Implementation of the Short-Form 36 (SF-36) in childhood cancer survivors: An analysis on measurement properties across 5 European countries

1 Version 4 January 22, 2024 FULL TITLE 1 Implementation of the Short-Form 36 (SF-36) in childhood cancer survivors: An analysis on 2 measurement properties across 5 European countries 3 SHORT TITLE 4 Measurement properties of the SF-36 in European childhood cancer survivors 5 AUTHORS 6 Katja Baust1, Gabriele Calaminus1,2, Claire Berger 3,4, Julianne Byrne5, Desiree Grabow6, Marry M. van 7 den Heuvel-Eibrink7,8,9, Melanie Kaiser6, Tomas Kepak10, 11, Leontien C.M. Kremer7,8, Jarmila 8 Kruseova12, Rahel Kuonen13,14, Katharina Roser15, Claudia Spix6, Heleen Maurice-Stam7, Claudia E. 9 Kuehni13,16, Grit Sommer13,17 10 11 1Department of Paediatric Haematology and Oncology, University Hospital Bonn, Germany, 12 [email protected], [email protected] 13 2Department of Paediatric Haematology and Oncology, University Hospital Münster, Germany, 14 [email protected] 15 3Department of Paediatric Hematology and Oncology Unit, University Hospital of Saint-Étienne, Saint16 Étienne, France, [email protected] 17 4Jean Monnet University of Saint-Étienne, Sainbiose, U1059, INSERM, F-42023 Saint-Étienne Cedex, 18 France, [email protected] 19 5Boyne Research Institute, 5 Bolton Square, East, Drogheda, Co. Louth A92 RY6K, Ireland, 20 [email protected] 21 6German Childhood Cancer Registry (GCCR), Division of Childhood Cancer Epidemiology (EpiKiK), 22 Institute of Medical Biostatistics, Epidemiology and Informatics, University Medical Center, Mainz, 23 Germany, [email protected], [email protected], [email protected] 24 7Princess Máxima Center for Pediatric Oncology, Utrecht, The Netherlands, m.m.vandenheuvel25 [email protected], [email protected], H.Maurice26 [email protected] 27 8University Medical Center Utrecht, Wilhelmina Children's Hospital, Utrecht, The Netherlands 28 [email protected], [email protected] 29 9ErasmusMC-Sophia Childrens Hospital, Rotterdam, The Netherlands, m.m.vandenheuvel30 [email protected] 31 10University Hospital Brno, Masaryk University, Brno, Czech Republic, kepak[email protected]z 32 11International Clinical Research Center (FNUSA-ICRC), Masaryk University, Brno, Czech Republic, 33 kepak[email protected]z 34 12Dept. of Paediatric Haematology/Oncology, Second Faculty of Medicine, Charles University, Prague, 35 Czech Republic, [email protected] 36 13Childhood Cancer Registry, Institute of Social and Preventive Medicine, University of Bern, Bern, 37 Mittelstrasse 43, 3012 Bern, Switzerland, gritti.somm[email protected], [email protected], 38 claudia.kuehn[email protected] 39 14Department of Pediatrics, Inselspital, Bern University Hospital, University of Bern, Bern, Switzerland, 40 [email protected] 41 15Faculty of Health Sciences and Medicine, University of Lucerne, Lucerne, Switzerland, 42 k[email protected] 43 16Pediatric Oncology, Inselspital, Bern University Hospital, University of Bern, Switzerland 44 17Pediatric Endocrinology, Diabetology and Metabolism, Department of Pediatrics, Inselspital, Bern 45 University Hospital, University of Bern, Switzerland 46 CORRESPONDENCE TO 47 Grit Sommer, PhD 48 Institute of Social and Preventive Medicine 49 University of Bern 50 Mittelstrasse 43 51 3012 Bern 52 Switzerland 53 Tel +41 31 684 33 47 54 E-Mail: gritti.somm[email protected] 55 . CC-BY-NC-ND 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted January 23, 2024. ; https://doi.org/10.1101/2024.01.23.24301414doi: medRxiv preprint NOTE: This preprint reports new research that has not been certified by peer review and should not be used to guide clinical practice. 2 Version 4 January 22, 2024 AUTHOR CONTRIBUTIONS 56 KB and GS contributed both substantially to this work. KB coordinated the study, contributed to the 57 design, implementation, data collection and harmonization, data preparation, and commented on the 58 data analysis. GS prepared the final data set, performed the data analysis. Both KB and GS 59 interpreted the data and wrote the manuscript. GC and CEK supervised the design, implementation 60 and coordination of the study and contributed to data collection and harmonization and analysis. They 61 also supervised data analysis and interpretation and contributed to writing of the manuscript. JB, DG 62 and MK and contributed to the implementation and coordination of the study; they also supervised 63 data collection and harmonization and data maintenance. CB, MMvdHE, TK, LCMK, JK, RK and HMS 64 conducted the study on site in the different countries and contributed to data collection and 65 harmonization. KR and CS interpreted the data and were involved in writing the manuscript. KR also 66 contributed to data analyses, and CS supervised harmonization of the data. All co-authors provided 67 feedback on the manuscript and approved its final version. 68 COUNTS 69 Keywords: SF-36; psychometrics; neoplasms; health status; Europe; survivors of childhood cancer 70 Abstract: 247/250 71 Plain summary (112/200) 72 Manuscript: 3706/4000 (excl. abstract, tables, figure legends, table footnotes, references, funding, 73 ethics declaration, conflict of interest statement, abbreviations, acknowledgements) 74 References: 38 75 Display items: 4 tables / 1 figure 76 Supplemental Tables: 2 77 . CC-BY-NC-ND 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted January 23, 2024. ; https://doi.org/10.1101/2024.01.23.24301414doi: medRxiv preprint 3 Version 4 January 22, 2024 ABSTRACT 78 Purpose: The Short Form-36 (SF-36) is widely used in many research contexts and cultures to 79 assess health-related quality of life (HRQOL). We investigated the measurement properties of the SF80 36 in a large cohort study among childhood cancer survivors living in 5 European countries. 81 Methods: The PanCareLIFE project includes adult survivors of childhood cancer living in the Czech 82 Republic, France, Germany, The Netherlands, and Switzerland. We invited 19,268 survivors aged >18 83 years, and 10,077 (53%) returned questionnaires. Of these, 9,871 (98%) were included in the 84 analyses. We assessed HRQOL with the SF-36 version 1 (V1) or version 2 (V2) and investigated its 85 performance. We analysed data completeness, floor and ceiling effects, item-internal consistency, 86 item-discriminant validity, reliability, and scaling assumptions focusing on countryand version-specific 87 differences. 