Systematic review of post-COVID-19 syndrome rehabilitation guidelines

Abstract

Introduction

Post-COVID-19 syndrome is associated with significant health and potential socioeconomic burden. Due to its novel nature, there is a lack of clarity over best practice for the rehabilitation of patients with ongoing or new symptoms following acute COVID-19 infection. We conducted a systematic review of clinical and service guidelines for post-COVID-19 syndrome rehabilitation.

Methods

This review was registered on PROSPERO and is reported according to Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. We included guidelines formally published or endorsed by a recognised professional body, covering rehabilitation of people with symptoms following resolution of acute COVID-19 infection. We searched Medline, Embase, PsycINFO, CINAHL, Web of Science, NHS Evidence, MedRxiv, PsyArXiv and Google for terms related to COVID-19, rehabilitation and guideline. Two reviewers independently screened articles for inclusion, data extracted and quality assessed using the AGREE II and AGREE-REX tools for clinical guidelines and the AGREE-HS tool for service guidelines. We included guidelines of sufficient quality in a narrative synthesis.

Results

We identified 12 790 articles, of which 37 guidelines (19 clinical only, 7 service only and 11 combined clinical and service) were included. Guidelines covered a range of countries, rehabilitation types, populations and rehabilitation settings. Synthesis of clinical guidelines (n=4) was structured following the patient pathway, from identification, to assessment, treatment and discharge, with consideration of specific patient groups. Synthesis of service guidelines (n=8) was structured according to the Donabedian framework.

Discussion

Though the available post-COVID-19 syndrome rehabilitation guidelines were generally of poor quality, there was a high degree of consensus regarding the breadth of symptoms, the need for holistic assessment by a broad multidisciplinary team and person-centred care. There was less clarity on management options, measuring outcomes and discharge criteria.

PROSPERO registration number

CRD42021236049.

WHAT IS ALREADY KNOWN ON THIS TOPIC.

• Post-COVID-19 syndrome (commonly known as ‘long COVID’) describes ongoing or new symptoms related to COVID-19 more than 12 weeks postacute infection.

• Due to the novelty of the condition, there is lack of clarity surrounding best practice for the rehabilitation of these patients.

• Several clinical and service-level guidelines have been produced internationally to guide COVID-19 rehabilitation processes.

WHAT THIS STUDY ADDS

• This is the first systematic review of post-COVID-19 syndrome rehabilitation guidelines.

• We identify 37 guidelines from a range of countries, covering a range of types of rehabilitation.

• There is consensus regarding the assessment of patients and the need for individualised, multidisciplinary care, but a lack of clarity on onwards referral, discharge criteria and assessment of patient-level and service-level outcomes.

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

• Patients with post-COVID-19 syndrome should have access to rehabilitation services that are multidisciplinary, individualised and provide additional support for vulnerable groups.

• Services should be set up in a way that allows robust patient-level and service-level evaluation.

• More research and guidance are needed to give a clearer indication of management and follow-up of these patients, to identify and mitigate inequalities, and to address workforce challenges in providing care to patients, including health professionals themselves, with post-COVID-19 syndrome.

Introduction

In the month prior to 6 November 2022, an estimated 2.2 million people in the UK were experiencing symptoms persisting more than 4 weeks after suspected COVID-19 that were not explained by another condition (commonly known as ‘long COVID-19’), including 1.9 million people more than 12 weeks postacute infection (post-COVID-19 syndrome).¹ Many of these people were experiencing symptoms long after acute infection: 1.2 million and 600 000 people at least 1 year and 2 years after acute COVID-19 infection, respectively.¹ Three-quarters of people with self-reported long COVID-19 reported that symptoms were affecting their day-to-day functioning, and approximately one-fifth reported that they were severely affected.¹ The prevalence of long COVID-19 is highest in those of working age,¹ meaning that in addition to the health burden on the individuals, this syndrome has the potential to be associated with significant socioeconomic damage to the country. However, there is also a significant burden of disease in children and young people with approximately 70 000 individuals under the age of 17 self-reporting long COVID-19.¹

Due to the novel nature of COVID-19 infection, there is a lack of clarity over best practice for post-COVID-19 syndrome rehabilitation. We have systematically reviewed and summarised clinical and service guidelines for post-COVID-19 syndrome rehabilitation with the aim of informing clinical leads and commissioners what good practice looks like for the rehabilitation of patients affected by post-COVID-19 syndrome.

