Responsiveness and Minimal Important Change of the Haemophilia Activities List in Patients With Inherited Bleeding Disorders Authors: Johan Blokzijl, Isolde A R Kuijlaars, Martijn F Pisters, Cindy Veenhof, Janjaap van der Net, Roger E G Schutgens, Kathelijn Fischer, Merel A Timmer Published: 2026 ABSTRACT ABSTRACTIntroductionThe Haemophilia Activities List (HAL) is a disease‐specific patient‐reported outcome measure (PROM) that is widely used in both healthcare and research in people with inherited bleeding disorders (PWBD), but information on responsiveness and minimal important change (MIC) is lacking.AimThe aim of this study is to determine the responsiveness and MIC of the HAL to measure self‐reported limitations in activities in PWBD.MethodsPWBD on the waiting list for orthopaedic surgery were invited to participate. Participants completed the HAL, reference PROMs regarding pain, physical functioning and ADL activities at baseline, 3 months and 12 months after surgery. The general perceived effect was completed at 3 and 12 months after surgery.ResultsA total of 32 surgeries performed in 28 participants were included. Three months after surgery the change of the HAL sum score showed moderate to strong correlation with the change on the physical functioning PROMs (r: 0.56) and the change on the ADL PROMs (r: 0.65) and weak correlation with pain scores. The HAL was well able to discriminate participants that improved in self‐reported limitations in activities and participation from those that did not improve (AUC: 0.86 (95%CI: 0.71–1.0)). The MICimprovement was −2.3 (95%CI: −6.0;1.4) and the MICdeterioriation −4.9 (95%CI: −8.4;−2.5).ConclusionAs the MIC is smaller than the smallest detectable change (SDC) of the HAL, it can be assumed that if a person has improved or deteriorated more than the known SDC of 10.2 (previous research), they are likely to experience a meaningful change over time.Keywords: activities of daily living, haemophilia, minimal important change (MIC), patient‐reported outcomes measures (PROMs), responsiveness Introduction Inherited bleeding disorders, including haemophilia, von Willebrand's Disease and other rare factor deficiencies, to prolonged bleeding after surgery, increased susceptibility to bruising, and an increased risk of bleeding into joints and muscles [1]. Despite advances in treatment, such as prophylactic factor replacement and alternative non‐replacement approaches, people with inherited bleeding disorders (PwBD) still experience bleeding episodes. These bleeds in joints and muscles contribute to acute pain and functional limitations. Repeated episodes of joint bleeding eventually lead to the development of chronic joint disease, including haemophilic arthropathy [1, 2]. As a result, people with haemophilic arthropathy face limitations in their daily activities and ability to participate in society [3, 4]. To enable healthcare providers to assess the limitations in activities in adult PwBD the Haemophilia Activities List (HAL) was developed [5]. The HAL is a disease specific patient‐reported outcome measure (PROM) which measures self‐reported limitations in activities and participation [5, 6]. A systematic literature review in 2018 indicated the measurement properties of the HAL together with the Functional Independence Score in Haemophilia (FISH) to be the most satisfactory to assess limitations in daily life activities [7]. The PROM is used in both clinical care and research and is available in multiple languages [5, 8, 9, 10]. In 2021, the original HAL was shortened (HALshort) from 42 to 18 questions to reduce patient burden [11]. Responsiveness, minimal important change (MIC) and smallest detectable change (SDC) are important properties of PROMs to evaluate interventions. Responsiveness is defined as ‘the ability of a Health‐Related Patient‐Reported Outcome instrument to detect change over time in the construct to be