88 Results: Data completeness was high but differed between countries (90% in the Czech Republic to 89 96% in France). Floor effects were negligible, but ceiling effects in 5 out of 8 SF-36 scales were 90 present (50-70%). V2 had slightly fewer ceiling effects than V1 in 2 scales. Item-internal consistency, 91 item-discriminant validity, and reliability were good. Scaling assumptions were met, with 4 scales 92 representing strong mental health content, and the other 4 representing strong physical health 93 content. 94 Conclusion: The SF-36 (V1 and V2) provided robust measurement properties among childhood 95 cancer survivors, and differences in study design across the 5 European countries did not affect 96 measurement properties. Researchers need to take into account that ceiling effects exist which might 97 limit sensitivity among participants who fared particularly well. 98 PLAIN ENGLISH SUMMARY 99 This study investigated how well the Short Form-36 (SF-36), a commonly used tool for assessing 100 health-related quality of life, performs in childhood cancer survivors from five different European 101 countries. We looked at the amount of survivors answering to all of the SF-36 questions, how many 102 indicated the most positive or the most negative answer category, whether the questions were 103 consistent within the different aspects of the questionnaire and how consistent and dependable it 104 measured quality of life in the survivors. We found that overall, the SF-36 measured health-related 105 quality of life well in all the countries. This is reassuring for researchers considering its use in studies 106 involving multiple centers across Europe. 107 MAIN TEXT 108 INTRODUCTION 109 Late effects after childhood cancer and its treatment can affect long-term health-related quality of life 110 (HRQOL) [1-10]. The Short-Form 36 (SF-36) is a widely used instrument to assess HRQOL [11-13]. The 111 childhood cancer survivor studies in the United States, the United Kingdom, Switzerland, and The 112 Netherlands implemented the SF-36 as a long-term outcome measure [3; 6; 7; 9]. Most studies used 113 version 1 (V1) of the SF-36 [13] and a few studies used version 2 (V2) [11] since its introduction in the 114 late 1990s. Versions differ mainly by the number of response categories and wordings of several 115 questions [11]. 116 117 Since SF-36 measurement properties vary by version and cultural background of the population, 118 researchers should evaluate the structural validity for a specific study population before interpreting 119 questionnaire results [14; 15]. Both V1 and V2 of the SF-36 showed good structural validity among 120 general population samples and a broad range of cohorts with chronic diseases, including multiple 121 sclerosis, rheumatoid conditions, and HIV [15-24]. Among more than 10,000 adult survivors from the 122 British Childhood Cancer Survivor Study (BCCSS), the SF-36 V1 provided overall valid and reliable 123 psychometric criteria [25]. Such reassuring results prompted the European Seventh Framework project 124 PanCareLIFE collaborators to use the SF-36 (either V1 or V2) for assessing HRQOL in a pooled sample 125 of adult survivors of childhood cancer from 5 European countries: Czech Republic, France, Germany, 126 The Netherlands, and Switzerland [26]. In this study, we evaluated measurement properties of the SF127 36 for a cohort of nearly 10,000 survivors of childhood cancer in Europe, since cultural differences 128 between countries and different SF-36 versions possibly affect HRQOL result interpretation. We aimed 129 to compare differences between countries, including data completeness, floor and ceiling effects, item130 internal consistency, item-discriminant validity, reliability, and scaling assumptions focusing on country131 and version-specific differences. 132 133 . CC-BY-NC-ND 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted January 23, 2024. ; https://doi.org/10.1101/2024.01.23.24301414doi: medRxiv preprint 4 Version 4 January 22, 2024 METHODS 134 Study and study population 135 We used data from the European PanCareLIFE framework (grant agreement no. 602030). Five 136 European countries (Czech Republic, France, Germany, The Netherlands, and Switzerland) contributed 137 HRQOL and clinical data from adult survivors of childhood cancer. Eligible participants were 1) younger 138 than age 18 at the time of diagnosis; 2) residents of the respective participating country at the time of 139 diagnosis; 3) diagnosed with cancer according to the International Classification of Childhood Cancer, 140 3rd edition (ICCC-3) [27] or Langerhans cell histiocytosis; 4) who were alive at least 5 years after cancer 141 diagnosis; 5) age 18 or older when they completed questionnaires; and 6) not undergoing treatment for 142 cancer at time of study [26]. All survivors participated in regional or national population–based cohort 143 studies. In the Czech Republic, data came from the university hospitals in Prague and Brno, which cover 144 childhood cancer survivors in 95% of the country and in France from the regional population–based 145 registry of the Rhone-Alps region. Germany, The Netherlands, and Switzerland collected data through 146 their own population– and hospital–based cohort studies with >95% national coverage: the German 147 VIVE cohort [8], the Dutch Childhood Cancer Survivor Study (DCCSS LATER, previously named DCOG 148 LATER) [28], and the Swiss Childhood Cancer Survivor Study (SCCSS) [29]. Study design, patient 149 characteristics, and sampling methods for the overall cohort and for each country were published 150 elsewhere [26]. 151 152 In brief, participants completed the questionnaires between 2005–2017. Eligible childhood cancer 153 survivors received questionnaires including the SF-36 and sociodemographic questions as part of their 154 scheduled follow-up consultations (Czech Republic) or by postal mail in other participating countries. 155 SF-36 questions were asked either at beginning of longer questionnaires (France, The Netherlands, and 156 Switzerland) or at the end (Czech Republic and Germany). The Czech Republic, Germany, and The 157 Netherlands used V1 of the SF-36; France and Switzerland used V2. Survivors who did not return 158 completed questionnaires were reminded at least once. Clinical baseline data came from respective 159 cancer registries or hospital databases of each country. Supplemental table 1 describes participant 160 characteristics stratified by SF-36 version. Country-specific characteristics were published elsewhere 161 [26]. All data contributed to PanCareLIFE were pseudonymized. All survivors provided either written 162 informed consent or implied consent or assent by returning the questionnaire. 