Impact of the COVID-19 pandemic on people in need of care or support: protocol for a SARS-CoV-2 registry

Abstract

Introduction

People in need of care or support are severely affected by the COVID-19 pandemic. We lack valid data of long-term assessments. We present a register study to detect the physical and psychosocial impact of the COVID-19 pandemic on people in need of care or support in Bavaria, Germany. To describe the persons’ life conditions comprehensively, we assess the perspectives and needs of the respective care teams too. Results will serve as evidence-based source to manage the pandemic and long-term prevention strategies.

Methods and analysis

The ‘Bavarian ambulatory COVID-19 Monitor’ is a multicentre registry including a purposive sample of up to 1000 patient–participants across three study sites in Bavaria. The study group consists of 600 people in need of care with a positive SARS-CoV-2 PCR test. Control group 1 comprises 200 people in need of care with a negative SARS-CoV-2 PCR test, while control group 2 comprises 200 people with a positive SARS-CoV-2 PCR test but are not in need of care. We assess the clinical course of infection, psychosocial aspects and care needs using validated measures. Follow-up is every 6 months for up to 3 years. Additionally, we assess up to 400 people linked to these patient–participants (caregivers, general practitioners (GPs)) for their health and needs. Main analyses are stratified by level of care I–V (I=minor/V=most severe impairment of independence), inpatient/outpatient care setting, sex and age. We use descriptive and inferential statistics to analyse cross-sectional data and changes over time. In qualitative interviews with 60 stakeholders (people in need of care, caregivers, GPs, politicians), we explore interface problems of different functional logics, of everyday and professional perspectives.

Ethics and dissemination

The Institutional Review Board of the University Hospital LMU Munich (#20-860) and the study sites (Universities of Wurzburg and Erlangen) approved the protocol. We disseminate the results by peer-reviewed publications, international conferences, governmental reports, etc.

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• This large, multicentre registry fills an evidence gap in COVID-19 research by focusing on a vulnerable, under-represented group of people, who are in need of care in ambulatory settings (long-term care facilities, informal/family care, etc) and survived COVID-19 infection.

• A 36-month follow-up provides data on long-term clinical course and sequels.

• A multiprofessional research team (ie, general practice, nursing, sociology and infectology) and a triangulated research approach combining quantitative and qualitative methods provide multiple perspectives and comprehensive analyses.

• Prestatus (before the pandemic) of the study population is not available; patient-reported outcomes and interviews are at risk for recall bias and social desirability.

• Due to limited life expectancy of the predominantly frail participants, we include additional participants over time (open registry).

Introduction

By February 2022, the WHO noted 433 million cases and almost six million deaths from COVID-19. Many COVID-19 survivors will be affected by long-lasting and debilitating sequelae.1 One of the most vulnerable and affected groups by the COVID-19 pandemic are people in need of care or support, including older and frail people.2–6 In Germany, until end of 2022, people aged 60 years and older accounted for more than two-thirds (71.4%) of COVID-19 cases in old people’s and nursing homes,7 and in the first year of the pandemic, the majority of the COVID-19 deaths involved people under care of an outpatient care service or living in a long-term care facility.8

Apart from a higher risk of death from COVID-19 for the older, the physical, psychological and social impact of the COVID-19 pandemic and the subsequent needs of COVID-19 survivors may differ between younger and frail older survivors of COVID-19 for a number of reasons. Multimorbidity is the rule rather than an exception among older people and COVID-19 may exacerbate both general frailty and specific comorbidities that are particularly affected by COVID-19 (such as respiratory and cardiovascular disease).2 9 All of these factors may prolong recovery, increase the likelihood of long-COVID/post-COVID syndrome and increase dependency.10 11 In addition, long-term care facilities have frequently implemented drastic infection control measures, with external and internal contact restrictions aggravating feelings of loneliness and isolation among their residents (irrespective of age or frailty status), which may have long-term consequences for mental and physical health.12–15

The additional care needs of often frail COVID-19 survivors also placed a further burden on formal and on informal caregivers, who had already been physically and psychologically challenged by staff shortages, fear of infection and frequent encounters with death.11 16 17 Furthermore, the pandemic was also a disruption to the provision of routine primary care—for example, in Germany, general practitioners (GPs) cared for 90% of COVID-19 patients.18 19

Against this background, it appears likely that the COVID-19 pandemic has and will continue to have a relevant impact on the physical and psychosocial health of people in need of care or support as well as on those caring for them, including formal (eg, nursing stuff) and informal (eg, family members /relatives) caregivers and GPs.

