Disease trajectories and end of life care in a Norwegian ALS cohort

Abstract

Objective

This retrospective cohort study aims to provide comprehensive data on the disease trajectory and end-of-life care in patients dying from ALS.

Methods

The study presents detailed information on a cohort dying from ALS in the area served by Akershus University Hospital from 2011 to 2022. Data was obtained from the patient medical records.

Results

118 patients were included. 65 patients (55%) were referred to the Department of Palliative medicine for specialist palliative care, with a median duration from referral to death of two months and one third of patients not being able to communicate verbally at the time of referral. The most common life-sustaining treatment was non-invasive ventilation and percutaneous endoscopic gastrostomy, whereas most prevalent limitation of life-sustaining treatment was withholding of tracheostomy with invasive ventilation. Hospital was the most common place of death with 54 (46%) of all deaths. For in-hospital-deaths, the most common place of death was the palliative care ward, with 21 patients (39%), but a substantial proportion (17%) died at intensive or intermediate care units. Overall, 37% of in-hospital deaths had been admitted to either intensive or intermediate care units during their last week of life, which may suggest these patients deteriorated rapidly without having documented advanced care planning and therefore received resource demanding and futile treatment during the last week of life.

Conclusion

Many patients lived in their own homes until admitted to their last hospital stay in life. However, these last hospital stays were frequently characterized by insufficient advanced care planning, leading to futile overtreatment.

Trial registration

Retrospectively registered.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12883-026-04698-8.

Background

Amyotrophic lateral sclerosis (ALS) is a fatal neurodegenerative disease causing loss of motor neurons and subsequent muscle weakness and atrophy [1, 2]. It often has a focal onset but is relentlessly progressive and disseminates to different body regions. In the absence of life-sustaining treatments respiratory failure and dysphagia with impaired oral nutrition/hydration limit survival to 2–5 years after diagnosis [3, 4].

Even though incidence of ALS is increasing it is still a rare disease with an annual incidence of 3 to 5 per 100.000 in Europe and North America [1, 5, 6]. Due to the fatal nature and often rapid progression many patients experience significant physical, emotional, spiritual and existential distress, with a great need for palliative care almost from the time of diagnosis [7]. The most prevalent symptoms of ALS are paresis, spasticity, dyspnea, dysarthria, difficulty chewing, dysphagia, thick mucus, and sialorrhea, all as a result of the motor neuron affection [1]. Other prevalent symptoms are pain, sleep disturbance and cognitive changes [8]. Patients have increased risk of developing depression both before and after the diagnosis [9]. Even though there are several pharmacological options for symptom management in ALS [7, 10], there is a lack of robust evidence for efficacy and patient selection for these interventions [11, 12]. Traditionally ALS patients are followed by both the primary health care services and by neurologists in a multidisciplinary team often consisting of pulmonologist, nurse, physiotherapist, nutritionist and speech therapist. The multidisciplinary team addresses disease and symptom management, psychosocial support, advance care planning (ACP) and end of life planning. There is increasing evidence that specialist palliative care (SPC) improves patient care in ALS with improved symptom assessment and management, ACP and discussion about goals of care [2, 13–15]. During the disease trajectory, it is necessary to decide whether life-sustaining treatments such as non-invasive ventilation (NIV), tracheostomy with invasive ventilation (TIV), parenteral nutrition and gastrostomy tubes should be initiated [7, 10]. Because these interventions may also provide symptom relief and improve quality of life, they are sometimes initiated without life-prolonging intent [16–19]. These treatments are in some cases terminated at the patient’s request if quality of life is poor despite treatment. The ethics of TIV has been questioned with particular reference to non-maleficence and justice [20]. In Norway, there is currently no established clinical practice or national protocol for organ donation from ALS patients, including donation after controlledcirculatory death.

According to studies in America and Europe the majority of ALS patients die in their own home or in a palliative care unit, but with substantial variation between countries [21–25]. There are few data available on which symptomatic, life prolonging and end-of-life treatments are provided during the disease trajectory. This is a study of a cohort dying from ALS in the area served by Akershus University Hospital (abbreviated as Ahus) from 2011 to 2022. The overall aim is to describe life prolonging and symptomatic treatments during the disease trajectory, and the end-of-life care, including place of death and the role of SPC in patients dying from ALS in a large Norwegian hospital.

