Home-based testing protocol to measure physiological responses to everyday activities in ME: a feasibility study

ABSTRACT Background and objectives Individuals with Myalgic Encephalomyelitis (ME) have shown altered physiological responses during maximum cardiopulmonary exercise testing. However, maximal testing is not representative of the everyday activities reported to cause or increase symptoms in ME, and is not accessible for those with severe or very severe illness. The aim of this study was to assess the feasibility and acceptability of a home-based testing protocol to measure physiological responses in ME to everyday activity. Methods Researchers attended participants’ homes to collect data and provide equipment for independent testing. Adults with ME who met the International Consensus Criteria wore a portable metabolic assessment system and a physiological stress monitor. Blood pressure, heart rate, oxygen saturation and lactic acid were assessed during a range of everyday positions and activities in their own homes. Results Online recruitment yielded 70 volunteers in 24 h. 17 eligible individuals reflecting a range of illness severities were enrolled. All participants found the procedures acceptable with 12 (70%) subjects completing every listed activity. Apparent physiological abnormalities were identified in all participants. Conclusion Physiological measurement during everyday activities was feasible for our participants who represented a range of ME severities. Activities must be adapted for different levels of severity to avoid significant symptom exacerbation. Further research is needed to develop home-based assessment protocols to advance the biobehavioral understanding of ME. Trial registration number: ISRCTN78379409.


Introduction
Myalgic Encephalomyelitis (ME) is a complex, acquired multi-systemic disease that may reflect dysfunction of the neurological control system resulting in faulty interactions between the central nervous system and major body systems [1].Diagnostic criteria for ME list a range of symptoms including cognitive impairments, orthostatic intolerance, pain, fatigue, and difficulty sleeping [1].The hallmark characteristic of the illness is post-exertional malaise (PEM), which is an abnormal response to physical, cognitive, social and/or emotional exertion leading to additional symptoms or an exacerbation of ongoing symptoms [2].This response may occur within hours post-exertion, or there may be a delay of over 24 h, with the aggravated symptoms lasting for hours, days, months or more [2].Current treatments show limited effectiveness in managing symptoms or ameliorating pathology, partly attributable to a lack of understanding of underlying mechanisms that drive symptoms and impairments triggered by performance of everyday tasks.
Individuals with ME have shown altered responses during and recovering from maximum cardiopulmonary exercise testing (CPET), including higher lactic acid levels at submaximal effort, a lower anaerobic threshold (AT), a blunted rise in heart rate (chronotropic intolerance), and abnormally low oxygen consumption (VO 2 ) [3][4][5][6][7][8][9][10][11][12][13][14].When the CPET is repeated 24 h later, a further deterioration in cardiopulmonary function has been found.This response pattern differs from both matched healthy controls, and people with other chronic conditions such as cardiopulmonary disease, multiple sclerosis and HIV, who reproduce or improve their performance on the second test [15].This abnormal response pattern in ME is an objective demonstration of PEM and is sometimes used as a diagnostic test for ME [16].
While maximal exercise testing has revealed abnormal physiology in ME, the method has several limitations.The collected data reflects the person's capacity during the specific testing conditions and cannot be generalised to performance in daily life.Maximal testing is also not representative of the conditions which lead to PEM in everyday life, such as routine daily physical activities, postural changes, and cognitive or sensory exertion [2].Notably the physiological effects of non-physical activities on PEM have received only limited attention.Furthermore, CPET may exclude people with severe or very severe illness who are unable or unwilling to undergo the procedure, e.g. the severity of symptoms precludes travel to a testing site [17].Finally, CPET is not an appropriate assessment for PEM in relation to everyday tasks as it lacks ecological validity nor is it a feasible clinical outcome measure.
The aim of this study was to assess the feasibility and acceptability of a testing protocol to measure physiological responses of individuals with ME in association with everyday activities in their own homes.Our long-term goal is to identify potential biobehavioral mechanisms in real time that can potentially inform diagnostic assessments and outcome measure (s).Our testing protocol assessed cardio-pulmonary function in individuals with ME including rate of oxygen consumption, heart rate, heart rate variability, blood pressure, oxygen saturation, and lactic acid levels.Feasibility was assessed by: (1) Ability to recruit subjects; (2) Acceptability of our methods and adherence to our protocol by participants; (3) Usability of outcome measures; and (4) Measure sensitivity to physiological parameters.

