A cross‐sectional follow‐up study of physical activity in adults with moderate and severe haemophilia

To conduct a cross‐sectional follow‐up assessment of physical activity (PA) in people with moderate and severe haemophilia (PwMSH) from the Irish Personalised Approach to the Treatment of Haemophilia (iPATH) study.


INTRODUCTION
Regular physical activity (PA) is associated with multiple health benefits, including a reduced risk of cardiometabolic diseases, certain cancers, physical disability, depression, and anxiety. 1 The World Health Organisation recommends at least 150-minutes of moderate intensity aerobic PA per week for substantial health benefits. 2 Recent guidelines advise that even light intensity PA has relative health benefits for people with chronic health conditions. 2 The World Federation of Hemophilia recommends regular PA and exercise as important components of haemophilia treatment and management. 3 Unfortunately, achieving regular PA can be difficult for many people with moderate and severe haemophilia (PwMSH) due to pain and physical disability associated with bleeds and resultant haemophilic arthropathy.
Haemophilic arthropathy is associated with reduced physical functioning and PA, although multiple factors may contribute towards variation in these physical domains, including age, clinical phenotype, treatment regimen and comorbidities, as well as socio-economic, cultural, environmental, and personality or behavioural influences. [4][5][6][7] The first phase of research conducted as part of the Irish Personalised Approach to the Treatment of Haemophilia (iPATH) study, found that adult PwMSH, who were predominantly treated with secondary or tertiary prophylaxis, were less physically active than the general population in Ireland. 8 This has also been echoed in other studies using similar methodologies to assess PA. 5,9 The iPATH cohort also demonstrated significant arthropathy and high levels of associated chronic pain and functional disability, which may be potential influencers of PA in PwMSH; although the relationship between chronic pain and its influence on PA participation is not fully understood. 10 Additional psychosocial barriers to PA, such as a lack of time and motivation for PA, were also identified. 11 Considering the findings of cross-sectional studies may be inherently limited by the sole opportunity for outcome assessment at only one time point, 12 the second phase of this study was to conduct a follow-up assessment of PA in this cohort, to determine the consistency of the original findings. Participation in the original assessment may have influenced self-awareness of PA amongst the iPATH cohort, therefore an examination of PA self-awareness amongst the group was also carried out. It was anticipated that the overall findings would be further useful for informing personalised PA and health interventions for adults with haemophilia in Ireland.
The ability to conduct the follow-up assessment of PA in the iPATH cohort was significantly impacted by the Covid-19 pandemic during the interim assessment period, therefore an additional aim was included to determine the impact of the pandemic on PA and other physical health variables such as pain, mobility and function. Major restrictions to daily life, commonly known as 'lockdown' , came into effect in Ireland on the 27 March 2020. Citizens were advised to 'stay at home' except for essential purposes, and to exercise only within a two kilometre (km) radius of their home. 13 Non-essential healthcare procedures were postponed, and access to clinical sites was restricted to essential frontline healthcare staff, and patients in direct care. 14 Many hospital services and clinics were adapted to telephone and virtual consultations, including the Irish haemophilia healthcare service at the National Coagulation Centre, St. James's Hospital, Dublin. 15 Ultimately, it was not feasible to conduct this follow-up assessment until the third lockdown in Ireland was over.
The primary aims of this study were: 1. To conduct a follow-up assessment of PA in participants from the iPATH study in order to determine whether their PA levels had changed since the original assessment.

Study design and setting
Recruitment and data collection for this cross-sectional follow-up study took place between June-December 2021, as lockdown restrictions were gradually eased from April 2021 in Ireland ( Figure 1).
All participants recruited for this follow-up study had taken part in the original iPATH study PA assessment, which took place between March 2018-March 2020. 8

Physical activity
PA was measured prospectively over one week using the ActiGraph wGT3X-BT accelerometer (ActiGraph Corp, Pensacola, Florida, USA) attached to an elasticated belt over the right hip. The ActiGraph has been deemed both a valid and reliable method of PA measurement in laboratory and field settings. [16][17][18] Participants were instructed to remove the monitor during sleep and water-based activities. They had previously been instructed in person about how to wear the accelerometer during the original assessment. 8 They additionally received an information document about correct use at follow-up.
A diary was also provided to record hours of daily wear-time for cross-validation with raw accelerometer output. The ActiLife software (Version 6.13.4) was used to download and process raw data. Raw data were analysed in one-minute epochs. Wear-time validation algorithms by Choi et al. were applied 19 and PA intensity was classified as light, moderate, vigorous and combined moderate-vigorous PA (MVPA) according to cut-points established by Troiano et al. 20 Bouts of MVPA sustained for ≥10-minutes (i.e. Freedson bouts) were also analysed. 21 Data were categorised according to the achievement of PA guidelines of at least 150-minutes of moderate PA per week. 2 Accelerometer data was compared within the group with their previously collected data. 8

Longitudinal follow-up survey
PA self-awareness and the impact of the Covid-19 pandemic on PA were determined using a retrospective survey designed by the research team. In the survey, participants were asked about selfperceived awareness of their own PA levels, a desire to change PA behaviour, and their knowledge of the PA guidelines. They were also asked to rate any self-perceived changes in PA compared to normal throughout the various phases of the pandemic ( Figure 1)

Recruitment flow and demographics
Of the original cohort (n = 54), 49 were contactable for follow-up and 33 were enrolled, representing a 67.3% follow-up rate. Full details of recruitment flow are presented in Figure 2.

