Received: 31 August 2018; Accepted: 1 October 2018; Published: 7 October 2018

Transcription

1 nutrients Article Develoment Validation of a New Tool to Assess Burden of Dietary Sodium Restriction in Patients with Chronic Heart Failure: The BIRD Questionnaire Etienne Audureau 1, Aziz Guellich 2, Esr Guéry 1, Florence Canouï-Poitrine 1, Véronique Benedyga 3, Hélène Duchossoir 3, Charles Taieb 4 Thibaud Damy 2, * 1 Public Health Deartment, CEiA (EA4393), Créteil, France; etienne.audureau@ah.fr (E.A.); esr.guery@gmail.com (E.G.); florence.canoui-oitrine@ah.fr (F.C.-P.) 2 Heart Failure Unit, Deartment of Cardiology, INSERM U955, ARI, DHU ATVB, AP-HP Henri Mondor Hosital, University Paris Est Créteil (UPEC), Créteil, France; guelliaz@gmail.com 3 Dietetic Unit, AP-HP Henri Mondor Hosital, University Paris Est Créteil (UPEC), Créteil, France; veronique.benedyga@ah.fr (V.B.); helene.duchossoir@ah.fr (H.D.) 4 EMMA, Euroean Market Maintenance Assessment, Vincennes, France; charles.taieb@ah.fr * Corresondence: thibaud.damy@ah.fr; Tel.: +33(0) ; Fax: +33(0) Received: 31 August 2018; Acceted: 1 October 2018; Published: 7 October 2018 Abstract: (1) Background: Burden s are useful in estimating imact of interventions from atients ersectives. This is overlooked in sodium diet/heart failure (HF). The aim of this study is to develo validate a secific tool to assess burden associated with low-sodium diets in HF: Burden In Restricted Diets (BIRD). (2) Methods: Based on literature reorts from atients, 14 cidate items were identified following dietary-related domains: organization, leasure, leisure, social life, vitality, self-rated health. The validation study was conducted rosectively. The questionnaire was refined via item reduction according to inter-item correlations exloratory factor analysis. Internal consistency was determined using Cronbach s alha (Cα) convergent validity by assessing correlations between BIRD health-related quality of life (HRQoL) Minnesota Living with HF questionnaire (MLHF). (3) Results: Of 152 invited atients, 96 (63%) returned questionnaire. The median score was 6.5 (IQR ). The results showed good accetability (non-resonse rates/item from 2.0% to 12.1%), an excellent internal consistency (Cα 0.903) a good convergent validity (rhos 0.37 (hysical), 0.4 (mental), 0.45 (global); all < 0.05). (4) Conclusions: BIRD demonstrates good sychometric roerties is useful to quantify burden associated with sodium restriction. It may hel otimize dietary interventions imrove overall management of atients with HF. Keywords: heart failure; questionnaire; sodium restriction; diet; burden 1. Introduction Chronic heart failure (HF) is a major ublic health social roblem. Its revalence increases with age, reaching aroximately 10% after 70 years [1,2]. This high revalence is exected to evolve furr as a result of continuous aging of oulation, imroved survival of atients with different heart diseases, effective treatments HF. Desite significant rogress in diagnosis treatment of HF, this condition remains a major cause of morbidity mortality, with a 5-year mortality rate of ~50% after first onset of symtoms [3]. Management of HF includes both harmacological non-harmacological interventions. Among non-harmacological interventions, dietary sodium restriction is commonly recommended is endorsed by most international guidelines [4,5]. However, se recommendations are based on limited evidence with Nutrients 2018, 10, 1453; doi: /nu

