A critical focus of pediatric hospital systems is to improve the quality of hospital-to-home transitions. Although validated patient-reported measures evaluating these improvement efforts exist for English-speaking families, a comprehensive measure to assess transition quality among families who speak a language other than English does not yet exist.
We used a team consensus translation approach to translate and culturally adapt the previously validated Pediatric Transition Experience Measure (P-TEM), a caregiver-reported hospital-to-home transition quality measure, from English to Spanish. We describe our rigorous translation approach, which involved a series of steps to preserve the original meaning of the P-TEM through careful team-based linguistic and cultural adaptation of the measure into Spanish. During this process, we also found additional opportunities to improve the understandability and content validity of the original English version of P-TEM. We then pilot tested the new Spanish P-TEM with 36 parents and administered the revised English P-TEM with 125 caregivers (ie, parents/legal guardians).
In pilot testing, none of the Spanish-speaking parents reported difficulty understanding questions, though 6% (2/36) expressed difficulty with understanding the response scale, prompting a change to present clearer scale anchors. Mean scores on the Spanish P-TEM were 95.4 (SD, 9.6) for the total score. Mean scores on the revised English P-TEM were 88.6 (SD, 15.6; total).
Using a team consensus translation approach is a comprehensive and collaborative approach that allows for translation of measures originally developed for English-speaking families to be translated in a way that is reliable, accurate, and culturally appropriate.
Language barriers and inadequate interpreter use create barriers for families to engage in bidirectional communication, leading to increased risk for serious adverse events postdischarge.1,2 Families with a language other than English, a structurally marginalized group, are at risk for worse outcomes during hospital-to-home transition with higher rates of confusion about medication instructions, difficulty accessing timely care, and higher rates of emergency department revisits.3–5 Given that 1 in 10 children in the United States have a parent who use a language other than English, it is important to address these risks.3,4
There is national recognition of the essential role families play in hospital-to-home transitions, and family engagement in discharge planning is necessary.6,7 We need to develop and evaluate equitable and patient-centered care transition interventions to support all families. Using patient-centered care transition quality measures helps ensure that health systems are addressing the needs of patients and families.8 However, to our knowledge, a Spanish version of a validated questionnaire that measures the quality of hospital-to-home transitions does not yet exist.
Previously, our group described the development of the Pediatric Transition Experience Measure (P-TEM), which is the first validated caregiver-reported outcome measure that fully captures the entire transition process from discharge through follow-up.8,9 The development of P-TEM was grounded in qualitative interviews with English-speaking families to understand what constitutes successful, high-quality transitions from hospital to home from the perspective of patients and families.10 Psychometric testing of P-TEM was also only conducted with English-speaking families. Although these limitations may influence the P-TEM’s relevance to other groups, we sought to undertake a high-quality process of translation and cultural adaptation and to maximize its relevance and understandability within those preexisting constraints. Therefore, our group elected to use a team consensus translation process to develop a Spanish version of P-TEM that could be used for future multilingual improvement efforts related to hospital-to-home transitions.11
Methods
We assembled a group of clinicians (including the measure developer), bilingual Spanish and English translators, and research coordinators with an understanding of the culture and linguistics used among our Spanish-speaking populations to develop a Spanish version of the P-TEM. This approach included a series of steps that used a team consensus translation approach incorporating principles previously described by the Census Bureau Guideline for the Translation of Data Collection Instruments and Supporting Materials.11 This process occurred between April 2021 and June 2021 at a tertiary children’s hospital in Seattle, WA.
Although other translation methods exist (such as forward to back translation only), our group chose the team consensus translation approach because it creates a higher quality translation, incorporating careful discussion of how to best translate each idea, cultural adaptation, and pretesting (Fig 1). Consistent with this approach, study investigators (including the measure developer) met to review the P-TEM measure and its meaning in English with 2 professional English-Spanish translators, who then conducted independent forward translations of the measure. These translators then met with the full translation team, including 2 bilingual, bicultural research coordinators, the measure developer, and an investigator with translation-related expertise, to discuss each item, resolve differences in the 2 translations, propose additional revisions, and develop consensus. A strength of this method is that it allows for direct opportunities throughout the process for translators to ask the measure developer questions to best preserve the construct validity of the original measure. During this rigorous translation process, we also assessed if items required cultural adaptation, but ultimately did not find any items that required cultural adaptation. Furthermore, this group consensus translation process identified areas to refine the original English P-TEM measure to optimize clarity and understandability.