163 164 Short Form-36 (SF-36) 165 The SF-36 includes 35 items covering 8 scales: physical functioning (PF, 10 items), role-limitations due 166 to physical problems (role physical, RP, 4 items), bodily pain (BP, 2 items), general health (GH, 5 items), 167 vitality (VT, 4 items), social functioning (SF, 2 items), role-limitations due to emotional problems (role 168 emotional, RE, 3 items) and mental health (MH, 5 items). One additional SF-36 item describes a 169 subjective change in health over a one year time period, but it is not part of this study. From the 8 scales, 170 2 summary scores—physical component summary (PCS) and mental component summary (MCS)— 171 are derived by factor analysis. Items use a Likert scale with a maximum range of possible answers from 172 1 to 6. The 2 SF-36 versions differ in layout, font size and bolding, wording, and number of response 173 choices. Wording changes from V1 to V2 include shortening and simplifying instructions and item text. 174 Response scale changes refer to replacing dichotomous (RP and RE) and some 6-level response 175 choices (MH and VT) with 5-level response choices [11]. 176 177 According to the User’s Manual of the SF-36 V2, we translated answers into percentage scores ranging 178 from 0–100 [11]. Higher scores indicate better HRQOL. We transformed all scales from V2 to V1 in line 179 with the User’s Manual [11]. Since normative data were not available for all participating countries, we 180 used normative data from a representative population in Germany from 1998 [30]. This is a well181 described sample including the largest number of participants from a country participating in our study 182 and it used V1, which is also mostly used in our sample. It also allowed comparing measurement 183 properties between countries. We converted raw scores into T-scores (mean=50, SD=10), according to 184 the German normative data stratified for age and gender from 1998 [30]. 185 Assessed criteria 186 Terwee et al. proposed quality criteria for evaluating health status questionnaires [31]. Based on these 187 recommendations, we evaluated a) data completeness; b) floor and ceiling effects; c) item-internal 188 consistency; d) item-discriminant validity; e) reliability; and f) checked scaling assumptions of the 2 SF189 36 summary scores, MCS and PCS. We stratified analyses for data completeness by country, for all 190 . CC-BY-NC-ND 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted January 23, 2024. ; https://doi.org/10.1101/2024.01.23.24301414doi: medRxiv preprint 5 Version 4 January 22, 2024 other criteria by version, country, gender, cancer diagnosis and age at survey, and for criterion f) by 191 version of the SF-36. 192 Statistical analyses 193 Data completeness: We described missing data by tabulating frequencies of survivors with unanswered 194 single items on the SF-36 and calculating percentages of missing data for each item. 195 196 Floor and ceiling effects: We measured the extent responses reached the upper and lower limit of the 197 measurement range by calculating percentages of survivors with the lowest and highest scores for each 198 scale. Floor and ceiling effects were present if >15% of participants responded with the lowest or highest 199 score, respectively [31]. 200 201 Item-internal consistency: We assessed whether items measured similar constructs as presumed by the 202 developers of the SF-36. Within each of the 8 SF-36 scales, we calculated correlations between each 203 item and its respective related scale (item-rest correlations). We considered correlations >0.4 204 satisfactory for item-internal consistency [13]. 205 206 Item-discriminant validity: To determine item-discriminant validity, we checked whether the correlation 207 between an item and its hypothesized scale was higher than the correlations between that item and all 208 other scales [32]. We subtracted correlations between each item and their unrelated scales (item-scale 209 correlations) from the item-rest correlations of the same item calculated. We considered differences of 210 >2 standard errors between item-rest correlations and item-scale correlations as scaling success, 211 indicating the item correlated higher with its hypothesized scale than with the other—supposedly 212 unrelated—scales. We calculated success rates as the proportion of comparisons with scaling success 213 divided by the total number of comparisons [13]. 214 215 Scale reliability: We calculated Cronbach’s alpha for all scales including the respective items of the 216 scale. Terwee et al. proposed a Cronbach’s alpha coefficient between 0.7 and 0.95 as satisfactory for a 217 good internal consistency [31], and the User’s Manual of the SF-36 V2 a threshold of ≥0.7 for acceptable 218 reliability [11]. 219 220 Scaling assumptions of the summary scores: We performed principal component analyses and 221 subsequent orthogonal varimax rotation and calculated factor loadings—correlations of each scale with 222 the PCS and MCS—and estimated communality and uniqueness for V1 and V2 [23]. Communality is 223 the variance of each scale explained by the PCS and MCS; uniqueness is the unexplained variance. 224 We considered correlations of r>0.7 as strong and of r<0.3 as weak [13]. We expected strong 225 correlations of the physical health scales PF, RP, and BP with the PCS and weak correlations with the 226 MCS and strong correlations of the mental health scales MH and RE with the MCS and weak correlations 227 with the PCS. We expected GH, VT, and SF scales with values in between. 228 RESULTS 229 Data completeness 230 Out of 24,993 eligible survivors, we contacted 19,268 survivors and 10,077 (53%) returned 231 questionnaires (Table 1). Details on the eligible survivors and on those who were contacted for the study 232 were published elsewhere [26]. In brief, the largest cohort of eligible survivors was in Germany 233 (n=12,144); the smallest in France (n=1,060). Participation ranged from 65% (1,029/1,576) in the Czech 234 Republic to 45% (385 out of 851 of contacted survivors) in France. Of the 10,077 survivors who returned 235 questionnaires, 9,300 (92%) answered all items of the SF-36 and 9,871 (98%) answered at least half of 236 the SF-36 items within a scale, which sufficed for scoring the SF-36 [13]. We found differences in 237 completeness between countries (Table 1). France had the highest percentage of survivors who 238 completed all items (377/391, 96%) and the Czech Republic had the lowest (979/1,085, 90%). In 239 Switzerland, 0.3% (5/1,590) of participants answered less than half of the questions in at least 1 scale; 240 in the Czech Republic, the figure was highest with 5% (56/1085). Data completeness was generally 241 high, and the percentage of missing data never exceeded 4% for a single item. Items 8 (BP), 11b and 242 11d (GH), and 9a (VT) had the most missing values, while items 3h and 3i (PF) had the fewest missing 243 values. We did not observe specific patterns in completeness for the two different versions of the SF244 36. 