Methods

This systematic review was registered on PROSPERO and is reported according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses checklist.²

Eligibility criteria

Eligible guidelines were formal, written guidelines, published or endorsed by a recognised professional body or organisation. We included clinical and service guidance which reported on rehabilitation of people of any age who had recovered from acute COVID-19 infection but still had residual symptoms, or developed new symptoms related to COVID-19 following resolution of acute infection. We included guidance on rehabilitation interventions for physical and mental health needs.

We used the WHO definition of rehabilitation ‘a set of interventions designed to optimise functioning and reduce disability in individuals with health conditions in interaction with their environment’³; and the National Institute for Health Research definition of a guideline ‘a systematically developed statement for practitioners and participants about appropriate healthcare for specific clinical circumstances’.⁴

We excluded guidelines on rehabilitation of exclusively non-COVID-19 patients or those with acute COVID-19, and guidelines recommending exclusively complementary or complementary medicine interventions. We did not apply any language restrictions, but only used English search terms.

Search strategy

We developed our search strategy (online supplemental appendix 1) with a medical librarian. We conducted preliminary searches on 3 December 2020. On 5 February 2021, we searched for articles from 2020 onwards from: Medline, Embase, PsycINFO, CINAHL, Web of Science, MedRxiv and PsyArXiv; and from NHS Evidence and Google for grey literature. We used search terms relating to COVID-19, rehabilitation and guideline.

Deduplicated titles and abstracts of articles from the database searches were imported into the Rayyan QCRI webtool.⁵ Two reviewers independently screened titles and abstracts, retrieved full texts of potentially relevant articles, and assessed eligibility for inclusion. Articles not published in English were translated using Google Translate. For guidelines identified via the Google search, the website of the professional body or organisation was checked to identify the most up-to-date version.

Data extraction

Two reviewers independently performed data extraction using a preagreed extraction template which collected information on type of guideline (clinical level, service level or both), publication date, rehabilitation type, country, patient population and rehabilitation setting.

Quality assessment

Quality assessments were performed using the AGREE II⁶ and AGREE-REX⁷ tools for clinical guidelines, and the AGREE-HS tool⁸ for service guidelines. Sections of guidelines applicable to post-COVID-19 syndrome rehabilitation were assessed as a whole, rather than at the level of individual recommendations. Guidelines were scored independently by two reviewers, before final scores for each question were ascertained by consensus. Domain scores (for AGREE II and AGREE-REX), item scores (for AGREE-HS), and an overall score (for all three tools) were calculated as a percentage, accounting for the maximum and minimum scores possible.

Discrepancies arising during study selection, data extraction, and risk of bias assessment were resolved by discussion in the first instance and by a third reviewer where necessary.

Narrative synthesis

We undertook two narrative syntheses of the guidelines: clinical level and service level. We included only guidelines of a certain quality in the synthesis. In the absence of a recommended quality threshold,⁶ we identified a threshold of 60% as a commonly used threshold from published literature.^{9 10} Given that COVID-19 is a novel disease that has created an urgent health crisis, we expected available guidelines to have been produced in a rapid way and therefore unlikely to meet this threshold. We; therefore, applied a threshold of 40% for the overall quality score (AGREE II for clinical guidelines and AGREE-HS for service guidelines) for inclusion in the data synthesis. Some of the guidelines were written at both clinical and service level and were included in both syntheses if they met the quality threshold.

The guidelines were initially reviewed to determine themes, and then synthesis was performed according to the identified themes. For the clinical guidelines, the themes reflected a typical patient journey including: identification of suitable patients, setting/context of rehabilitation, investigation and assessment, management, evaluation of patient outcomes, and considerations for specific patient groups. For the service-level guidance, the themes reflected the Donabedian framework¹¹ encompassing structure, process and outcome.

Results

We identified 12 790 articles (figure 1). After removing 7122 duplicates from our database searches, 5518 unique records were screened for inclusion. Of the 201 full-text records reviewed, 14 studies met the inclusion criteria. Of the 150 articles identified from advanced Google searches, 23 met the inclusion criteria. In total, 37 guidelines were included.^12–50

Figure 1.

PRISMA 2020 flow diagram for new systematic reviews which included searches of databses, registers and other sources.

