Results
The result of the translation and cross-cultural adaptation was a Swedish FACIT-CD that is conceptually and semantically equivalent to the English version and is linguistically valid. See Tables
1,
2,
3,
4 and
5 for the results of the translation and cross-cultural adaptation of each item. The translation and cross-cultural adaptation and the pilot test are presented separately.
Table 1
Results of translation and cross-cultural adaptation, and Cronbach’s alpha coefficient, and corrected item-total correlation for the Swedish FACIT-CD subscale Physical well-being (n = 34)
| Physical well-being/Fysiskt välbefinnande | 0.71 | | |
CD1 | I have discomfort in my pelvic area | | | |
Jag känner obehag i mitt underliv | | 0.78 | 0.59 |
CD2 | I have pain in my pelvic area | | | |
Jag känner smärta i mitt underliv | | 0.55 | 0.65 |
CD3 | I have cramping in my pelvic area | | | |
Jag har kramper i mitt underliv | | 0.46 | 0.68 |
Cx1 | I am bothered by discharge or bleeding from my vagina | | | |
Jag besväras av flytningar eller blödningar från slidan | | 0.34 | 0.71 |
GP5 | I am bothered by side effects of treatment | | | |
Jag besväras av biverkningar av behandlingen | | 0.11a
| 0.73 |
ES8 | I have pain or discomfort with intercourse | | | |
Jag känner smärta eller obehag vid samlag | | 0.55 | 0.65 |
CD4 | I have to limit my sexual activity because of the infection | | | |
Jag måste begränsa min sexuella aktivitet på grund av infektionen | | 0.42 | 0.68 |
CD5 | I worry about spreading the infection | | | |
Jag oroar mig för att sprida infektionen | | 0.08a
| 0.75 |
Table 2
Results of translation and cross-cultural adaptation, and Cronbach’s alpha coefficient, and corrected item-total correlation for the Swedish FACIT-CD subscale Treatment satisfaction (n = 34)
| Treatment satisfaction/Tillfredställelse med behandling | 0.81 | | |
GR1 | I have confidence in my doctor(s) | | | |
Jag har förtroende för min(a) läkare | | 0.55 | 0.81 |
CD6 | I feel that I received the treatment that was right for me | | | |
Jag känner att jag har fått den behandling som var rätt för mig | | 0.69 | 0.74 |
CD7 | My doctor gave me explanations that I could understand | | | |
Min läkare gav mig förklaringar (information) som jag kunde förstå | | 0.79 | 0.68 |
CD8 | My doctor explained the possible benefits of my treatment | | | |
Min läkare förklarade de eventuella fördelarna med min behandling | | 0.62 | 0.81 |
Table 3
Results of translation and cross-cultural adaptation, and Cronbach’s alpha coefficient, and corrected item-total correlation for the Swedish FACIT-CD subscale General perceptions (n = 34)
| General perceptions/Allmänna uppfattningar | 0.74 | | |
GF1 | I am able to work (include work at home) | | | |
Jag kan arbeta (innefattar även arbete i hemmet) | | 0.13a
| 0.76 |
GF3 | I am able to enjoy life | | | |
Jag kan njuta av livet | | 0.47 | 0.72 |
HI11 | I am hopeful about the future | | | |
Jag är hoppfull inför framtiden | | 0.62 | 0.70 |
Sp9 | I find comfort in my faith or spiritual beliefs | | | |
Jag finner tröst i min tro eller inre övertygelse | | 0.41 | 0.75 |
GF7 | I am content with the quality of my life right now | | | |
Jag är nöjd med min livskvalitet just nu | | 0.57 | 0.69 |
CD9 | I feel that I can manage things that come up around this infection | | | |
Jag känner att jag kan hantera saker som dyker upp runt den här infektionen | | 0.67 | 0.65 |
CD10 | I have accepted that I have this infection | | | |
Jag har accepterat att jag har den här infektionen | | 0.57 | 0.68 |
Table 4
Results of translation and cross-cultural adaptation, and Cronbach’s alpha coefficient, and corrected item-total correlation for the Swedish FACIT-CD subscale Emotional well-being (n = 34)
| Emotional well-being/Känslomässigt välbefinnande | 0.79 | | |
CD11 | I worry that the infection will get worse | | | |
Jag oroar mig för att infektionen kommer bli värre | | 0.39 | 0.78 |
CD12 | I have hidden this problem so others will not notice | | | |
Jag har dolt detta problem så att andra inte kommer att märka det | | 0.48 | 0.77 |
