Background: Home enteral nutrition (HEN) is a therapeutic method used in patients who are unable to ingest the required amounts of nutrients but retain a functional gastrointestinal tract. The objective of this study was to compose a specific questionnaire for measuring health-related quality of life (HRQoL) in HEN patients irrespective of their underlying condition and HEN route of administration. Methods: Literature review, focus groups and semi-structured interviews were used to propose an initial version of the questionnaire which was answered by 165 participants. The responses were analyzed using the Rasch methodology. Firstly, the appropriateness of response options was assessed. Then, the differential item functioning (DIF) was evaluated. Finally, the item fit statistics, infit and outfit, were determined. Results: Rasch analysis was performed on the responses given to the 43 items included in the initial questionnaire. Four items were excluded because more than 50% of respondents answered that the situation proposed did not apply to them. Seven items that showed overlapping and disordered categories were also removed. Pairwise DIF analysis were performed in subgroups defined by underlying disease and administration route. Eleven items presented DIF and were eliminated from the questionnaire. Finally, four items were deleted after analyzing the fit statistics, three of which did not fit the Rasch model and one did not belong to either of the dimensions. The final version of NutriQoL® includes 17 items. Conclusions: NutriQoL® is a useful instrument to assess the HRQoL of HEN patients with any disease and any administration route. Keywords: Home enteral nutrition, Health-related quality of life, Rasch analysis Background patients to remain in their social and family environment, Home enteral nutrition (HEN) is a therapeutic method reducing the probability of complications related to hospi- used in patients who are unable to ingest the required talizations, decreasing direct costs, and improving their amounts of nutrients but retain a functional gastrointes- health-related quality of life (HRQoL) . tinal tract. The main purpose of HEN is to achieve the HRQoL is a dynamic variable that evaluates the sub- caloric needs for the organism, providing the proper nutri- jective influence of health status, health care and health ents by means of the digestive tract [1, 2]. HEN allows prevention activities on patients’ capability of achieving and maintaining a functional status that permits attain- * Correspondence: firstname.lastname@example.org ing vital objectives and which reflects general wellbeing. Outcomes’10, Universitat Jaume 1, Parc Cientific, Tecnologic i Empresarial, HRQoL is a multidimensional concept whose essential Edificio Espaitec 2, Campus del Riu Sec, Avenida Sos Baynat s/n, 12071 dimensions are: physical, psychological and cognitive Castellon de la Plana, Castellon, Spain Departamento de Medicina, Universidad Jaime I, Campus Riu Sec, Avenida and social functioning . The measurement of HRQoL Sos Baynat s/n, 12071 Castellón de la Plana, Spain Full list of author information is available at the end of the article © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. Apezetxea et al. Journal of Patient-Reported Outcomes (2018) 2:25 Page 2 of 11 is useful to assess changes resulting from therapeutic in- to the importance given to situations mentioned in the terventions or the course of the disease . part ‘a’. This second part of the items was included to Several studies have assessed HRQoL in patients with consider that all aspects of life are not equally important HEN [6–11]. The majority of publications evaluated for everyone. This is the undelaying idea of question- HRQoL using generic instruments such as EQ-5D, SF- naires like the SEIQoL, a method which values the 12 and SF-36 [6, 7, 9–11]. A major limitation of these HRQoL from the individual perspective . tools is that in spite of providing an overall assessment A second focus group meeting with the same previously of HRQoL and making it possible to compare results consulted experts was carried out to evaluate the items in- with other populations, they are too insensitive to estab- cluded in NutriQoL® version 0. For each item, participants lish and measure the influence of aspects directly related rated the clarity of the wording, the frequency of occur- to HEN [4, 6]. Combining this limitation with the fact rence in routine clinical practice, the importance of the that aspects such as mental health and emotions are less item in the HRQoL and the relevance of the item in the affected in patients receiving HEN [12, 13] would sug- therapeutic decisions. Participants scored clarity, fre- gest that HRQoL results in patients receiving HEN eval- quency, importance and relevance on a 5-point Likert scale uated with generic questionnaires may not represent the (1 = not clear/ frequent/important/relevant and 5 = very reality. On the other hand, specific questionnaires are clear/ frequent/important/relevant). The items remained in available only for patients with a specific disease and a the questionnaire if clarity, frequency, relevance and im- specific type of HEN . In particular, one tool was portance were high, or if clarity, relevance and importance identified that