Guideline characteristics

The characteristics of the 37 guidelines are shown in online supplemental table 1, organised by guideline type—clinical only (n=19), service only (n=7) and combined clinical and service guidelines (n=11). Most of the guidelines were published in the summer of 2020, shortly after the first wave in Europe. We included guidelines from the UK (n=16), other European countries (n=8), China (n=4), Australia (n=3), New Zealand (n=1), the Americas (n=1), a European/USA joint task force (n=1) and global organisations (n=3). The guidelines covered a broad range of different types of rehabilitation and patient populations, including patients recovering from severe illness and those never hospitalised.

Guideline quality

Risk of bias is summarised in online supplemental table 2. Overall, the average quality of the clinical guidelines was poor (median 31%, IQR 28%–35%, range 14%–75%). The guidelines generally scored best on scope and purpose (median 64%, IQR 56%–72%) and clarity of presentation (median 56%, IQR 50%–71%). They scored most poorly on rigour of development (median 19%, IQR 13%–31%) and applicability (median 15%, IQR 9%–21%). Service guidelines scored slightly better (median overall score 37%, IQR 28%–40%, range 17%–63%), scoring best on defining the topic (median 67%, IQR 50%–67%) and scoring least well for methodology (median 17%, IQR 17%–33%). The NICE guideline³⁹ scored significantly higher than other clinical or service guidelines.

Narrative synthesis of clinical guidelines

Of the 30 clinical guidelines, 4 met the quality threshold for inclusion in the synthesis.20 21 33 39

Identifying patients in need of post-COVID-19 syndrome rehabilitation services

Symptoms

The guidelines describe post-COVID-19 syndrome as a multisystem disorder, with a wide range of symptoms that may change in nature or severity over time. While some symptoms are predictable from the nature of the acute disease (eg, postintensive care syndrome), others are more difficult to predict, especially in older people and children.

Case finding

There was little consensus between the guidelines about which patient cohorts should be screened for ongoing rehabilitation needs. Two guidelines^{21 39} recommended follow-up of all hospitalised COVID-19 patients around 6 weeks after discharge. This could be achieved using a screening tool, for example, the COVID-19 Yorkshire Rehab Screen,^{33 39} in conjunction with clinical assessment via phone or video consultation.³⁹ One guideline³⁹ additionally recommended active primary care follow-up of non-hospitalised COVID-19 patients in high-risk or vulnerable groups. One guideline³⁹ specified explicitly that a positive test during acute SARS-CoV-2 infection should not be required for access to rehabilitation services; and differentiated between ongoing symptomatic COVID-19 (4–12 weeks postacute illness) and post-COVID-19 syndrome (≥12 weeks).

Appropriate settings and contexts for post-COVID-19 syndrome rehabilitation

Infection prevention and control

There was broad consensus between the guidelines that local infection control protocols, for example, personal protective equipment, must be followed during rehabilitation. One guideline²¹ specifically highlighted that respiratory function and exercise capacity tests could be aerosol generating, and would need particular consideration. Another³³ linked to a checklist for reopening pulmonary rehabilitation services.

Modality and setting

There was consensus regarding the need for personalised care, including shared decision making regarding the modality and setting of rehabilitation. There was recognition that various settings could be appropriate for post-COVID-19 syndrome rehabilitation depending on the level of complexity, including: bedded rehabilitation units, community outpatient rehabilitation settings and home-based rehabilitation.^{20 21 33} There was also recognition that multiple rehabilitation modalities could be useful and should be tailored to the individual patient’s needs. These included: face-to-face consults, tele-rehabilitation (eg, telephone or video consultations) and self-management.^{33 39} One guideline³³ linked to specific guidance on delivering pulmonary rehabilitation via telehealth. It was noted that home-based and virtual rehabilitation modalities have the added benefit of minimising infection prevention and control concerns.³³

Investigation and assessment of patients with post-COVID-19 syndrome

Investigations

One guideline²¹ suggested that hospitalised patients should specifically be screened with assessment of respiratory function and exercise capacity at 6–8 weeks postdischarge. Another guideline³⁹ advised offering blood tests (eg, full blood count, renal function, liver function, C reactive protein, ferritin, B-type natriuretic peptide and thyroid function) and an exercise tolerance test (if appropriate) for all patients presenting for assessment, with further investigation (eg, chest X-ray, lying-standing BP, 3 min active stand test) according to symptoms. A third guideline²⁰ recommended investigations should be tailored to each patient’s specific symptoms (eg, if cardiac symptoms: cardiac blood panel, ECG, echocardiogram, cardiopulmonary exercise testing and/or cardiac MRI). It was noted that a plain chest X-ray may not be sufficient to rule out lung disease.³⁹

Assessment

There was consensus that patients with ongoing symptoms should have a holistic clinical consultation to identify ‘treatable traits’ amenable to rehabilitation,^{21 33} by completing a comprehensive assessment (including medical history, assessment of symptoms including mental health, functional assessment, quality of life, self-management skills and what the patient considers to be their main problem).³³ One guideline³³ provided a comprehensive list of assessment tools, organised by treatable trait (table 1).