Aim and objectives

The aim of the ‘Bavarian ambulatory COVID-19 Monitor (BaCoM)’ is, therefore, to conduct a systematic assessment of the physical, psychological and social long-term outcomes and sequels of the COVID-19 pandemic on people in need of care or support as well as their care needs and the needs of care providers. The findings should support the development and implementation of long-term prevention and aftercare strategies. The specific study objectives are:

1. To examine clinical parameters, psychosocial burden and care needs in people dependent on care or support.

2. To examine long-term sequels in people in need of care or support.

3. To examine the needs of formal and informal caregivers

Methods and analysis

Study design and setting

BaCoM is a multicentre, open registry study in the State of Bavaria (Southern Germany). For objectives 1 and 2, we include patient-participants in one study group (SG) and two control groups (CG1 and CG2). The SG comprises people with evidence of a previous SARS-CoV-2 infection, who were in need of care or support at the time of infection and survived COVID-19. In order to examine the impact of COVID-19 on clinical parameters and psychosocial burden, participants in CG1 comprise people in need of care or support during the COVID-19 pandemic but without evidence of SARS-CoV-2 infection. In order to examine effect modification of the COVID-19 impact by need of care or support, participants in CG2 comprise people with evidence of a previous SARS-CoV-2 infection who were not in need of care or support at the time of infection. For objective 3 we also collect information from formal and informal caregivers of participants included in SG, CG1 and CG2 as well as their GPs.

Study population

Eligibility criteria for patients

The inclusion/exclusion criteria of the SGs and CGs are provided in table 1. All adult residents of the State of Bavaria who are 18 years or older at the time of recruitment and have had at least one SARS-CoV-2 test are eligible for inclusion in BaCoM. In order to determine COVID-19 status (for assignment to SG and CG2 vs CG1), we consider the results of PCR tests, where people with at least one previous positive test result (not older than 1 March 2020) are assigned to SG or CG2, respectively, and otherwise to CG1. In cases, where PCR test results are not available (which is commonly the case in later stages of the pandemic), people with rapid SARS-CoV-2 antigen test results (not older than 6 months) can be enrolled. Test results from antigen tests are interpreted in combination with any evidence of nucleocapsid antibodies measured as part of the study protocol (see the Data collection section). The assessment of nucleocapsid antibodies serves as a further means to verify any previous infection with SARS-CoV-2, which is not influenced by exposure to vaccines (since vaccines only trigger antibodies against the spike protein). Patients, who have previously been allocated to CG1, but who subsequently test positive for nucleocapsid antibodies, are classified as group-switchers and are reallocated to the SG accordingly.

Table 1.

Inclusion/exclusion criteria of patient-participants

Study group	Control group 1	Control group 2
Inclusion criteria:
Signed informed consent from the participant or a legal guardian Age≥18 years Sufficient knowledge of German to give consent/answer questionnaires or possibility of translation by an interpreter residence in Bavaria
Existing need for care (care level I–V*) or support (according to the clinical judgement of the recruiting doctor: current need for care or expected need in the near future (Clinical Frailty Scale≥5 and <9)21 22 Positive SARS-CoV-2 PCR test (maximum backdated to 1 March 2020)	Existing need for care (care level I–V*) or support (according to the clinical judgement of the recruiting doctor: current need for care or expected need in the near future (Clinical Frailty Scale≥5 and <9) Negative SARS-CoV-2 PCR test (maximum backdated to 1 March 2020) with respiratory infection	No existing need for care (no care level I–V*) or support (according to the clinical judgement of the recruiting doctor: no current need for care or expected need in the near future (Clinical Frailty Scale<5)) Positive SARS-CoV-2 PCR test (maximum backdated to 1 March 2020)
Exclusion criteria:
Refugees/asylum seekers Life expectancy<6 months (clinical judgement of the recruiting doctor) Persons without health insurance

The levels of care I–V are:

Care level I: minor impairment of independence.