assesses the impact of enteral feeding were high even though the frequency was low. In cases of tubes in HRQoL in patients diagnosed with head and high score for frequency, relevance and importance with a neck cancer, the QOL-EF . This questionnaire is not low score for clarity, the wording of the items was modified applicable to other HEN administration methods or to improve this feature [15, 17]. High scores were those other underlying conditions requiring enteral feeding. above the 25th percentile. Based on this assessment, Rasch analysis is a widely used method for the develop- version 1 of the questionnaire was issued. ment and reduction of questionnaires [14–16] in a wide Finally, 19 participants (patients and caregivers) range of disciplines, including health [17–19], education responded to questionnaire version 1 face validity and  and psychology . It is a methodology that belongs feasibility measures. They gave their opinion about the to the item response theory (IRT) and allows obtaining a clarity of wording and comprehension, the importance questionnaire in which each item response is taken as an of items, the general appearance of the questionnaire and outcome of the independent interaction between the re- the time required for completion. The wording of items spondent’s abilities and the item difficulty . The that received unfavorable opinion in clarity and compre- method provides an alternative approach that overcomes hension by at least 5% of the participants was changed. the limitations of the classical test theory (CTT) [23, 24]. Items which receive negative opinion in importance for at Rasch methodology also allows for analyzing differential least 10% of participants were removed from questionnaire. item functioning (DIF) which identifies items that do not The general presentation of the questionnaire was modi- function equally in different groups of participants . fied if at least 10% of the participants gave an unfavorable The aim of this study is to develop a specific question- opinion. Based on these results, we drew up version 2 of naire to assess HRQoL in patients receiving HEN re- the questionnaire to be used in the Rasch analysis. gardless of the disease and the administration route. Rasch analysis Methods In order to scale and reduce the questionnaire items, we NutriQoL® was developed between 2011 and 2012. To performed a Rasch analysis using the Partial Credit carry out this development, a comprehensive literature Model because it is a proper model from the IRT family review, an expert focus group (n = 7) and a series of for items with polytomous responses with ordered cat- semi-structured interviews with patients (n = 21) and egories . For this purpose, an observational, pro- their primary caregivers (n = 10) were performed to spective and non-controlled pilot study was carried out. identify the main dimensions of HRQoL in patients with One hundred and sixty-five participants (150 required HEN. This allowed drawing up an initial draft of the by the analysis  and an additional 10% for possible questionnaire, which we refer to as NutriQoL® version 0. loss of data) answered version 2 of the NutriQoL® ques- Each item was structured in two parts: ‘a’ and ‘b’. Part a’ tionnaire. Participant selection criteria included having asked about the frequency of which HEN affects particu- received HEN during at least one month and having given lar aspects of their HRQoL. This part was written trying their written consent for participation. The sample selected to reflect literally the information provided by the pa- was non-random, intentional, consecutive, proportional ac- tients in the semi-structured interviews. Part ‘b’ referred cording to the main epidemiology of HEN (oncologic, Apezetxea et al. Journal of Patient-Reported Outcomes (2018) 2:25 Page 3 of 11 Table 1 Items description Table 1 Items description (Continued) Wording of part a Wording of part a Item 1 I keep my habitual mealtimes because of HEN Item 29 I find it unpleasant to eat with other people (e.g. breakfast, lunch, and dinner) because of HEN. Item 2 I can adjust HEN to unexpected events Item 30 HEN lets me eat with my family. (e.g. leaving the house suddenly, an Item 31 I limit activities with my friends to those that are unscheduled visit). not food-related because of HEN. Item 3 I can feed myself with HEN and I do not Item 32 I sense that others feel bad about me because have to be hospitalized for that. of HEN. Item 4 I am forced to spend most of the day sitting Item 33 Since receiving HEN, my family is more worried or lying down for food because of HEN. about my health. Item 5 HEN suits my preferences (e.g. smell, temperature, Item 34 Since receiving HEN, my family watches over and flavor). my food. Item 6 I find HEN products suit my other health Item 35 Since receiving HEN, I have to explain how I feed. problems (e.g. products without sugar because I am diabetic, without salt because I am Item 36 Receiving HEN causes me an added problem. hypertensive). Item 37 Since receiving HEN, I feel stronger and Item 7 Since I receive HEN, I move more easily; I feel more energetic. more agile. Item 38 Since receiving HEN, I am more worried about Item 8 HEN damages my skin (e.g. dryness, my health. irritation, infection). Item 39 Since receiving HEN, I feel