245 . CC-BY-NC-ND 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted January 23, 2024. ; https://doi.org/10.1101/2024.01.23.24301414doi: medRxiv preprint 6 Version 4 January 22, 2024 246 Table 1: Missing data of SF-36 items for all survivors who answered at least 1 item of the SF-36 for all countries 247 combined and stratified by country. 248 All countries combined Czech Republic France Germany The Netherlands Switzerland N=10,077 N=1085 N=391 N=4835 N=2176 N=1590 n % n % n % n % n % n % Physical functioning (PF) Item 3a 83 0.8 37 3.4 4 1 29 0.6 11 0.5 2 0.1 Item 3b 68 0.7 32 3 3 0.8 22 0.5 8 0.4 3 0.2 Item 3c 74 0.7 36 3.3 3 0.8 25 0.5 7 0.3 3 0.2 Item 3d 67 0.7 30 2.8 3 0.8 23 0.5 7 0.3 4 0.3 Item 3e 66 0.7 32 3 3 0.8 23 0.5 7 0.3 1 0.1 Item 3f 70 0.7 32 3 3 0.8 25 0.5 9 0.4 1 0.1 Item 3g 67 0.7 32 3 3 0.8 23 0.5 8 0.4 1 0.1 Item 3h 63 0.6 31 2.9 3 0.8 23 0.5 6 0.3 0 0 Item 3i 63 0.6 31 2.9 3 0.8 22 0.5 6 0.3 1 0.1 Item 3j 67 0.7 33 3 4 1 22 0.5 8 0.4 0 0 Role physical (RP) Item 4a 83 0.8 29 2.7 7 1.8 40 0.8 7 0.3 0 0 Item 4b 83 0.8 29 2.7 5 1.3 39 0.8 10 0.5 0 0 Item 4c 93 0.9 27 2.5 5 1.3 45 0.9 11 0.5 5 0.3 Item 4d 97 1 27 2.5 6 1.5 46 1 13 0.6 5 0.3 Bodily pain (BP) Item 7 111 1.1 27 2.5 5 1.3 73 1.5 5 0.2 1 0.1 Item 8 133 1.3 27 2.5 7 1.8 83 1.7 12 0.6 4 0.3 General health (GH) Item 1 66 0.7 18 1.7 2 0.5 43 0.9 2 0.1 1 0.1 Item 11a 113 1.1 28 2.6 3 0.8 57 1.2 20 0.9 5 0.3 Item 11b 126 1.3 30 2.8 3 0.8 67 1.4 21 1 5 0.3 Item 11c 100 1 27 2.5 3 0.8 52 1.1 14 0.6 4 0.3 Item 11d 133 1.3 30 2.8 3 0.8 72 1.5 16 0.7 12 0.8 Vitality (VT) Item 9a 128 1.3 28 2.6 5 1.3 66 1.4 20 0.9 9 0.6 Item 9e 114 1.1 30 2.8 4 1 65 1.3 13 0.6 2 0.1 Item 9g 109 1.1 31 2.9 4 1 61 1.3 11 0.5 2 0.1 Item 9i 104 1 33 3 4 1 55 1.1 8 0.4 4 0.3 Social functioning (SF) Item 6 86 0.9 27 2.5 4 1 46 1 7 0.3 2 0.1 Item 10 110 1.1 29 2.7 5 1.3 64 1.3 8 0.4 4 0.3 Role emotional (RE) Item 5a 90 0.9 27 2.5 7 1.8 47 1 8 0.4 1 0.1 Item 5b 88 0.9 25 2.3 6 1.5 46 1 9 0.4 2 0.1 Item 5c 101 1 26 2.4 6 1.5 54 1.1 14 0.6 1 0.1 Mental health (MH) Item 9b 99 1 29 2.7 4 1 54 1.1 3 0.4 3 0.2 Item 9c 102 1 29 2.7 4 1 53 1.1 14 0.6 2 0.1 Item 9d 100 1 31 2.9 4 1 48 1 13 0.6 4 0.3 Item 9f 98 1 28 2.6 4 1 50 1 13 0.6 3 0.2 Item 9h 109 1.1 31 2.9 4 1 62 1.3 9 0.4 3 0.2 Incomplete in ≥1 item 777 7.7 106 9.8 14 3.6 469 9.7 114 5.2 74 4.7 All items complete 9300 92.3 979 90.2 377 96.4 4366 90.3 2062 94.8 1516 95.4 ≥50% missings in ≥1 scale 206 2 56 5.2 6 1.5 109 2.3 30 1.4 5 0.3 <50% missings per scale 9871 98 1029 94.8 385 98.5 4726 97.8 2164 98.6 1585 99.7 Abbreviations: BP: bodily pain, GH: general health, MH: mental health, PF: physical functioning, RE: role emotional, 249 RP: role physical, SF: social functioning, SF-36, Short Form-36. 250 . CC-BY-NC-ND 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted January 23, 2024. ; https://doi.org/10.1101/2024.01.23.24301414doi: medRxiv preprint 7 Version 4 January 22, 2024 Ceiling and floor effects 251 Ceiling effects were present in the 5 scales of PF, RP, BP, SF, and RE, yet not in the 3 scales of GH, 252 VT and MH (Table 2). Among all scales, RP and RE had the highest ceiling effects (71% and 70%) 253 (Table 2). Ceiling effects differed by version, country, gender, age at survey, and diagnostic group of 254 the participants. Ceiling effects were higher in V1 (RP, 76%; RE, 74%) compared with V2 (RP, 63%; 255 RE, 58%). Of all countries, Switzerland (PF, 71%; BP, 66%; SF, 59%) and Germany (RE and RP, 75%) 256 had the highest ceiling effects. Among men ceiling effects were higher when compared with women 257 (range 58%-77% vs. 55%-71%) (Table 2). Among all cancer diagnostic group, ceiling effects were high 258 in the 5 scales of PF, RP, BP, SF, and RE, but less pronounced in survivors of CNS, hepatic and bone 259 tumours. Ceiling effects were particularly high for young participants aged 25–35 years at survey and 260 lower among older adults. Floor effects were rare (Supplemental Table 2). We found them only in the 261 scale of RP for bone tumour survivors and for survivors older than 40 years. 262 . CC-BY-NC-ND 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted January 23, 2024. ; https://doi.org/10.1101/2024.01.23.24301414doi: medRxiv preprint 8 Version 4 January 22, 2024 Table 2: Ceiling effects: Proportions of survivors with the highest possible score by country, SF-36 version, gender, cancer diagnosis, and age at survey. 263 n PF(%) RP(%)* BP(%) GH(%) VT(%)* SF(%) RE(%)* MH(%)* V1 V2 V1 V2 V1 V2 V1 V2 Overall cohort 9871 51.26 72.54 62.54 56.28 8.72 2.44 3.35 53.01 73.52 58.12 3.72 4.57 Country Czech Republic 1029 45.38 70.36 n/a 55.30 5.64 1.36 n/a 47.81 69.10 n/a 2.04 n/a France 385 51.69 n/a 55.06 40.78 5.45 n/a 1.82 36.62 n/a 48.31 n/a 2.08 Germany 4726 52.31 75.43 n/a 60.85 8.72 2.20 n/a 56.79 74.63 n/a 3.24 n/a The Netherlands 2146 44.69 67.19 n/a 41.94 6.52 3.49 n/a 45.90 73.21 n/a 5.59 n/a Switzerland 1585 60.76 n/a 64.35 66.44 14.51 n/a 3.72 58.74 n/a 60.50 n/a 5.17 Version 1 7901 49.34 72.54 n/a 54.99 7.72 2.44 n/a 52.66 73.52 n/a 3.72 n/a 2 1970 58.98 n/a 62.54 61.42 12.74 n/a 3.35 54.42 n/a 58.12 n/a 4.57 Gender Male 4725 58.48 76.30 65.84 61.38 10.60 3.51 4.74 57.86 76.50 65.59 4.58 5.65 Female 5146 44.64 69.16 59.14 51.99 7.00 1.49 1.94 48.56 70.53 50.56 2.95 3.47 Cancer diagnosis (ICCC-3) I Leukaemias 3157 58.12 75.01 71.54 59.71 8.93 2.81 3.86 56.16 73.43 63.62 4.05 4.07 II Lymphomas 2075 57.20 77.26 67.51 59.81 9.83 2.62 5.07 55.81 74.34 62.90 3.53 5.99 III CNS tumours 1356 39.01 62.50 51.81 52.29 6.64 2.44 2.11 44.25 70.02 49.40 3.61 4.22 IV Neuroblastoma 440 50.91 74.48 66.02 54.55 10.91 2.67 2.91 53.18 75.37 64.80 2.67 3.88 V Retinoblastoma 172 64.16 79.39 53.66 65.32 7.51 6.11 2.44 57.56 77.10 60.98 9.16 12.20 VI Renal tumours 738 54.88 78.22 64.39 57.99 9.89 1.32 3.03 57.32 76.40 56.82 2.15 2.27 VII Hepatic tumours 61 52.46 74.42 50.00 49.18 3.28 0.00 5.56 44.26 62.79 44.44 0.00 5.56 VIII Bone tumours 588 15.65 59.48 41.13 34.18 6.63 1.51 0.00 45.92 72.84 48.39 3.66 1.61 IX Soft tissue sarcomas 645 47.93 72.78 60.80 53.14 8.73 1.51 0.80 50.61 73.35 56.00 4.91 4.00 X Germ cell tumours 386 50.26 73.31 57.33 54.14 7.51 0.96 9.33 50.52 76.53 50.67 2.25 10.67 XI