Age and Ageing 2019; 48: 506–512 © The Author(s) 2018. Published by Oxford University Press on behalf of the British Geriatrics doi: 10.1093/ageing/afz047 Society. All rights reserved. For permissions, please email: journals.permissions@oup.com Published electronically 13 May 2019 Controlled trial of balance training using a video game console in community-dwelling older adults PILAR MONTERO-ALÍA1,2,3, RAMÓN MIRALLES-BASSEDA3,4, TOMÁS LÓPEZ-JIMÉNEZ5,3, LAURA MUÑOZ-ORTIZ6, MERCÈ JIMÉNEZ-GONZÁLEZ2, JOSEP PRAT-ROVIRA7, JOSÉ LUÍS ALBARRÁN-SÁNCHEZ2, JOSEP MARIA MANRESA-DOMÍNGUEZ1, CELIA MARIA ANDREU-CONCHA2, M. CARMEN RODRÍGUEZ-PÉREZ2, JUAN JOSÉ MARTÍ-CERVANTES2, LIDIA SAÑUDO-BLANCO2, CARLOS ANDRÉS SÁNCHEZ-PÉREZ2, SÒNIA DOLADER-OLIVÉ2, PERE TORÁN-MONSERRAT1 1 Unitat de Suport a la Recerca Metropolitana Nord (GRIDAES), Fundació Institut Universitari per a la recerca a l’Atenció Primària de Salut Jordi Gol i Gurina (IDIAPJGol), Mataró, Spain 2 Centre d’Atenció Primària La Riera (Mataró 1), Institut Català de la Salut, Barcelona, Spain 3 Department of Medicine, Universitat Autònoma de Barcelona, Bellaterra (Cerdanyola del Vallès), Spain 4 Direcció Clínica Territorial de Cronicitat, Gerència Territorial Metropolitana Nord, Institut Català de la Salut, Barcelona, Spain 5 Fundació Institut Universitari per a la recerca a l’Atenció Primària de Salut Jordi Gol i Gurina (IDIAPJGol), Barcelona, Spain 6 Agència de Qualitat i Avaluació Sanitàries de Catalunya (AQuAS), Barcelona, Spain 7 Residència de Gent Gran de Mataró. Departament de Treball, Afers Socials i Famílies, Generalitat de Catalunya, Spain Address correspondence to: Pilar Montero-Alía. Email: pmontero.bnm.ics@gencat.cat Abstract Background: gamification is a potentially attractive option for improving balance and reducing falls. Objectives: to assess the effect of balance training using the NintendoTM Wii game console on balance (primary outcome), falls and fear of falling. Design: quasi-randomised, open-label, controlled clinical trial in parallel groups, carried out on community-dwelling patients over 70 years, able to walk independently. Participants were assigned 1:1 to the intervention or control group. Balance train- ing was conducted using the Nintendo WiiFitTM twice a week for 3 months. Balance was assessed using the Tinetti balance test (primary outcome), the unipedal stance and the Wii balance tests at baseline, 3 months and 1 year. Falls were recorded and Fear of falling was assessed by the Falls Efficacy Scale (Short-FES-I). Results: 1,016 subjects were recruited (508 in both the intervention and the control group; of whom 274 and 356 respect- ively completed the 3-month assessment). There was no between-group difference in the Tinetti balance test score, with a baseline mean of 14.7 (SD 1.8) in both groups, and 15.2 (1.3) at 3 months in the intervention group compared to 15.3 (1.7) in controls; the between-group difference was 0.06 (95% CI 0.30–0.41). No differences were seen in any of the other bal- ance tests, or in incident falls. There was a reduction in the fear of falling at 3 months, but no effect at 1 year. Conclusions: the study found no effect of balance training using the NintendoTM Wii on balance or falls in older community-dwelling patients. The study protocol is available at clinicaltrials.gov under the code NCT02570178. Keywords quasi-randomized clinical trial, balance, falls, Nintendo WiiFitTM, older adults, community-dwelling Key points • No improvements were detected on any of the balance tests at the end of training. • No fall reduction was detected after balance training using the NintendoTM Wii console in older adults. 