File
Melman A1, Moseley A1, Elkins M1, Kamper SJ1
1Institute for Musculoskeletal Health, School of Public Health, University of Sydney, Physiotherapy Evidence Database, Sydney, Australia
Background: 90% of trials, reviews and guidelines evaluating physiotherapy interventions are published in English but only 20% of the world's population speaks English, and 3% have English as their first language. To help reduce this language barrier, the Physiotherapy Evidence Database (PEDro) web-site and/or training videos are available in 13 languages: English, Chinese simplified characters, Chinese traditional characters, French, Dutch, German, Italian, Japanese, Korean, Spanish, Portuguese, Tamil and Turkish. Speakers of other languages sometimes use an online translation tool (e.g., Google Translate), making it possible to identify their preferred language. Analysis of usage of PEDro by language could identify underutilised segments and additional languages required.
Purpose: The aim of this paper was to quantify usage of the PEDro web-site and training videos by language.
Methods: The number of times each page was accessed by a user (page views) on PEDro was downloaded using Google Analytics for the period 1 July 2017 to 30 June 2018. Pages containing error messages, uploads, or viewed for 5 seconds were excluded from the analysis (these represented 0.3% of page views). The remaining page views were categorised by language, and proportion of total page views was calculated. Additionally, pages viewed using an online translation tool were analysed by language. The number of views of training videos in each language in the past 3 years was downloaded from YouTube and analysed as the proportion of total views.
Results: Among the 2,840,053 page views, the proportion for each language was: English 59%, Portuguese 16%, Spanish 12%, German 4%, French 3%, Japanese 2%, Italian 1%, Chinese traditional characters 0.8%, Turkish 0.8%, Korean 0.7%, Chinese simplified characters 0.3%, and Tamil 0.2%. Users translated pages into 29 languages (0.02% of views), the top seven being Greek, Polish, Indonesian, Russian, Croatian, Dutch and Arabic. Amongst these, Arabic and Russian were identified as populous world languages with concurrent lower levels of English proficiency (when compared against the English Proficiency Index).
Among the 78,150 views of training videos, the proportion for each language was: English 59%, Portuguese 20%, Spanish 6%, French 6%, German 4%, Italian 3%, Tamil 0.6%, Japanese 1%, Korean 0.4%, Chinese traditional characters 0.3%, Chinese simplified characters 0.3%, and Dutch 0.2%.
Conclusion(s): Substantial utilisation of the non-English pages and videos suggests that translation has reduced the language barrier for many users, particularly Portuguese and Spanish speakers. Lower utilisation rates for some languages (particularly Chinese simplified characters) highlight the need to work collaboratively with WCPT member organisations to promote this valuable resource. Consideration of the online translation data in conjunction with the world's most populous languages and corresponding levels of English proficiency suggests Arabic and Russian as potential choices for the expansion of PEDro.
Implications: PEDro will direct future language resources based on this utilisation and needs-based analysis.
Keywords: Evidence Based Practice (EBP), Language, Barriers
Funding acknowledgements: PEDro is funded by industry groups: the APA, MAIC, TAC, Cerebral Palsy Alliance, CSP, and 46 other WCPT member organisations.
Purpose: The aim of this paper was to quantify usage of the PEDro web-site and training videos by language.
Methods: The number of times each page was accessed by a user (page views) on PEDro was downloaded using Google Analytics for the period 1 July 2017 to 30 June 2018. Pages containing error messages, uploads, or viewed for 5 seconds were excluded from the analysis (these represented 0.3% of page views). The remaining page views were categorised by language, and proportion of total page views was calculated. Additionally, pages viewed using an online translation tool were analysed by language. The number of views of training videos in each language in the past 3 years was downloaded from YouTube and analysed as the proportion of total views.
Results: Among the 2,840,053 page views, the proportion for each language was: English 59%, Portuguese 16%, Spanish 12%, German 4%, French 3%, Japanese 2%, Italian 1%, Chinese traditional characters 0.8%, Turkish 0.8%, Korean 0.7%, Chinese simplified characters 0.3%, and Tamil 0.2%. Users translated pages into 29 languages (0.02% of views), the top seven being Greek, Polish, Indonesian, Russian, Croatian, Dutch and Arabic. Amongst these, Arabic and Russian were identified as populous world languages with concurrent lower levels of English proficiency (when compared against the English Proficiency Index).
Among the 78,150 views of training videos, the proportion for each language was: English 59%, Portuguese 20%, Spanish 6%, French 6%, German 4%, Italian 3%, Tamil 0.6%, Japanese 1%, Korean 0.4%, Chinese traditional characters 0.3%, Chinese simplified characters 0.3%, and Dutch 0.2%.
Conclusion(s): Substantial utilisation of the non-English pages and videos suggests that translation has reduced the language barrier for many users, particularly Portuguese and Spanish speakers. Lower utilisation rates for some languages (particularly Chinese simplified characters) highlight the need to work collaboratively with WCPT member organisations to promote this valuable resource. Consideration of the online translation data in conjunction with the world's most populous languages and corresponding levels of English proficiency suggests Arabic and Russian as potential choices for the expansion of PEDro.
Implications: PEDro will direct future language resources based on this utilisation and needs-based analysis.
Keywords: Evidence Based Practice (EBP), Language, Barriers
Funding acknowledgements: PEDro is funded by industry groups: the APA, MAIC, TAC, Cerebral Palsy Alliance, CSP, and 46 other WCPT member organisations.
Topic: Professional issues; Globalisation: health systems, policies & strategies; Education: continuing professional development
Ethics approval required: No
Institution: USYD
Ethics committee: USYD
Reason not required: Ethics approval was not required as the project consisted of analysing anonymous user data obtained via Google Analytics.
All authors, affiliations and abstracts have been published as submitted.