Method
This observational study design followed STROBE guidelines for cross sectional studies [18].Adults with self-reported ME were recruited online in May 2021 via Facebook, Twitter and ME related websites.The inclusion criteria required participants to be aged over 18 years (with no upper age limit), to meet the International Consensus Criteria [1] (ICC) for diagnosis of ME which requires the presence of post-exertional neuroimmune exhaustion (defined in other criteria as post-exertional malaise), at least one neurological impairment (e.g.cognitive impairment, pain, sleep dysfunction), at least one immune/ gastro-intestinal/genitourinary impairment (e.g.recurrent flu-like symptoms, irritable bowel syndrome) and at least one energy transport/ion transport impairment (e.g.orthostatic intolerance, loss of thermostatic stability).Also, participants needed to reside in proximate locations that the researcher could visit in person.Bedbound status was an exclusion because it precluded (out-of-bed) testing procedures.Participants were selected to ensure a wide range of ME severities (as measured with the CFIDS Disability Scale [19]).Study screening was completed online or by telephone in which the researcher confirmed ICC criteria for ME and participant agreement to undergo study procedures.To minimise cognitive effort, the patient information sheet was sent before the call, so participants could read and digest the information at their own pace.
The information sheet provided details about the purpose of the study, procedures involved, risks and benefits of taking part, the right to withdraw at any time, data protection and confidentiality.Also included was identification of the study investigators and the study funders, as well as contact information for concerns and complaints.Information on future dissemination of study results was also provided.All interested individuals provided signed, fully informed consent on the day of assessment after reviewing the patient information sheet sent in advance via email.In addition to individuals with ME, five healthy participants were recruited to carry out the portable cardiopulmonary testing and wear the physiological stress monitor for a further five days to provide comparable data.Healthy participants had to report no medical conditions and all gave informed consent.Each healthy volunteer was matched to a participant with ME in terms of age and sex.