sad. Item 9 HEN lets me eat away from home. Item 40 I would change the HEN administration route Item 10 Since receiving HEN, I need more help with (e.g. From nasogastric tube to oral administration). my personal hygiene (e.g. showering, dressing). Item 41 I trust I am well-nourished because of HEN. Item 11 HEN lets me keep doing housework Item 42 I have gained weight because of HEN. (e.g. cooking, ironing, cleaning). Item 43 I have adapted to receiving HEN. Item 12 HEN lets me enjoy my hobbies. Item 13 HEN hinders having sex. neurological and malabsorption) and administration route Item 14 Since receiving HEN, I have problems to travel several days. (oral, ostomy and nasogastric tube) in Spain . Item 15 After some time receiving HEN, I can think about Rasch analysis constitutes a scaling and reduction pro- going back to work. cedure with the following steps. Firstly, the appropriate- Item 16 HEN lets me go out with my friends. ness of response options is assessed. Rasch models enable ensuring that the features of the parameters of Item 17 Since receiving HEN, I have stopped going to family celebrations (e.g. weddings, birthdays). the respondents and the questionnaire items are inde- Item 18 HEN prevents me from sleeping well. pendent, i.e. each respondent’s estimated ability does not depend on the number and type of items they have an- Item 19 I have a good appetite because of HEN. swered nor does the difficulty of the items depend on Item 20 Since receiving HEN, I fend for myself; I can do the number and type of respondents who have answered more things without help. them [22, 29], thus ensuring that all items work in the same Item 21 Since receiving HEN, my physical appearance is improving (e.g. I look healthier). way for all respondents completing them . This makes Rasch analysis different from CTT, the most widely used Item 22 I worry that my body is adapting to HEN and I will not be able to feeding as before. measurement model in health sciences in the last century, in which conversely a respondent’s observed score depends Item 23 I miss chewing and tasting food because of HEN. on the number and difficulty of the items in the test. Thus, Item 24 Since receiving HEN, I find it a sacrifice to prepare food for others. the number and the particular skills of the respondents in- fluence the final score in the questionnaire . Item 25 I need help to eat because I cannot do it alone with HEN. Rasch estimations are represented graphically by cat- egory probability curves [16, 19]. A requirement of Rasch Item 26 My food preparation is simpler because of HEN. is that response categories must be ordered increasingly Item 27 I have physical discomfort because of HEN (so the lowest category shows the worst HRQoL and the (e.g. bloating, heartburn, dry mouth, regurgitation). highest the best HRQoL), so the curves should be dis- Item 28 Getting HEN products is simple (i. e. they are played in the same order. Item curves with overlapped cat- available in pharmacies and I can easily get egories suggest an excess of response options that should the preparation). be reduced, while disordered curves indicate that the Apezetxea et al. Journal of Patient-Reported Outcomes (2018) 2:25 Page 4 of 11 categorization of responses does not work as intended . (v2) whose wording is showed in Table 1. Items that form After the reduction of response options, items whose the final version have been indicated in italics. curves remained disordered were deleted from NutriQoL®. The second step in the reduction procedure is the assess- Rasch results ment of DIF. The analysis of DIF is a measurement bias Rasch analysis was performed with the answers of the 43 and refers to differences in the probability of giving a cer- specific items from version 2 of NutriQoL® provided by tain response between groups [16, 18, 25, 30]. The assess- 165 participants. The main characteristics of the 165 par- ment was carried out between pairs of groups defined by ticipants (141 patients and 24 caregivers that responded the underlying disease (oncologic, neurological and malab- on behalf of patients) are presented in Table 2. sorption/others) and the administration route of HEN (oral, The percentage of answers by category response in each nasogastric tube, and ostomy) in order to obtain a ques- item was calculated (Table 3). Items with a response rate of tionnaire whose measures were independent of those char- greater than 50% to the option Does not apply to my acteristics. Within the Rasch modeling, there are different current situation were eliminated as they were considered methods for DIF analysis, one of which is Andersen’s Likeli- not to represent the reality of the patient. Applying this cri- hood Ratio Test  that was used in this study. This DIF terion, items 6 (53.21%), 13 (52.56%), 15 (67.31%) and 24 analysis consists of comparing Rasch analysis estimations (55.13%) were removed from the questionnaire. between pairs of groups. The graphical representation of One of the requirements of Rasch methodology is that each comparison using this test identified items with DIF as item response categories must be ordered in increasing the ones located outside confidence intervals. Items with level of HRQoL. To accomplish this criterion where a DIF were deleted from the questionnaire. Table 