Epithelial neoplasms & melanomas 130 50.77 72.34 58.33 58.46 5.38 3.19 0.00 46.15 68.09 50.00 4.26 0.00 Other malignant neoplasms# 114 63.16 69.64 68.97 66.67 14.91 7.14 1.72 55.26 75.00 60.34 5.36 3.45 Age at survey (years) 18-<25 1636 56.91 70.51 62.50 56.91 10.70 3.46 2.10 52.57 70.77 57.71 4.62 4.21 25-<30 2997 56.26 76.40 64.99 62.30 10.14 3.00 4.02 57.26 74.28 56.74 3.56 4.43 30-<35 2463 53.76 75.05 61.33 58.02 9.38 2.39 3.93 53.02 74.44 59.21 4.17 3.32 35-<40 1437 46.97 70.75 63.13 44.30 6.47 1.43 4.47 52.12 74.04 62.57 2.78 6.15 40-<45 833 37.45 65.05 55.07 33.05 4.20 1.57 5.80 46.58 71.07 56.52 2.49 8.70 >=45 505 26.14 60.81 52.63 56.28 4.55 2.14 7.89 42.38 72.38 57.89 5.57 10.53 Abbreviations: BP: bodily pain, GH: general health, ICCC-3: International Classification of Childhood Cancer, 3rd edition, MH: mental health, n/a: not applicable, PF: 264 physical functioning, RE: role emotional, RP: role physical, SF: social functioning, V1: Version 1, V2: Version 2, VT: vitality, SF-36: Short Form-36. 265 * Values calculated separately for V1 and V2, as number of response choices differs between versions. 266 # ICCC-3 main group XII (Other and unspecified malignant neoplasms) and Langerhans cell histiocytosis but not benign and in situ tumours and tumour-like lesions 267 or unclassified survivors. 268 . CC-BY-NC-ND 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted January 23, 2024. ; https://doi.org/10.1101/2024.01.23.24301414doi: medRxiv preprint 9 Version 4 January 22, 2024 Item-internal consistency 269 In the total cohort and stratified by country, version, gender, and age at survey, all correlations were 270 above the threshold of 0.4, indicating satisfactory consistency between items within a scale (Table 3). 271 We found the lowest correlation coefficients among survivors in the Czech Republic for 5 out of 8 scales 272 (PF, RP, BP, SF, RE) and the highest coefficients in Germany (PF, BP, SF) and France (RP, GH, VT, 273 RE). Item-rest correlations were lower in V1 than in V2 for RP (0.67–0.75 vs. 0.80–0.84) and RE (0.65– 274 0.70 vs. 0.75–0.82). For survivors of some cancers, correlations were below 0.4: for survivors of hepatic 275 tumours and neuroblastoma, correlations within PF were 0.35 for item 3f and 0.38 for item 3j, 276 respectively. For survivors of epithelial neoplasms and melanomas, and other malignant neoplasms, 277 correlations within MH were 0.36 for item 9b (V2), and 0.36 for item 9d (V2). 278 Item-discriminant validity 279 Overall, all items correlated better with the items in their hypothesized scale than with the items in 280 unrelated scales (data not shown). However, we found exceptions when we stratified the sample by 281 country or cancer diagnosis. For example, in Switzerland and France—the countries using V2—item 9h 282 (MH) correlated slightly better with VT; in France and The Netherlands, item 9a (VT) correlated slightly 283 better with MH. Among survivors of retinoblastoma, hepatic tumours, epithelial neoplasm and 284 melanoma, and other tumours, few items correlated better with unrelated scales than with their 285 hypothesized scale. In retinoblastoma survivors, item 1 (GH) correlated slightly higher with SF; item 3a 286 (PF) with BP, SF, and GH; and item 9g (VT) with MH. In survivors of hepatic tumours, item 3g (PF) 287 correlated higher with SF; and items 9a, 9e (VT) with MH. In survivors of epithelial neoplasm and 288 melanoma, item 9a (VT) correlated higher with MH. In survivors of other tumours, item 3a (PF) correlated 289 higher with GH; and item 9a (VT) with MH. These differences never exceeded 2 standard errors, so we 290 concluded 100% scaling success for all items in all subgroups. 291 . CC-BY-NC-ND 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted January 23, 2024. ; https://doi.org/10.1101/2024.01.23.24301414doi: medRxiv preprint 16 Version 4 January 22, 2024 REFERENCES 422 1. Belson, P. J., Eastwood, J. A., Brecht, M. L., Hays, R. D., & Pike, N. A. (2020). A 423 Review of Literature on Health-Related Quality of Life of Retinoblastoma Survivors. J 424 Pediatr Oncol Nurs, 37(2), 116-127. https://doi.org/10.1177/1043454219888805 425 2. Calaminus, G., Dorffel, W., Baust, K., Teske, C., Riepenhausen, M., Bramswig, J., 426 Flechtner, H. H., Singer, S., Hinz, A., & Schellong, G. (2014). Quality of life in long427 term survivors following treatment for Hodgkin's disease during childhood and 428 adolescence in the German multicentre studies between 1978 and 2002. Support Care 429 Cancer, 22(6), 1519-1529. https://doi.org/10.1007/s00520-013-2114-y 430 3. Hudson, M. M., Ness, K. K., Gurney, J. G., Mulrooney, D. A., Chemaitilly, W., Krull, 431 K. R., Green, D. M., Armstrong, G. T., Nottage, K. A., Jones, K. E., Sklar, C. A., 432 Srivastava, D. K., & Robison, L. L. (2013). Clinical ascertainment of health outcomes 433 among adults treated for childhood cancer. JAMA, 309(22), 2371-2381. 434 https://doi.org/10.1001/jama.2013.6296 435 4. Michel, G., Brinkman, T. M., Wakefield, C. E., & Grootenhuis, M. (2020). 436 Psychological Outcomes, Health-Related Quality of Life, and Neurocognitive 437 Functioning in Survivors of Childhood Cancer and Their Parents. Pediatr Clin North 438 Am, 67(6), 1103-1134. https://doi.org/10.1016/j.pcl.2020.07.005 439 5. Nathan, P. C., Ness, K. K., Greenberg, M. L., Hudson, M., Wolden, S., Davidoff, A., 440 Laverdiere, C., Mertens, A., Whitton, J., Robison, L. L., Zeltzer, L., & Gurney, J. G. 441 (2007). Health-related quality of life in adult survivors of childhood Wilms tumor or 442 neuroblastoma: A report from the childhood cancer survivor study. Pediatr Blood 443 Cancer, 49(5), 704-715. https://doi.org/10.1002/pbc.20949 444 6. Reulen, R. C., Winter, D. L., Lancashire, E. R., Zeegers, M. P., Jenney, M. E., Walters, 445 S. J., Jenkinson, C., & Hawkins, M. M. (2007). Health-status of adult survivors of 446 childhood cancer: a large-scale population-based study from the British Childhood 447 Cancer Survivor Study. Int J Cancer, 121(3), 633-640. https://doi.org/10.1002/ijc.22658 448 7. Rueegg, C. S., Gianinazzi, M. E., Rischewski, J., Beck Popovic, M., von der Weid, N. 449 X., Michel, G., & Kuehni, C. E. (2013). Health-related quality of life in survivors of 450 childhood cancer: the role of chronic health problems. J Cancer Surviv, 7(4), 511-522. 451 https://doi.org/10.1007/s11764-013-0288-4 452 8. Scholtes, C., Baust, K., Weinhold, L., Creutzig, U., Gnekow, A., Hinz, A., Kaatsch, P., 453 Kreitz, K., Langer, T., Rutkowski, S., Singer, S., Spix, C., Teske, C., Schmid, M., Dilloo, 454 D., & Calaminus, G. (2019). Health status, health-related quality of life, and 455 socioeconomic outcome in childhood brain tumor survivors: a German cohort study. 