506 Controlled trial of balance training using a video game console in community-dwelling older adults • A reduced fear of falling was found after balance training using the NintendoTM Wii console in older adults at 3 months, but no longer at 1 year. This may be a chance finding. Introduction WBB in the Wii balance test from 0 to 100 points (higher scores indicate better balance) [15], fear of falling and self- Background efficacy regarding falls through the Falls Efficacy scale or Falls and their consequences constitute a major health Short-FES-I (7–28 points, lower scores indicating less fear of problem for older adults, due to both potential physical falling) [16], and the number of falls during the study period. injuries as well as psychological trauma with fear of falling, The study protocol was approved by the ANONYMIZED. loss of self-confidence, loss of autonomy and reduced qual- All data were anonymized and the confidentiality of EHR ity of life [1]; there is a substantial financial burden asso- was respected at all times, in accordance with national and ciated with hospitalisation, institutionalisation and even international law. death [2]. Over the course of a year, 30% of adults over 65 years of age fall at least once [3]. It has been suggested that balance training using home Participants video consoles may be an effective intervention for older adults Inclusion criteria to improve balance and reduce falls [4, 5]. The NintendoTM Patients 70 years of age or older who were able to walk Wii console, together with the Nintendo WiiFitTM game and independently (with or without walking aids) attending one Wii Balance Board, offers the possibility of following a balance of five primary care centres. training exercise programme [6–9]. This is a low-cost technol- ogy that can be used recreationally both on an individual and group basis. However there is no consensus on what exercises, Exclusion criteria doses or maintenance times should be used [10, 11], and the Home care patients, patients receiving rehabilitation treat- largest clinical trials published to date have been conducted on ment for gait, patients with moderate cognitive impairment samples of just 40–60 people, with uncertainty about any bene- (Pfeiffer’s test ≥5 [17]), patients in terminal phases of dis- fits [4, 5]. ease, patients without a telephone, patients with difficulties communicating (sensory impairment and language barriers). Objectives The primary objective of this study was to determine Interventions whether balance training, delivered using the Nintendo Balance training was performed by using the NintendoTM Wii WiiFitTM video console, improved balance in older adults. console and the WBB. We used eight out of the nine exercises Secondary objectives included assessment of effect on falls in the balance category of the Nintendo WiiFitTM game (bal- and fear of falling. ance bubble, soccer heading, ski jump, table tilt, ski slalom, penguin slide, snowboard slalom, tightrope walk) [18]. Two Methods 30-min sessions per week during three months were carried out. Training was delivered in groups of four persons and the Trial design same exercises at the same time were performed. The groups Quasi-randomized, open (non-blinded), controlled clinical were guided by support monitors who had received standar- trial in parallel groups (allocation ratio 1:1) that compared an dized training. During the training sessions, participants stood intervention group, which received balance training using the barefoot on the WBB and performed the various exercises Nintendo WiiFitTM, with a control group which received included in the balance category of the Nintendo WiiFitTM usual care. Potential participants were identified from the pri- game following the instructions displayed on the screen. mary care lists, and recruited via telephone calls by clerical Chairs were available for anyone who wanted one to guarantee staff. A detailed description of the study protocol was pub- their safety and catch themselves in case they lose their bal- lished at clinicaltrials.gov under the code NCT02570178, as ance. The monitor was in charge of handling the consoles, col- of 2 July 2015 [12]. lecting data, keeping order and managing the time so that all The primary outcome was the Tinetti’s balance test exercises were completed in each session. The number of (scored from 0 to 16 points; higher scores indicate better repetitions of each exercise varied depending on the partici- balance); secondary outcomes included the Tinetti gait test pants’ agility, but the total time dedicated to each exercise was (0–12 points), total Tinetti score (balance plus gait; 0–28 the same for