������Ƶ

PDF [ - 94.83 KB]

[1]

Introduction

This document addresses the growing need for translating CAHPS surveys from English into a variety of languages in a cost-effective, yet culturally competent manner. It provides the ������Ƶ (AHRQ) and users of CAHPS surveys with a set of guidelines for translating CAHPS instruments that aims to standardize the translation approach and ensure the quality of translations obtained.

The guidelines presented here are based on a review and analysis of the translation approach used by the CAHPS Cultural Comparability Team in translating CAHPS surveys into Spanish. This approach involves using two translators to each produce a forward translation and then having the two forward translations reviewed (by a separate bilingual reviewer) against each other and compared to the original English survey. The advantages of the approach advocated in this document (compared to a simple translation/back-translation approach, for example) include:

• Increased ability to identify and resolve translation errors (i.e., errors in syntax, grammar, or meaning).
• Increased ability to identify issues related to variations in terms or expressions used by subgroups of people in the target language,
• Increased ability to produce a translation that uses language that is more easily understood by a wide variety of speakers of the target language, and
• Increased ability to identify and resolve problems with the readability level of the translation.

The CAHPS Consortium has successfully used the approach recommended in these guidelines to produce Spanish translations of the CAHPS Hospital Survey, the CAHPS In-Center Hemodialysis Survey, the CAHPS Clinician & Group Survey, and various supplemental item sets for these surveys.

Please note: Nearly all CAHPS instruments are available in Spanish. If you cannot find a Spanish translation in the CAHPS Kits, contact the CAHPS User Network at 800-492-9261 or cahps1@ahrq.gov to find out if one will be available.

If you need a survey instrument in a language other than English or Spanish, we recommend the following process.

Obtain Forward Translations

• Obtain the most recent English-language version of the CAHPS survey (available on http://www.ahrq.gov/cahps/index.html).
• Review ������Ƶguidelines for selecting translators and translation reviewers (available on http://ahrq.gov/cahps/surveys-guidance/helpful-resources/translating/index.html) and select two translators and a translation reviewer. Provide the translators and reviewer with background information on the translation task:
  • Mode of survey administration,
  • Socio-demographic characteristics of the target population (e.g., whether the sample includes elderly respondents, non-elderly adults, or minors; average educational level or socioeconomic status of the target respondents; country or region of origin),
  • Target reading level (6th to 8th grade),
  • Region(s) of the country where the survey will be fielded, and
  • The type of language they should use (“broadcast” or standard language vs. sub-group or region-specific language).

What’s “Broadcast” Language?

In order to meet the needs of a broad range of speakers, the CAHPS Consortium generally strives to use “broadcast” or standard language (typically used in broadcast journalism) in both its English versions of surveys as well as in its translations. By “broadcast” or standard language, we mean the use of language that has a basic vocabulary, grammar, and syntax that cannot be attributed to a single region/country and is understood by the majority of speakers in the target language.

However, if a survey translation is going to be used in a particular country or region, it may be acceptable to use a translation that is appropriate for that particular country or region, provided that you maintain equivalence with the English version of the survey.

• Provide the translators and reviewer with a copy of another CAHPS survey translated into the target language, if available. This enables the translators to check for and continue to use standardized translations for response scales and other elements that are common across surveys, for example, translating “medicines” into Spanish as “medicinas”, rather than “medicamentos”. You may also want to provide a link to the Frequently Asked Questions on the CAHPS Web site (http://www.ahrq.gov/cahps/faq/index.html), which offer information on the principles underlying the design and content of CAHPS instruments.
• Obtain two (2) independent forward translations in the target language.

Conduct Review of Translations

• Produce a master table (see Appendix A for sample) that includes:
  • The original English version of the survey, including all instructions and all items,
  • Translation 1,
  • Translation 2,
  • A column for the bilingual reviewer to insert a third version of the translation of each survey item, and
  • A column for the bilingual reviewer to comment on any issues that should be addressed or considered in revising the translation.

We recommend using each row in the table for a single set of instructions or a single survey item with the corresponding response options.

• Provide the bilingual reviewer with the instructions and background information the translators received.
• Have one (1) bilingual reviewer review the master table, indicating which version of the translation should be used for each survey item or producing a reconciled version. The proposed reconciled translation should be entered into the master table.

Reconcile Translations by Committee Consensus

• Have the two translators and the reviewer meet as a committee (referred to as the “translation committee”) to review the findings of the bilingual review, to discuss any issues or problems identified by the reviewer, and to produce a reconciled version of the translations. The choice of final wording to be used should be arrived at by agreement of the committee. (Note: Meetings can be either in person or by telephone).
• Review the master table by having the bilingual reviewer take the lead in chairing the committee review meetings and keeping track of the agreed upon changes using the master table. The reconciled version from the original review should be modified as needed based on the committee’s decision and entered into the master table
• Document any concerns that the team cannot resolve and contact the CAHPS User Network to request additional guidance and to keep ������Ƶinformed of issues such as the following:
• Cultural or conceptual differences between the English source document and the translated version that cannot be resolved by modifying the translation (or at least not in a way that doesn’t substantially change the meaning of the original item);
• Questions about the specific intent of a survey item, word, or phrase; or
• Issues related to the translation of the response options.

Produce Final Version of Translated Survey

• Produce a final version of the translation once the translation committee has completed their review of the translations and has come to a consensus on final wording for the instructions and each item and corresponding responses.

Submit Translated Survey to AHRQ

• Submit a copy of the final translation to AHRQ’s CAHPS User Network (cahps1@ahrq.gov). Please indicate whether ������Ƶmay share your translation with other users.
• Submit a description of the translation process and a copy of the final version of the master table that includes the two forward translations, reviewer comments and suggested changes or edits, and the final reconciled version of the translation, if available. This documentation will enable the Consortium and other users to confirm the appropriateness of your translation.

If you have any questions or comments about these guidelines, please contact AHRQ’s CAHPS User Network via e-mail at CAHPS1@ahrq.gov or by phone at 1-800-492-9261.

APPENDIX A: SAMPLE MASTER TABLE

[1] These guidelines were prepared by the CAHPS II Cultural Comparability Team. The lead author was Beverly Weidmer, RAND; collaborating authors were Margarita Hurtado (AIR), Robert Weech-Maldonado and Quyen Ngo-Metzger (RAND), and Karen Bogen (University of Massachusetts/Harvard).