2 Main characteristics of participants interviewed for The last step consisted of determining the item fit statis- Rasch analysis tics: infit and outfit. These statistics are based on residuals Characteristic Description (difference between observed and predicted responses). Sex [n(%)] They indicate how well each item fits the Rasch model Men 104 (64.20%) and let us to assess the unidimensionality of the set of items analyzed. Outfit is more sensitive to unexpected re- Women 58 (35.80%) sponses in items which are far from person measure, Age [Mean (SD)] 61 (15) whereas Infit is more sensitive to unexpected responses in Charlson index [Mean (SD)] 3.26 (2.43) items which are close to person measure . The infit Karnofsky index [Mean (SD)] 70.41 (16.99) and outfit mean squares were converted to an approxi- Main disease [n(%)] mately normalized t-statistic by the Wilson-Hilferty Oncologic 93 (56.36%) method to be more conveniently represented. The interval of values that determined a good fit was [− 2, 2] [33–35]. Neurological 31 (18.79%) We considered that belonged to the same dimension Malabsorption 41 (24.85%) those items whose infit and outfit t values were inside HEN duration [n(%)] [− 2, 2]. Items whose infit t and outfit t values were lo- 1-3 months 59 (35.76%) cated outside the mentioned interval were analyzed 4-6 months 34 (20.61%) separately, as part of another dimension, or removed. 7-9 months 15 (9.09%) All calculations involved in Rasch analysis were made using the eRm package of R statistical software [36, 37]. 10-12 months 15 (9.09%) More than 12 months 42 (25.45%) Results HEN function [n(%)] During NutriQoL® development, the number and phrasing Sole source of nutrition 105 (63.64%) of the items was progressively modified according to partic- Nutritional supplement 60 (36.36%) ipants’ responses and statistical analysis. Initially 52 items Administration route [n(%)] were included in the first version of the questionnaire (v0), which were taken from the literature review, a focus group Oral 92 (55.76%) with experts and semi-structured interviews with patients. Ostomy 50 (30.30%) Subsequently, during the second focus group meeting with Nasogastric tube 17 (10.30%) experts, the number of questionnaire items was reduced to Oral and ostomy 4 (2.42%) 46 (v1), based on item frequency, importance and clarity. Oral and nasogastric tube 1 (0.61%) After the face validity and feasibility steps, 3 items were Nasogastric tube and ostomy 1 (0.61%) eliminated because of problems with clarity and compre- hension, so the number of items was further reduced to 43 Data unavailable for 3 participants Apezetxea et al. Journal of Patient-Reported Outcomes (2018) 2:25 Page 5 of 11 Table 3 Percentage of answers by category response in each item Items Never (%) Sometimes (%) Usually (%) Always (%) Does not apply to my No answer (%) current situation (%) 1 I keep my habitual mealtimes because of HEN 3.84% 6.41% 32.05% 51.92% 5.13% 0.64% (e.g. breakfast, lunch, and dinner) 2 I can adjust HEN to unexpected events (e.g. leaving 6.41% 25.64% 17.31% 39.74% 9.61% 1.28% the house suddenly, an unscheduled visit). 3 I can feed myself with HEN and I do not have to be 3.85% 1.28% 6.41% 71.15% 15.38% 1.92% hospitalized for that. 4 I am forced to spend most of the day sitting or 39.74% 10.90% 14.74% 12.82% 19.87% 1.92% lying down for food because of HEN. 5 HEN suits my preferences (e.g. smell, temperature, 6.41% 12.82% 23.72% 31.41% 25.00% 0.64% and flavor). 6 I find HEN products suit my other health problems 9.62% 1.92% 8.33% 24.36% 53.21% 2.56% (e.g. products without sugar because I am diabetic, without salt because I am hypertensive). 7 Since I receive HEN, I move more easily; I feel more 10.26% 16.03% 21.79% 28.85% 21.15% 1.92% agile. 8 HEN damages my skin (e.g. dryness, 52.56% 12.82% 1.92% 1.28% 30.13% 1.28% irritation, infection). 9 HEN lets me eat away from home. 17.95% 18.59% 7.69% 39.10% 14.74% 1.92% 10 Since receiving HEN, I need more help with my 53.85% 6.41% 5.13% 6.41% 27.56% 0.64% personal hygiene (e.g. showering, dressing). 11 HEN lets me keep doing housework (e.g. cooking, 5.13% 14.10% 10.90% 31.41% 37.18% 1.28% ironing, cleaning). 12 HEN lets me enjoy my hobbies. 5.77% 18.59% 16.03% 38.46% 19.23% 1.92% 13 HEN hinders having sex. 32.05% 4.49% 2.56% 5.77% 52.56% 2.56% 14 Since receiving HEN, I have problems to travel 42.95% 8.33% 3.21% 15.38% 28.85% 1.28% several days. 15 After some time receiving HEN, I can think about 13.46% 8.33% 2.56% 5.77% 67.31% 2.56% going back to work. 16 HEN lets me go out with my friends. 10.26% 13.46% 14.10% 44.23% 16.67% 1.28% 17 Since receiving HEN, I have stopped going to 50.00% 12.82% 5.77% 12.82% 16.67% 1.92% family celebrations (e.g. weddings, birthdays). 18 HEN prevents me from sleeping well. 69.23% 12.18% 0.64% 1.92% 14.74% 1.28% 19 I have a good appetite because of HEN. 16.67% 28.85% 19.23% 23.72% 10.90% 0.64% 20 Since receiving HEN, I fend for myself; I can do 9.62% 13.46% 25.00% 24.36% 26.92% 0.64% more things without help. 21 Since receiving HEN, my physical appearance is 6.41% 19.87% 25.00% 42.95% 5.13% 0.64% improving (e.g. I look healthier). 22 I worry that my body is adapting to HEN and I 21.15% 17.31% 7.05% 22.44% 30.77% 1.28% will not be able to feed as before. 