measured’ [12]. MIC is defined as ‘the smallest change in score in the construct to be measured which patients perceive as important’ and is a measure of interpretability [13]. In contrast, SDC refers to measurement error and represents the smallest amount of change that is considered a real, true difference, rather than random error from the instrument itself. Whereas a SDC of 10.2 of the HAL was identified in a previous study, responsiveness and MIC are not yet studied for either the HAL or the HALshort. The aim of this study is to determine the responsiveness and MIC of the HAL to measure self‐reported limitations in activities in adult PwBD, according to the COSMIN recommendations [13]. Methods Study Design A single‐centre longitudinal study was performed in patients with haemophilia or Von Willebrand's disease (VWD) undergoing total hip replacement, total knee replacement or ankle arthrodesis. After providing written informed consent, participants were asked to complete questionnaires at baseline (T0), 3 months after surgery (T1) and 1 year after surgery (T2). The study was approved by the Institutional Review Board of the UMC Utrecht (IRB number: 17–752). Study Population Individuals with haemophilia or VWD with planned total hip replacement, total knee replacement or ankle arthrodesis at the Van Creveldkliniek, University Medical Centre Utrecht between June 2017 and March 2024, were invited to participate in the current study. Individuals with multiple joint procedures and/or revision surgery were also eligible. Outcome Measures HAL The HAL contains 42 items across 7 domains (lying down/sitting/kneeling/standing, functions of the legs, functions of the arms, use of transportation, self‐care, household tasks and leisure activities and sports). Participants scored the items on a 6‐point Likert scale (‘impossible’, ‘always’, ‘usually’, ‘sometimes’, ‘almost never’, ‘never’) and the option to reply ‘not applicable’. A summary score as well as component scores (upper extremity, basic lower extremity and complex lower extremity) were calculated. The scores were converted to a normalized domain score and a HALshort score was calculated using the method described by Kuijlaars et al. [11]. The scores range from 0 to 100, where higher scores represent a better functional status. Anchor and Reference PROMs The general perceived effect questionnaire was used as an anchor questionnaire. Questionnaires measuring physical functioning, activities of daily living and pain were used as reference PROMs. General Perceived Effect At 3 and 12 months after surgery participants were asked to rate their perceived change through a General Perceived effect question. Perceived change was rated on a seven‐point scale (‘very much worst’, ‘much worse’, ‘a little worse’, ‘no change’, ‘a little better’, ‘much better’, ‘very much better’). Physical Functioning Knee and hip functioning are measured using the Knee disability and Osteoarthritis Outcome Score‐Physical Function Short‐form (KOOS‐PS) or Hip disability and Osteoarthritis Outcome Score‐Physical Function Short‐form (HOOS‐PS). The PROMs consist of seven and five items respectively and the scores range from 0 to 100, with zero representing no difficulty. Ankle physical functioning was assessed with the Ankle Osteoarthritis Scale (AOS). The AOS is a validated PROM for ankle osteoarthritis, assessing the level of pain and disability on a 100 mm line for nine conditions each. Items’ responses in the domain were averaged to calculate a domain score. The overall score ranges from 0 (indicating best ankle function) to 100 (worst ankle function). Activities of Daily Living (ADL) ADL functioning in patient with knee or hip surgery is measured using the Oxford Knee Score (OKS) or Oxford Hip Score (OHS). Both PROMs consist of 12 questions on pain and function in daily activities. The overall score ranges from 0 (most severe symptoms) to 48 (least