Care level II: significant impairment of independence.

Care level III: severe impairment of independence.

Care level IV: most severe impairment of independence.

Care level V: most severe impairment of independence with special requirements for nursing care.

*The degree of independence of the person in need of care is decisive for classification into the care levels.

Eligibility criteria for participating caregivers

Formal and informal caregivers are eligible for recruitment, if they are involved in the care or support of a recruited patient-participant.

Eligibility criteria for recruiting GPs

GPs are eligible for recruitment, if (a) they offer statutory health insurance service, (b) they care for COVID-19 patients, (c) they offer a primary healthcare service open for all patient groups.

Participant recruitment

Up to 1000 patient-participants (n=600 in SG and 200 in each of CG1 and CG2) are recruited at three study sites in Bavaria (Munich, Erlangen and Wurzburg). In addition, we recruit up to 200 formal caregivers, up to 100 informal caregivers and up to 100 GPs. Patient-participants can be recruited in inpatient (long-term care facilities) or outpatient care settings (home care provided by informal caregivers and/or outpatient care services). In order to maximise the geographical spread of study participants, we implement a Bavarian-wide recruitment campaign with broad publicity. Recruitment of patient-participants can take place at any time after a SARS-CoV-2 PCR test result (subsequently referred to as the ‘index test’). The index test is defined as the first positive SARS-CoV-2 test result (for SG and CG2) or otherwise the latest negative SARS-CoV-2 test result without any previous infection (for CG 1). Patient-participants are identified via their GP, the long-term care facility they live in, via outpatient care services or informal caregivers, or via self-referral. Irrespective of how prospective patient-participants are identified, they are either recruited by their GP or (if not available) a study physician.

The recruitment of GPs is carried out via in total 240 GCP-qualified practices of the Bavarian Research Practice Network and cooperating teaching practices. Further eligible general practices with a past or current focus on managing patients with COVID-19 are identified.

The participating GP’s receive compensation for their work within the framework of the study (participant inclusion and information, baseline examination, follow-up (FU) surveys). For the recruitment from inpatient and outpatient care facilities, we use a list of about 700 eligible facilities in Bavaria with documented COVID-19 outbreaks who file their interest in participating in BaCoM via a reporting system to the Bavarian State Office for Health and Food Safety (Bayerisches Landesamt für Gesundheit und Lebensmittelsicherheit.

Data collection

Baseline data collection began on 1 March 2021 and enrolment will continue until the end of 2023. Data collection methods include blood samples, clinical investigations, data abstraction from clinical data sources, surveys as well as semistructured interviews (see tables 2–5). Appropriately trained study staff conduct all data collection, so that inter-rater reliability for all assessments (including questionnaires and clinical tests) can be ensured. It can be assumed that a certain proportion of the study participants will not be able to provide self-disclosure (eg, in case of cognitive impairment). In these cases, the information collection is to be ensured by relatives or caregivers who are asked instead. In order to ensure that the cognitive status is determined, the ‘six-item screener’20—a cognitive short test—is administered (0–6 points). If the ‘six-item screener’ is not successfully completed (<4 points), the information collection of the self-reports will be ensured according to the substitution principle mentioned above.

Table 2.