456 Neuro Oncol, 21(8), 1069-1081. https://doi.org/10.1093/neuonc/noz044 457 9. van Erp, L. M. E., Maurice-Stam, H., Kremer, L. C. M., Tissing, W. J. E., van der Pal, 458 H. J. H., de Vries, A. C. H., van den Heuvel-Eibrink, M. M., Versluys, B. A. B., Loonen, 459 J. J., Bresters, D., Louwerens, M., van der Heiden-van der Loo, M., van den Berg, M. 460 H., Ronckers, C. M., van der Kooi, A., van Gorp, M., van Dulmen-den Broeder, E., & 461 Grootenhuis, M. A. (2021). Health-related quality of life in Dutch adult survivors of 462 childhood cancer: A nation-wide cohort study. Eur J Cancer, 152, 204-214. 463 https://doi.org/10.1016/j.ejca.2021.04.033 464 10. Vetsch, J., Wakefield, C. E., Robertson, E. G., Trahair, T. N., Mateos, M. K., 465 Grootenhuis, M., Marshall, G. M., Cohn, R. J., & Fardell, J. E. (2018). Health-related 466 quality of life of survivors of childhood acute lymphoblastic leukemia: a systematic 467 review. Qual Life Res, 27(6), 1431-1443. https://doi.org/10.1007/s11136-018-1788-5 468 11. Maruish, M. E., Kosinski, M., Bjorner, J., Gandek, B., Turner-Bowker, D. M., & Ware, 469 J. E., Jr. (2011). User's Manual for the SF-36v2 Health Survey. Quality Metric 470 Incorporated. https://books.google.de/books?id=a0vYnQEACAAJ 471 . CC-BY-NC-ND 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted January 23, 2024. ; https://doi.org/10.1101/2024.01.23.24301414doi: medRxiv preprint 17 Version 4 January 22, 2024 12. Ware, J. E., Jr., & Gandek, B. (1998). Overview of the SF-36 Health Survey and the 472 International Quality of Life Assessment (IQOLA) Project. J Clin Epidemiol, 51(11), 473 903-912. https://doi.org/10.1016/s0895-4356(98)00081-x 474 13. Ware, J. E., Jr., Kosinski, M., & Gandek, B. (1993). SF-36 health survey: Manual and 475 interpretation guide. The Health Institute, New England Medical Center. 476 14. McHorney, C. A., Ware, J. E., Jr., Lu, J. F., & Sherbourne, C. D. (1994). The MOS 36477 item Short-Form Health Survey (SF-36): III. Tests of data quality, scaling assumptions, 478 and reliability across diverse patient groups. Med Care, 32(1), 40-66. 479 https://doi.org/10.1097/00005650-199401000-00004 480 15. Hobart, J., Freeman, J., Lamping, D., Fitzpatrick, R., & Thompson, A. (2001). The SF481 36 in multiple sclerosis: why basic assumptions must be tested. J Neurol Neurosurg 482 Psychiatry, 71(3), 363-370. https://doi.org/10.1136/jnnp.71.3.363 483 16. da Costa Vieira, V., Lins, L., Sarmento, V. A., Netto, E. M., & Brites, C. (2018). Oral 484 health and health-related quality of life in HIV patients. BMC Oral Health, 18(1), 151. 485 https://doi.org/10.1186/s12903-018-0605-4 486 17. ten Klooster, P. M., Vonkeman, H. E., Taal, E., Siemons, L., Hendriks, L., de Jong, A. 487 J., Dutmer, E. A., van Riel, P. L., & van de Laar, M. A. (2013). Performance of the Dutch 488 SF-36 version 2 as a measure of health-related quality of life in patients with rheumatoid 489 arthritis. Health Qual Life Outcomes, 11, 77. https://doi.org/10.1186/1477-7525-11-77 490 18. Alzahrani, H., Alshehri, Y. S., Barcaccia, B., Alshehri, M. A., Alzhrani, M., & Bjorner, 491 J. B. (2023). Health-related quality of life in Welsh adults: psychometric properties of 492 the SF-36v2 and normative data. Public Health, 214, 153-162. 493 https://doi.org/10.1016/j.puhe.2022.11.010 494 19. Frieling, M. A., Davis, W. R., & Chiang, G. (2013). The SF-36v2 and SF-12v2 health 495 surveys in New Zealand: norms, scoring coefficients and cross-country comparisons. 496 Aust N Z J Public Health, 37(1), 24-31. https://doi.org/10.1111/1753-6405.12006 497 20. Gandek, B., Ware, J. E., Jr., Aaronson, N. K., Alonso, J., Apolone, G., Bjorner, J., Brazier, 498 J., Bullinger, M., Fukuhara, S., Kaasa, S., Leplege, A., & Sullivan, M. (1998). Tests of 499 data quality, scaling assumptions, and reliability of the SF-36 in eleven countries: results 500 from the IQOLA Project. International Quality of Life Assessment. J Clin Epidemiol, 501 51(11), 1149-1158. https://doi.org/10.1016/s0895-4356(98)00106-1 502 21. Garratt, A. M., & Stavem, K. (2017). Measurement properties and normative data for 503 the Norwegian SF-36: results from a general population survey. Health Qual Life 504 Outcomes, 15(1), 51. https://doi.org/10.1186/s12955-017-0625-9 505 22. Jenkinson, C., Stewart-Brown, S., Petersen, S., & Paice, C. (1999). Assessment of the 506 SF-36 version 2 in the United Kingdom. J Epidemiol Community Health, 53(1), 46-50. 507 https://doi.org/10.1136/jech.53.1.46 508 23. Roser, K., Mader, L., Baenziger, J., Sommer, G., Kuehni, C. E., & Michel, G. (2019). 509 Health-related quality of life in Switzerland: normative data for the SF-36v2 510 questionnaire. Qual Life Res, 28(7), 1963-1977. https://doi.org/10.1007/s11136-019511 02161-5 512 24. Taft, C., Karlsson, J., & Sullivan, M. (2004). Performance of the Swedish SF-36 version 513 2.0. Qual Life Res, 13(1), 251-256. 514 https://doi.org/10.1023/B:QURE.0000015290.76254.a5 515 25. Reulen, R. C., Zeegers, M. P., Jenkinson, C., Lancashire, E. R., Winter, D. L., Jenney, 516 M. E., & Hawkins, M. M. (2006). The use of the SF-36 questionnaire in adult survivors 517 of childhood cancer: evaluation of data quality, score reliability, and scaling 518 assumptions. Health Qual Life Outcomes, 4, 77. https://doi.org/10.1186/1477-7525-4519 77 520 26. Calaminus, G., Baust, K., Berger, C., Byrne, J., Binder, H., Casagranda, L., Grabow, D., 521 Grootenhuis, M., Kaatsch, P., Kaiser, M., Kepak, T., Kepakova, K., Kremer, L. C. M., 522 . CC-BY-NC-ND 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted January 23, 2024. ; https://doi.org/10.1101/2024.01.23.24301414doi: medRxiv preprint 18 Version 4 January 22, 2024 Kruseova, J., Luks, A., Spix, C., van den Berg, M., van den Heuvel-Eibrink, M. M. M., 523 van Dulmen-den Broeder, E., Kuonen, R., Sommer, G., & Kuehni, C. (2021). Health524 Related Quality of Life in European Childhood Cancer Survivors: Protocol for a Study 525 Within PanCareLIFE. JMIR Res Protoc, 10(1), e21851. https://doi.org/10.2196/21851 526 27. Steliarova-Foucher, E., Stiller, C., Lacour, B., & Kaatsch, P. (2005). International 527 Classification of Childhood Cancer, third edition. Cancer, 103(7), 1457-1467. 