symptoms). The ADL functioning in patient who underwent ankle surgery was assessed with the Manchester Oxford Foot Questionnaire (MOxFQ). The MOxFQ assesses how foot problems impair health‐related quality of life. The MOxFQ is a 16‐item instrument answered on a five‐point Likert scale (each item is scored from 0 to 4, with 4 denoting ‘most severe’). Scores for each item are summed to form three separate subscales representing walking/standing problems (seven items), foot pain (five items), and issues related to social interaction (four items), besides the sum score. To allow comparison of instruments with different scales, scores were normalized between 0 and 100 with higher scores reflecting better functioning. Pain Pain was assessed in all participants with the Numeric Pain Rating Scale (NPRS) during rest and movement. Patients scored the level of pain on a Likert scale from 0 (no pain) to 10 (very severe pain). Statistical Analyses Participant Characteristics Participant characteristics were presented as proportions or median with interquartile range (IQR). Descriptive analyses (mean and standard deviation) were presented for the scores at baseline and follow‐up. Responsiveness Responsiveness was assessed in two ways. First, a‐priori determined hypotheses based on previous research about relationships between the changes scores (Δ) of the HAL/HALshort and reference PROMs were tested [14]. Correlations were calculated for ankle arthrodesis, total knee replacement and total hip replacement together. Correlations of 0.80–1 were considered as ‘very strong’, 0.60–0.79 as ‘strong’, 0.40–0.59 as ‘moderate’ and 0.20–0.39 as ‘weak’ [15]. Correlations between ΔHAL/HALshort and Δreference PROMs of >0.6 were hypothesized. Second, the perceived change after treatment was used to determine the area under the curve (AUC) of a receiver operating characteristic (ROC) curve. The ROC curve was plotted by dividing patients into two groups: ‘improved’ and ‘unchanged’ based on the responses on the general perceived effect questionnaire. An AUC of at least 0.70 is recommended for sufficient responsiveness [14]. MIC The MIC was defined as the smallest change of the HAL/HAL short score above which the patient perceived themselves to have changed importantly over time [16]. The response to the general perceived effect question was used as an anchor to determine whether a person perceived themselves to have changed. Both the ROC method and predictive modelling were used to calculate the MIC. The predictive modelling method was presented as the results from this method are more accurate and the methods allow researchers to adjust the results if the percentage of changed patients is not 50%. A minimum correlation between the anchor and the change in HAL scores of 0.3 is required to ensure the validity of the results. If the percentage of improved patients is not 50%, the adjusted MIC is presented. Bootstrapping with 1000 samples was used to determine the 95% confidence interval. Complete case analysis was performed for every analysis. Data analysis was performed using RStudio (version 4.3.1). Results Participant Characteristics and Outcome Measures In the current study 32 surgeries performed in 28 participants were included, detailed information on participant inclusion is presented in appendix 1. Participants had a median age of 59.5 [IQR: 14] at baseline. Majority of the participants were male and had severe haemophilia A. A complete overview of participant characteristics is shown in Table 1. Surgeries included 14 total knee replacements (of which 3 revision surgeries), 8 total hip replacements (of which 2 revision surgeries) 5 ankle arthrodeses and 5 multiple joint procedures. Multiple joint procedures included 1 bilateral knee replacement (both knees included as separate trajectories), 1 bilateral hip replacement (only data of the right hip available) and 3 combinations of total