy in high‐risk, multimorbid patients would be beneficial. ¹⁷ In the United States, vascular medicine, which was traditionally a surgical specialty, is an emerging area of subspecialty training after internal medicine to meet the growing needs of patients with complex vascular disease that warrants both medical and surgical interventions. ¹⁸

Over time, more areas are being identified for further training of GIM physicians based on clinical interests and identified gaps in existing clinical services. Unlike procedural subspecialties, GIM may have greater capacity to develop new clinical services based on local/institutional needs analyses. Examples may range from services for common ailments seen in general medical patients such as allergies and dermatological conditions, chronic pain and psychiatric illness, palliative care services in an ageing population with various forms of life‐limiting disease outside the oncological realm, and medical care for complex surgical and obstetric patients. To provide these clinical services competently, further training and accreditation for GIM trainees are typically required after completion of their residency programme. For instance, in places like the United States and Brazil, Allergy and Clinical Immunology is offered as a 2‐year training programme that can be undertaken after completion of internal medicine residency training. ¹⁹ Diploma programmes and other forms of professional accreditation for specific competencies in chronic pain management, mental health and palliative care are also increasingly available.

In places like the United Kingdom, dual specialty accreditation with AIM and another internal medicine subspecialty such as intensive care, infectious diseases or respiratory medicine is an official training pathway for AIM residents. ²⁰ Procedure skills‐specific credentialling such as in point‐of‐care ultrasound (POCUS) is particularly useful for general internists managing acutely ill patients. For example, POCUS is useful in evaluating patients in shock – both to diagnose the underlying mechanism/aetiology and guide/monitor response to therapy. ²¹ Moreover, POCUS is also increasingly adopted to guide bedside procedures such as thoracocentesis. ²¹ Recently, the consensus statement released from the Internal Medicine Society of Australia and New Zealand POCUS special interest group endorsed eight clinical scenarios for the use of POCUS by internists: fluid status assessment, undifferentiated shock, dyspnoea/respiratory failure, acute renal impairment, suspected venous thromboembolism, assessment of joint effusions and guidance in invasive procedures or adjustments to diuretic therapy in heart failure patients. ²²

Developing robust hospital‐at‐home programmes

The development of hospital‐at‐home (HaH) services has accelerated since the recent COVID‐19 pandemic. Patients admitted to HaH can continue to receive high‐quality care from the comfort of their homes, including remote vital signs monitoring, daily teleconsultation with physicians (with home visits for physical examination if deemed necessary), nursing support for caregiver training, blood‐taking and administration of intravenous treatments, allied health involvement and arranging transportation to hospitals for investigations such as diagnostic imaging if required. ²³ Successful implementation of HaH programmes helps to alleviate ED bed crunch and waiting times and reduce healthcare costs and is associated with positive patient outcomes ²⁴ and experiences. ²⁵

Implementation of HaH by GIM‐led services ²⁶ carries an advantage of robust service infrastructure with broad‐based clinical expertise to treat conditions ranging from straightforward ‘bread‐and‐butter’ general medical cases to cardiology, oncology, perioperative cases and even complex, ‘high‐risk’ patients such as haemopoietic transplant recipients. ²⁷ To optimise resource utilisation, hybrid models involving both in‐person visits and remote/virtual consultations can be implemented and tailored according to contextual/situational needs. ²⁸

Championing medical education, quality improvement and effective hospital administration

General medicine training is crucial in providing a solid internal medicine foundation for medical students, where they gain adequate exposure to ‘bread‐and‐butter’ cases and fulfil core entrustable professional activities. Given the breadth of undifferentiated diagnostic cases, clinical reasoning can be explicitly cultivated through having trainees form their impression with problem representation statements and semantic qualifiers, asking probing questions and modelling metacognitive practice. Moreover, real‐life contributors to suboptimal clinical decision‐making such as cognitive biases, ‘noise’ (unacceptable inter‐ or intra‐provider variability in clinical judgements) and over‐investigation could be explored. Importantly, as generalists are highly attuned to the management of hospital bed capacity and patient flow, they can educate the junior trainees on grasping the complexities of hospital bed management, prioritisation and development of pathways to expedite discharge, avoid admission or step‐down care.

Besides education, generalists tend to be heavily involved in quality improvement projects (QIPs), which are found to be the most common research undertaken by general internists in Australia and New Zealand. ²⁹ Indeed, generalist‐led QIPs have included endeavours to promote diagnostic excellence, ³⁰ reduce unnecessary testing, ³¹ enhance the quality of clinical documentations (e.g. discharge summaries) ³² and integrate artificial intelligence platforms in clinical care. ³³ In Japan, a Diagnostic Excellence Team led by general internists was formed to champion research work in clinical reasoning, propose best practices and policy changes to strengthen diagnostic accuracy, and host academic conferences for cross‐disciplinary knowledge sharing and cultivation of collective metacognition. ³⁴

Last, hospital administration is essential to the proper functioning of healthcare institutions and high‐quality patient care. For general internists, the cross‐disciplinary expertise and on‐the‐ground experience in operating typically the largest clinical service within a hospital puts them in good stead to understand the complexities and challenges in healthcare delivery. In the United States, GIM training explicitly prepares its trainee physicians for hospital administrative and leadership roles, including medical directorship.

Developing the value of general internal medicine through mindset change

General internists, like general practitioners in primary care, have traditionally been subject to specialty disrespect. For example, medical trainees may encounter seniors who view joining a generalist track to be a back‐up plan for those whose dreams of entering into subspecialty training do not materialise. ³⁵ Moreover, systemic factors such as lack of organisational priority for general medicine and permitting subspecialty teams to re‐direct admissions from the ED to general medicine further perpetuate its negative stereotypes. ⁶

Therefore, a mindset shift at the individual level and cultural change at the institutional/systemic level are necessary to accord the specialty the respect it deserves. Individual physicians should call out those who engage in specialty disrespect at all levels. Likewise, general internists must also champion their specialty and highlight its inherent strengths. Transformative leadership at healthcare institutions is necessary to properly recognise the scope and objective of general medicine services, acknowledge its value proposition and eliminate structural biases that exist against this specialty. With proper recognition of its roles in the hospital system, audits on the efficacy of differing models of practice of general medicine can then be performed to determine whether the services rendered are achieving their intended purpose.

Conclusion

In summary, GIM is at a crossroads, on the cusp of transitioning from a historically ambiguous, heterogeneous area of practice to solidifying its position as an integral hospital‐based specialty that provides clinical care to undifferentiated, complex and multimorbid patients and adapts its offered services based on existing care needs. Beyond clinical duties, general internists can spearhead innovative care models such as the HaH and digital health systems and engage in wide‐ranging quality improvement and educational endeavours.

Data availability statement

Data sharing is not applicable to this article as no datasets were generated or analysed during the current study.