CD13 | I have concerns about my ability to become pregnant | | | |
Jag är bekymrad över min förmåga att bli gravid | | 0.40 | 0.78 |
BMT18 | The cost of my treatment is a burden on me or my family | | | |
Kostnaden för min behandling är en börda för mig eller min familj | | 0.42 | 0.77 |
CD14 | I worry about other people’s attitudes towards me | | | |
Jag oroar mig för andra människors attityder till mig | | 0.42 | 0.77 |
CD15 | I feel embarrassed about the infection | | | |
Jag känner mig generad över infektionen | | 0.65 | 0.75 |
CD16 | I tend to blame myself for the infection | | | |
Jag har en tendens att klandra mig själv för infektionen | | 0.68 | 0.74 |
CD17 | I was careful who I told about the infection | | | |
Jag var försiktig med för vem jag berättade om infektionen | | 0.34 | 0.78 |
CD18 | I have had difficulty telling my partner/spouse about the infection | | | |
Jag har haft svårt att berätta för min partner/make om infektionen | | 0.36 | 0.78 |
CD19 | I am frustrated by the infection | | | |
Jag är frustrerad över infektionen | | 0.49 | 0.77 |
CD20 | I am depressed about the infection | | | |
Jag är nedstämd av infektionen | | 0.39 | 0.78 |
Table 5
Results of translation and cross-cultural adaptation, and Cronbach’s alpha coefficient, and corrected item-total correlation for the Swedish FACIT-CD subscale Relationships (n = 34)
| Relationships/Relationer | 0.67a
| | |
Q9 | I have told my partner/spouse about my infection | |
b
|
b
|
Jag har berättat för min partner/make om min infektion | |
b
|
b
|
CD21 | I get emotional support from my partner/spouse | | | |
Om ja: Jag får känslomässigt stöd från min partner/make | | 0.66 | 0.43 |
Q10 | I have told family members about my infection | |
b
|
b
|
Jag har berättat för familjemedlemmar om min infektion | |
b
|
b
|
CD22 | I get emotional support from family members | | | |
Om ja: Jag får känslomässigt stöd från familjemedlemmar | | 0.75 | 0.35 |
GS1 | I feel close to my friends | | | |
Jag känner närhet till mina vänner | | 0.01c
| 0.82 |
HI3 | I have people to help me if I need it | | | |
Det finns personer som hjälper mig om jag behöver det | | 0.67 | 0.58 |
The result section ends with evaluation of the reliability, which indicated that the Swedish FACIT-CD exhibited good internal consistency reliability of the total scale, and that the majority of the items in the respective subscales exhibited acceptable corrected item-total correlations.
Translation and cross-cultural adaptation
The translation and cross-cultural adaptation went smoothly without any problems, except for the three items that consisted of the term “pelvic area” (CD1, CD2 and CD3). Initially, this term was translated by two of the forward translators and the reconciler into Swedish as “bäckenområde”. However, these three individuals had no previous professional contact with women who had been diagnosed with cervical dysplasia. The remaining forward translator was a midwife and familiar with this type of diagnoses, and she translated the term “pelvic area” into Swedish as “underliv”.
During the translation and cross-cultural adaptation process, the Swedish terms “bäckenområde” and “underliv” were discussed extensively between the FACIT project manager and the reviewer/language coordinator. The latter strongly felt that the term “underliv” was the preferable term in the context of cervical dysplasia. The FACIT project manager preferred the Swedish term “bäckenområde”, since it was a more literal translation of the “pelvic area”, closer to the source. As a result, the test version was finalised with the term “bäckenområde”, but the term “underliv” was used as an alternative translation, discussed in the cognitive debriefing interviews. The interviews determined which Swedish term should be used.
Additionally, one item (Sp9) changed the source word “spiritual beliefs” in Swedish to use “inre övertygelse”, which could be back translated into English as “inner beliefs”. This change was based on additional review in a separate project from this study, and during the same time period.