knee replacement and ankle arthrodesis (of 1 patient data of the ankle and knee were included as separate trajectories, of 2 patients only ankle data were available). TABLE 1Participant characteristics (n = 28).Median [Q1–Q3] N (%) Age (years) 59.5 [51.0–65.0] Gender Male25 (89.3%)Female3 (10.7%) HJHS*39.0 [29.0–49.3] Disease Haemophilia A21 (75.0%)Haemophilia B3 (10.7%)Von Willebrand Disease4 (14.3%) Haemophilia type (n = 24) Severe (<1%)19 (79.2%)Moderate (1%–5%)4 (16.7%)Mild (6%–40%)1 (4.2%) Von Willebrand Disease type (n = 4) Type 11 (25.0%) Type 2 Type 3 2 (50.0%) 1 (25.0%) Note: *Available in 19 of the 28 participants. Participant characteristics (n = 28). Type 2 Type 3 2 (50.0%) 1 (25.0%) Note: *Available in 19 of the 28 participants. The sum score of the HAL changed from 42.95 ± 14.7 at baseline to 43.26 ± 17.87 (p = 0.87) after 3 months and 54.80 ± 20.95 (p < 0.01) after 12 months. The HALshort score changed from 39.02 ± 14.64 at baseline to 40.18 ± 18.16 (p = 0.60) after 3 months and 51.66 ± 22.22 after 12 months (p < 0.01). The proportion of participants that indicated they had improved on the anchor question was 18/31 (58.0%) after 3 months and 24/30 (80.0%) after 12 months. A complete overview of change scores is presented in Table 2. TABLE 2Outcome measurements at different time frames.Baseline (T0)3 months (T1)12 months (T2)Change T0–T1Change T0–T2 N Mean ± sd N Mean ± sd N Mean ± sdMean ± sdMean ± sd HAL Sum score3241.62 ± 14.03241.52 ± 17.073052.23 ± 21.150.31 ± 11.3212.72 ± 12.95 c Short score3239.02 ± 14.643240.18 ± 18.153051.6 ± 22.221.16 ± 12.3213.41 ± 14.02 c Upper extremity3265.79 ± 24.253265.49 ± 24.983070.22 ± 26.91−0.30 ± 12.384.12 ± 12.19 c Lower extremity—Basic3230.00 ± 18.533234.19 ± 23.063054.27 ± 24.605.16 ± 17.7626.27 ± 20.68 c Lower extremity—Complex3219.81 ± 13.853220.90 ± 15.203027.58 ± 21.191.09 ± 10.748.89 ± 13.25 c NPRS Rest293.80 ± 2.50311.35 ± 1.20290.59 ± 1.40−2.45 ± 2.78 c −3.3 ± 2.57 c During load295.97 ± 2.37311.93 ± 1.81290.86 ± 0.95−4.10 ± 2.81 c −5.30 ± 2.30 c ADL—Normalized a 2937.54 ± 15.593056.06 ± 19.902976.04 ± 18.2119.55 ± 20.75 c 37.52 ± 18.13 c Physical functioning b 2955.90 ± 14.872943.67 ± 17.502829.73 ± 18.28−12.20 ± 20.57 c −28.32 ± 21.19 c Abbreviations: ADL, Activities of daily living; HAL, Haemophilia Activities List; NPRS, Numeric Pain Rating Scale. a PROMs included were OKS/OHS/MOXFQ. normalized from 0–100 with higher scores representing better functioning, as questionnaires had different scales. b PROMs included were KOOS/HOOS/AOS and a lower score indicates better function. c p < 0.05. Outcome measurements at different time frames. Abbreviations: ADL, Activities of daily living; HAL, Haemophilia Activities List; NPRS, Numeric Pain Rating Scale. PROMs included were OKS/OHS/MOXFQ. normalized from 0–100 with higher scores representing better functioning, as questionnaires had different scales. PROMs included were KOOS/HOOS/AOS and a lower score indicates better function. p < 0.05. Responsiveness The correlation coefficient calculated on the delta of the HAL total score, HALshort score and lower extremity subscales between baseline and 3 months showed strong correlations with the normalized ADL PROMs. The physical functioning PROMs showed a moderate correlation with most of the HAL scores, and a strong correlation with the lower extremity complex subscale. The anchor question showed a strong correlation with the HAL sum score, HALshort score and lower extremity basic subscale and a moderate correlation with the delta of the lower extremity complex subscale. Change in pain scores from baseline to 3 months was not correlated with the HAL scores. A complete overview of the correlations is presented in Table 3. TABLE 3Correlation coefficients between difference scores in HAL subscales and reference PROMs between (a) baseline (T0) and 3 months (T1).