Measures
Fatigue Severity Scale (FSS): This is a nine-item self-report questionnaire that measures the effect of fatigue on function [20].Each item is rated on a Likert-scale ranging from 1 (no impairment) to 7(severe impairment).Answer choices for each item are summed and divided by the number of items answered.Scores range from 1.00 to 7.00, with greater scores suggesting greater fatigue severity.To limit cognitive exertion, this questionnaire was sent before the start of data collection so participants could familiarise themselves with the questions.
Portable Cardiopulmonary-Monitoring System (MetaMax 3B, Cortex Biophysik GmbH, Leipzig, Germany; Figure 1) was calibrated prior to each test, which measured: anaerobic metabolism).Thus, a score over 1 indicates the participant has exceeded their anaerobic threshold.
Portable cardiopulmonary monitoring systems have been successfully used in healthy populations to determine oxygen consumption (VO 2 ) during normal activities of daily living [21] and in people with physical disabilities such as spinal cord injury [22].Portable metabolic devices used for cardiopulmonary exercise testing in a mobile setting have been shown to correlate with standard cardio-pulmonary exercise tests in healthy populations [23]; however, the system has not been validated specifically for use in a mobile setting with people who have ME.
Physiological stress: The FirstBeat bodyguard 2 (Figure 2), uses heart rate, breathing frequency, heart rate variability, and oxygen uptake to identify periods of 'physiological stress' (suggesting sympathetic nervous system dominance), 'recovery' (parasympathetic nervous system) and physical activity.Data are presented as % time spent in each state.The FirstBeat bodyguard 2 has been shown to be valid for healthy participants when compared with electrocardiogram recordings during everyday activity [24]; however, the validity of the measure has not been described in people with ME.Heart rate variability (HRV) is the variation in time between consecutive heartbeats which is regulated by the autonomic nervous system.A high HRV indicates a healthy balance between the sympathetic and parasympathetic systems, whereas a low HRV may be disproportionately influenced by an overactive sympathetic system [25,26].As these measurements are more meaningful over time, the physiological stress measurements were collected after the period of independent testing.Lactic Acid.Lactic acid was measured with a handheld device (Cosmed Lactate Pro 2) to analyse a finger prick blood sample.Lactic acid is an indication of anaerobic metabolism, where energy is produced in the absence of oxygen.High lactic acid levels increase the acidity of muscle cells, causing a burning sensation in the working muscle.Normal lactic acid values range between 0.5 and 2.2 mmol/L during submaximal activities [27].People with ME have shown higher levels during maximal exertion compared to healthy controls [3].The validity of the Cosmed Lactate Pro2 has been tested in healthy athletes.Compared to a laboratory-based analyser it was found to be valid [28].The validity of this equipment has not been reported in ME studies.Blood pressure (BP).BP was measured with a handheld automatic sphygmomanometer cuff on the arm (OMRON M2 Basic).The validity of automatic BP measurement compared to a standard mercury sphygmomanometer method was established with a large population-based survey of 37,816 healthy people and people with hypertension [29].
Oxygenation levels and heart rate.Heart rate and oxygenation level was measured with a finger pulse oximeter (OxiPro 2).The pulse oximeter measurement for measuring heart rate in healthy participants has been compared to physical radial pulse measurement and was found to be valid [30].

Testing procedure
On the day of testing, participants were seen in their homes on waking with the time previously arranged during the screening telephone call.Some participants preferred an early morning hour and others with sleep reversal patterns agreed to an afternoon visit.Following fully informed consent, participants were asked to follow their normal routines for medication and other therapies, but not to eat after waking until testing was complete, although drinking water was allowable if needed.The FSS was completed before physical activity testing.The testing equipment was fitted and then the participant was asked which of the everyday activities they felt able to complete from a standardised list created by the research team based on typical postures and activities of daily living.The activities were: . Lying for 5 min .Sitting for 5 min .Standing for up to 5 min .Typical bathroom activities (e.g.washing, grooming, toileting) for up to 5 min .Walking downstairs .Typical kitchen activities (e.g.preparing breakfast, hot drinks) for up to 5 min .Walking upstairs .A formal cognitive activity (serial sevens test [31]) and other standardised mental arithmetic problems were performed while sitting for up to 5 min.
Participants selected activities from the above list to do at their own pace with rest breaks as needed.All participants wore the portable cardiopulmonary monitoring system and physiological stress monitor, which collected data continually for the duration of the testing process.Blood pressure, oxygen saturation and heart rate were measured before and after each activity.Lactic acid was measured once at the beginning and once at the end of the full set of activities.At the end of the testing procedure, the portable cardiopulmonary monitoring system was removed.Participants continued to wear the FirstBeat bodyguard 2 device and recorded their BP, HR, O 2 and LA three times a day, in the morning, during the day and once at the end of the day for the next six days.
Following the extended data collection period, the researcher returned to collect the equipment and gather informal verbal and written opinions from the participants about their testing experience, either by note taking during conversation or in follow-up email correspondence.Participants were asked their opinions on the feasibility and acceptability of the testing process, particularly wearing and using the equipment; the activities tested; their responses to the physical and cognitive efforts of completing the tasks both during and after testing, and any additional information they wanted to add to inform the process.