Schedule of enrolment and assessments in BaCoM: study group and control groups

Construct	Timepoint	Baseline	FU1–FUx
	Instrument	Enrolment	6 months after PCR test
Sociodemographic information*†	Questionnaire
Age, sex, migration background, educational level, professional life/activity, income, marital status, health insurance, insurance number		x
Care-specific parameters with reference to the need for care†*	Questionnaire
Care level (I–V)/frailty level and onset date of the need for care/frailty	Clinical Frailty Scale21 22	x	x
Diagnosis justifying care		x	x
Capacity for giving consent/legal guardian		x	x
Care setting (outpatient/inpatient care)		x	x
Change of care level/frailty level and care setting since start of the pandemic		x	x
Provision of aids/therapies (Which are needed? Which ones are not supplied/prescribed due to the pandemic, etc? Which individual solution is used?)		x	x
Measures involving deprivation of freedom (before pandemic/during pandemic)		x	x
Pressure ulcer (before pandemic/during pandemic)		x	x
respiration		x	x
Self-rated pain levels	Rating scale1–10	x	x
Activities of daily living (ADL)	Barthel Index35	x	x
Nursing diagnoses in terms of NANDA-I (Definitions and Classification 2018–2020)34 *†	Questionnaire
Stability of respiratory parameters38	Likert Scale
Impaired gas exchange		x	x
Impaired spontaneous breathing		x	x
Ineffective airway clearance		x	x
Sense of smell and taste	Likert Scale
Smells all primary odours		x	x
Tasting all substance spectra		x	x
Social interaction34 39	Likert Scale
Mobility, ability to walk		x	x
Communication verbal/via electronics   (impaired/reports needs)		x	x
Family processes (before and since the beginning of the pandemic)	Likert Scale
Continuous family processes (before and since the beginning of the pandemic)		x	x
Interrupted family processes (before and since the beginning of the pandemic)		x	x
Social isolation34 39	Likert Scale
Feeling of being alone (during the COVID-19 pandemic, as imposed by others)		x	x
Feeling of being alone (before and since the beginning of the pandemic)		x	x
Can explain current situation of the pandemic		x	x
Can place current challenges in the context of the pandemic		x	x
Can get help to cope with current life situation		x	x
Can cope with tasks and challenges themselves		x	x
Feeling of powerlessness/helplessness		x	x
Physical health status†
Height, weight, body mass index, smoking status	Questionnaire	x	x
Blood sampling: laboratory parameters, serostatus survey, long-term storage24 40	Measurement	x	x
Vital parameters (blood pressure, pulse, breath rate, O2-saturation, body temperature)	Measurement	x	x
Identification of patients at risk from sepsis	Quick sepsis-related organ failure assessment (qSofa)41 42	x	x
Measurement of pulmonary function  PEF, FEV1, FVC, FEV0.75, FEV0.5, FEV1/FVC, FEF75, FEF25-75, FEF50, FV-curve	Spirometry (mySpiroSense, mobile spirometer)	x	x
Cognitive short test	Six-Item-Screener20	x	x
Cognitive Impairment	MoCA-BLIND23	x	x
Mobility	Timed-Up&Go Test43	x	x
Medication	Medication list	x	x
Diagnosis	Diagnosis list	x	x
Vaccination status (COVID-19/influenza/pneumococcus)	Vaccination certificate	x	x
Characteristics of a COVID-19 disease course (Study+Control group 2)/Characteristics of the course of the respiratory infection (control group 1)*†‡	Questionnaire
Date of positive/negative: SARS-CoV-2 PCR test / rapid SARS-CoV-2 antigen test + nucleocapsid antibodies		x	x
Symptomatic/asymptomatic infection		x	x
Time of onset of COVID-19 symptoms/symptoms of the respiratory infection
COVID-19 symptoms/symptoms of the respiratory infection24 §		x	x
Duration of symptoms (long-COVID/post-COVID)¶		x	x
COVID-19 specific clinical complications/complications of the respiratory infection (diagnoses)24		x	x
New medication since PCR Test		x	x
Mortality (time/cause of death, autopsy findings)
Use of medical care facilities (since PCR test)*†‡	Questionnaire
Inpatient medical care (days/diagnosis): Hospitalisation (with intensive care unit)/rehabilitation/ psychatry  Number of general practitioner/other specialists contacts		x	x
Ambulatory medical care (days/diagnosis): Number of general practitioner/other specialists contacts /treatment in emergency rooms		x	x
Psychosocial health status*	Questionnaire
Health-related generic quality of life	EQ-5D-5L and EQ-VAS25–27	x	x
Depressiveness	PHQ-9(29)	x	x
Post-traumatic stress disorder (PTSD)	Impact of Event Scale (IES-R)31	x	x
Anxiety	GAD-7(30)	x	x
Health literacy	HLS-EU-Q1633	x	x
Coping/self-management/self-efficacy	SES6G32	x	x
Healthcare utilisation	Claims data**
Medical diagnoses	ICD-10 codes	x	x
Planned and emergency hospital admissions	ICD-10 codes	x	x
Medication dispensed by community pharmacies	ATC codes	x	x
Level of care applications/assessments	Standardised assessment templates	x	x

*Survey of the person in need of care or support.