Finally, the result of the translation and cross-cultural adaptation process suggests that the Swedish FACIT-CD is conceptually and semantically equivalent to the English version and linguistically valid.
Pilot test
Women’s socio-demographic characteristics
In total, ten women were interviewed. The mean age was 33 years (range 25–46 years). All of the women were native Swedish speakers and none were receiving any treatment at the time of the interviews. The average time between when the women were diagnosed and when the interviews took place was 8 months (range 1–18 months).
Cognitive debriefing interviews
The women displayed good understanding of the items, and the responses they selected corresponded with the reasons they provided for choosing those answers. According to the women, all of the items were relevant to their diagnosis, the instructions were easy to understand and the response categories were unambiguous and comprehensive. The women also reported that the Swedish FACIT-CD was easy to complete in general.
Results from the cognitive debriefing interviews led to some changes in eight of the items in the Swedish FACIT-CD. The items CD1, CD2 and CD3 were changed to use the term “underliv”. One of the items had the Swedish word “information” added in parentheses (CD7). Four of the items had structural change, where the phrase “if yes” was moved from the question introduction (Q9 and Q10) and placed instead before the respective follow-up questions (CD21 and CD22).
Evaluation of the reliability
Women’s socio-demographic characteristics
In total, 34 women completed the Swedish FACIT-CD without skipping data. The mean age was 36 years (range 23–64 years). Of these women, 14 had completed senior high school and 16 college or higher education. Twenty-four women had a partner, and 20 had one or more children. One of the women was pregnant at the time that she completed the Swedish FACIT-CD. Three of the women were born abroad, either in Kosovo, Spain or Thailand, and all of the other women were Swedish born. All the women completed the questionnaire within a year after the diagnosis.
Internal consistency reliability and homogeneity
The total scale of the Swedish FACIT-CD exhibited good internal consistency reliability with a Cronbach’s alpha coefficient of 0.84. Cronbach’s alpha coefficient was acceptable on four out of five of the subscales: Physical well-being α = 0.71, Treatment satisfaction α = 0.81, General perceptions α = 0.74 and Emotional well-being α = 0.79. Nevertheless, the subscale Relationships had a Cronbach’s alpha coefficient of 0.67, which is below the acceptable value of ≥ 0.70. The corrected item-total correlations for each subscale are shown in Tables
1,
2,
3,
4 and
5. The majority of the items exhibited acceptable corrected item-total correlations, with the following exceptions in the respective subscales: Physical well-being subscale 0.11 (if item deleted α = 0.73) and 0.08 (if item deleted α = 0.75), Relationships subscale 0.01 (if item deleted α = 0.82) and General perceptions subscale 0.13 (if item deleted α = 0.76).
Discussion
The purpose of this study was to translate and cross-culturally adapt the Functional Assessment of Chronic Illness Therapy – Cervical Dysplasia (FACIT-CD) into a Swedish context, and evaluate the linguistic validity and reliability. To the best of our knowledge, the FACIT-CD is the first Swedish disease-specific instrument that assesses HRQoL associated with cervical dysplasia. The results indicated that the Swedish FACIT-CD is conceptually and semantically equivalent to the English version and linguistically valid; further, it exhibits good internal consistency reliability.
In this study, the translation and cross-cultural adaptation went smoothly without any problems for the majority of the items. One reason may be because the FACIT-CD is developed in the US [
15], and American and Swedish cultures seem to be relatively similar; additionally, they are both Germanic languages. This facilitates the possibility to reach semantic and conceptual equivalence between the original and the target version of the instrument. It is essential to reach semantic and conceptual equivalence, as it is the basis for determining whether an instrument can be adapted to the target culture [
25]. Although it is impossible to attain 100% equivalence, it is essential to strive to minimise bias and attain as strong equivalence as possible [
21]. In order to increase the possibility of attaining strong equivalence, it is essential to use a rigorous translation and cross-cultural adaptation method [
16]. In this study, the FACIT translation methodology was used, which seems to be a more rigorous version of the double-back-translation method [
26]. The FACIT translation methodology is considered more rigorous because it has a multi-step approach, including cognitive debriefing interviews in the target language [
21]. Furthermore, the methodology provided opportunities for dialogue between the reviewer/language coordinator and the FACIT project manager, where the items were discussed and further probed. This allowed us to ensure appropriate decision-making regarding the translation and cross-cultural adaptation of each item. This study encountered specific issues during the translation and cross-cultural adaptation process. There were some discussions between the FACIT project manager and the reviewer/language coordinator regarding which Swedish term should be used for the term pelvic area in three of the items. These items were debriefed in the cognitive interviews, which provided information to clarify which Swedish term should finally be used in the FACIT-CD. The cognitive debriefing interviews were a very important step in finalising items in the translation and cross-cultural adaptation process.