Methods

Study design

This study was a retrospective observational cohort study of deceased patients with ALS.

Study population

The study was conducted at Ahus, in the Greater Oslo Region, Norway. As a publicly funded hospital Ahus serves approximately 618 000 inhabitants, which is slightly above 10% of the Norwegian population. Ahus is the only local hospital in the region and serves an unselected population of ALS patients with multidisciplinary care. In addition to specialist health care services, most ALS patients receive primary health services.

During the study period, there were no standardized care pathways involving SPC for ALS patients. The Department of Palliative Medicine, founded in 2013, is the sole provider of SPC for adults in the region. Before 2013 SPC was provided by a smaller palliative team. Patients can be referred to SPC regardless of diagnosis. Patients are referred to the department of palliative medicine, which then determines whether the initial assessment should take place in an outpatient clinic and subsequently admitted to the palliative care ward if their condition requires more comprehensive palliative measures. Formal co-management of ALS patients began around 2017, but the disease-specific referral criteria for SPC were established only after the study period.

Data collection and management

All patients being registered in hospital records with the diagnostic code for ALS (G12.2) according to the International Classification of Diseases, 10th revision, clinical modification (ICD-10CM) were identified. Both inpatients and outpatients were eligible if they were treated for ALS from 2011 and forward and were deceased by December 31st 2022.

Pre-determined variables were collected from patient records by a single investigator. If there was uncertainty about inclusion criteria or variable coding, a co-author was consulted. For symptomatic treatment, the primary documented indication in the patient`s medical record was used, even if the drug had multiple indications for the same patient.

Statistical analysis

IBM^® Statistical Package for Social Sciences (SPSS), version 25 (SPSS^® Inc., Chicago, Illinois) was used for data management and descriptive analyses. Because no hypotheses-testing was performed sample size calculations and estimation of p-values were not relevant.

Results

Out of 127 identified patients, nine were excluded (five misclassified, three atypical/protracted disease trajectory, one lack of consent from next of kin), resulting in 118 included patients deceased from ALS.

Cohort characteristics and disease trajectory

Mean age at the time of diagnosis was 67,5 years (SD 10,4), and 68 patients (58%) were male. 83 patients (70%) were cohabitating with partner at the time of diagnosis. Median duration from symptom debut to diagnosis was 11 months (IQR 7–19), and median duration from diagnosis to death was 12 months (IQR 4–21).

65 patients (55%) were referred to the Department of Palliative medicine to SPC, of whom 78% were referred during the last half of the study period. Referral physician for 52% of patients were neurologists, 18% primary physicians, 14% pulmonologists, 12% nursing home physicians, 2% oncologists and internal medicine physicians. Median duration from referral to SPC to death was 2 months (IQR 1–9).

Functional status at time of referral to SPC

At the time of referral to SPC 26 of 65 patients (42%) were bed- or chairbound, 23 patients (36%) received parenteral or enteral nutrition, and 22 patients (34%) were not able to communicate verbally (Table 1).

Table 1.

Functional status for ALS patients at time of referral to specialist palliative care

Mobility	Total n = 65 (%)
Walking independent	20 (31)
Able to walk with frame, stick, tripod	11 (17)
Walking with assistance from 1–2 persons	6 (9)
Bedbound or chairbound	27 (42)
Unknown	1 (1)
Nutrition
Feed unaided	12 (18)
Feed unaided with specially prepared meal	20 (31)
Needs help or supervision	7 (11)
Must be fed	0
Only parenteral/enteral nutrition	23 (35)
Unknown	3 (5)
Speech
Normal speech	14 (22)
Slurred but intelligible	24 (37)
Unintelligible speech/groans	4 (6)
No speech	22 (34)
Unknown	1 (1)

Symptomatic treatment

The most common symptomatic treatments provided during the disease trajectory were pain pharmacotherapy for 70 patients (59%), anxiety pharmacotherapy for 69 patients (59%) and sleep disturbance pharmacotherapy for 61 patients (52%) (Table 2).

Table 2.

