Essential methodological steps
The
pre-test procedure represents the true crucial step in the process of developing a new questionnaire. Usually, the real effectiveness of this instrument is assessed by administering the original list of questions to a limited sample of subjects (usually people who are easily and quickly available, of both genders, with different cultural levels, and not necessarily participating in the real survey) during the
pre-test phase [
9‐
16]. Actually, a substantial proportion of mistakes are only identified thanks to the contribution of these potential respondents when they face the list of questions for the first time [
16‐
21].
All questions included in the questionnaire should be checked in terms of their
validity and reliability. The former one identifies the instrument’s ability to effectively measure one aspect with an accepted degree of precision, while the latter consists in its ability to provide reproducible measurements of the same topic, in similar conditions, over time [
13,
14,
18‐
21].
When producing an effective investigational instrument in two languages, it is much more convenient to consider the two versions of the questionnaire [
15,
22,
23]. One of the most widely used methods for “translating” a questionnaire from one language to another is the “translating/back-translating method”, which consists of four different operational phases [
22,
23]: 1) the initial translation from the research language to the foreign language by a bilingual translator; 2)
back-translation: in other words, a new translation, this time from the foreign language to the research language by a translator different from that one who performed the first translation, and who is not familiar with the original version of the questionnaire; 3) comparison between the two versions of the questionnaire written in the research language; 4) in the event of substantial differences between the two versions, the preparation of a new draft translation containing the necessary changes is needed. This procedure was strictly adopted for translating the NHQ-2 from Italian into English language.
The essential methodological steps for a questionnaire validation are summarized in Table
1.
Table 1
Methodological steps for a questionnaire validation in two languages
Pre-test |
administering the original list of questions to a limited, unselected sample of subjects; | |
identification of mistakes; | |
administering the revised list of questions; | |
Validation |
check of each question’s validity (precision) and reliability (reproducibility); | |
Translation of the definitive version of questions |
initial translation from the research language to the foreign language by a bilingual translator; | |
back-translation: a new translation, from the foreign language to the research language by a different translator; | |
comparison between the two translated versions of the questionnaire; | |
proceed in order to obtain two corresponding versions, equally understandable, and a comprehension rate >90 % for each question. | |
The operational steps towards the definitive version of the novel questionnaire
The
NHQ-2 questionnaire deeply involves both patients and nurses. For this reason, according to the accepted validation procedures, the original Italian version of the questionnaire was firstly submitted to a sample of 26 individuals (the usual sample size for a pilot test) not educated in respirology (whose gender, age, and educational level are reported in Table
2), and to a limited sample of 6 expert respiratory nurses (4 females, aged 29–41 years). As previously claimed, the aim in this phase was to detect any structural or conceptual bias within the questionnaire, and/or the existence of linguistic errors and/or mis-understandings which should affect the effectiveness of questions.
Table 2
Characteristics of the sample of subjects involved in the pre-test phase
Age | <19 | 1 | 1 | 2 | 1 | 5 |
| 19–60 | 3 | 4 | 4 | 3 | 14 |
| >60 | 2 | 1 | 2 | 2 | 7 |
Total | 6 | 6 | 8 | 6 | 26 |
Possible devices to compare by means of the NHQ-2 questionnaire during a single session were limited to a maximum of four, belonging to any family of inhalation devices, and they are generically indicated as A, B, C, and D.
All questions had been written in a single line in order to facilitate both their reading and comprehension. The great majority of questions were closed questions, except the age, those concerning possible previous education providers, and those of items 1.a; 1.c.1; 2.b, and 2.d within the Assessing Track.
Subject were required to indicate (by checking with an X), or to describe their choice in the appropriate spaces: this procedure was clearly reported on the top of the questionnaire, in the second line, just under the name of the questionnaire.
Information concerning any previous experience with different families of inhalation devices are directly required to patients. In the section of the Assessing Track, all items in group 1 and 5, items 2.a; 2b in group 2 are involving the patients directly, differently from all other remaining items which involve the nurse only.
The NHQ-2 Questionnaire did not show any structural flaw in its original version; nonetheless some changes and corrections were needed. Firstly, a much more appropriate numbering system was introduced for some items (namely, items # 1; 2; 3, and 5), in order to clarify to respondents that each battery of these questions was dependent of the main previous comprehensive item. As some items are specifically addressed to patients, while other items to the nurse, whom the item is addressed to was clearly reported for all items.
The most difficult item in terms of comprehension was item # 2, because it consists of two sub-items addressed to patients (2.a and 2.b), and the remaining two to the nurse (2.c and 2.d), exclusively.
Moreover, some English terms had been explained and implemented in the Italian version. Namely, the original terms “DPI” (Dry Powder Inhaler); “MDI” (Metered Dose Inhaler), and “SMI” (Soft Mist Inhaler), well known to professionals, were scarcely known by patients or general population, and then they were implemented with the following descriptions, namely: “Inalatore per polveri secche”; “Aerosol predosato”, and “Erogatore per nebbia inalatoria”, respectively, because otherwise not understood by the great majority of subjects ([26.9, 53.8, and 19.2 %, respectively].
Moreover, also the term “nurse”, which was reported several times in the first page of the original version of NHQ-2 (see Assessing Track and Item # 1), was changed. Even if this term is actually commonly used also in Italian, it was effectively understood by only 34.6 % of subjects (9/26), likely due to the heterogeneity of the testing sample in terms of age and education level. The same fate had the term “device”, which was also repeated several times in all the items of the original Italian version of the questionnaire. These two terms were replaced with their Italian equivalents, namely “tecnico/a” and “erogatore”, respectively.
Questions corresponding to items 1.b; 1.c, and 1.c.1 of the validated version were reworded because they resulted mis-leading in their original version, and only a poor proportion of patients perceived their meaning properly (such as: 80.7 % (21/26); 76.9 % (20/26), and 73.13 % (19/26), respectively).
Moreover, the term “in ordine crescente” has been changed into “da meno a più” in the Italian version, because the original wording had been mis-understood by 61.5 % (16/26) of subjects. In particular, in items 2.a and 2.c the grading sequence of devices was facilitated by using ordinal numbers. This new formulation tended to minimize the possibility that the respondent may attribute the opposite meaning to his own response.
Other minor changes were made to the original version of the questionnaire. The items concerning the age, the sex, and the education degree of patients were shifted to the final section of the questionnaire. As they deal with personal details, this kind of questions might be perceived by subjects as irrespective of their privacy [
8,
9].
Moreover, in order to emphasize the value of the patient role, a short thanking sentence was included at the end of the questionnaire.
The English version of the tested questionnaire was carried out by using the technique of translation and back-translation, with the contribution of human science professionals. The English translation of the original researcher’s version of the Handling Questionnaire and both the final validated versions in Italian and in English are reported in the Additional files
1 and
2, together with the corresponding percent rate of comprehension achieved for each item.
In order to better perceive and compare the different validation steps and their relevance in terms of comprehension, the
New Handling Questionnaire (NHQ-2) was reported in Additional file
3 in its original Italian version, together with the percent rate of comprehension for each question measured at the first reading of the questionnaire. Finally, the definitive Italian and English versions of the questionnaire were reported in Additional files
1 and
2, together with their corresponding comprehensibility rates.