all participants in the group. The intervention points) [13]; the unipedal stance test (patient able to stand up was considered to be complete if participants attended at least on one foot without any support for 5 s [14], assessment of 80% of the sessions. The second visit was performed after the balance through the balance percentage calculated by the 3-month period (up on completing the intervention) and the 507 P. Montero-Alía et al. Assessed for selection N = 1236 Excluded (n = 31) Refused (n = 228) Group allocation n = 977 INTERVENTION GROUP CONTROL GROUP n = 508 Baseline assessment 39 refused to proceed with the study after group allocation Drop outs Drop outs n = 508 n = 234 n = 113 Baseline assessment Unreachable (24) Unreachable (18) n = 469 Exitus (2) Exitus (1) Lack of interest (55) Lack of interest (62) Lack of time (26) Lack of time (5) Distance (6) Distance (1) Health problems (77) Health problems (14) 3 months Family problems (9) Family problems (4) 3 months Moved (8) Moved (3) n = 274 Training problems (17) Other (5) n = 356 Other (10) Drop outs Drop outs n = 55 n = 64 Unreachable (2) Unreachable (7) 1 year Exitus (1) Exitus (0) Lack of interest (7) Lack of interest (18) 1 year Lack of time (0) Lack of time (0) n = 236 Distance (3) Distance (0) n = 302 (17 participants who Health problems (5) Health problems (7) (10 participants who dropped out at 3 months Family problems (1) Family problems (0) dropped out at 3 months could be evaluated at 1 Moved (1) Moved (1) could be evaluated at 1 Training problems (2) Other (31) year) year) Other (3) Figure 1. Flow chart. third visit 1 year after the beginning of the study (9 months inflated to allow for loss of up to 20% of patients in post-intervention). Thus, the two groups participated in the follow-up. More detailed information regarding sample size study for one year. Balance tests were repeated at every visit can be found in the study protocol [12]. for both groups. Participants were also asked for new falls in every visit. The intervention was performed in a nursing home and the follow-up visits in five primary Healthcare Randomisation centres. At the initial visit, having revised the inclusion/exclusion criteria and signing the informed consent, each participant was assigned an identification code. This code contained Outcomes four digits: the first referred to their primary healthcare cen- The primary outcome was the change in the Tinetti balance tre and the rest were sequential numbers. Participants were test from baseline to the 3-month follow-up visit. Secondary assigned by the care provider (nursing staff) to the interven- outcomes included the unipedal stance test and Wii balance tion or control group based on whether or not their code test, falls and fear of falling. was odd or even (odds were the intervention group and evens were the control group). Sample size We calculated that 380 patients per group were required to Statistical analysis give 80% power at P = 0.05 (2-tailed) to detect a between- Descriptive statistics were used to summarize overall data, group difference of 0.5 points in the sub-scale of Tinetti’s describing demographic characteristics of the groups. stationary balance test (score range 0–16 points), assuming Categorical variables were expressed as absolute number a standard deviation of 2.2. The study numbers were and percentages and continuous variables as means or 508 Controlled trial of balance training using a video game console in community-dwelling older adults Table 1. Baseline data. Whole cohort study Total Intervention Group Control Group N = 977 N = 508 N = 469 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Sex, n/N (%) Male 400/977 (40.9) 192/508 (37.8) 208/469 (44.3) Female 577/977 (59.1) 316/508 (62.2) 261/469 (55.7) Age, median (p25–p75) 75.3 (72.6–78.5) 75.1 (72.6–78.7) 75.4 (72.7–78.6) BMI, median (p25–p75) 27.8 (25.1–30.8) 27.8 (25.5–30.8) 27.5 (24.8–30.8) Underweight, n/N (%) 6/963 (0.6) 1/499 (0.2) 5/464 (1.1) Normal weight, n/N (%) 225/963 (23.4) 103/499 (20.6) 122/464 (26.3) Overweight, n/N (%) 432/963 (44.9) 239/499 (47.9) 193/464 (41.6) Obesity, n/N (%) 300/963 (31.1) 156/499 (31.3) 