Recruitment
Within four hours of posting the invitation to participate online, 20 individuals with ME had volunteered to take part.After 24 h, 70 volunteers were identified, recruitment was closed, and the screening process began.Between May and September 2021, 17 eligible participants were consented, enrolled, and tested.Twelve (71%) were women, 95% were white, and the mean age was 44.5 years (SD 14.8).The mean duration of ME was 15.3 (SD 10) years and most (n = 9, 53%) were moderately impacted (based on CFIDS disability scores between 40 and 60) with high fatigue scores on the Fatigue Severity Scale of (Mean = 6.4;SD 0.75).Self-report co-morbidities were common (Table 1).In addition, five healthy participants were recruited, four (80%) were women, 100% were white and the mean age was 38 years (SD 18.8).

Adherence/feasibility of the activities tested
Table 2 details the activities that participants were able to complete.The total time to perform all activities listed was 30 min plus the time required to ascend and descend the stairs which was not time-restricted.All participants with severe ME (as measured by the CFIDS disability score) and two with moderate ME needed significant rest between each activity.At least some people in each severity group managed to complete every activity.Those who were unable to complete all the testing had moderate (n = 2; 33%) or severe 3 (n = 3; 60%) ME.All participants completed the lying and sitting tasks and all but one participant (who had moderate ME) completed the cognitive task.Standing was attempted by all participants, with 13 (76%) able to do so for the full 5 min.The remainder (n = 4) stopped before 5 min due to symptom exacerbations including heart rate increases from 10 to 21 bpm, dizziness, nausea, feeling weak, legs and arms shaking and self-reported cognitive difficulties.Three subjects who could not complete the standing task had severe ME and a co-morbid diagnosis of orthostatic intolerance, and one was moderately affected without a formal diagnosis.Orthostatic intolerance was a co-morbidity for 7 (41%) participants, 4 of whom were able to stand for the full 5 min but experienced symptoms during testing.

Sensitivity of outcome measures
Data from the Portable Cardiopulmonary-Monitoring System indicated that participants often exceeded their anaerobic threshold during these activities (defined as a respiratory exchange ratio (RER) exceeding 1.0).For two subjects (11%), this occurred while lying down.The activity which caused most participants to exceed their anaerobic thresholds was the cognitive task (n = 10, 62%).One third of participants exceeded this threshold during the more physically demanding tasks (bathroom activities and walking upstairs, Table 3).Healthy participants did not exceed their RER during any activities.We found no published data specifically on RER for healthy participants during everyday activities, although oxygen consumption (VO 2 ) and calculated Metabolic equivalents (METs) have been published during everyday activity in healthy participants [32] and people with spinal cord injury [33] using portable metabolic systems.Physiological data collected with Firstbeat Bodyguard for up to six days after the testing day indicated that all participants spent some time (29-95%) in cardiorespiratory 'physiological stress' (suggesting sympathetic nervous system dominance).The time in 'physiological stress' was not balanced by the time 'in recovery' (suggesting Table 2. Activities completed by participants, according to severity of ME. parasympathetic nervous system activity) which ranged from 0% to 67% of the time monitored.The proportion of time in 'physiological stress' tended to increase with the severity of ME.By comparison, healthy participants exhibited less time in physiological stress on average (30% Healthy: 63% ME) and more time in recovery (55% Healthy: 16% ME).Research using Firstbeat technology with 20,000 participants has shown that 'a typical 24-hour measurement day was classified as "physiological stress" 51% of the time (12 h 14 min) and as "recovery" 49% of the time (6 h 14 min)' [34].
Lactic acid levels for four participants (2 mild, 1 moderate and 1 severe) were greater than 4 mmol/L at the start of the day of testing, suggesting lactic acidosis [27].The 1-hour testing procedure increased lactic acid in 5 people (1 moderate and 4 severe), and reduced lactic acid in 11 participants, reflecting the majority of participants with ME.The procedure did not cause excessive anaerobic metabolic changes.Over the followup days, high levels of lactic acid (>4 mmol/L) were associated with participant-reported high levels of leg/muscle pain (22 occasions), fatigue (15), cognitive difficulties/brain fog (8), headache (8), dizziness (5), tinnitus (3) and nausea (2).One person with mild ME had consistently high levels of lactic acid rising to 17.6 mmol/L on one occasion.This participant always had significant leg pain and stated that they struggled to pace their activities due to childcare commitments.Normal lactate levels range from 0.5 to 2.2 mmol/L [27].
In addition, the physiological responses to assigned tasks revealed: . Changes in blood pressure during activity.One person showed a large increase on standing (from 109/75 while sitting to 144/77 after 5 min standing and continued to rise to 161/85 for a further 15 min in lying with head propped) while four others showed a reduction in blood pressure (mean 13.75 mmHg (range 10-17) in systolic BP).The remaining participants and activities did not show marked changes in blood pressure.