Schematic overview of team consensus translation approach. a Spanish translators originated from various parts of Latin America and had experience translating for Spanish speakers from different Spanish-speaking countries. b Team members met 8 times (1 hour each) over a 2-month period.
Schematic overview of team consensus translation approach. a Spanish translators originated from various parts of Latin America and had experience translating for Spanish speakers from different Spanish-speaking countries. b Team members met 8 times (1 hour each) over a 2-month period.
Next, we pilot-tested the Spanish survey items with a sample of families through our institution’s patient-reported outcomes (PRO) program to verify face validity and caregiver (ie, parents and/or legal guardians) comprehension of items. Pilot testing occurred between March 10, 2021 and March 16, 2022. The PRO program routinely collects quality-of-life outcomes data from a sample of families during (admission survey) and after a hospital admission (follow-up survey).12 P-TEM items were added to the follow-up survey administered 5 to 7 weeks after the child’s discharge from the hospital. Formal audio-recorded cognitive interviews were not conducted because of resource limitations. Instead, program coordinators pilot tested revised questions by administering the follow-up survey by phone to all Spanish-speaking families. Coordinators asked questions about their understanding of the survey items and wording, as well as why they chose specific response options. They recorded responses to these questions verbatim and noted when parents hesitated when answering a P-TEM survey item. The revised English P-TEM measure was also administered to a sample of English-speaking families by phone or online through the PRO program during the same time frame; however, no additional questions to assess understandability or response processes were asked of English-speaking families.
Results
Team Consensus Translation of P-TEM in Spanish
Over the course of 8 meetings over 2 months, we developed a team consensus translation of the P-TEM in Spanish, along with a revised English version intended to match the Spanish version (Tables 1 and 2). Some key revisions that we made to the P-TEM survey items during the team consensus translation process included rephrasing screening items, so they are written as statements rather than questions to improve flow and consistency. We also discovered that a question pertaining to “caring for the child at home” required additional clarity because this could be misinterpreted as “general needs of the child” in Spanish. Another question about having “everything I needed” required additional specification to improve understandability; therefore, the item was revised to include examples of which types of needs respondents should consider when answering this question (eg, information, medications, medical equipment/devices, support from providers, support from loved ones).
Pilot Testing of Pediatric Transition Experience Measure (P-TEM), Spanish Version
| No. Respondents | Mean Score (SD) | Minimum Score | Median Score | Maximum Score | Responses in Top Box, % | ||
|---|---|---|---|---|---|---|---|
| P-TEM total scorea | 36 | 95.4 (9.6) | 50 | 100 | 100 | 86 | |
| Transition preparation domain | 36 | 94.9 (13.3) | 37.5 | 100 | 100 | 85 | |
| 1. | Entendí claramente todas las instrucciones del alta que recibí antes de irnos del hospital. | 36 | 96.1 (11) | 50 | 100 | 100 | 83 |
| 2. | Me sentí preparado/a para atender las necesidades médicas de mi hijo/a en casa. | 36 | 94.2 (19) | 20 | 100 | 100 | 86 |