144/464 (31.0) Cognitive impairment, n/N (%) Without 902/934 (96.6) 461/475 (97.1) 441/459 (96.1) Mild 32/934 (3.4) 14/475 (2.9) 18/459 (3.9) IADL Scale, n/N (%) 204/970 (21.0) 106/503 (21.1) 98/467 (21.0) Dependent (0–7) 766/970 (79.0) 397/503 (78.9) 369/467 (79.0) Independent (8) Tinetti’s test, median/N (p25–p75) Balance 16/963 (14–16) 15/499 (14–16) 16/464 (14–16) Gait 12/962 (11–12) 12/499 (11–12) 12/463 (12-12) Total 27/960 (26–28) 27/497 (25–28) 28/463 (26–28) Unipedal stance test, n/N (%) Is able 762/962 (79.2) 392/501 (78.2) 370/461 (80.3) Is unable 200/962 (20.8) 109/501 (21.8) 91/461 (19.7) Balance Wii test %, median/N (p25–p75) 70/965 (37–87) 70/502 (35–87) 71/463 (39–86) Patients who had fallen during past year, n/N (%) No 658/970 (67.8) 312/504 (61.9) 346/466 (74.2) Yes 312/970 (32.2) 192/504 (38.1) 120/466 (25.8) Has received rehabilitating treatment for gait during the past year, n/N (%) No 907/938 (96.7) 466/485 (96.1) 441/453 (97.4) Yes 31/938 (3.3) 19/485 (3.9) 12/453 (2.6) Needs technical assistance to walk, n/N (%) No 844/965 (87.5) 435/502 (86.7) 409/463 (88.3) Yes 121/965 (12.5) 67/502 (13.3) 54/463 (11.7) Fear of falling, median/N (p25–p75) 8/934 (7–10) 8/475 (7–11) 8/459 (7–10) medians and standard deviation (SD). To compare the dif- technical assistance to walk and had better scores on the ferences between groups before and after intervention, fear of falling assessment (median Short-FES-I = 8). A Student’s T-test and McNemar’s test were used. A second- record of falls in the previous year was observed in 192 ary analysis was carried out on those patients who com- (38%) of the intervention group compared to 120 (26%) of pleted the study intervention (per-protocol analysis). All the controls, [192 (38%) vs. 120 (26%)] (Table 1 and statistical tests were two-sided and the level of significance Appendix 1 available in Age and Ageing online). was set at 5%. Statistical analyses were performed using Stata/SE, version 14 for Windows (Stata Corp. LP, College Station, TX, USA). Outcomes We observed no difference between groups on any of the balance tests post-intervention (at 3-month and 1-year Results follow-up visits), even though the intra-group balance A total of 1016 participants were recruited between scores improved at 3 months and 1 year. We found a October 2013 and November 2016; 508 were assigned to reduction fear of falling in 0.63 points (P = 0.008) at 3 the intervention group and 508 to the control group, of months in the intervention group compared to controls whom 469 proceeded with the study (Figure 1). that was not maintained in the 1-year follow-up visit (Table 2). The per-protocol analysis gave very similar results (see Appendix 2 available in Age and Ageing online). Baseline data We found 92 participants who fell at 3 months [28 Five hundred and seventy-seven (59%) participants were (12.0%) vs. 64 (17.6%)] and 126 at 1 year [47 (23.6%) vs. female; the median age was 75 (73–79); 121 (13%) required 79 (24.8%)]. 509 P. Montero-Alía et al. Table 2. Outcomes between groups at 3 months and 1 year. Whole cohort study. Intervention Control Differences in groups Differences between groups Mean (SD) or n Mean (SD) or n Difference in % IG (95% Difference in % CG (95% Change difference % IG–CG P- (%) (%) CI)/N CI)/N (95%CI) value . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Tinetti’s test, balance Baseline 14.9 (1.7)/499 14.8 (2.0)/464 3 months 15.2 (1.4)/293 15.1 (1.9)/342 0.28 (0.09; 0.47)/293 0.30 (0.10; 0.50)/342 −0.02 (−0.30; 0.26) 0.878 1 year 14.9 (1.7/250) 15.1 (1.9)/295 0.08(−0.16; 0.32)/250 0.25 (0.04; 0.46)/295 −0.17 (−0.47; 0.14) 0.280 Tinetti’s test, gait Baseline 11.4 (1.3)/499 11.4 (1.8)/463 3 months 11.5 (1.3)/288 11.4 (1.9)/336 0.16 (−0.03; 0.35)/288 0.06 (−0.13; 0.25)/336 0.10 (−0.17; 0.37) 0.466 1 year 11.1 (2.0)/247 11.4 (1.9)/294 −0.15 (−0.44; 0.14)/247 0.07 (−0.17; 0.32)/294 −0.25 (−0.67; 0.17) 0.250 Tinetti’s test, total Baseline 26.2 (2.6)/497 26.2 (3.5)/463 3 months 26.7 (2.4)/288 26.5 (3.5)/333 0.43 (0.10; 0.76)/288 0.38 (0.04; 0.72)/333 0.05 (−0.43; 0.53) 0.835 1 year 26.1 (3.3)/246 26.4 (3.4)/294 −0.05 (−0.50; 0.40)/246 0.33 (−0.07; 0.73)/294 −0.38 (−0.98; 0.23) 0.220 Unipedal stance test, if able Baseline 231 (79.7)/501 276 (81.7)/461 3 months 251 (86.6)/290 292 (86.4)/338 6.90 (1.20; 12.59)/290 4.73 (0.54; 8.93)/338 2.16 (−4.36; 8.69) 0.515 1 year 190 (76.3)/249 231 (79.1)/292 −8.03 (−7.60; 6.00)/249 −2.05 (−6.84; 2.73)/292 1.25 (−6.40; 8.90) 0.748 Balance Wii test % Baseline 62.9 (29.3)/502 63.8 (28.4)/463 3 months 52.1 (32.3)/238 53.5 (30.2)/324 −10.8 (−15.9; −5.7)/238 −10.4 (−14.7; −6.1)/324 −0.4 (−7.1; 6.2) 0.895 1 year 49.5 (31.5)/258 58.8 (69.1)/293 −8.74 (−13.9; −3.59)/258 −2.18 (−10.6; 6.22)/293 −6.6 (−16.7; 3.6) 0.205 Fear of falling Baseline 9.5 (3.2)/475 9.0 (3.1)/459 3 months 8.9 (2.4)/269 9.0 (3.0)/339 −0.61 (−0.89; −0.32)/269 0.06 (−0.19; 0.31)/339 −0.66 (−1.05; −0.28) 0.001 1 year 9.3 (2.7)/177 8.8 (3.1)/249 −0.60 (−1.06; −0.15)/177 −0.42 (−0.73; −0.11)/249 −0.18 (−0.71; 0.35) 0.499 IG, intervention group; CG, control group; SD, standard deviation; CI, confidence interval. Drop outs and harms To date, group and home exercise programmes contain- Three hundred and forty-seven (36%) participants dropped ing some balance and strength training exercises, such as out of the study at 3 months, 234 (46%) in the intervention Tai Chi, are known to be effective in reducing falls accord- group vs. 113 (24%) of the controls. The most common rea- ing to the Cochrane review [21]. For multifactorial interven- sons for dropping out were: lack of interest [117 (33.7%); 55 tions, it is not known whether their efficacy may depend on (23.5%) vs. 62 (54%)]; health problems [91 (26%); 77 (33%) factors that have not yet been determined. vs. 14 (12.4%)]; lack of time [31 (9%); 26 (11%) vs. 5 (4.4%)], It is possible that the balance variables analysed were com- unreachable [42 (12%); 24 (10%) vs. 18 (16%)], problems promised by the ceiling effect [22], making it impossible to with the training programme [17 (4.8%)]. The reasons for observe improvements. It must be considered that the study dropping out of study due to the training itself can be cate- population consisted of healthy, ‘robust’ older adults who live gorised in three groups: pain (n = 9), difficulty performing in the community and who already presented good results on exercises (due to difficulties with mobility n = 4 due to a sen- all the balance tests at the baseline visit. The percentages of sation of instability n = 3) and exhaustion (n = 1). (Figure 1 falls in both groups (38% intervention group and 26% control and Appendix 3). group) are similar to other studies in this age group; in our region, the percentage of falls in patients older than 65 ranges from 14% to 46% [3]. Discussion It is worth noting that there were baseline imbalances Our study showed no beneficial effect on balance or falls of between our two groups. The intervention group had more 3 months of balance training using the Nintendo Wii video women and a higher percentage with previous falls. This console. We found a reduction in fear of falling at 3 months could be a source of bias. but this was no longer present at 1 year; this is most likely to be a chance finding. To date, this is the largest clinical trial of its kind [10, 11, 19, 20]. Exercises were performed Limitations safely; no accidents were recorded during training and the Our trial had a number of limitations. First, the non-random monitors did not have to prevent any falls. We did not per- method of group allocation using odd and even numbers is form warm-up exercises prior to the training. prone to bias, as the allocation of patients can be anticipated. 510 Controlled trial of balance training using a video game console in community-dwelling older adults Blinding of patients was not possible, and there was no for- revisión sistemática, 2007. Rev Esp Salud Pública 2008; 82: mal blinding of outcome assessors. It is also worth noting 43–55. that there were baseline imbalances between our two groups. 