Table 1.

Assessment tools by treatable trait, adapted from New South Wales Gov, ‘rehabilitation following COVID-19 in the pulmonary rehabilitation setting’³³

Treatable trait	Assessment tools
General fatigue	Fatigue Severity Scale
	FACIT-F
	PROMIS-29 (fatigue questions 13–16)
	Brief Fatigue Inventory
Exertional fatigue	Modified 0–10 Borg Rate of Perceived Exertion scale
Breathlessness	Modified 0–10 Borg Dyspnoea Scale
	Modified Medical Research Council Dyspnoea Scale
Exertional oxygen desaturation	Pulse oximetry
Decreased exercise capacity	6MWT
	Cardiopulmonary exercise testing
	30 s sit-to-stand (STS) test
	1 min STS test
	2 min walk test
	40 step test (not validated)
Muscle weakness	Functional Assessment: 5STS
	Muscle Strength Assessment: UK Medical Research Council Test Manual Muscle Test Isokinetic Muscle Testing Grip Strength
	Respiratory Muscle Strength: Maximum inspiratory pressure
Balance	Short Physical Performance Battery
	Berg Balance Scale
	Mini-BESTest
	Time Up and Go
Reduced physical activity	Accelerometers for example, ActivPAL, ACTi Graph
	Pedometers
	Personal fitness trackers, for example, Fitbit, Garmin, Apple Watch
	Mobile phones: apps and built-in GPS tracking systems
Reduced health-related quality of life	PROMIS-29
	SF-36
	EQ-5D
Sputum	Description of sputum – colour, amount, consistency, ease of expectoration, change from ‘normal’
Psychological disorders	Hospital anxiety and depression scale
	Depression, Anxiety and Stress Scale -21
Poor self-management skills	Patient Activation Measure
	Pulmonary Rehabilitation Adapted Index of Self- Efficacy Tool

EQ-5D, EuroQol five-dimension scale questionnaire; FACIT-F, The Functional Assessment of Chronic Illness Therapy – Fatigue; 6MWT, The six minute walking test; PROMIS-29, Patient Reported Outcome Measures Information System-29; SF-36, 36-Item Short Form Health Survey.

One guideline³⁹ warned that patients presenting with respiratory failure, severe lung disease, cardiac chest pain or a multisystem inflammatory syndrome may require prompt referral to acute care.

Management of patients with post-COVID-19 syndrome

Management pathways

A range of management options were presented, including in-house multidisciplinary rehabilitation,³⁹ self-management^{21 33 39} and onward referral to existing rehabilitation or other specialist services.20 21 33 39 Self-management should include aspects of patient education and goal setting.³⁹ This could be via an online platform (eg, Your COVID-19 Recovery)^{33 39} or through printed material (eg, Better Living with Exercise Booklet).³³ Common routes of onward referrals included: general rehabilitation services,³³ pulmonary rehabilitation,^{20 21 33} cardiac rehabilitation,^{20 33} neurological rehabilitation,³³ psychiatry,³⁹ rheumatology,³³ paediatrics³⁹ and allied health professional-led clinics such as speech and language therapy (SLT) or dietetics.³³

Patient education

Patient education was a common theme across all the guidelines. This should include: the most common post-COVID-19 syndrome symptoms, strategies to self-manage symptoms, when to contact a health professional,³⁹ and reassurance that symptoms like loss of taste and smell are likely to resolve with full recovery.²⁰ One guideline³⁹ included a patient information booklet and highlighted the importance of sharing care plan information with patients.