| 3. | Las instrucciones escritas del alta que recibí fueron útiles para atender las necesidades médicas de mi hijo/a en casa. | 36 | 94.2 (16.6) | 30 | 100 | 100 | 86 |
| 4. | Sentí que al irnos del hospital tenía todo lo que necesitaba (p. ej., información, medicamentos, aparatos o dispositivos médicos, apoyo emocional y comunitario, etc.) para atender las necesidades médicas de mi hijo/a en casa. | 36 | 95 (13) | 50 | 100 | 100 | 83 |
| Transition support domain | 32 | 97.2 (5.8) | 75 | 100 | 100 | 90 | |
| Screener 1 | Tuve que contactar a los médicos del hospital de mi hijo/a (p. ej., especialistas) para hacerles preguntas después del alta. | ||||||
| 5. | Me fue fácil contactar a los médicos del hospital de mi hijo/a (p. ej., especialistas) para hacerles preguntas después del alta. | 9 | 100 (0) | 100 | 100 | 100 | 100 |
| Screener 2 | Se tuvo que programar una consulta de seguimiento para mi hijo/a después del alta | ||||||
| 6. | Se tomaron en cuenta mis preferencias (fechas y horarios, lugar, médicos) al programar las consultas de seguimiento para mi hijo/a. | 26 | 94.6 (12.1) | 50 | 100 | 100 | 76 |
| Screener 3 | Mi hijo/a ya ha tenido una consulta de seguimiento con algún proveedor médico. | ||||||
| 7. | Todos los proveedores médicos en las consultas de seguimiento estaban informados | 28 | 95.3 (13.4) | 50 | 100 | 100 | 85 |
| 8. | Todos los proveedores médicos en las consultas de seguimiento supieron cómo brindarle atención médica a mi hijo/a. | 28 | 100 (0) | 100 | 100 | 100 | 100 |
| No. Respondents | Mean Score (SD) | Minimum Score | Median Score | Maximum Score | Responses in Top Box, % | ||
|---|---|---|---|---|---|---|---|
| P-TEM total scorea | 36 | 95.4 (9.6) | 50 | 100 | 100 | 86 | |
| Transition preparation domain | 36 | 94.9 (13.3) | 37.5 | 100 | 100 | 85 | |
| 1. | Entendí claramente todas las instrucciones del alta que recibí antes de irnos del hospital. | 36 | 96.1 (11) | 50 | 100 | 100 | 83 |
| 2. | Me sentí preparado/a para atender las necesidades médicas de mi hijo/a en casa. | 36 | 94.2 (19) | 20 | 100 | 100 | 86 |
| 3. | Las instrucciones escritas del alta que recibí fueron útiles para atender las necesidades médicas de mi hijo/a en casa. | 36 | 94.2 (16.6) | 30 | 100 | 100 | 86 |
| 4. | Sentí que al irnos del hospital tenía todo lo que necesitaba (p. ej., información, medicamentos, aparatos o dispositivos médicos, apoyo emocional y comunitario, etc.) para atender las necesidades médicas de mi hijo/a en casa. | 36 | 95 (13) | 50 | 100 | 100 | 83 |
| Transition support domain | 32 | 97.2 (5.8) | 75 | 100 | 100 | 90 | |
| Screener 1 | Tuve que contactar a los médicos del hospital de mi hijo/a (p. ej., especialistas) para hacerles preguntas después del alta. | ||||||
| 5. | Me fue fácil contactar a los médicos del hospital de mi hijo/a (p. ej., especialistas) para hacerles preguntas después del alta. | 9 | 100 (0) | 100 | 100 | 100 | 100 |
| Screener 2 | Se tuvo que programar una consulta de seguimiento para mi hijo/a después del alta | ||||||
| 6. | Se tomaron en cuenta mis preferencias (fechas y horarios, lugar, médicos) al programar las consultas de seguimiento para mi hijo/a. | 26 | 94.6 (12.1) | 50 | 100 | 100 | 76 |
| Screener 3 | Mi hijo/a ya ha tenido una consulta de seguimiento con algún proveedor médico. | ||||||
| 7. | Todos los proveedores médicos en las consultas de seguimiento estaban informados | 28 | 95.3 (13.4) | 50 | 100 | 100 | 85 |
| 8. | Todos los proveedores médicos en las consultas de seguimiento supieron cómo brindarle atención médica a mi hijo/a. | 28 | 100 (0) | 100 | 100 | 100 | 100 |
Scores are on a 0-100 scale; higher scores reflect better quality.