528 https://doi.org/10.1002/cncr.20910 529 28. Teepen, J. C., Kok, J. L., Feijen, E. A. M., Loonen, J. J., van den Heuvel-Eibrink, M. 530 M., van der Pal, H. J., Tissing, W. J. E., Bresters, D., Versluys, B., Grootenhuis, M. A., 531 Louwerens, M., Neggers, S., van Santen, H. M., de Vries, A., Janssens, G. O., den 532 Hartogh, J. G., van Leeuwen, F. E., Hollema, N., Streefkerk, N., Kilsdonk, E., van der 533 Heiden-van der Loo, M., van Dulmen-den Broeder, E., Ronckers, C. M., Kremer, L. C. 534 M., & Group, D. L. S. (2023). Questionnaireand linkage-based outcomes in Dutch 535 childhood cancer survivors: Methodology of the DCCSS LATER study part 1. Cancer 536 Med, 12(6), 7588-7602. https://doi.org/10.1002/cam4.5519 537 29. Kuehni, C. E., Rueegg, C. S., Michel, G., Rebholz, C. E., Strippoli, M. P., Niggli, F. K., 538 Egger, M., von der Weid, N. X., & Swiss Paediatric Oncology, G. (2012). Cohort profile: 539 the Swiss childhood cancer survivor study. Int J Epidemiol, 41(6), 1553-1564. 540 https://doi.org/10.1093/ije/dyr142 541 30. Ellert, U., & Bellach, B. M. (1999). [The SF-36 in the Federal Health Survey-- 542 description of a current normal sample]. Gesundheitswesen, 61 Spec No, S184-190. 543 https://www.ncbi.nlm.nih.gov/pubmed/10726419 (Der SF-36 im Bundes544 Gesundheitssurvey--Beschreibung einer aktuellen Normstichprobe.) 545 31. Terwee, C. B., Bot, S. D., de Boer, M. R., van der Windt, D. A., Knol, D. L., Dekker, J., 546 Bouter, L. M., & de Vet, H. C. (2007). Quality criteria were proposed for measurement 547 properties of health status questionnaires. J Clin Epidemiol, 60(1), 34-42. 548 https://doi.org/10.1016/j.jclinepi.2006.03.012 549 32. Campbell, D. T., & Fiske, D. W. (1959). Convergent and discriminant validation by the 550 multitrait-multimethod matrix. Psychol Bull, 56(2), 81-105. 551 https://www.ncbi.nlm.nih.gov/pubmed/13634291 552 33. Beullens, K., Loosveldt, G., Vandenplas, C., & Stoop, I. (2018). Response Rates in the 553 European Social Survey: Increasing, Decreasing, or a Matter of Fieldwork Efforts? 554 Survey Methods: Insights from the Field. https://doi.org/10.13094/SMIF-2018-00003 555 34. Ohlsson-Nevo, E., Hiyoshi, A., Noren, P., Moller, M., & Karlsson, J. (2021). The 556 Swedish RAND-36: psychometric characteristics and reference data from the Mid557 Swed Health Survey. J Patient Rep Outcomes, 5(1), 66. https://doi.org/10.1186/s41687558 021-00331-z 559 35. Hawthorne, G., Osborne, R. H., Taylor, A., & Sansoni, J. (2007). The SF36 Version 2: 560 critical analyses of population weights, scoring algorithms and population norms. Qual 561 Life Res, 16(4), 661-673. https://doi.org/10.1007/s11136-006-9154-4 562 36. Ryder-Burbidge, C., Diaz, R. L., Barr, R. D., Gupta, S., Nathan, P. C., McKillop, S. J., 563 & Fidler-Benaoudia, M. M. (2021). The Burden of Late Effects and Related Risk 564 Factors in Adolescent and Young Adult Cancer Survivors: A Scoping Review. Cancers 565 (Basel), 13(19). https://doi.org/10.3390/cancers13194870 566 37. Morfeld, M., Bullinger, M., Nantke, J., & Brahler, E. (2005). [The version 2.0 of the 567 SF-36 Health Survey: results of a population-representative study]. Soz Praventivmed, 568 50(5), 292-300. https://doi.org/10.1007/s00038-005-4090-6 (Die Version 2.0 des SF-36 569 Health Survey--ergebnisse einer bevolkerungsreprasentativen Studie.) 570 38. Kaatsch, P., Byrne, J., Grabow, D., & PanCare, L. C. (2021). Managing a Pan-European 571 Consortium on Late Effects among Long-Term Survivors of Childhood and Adolescent 572 . CC-BY-NC-ND 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted January 23, 2024. ; https://doi.org/10.1101/2024.01.23.24301414doi: medRxiv preprint 19 Version 4 January 22, 2024 Cancer-The PanCareLIFE Project. Int J Environ Res Public Health, 18(8). 573 https://doi.org/10.3390/ijerph18083918 574 575 . CC-BY-NC-ND 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted January 23, 2024. ; https://doi.org/10.1101/2024.01.23.24301414doi: medRxiv preprint 20 Version 4 January 22, 2024 FUNDING 576 The authors claim responsibility for the presented materials and expressed. The European Union 577 Commission takes no responsibility for any use made of published information. Our project received 578 funding from the European Union’s Seventh Framework Programme (FP7) for research, technological 579 development, and demonstration under grant agreement no 602030. 580 581 In the Czech Republic, we received funding from the University Hospital of Brno from the Ministry of 582 Education, Youth and Sports of the Czech Republic, under grant agreement No. 7E13061; and the 583 University Hospital of Prague from the foundation Národ dětem (www.naroddetem.cz). In France, we 584 received funding from The Wyeth Foundation; the French National Institute of Cancer (INCa); and the 585 French League against Cancer. In Germany, we received funding from the German Cancer Aid (Grant 586 No. 110298). In Switzerland, we also received funding from the Swiss Cancer League and Swiss Cancer 587 Research (Grant no. KLS-3412-02-2014, HSR‐4951‐11‐2019, KFS-5302-02-2021, KLS/KFS5711588 01-2022); the Bernese Cancer League, Kinderkrebs Schweiz, and Stiftung für krebskranke Kinder Regio 589 Basiliensis. The work of the Childhood Cancer Registry in Switzerland is supported by the Federal Office 590 of Public Health. 591 DATA CONTRIBUTIONS 592 Data for this sub-project were provided by Academisch Medisch Centrum bij de Universiteit van 593 Amsterdam, Netherlands (Prof. Dr. LCM Kremer), Stichting VU-VUMC, Amsterdam, Netherlands (Dr. E 594 van Dulmen-den Broeder, Dr. MH van den Berg), Erasmus Medisch Centrum Rotterdam, Netherlands 595 (Prof. Dr. MM van den Heuvel-Eibrink), Prinses Máxima Centrum, Netherlands (Prof. Dr. MM van den 596 Heuvel-Eibrink, Prof. Dr. LCM Kremer, Dr. E van Dulmen-den Broeder), on behalf of the DCOG LATER 597 Study centres, Universität Bern, Switzerland (Prof. Dr. CE Kuehni), Fakultni nemocnice Brno, Czech 598 Republic (Dr. T Kepak), Centre Hospitalier Universitaire Saint Étienne, France (Dr. C Berger), Fakultni 599 nemocnice v Motole, Prague, Czech Republic (Dr. J Kruseova) and Universitaetsklinikum Bonn, Bonn, 600 Germany (Dr. G Calaminus). 601 DATA AVAILIBILITY STATEMENT 602 The data that support the information of this manuscript were accessed on secured servers of the 603 Institute of Social and Preventive Medicine at the University of Bern. Individual-level sensitive data can 604 only be made available for researchers who fulfil the respective legal requirements. All data requests 605 should be communicated to the corresponding author. 