After an instrument has been translated and cross-culturally adapted, it is highly recommended to statistically evaluate that version produced [
16]. Accordingly, statistical analyses were performed on the Swedish FACIT-CD. With regard to the internal consistency reliability of the FACIT-CD subscales, the Cronbach’s alpha assessment demonstrated acceptable (α coefficients ≥ 0.70) to good (α coefficients ≥ 0.80) internal consistency reliability, except for the Relationships subscale, which had a somewhat low Cronbach’s alpha coefficient (0.67) [
24]. This low value for the Relationships subscale could be explained by the fact that the item
I feel close to my friends (GS1) exhibited significantly low corrected item-total correlation of 0.01. Our findings suggest that the Relationships subscale could be improved by deleting the item GS1 (deleted α = 0.82), as it did not seem to be measuring the same construct as the other items. Furthermore, each of the Physical well-being and General perceptions subscales had at least one item, respectively, which demonstrated significantly low corrected item-total correlation: specifically, 0.11 on item
I am bothered by side effects of treatment (GP5) and 0.13 on item
I am able to work (include work at home) (GF1). Nevertheless, even with the items deleted, none of them dramatically increased the respective subscales’ value of Cronbach’s alpha. The lack of studies evaluating the psychometric properties of the FACIT-CD made it impossible to perform some comparisons. However, the items mentioned above have previously been psychometrically evaluated in a study [
27] with samples from five Spanish-speaking countries using the Functional Assessment of Cancer Therapy – Gastric Cancer (FACT-Ga). In all of the Spanish countries, these items demonstrated corrected item-total correlations over the acceptable value of ≥ 0.20 [
21]. However, one of the Spanish countries demonstrated low corrected item-total correlation on the item
I feel close to my friends (GS1), and two countries on the item
I am able to work (include work at home) (GF1) [
27], which could correspond to our results. Attention must be paid to the fact that the FACIT-CD and FACT-Ga subscales differ, which affects the corrected item-total correlation in the respective subscales. Furthermore, the Physical well-being subscale also demonstrated a corrected item-total correlation of 0.08 for the item
I worry about spreading the infection (CD5), but if the item was deleted, the Cronbach’s alpha coefficient increased to 0.75. In order to improve the physical well-being subscale, our findings suggest that this item might be deleted unless there are theoretical reasons for not doing so [
17]. However, this item relates to a physical situation, about spreading the disease, which does not necessarily reflect on the emotional element about how you feel about yourself or your situation.
There were some limitations to this study. One possible limitation is related to the small sample size (
n = 34) regarding the evaluation of the internal consistency reliability. However, according to Eremenco [
21], a sample size of 15–30 is sufficient to provide preliminary evidence for internal consistency reliability using Cronbach’s alpha. Another limitation might be that the cut-off at ≥ 0.20 is used for an acceptable corrected item-total correlation, while the most common cut-off is ≥ 0.30 [
24]. However, we followed the FACIT translation methodology recommendation of the cut-off at ≥ 0.20, according to Eremenco [
21]. Furthermore, the stability of the Swedish FACIT-CD was not evaluated, as no data were collected a second time.
Further studies are recommended to conclude that the Swedish FACIT-CD is valid and reliable with a larger sample, and this is warranted before any decision on whether or not to delete the items discussed above. To confirm the original factor structure of the Swedish FACIT-CD, an exploratory factor analyses can be used. Finally, we plan to use the Swedish FACIT-CD in an intervention study with a larger sample.
Acknowledgements
We are grateful to Anna Carin Aho and Tobias Ek for their contributions in the translation and cross-cultural adaptation process. Special thanks to Associated Professor Andrejs Ozolins for statistical advice. We also thank the women who participated in the study and the healthcare professionals at the women’s health clinic for their contribution to the study.