Pilot Testing of Pediatric Transition Experience Measure (P-TEM), English Version
| No. Respondents | Mean Score (SD) | Minimum Score | Median Score | Maximum Score | Responses in Top Box, % | ||
|---|---|---|---|---|---|---|---|
| P-TEM total scorea | 125 | 88.6 (15.6) | 15 | 94.3 | 100 | 68 | |
| Transition preparation domain | 125 | 89.3 (16) | 20 | 97.5 | 100 | 68 | |
| 9. | I had a clear understanding of all the discharge instructions I received before we left the hospital. | 125 | 91.4 (16.7) | 10 | 100 | 100 | 69 |
| 10. | I felt prepared for taking care of my child’s medical needs at home. | 125 | 86.8 (21.7) | 0 | 100 | 100 | 59 |
| 11. | The written discharge instructions I was given were helpful for taking care of my child at home. | 122 | 90.1 (18.9) | 0 | 100 | 100 | 70 |
| 12. | I felt I had everything I needed (eg, information, medications, medical equipment/devices, support from providers, support from loved ones, etc.) to take care of my child’s medical needs at home. | 125 | 89.3 (22.2) | 0 | 100 | 100 | 72 |
| Transition support domain | 103 | 85.9 (22) | 0 | 96.7 | 1100 | 66 | |
| Screener 1 | I needed to contact my child’s hospital doctors (eg, specialists) with questions after we left the hospital. | ||||||
| 13. | I was able to easily contact my child’s hospital doctors (eg, specialists) with questions after we left the hospital. | 51 | 83.3 (27) | 0 | 100 | 100 | 59 |
| Screener 2 | My child had one or more follow-up appointments that needed to be scheduled after his/her hospitalization. | ||||||
| 14. | My preferences (dates and times, location, doctors) were considered when scheduling follow-up appointments for my child. | 84 | 82.4 (27.3) | 0 | 100 | 100 | 59 |
| Screener 3 | My child already had one or more follow-up appointments with a medical provider. | ||||||
| 15. | All the medical providers in the follow-up visits knew what medical care my child received in the hospital. | 88 | 90 (22.2) | 0 | 100 | 100 | 75 |
| 16. | All the medical providers in the follow-up visits knew how to manage my child’s medical care. | 88 | 89.3 (20) | 0 | 100 | 100 | 63 |
| No. Respondents | Mean Score (SD) | Minimum Score | Median Score | Maximum Score | Responses in Top Box, % | ||
|---|---|---|---|---|---|---|---|
| P-TEM total scorea | 125 | 88.6 (15.6) | 15 | 94.3 | 100 | 68 | |
| Transition preparation domain | 125 | 89.3 (16) | 20 | 97.5 | 100 | 68 | |
| 9. | I had a clear understanding of all the discharge instructions I received before we left the hospital. | 125 | 91.4 (16.7) | 10 | 100 | 100 | 69 |
| 10. | I felt prepared for taking care of my child’s medical needs at home. | 125 | 86.8 (21.7) | 0 | 100 | 100 | 59 |
| 11. | The written discharge instructions I was given were helpful for taking care of my child at home. | 122 | 90.1 (18.9) | 0 | 100 | 100 | 70 |
| 12. | I felt I had everything I needed (eg, information, medications, medical equipment/devices, support from providers, support from loved ones, etc.) to take care of my child’s medical needs at home. | 125 | 89.3 (22.2) | 0 | 100 | 100 | 72 |
| Transition support domain | 103 | 85.9 (22) | 0 | 96.7 | 1100 | 66 | |
| Screener 1 | I needed to contact my child’s hospital doctors (eg, specialists) with questions after we left the hospital. | ||||||
| 13. | I was able to easily contact my child’s hospital doctors (eg, specialists) with questions after we left the hospital. | 51 | 83.3 (27) | 0 | 100 | 100 | 59 |
| Screener 2 | My child had one or more follow-up appointments that needed to be scheduled after his/her hospitalization. | ||||||
| 14. | My preferences (dates and times, location, doctors) were considered when scheduling follow-up appointments for my child. | 84 | 82.4 (27.3) | 0 | 100 | 100 | 59 |
| Screener 3 | My child already had one or more follow-up appointments with a medical provider. | ||||||
| 15. | All the medical providers in the follow-up visits knew what medical care my child received in the hospital. | 88 | 90 (22.2) | 0 | 100 | 100 | 75 |
| 16. | All the medical providers in the follow-up visits knew how to manage my child’s medical care. | 88 | 89.3 (20) | 0 | 100 | 100 | 63 |
Scores are on a 0-100 scale; higher scores reflect better quality.