Acceptability of the testing equipment
The Portable Cardiopulmonary-Monitoring System (worn throughout the testing activities) and the handheld sphygmomanometer cuff and pulse oximeter (used before and after each activity to assess blood pressure, oxygen saturation and heart rate) were well tolerated, except for one participant who found the mask (part of the cardiopulmonary monitoring system) very uncomfortable to wear.Another individual felt that they breathed differently than usual when wearing the mask but were able to tolerate it.One participant found the physiological stress monitor comfortable but rather obtrusive and 'obvious' in appearance.

Acceptability of the independent testing
All participants were able to complete independent testing of monitoring physiological stress, blood pressure, oxygen saturation and lactic acid every morning and evening.While the aim was to complete six days of testing, the timeframe was not always met due to logistics of the researcher planning dates to return to collect the equipment.One participant reported some skin irritation from the pads worn with the Firstbeat monitor, but was able to change the location of the pads to manage this.The Firstbeat monitor malfunctioned for another participant and their physiological stress data could not be collected.One participant with moderate/severe ME needed assistance from another person due to cognitive difficulties and several noted that data collection was difficult when they were experiencing post-exertional malaise.All participants were positive about taking part in the testing, even those with severe ME.One subject stated: It's been a real privilege to be part of the study … thank you for allowing me to participate (especially at severe).
All participants indicated that they would volunteer for the study again and would recommend it to others.They suggested the use of videos and more instructions about how to take the measurements for use when collecting data independently.Participants also requested a copy of their individual results which could be made available to their general practitioners.Some subjects said that they would like to monitor their physiology in more detail and suggested more frequent measurements for longer periods (up to one month), while others suggested additional measurements including body temperature, menstrual cycle, blood sugar levels and a food diary.