Physical activity
Accelerometer wear-time was similar between assessments (Table S1).
There was no significant difference within the group between the  (Table S2).

Physical activity self-awareness
At follow-up, 76.7% of participants reported to be more aware of their PA behaviour. Additionally, 66.7% reported a desire to become more physically active. New exercise programmes or sports were commenced by 30.0% since the original assessment, and involved a variety of types of PA, including sea swimming and online exercise classes.
Knowledge of the PA guidelines was claimed by 36.7% (n = 11). One participant cited the guidelines correctly in full, whilst the remaining 10 partially cited the guidelines correctly.

Impact of the Covid-19 pandemic on physical activity
The impact of the Covid-19 pandemic on self-reported PA is presented in Figure 4. The majority of participants reported no difference in PA with mobility (20.0%) during lockdown, whilst the majority denied any differences in these domains (Table S3).
Lastly, participants were asked about their concerns for PA and health beyond the pandemic over the subsequent year. The most common concern reported was pain (33.3%). Access to various exercise resources (30.0%), as well as desires to increase and maintain PA

DISCUSSION
The aim of this study was to conduct a follow-up assessment of PA amongst the iPATH study cohort. There were no significant dif-ferences in objectively measured PA between the two assessment time-points, although the number of participants who achieved PA guidelines slightly increased at follow-up. PA self-awareness had increased in the majority of respondents since the original assessment, and a desire to become more active was reported by most. The majority of participants reported no differences or lower levels of PA compared to normal during three waves of lockdown restrictions, however by the third phase of eased restrictions, the majority reported increased PA. A variety of concerns related to PA and health beyond the pandemic, when restrictions were eased, were also highlighted by participants, especially in relation to pain and access to exercise resources.
The duration of time spent in all parameters of objectively measured PA was not significantly different in this cohort at follow-up.
This was despite reports of increased self-awareness of PA and desires to become more physically active by the majority of the group. These findings support the evidence regarding limited PA amongst adult PwMSH from the original assessment, 8 and demonstrate that limited PA is still a concern amongst this cohort approximately three years PA engagement was positively correlated with better self-efficacy to participate in PA amongst adults with haemophilia in a study by Goto et al., 24 and a lack of self-awareness of health risk behaviours, including PA, has been associated with a reduced perception of a need to change such behaviours. 25 27 and PA levels decreased thereafter during the second wave of the pandemic. 28 It therefore appears that adult PwMSH may have been less physically active than the general Irish population during the initial restrictions, but levels may have been similar as the pandemic progressed. This is not overly surprising considering medically vulnerable populations were at a greater risk of serious illness or fatality from Covid-19, 29 and therefore may have been less likely to go out for exercise in public places, especially in the absence of vaccinations at the beginning of the pandemic. Decreased sports participation compared to the general population during the pandemic has also been reported in a Dutch cohort of adults with haemophilia. 30 Encouragingly in the present cohort, increased PA was reported by the majority after the third lockdown was over, although a number of ongoing concerns for PA beyond the pandemic were also highlighted.
Pain was the most commonly reported concern for PA beyond the pandemic, including concerns regarding haemophilic arthropathy (∼43.3%). Pain associated with haemophilia is complex and multifaceted, including presentations of both acute pain arising from bleeding episodes and musculoskeletal injuries, as well as the development of chronic pain due to haemophilic arthropathy. 31  and management of people with haemophilia will certainly play a key role in the design and delivery of such interventions.
Despite the best efforts possible with recruitment, study findings may have been limited by the small sample size who agreed to followup, therefore the risk of a type II error in statistical analyses must be acknowledged. Additionally, observation bias may have impacted the assessment of PA using the ActiGraph accelerometer, although it was emphasised to participants to maintain their habitual PA during the study period. Accelerometers are also limited in their potential to differentiate various types and domains of PA engagement, and also cannot measure water-based activity (e.g., swimming, cycling, occupation-related PA, etc.). It was also not feasible to obtain a measurement of PA from a control group during the project time-frame.
Anecdotally, a small number of participants had self-reportedly undergone surgery during the previous three years until follow-up. Direct access to clinical records was restricted to essential clinical staff only during the assessment period due to the pandemic. It was therefore not possible to determine whether any potential history of trauma or surgery affected PA. Furthermore, due to the cross-sectional nature of data collection for this follow-up study, it is not possible to determine cause and effect between exposures and outcomes during the study period. Lastly, the self-reported, retrospective nature of the follow-up survey may have increased a risk of recall bias in the results.

CONCLUSION
Self-reported PA throughout the pandemic was variable amongst the iPATH study cohort, whilst objectively measured PA did not significantly change between the original and follow-up assessments. This was despite reports of an increased self-awareness of PA and a desire to become more physically active amongst the majority of participants. Evidently, the clinical need for personalised, multi-disciplinary PA and health interventions has persisted over a three-year follow-up period in this cohort. Further qualitative research to determine barriers and facilitators to PA, as well as the feasibility and acceptability of personalised PA and health interventions is warranted. Future qualitative studies of this nature should capture the perspectives of patients, their families and caregivers, as well as those of multi-disciplinary haemophilia healthcare providers.