†Survey by medical/nursing staff, (caring) relatives.

‡Patient record/care record.

§Main symptoms according to the National Research Network of University Medicine on COVID-19: German Corona Consensus Data Set24: Disturbance of the sense of smell and/or taste, abdominal pain, disturbance of consciousness/confusion, diarrhoea, vomiting, cough, shortness of breath (dyspnoea), nausea, fever, headache, fatigue, etc.

¶Long-COVID syndrome is defined as health complaints that persist beyond the acute illness phase of a SARS-CoV-2 infection of 4 weeks or are new. Post-COVID syndrome refers to symptoms that persist for more than 12 weeks after the onset of SARS-CoV-2 infection and cannot be explained otherwise.44 45

**Data are provided by the statutory health insurance of study participants and linked nursing care assessment services (Medizinischer Dienst). Data linkage is provided by a dedicated trust centre.

BaCoM, Bavarian ambulatory COVID-19 Monitor; EQ-5D-5L, EuroQol; EQ-VAS, EQ visual analogue scale; FEF, Forced expiratory flow; FEV, Forced expiratory volume; FU, follow-up; FVC, Forced vital capacity; FV-curve, Flow volume-curve; GAD-7, Generalised Anxiety Disorder Screener; HLS-EU-Q16, European Health Literacy Survey; NANDA, North American Nursing Diagnosis Association; PEF, Peak expiratory Flow; PHQ-9, Patient Health Questionnaire; SES6G, Six-item Self-Efficacy Scale.

Table 3.

Schedule of enrolment and assessments in BaCoM: formal caregivers and care facilities (inpatient/outpatient)

Construct	Timepoint	Baseline	FU1–FUx
	Instrument	Enrolment	6 months
Sociodemographic information	Questionnaire
Age, sex, ethnicity, migration background, educational level, professional life/activity, income, marital status		x
SARS-CoV-2 infection	Questionnaire
SARS-CoV-2 infection in the past		x	x
Date of positive SARS-CoV-2 PCR tests		x	x
COVID-19-Infection symptomatic/asymptomatic		x	x
Care facility parameters	Questionnaire
Inpatient care facilities and other forms of housing (provider: non-profit/private/public; group of persons: older people/disabled/mentally ill/palliative; organisation: long-term/short-term/day/night care; number of beds; nursing ratio; staffing: specialist ratio/qualification/employment ratio/case numbers COVID-19/vaccination ratio/visit management/workload)		x	x
Outpatient care facilities (provider: non-profit/private/public; group of persons: older people/disabled/psychologically ill/palliative; care performance; care ratio; staffing: skilled worker ratio/qualification/employment ratio/case numbers COVID-19/vaccination rate/visit management/workload)		x	x
SARS-CoV-2 vaccination	Questionnaire
Psychological factors influencing the decision to vaccinate against SARS-CoV-2	5 C46	x	x
Psychosocial health status	Questionnaire
Depressiveness	PHQ-929	x	x
Burnout	Maslach Burnout Inventory (MBI)36	x	x

BaCoM, Bavarian ambulatory COVID-19 Monitor; FU, follow-up; PHQ-9, Patient Health Questionnaire.

Table 4.

Schedule of enrolment and assessments in BaCoM: informal/family caregivers

Construct	Timepoint	Baseline	FU1–FUx
	Instrument	Enrolment	6 months
Sociodemographic information	Questionnaire
Age, sex, ethnicity, migration background, educational level, professional life/activity, income, marital status		x
SARS-CoV-2 infection	Questionnaire
SARS-CoV-2 infection in the past		x	x
Date of positive SARS-CoV-2 PCR tests		x	x
COVID-19-Infection symptomatic/asymptomatic		x	x
Care burden situation	Questionnaire
Duration and onset of informal/family care		x	x
Support through outpatient care service		x	x
Support through care allowance		x	x
Use of other support services		x	x
Burden Scale for Family caregivers  Caregivers	Häusliche Pflegeskala (HPS)	x	x
SARS-CoV-2 vaccination	Questionnaire
Psychological factors influencing the decision to vaccinate against SARS-CoV-2	5 C46	x	x
Psychosocial health status	Questionnaire
Depressiveness	PHQ-9(29)	x	x
Burnout	Maslach Burnout Inventory (MBI)36	x	x

BaCoM, Bavarian ambulatory COVID-19 Monitor; FU, follow-up; PHQ-9, Patient Health Questionnaire.