Symptomatic treatment during the disease trajectory in deceased ALS patients

	Total n = 118 (%)
Pain pharmacotherapy a	70 (59)
Anxiety pharmacotherapy b	69 (59)
Sleep disturbance pharmacotherapy c	61 (52)
Communication tool d	51 (43)
Cough machine	42 (36)
Depression pharmacotherapy e	35 (30)
Sialorrhea pharmacotherapy f	33 (28)
Spasticity pharmacotherapy g	18 (15)
Sialorrhea Botulinum toxin therapy	17 (14)
Pseudobulbar affect pharmacotherapy h	5 (4)
Sialorrhea radiotherapy	4 (3)
Spastitcity Botulinum toxin therapy	4 (3)

^aParacetamol, NSAIDs, opiods, Gabapentin, Pregabali

^bSelective Seretonin Reuptake inhibitors, Benzodiazepine

^cBenzodiazepines, benzodiazepine hypnotics, Amitriptyline off-label, Mirtazapine off-label

^dCommunication tools ranging from pointing boards to speech generating eye-tracking system

^eSelective Seretonin Reuptake inhibitors, Amitriptyline, Tetracyclic antidepressants

^fAmitriptyline off-label, scopolamine patch

^gBaclofen, Benzodiazepines

^hAmitriptyline, Citalopram

Life-sustaining treatment provided during disease trajectory

The most common life-sustaining treatments provided during the complete disease trajectories were NIV treatment for 70 patients (59%) and percutaneous endoscopic gastrostomy (PEG) for 67 patients (57%). Five patients (4%) received TIV. In addition to the patients receiving TIV, five patients were intubated and received invasive ventilation due to acute deterioration, of whom three were successfully weaned from respirator, one died due to withdrawal of treatment, and one was converted to TIV. Intensive or intermediate care unit treatment (ICU/IMCU) was provided for 29 patients (25%) and parenteral nutrition for 10 patients (9%).

Do-not-resuscitate (DNR) order

A DNR order had been documented for 104 patients (88%) in hospital medical records, 94% of patients referred to SPC and 81% of patients in the group not referred. Neurologists were responsible for 82 (79%) of the DNR orders. Information to patients or next of kin about DNR was documented in 87 (84%) of the DNR orders. It was documented in 55 of 104 patients (53%) that DNR orders were in accordance with the patient’s expressed preference. Median duration from DNR order to death was 76 days (IQR 13–244), longer in patients referred to SPC (113 days (IQR 17–339)), than in those not referred (68 days (IQR 7-114)).

Decisions on limitation of life-sustaining treatment

The most prevalent limitation of life-sustaining treatment was withholding of TIV in 104 patients (88%) (Table 3).

Table 3.

Life-sustaining treatment provided and decided withheld during disease trajectory of ALS patients

Total n = 118	Provided	Withhelda	Timeb
	n (%)	n (%)	months (IQR)
Tracheostomy invasive ventilation	5 (4)	104 (88)	8 (1–12)
Non-invasive ventilation	70 (59)	35 (30)	3 (0–2,5)
Perctuaneous enteral tube feeding	67 (57)	23 (20)	4 (0–6)
Parenteral nutrition	10 (9)	16 (14)	0,5 (0–1,5)
Caldiopulmonary resuscitationc, e	-	104 (88)	75,5 (13–244)d
Antimicrobial agents e	-	22 (17)	0 (0–1)

^a Documented decision on withholding treatment in medical record

^b Median duration from documented decision to death

^c Documented DNR order

^d Duration in days

^e Sufficient data not available through medical records

Limitation of other life-sustaining treatments was recorded in 30% or less of the patients. Median duration from decision to withhold TIV to death was eight months. Decisions on limitation of other life sustaining-treatments were made closer to death. For patients referred to SPC withholding NIV was documented for seven patients (11%) before referral and 18 patients (28%) after referral. Documented decision on no PEG placement for eight patients (12%) before referral and 13 patient (20%) after referral. Documented decision on no parenteral nutrition for three patients (5%) before referral and eight patients (12%) after referral. Documented decision on no parenteral antibiotic therapy for two patients (3%) before referral and 15 patients (23%) after referral. Decisions on no hospital admission were documented for two (2%) before referral and documented for five patients (8%) after referral. Among patients referred to SPC, decisions to withhold non-invasive ventilation (NIV) were documented in 11% prior to referral and increased to 28% following referral. Similarly, documented decisions against PEG placement rose from 12% before referral to 20% after referral. Decisions to forego parenteral nutrition were noted for 5% of patients prior to referral and 12% post-referral. Withholding of parenteral antibiotic therapy was documented in 3% before referral and in 23% after referral.