Monitoring and safety

There was broad consensus that patients require monitoring (eg, blood pressure, heart rate, oxygen saturation) during exercise to ensure patient safety.^{20 33 39} One guideline³³ provided a detailed table detailing what monitoring should be undertaken prior to, and during, exercise therapy of different modalities including face to face, video and telephone; and the exclusion and termination criteria for exercise-based therapy. Exclusion criteria were based on either observations (eg, resting HR >100 bpm, BP <90/60 mm Hg or >140/90 mm Hg, oxygen desaturation of ≥3% during exercise, temperature >37.2°C) or comorbidities.³³

Patient tailoring

There was consensus that rehabilitation should be individualised, holistic and include shared decision making. Factors to consider included patient preference,²¹ patient need,^{20 21} age, comorbidities, length of hospital stay and progress made following discharge from hospital.³³ It was recognised that returning to work would be a likely goal for some patients, and recommendations included tailoring physical, cognitive and functional assessments according to occupation,²⁰ receiving occupational therapy support³³ and supporting patients to discuss return to work or education, including adaptations such as phased return, with employers or educational settings.³⁹ Guidelines also recommended support to return to daily activities to improve function,²¹ access to support groups³⁹ and support with housing, employment, finances and social care³⁹ as appropriate.

Monitoring and outcomes

There was consensus on the requirement to review patient outcomes throughout the rehabilitation process, for example at baseline,^{21 33 39} 6–8 weeks posthospital discharge,²¹ and at the final assessment.³³ One guideline³³ recommended that outcome measures should reflect the individual’s rehabilitation focus (eg, reversing deconditioning, reducing fatigue, increasing strength, functional independence); and another recommended that the monitoring strategy should be selected with individual patients through shared decision making.³⁹ One guideline²¹ recommended the use of patient-reported outcome measures, and another³³ recommended collecting patient feedback on the rehabilitation programme. Recommendations for specific measurement tools included: functional assessments³⁹ (eg, the Post-COVID-19 Function Status Scale)³³; oxygen saturations and HR³⁹ and for post-acute respiratory distress syndrome patients, the EuroQol five-dimension scale questionnaire, Hospital Anxiety and Depression Scale and revised Impact of Events Scale.²¹

Considerations for specific patient groups

One guideline²¹ highlighted that disadvantaged communities often have poor access to rehabilitation.

One guideline²⁰ focused explicitly on ‘active populations’ and advised that patients with myocarditis required 3–6 months of complete rest prior to returning to high level sport or a physically demanding job. The period of rest depended on the severity and duration of the acute illness, left ventricular function and the amount of inflammation seen on cardiac MRI. People could return to training provided their left ventricular function, cardiac biomarkers and 24-hour ECG were normal, but they must be reassessed over a 2-year period.²⁰

Narrative synthesis of service guidelines

Of the 18 service guidelines, 8 met the minimum quality threshold for inclusion in the synthesis.12 27 28 38 39 44–46

Structure

Multidisciplinary team

There was broad consensus that the multidisciplinary team (MDT) should include consultant physicians (eg, rehabilitation medicine, respiratory medicine), therapists (including physiotherapy, occupational therapy, SLT), dieticians, psychologists and social workers.

Resources

Several guidelines concluded that existing capacity within rehabilitation services would be insufficient to meet the needs of the post-COVID-19 syndrome cohort, especially at the community level.12 27 39 45 The need for increased telerehabilitation capacity,²⁷ equitable distribution of specialised SLT equipment⁴⁴ and reconfiguration of mental health services⁴⁶ were highlighted as priority actions in guidelines focussing on telerehabilitation, SLT and mental health, respectively. However, no guideline quantified the funding required to bridge the perceived resource gap.

Setting

There was broad recognition that rehabilitation services are required to support people living at home (including those never hospitalised and those discharged home), accessing primary and community-level care, as well as those in bedded rehabilitation services. There was particular focus on the need for these tiers of rehabilitation to be well integrated and flexible, and movement between them to be seamless and promote continuity of care. ‘Rehabilitation Prescriptions’, in which the rehabilitation needs and preferences of the patient are documented, were recommended as a mechanism to achieve this.^{12 39}

Process

Inward referrals

Two mechanisms for identifying patients who should be referred into rehabilitation services were described.

• Symptom screening (eg, the COVID-19 Yorkshire Rehabilitation Screen) applied at routine follow-up around 6 weeks postdischarge for all patients hospitalised with COVID-19.

• Patients presenting to primary care, where the GP assesses that the patient has symptoms consistent with post-COVID-19 syndrome severe enough to require specialist support.