Table 5.

Schedule of enrolment and assessments in BaCoM: GPs and practices

Construct	Timepoint	Baseline	FU1–FUx
	Instrument	Enrolment	6 months
Sociodemographic information	Questionnaire
Age, sex, professional experience		x
Practice-specific parameters	Questionnaire
Single/joint practice		x	x
Number of GP’s, number of medical assistants		x	x
Number of patients per quarter		x	x
Use of other support services		x	x
Number of COVID-19 patients per quarter		x	x
Number of deceased COVID-19 patients per quarter		x	x
SARS-CoV-2 vaccination	Questionnaire
Psychological factors influencing the decision to vaccinate against SARS-CoV-2	5 C46	x	x

BaCoM, Bavarian ambulatory COVID-19 Monitor; FU, follow-up; GPs, general practitioners.

Baseline and FU assessments

The baseline data collection occurs within 4 weeks after recruitment. For all patient-participants, FU assessments are conducted at 6-month intervals after the date of the index test for a period of up to 3 years in order to be able to observe the development of physical and mental health as well as provider and care needs over an extended time period. Depending on the date of enrolment, the number of FU’s will, therefore, range between one and five FUs (3 years). For formal and informal caregivers and GPs, FU’s intervals are similar. The parameters and constructs of interest as well as their corresponding data collection methods are provided in tables 2–5 and their rationale is briefly described below. The qualitative interviews with stakeholder are conducted only once per participant.

Clinical parameters of patientparticipants

Table 2 shows that clinical parameters of interest include physical status, laboratory, medication, comorbidities, body mass index and vaccination status. Tests for frailty (Clinical Frailty Scale21 22) and for cognitive impairment (MoCA-Blind23) are performed. In addition, the use of healthcare facilities, COVID-19-specific complications and symptoms (long-Covid/post-Covid) are used to characterise the course of the disease. Many target variables are collected in accordance with the ‘German Corona Consensus Data Set’ of the National Research Network of University Medicine on COVID-19.24 The attending GP or the study nurse performs a brief physical examination with measurement of the vital parameters and takes a venous blood sample. By this, an antibody test for SARS-CoV-2 and, thus, the influence of COVID-19 disease on the immune response can be measured. For the remaining part of the blood samples, an immediate laboratory analysis of a complete blood count, a differential blood count and 26 organ-specific parameters relevant for COVID-19 disease are carried out. To cover future research questions, serum and whole blood samples are transferred to quality-controlled long-term storage at −80°C in the Institute of Laboratory Medicine.

Psychosocial parameters of patient participants

Table 2 shows psychological parameters of interest that include the mental health status among others. The COVID-19 pandemic may cause severe psychosocial stress among people in need of care or support at different ages and life situations. To recognise these burdens and identify possible protective factors or risk factors, participant questionnaires with validated measurement instruments are used. This includes health-related quality of life (EuroQol25–27 (Eq-5D-5L)), symptoms of depression (Patient Health Questionnaire28 29 (PHQ-9)), anxiety (Generalised Anxiety Disorder Screener30) and post-traumatic stress disorders (Impact of Event Scale revised31). The aim is to identify possible resources (Six-item Self-Efficacy Scale32) and strategies that can contribute to convalescence on the one hand and address the specific care needs of people in need of care or support for sustainable prevention on the other.

Care needs of patient participants

Special medical and care needs among this group of people in the different care settings are largely unknown so far. Table 2 shows the care parameters of interest, which include factors such as deprivation. In addition to the mobility and social participation, such an assessment should also include care services already used, health literacy (European Health Literacy Survey33), individual coping strategies, physical function and frailty, care diagnoses of the North American Nursing Diagnosis Association (NANDA)34 and a geriatric assessment (Barthel-Index35). From these findings, it is to be derived which services may be necessary for sustainable prevention in long-term care in order to be able to contribute to an improvement of resilience, so that an individual, self-determined life and living-oriented towards the principle of normality is still possible.