(b) baseline (T0) and 12 months (T2).HAL sum scoreHAL short scoreHAL lower extremity—BasicHAL lower extremity—Complex Correlation between difference scores at baseline (T0) and 3 months (T1) Physical functioning b −0.57 [−0.78; –0.24] −0.59 [−0.79; −0.28] −0.43 [−0.70; −0.05] −0.62 [−0.81; −0.32] ADL—Normalized a 0.65 [0.36; 0.82] 0.63 [0.34; 0.81] 0.60 [0.29; 0.80] 0.59 [0.28; 0.79] NPRSrest−0.30 [−0.60; 0.07]−0.37 [−0.65; 0]0.01 [−0.38; 0.36]−0.23 [−0.55; 0.15]load−0.27 [−0.58; 0.11]−0.15 [−0.49; 0.23]−0.16 [−0.50; 0.22]−0.26 [−0.57; 0.12]Anchor (T1) 0.72 [0.58; 0.86] 0.63 [0.35; 0.80] 0.59 [0.30; 0.79] 0.40 [0.06; 0.66] Correlation between difference scores at baseline (T0) and 12 months (T2) Physical functioning−0.29 [−0.60; 0.11]−0.41 [−0.69; 0.03]−0.19 [−0.54; 0.21] −0.49 [−0.74; −0.13] ADL—Normalized0.31 [−0.09; 0.62] 0.45 [0.08; 0.71] 0.34 [−0.05; 0.64] 0.47 [0.10; 0.72] NPRSrest−0.16 [−0.51; 0.23]−0.25 [−0.57; 0.15]0.02 [−0.36; 0.40]−0.27 [−0.59; 0.12]load−0.28 [−0.60; 0.11]−0.25 [−0,58; 0.14]<0.01 [−0.38; 0.38]−0.25 [−0.58; 0.14]Anchor (T2) 0.44 [0.09; 0.69] 0.47 [0.12; 0.71] 0.54 [0.22; 0.76] 0.39 [0.03; 0.66] Note: All correlation coefficients are calculated using a Pearson correlation and 95% CI, Correlation that are significant (p < 0.05) are presented in bold.Abbreviations: HAL, Haemophilia Activities List; NPRS, Numeric Pain Rating Scale; ADL, Activities of daily living. a PROMs included were OKS/OHS/MOXFQ. normalized from 0–100 with higher scores representing better functioning, as questionnaires had different scales. b PROMs included were KOOS/HOOS/AOS and a lower score indicates better function. c PROMs included were OKS/OHS/MOXFQ. normalized from 0–100 with higher scores representing better functioning, as PROMs had different scales. Correlation coefficients between difference scores in HAL subscales and reference PROMs between (a) baseline (T0) and 3 months (T1).(b) baseline (T0) and 12 months (T2). Note: All correlation coefficients are calculated using a Pearson correlation and 95% CI, Correlation that are significant (p < 0.05) are presented in bold. Abbreviations: HAL, Haemophilia Activities List; NPRS, Numeric Pain Rating Scale; ADL, Activities of daily living. a PROMs included were OKS/OHS/MOXFQ. normalized from 0–100 with higher scores representing better functioning, as questionnaires had different scales. b PROMs included were KOOS/HOOS/AOS and a lower score indicates better function. c PROMs included were OKS/OHS/MOXFQ. normalized from 0–100 with higher scores representing better functioning, as PROMs had different scales. The ROC curve of the change scores after 3 months and after 12 months is presented for both the HAL sum score and HALshort score. The area under the curve (AUC) of the HAL sum score after 3 months was 0.86 (95%CI: 0.71–1.0) and after 12 months was 0.89 (95%CI: 0.75–1.0). The AUC of the HALshort score after 3 months was 0.86 (95%CI: 0.72–1.0) and after 12 months was 0.87 (95%CI: 0.73–1.0). ROC curves are presented in Figure 1A and 1B. These AUC's indicate that both the HAL and HALshort are able to distinguish participants that improved from those that did not improve 3 months and 12 months after surgery. ROC curve of change in HAL sum (A) and HAL short (B) from baseline (T0) to 3 months (T1) and 12 months (T2) after surgery. MIC The median change of the HAL sum score between baseline and 3 months after surgery of patients that improved based on the anchor question was 5.0 [IQR: 7.39]. Patients that did not improve showed a median change in HALsum score of −9.5 [IQR: 9.59]. The median change in the HALshort score was 6.0 [IQR: 13.9] for the improved group and −8.2 [IQR: 14.6] for the unimproved group. A complete overview of the change scores is presented in Table 4. The MICimproved based on adjusted predictive modelling was −2.3 (95%CI: −6.0;1.4) for the HAL score and −0.8 (95%CI: −4.7;2.5) for the HALshort score 3 months after surgery. Twelve months after surgery the