Translation, Cross-Cultural Adaptation and Measurement Properties of Three Implementation Measures FIM-AIM-IAM into Bangla

This study successfully translated, culturally adapted, and validated the Bangla versions of the Acceptability, Appropriateness, and Feasibility Implementation Measures (AIM, IAM, and FIM) among 549 pregnant women in Bangladesh, demonstrating that the adapted scales possess excellent internal consistency, construct validity, and convergent validity for use in maternal and child health research.

The Gist

Imagine you have a very special, high-quality measuring tape designed in the United States to check how people feel about new medical ideas. This tape has three specific rulers on it: one for Acceptability (Do people like it?), one for Appropriateness (Does it fit the situation?), and one for Feasibility (Can people actually do it?).

The problem is, this tape is written in English. If you try to use it on someone who only speaks Bangla, the numbers might not make sense, or the words might feel awkward, like trying to wear a shoe made for a different foot. You can't just translate the words; you have to make sure the feeling of the measurement stays the same.

This paper is the story of how the authors took that English "measuring tape" and remade it perfectly for Bangla-speaking pregnant women in Bangladesh.

Here is the breakdown of their journey, using simple analogies:

1. The Mission: Why Remake the Tape?

The authors noticed a gap. In healthcare, there is often a difference between what research says should work and what actually happens in real life. To fix this, doctors need to know if patients actually like a treatment, if it fits their lives, and if it's doable.

The original tools (AIM, IAM, and FIM) were great in America, but they hadn't been tested in Bangladesh. Using the English version directly would be like trying to read a map written in a language you don't know; you might get lost. The goal was to create a "Bangla version" that is just as accurate as the original.

2. The Process: How They Remade It

They didn't just hire one person to translate. They treated it like a team of architects building a bridge:

• The Translators: Six native speakers translated the English questions into Bangla independently.
• The Back-Translators: Two other people, who didn't know the original English questions, translated the new Bangla version back into English.
• The Comparison: They compared the new English version with the original. If the meaning drifted, they fixed it.
• The Experts: A panel of 10 experts (doctors, researchers, and specialists) reviewed the questions to make sure they made sense culturally.
• The Test Run: They asked 20 pregnant women to read the questions to see if they were easy to understand.

3. The Test: Putting the New Tape to Work

Once the new "Bangla Measuring Tape" was ready, they took it to a hospital in Rajbari, Bangladesh.

• The Participants: They asked 549 pregnant women to fill out the survey.
• The Check-Up: To make sure the tape didn't stretch or shrink, they asked 50 of those women to take the test again one week later to see if they gave the same answers.

4. The Results: Did It Work?

The results were very promising. Here is what the data told them:

• It Sticks Together (Reliability): When they looked at the answers, the questions on each ruler (Acceptability, Appropriateness, Feasibility) were consistent. If a woman liked the idea, she tended to answer similarly across all questions for that topic. The "stickiness" score (Cronbach's alpha) was good, meaning the questions worked well together as a team.
• It Measures One Big Thing (Validity): This is the most interesting part. In the original American study, the three rulers were thought to be three separate things. However, when the Bangla women answered, the math showed that Acceptability, Appropriateness, and Feasibility felt like one single concept to them.
  • The Analogy: Imagine you ask someone, "Is this cake tasty? Is it the right cake for the party? Can we eat it?" In this specific culture, the women seemed to answer "Yes" to all three as one big "Yes, this is a good idea." The math showed these three ideas were so tightly linked in their minds that they acted like a single ruler.
• It Fits the Data: The statistical models showed the new tape fit the Bangla population perfectly, with very little "wiggle room" for error.

5. The Conclusion

The authors concluded that they successfully built a valid and reliable Bangla version of these three implementation tools.

They found that for the pregnant women in their study, the concepts of "liking it," "fitting it," and "doing it" were so closely connected that they could be measured as a single, unified idea. This new tool is now ready to help researchers in Bangladesh understand how well new health programs are being received, ensuring that evidence-based care actually reaches the people who need it.

In short: They took a complex American measuring tool, carefully rebuilt it for a Bangla audience, tested it rigorously, and proved it works just as well as the original, revealing that for these women, the three key questions about a medical intervention are all part of the same big picture.