Sociodemographic differentiations also play a role in all these dimensions and educational level, former, professional background, income class and family situation are collected for the distillation of at least trend statements.

Needs of formal and informal caregivers and GPs

Tables 3–5 show that parameters of formal (table 3) and informal caregivers (table 4) from outpatient and inpatient care as well as GPs (table 5) are collected with regards to coping with the burdens of the pandemic to enable addressing any deficits. Contextual information on sociodemography, structural information on the care facility or the GP practice as well as data on the psychosocial health (PHQ-9) and stress situation (Maslach-Burnout-Inventory36) are collected. In addition, the formal and informal caregivers will be asked about their own SARS-CoV-2 infection and about their vaccination decision.

Qualitative substudy: stakeholder interfaces in long-term care

The field of long-term care in the context of the COVID-19 pandemic is characterised by a multitude of stakeholders: people in need of care, their professional and informal carers, home managers, relatives, GPs and those responsible in politics and administration. The constellation of stakeholders is complex, generates interactions and must be managed accordingly. Administrative requirements, for example, not only have to be integrated into the everyday care of professionals but also coordinated with the habits and expectations of those in need of care and their relatives. Hence, interface problems arise. A total of approximately 60 expert interviews will be conducted in form of semistandardised, guideline-supported interviews, which are primarily aimed at the perspective view of the interfaces of the different stakeholders. Patients, GPs, formal and informal carers will be recruited primarily from the study participants of the quantitative study. Based on interviews already conducted, further relevant actors will be identified and requested as interview partners. Here, the aim is to achieve the greatest possible variance with regards to the characteristics of the level of care, the place of residence in urban or rural regions and the presence or absence of family connections. The group of caregivers is also differentiated according to the care setting as well as according to the level of education and function. We will form conceptual categories (deductive—based on the theory of functional differentiation) for the computer-assisted coding and evaluation of the interviews with MAXQDA, which will be adapted, refined and supplemented in an iterative process during analyses (inductive). All interviews will be conducted by appropriately trained staff with a background in sociology (KM), who will also conduct or supervise all qualitative analyses.

The aim of the study is to name not only the challenges and needs of the various stakeholders but also their structural conditionality in a highly differentiated field. The qualitative results, thus, not only complement the quantitative research approach on the one hand but also offer an extended interpretive framework for the quantitative results.

Sample size calculation and stratification

Based on 600 (SG) and 200 (each CG) persons recruited for the registry, we simulate minimal detectable (statistical) difference for major outcomes (age, comorbidities and mortality). Comparing the SG and the controls using a two-tailed t-test or log-rank test, with the assumptions for the significance level α=0.05 and the power β=0.8 and given SD, the detectable differences for the following variables are obtained: age: SD=10.0; detectable difference of −2.29 or 2.29; comorbidities: SD=3.1; detectable difference of −7.10 or 7.10; mortality: median survival time=4.0; detectable difference of 2.66 or 6.56; EQ-5D-5L: SD=0.29; detectable difference of −0.07 or 0.07. With respect to the limited life expectancy of care recipients, we conservatively estimate that after 4 years about 30% of the study participants, across all levels of care, will still be alive.37 The registry will, therefore, be expanded with additional participants at regular intervals and evaluated separately in subgroups. The aim is to achieve a relative distribution of the persons in need of care or support in outpatient care (50%) and inpatient care (50%). In order to account for the differences in medical infrastructures, population density and regional differences in infection incidence, we are aiming at an equal stratification according to the seven administrative districts in Bavaria.

Statistical analysis

All collected parameters of the study participants are analysed descriptively. Analyses are performed for the entire population and stratified by level of care, outpatient and inpatient care, gender and age groups. For group comparisons between patients with positive SARS-CoV-2 PCR test and controls, the χ² test or Fisher’s exact test are used for categorical variables, the t-test (normally distributed variables) or Mann-Whitney-U test (non-normally distributed variables) for metric variables and the log-rank test for survival times. All p values are purely exploratory. Regression models are used to identify, among other things, risk factors that predict a severe course, occurrence of long-term consequences of a COVID-19 infection or a deterioration of the quality of life due to a COVID-19 infection. The survival of the study and CGs is shown graphically using Kaplan-Meier curves. Missing data for the study population are imputed according to multiple imputation where appropriate.