Discussion
The findings of this study indicated that it is feasible and acceptable for individuals with ME to assess physiological parameters during everyday activity using portable equipment and protocols that are traditionally confined to a clinic or laboratory.Furthermore, our findings revealed apparent physiological abnormalities during these activities that may be useful for understanding real-time biological dynamics in ME that perhaps could inform diagnosis and assessment as well as development of novel therapies to manage symptoms.Feasibility was considered in terms of the ability to recruit participants, acceptability of the testing protocol, fidelity to procedures, and sensitivity of outcome measures.
Recruitment was exceptionally rapid and completed within 24 h.This demonstrates that people with ME, even those who are severely affected, are keen to be involved in research.Social media was effective for recruitment obviating the need to use more traditional approaches such as referrals from clinical services and primary care.However, our apparent success meant that some people, especially those with severe ME, did not have time to read and process the participant information and respond before the study closed.It may be helpful to extend recruitment periods specifically for people with severe illness to allow this cohort greater opportunity to respond.
The testing protocol was feasible but challenging for both participants and the researcher.Because the researcher visited each participant's home in the morning timed to arrive shortly after waking, recruitment was geographically limited to a reasonable travel distance, and to people who could arrange for someone to let the researcher in (or for the researcher to let herself in).Timing a repeat visit to collect equipment also impacted on the length of time that independent testing could continue.The optimal duration of data collection needs to be established to minimise assessment burden for both participants and researchers.However, testing in participants' homes greatly improved the equity of access to research experienced by many, particularly those with severe ME, who often cannot travel to attend a laboratory or clinic.This should be considered a useful approach for ME research to collect real-time biological data, although at greater cost in terms of researchers' time and travel and the amount of equipment needed.
An important challenge in developing a feasible testing protocol was to minimise potential harms to participants from triggering PEM while they performed sufficient activity to detect abnormalities, if present.As expected, the severity of participants' illness influenced the activities completed, with the more severely affected completing fewer and less demanding activities.All participants demonstrated some physiological abnormalities during testing, which supports preliminary success to further develop this home-based assessment approach.However, there was no single activity that all participants could complete that caused roughly similar abnormal response(s) in all participants.Thus, future work to produce a standardised testing protocol that yields broadly informative findings on ME pathophysiology is likely to require additional study with larger samples.Perhaps surprisingly, the activity that caused most participants to exceed their anaerobic threshold was the cognitive task, which all but one participant was able to complete.Thus, cognitive as well as physical activities may show promise for further exploration of potential markers.
All participants were positive about the testing process and would recommend it to others.Although we strove to avoid asking participants to do anything that may trigger more than minimal PEM, some wanted to challenge themselves to see its impact on their objective measurements.This raises several important issues.Firstly, the challenge of fully standardising any testing protocol, and secondly the importance of careful discussion and shared decision-making with study participants regarding testing.Finally, we recommend that research projects with ME patients should offer feedback about individualised results, a tangible benefit to participation that may empower subjects to better understand and manage their condition.For example, during the standing task, 11 participants experienced short-lived symptom exacerbation indicative of orthostatic intolerance, three of whom had no pre-existing diagnosis.Their results were forwarded to their GP with recommendations for further investigation and treatment.This indicates how physiological assessment may have a role in the diagnosis and management of ME and related co-morbidities.Furthermore, our finding that people with ME may exceed their anaerobic thresholds during everyday activity supports the potential for heart rate monitoring to be used as part of an energy management strategy which warrants further investigation.This is the first study as far as we are aware to use a portable cardiopulmonary monitoring system in individuals with ME during everyday activities.More research is needed to assess not only the reliability and validity of the equipment and testing protocols when used outside a laboratory environment, but also the impact of the fluctuating nature of ME on measurement error and sensitivity to change.Furthermore, comparative biobehavioral data from healthy volunteers and individuals with other illness conditions in the home environment are needed to establish whether people with ME are demonstrating specific abnormalities that could inform diagnostic assessments and outcome measures.

Limitations
The study has several limitations.As the first attempt to investigate physiological testing during everyday activity, our sample size was small and based on convenience.Although we were able to increase volunteer access to a research study, particularly for the more severely affected, recruitment was limited to those within a specific geographical area who were willing/able to arrange for the researcher to access their property.
A common potential confound in clinical exercise research is the physiological effect of medication.Participants in this study were taking a number of medications which may alter metabolic pathways affecting responses to activities.However, this is a first step in the exploration of objective testing during everyday activities and an important finding was the putatively altered metabolic response during everyday activities.

Conclusions
Our home-based physiological testing conducted during everyday activity revealed potential abnormalities that varied between individuals that have previously not been reported.The protocol was found to be feasible and acceptable for people with mild to severe ME.This supports the need for further work to develop a reliable, valid and accurate testing protocol to assess, diagnose and monitor ME.Given the variability in participants' responses to activities, the testing protocols will need to be individualised to participants' level of ability and preferences.All subjects demonstrated altered metabolic responses during a number of activities and exceeded their anaerobic threshold during at least one of the everyday activities tested.Our findings suggest the need to further explore metabolism during everyday tasks in larger samples and more specifically to explore the potential for using heart rate as a surrogate marker to monitor metabolic effort.

Table 1 .
Levels of severity and common co-morbidities of participants.