23 I miss chewing and tasting food because of HEN. 14.74% 12.82% 5.77% 32.05% 33.33% 1.28% 24 Since receiving HEN, I find it a sacrifice to prepare 26.28% 8.33% 3.85% 4.49% 55.13% 1.92% food for others. 25 I need help to eat because I cannot do it alone 47.44% 4.49% 3.85% 10.90% 32.05% 1.28% with HEN. 26 My food preparation is simpler because of HEN. 14.10% 10.90% 5.77% 26.28% 39.74% 3.20% 27 I have physical discomfort because of HEN 46.15% 35.26% 5.77% 7.05% 5.13% 0.64% (e.g. bloating, heartburn, dry mouth, regurgitation). 28 Getting HEN products is simple (i. e. they are 5.77% 5.77% 15.38% 67.31% 4.49% 1.28% available in pharmacies and I can easily get the preparation). 29 I find it unpleasant to eat with other people 53.85% 9.62% 3.85% 8.97% 21.79% 1.92% because of HEN. Apezetxea et al. Journal of Patient-Reported Outcomes (2018) 2:25 Page 6 of 11 Table 3 Percentage of answers by category response in each item (Continued) Items Never (%) Sometimes (%) Usually (%) Always (%) Does not apply to my No answer (%) current situation (%) 30 HEN lets me eat with my family. 14.74% 13.46% 8.97% 46.15% 14.10% 2.56% 31 I limit activities with my friends to those 38.46% 10.90% 8.97% 14.10% 25.00% 2.56% that are not food-related because of HEN. 32 I sense that others feel bad about me 50.64% 14.74% 7.05% 7.05% 18.59% 1.92% because of HEN. 33 Since receiving HEN, my family is more 44.23% 17.31% 10.26% 19.23% 7.69% 1.28% worried about my health. 34 Since receiving HEN, my family watches 14.10% 19.23% 14.10% 44.23% 7.69% 0.64% over my food. 35 Since receiving HEN, I have to explain how 37.82% 31.41% 8.33% 5.13% 15.38% 1.92% I feed. 36 Receiving HEN causes me an added problem. 60.90% 17.31% 7.05% 6.41% 7.05% 1.28% 37 Since receiving HEN, I feel stronger and 7.05% 23.72% 23.72% 39.10% 5.13% 1.28% more energetic. 38 Since receiving HEN, I am more worried about 43.59% 22.44% 8.97% 11.54% 10.90% 2.56% my health. 39 Since receiving HEN, I feel sad. 60.26% 21.15% 3.85% 3.85% 8.97% 1.92% 40 I would change the HEN administration route 46.79% 5.13% 0.64% 9.62% 34.62% 3.20% (e.g. From nasogastric tube to oral administration). 41 I trust I am well-nourished because of HEN. 1.28% 2.56% 12.82% 80.14% 1.92% 1.28% 42 I have gained weight because of HEN. 11.54% 28.21% 17.31% 36.54% 3.21% 3.20% 43 I have adapted to receiving HEN. 2.56% 6.41% 23.72% 64.10% 1.28% 1.92% ab Fig. 1 Category probability curves for items 4 (a) and 5 (b) Apezetxea et al. Journal of Patient-Reported Outcomes (2018) 2:25 Page 7 of 11 higher score refers to higher HRQoL, category responses Items 4, 8, 10, 14, 17, 18, 22, 23, 25, 27, 29, 31, of the 43 items were codified as follows: 32, 33, 34, 35, 36, 38, 39 and 40: Never = 3, Sometimes = 2, Usually = 1 and Always = 0. Items 1, 2, 3, 5, 7, 9, 11, 12, 16, 19, 20, 21, 26, 28, 30, 37, 41, 42 and 43: Never = 0, Sometimes = 1, Category probability curves were represented for the Usually = 2 and Always = 3. remaining 39 items (43 minus 4) to assess the appro- priateness of response options. Figure 1 shows the es- timates of the thresholds corresponding to items 4 Fig. 2 Differential item functioning in groups defined by the underlying Fig. 3 Differential item functioning in groups defined by administration disease (oncologic vs. neurological, malabsorption vs. oncologic route (nasogastric tube vs. oral, ostomy vs. nasogastric tube and ostomy and malabsorption vs. neurological disease) vs. oral) Apezetxea et al. Journal of Patient-Reported Outcomes (2018) 2:25 Page 8 of 11 (Fig. 1a)and 5(Fig. 1b). Curves of item 4 show over- disease and administration route, respectively. Eleven items lapping, however in item 5, category curves are ordered (4, 9, 12, 14, 19, 20, 26, 30, 35, 37 and 43) were located out- correctly without overlapping, thus fulfilling the require- side the confidence intervals (blue lines), i.e. the ones with ments of Rasch analysis. As most of items showed over- DIF, and were eliminated from the questionnaire. lapping, we decided to reduce the number of category The last stage involved in Rasch analysis was the as- responses, so options Sometimes and Usually were unified sessment of item fit statistics (infit t and outfit t). A new in one category. The new categorization was: Rasch analysis was performed with the 21 remaining items. As a result, 14 items (1, 5, 7, 8, 11, 18, 21, 22, 23, Items 1, 2, 3, 5, 7, 9, 11, 12, 16, 19, 20, 21, 26, 27, 28, 34, 31, 41 and 42) showed values of infit t and 28, 30, 37, 41, 42 and 43: Never = 0, Sometimes = outfit t within the range [− 2, 2]. Dimension 1 of the Usually =1, and Always = 2. questionnaire was constituted by this group of 14 items. Items 4, 8, 10, 14, 17, 18, 22, 23, 25, 27, 29, 31, Another Rasch model was applied with the seven 32, 33, 34, 35, 36, 38, 39 and 40: Never = 2, items that showed misfit in the previous analysis. The Sometimes = Usually 1, and Always = 0. goodness of fit of the new analysis presented accept- able values of infit t and outfit t statistics in three Another Rasch analysis was performed to assess the items (16, 31 and 38) which constituted dimension 2. new categorization of items and their results showed A new analysis showed that the four remaining items that 32 of the 39 items presented the curves ordered did not fit the Rasch model with the exception of item correctly. The 7 items that showed overlapping and dis- 2, though it did not belong to any dimension (Table 4). ordered categories were removed from NutriQoL®. These four items were deleted from NutriQoL®. Analysis of DIF was performed with the remaining items As a result of the reduction procedure by means of (32 items). Figures 2 and 3 show the graphical representa- Rasch analysis, the final version of NutriQoL® was com- tion of pair comparisons of groups defined by underlying posed of 17 pairs of items divided into two dimensions Table 4 Infit and outfit statistics Items Dimension Infit t Outfit t 1 I keep my habitual mealtimes because of HEN (e.g. breakfast, lunch, and dinner) 1 −1.04 −1.03 5 HEN suits my preferences (e.g. smell, temperature, and flavor). 