MICimproved for the HAL and HALshort were respectively 4.3 (95%CI: −2.1;7.6) and 3.9 (95%CI: −1.6;7.3). The MICdeterioriation at T1 was −4.9 (95%CI: −8.4;‐2.5) for the HAL score and −2.9 (95%CI: −7.4;0.5) for the HALshort score. At T2 the MICdeterioriation was −3.7 (95%CI: −5.3; 7.9) for the HAL score and 3.0 (95%CI: −14.3;7.6) for the HALshort score. Results of the ROC method and the unadjusted MIC scores did not provide any major differences. Results of MIC analysis are provided in appendix 2. TABLE 4Mean change of HAL scores of the improved and not improved group based on the general perceived effect.HAL sumHALshort HAL sumHALshort N Median [IQR]Median [IQR] N Median [IQR]Median [IQR] 3 months after surgery (T1) Very much worse2−15.7 [3.84]−11.5 [2.62]13−9.5 [9.59]−8.2 [14.60]Much worse4−10.5 [3.88]−7.1 [13.6]A little worse4−5.4 [4.5]−6.4 [6.49]No change32.8 [14.5]2.4 [8.47]A little better35.7 [8.33]4.4 [14.3]185.0 [9.26]6.0 [14.9]Much better132.5 [7.28]6.0 [7.45]Very much better221.8 [0.97]21.3 [1.78] 12 months after surgery (T2) Very much worse17.17.261.2 [10.0]3.2 [7.43]Much worse1−1.53.3A little worse2−1.7 [6.42]−7.9 [10.9]No change2−1.5 [5.38]−0.5 [4.9]A little better14.311.12415.8 [15.4]Much better1615.8 [11.3]14.6 [16.9]14.6 [21.4]Very much better716.7 [17.3]20.0 [21.4] Mean change of HAL scores of the improved and not improved group based on the general perceived effect. Discussion In the current study we evaluated the responsiveness and MIC of the HAL and HALshort PROM in adult patients with haemophilia and VWD who had lower extremity orthopaedic surgery. Three months after surgery the HAL and HALshort change scores showed moderate to strong correlation with the physical functioning PROMs and the ADL PROMs and weak correlation with pain scores. ROC curves showed that the HAL and HALshort are able to distinguish participants that improved from those that did not improve. The MICimprovement and MICdeterioration were both smaller than the known SDC of the HAL. Moderate to strong correlations in delta HAL scores between baseline and 3 months post‐surgery with reference PROMs are in agreement with good to excellent correlations between the HAL and the Dutch Arthritis Impact Measurements Scales‐2 (Dutch‐AIMS2) and the Impact on Participation and Autonomy questionnaire (IPA) in a previous cross‐sectional validity study [6]. Correlations between the delta HAL scores between baseline and 12 months post‐surgery with reference PROMs were lower than hypothesized. This could be explained by the slightly different construct measured by the different PROMs. The physical functioning and ADL PROMs focus on one specific (operated) joint, whereas the HAL evaluates any limitations experienced in the last 6 weeks due to haemophilia. Shortly after surgery the operated joint is expected to impact physical functioning primarily. One year post‐surgery physical functioning is likely more influenced by other affected joints, as multi‐joint arthropathy is common in haemophilia [17]. The limited correlation between HAL and pain changes scores was unexpected, as previous studies in osteoarthritis showed that worsening knee pain is associated with worsening self‐reported physical functioning [18, 19, 20]. PwBD's lifelong pain experiences in multiple joints might influence the extent to which pain affects their perceived ability to perform activities, different from osteoarthritis [21]. In this study, even a minor decrease in HAL was identified as meaningful improvement. The MICimproved (−2.3) and the MICdeteriorated (−4.9) are both smaller than the known SDC of 10.2 [22]. The HAL is therefore not able to identify minor, but clinical relevant, changes in physical functioning. A change in HAL score larger than 10.2 can be interpreted as a real and clinically relevant change. An explanation for the lower MIC compared to the SDC could be the diluting effect of the questions related