606 ETHICS DECLARATION 607 PanCareLIFE is a multinational collaborative research project that has harmonized and combined data 608 from regional and national cohorts across Europe. Data collection and analysis for each cohort was 609 approved by the responsible ethics committee in each participating country. For Germany this is the 610 Ethics Committee of the Medical Association of Westphalia-Lippe and the Medical faculty of the 611 Westphalian Wilhelms University (2012-530-f-S), for Switzerland the Cantonal Ethics Committee of the 612 Canton of Bern (KEK-BE: 166/2014; 2021-01462), for the Czech Republic the Ethics Committee for 613 Multi-Centric Clinical Trials of the University Hospital Motol (EK-1723/13) and the Multi-Centric Ethics 614 Committee of the University Hospital Brno (approval date: 2014/10/22), for France the Personal 615 Protection Committee South East 1 (CPP: 2015-23) and for the Netherlands the Medical Ethical Review 616 Committee of the Academic Medical Center, University of Amsterdam (2010_332#B2015108). 617 CONFLICT OF INTEREST 618 The authors have no relevant conflicts of interest to declare. 619 . CC-BY-NC-ND 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted January 23, 2024. ; https://doi.org/10.1101/2024.01.23.24301414doi: medRxiv preprint 21 Version 4 January 22, 2024 ABBREVIATIONS 620 BCCSS: British Childhood Cancer Survivor Study 621 BP: Bodily pain 622 CCSS: Childhood Cancer Survivor Study 623 CNS: Central Nervous System 624 CHU-SE: Centre Hospitalier Universitaire de Saint-Étienne, St. Étienne, France 625 DCOG LATER: Dutch Childhood Oncology Group Survivor study 626 FNM: Motol Teaching Hospital, Prague, Czech Republic 627 GCCR: German Childhood Cancer Registry 628 GH: General Health 629 HRQOL: Health-related Quality of Life 630 ICCC-3: International Classification of Childhood Cancer, 3rd edition 631 MCS: Mental Component Summary 632 MH: Mental health 633 n/a: not applicable 634 PCS: Physical Component Summary 635 PF: Physical functioning 636 RE: Role-limitations due to emotional problems 637 RP: Role-limitations due to physical problems 638 SCCR: Swiss Childhood Cancer Registry 639 SCCSS: Swiss Childhood Cancer Survivor Study 640 SF: Social functioning 641 SF-36: Short-Form 36 642 UHB: University Hospital Brno, Czech Republic 643 UKB: Universitätsklinikum Bonn, Germany 644 UKM: Universitätsklinikum Münster, Germany 645 UNIBE: University of Bern 646 UK: United Kingdom 647 US: United States 648 VIVE: First Basic Survey on Life Situation, State of Health, and Quality of Life of Childhood Cancer 649 Survivors in Germany 650 VT: Vitality 651 652 ACKNOWLEDGEMENTS 653 We gratefully acknowledge the participation of survivors of childhood and adolescent cancer and their 654 families in this research. 655 656 PanCareLIFE (Grant no. 602030) is a collaborative project in the 7th Framework Program of the 657 European Union. Project partners are Universitätsmedizin der Johannes Gutenberg-Universität Mainz, 658 Germany (PD Dr P Kaatsch, Dr D Grabow); Boyne Research Institute, Drogheda, Ireland (Dr J Byrne, 659 Ms H Campbell); Pintail Ltd., Dublin, Ireland (Mr C Clissmann, Dr K O’Brien); Academisch Medisch 660 Centrum bij de Universiteit van Amsterdam, the Netherlands (Prof Dr LCM Kremer); Universität zu 661 Lübeck, Germany (Prof T Langer); Stichting VU-VUMC, Amsterdam, the Netherlands (Dr E van Dulmen662 den Broeder, Dr MH van den Berg); Erasmus Universitair Medisch Centrum, Rotterdam, the 663 Netherlands; and Princess Maxima Center for Pediatric Oncology, Utrecht, the Netherlands (Prof Dr MM 664 van den Heuvel-Eibrink); Charité-Universitätsmedizin Berlin, Germany (Prof Dr A Borgmann-Staudt); 665 Westfälische Wilhelms-Universität Münster, Germany (Prof Dr A am Zehnhoff-Dinnesen); Universität 666 Bern, Switzerland (Prof Dr CE Kuehni); Istituto Giannina Gaslini, Genoa, Italy (Dr R Haupt); Fakultni 667 nemocnice Brno, Czech Republic (Dr T Kepak); Centre Hospitalier Universitaire Saint-Étienne, Saint668 Étienne, France (Dr C Berger); Kraeftens Bekaempelse, Copenhagen, Denmark (Dr JF Winther); 669 Fakultni nemocnice v Motole, Prague, the Czech Republic (Dr J Kruseova); Universitaetsklinikum Bonn, 670 Bonn, Germany (Dr G Calaminus, K Baust); and University Hospital Essen, Essen, Germany (Prof U 671 Dirksen). The EU Commission takes no responsibility for any use made of the information set out. 672 673 In the Czech Republic, at the University Hospital of Brno we thank the team of collaborators from the 674 Children’s Hospital in Brno, the Masaryk University in Brno and Brno survivors’ association Together 675 Towards a Smile: H. Hrstková, V. Bajčiová, D. Hošnová, Z. Kuttnerová, M. Blanářová, R. Mazúr, I. 676 Mikulec, E. Bučková, L. Červinková, E. Bařinová, I. Krupková, E. Novotná, P. Chloupková, Z. 677 Wimmerová, L. Štrublová and numerous other external collaborators and supporters. At the University 678 . CC-BY-NC-ND 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted January 23, 2024. ; https://doi.org/10.1101/2024.01.23.24301414doi: medRxiv preprint 22 Version 4 January 22, 2024 Hospital of Prague LTFU care registry, team of the Prague Childhood Cancer Survivor Study: Keslová 679 P, Ganevová M, Reichlová V, Bašeová J, Nováková L, Douchová M, Čepelová M. 680 681 In France, the team of the long-term follow-up study (SALTO): Faure-Conter C. (Lyon), Corradini N. 682 (Lyon), Plantaz D. (Grenoble), Tarral E. (Grenoble), Durieu I. (Lyon), Guichard I. (Saint-Etienne), Odier 683 F. (Saint-Etienne), Gauthier N. (Lyon), Métral P. (Lyon), Mercier M. (Lyon), Billet S. (Grenoble), Celette 684 C. (Grenoble), Schiff I. (Grenoble), Loubier A. (Saint-Étienne). 685 686 In Germany, we thank the VIVE group: G. Calaminus (Bonn) U Creutzig (Hannover), P Kaatsch (Mainz), 687 T Langer (Lübeck), K. Baust (Bonn), J. Dobke (Berlin) I. Jung (Mainz), I. Kerenyi (Mainz), C. Spix 688 (Mainz), C. Teske (Bonn), M. Zimmermann (Hannover). 689 690 In The Netherlands, we thank the DCOG LATER consortium: L.C.M. Kremer, E. van Dulmen-den 691 Broeder, M.M. van den Heuvel-Eibrink, W.J.E. Tissing, J. Loonen, D. Bresters, B. Versluijs, M.A. 692 Grootenhuis, M. van der Heiden-van der Loo, F. van Leeuwen, S.J.C. Neggers, H.J.H. van der Pal, C.M. 693 Ronckers, A.C.H. de Vries, G.O.R Janssens, J. den Hartogh. H.M. van Santen, M.A. Veening, M. 694 Louwerens, S.M.F. Pluijm. 695 696 In Switzerland, we thank the Swiss Childhood Cancer Survivor Study team, the Swiss Paediatric 697 Oncology Group data managers, and the Swiss Childhood Cancer Registry study team. We also thank 698 Marcel Zwahlen for statistical advice. We thank Kristin Marie Bivens for her editorial guidance and 699 suggestions. 700 701 702 703 704 . CC-BY-NC-ND 4.0 International licenseIt is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint The copyright holder for thisthis version posted January 23, 2024. ; https://doi.org/10.1101/2024.01.23.24301414doi: medRxiv preprint