Mode of delivery

Owing to the often complex symptomatology of post-COVID syndrome, several guidelines^{39 44 46} recommended that rehabilitation services operate as ‘one-stop’ services in which all members of the MDT are available during the care episode. There was complete consensus between the guidelines that the symptom screen for postdischarge patients should take place remotely either via a telephone or a video call; and that telerehabilitation options should be incorporated into each part of the rehabilitation pathway, provided it is appropriate for the person. Several guidelines28 38 39 44–46 also recommended the use of self-management tools, for example, Your COVID-19 Recovery, provided this is deemed appropriate following clinical assessment of the patient.

Onward referrals

The mechanism for discharging patients, or making onward referrals, was not always clear. Some guidelines^{39 44} favoured services acting primarily as assessment clinics, with initial treatment and advice provided, before onward referral to established clinics as appropriate. On the other hand, one guideline⁴⁶ recommended a ‘no-discharge policy’ in order to support patient confidence, which the authors argued could paradoxically lead to decreased service utilisation.

Equity

Many of the guidelines discussed the need to consider digitally excluded populations when adopting virtual modes of rehabilitation delivery,27 38 39 45 for example, by using a telephone-based approach for those without internet access,²⁷ and making available a paper workbook which replicated the content of the online platform.³⁸

Individual guidelines also recommended the provision of additional support (eg, accessible formats and longer appointments) to vulnerable groups such as: those with learning difficulties and/or autism,³⁸ older people, children and those for whom English is not their first language,³⁹ and issues such as addictions, offending, homelessness and domestic violence.⁴⁵

Outcome

Auditing and monitoring criteria

Recommendations on service outcome measures were not always included, and there was little consensus between the guidelines on how to measure patient outcomes. One guideline³⁹ included an evaluation tool to assist services in assessing their implementation of the guideline recommendations, which the authors suggested could be repeated at regular intervals to monitor progress. One guideline⁴⁶ recommended a series of evaluation metrics comprising a mixture of service and patient outcome measures. Two guidelines^{38 39} recommended the use of self-monitored symptom trackers which the patient should fill in themselves remotely, for example, using a smartphone app. Two guidelines^{12 44} recommended minimum clinical datasets which should be completed, but these tended towards capturing episodes of acute care, rather than rehabilitation outcomes.

Data governance

One guideline³⁸ reported that data captured by an online platform should be encrypted to AES-256 standard.

Discussion

Key findings

This systematic review summarises clinical and service guidelines, published up to February 2021, for the rehabilitation of people affected by post-COVID-19 syndrome. The systematic review summarises the recommendations made my professional bodies soon after the emergence of post-COVID-19 syndrome and is not intended to represent guidance in itself. Of the 37 guidelines identified, 420 21 33 39 clinical and 812 27 28 38 39 44–46 service guidelines met the quality threshold and were included in our synthesis. Generally, guidelines were of poor quality, scoring particularly poorly on rigour of development, implementability and applicability to local settings.

The key areas of consensus between the guidelines included: the broad range of post-COVID-19 syndrome symptoms, the need for appropriate infection prevention and control procedures, the need for personalised care and shared decision making, the need for flexible rehabilitation across multiple settings (including inpatient and outpatient settings) with good integration between tiers of the system, the need for a hybrid model which incorporates face-to-face and virtual interactions as appropriate, the need for holistic assessment of patient needs to identify ‘treatable traits’ amenable to meaningful rehabilitation outcomes, the incorporation of self-management options and patient education, the need for adequate monitoring during exercise to ensure patient safety, the need for a broad MDT, the need for referral routes from primary and secondary care, and the need to provide additional support to specific patient groups.

In contrast, we found a lack of consensus as to whether case finding approaches should be applied in the identification of patients, and about which initial investigations should be performed. There was some consensus that clinics should establish themselves as ‘one stop’ clinics with access to all necessary parts of the MDT in one visit, but there was less clarity on what the treatment should consist of, though most guidelines tended towards a view that adaptation of existing treatments was appropriate to the post-COVID-19 syndrome patient (eg, pulmonary rehabilitation or cardiac rehabilitation programmes). However, there was a lack of consensus about how much treatment should be provided within the service or whether the focus should be on onward referral to existing external services. It was clear that patient outcomes should be regularly monitored during treatment, but even the best quality included guidelines lacked detail on how this should be done in practice. No guidelines provided discharge criteria, but one guideline suggested adopting a ‘no discharge policy’ to support patient confidence.⁴⁶ Finally, while several guidelines called for increased resources to meet the scale of post-COVID-19 syndrome, there was minimal detail on what form that resource should take and how much it would cost.