1 0.02 0.61 7 Since I receive HEN, I move more easily; I feel more agile. 1 −1.84 −1.77 8 HEN damages my skin (e.g. dryness, irritation, infection). 1 1.19 0.49 11 HEN lets me keep doing housework (e.g. cooking, ironing, cleaning). 1 −1.57 −1.67 18 HEN prevents me from sleeping well. 1 0.15 −0.93 21 Since receiving HEN, my physical appearance is improving (e.g. I look healthier). 1 −1.84 −1.68 22 I worry that my body is adapting to HEN and I will not be able to feed as before. 1 0.30 0.49 23 I miss chewing and tasting food because of HEN. 1 0.65 0.79 27 I have physical discomfort because of HEN (e.g. bloating, heartburn, dry 1 −1.95 −1.45 mouth, regurgitation). 28 Getting HEN products is simple (i. e. they are available in pharmacies and 1 1.25 1.42 I can easily get the preparation). 34 Since receiving HEN, my family watches over my food. 1 −1.38 −1.15 41 I trust I am well-nourished because of HEN. 1 0.51 0.61 42 I have gained weight because of HEN. 1 −1.13 −1.40 16 HEN lets me go out with my friends. 2 −1.24 −1.36 31 I limit activities with my friends to those that are not food-related because 2 −1.32 −1.43 of HEN. 33 Since receiving HEN, my family is more worried about my health. 2 −1.05 −1.32 2 I can adjust HEN to unexpected events (e.g. leaving the house suddenly, Without dimension 1.86 1.81 an unscheduled visit). 32 I sense that others feel bad about me because of HEN. Without dimension −3.57 −4.59 36 Receiving HEN causes me an added problem. Without dimension −3.64 −4.37 39 Since receiving HEN, I feel sad. Without dimension −3.00 −2.92 Apezetxea et al. Journal of Patient-Reported Outcomes (2018) 2:25 Page 9 of 11 whose name was established according to the content of was also used in a study that aimed to assess HRQoL in the items that were part of each one: 1) physical func- patients with home enteral tube feeding . Authors tioning and activities of daily living, and 2) social life as- measured patients’ HRQoL at two and ten weeks after pects. These are usual dimensions of HRQoL in HEN discharge, however, the results did not show differences patients . between the two time points. Furthermore, the results The different stages of the questionnaire development and findings of the study could not be generalizable due and item reduction are detailed in Fig. 4. to the small sample size. Another study used the ques- tionnaires SF-36 and EQ-5D to evaluate the HRQoL in patients using long term HEN . Subgroup analyses Discussion were performed based on the age and cancer diagnosis In this study, the investigators present the use of Rasch of patients. Results from EQ-5D did not reflect differ- analysis to develop NutriQoL®, a novel questionnaire to ences between subgroups, while results from SF-36 measure HRQoL in patients with HEN, irrespective of only showed statistical differences in physical function- their underlying condition and route of administration. ing and role-emotional in age and cancer subgroups, Throughout a literature review, only one specific tool, respectively. Authors attributed the lower sensitivity of the QOL-EF questionnaire, was identified to assess the EQ-5D to the smaller number of items and the small impact of enteral feeding tubes in HRQoL. However, sample size, in fact, they stated that their results did the QOL-EF is specific for patients with head and neck not represent the HRQoL of HEN patients. The ques- cancer . tionnaire EQ-5D was also used in another study that Previous studies have assessed the HRQoL on patients assessed the HRQoL in patients with HEN . Au- receiving HEN by means of the widely used generic thors highlighted the lack of specific and relevant vali- questionnaires SF-12, EQ-5D or the SF-36 [6, 7, 9–11]. dated measurement tools to evaluate the HRQoL in The SF-12 questionnaire was used to measure the this type of patients. They stated that the measures of HRQoL in patients with percutaneous endoscopic gas- HRQoL obtained by means of specific tools would be trostomy . The results showed that the questionnaire necessary to detect specific aspects of illnesses or treat- scores did not reflect some of the patients’ problems, ments. These findings highlight the need for an instru- such as the nausea they constantly experienced. SF-12 ment like NutriQoL to measure HRQOL in patients Fig. 4 Development of the NutriQoL questionnaire Apezetxea et al. Journal of Patient-Reported Outcomes (2018) 2:25 Page 10 of 11 receiving HEN irrespective of their underlying condi- Publisher’sNote Springer Nature remains neutral with regard to jurisdictional claims in tion and route of