Place of death

Hospital was the most common place of death with 54 (46%) of all deaths. It was recorded that 34 patients (29%) died at a nursing home and 14 patients (12%) died in their own home, whereas data on whether death was at home or nursing home was not available for 16 patients (14%). Among patients referred to specialized palliative care (n = 65), 33 patients (51%) died in hospital, 18 patients (28%) in nursing homes, and 13 patients (20%) at home, with one patient (1.5%) unknown. In contrast, among patients not referred to specialized palliative care (n = 53), 21 patients (40%) died in hospital, 16 patients (30%) in nursing homes, and only one patient (2%) at home, while the place of death was unknown for 15 patients (28%) of patients. For in-hospital deaths, the most common place of death was the palliative care ward, with 21 patients (39%). 11 patients (20%) died at a pulmonology ward, nine patients (17%) died at ICU/IMCU and six patients (11%) died at the neurology ward. Nine patients died in the ICU or IMCU. One of these patients had documented treatment limitation decisions prior to the final hospital admission, consisting of a DNR order and a decision to withhold invasive ventilation. The timing of DNR decisions among the remaining patients ranged from the day of death to seven days prior to death, with tree decisions documented on the day of death.

Cause of in-hospital deaths

Pneumonia was the most common cause of in-hospital death with 15 of 54 patients (28%). 20 patients in total, including 15 with pneumonia, had a suspected end-of-life infection, and 80% of them received intravenous antibiotics. In the nine patients (8%) who died due to withdrawal of life-sustaining treatment the median duration from withdrawal to death was 30 min (range 2–135 min).

Last hospital stay, in-hospital-deaths

Of patients dying in hospital 47 patients (91%) were admitted from their own home and 5 patients (9%) were admitted from nursing homes. The median duration of the end-of-life hospital stay was 4 days (IQR 2–7 days).

Medical treatment last week of life, in-hospital-deaths

In the 54 in-hospital deaths the prevalence of parenteral hydration and antibiotic therapy during the last week of life was 56% and 32% respectively. NIV was initiated in 18 (33%) during the last week of life, and 8 (15%) and 12 (22%) were respectively admitted to ICU and IMCU during the last week of life (Table 4).

Table 4.

Medical treatment during the last week of life for ALS patients who died in hospital

	Total n = 54 (%)
Parenteral hydration	30 (56)
Continued enteral tube feeding	29 (54)
Initiated non-invasive ventilation	18 (33)
Parenteral antimicrobial agents	17 (32)
Intermediate care unitb	12 (22)
Intensive care unita	8 (15)
Initiated invasive ventilation	4 (7)
Initiated parenteral nutritionc	3 (6)

^a Patients with TIV as a reason for requiring intensive care unit admission excluded n = 2

^b Patients with NIV as a reason for requiring intermediate care unit admission excluded n = 2

^c Patient who continued parenteral nutrition excluded n = 1

Palliative pharmacotherapy last 48 h, in-hospital-deaths

40 patients (77%) received morphine and benzodiazepine. Eight patients (17%) received glycopyrrolate. Four patients (8%) received propofol in the context of palliative sedation. Median parenteral morphine dosage per 24 h was 39,5 mg and median parenteral midazolam dosage per 24 h was 22 mg.

Discussion

The clinically most important finding in this study was that a substantial proportion of ALS patients dying in hospital received resource-demanding medical treatments including antibiotics and ICU/ICMU with invasive and non-invasive ventilation initiated during the last week of life. Such interventions in the terminal phase are likely to be futile and may prolong suffering rather than improve quality of life. This underlines the importance of discussing ACP and end-of-life care earlier in the disease trajectory. Based on our findings it appears that involvement of specialist palliative care might be beneficial for such ACP, and we speculate that standardized care pathways might further improve ACP and end-of-life care.

In Norway, there is no nationally implemented advance directive document that is systematically integrated into the standard patient pathway within the public health care system. While patients may express their preferences regarding end-of-life care through conversations with health care professionals regarding advance care planning, these preferences are typically documented narratively in the medical record and registered in the patients “critical information” in their medical records, rather than through a standardized advance directive form. Avoidance and delayed ACP have been shown to lead to emergency intubations, sometime against patient’s wishes [26]. If possible it is important to start ACP while most patients can still express their wishes verbally and are not affected by cognitive impairment, and re-evaluate in settings with triggers, clinical deterioration, recurrent hospitalizations or caregiver distress [27]. Half of the patients were referred for SPC 0–2 months before death, and a third of patients could not communicate verbally at the time of referral, making communication around life-sustaining-treatment and other preferences more challenging.