Non-responder analysis

As in most research in the ambulatory care environment, the external validity of our findings is vulnerable to participation bias. For example, it is conceivable that non-responding institutions are particularly burdened by the pandemic. To understand better who does and does not participate, we will conduct an analysis of a subsample of the non-responding care facilities or practices via telephone or postal surveys scheduled 6 and 12 months after first contact, in order to elicit structural and contextual information about the facilities.

Patient and public involvement

Members of the BaCoM advisory board (listed twww.bacomstudie.de) represent a broad expertise in the field: science, patient advocacy, health assurances, health authorities, institutional facilities, etc. The board commented and approved the protocol and comment on the results (public outcome symposium). In addition, we present the protocol and results to a primary care-related citizen forum (‘Bürgerforum’) in Wurzburg and Erlangen.

Ethics and dissemination

Informed consent

All participants provide written informed consent to participate. If a treating GP participates in BaCoM, he/she informs his/her patient about the study. Otherwise, the enrolment and information are provided by the doctor of the study team. If the person in need of care or support is not capable of giving consent him or herself (eg, dementia, cognitive impairment), consent can be given by the legal guardian. In case a person who initially gives consent subsequently loses capacity at one or more of the data collection time points, we only carry out further surveys if the consent to the study also signed by the legal guardian.

We made a conscious decision in the study design not to exclude people with severe cognitive impairment, as these groups (such as people with dementia), may have suffered particularly from the effects of the pandemic (eg, through isolation rules, etc). In order to mitigate undue distress to this vulnerable group, the length of the survey is reduced because some questionnaires are not applicable (eg, health literacy) or the information is collected through an external survey of relatives or carers where this has previously been shown to be possible (eg, PHQ-9). Staff responsible for data collection are instructed to interrupt or end interviews with participants if they notice signs of distress.

The BaCoM Team pursues all measures to protect the interests of participants who are unable to consent.

Study registration and ethics

BaCoM is registered at the German Clinical Trials Register (DRKS) ID: DRKS00026039. The conduct of the study is in accordance with the principles of the Declaration of Helsinki.

The responsible Institutional Review Board of the coordinating study centre of BaCoM (Ethics Committee of the Medical Faculty of the University Hospital of LMU Munich; ethical vote number: #20–860) and of all participating study sites (Ethical Committees at the Medical Faculties of the University of Würzburg and Friedrich-Alexander-University of Erlangen-Nuremberg) approved BaCoM study procedures.

Data access and protection

All data are collected with pseudonyms (ID) first on paper-based case report forms and then transferred in electronic case report forms (double data entry). Data entry takes place on the servers of University Hospital of the LMU with ‘LibreClinica’, an open source validated study management software. To ensure a pseudonymised analysis of data, each participant data set is given a unique participant identification number (ID) when being entered into the study database. The anonymity of the data in the context of evaluations is ensured. The allocation between study participant and participant ID takes place in the study centre through the password-protected allocation lists of the study participants. This information is stored separately and not in the database. By using a hierarchical access concept, unauthorised access to the pseudonymised patient data in the database is impossible.

Storage resources for the data are available in the personal cloud storage of the Leibnitz Rechenzentrum (LRZ). For long-term archiving, the Archive and Backup Service offered by the LRZ based on the IBM Spectrum Protect software is used. Copies of all data in the archive are made on separate tapes to increase security. Data quality is checked for errors electronically and on-site by experienced monitors. Data access to the final data set is provided to the BaCoM Study Group along with written use and access rules.

Dissemination

As an instrument for optimising ambulatory COVID-19 care in Bavaria, the results of the interdisciplinary monitor are presented in regular progress reports and discussed with other (external) experts at symposia. On this basis, further questions or measurement instruments can be systematically included, or less meaningful variables are removed.

Findings are presented at scientific conferences and through peer-reviewed publications.

Ethics statements

Patient consent for publication

Not applicable.