administration. published maps and institutional affiliations. The main limitation of the study is the sample size. Al- though the sample size is appropriate for Rasch analysis Author details 1 2 Organización Sanitaria Integrada Bilbao-Basurto, Bilbao, Spain. Centro de , when the sample was divided into subgroups to Salud Victoria de Acentejo, Santa Cruz de Tenerife, Spain. Department of perform the DIF analysis, the size of them were very Medicine, Universidad de Santiago de Compostela, Complejo Hospitalario small and DIF have limitations in this kind of situations. Universitario de Santiago (CHUS); CIBER de Fisiopatologia Obesidad y Nutricion (CIBERobn), Instituto Salud Carlos III, Santiago de Compostela, This weakness will be the focus of future research in Spain. Hospital General Universitario Gregorio Marañón, Madrid, Spain. order to improve the NutriQoL® results and assess the 5 6 Hospital Clínico San Carlos, Madrid, Spain. Hospital Universitario Nuestra properties of the items that are part of it. Señora de Valme, Sevilla, Spain. Hospital Universitario de Bellvitge, L’Hospitalet de Llobregat, Barcelona, Spain. Nestlé Health Science, Barcelona, Additional analyses were carried out to ensure the re- Spain. Outcomes’10, Universitat Jaume 1, Parc Cientific, Tecnologic i producibility of the results (reliability), NutriQoL®s Empresarial, Edificio Espaitec 2, Campus del Riu Sec, Avenida Sos Baynat s/n, sensitivity to changes in patients’ health (responsiveness) 12071 Castellon de la Plana, Castellon, Spain. Departamento de Medicina, Universidad Jaime I, Campus Riu Sec, Avenida Sos Baynat s/n, 12071 and the reliability of answers between patient and Castellón de la Plana, Spain. caregiver (inter-observer reliability). These analyses were performed in a prospective study between 2013 and Received: 16 November 2017 Accepted: 24 April 2018 2014 whose results are detailed in other publication . Therefore, NutriQoL®, it is a useful instrument to assess the HRQoL of HEN patients with any disease and any References 1. Schneider, S. M., Raina, C., Pugliese, P., Pouget, I., Rampal, P., Hébuterne, X., administration route in a context were such an instrument et al. (2001). Outcome of patients treated with home enteral nutrition. was not previously available. Journal of Parenteral and Enteral Nutrition, 25, 203–209. 2. Guía de nutrición enteral domiciliaria en el Sistema Nacional de Salud. Ministerio de Sanidad y Consumo. 2008. https://www.msssi.gob.es/ Conclusion profesionales/prestacionesSanitarias/publicaciones/docs/guiaNED.pdf. Accesed 20 May 2011. A new tool has been developed to assess the impact of 3. McNamara, E. P., Flood, P., & Kennedy, N. P. (2001). Home tube feeding: An HEN in patients’ HRQoL (the NutriQoL® questionnaire). integrated multidisciplinary approach. Journal of Human Nutrition and Rasch methodology has allowed performing a short Dietetics, 14,13–19. questionnaire composed of 17 items able to measure 4. Lizán, L. (2009). La calidad de vida relacionada con la salud. Atencion Primaria, 41, 411–416. HEN-related HRQoL irrespective of the patients’ under- 5. Kaplan, R. M., & Ries, A. L. (2007). Quality of life: Concept and definition. lying disease and the route of administration. The Nutri- COPD, 4, 263–271. QoL® questionnaire provides a specific instrument that 6. Bannerman, E., Pendlebury, J., Phillips, F., & Ghosh, S. (2000). A cross- sectional and longitudinal study of health-related quality of life after may be used in clinical practice to adjust treatments ac- percutaneous gastrostomy. European Journal of Gastroenterology & cording to HRQoL results. Hepatology, 12, 1101–1109. 7. Jordan, S., Philpin, S., Warring, J., Cheung, W. Y., & Williams, J. (2006). Percutaneous endoscopic gastrostomies: The burden of treatment from a Acknowledgements patient perspective. Journal of Advanced Nursing, 56, 270–281. We thank Outcomes’10 for the coordination of the study. 8. Stevens, C. S. M., Lemon, B., Lockwood, G. A., Waldron, J. N., Bezjak, A., & Ringash, J. (2011). The development and validation of a quality-of-life Funding questionnaire for head and neck cancer patients with enteral feeding tubes: Nestlé Health Science. The QOL-EF. Support Care Cancer, 19, 1175–1182. 9. Schneider, S. M., Pouget, I., Staccini, P., Rampal, P., & Hebuterne, X. (2000). Quality of life in long-term home enteral nutrition patients. Clinical Nutrition, 19,23–28. Authors’ contributions 10. Wanden-Berghe, C., Nolasco, A., Sanz-Valero, J., Planas, M., & Cuerda, C. All authors took part in the development of the study, reviewed the (2009). Health-related quality of life in patients with home nutritional methodology, the analysis and interpretation of results. They significantly support. Journal of Human Nutrition and Dietetics, 22, 219–225. contributed to the conception, design, progress and conclusion of the study. 11. Bjuresäter, K., Larsson, M., Athlin, E., & Nordström, G. (2014). Patients living All authors critically reviewed preliminary drafts for important intellectual with home enteral tube feeding: Side effects, health-related quality of life content and contributed to its final version. They all gave their final approval and nutritional care. Clinical Nursing Studies, 2, 64. of the version being submitted. 