TIV may allow the patient to survive with increasing paresis, ultimately leaving the patient in a locked in state and totally dependent on help from others [20]. The use of TIV is therefore in Norway restricted to selected cases [28]. Accordingly, most patients had a documented decision on withholding TIV, in median eight months from death, making this the first and most frequent decision to withhold life-sustaining treatment during the disease trajectory. DNR orders being documented just above two and a half months from death is surprisingly close to death given the terminal nature of ALS and raises concerns related to whether a patient with advanced ALS can be weaned from the respirator following successful resuscitation. The finding is, however, in line with findings from a general death cohort were DNR decision often were not made at the time cardiopulmonary rescue (CPR) would be futile but was delayed until patients received end-of-life care [29]. In our cohort, patients referred to specialized palliative care had nearly twice as long median duration from DNR order to death. This might suggest earlier recognition of end-of-life needs but could be influenced by selection bias regarding which patients were referred to specialist palliative care. It could be speculated that referral to palliative care coincides with an advanced clinical stage at which patients experience substantial symptom burden and suffering, and where the focus of care has shifted away from life-prolonging interventions toward comfort-oriented goals, which may explain the higher rate of withholding antibiotic therapy among patients referred to specialist palliative care.

Delayed ACP and non-referral to SPC can represent reluctance in some patients to make such decisions or even talk about end-of-life topics. This represents an ethical challenge where honoring patient autonomy increases the risk of violating the principle of non-maleficence [20]. However, the finding that decisions regarding limitation of life-prolonging treatments are made late in the disease trajectory is in line with previous Norwegian findings from other patient populations, indicating that cultural and organizational issues influence the timing of these decisions [29].

Nearly half of our cohort died in hospital (46%), with a higher percentage dying at home (20%) among patients referred to SPC. These findings suggest that patients referred to SPC are likely to receive closer follow-up from specialized health services during the final phase of life. The higher rate of home deaths in the group referred to SPC might be influenced by selection bias, with either an expressed wish for home death or a general desire for discussing end-of-life care a reason for referral. In contrast, 19% in-hospital deaths and approximately 21% deaths in own home, have been reported from neighboring country Sweden [22], and in studies from the USA, Germany and the UK show hospital death rates of 0–22%, and home death rates of 52–64% [24, 25]. Variation likely reflects differences in health care system, culture, or access to palliative care units/hospices or SPC. It could, however, also be a result of lack of ACP directives, or a combination of lack of access to SPC and lack of ACP directives. In line with the present finding, 7% of patients in the German cohort [25] were admitted to ICU and several intubated shortly before death, emphasizing the urgent need of early discussions of advance directives for end-of-life care. For many palliative patients with other terminal diseases like cancer, dying in hospital is preferably avoided if the end of life is well planned, and for some countries this is also applicable for ALS [22]. We would, however, argue that the large proportion of hospital deaths in the present study also could be viewed as positive and beneficial to the patients as they lived in their own home until very close to death. This indicates that they received sufficient assistance and palliative care in their own homes until few days before death when they were admitted to hospital for end-of-life care. With improved ACP and standardized care pathways it is possible that an even higher proportion than the observed 40% of in-hospital deaths could have been admitted directly to the SPC ward for treatment focusing on holistic palliative care and symptom relief at the end of life, avoiding futile attempts at life prolonging treatment.

Given the inevitable terminal nature of ALS and uncertainty of the disease trajectory, ACP requires skillful communication around complex topics including treatment which may alter the disease trajectory [10]. Patients need time to overcome the existential shock associated with an ALS diagnosis, and also reflect on the challenges and opportunities posed by living with the disease [30]. They have different coping mechanisms, there are barriers to discussing sensitive topics, and the timing of discussions about ACP is challenging [10, 14, 31]. In Norway, neurologist are the frontline of palliative care for ALS patients and guide patients and their caregivers through discussions on end of life, and deciding life-sustaining measures throughout the disease trajectory [26]. A patient/caregiver survey in USA showed that patients were quick to think about end-of-life options and discussed these with their spouses already one month after the diagnosis [31]. Surveys and clinical experience further show that most patients feel comfortable and embrace the chance to discuss end-of-life issues [31, 32].