Technical Summary

Technical Summary: Translation, Cross-Cultural Adaptation, and Measurement Properties of FIM-AIM-IAM into Bangla

Problem Statement
Implementation science relies on valid measures to assess outcomes such as acceptability, appropriateness, and feasibility, which are critical for bridging the gap between evidence-based research and routine healthcare practice. While the Acceptability of Intervention Measure (AIM), Intervention Appropriateness Measure (IAM), and Feasibility of Intervention Measure (FIM) are widely used, they were originally developed and validated in high-income, English-speaking contexts. Direct application of these instruments in low- and middle-income countries (LMICs), specifically in Bangla-speaking populations, risks conceptual non-equivalence and measurement bias. Prior to this study, no validated Bangla versions of these tools existed, limiting the capacity to conduct rigorous implementation research in maternal and child health settings in Bangladesh.

Methodology
This study followed a methodological design adhering to the Consensus-based Standards for the Selection of Health Measurement Instruments (COSMIN) framework and STROBE guidelines. The research was conducted in three stages at Rajbari District Hospital, Bangladesh, between October and November 2025:

1. Translation and Cross-Cultural Adaptation: Six native translators independently translated the original English AIM, IAM, and FIM into Bangla. Two bilingual translators, blinded to the study objectives, performed back-translations. An expert panel (n=10) evaluated relevance, clarity, and construct consistency. Face validity was assessed via a convenience sample of 20 pregnant women.
2. Data Collection: A cross-sectional study was conducted with 549 pregnant women (aged 18–45) attending antenatal care. Data were collected using a pre-tested semi-structured questionnaire including socio-demographics and the adapted scales. A subsample of 50 participants completed the scales again after a 7-day interval for test-retest reliability.
3. Psychometric Evaluation: Data were analyzed using R software (v4.4.0). The study assessed interpretability (floor/ceiling effects), reliability (internal consistency via Cronbach's alpha, test-retest via Intraclass Correlation Coefficient, Standard Error of Measurement, and Smallest Detectable Change), and validity (construct validity via Exploratory and Confirmatory Factor Analysis, and convergent validity via Average Variance Extracted).

Key Results

• Sample Characteristics: The cohort (N=549) had a mean age of 23.39 years. The majority (98.29%) were not working, and most participants and their husbands had attained secondary or higher education.
• Item-Level Analysis: All items demonstrated high mean scores (4.14–4.39 on a 5-point scale) with low variability, indicating strong positive perceptions of interventions. Item-total correlations were strong (>0.60), and skewness/kurtosis values fell within acceptable ranges, suggesting normal distribution.
• Reliability:
  • Internal Consistency: The scales demonstrated excellent internal consistency (AIM: α = 0.78; IAM: α = 0.81; FIM: α = 0.77).
  • Test-Retest Reliability: Reliability over a 7-day interval was moderate, with ICCs of 0.41 (AIM), 0.52 (IAM), and 0.49 (FIM).
  • Measurement Error: Standard Error of Measurement (SEM) and Smallest Detectable Change (SDC) values were low, indicating the tools are precise enough to detect meaningful changes.
• Validity:
  • Construct Validity: Contrary to the original three-factor structure, both Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA) supported a robust unidimensional model for the combined 12-item scale. The first factor explained 48% of the variance.
  • Model Fit: The single-factor model demonstrated excellent fit indices (CFI = 0.92, RMSEA = 0.04, SRMR = 0.06, χ²/df = 2.01).
  • Convergent Validity: Average Variance Extracted (AVE) values were 0.51 (AIM), 0.52 (IAM), and 0.50 (FIM), meeting the threshold for adequate convergent validity.
  • Interpretability: No missing data were observed, and ceiling/floor effects were negligible (<1%), indicating high comprehensibility.

Significance and Claims
The authors claim that the Bangla versions of the AIM, IAM, and FIM are valid and reliable screening tools for implementation research within maternal and child health settings in Bangladesh. The study highlights that in this specific cultural context, the theoretical distinctions between acceptability, appropriateness, and feasibility may not be distinct, as participants perceived them as a single underlying construct.

The paper asserts that these adapted instruments facilitate high-quality implementation research in Bangla-speaking settings, thereby supporting the evidence-based adoption and sustained implementation of health interventions. By providing psychometrically sound tools, the study aims to address the equity gap in implementation science for LMICs, allowing for accurate assessment of early implementation outcomes without the bias introduced by direct translation of English instruments. The authors note that while the test-retest reliability was lower than in original English validations, this may reflect the dynamic nature of participant perceptions in treatment programs or specific population stability issues, rather than a failure of the tool itself.