Limitations of the evidence base

The primary limitation of the evidence base is the overall quality of the available guidelines. The guidelines also scored poorly on implementability and adaptability, which is further evidence of the immaturity of the body of literature.

Although three guidelines^{38 39 46} mentioned involvement of those with lived experience of post-COVID-19 syndrome in the formulation of their guidance, the majority of the guidelines appeared to have been created by professionals only. It has been claimed that long COVID-19 is the first illness to have been created following patients locating each other on Twitter,⁵¹ and the existence of prolonged symptoms related to COVID-19 infection was generally recognised by patients before it was accepted by clinicians and professional bodies such as the WHO.⁵¹ There is a strong argument for the ‘cocreation’ of guidance,⁵² which could be achieved through the addition of people with lived experience of post-COVID-19 syndrome onto stakeholder panels. However, the evidence base highlights that this is currently an exception rather than the rule.

A further limitation is the lack of available guidelines for groups recognised in the NHS England national commissioning guidance for post COVID-19 services as needing specialist services or management, such as children and young people and people in employment.⁵³

A final limitation is the lack of available guidelines from low-income and middle-income countries that have been particularly hard hit by the pandemic, for example, India and Brazil. It is not clear how applicable the included guidelines, which are predominantly from high-income countries, are to these settings.

Strengths and limitations of our study

The strengths of this study are that this is the first systematic review of post-COVID-19 syndrome rehabilitation guidelines, and we were able to include a good mix of both clinical and service guidelines. The main limitation of our pragmatic approach is that evidence on COVID-19 is accumulating at an unprecedented rate, and synthesising guidelines (which are themselves syntheses of primary evidence) means that we are one step behind the most recent data. Most of the included guidelines were produced early in the pandemic, so important aspects of the condition which are now better understood including the pathogenesis of post-COVID-19 syndrome, the occupational impacts and required adjustments in the workplace, and the requirements for specialist services for children and young people are not well covered. Also, several of the included guidelines referenced the need for them to be updated as new evidence emerges, so a repeated review may be warranted.

Implications for practice, policy and research

While academia catches up, health systems are having to face up to the immediate reality of large numbers of patients requiring support. In England, this need has been met by the NHS in the form of post-COVID-19 syndrome assessment clinics, which accept referrals from primary and secondary care from 4 weeks postacute infection, assess patients and provide initial treatment advice, before referring patients to self-management or onwards to existing rehabilitation or specialist services as appropriate.⁵³ The service specification of those assessment clinics is largely supported by the findings of the review. But the elephant in the room of this approach is that these clinics represent only the ‘front door’ for patients, who are then referred on to existing services that are stretched and have long waiting lists.

Some guidelines mention the need for services to take steps to ensure equity of access (through active case finding) or equity of treatment (eg, with longer clinic appointments for certain groups of patients). It is important that any inequalities created by these services are quantified, and steps taken to mitigate these inequalities evaluated, with learning disseminated. Equally, with the understandable emphasis on self-management and tele-rehabilitation approaches to match the scale of the crisis, services must take steps to avoid digital exclusion and ensure that appropriate safety-netting is in place on these pathways.

There was a lack of occupational health guidance for those affected by post-COVID-19 syndrome. Urgent research and policy action is needed to address this given the disproportionate burden of post-COVID-19 syndrome on the working population.¹ Research has shown that individuals with post-COVID-19 syndrome experience poor treatment in the workplace including disbelief about their symptoms, discrimination and even job loss,⁵⁴ highlighting the need for timely and tailored rehabilitation interventions to minimise ill health and maximise job retention.

Finally, there was little detail on the workforce challenge involved in providing high-quality rehabilitation services at the scale required. This workforce is in many cases the same groups who have worked tirelessly to provide acute care during the pandemic and need adequate rest and recuperation to ensure their own health does not suffer. In addition, many allied health professional groups have longstanding problems with high vacancy rates. Healthcare workers themselves also have a high prevalence of post-COVID-19 syndrome,¹ further compounding the staffing crisis. More money does not necessarily equate to recruitment of more staff if those extra out-of-work staff do not exist. System-level resource planning must be sufficient to meet the scale of challenge of post-COVID-19 syndrome but must also be realistic and deliverable.

Ethics statements

Patient consent for publication

Not applicable.