How to adapt a measure for use in a different population?
Background
Nutrition and dietary behaviors play an important role in the development of chronic diseases for Asian American adults, including those who are caregivers of Asian American children. However, food preferences and patterns are often different for Asian American sub-groups and thus require tailored instruments to capture nutrition and dietary behaviors. Developing high quality tools that capture culturally-specific and culturally-preferred foods among ethnic minority groups is important to reduce health inequities in the United States.
Glossary
Adaptation refers to the process of making thoughtful and planned alterations to the design or delivery of an evidence-based intervention or tool, with the explicit goal of improving fit or effectiveness in a specific context. Cultural adaptation, more specifically, refers to adaptations that pay attention to the importance of cultural factors and accounts for cultural patterns, meanings, and values within that context.
This case study describes how one team approached the process of adapting the 26-item Dietary Screener Questionnaire (DSQ) from the 2009–2010 National Health and Nutrition Examination Survey (NHANES) to be both relevant and culturally appropriate for English-speaking Asian American families. Specifically, the research team adapted the DSQ to better reflect the diets of Chinese, South Asian, Filipino, Korean, Vietnamese, and Japanese Americans, which represent 85% of Asian Americans. The research team planned to adapt the DSQ after searching the peer-reviewed literature and “grey” literature on Google Scholar and identified that currently, no dietary screeners exist that capture the variety of food preferences and patterns specific to Asian American sub-groups.
Considerations and challenges
Why is it important to develop and culturally adapt measures to be relevant to Asian American adults and children?
Asian Americans are an underserved yet understudied population. Applications of culturally adapted tools can help ensure more accurate data collection by providing specific examples of foods that are consumed by first-, second- and/or third- generation Asian ethnic subgroups. The overarching goal is to advance population health equity by filling gaps in knowledge about different Asian households and families.
What are the parameters for how the tool will be used?
In this case study, the selection of a specific tool is guided by 1) length—the tool must be short and validated, 2) delivery options—can it be delivered online, and 3) adaptability—does it allow for conversion of screener responses to dietary factors of interest. After examining the evidence-based tools listed in the National Cancer Institute’s Register of Validated Short Dietary Assessment Instruments, the research team makes the decision to adapt the DSQ. They recognize that the original DSQ did not capture foods typically consumed by Asian American families, including expectations of what a typical “plate” looks like in Asian American households—a lesson the research team learned from prior work to culturally adapt plate planning tools for multiple Asian subgroups. For example, among the many food options listed in the prompts/probes under each item in the DSQ, the majority 1) are not commonly found in Asian ethnic grocery stores, bakeries, coffeeshops, and household pantries, and 2) did not include traditional and popular items that are commonly found in Asian ethnic grocery stores, bakeries, coffeeshops, or household pantries.
Who is the target population, and what degree of adaptation is needed?
Practitioners and researchers should be specific in understanding who their target population is by age, ethnic subgroup, language preference, and acculturation level (i.e., adaptation to U.S. norms and values). These factors help determine the degree of cultural adaptation that is necessary and would be acceptable to the target population. For example, is a larger-scale “macro” adaptation required or would smaller-scale “micro” adaptation be more appropriate. For this case study, it is important to note that because the planned survey would be administered online and in English, the research team knows that they would likely be reaching Asian Americans with higher levels of acculturation. This bears mentioning, as the research team considers this level of acculturation when deciding how deeply to adapt some measures (e.g., consumption of fruits and vegetables did not include multiple ethnic options; retaining other common U.S. foods such as Mexican salsa).
Measure Selection
The research team’s adaptation process was driven by long-standing community partnerships, experience conducting community-based participatory research, previous survey work focused on diverse populations, and experience working with validated health status questions from national surveys. The community partnerships are sustained by the research team’s commitment to engage Asian American communities, mobilize large numbers of Asian Americans to participate in health research, and support the equitable translation of research findings into policy and practice recommendations. The research team’s key partnerships include those with a national partner, the Asian & Pacific Islander American Health Forum (APIAHF), a National Advisory Committee on Research and Development (NAC), and a Community Partner Network—comprised of more than 75 Asian American-serving community organizations, including public health departments, education, social service, and health care.
The first step in adapting the DSQ to be culturally relevant and culturally appropriate for English-speaking, Asian Americans was to identify the most popular Asian American foods in each food group listed in the DSQ for each Asian American subgroup. This involved 1) searching peer-reviewed literature for similar screeners that were adapted for specific Asian and/or Asian American subgroups; 2) searching for items included in food composition tables, including databases located here: http://www.fao.org/infoods/infoods/tables-and-databases/asia/en/; and 3) searching the Internet for relevant food blogs, social media accounts, etc.
The second step involved an iterative review process over multiple rounds to integrate critical feedback on the DSQ questions from a diverse group of fifteen Asian American-serving community leaders, partners, and staff, including several experienced bicultural community health workers. Key components of their feedback include whether the examples seemed appropriate and whether the research team missed any major examples or food groups/items.
During this phase, the research team had to decide which questions either should or should not be adapted, and whether to add questions. Not all items required adaptation, given the acculturation level anticipated for the sample. For example, no modifications were made to fruit or vegetable examples, or for chocolate and ice cream. Additionally, because seafood is such a prominent food group among Asian American caregivers and their children as indicated by the published literature, the team decided to add a question specifically addressing seafood intake. This added item mimics the pre-existing red and processed meat intake frequency questions, which also measures items per day. Decisions like these were then reviewed by community partners and leaders as member checks for relevance and accuracy.
To allow for broader applications of the adapted dietary screener, the team considers how to translate the instrument into multiple languages. Doing so would greatly enhance relevance and reach beyond just Asian Americans, who may be more acculturated, and would be an important next step towards advancing population health equity.
Lessons Learned
- Throughout the cultural adaptation process, learning should be bi-directional between the research team and key community partners. This requires working closely together to co-learn and co-produce knowledge and solutions. The research team gleaned valuable cultural insights from the community for each subgroup, and community partners learned about some of the practical realities of data collection (e.g., limiting examples for question brevity).
- Relatedly, balance should be maintained between providing a comprehensive and representative list of food examples, without overwhelming respondents.
- Cultural adaptation can complement and bolster existing literature and online databases of evidence-based tools.
- Identify and specify when to prioritize food examples for one sub-group but not another.
- Determine the “depth” of adaptation required based on the characteristics of the setting and target population. In this particular case study, changes were more conservative because the research team targeted those who speak English and may be more acculturated, and given the online delivery of the adapted tool, the team had to decide how to best display examples.
Related Resources
- Kwon SC, Patel S, Choy C, Zanowiak J, Rideout C, Yi SS, et al. Implementing health promotion activities using community-engaged approaches in Asian American faith-based organizations in New York City and New Jersey. Transl Behav Med. 2017;7(3):444-66. doi:10.1007/s13142-017-0506-0.
- Gore R, Patel S, Choy C, Taher M, Garcia-Dia MJ, Singh H, et al. Influence of organizational and social contexts on the implementation of culturally adapted hypertension control programs in Asian American-serving grocery stores, restaurants, and faith-based community sites: a qualitative study. Transl Behav Med. 2019. doi:10.1093/tbm/ibz106.
Acknowledgments
The authors would like to thank Stella Chong, Lily Divino, Mary Joy Garcia, Alka Kanaya, Simona Kwon, Stephanie Liu, Binh Lu, Deborah Min, Rhea Naik, Chorong Park, MD Taher, Sameer Talegawkar, Kosuke Tamura, Tracy Vo, and Jennifer Wong for their feedback on our survey questions related to acculturation, diet, and grocery shopping.