12. Reddy, P., & Malone, M. (1998). Cost and outcome analysis of home parenteral and enteral nutrition. Journal of Parenteral and Enteral Nutrition, 22, 302–310. Ethics approval and consent to participate 13. Brotherton, A. M., & Judd, P. A. (2007). Quality of life in adult enteral tube Not applicable feeding patients. Journal of Human Nutrition and Dietetics, 20, 513–522. 14. Badia, X., Prieto, L., Roset, M., Díez-Pérez, A., & Herdman, M. (2002). Competing interests Development of a short osteoporosis quality of life questionnaire by Nestlé Health Science sponsored the study. The authors Cuerda C, Apezetxea equating items from two existing instruments. Journal of Clinical A, Carrillo L, Casanueva F, Cuesta F, Irles JA and Virgilli N, state they have no Epidemiology, 55,32–40. conflict of interest. Layola M works at Nestlé Health Science. Lizan L works 15. Webb, S. M., Prieto, L., Badia, X., Albareda, M., Catalá, M., Gaztambide, S., et al. for an independent research organization (Outcomes’10, S.L.) which has (2002). Acromegaly quality of life questionnaire (ACROQOL) a new health- received fees for its contribution to the development and coordination of related quality of life questionnaire for patients with acromegaly: Development the original research project and to the writing of this manuscript. and psychometric properties. Clinical Endocrinology, 57,251–258. Apezetxea et al. Journal of Patient-Reported Outcomes (2018) 2:25 Page 11 of 11 16. Hagquist, C., Bruce, M., & Gustavsson, J. P. (2009). Using the Rasch model in nursing research: An introduction and illustrative example. International Journal of Nursing Studies, 46, 380–393. 17. Badia, X., Webb, S. M., Prieto, L., & Lara, N. (2004). Acromegaly quality of life questionnaire (AcroQoL). Health and Quality of Life Outcomes, 2, 13. 18. Djaja, N., Youl, P., Aitken, J., & Janda, M. (2014). Evaluation of a skin self examination attitude scale using an item response theory model approach. Health and Quality of Life Outcomes, 12, 189. 19. Vincent, J. I., MacDermid, J. C., King, G. J., & Grewal, R. (2015). Rasch analysis of the patient rated elbow evaluation questionnaire. Health and Quality of Life Outcomes, 13, 84. 20. Lambert, M. C., Cress, C. J., & Epstein, M. H. (2015). Rasch analysis of the patient rated elbow evaluation questionnaire. Health and Quality of Life Outcomes, 13, 84. 21. Hadzibajramovic, E., Ahlborg Jr., G., Grimby-Ekman, A., & Lundgren-Nilsson, Å. (2015). Internal construct validity of the stress-energy questionnaire in a working population, a cohort study. BMC Public Health, 15, 180. 22. Masters, G. N. (1982). A Rasch model for partial credit scoring. Psychometrika, 47, 149–174. 23. Magno, C. Demonstrating the difference between classical test theory and item response theory using derived test data. 2009. 24. Prieto, G., & Delgado, A. R. (2003). Análisis de un test mediante el modelo de Rasch. Psicothema, 15,94–100. 25. Tennant, A., & Pallant, J. F. (2007). DIF matters: A practical approach to test if differential item functioning makes a difference. Rasch measurement transactions., 20, 1082–1084. 26. Joyce, C. R. B., Hickey, A., McGee, H. M., & O’Boyle, C. A. (2003). A theory- based method for the evaluation of individual quality of life: The SEIQoL. Quality of Life Research, 12, 275–280. 27. Linacre, J. M. (1994). Sample size and item calibration stability. Rasch measurement transactions, 7, 328. 28. Luengo Pérez, L. M., Chicharro, M. L., Cuerda, C., García Luna, P. P., & Rabassa Soler, A. (2009). Romero, et al. Registro de nutrición enteral domiciliaria en España en el año 2007. Nutrición Hospitalaria, 24, 655–660. 29. Dols, O., & López, J. A. (2011). Análisis psicométrico de la subescala manipulación de objetos de la escala peabody de desarrollo motor segunda edición (PDMS-2) con el modelo de Rasch. Fisioterapia, 33,145–156. 30. Dencker, A., Sunnerhagen, K. S., Taft, C., & Lundgren-Nilsson, Å. (2015). Multidimensional fatigue inventory and post-polio syndrome–a Rasch analysis. Health and Quality of Life Outcomes, 13, 20. 31. Andersen, E. B. (1973). A goodness of fit test for the Rasch model. Psychometrika, 38, 123–140. 32. Anselmi, P., Vidotto, G., Bettinardi, O., & Bertolotti, G. (2015). Measurement of change in health status with Rasch models. Health and Quality of Life Outcomes, 13, 16. 33. Linacre, J. M. (2002). What do Infit and outfit, mean-square and standardized mean? Rasch Measurement Transactions., 16, 878. 34. Smith, A. B., Rush, R., Fallowfield, L. J., Velikova, G., & Sharpe, M. (2008). Rasch fit statistics and sample size considerations for polytomous data. BMC Medical Research Methodology, 8, 33. 35. Schulz, M. (2002). Standardization of mean-squares. Rasch Measurement Transactions, 16, 879. 36. Mair P, Hatzinger R, Maier JM. eRm: Extended Rasch modeling. R package version 2.14.0. 2012. http://CRAN.R-project.org/package=eRm. Accessed 15 Mar2012. 37. R Core Team. R: A language and environment for statistical computing R foundation for statistical computing. 2014. http://www.R-project.org. Accessed 15 Mar 2012. 38. Cuerda, M. C., Apezetxea, A., Carrillo, L., Casanueva, F., Cuesta, F., Irles, J. A., et al. (2016). Reliability and responsiveness of NutriQoL® questionnaire. Advances in Therapy, 33(10), 1728.
Journal of Patient-Reported Outcomes
– Springer Journals
Published: May 30, 2018