4. Rendon AA, Lohman EB, Thorpe D, Johnson EG, The intervention group had more women and a higher per- Medina E, Bradley B. The effect of virtual reality gaming on centage with previous falls. There was substantial attrition, dynamic balance in older adults. Age Ageing 2012; 41: 549–52. both immediately after allocation and during follow-up; 39 5. Fu AS, Gao KL, Tung AK, Tsang WW, Kwan MM. The participants in the control group refused to participate after effectiveness of exergaming training for reducting fall risk recruitment and assignation of codes. Third, adherence to and incidence among the frail older adults with a history of the training programme was a further problem (mostly due falls. Arch Phys Med Rehabil 2015; 96: 2096–102. to health problems or lack of interest). 6. Williams MA, Soiza RL, Jenkinson AM, Stewart A. In conclusion, our study failed to demonstrate benefit Exercising with Computers in Later Life (EXCELL) – pilot from balance training using the Nintendo Wii video console and feasibility study of the acceptability of the Nintendo® in terms of any change in the balance tests or incident falls; WiiFit in community-dwelling fallers. BMC Res Notes 2010; the reduction in fear of falling at 3 months may be a chance 3: 238. finding. 7. Clark RA, Bryant AL, Pua Y, McCrory P, Bennell K, Hunt M. Validity and reliability of the Nintendo Wii Balance Board for assessment of standing balance. Gait Posture 2010; 31: Supplementary Data: Supplementary data mentioned in 307–10. the text are available to subscribers in Age and Ageing online. 8. Park DS, Lee G. Validity and reliability of balance assessment Acknowledgements: We would like to thank the Mataró software using the Nintendo Wii Balance board: usability and validation. J Neuroeng Rehabil 2014; 11: 99. Nursing Home of the Ministry of Labor, Family and Social 9. Leach JM, Mancini M, Peterka R, Hayes TL, Horak FB. Affairs for allowing us to carry out the training programme Validating and calibrating the Nintendo Wii Balance Board to in their facilities. We would also like to thank all the partici- derive Reliable Center of pressure Measures. Sensors 2014; pants for collaborating on this project. 14: 18244–67. Declaration of Conflict of Interest: Nintendo lent the 10. Manlapaz DG, Sole G, Jayakaran P, Chapple CM. A narrative synthesis of Nintendo Wii Fit Gaming protocol in addressing equipment to carry out the training programme. This equip- balance among healthy older adults: What system works? ment was be returned up on completion of the study, as sti- Games Health J 2017; 6: 65–74. pulated in the contract. 11. Shier V, Trieu E, Ganz D. Implementing exercise programs Declaration of Sources of Funding: This project received to prevent falls: systematic descriptive review. Inj Epidemiol a grant for primary care research from the Fundació 2016; 3: 16. 12. Montero-Alía P, Muñoz-Ortiz L, Jiménez-González M et al. Institut Universitari per a la recerca a l’Atenció Primària de Study protocol of a randomized clinical trial evaluating the Salut Jordi Gol i Gurina (IDIAPJGol) in 2010 (Barcelona, effectiveness of a primary care intervention using the Spain) and a Goncal Calvo I Queralto grant from the NintendoTM Wii console to improve balance and decrease Acadèmia de Ciències Mèdiques i de la Salut de Catalunya i falls in elderly. BMC Geriatr 2016; 16: 8. Balears, Filial Maresme 2011 (Mataró, Spain). Additionally, 13. Tinetti ME. Performance-oriented assessment of mobility it received funding from the Instituto de Salud Carlos III problems in elderly patients. J Am Geriatr Soc 1986; 34: (Madrid, Spain) in the 2012 call for Strategic Action in 119–26. Health Grants, within the framework of the Spanish 14. Vellas B, Wayne SJ, Romero L, Baumgartner RN, Rubenstein National Plan for Scientific Research Development and LZ, Garry PJ. One-leg balance is an important predictor of Technical Innovation 2008–11; under file code PI12/ injurious falls in older persons. J Am Geriatr Soc 1997; 45: 01677, and cofounded by the European Union through the 735–8. 15. GuíasNintendo [http://www.guiasnintendo.com/2a_WII/ European Regional Development Fund (ERDF). wii_fit/wii_fit_sp/test_equilibrio.html] 16. Kempen GI, Yardley L, van Haastregt JC et al. The Short FES-I: a shortened version of the falls efficacy scale- References international to assess fear of falling. Age Ageing 2008; 37: 45–50. 