Refinement of English P-TEM
From this team consensus translation approach, we identified areas of the original English P-TEM measure that needed refinement and optimization. By questioning the meaning behind each word and idea during group discussions of the Spanish version, the team consensus translation process highlighted important terminology changes within the English version to enhance clarity and understandability. Examples of key text revisions within the English version are provided in the supplemental materials (Supplemental Table 3).
Pilot-Testing Both Versions
A total of 36 caregivers participated in the pilot test of the new Spanish P-TEM and 125 parents completed the revised English version of P-TEM (Supplemental Table 4). None of the Spanish-speaking caregivers reported difficulty understanding questions, though 6% (2/36) expressed difficulty with understanding the numeric response scale, prompting a change to present clearer scale anchors. Overall, caregivers who specifically commented on positive and negative experiences with care, similarly reported higher or lower scores on corresponding survey items. Mean scores on the Spanish version of P-TEM were 95.4 (SD, 9.6) for the total score, 94.9 (SD, 13.3) for the transition preparation domain, and 97.2 (SD, 5.8) for the transition support domain (Table 1). Mean scores on the revised English version of P-TEM were 88.6 (SD, 15.6; total), 89.3 (SD, 16.0; transition preparation domain), and 85.9 (SD, 22.0; transition support domain; Table 2).
Discussion
Using a team consensus translation approach, we developed a Spanish version of the P-TEM that is readily understandable and successfully preserves the meaning of the original measure. We believe that our work demonstrates that a team consensus translation approach with bilingual translators who have cultural insight into the population of interest allows for high-quality development of translated measures when in-depth measure development within a target population is not feasible.13 Given that this process illuminated areas to optimize the comprehension of the English version, our findings also suggest that conducting the initial survey development work of P-TEM with families who speak languages other than English would have helped to improve understandability and content validity of the measure for all languages.
Survey scores from Spanish- and English-speaking respondents demonstrate higher P-TEM scores among Spanish-speaking respondents. This is consistent with known response bias based on language for similar types of surveys measuring care experiences or surveys assessing health status, or when surveys are administered by phone.14–18 Some study limitations to note are that our sample size cannot detect a meaningful difference in these scores; therefore, future studies should consider larger sample sizes using similar survey administration modalities. Additionally, respondent scores are based on the original scale and future testing of the revised scale may influence differences in scores.
Given that many existing validated measures are not available in languages other than English, this study offers a high-quality process for translating measures originally developed among English-speaking families. In circumstances in which it is infeasible to conduct the formative survey development work in a language other than English, this process offers a more rigorous approach to linguistically and culturally adapt existing measures compared with simple forward and back translation.19 Future directions include translating P-TEM into additional languages using this approach. By increasing the number of measures accessible to researchers in languages other than English, we can continue to move the needle toward increased diversity and inclusivity within clinical research and improvement work.
Dr Squires participated in data analysis, drafted the initial manuscript, and critically reviewed and revised the manuscript; Dr Lion participated in design of the study, participated in data analysis, and critically reviewed and revised the manuscript; Ms Burgara collected data, participated in data analysis, and critically reviewed the manuscript and revised the manuscript; Ms Garcia collected data, participated in data analysis, and critically reviewed the manuscript and revised the manuscript; and Dr Desai conceptualized and designed the study, provided oversight of all study procedures including the analysis, and critically reviewed and revised the manuscript.
FUNDING: This study was funded in part by the Seattle Children’s Hospital, Center for Quality & Safety, Patient-Reported Outcomes Program.
CONFLICT OF INTEREST DISCLOSURES: The authors have indicated they have no potential conflicts of interest to disclose.
COMPANION PAPER: A companion to this article can be found online at www.hosppeds.org/cgi/doi/10.1542/hpeds.2023-007198.
Comments