Patients in our study were referred to SPC at the neurologists’ or other physicians`discretion based on joint decision making with patients. This is in line with previous reports [2]. However, the optimal timing of referral to SPC remains to be established. On one hand some patients appear to be reluctant to be referred early, while on the other hand it is easier for SPC to make plans in accordance with patients’ preferences and values if patients are referred before communication and cognition is severely impaired. A topic for future research should be whether more standardized care pathways could improve patient care without violating the autonomy and preferences of individual patients. Another important area for further investigation is the various barriers which may hinder referral to specialist palliative care.

Strengths and limitations

The major strength of this study is access to comprehensive details on the death setting and treatment for in-hospital deaths. Moreover, only one patient was excluded due to lack of consent from next-of-kin. Major weaknesses are the cohort size due to ALS being a rare disease, the lack of data on place of death for 14% of patients, and that data from out-of-hospital treatments were not available. The descriptive nature of the study does not allow conclusions regarding causality. Notably, SPC was introduced and became increasingly used at Ahus during the study period. Thus, we cannot exclude that differences in use and timing of ACP between those referred or not to SPC could represent selection bias or underlying temporal trends in the use of ACP coinciding with a more comprehensive SPC service at our hospital.

Conclusion

Many patients in the present cohort lived in their own homes until admitted to their last hospital stay in life. However, these last hospital stays in life were frequently characterized by insufficient previous discussions on ACP and life-sustaining treatment, leading to futile overtreatment. While early clarifications regarding tracheostomy invasive ventilation, DNR and nutrition therapy were made, directives for other life-sustaining treatments were often lacking. It may pose difficulties to find the right timing for referral to SPC for ALS patients, but it can be hypothesized that formal procedures for co-management or standardized care pathways could improve end-of-life care and reduce over-treatment during the last week of life.

Abbreviations

ALS

Amyotrophic lateral sclerosis

ACP

Advanced care planning

SPC

Specialist palliative care

NIV

Non–invasive ventilation

TIV

Tracheostomy with invasive ventilation

Ahus

Akershus University Hospital

SPSS

Statistical Package for Social Sciences

SD

Standard deviation

IQR

Interquartile range

ICU

Intensive care unit treatment

IMCU

Intermediate care unit treatment

DNR

Do–not–resuscitate

PEG

Percutaneous endoscopic gastrostomy (PEG)

CPR

Cardiopulmonary resuscitation

USA

United states of America

UK

United Kingdom

Biographies

Henriette Johansen

HJ is senior consultant at Department of Palliative medicine, Akershus University Hospital and a specialist in neurology.

Ola Nakken

ON is senior consultant and postdoc at Department of Neurology, Akershus University Hospital and specialist in neurology.

Trygve Holmøy

TH is section head and senior consultant at Department of Neurology at Akershus University Hospital and professor of neurology at Institute of Clinical Medicine, University of Oslo.

Olav M. S. Fredheim

OMSF is head of Department of Palliative Medicine, Akershus University Hospital, specialist in anaesthesiology and professor of palliative medicine at Institute of Clinical Medicine, University of Oslo.

Authors’ contributions

HJ collected and analyzed the patient data. If there was uncertainty about inclusion criteria or variable coding, OMSF was consulted. All authors discussed the results and contributed to the final manuscript. HJ revised the manuscript and OMSF supervised the project.

Funding

Open access funding provided by University of Oslo (incl Oslo University Hospital). This study received funding from the Division of Surgery, Akershus University Hospital.

Data availability

All data generated or analysed during this study are included in this published article.

Declarations

Ethics approval and consent to participate

The study was approved by the Regional Committee for Medical and Health Research Ethics for south-east Norway (Reference number 621459) and the institutional data protection officer at Ahus. Since the patients were deceased, informed passive consent was obtained from the next of kin. Written information about the study was sent to the next of kin, who were given the opportunity to decline participation on behalf of the patient. No patients were included if the next of kin opted out. The study was conducted in accordance with the ethical principles outlined in the Declaration of Helsinki.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.