1. World Health Organization. WHO Global Report on Falls 17. Pfeiffer E. A short portable mental status questionnaire for Prevention in Older Age. Geneva, Switzerland: World the assessment of organic brain deficit in elderly patients. Health Organization, 2008. http://www.who.int/ageing/ J Am Geriatr Soc 1975; 23: 433–41. publications/Falls_prevention7March.pdf. Available from. 18. Guías Nintendo [http://wiifit.com/training/balance-games. 2. Gillespie LD, Robertson MC, Gillespie WJ et al. Interventions html] for preventing falls in older people living in the community. 19. Padala KP, Padala PR, Lensing SY et al. Efficacy of Wii-Fit Cochrane Database Syst Rev 2012; 12: CD007146. on static and dynamic balance in community dwelling older 3. Da Silva Gama ZA, Gómez Conesa A, Sobral Ferreira M. veterans: a randomized controlled pilot trial. J Aging Res Epidemiología de caídas de ancianos en España. Una 2017; 2017: 4653635. 511 L. Rydén et al. 20. Roopchand- Martin S, Mc Lean R, Gordon C, Nelson G. Cochrane Database Syst Rev 2012; CD007146. doi:10.1002/ Balance training with Wii Fit plus for community-dwelling 14651858.CD007146.pub3. persons 60 years and older. Games Health J 2015; 4: 22. Mancini M, Horak FB. The relevance of clinical balance 247–52. assessment tools to differentiate balance deficits. Eur J Phys 21. Gillespie LD, Robertson MC, Gillespie WJ et al. Interventions Rehabil Med 2010; 46: 239–48. for preventing falls in older people living in the community. Received 7 June 2018; editorial decision 29 March 2019 Age and Ageing 2019; 48: 512–517 © The Author(s) 2019. Published by Oxford University Press on behalf of the British Geriatrics doi: 10.1093/ageing/afz033 Society. All rights reserved. For permissions, please email: journals.permissions@oup.com Published electronically 9 April 2019 Agreement between self-reports, proxy-reports and the National Patient Register regarding diagnoses of cardiovascular disorders and diabetes mellitus in a population-based sample of 80-year-olds LINA RYDÉN1, ROBERT SIGSTRÖM1, JOHAN NILSSON1, VALTER SUNDH1, HANNA FALK ERHAG1, SILKE KERN1, MARGDA WAERN1, SVANTE ÖSTLING1, KATARINA WILHELMSON2,3, INGMAR SKOOG1 1 Neuropsychiatric Epidemiology Unit, Department of Psychiatry and Neurochemistry, Institute of Neuroscience and Physiology, Sahlgrenska Academy, Centre for Ageing and Health (AgeCap) at the University of Gothenburg, Sweden 2 Department of Health and Rehabilitation, Institute of Neuroscience and Physiology, Sahlgrenska Academy, Centre for Ageing and Health (AgeCap) at the University of Gothenburg, Sweden 3 Department of Geriatrics, Sahlgrenska University Hospital, Gothenburg, Sweden Address correspondence to: Lina Rydén. Tel: +46 (0) 31 3438646; Email: lina.ryden@gu.se Abstract Background: cognitive impairment is common among older adults, necessitating the use of collateral sources in epidemio- logical studies involving this age group. The objective of this study was to evaluate agreement between self- and proxy- reports of cardiovascular disorders and diabetes mellitus in a population-based sample of 80-year-olds. Further, both self- and proxy-reports were compared with hospital register data. Methods: data were obtained from the Gothenburg H70 Birth Cohort Studies in Sweden. The study had a cross-sectional design and information was collected through semi-structured interviews in 2009–2012 from participants born in 1930 (N = 419) and their proxy informants. The National Patient Register provided diagnoses registered during hospital stays. Agreement was measured with Kappa values (K ). Results: agreement between self- and proxy-reports was substantial for diabetes mellitus (K = 0.79), atrial fibrillation (K = 0.61), myocardial infarction (K = 0.75), angina pectoris (K = 0.73) and hypertension (K = 0.62), and fair for intermittent claudication (K = 0.38) and heart failure (K = 0.40). Compared to the National Patient Register, a large proportion of those with a hospital discharge diagnosis were also self- and proxy-reported. Conclusions: proxy informants can be an important source of information, at least for well-defined conditions such as myocardial infarction, angina pectoris and diabetes mellitus. 512