to the upper extremity included in the HAL, which are not expected to change after the lower extremity orthopaedic surgeries. MIC values for other PROMs for physical functioning are not yet available in PWH. PROMIS‐PF PROMs in related musculoskeletal populations show comparable MIC 0.1–12 and SDC 6.9–10.6 values [16, 23, 24]. A strength of the study is the use of two methods calculating the MIC, as previous studies have described large differences in the MIC values depending on the method chosen [26]. Providing results of both methods will inform clinicians and researchers in the most transparent way. A limitation of the current study is the limited sample size, which is lower than the recommended number of 100 participants [16], affecting the precision of the results. Moreover, the included population is relatively old and has advanced arthropathy of multiple joints, leading to substantial limitations in activities. As a study population receiving orthopaedic surgery is most suitable to determine MIC, a more severely affected population is inevitable. However, this may limit the generalizability to younger and less affected PwBD. For clinical care and research purposes changes in HAL scores exceeding the SDC of 10.2 points should be interpreted as real and clinical relevant changes within groups. MIC values should only be used to interpret within group changes, not to interpret between group differences [16]. Furthermore, the value does not take into account costs, risks and inconveniences of an intervention and therefore cannot be used as a threshold for the smallest worthwhile effect [27]. In conclusion, the HAL has good responsiveness to detect change in self‐reported limitations in activities over time in PwBD. The MIC obtained for both improvement and deterioration is smaller than the previously available SDC. An improvement or deterioration exceeding the SDC of 10.2, is also a meaningful change over time. Author Contributions Isolde A.R. Kuijlaars, Janjaap van der Net, Kathelijn Fischer and Roger E.G. Schutgens were involved in the design of the study. Johan Blokzijl, Isolde A.R. Kuijlaars and Merel A. Timmer collected the data. Johan Blokzijl, Martijn F. Pisters and Merel A. Timmer analysed the data and wrote the initial version of the report. All authors were involved in data interpretation and review of the paper. All authors approved the final version of the paper. Ethics Statement The study was approved by the Institutional Review Board of the UMC Utrecht (IRB number: 17–752) and performed according to the Declaration of Helsinki and guidelines for Good Clinical Practice. Consent Written informed consent was obtained from all participants before inclusion Conflicts of Interest The institution of REG Schutgens has received speaker's fees and/or research grants from Bayer, CSL Behring, Hemab, Novartis, NovoNordisk, Octapharma, Roche, Sobi and Takeda. The institution of MA Timmer has received speaker's fees and/or research grants from Sobi and NovoNordisk. The institution of K Fischer has received speaker's fees and/or research grants from Bayer, Baxter/Shire, Sobi, CSL Behring, Novo Nordisk, Freeline, Roche. J Blokzijl, I Kuijlaars, J Net, MF Pisters en C Veenhof declare no conflicts of interest. Funding The authors have nothing to report. Supporting information Supporting Information: hae70327‐sup‐0001‐Appendix‐1.docx Supporting Information: hae70327‐sup‐0002‐Appendix‐2.docx Acknowledgements The authors have nothing to report. Data Availability Statement The data that support the findings of this study are available on request from the corresponding author. References Associated Data Supplementary Materials Supporting Information: hae70327‐sup‐0001‐Appendix‐1.docx Supporting Information: hae70327‐sup‐0002‐Appendix‐2.docx Data Availability Statement The data that support the findings of this study are available on request from the corresponding author.