2 Nutrients 2018, 10, of 10 inconsistent findings across studies [6]. Furrmore, available data suggest that less than half of atients with HF actually follow se recommendations [7], raising questions about comliance underlying reasons non-comliance. Comliance to diet is closely related to atients subjective ercetions exectations. In this regard, chronic heart failure is a comlex condition that requires imortant ersonal investment from atients to manage ir disease. However, alteration in ir long-sting ersonal habits lifestyle may lead to ir ercetion of dietary sodium restriction diet as a negative intervention [8]. Indeed, many atients with HF may face a variety of constraints due to this regimen that cover social, emotional, organizational, economic asects of daily life all limitations that can be catured by notion of burden. The current existing s evaluating quality of life in atients with HF are fundamentally limited assessing this burden, since y include few to no items inquiring about a atients diet, wher y be generic, like SF-36 [9]; disease-secific, like Minnesota living with heart failure questionnaire [10]; or diet-targeted but focused on conditions too secific to be generalizable to or contexts [11,12]. To best of our knowledge, re is no tool available evaluating burden exerienced by atients with heart failure on a low-sodium diet, so re is a ressing need accurate tools to measure this burden. Thus, develoing a dedicated instrument would be beneficial to ascertain concerns of HF atients ir hysicians alike identifying which atients are in need of more sustained monitoring or suort ir diet. The objective of resent study, is ree, to develo validate a new Burden In Restricted Diets (BIRD) to allow an adequate evaluation of burden associated with dietary sodium restriction in heart failure atients. 2. Methods The BIRD questionnaire was designed based on stardized health care, health-related quality of life (HRQoL) questionnaire develoment, validation methodology [13,14]. A multidiscilinary team was set u to ensure scientific clinical relevance of rocess, including cardiologists, dietitian nutritionists, exerts in questionnaire concetion quality-of-life indexes Develoment The first stage included creation of a verbatim reort based on review of literature, qualitative inmation collected during semi-/unstructured interviews with atients with HF to discuss ir comlaints distresses related to diet in HF. Based on this reort inut from multidiscilinary working grou, major identified concerns were meal rearation, leasure, budget, leisure, social life, consequences at work, vitality, self-rated health. An initial set of 14 cidate items were roduced groued according to ir content to constitute initial draft questionnaire (Table 1). Wording resonse modalities were ed in native French-seaking subjects during individual interviews to determine otential issues (ambiguity, misundersting, accetability). A five-oint Likert was used (not at all, just a little, somewhat, quite a lot, very much) with answers numbered from 0 to 4 accordingly. To limit missing data, modality Not alicable was also included items 4, 5, 7, 9, 12 related to meals away from home, grocery shoing, close relatives rearing meals, leisure activities, rofessional activity, resectively. A global score was calculated by summing individual item scores, where a higher total score reresents a higher symtom burden.

3 Nutrients 2018, 10, of 10 Table 1. Initial cidate items Burden In Restricted Diets (BIRD) questionnaire. Item 1 Item 2 Item 3 Item 4 Item 5 Item 6 Item 7 Item 8 Item 9 Item 10 Item 11 Item 12 Item 13 Item 14 On Account of My Diet, I am Not Living AS I Would Like, Because I have to limit my consumtion of my favorite dishes... my aetite is decreased... every meal is difficult me... having a meal away from home is comlicated... grocery shoing is comlicated... it results in additional exenses... I have imression of being a bor or a burden to those rearing my meals... it makes relationshis or activities with friends or family difficult... it makes my leisure activities difficult (favorite astimes, sorts)... it revents me from travelling, going on vacation... it makes me feel tired, weary or I lack energy... it is difficult to manage in my worklace/rofessional activity... it deresses me... it aggravates my health 2.2. Validation Study A validation study was conducted to assess sychometric roerties of. A cross-sectional survey was conducted in more than 100 adult atients with medically diagnosed HF using draft questionnaire. The questionnaire was reed in first 10 atients to evaluate comrehensibility, changes were made based on ir comments. To be eligible, atients had to be able to read, be able to underst seak French language, lack any cognitive imairment. Patients were rosectively enrolled in Cardiology Deartment of Henri-Mondor Hosital, Creteil, France. Given nature of study, without any change to atient care, need written inmed consent was waived under French regulations; all atients received an inmation sheet with contact details, authorization to use data could be withdrawn at any time. The study rotocol conms to ethical guidelines of 1975 Declaration of Helsinki Statistical Analysis Psychometric analysis included assessment of item characteristics, construct validity, internal consistency, known-grous validity, convergent validity. Descritive analyses were ermed to study distribution of individual items global score, to inm on accetability (% missing values), to identify otential ceiling /or floor effects when a majority of item resonses were distributed at eir end of. Searman s rank correlation coefficients (r s ) were comuted to examine homogeneity of (item-total correlation, with a minimal accetable level of r s 0.3) to identify wher highly correlated items should be omitted redundancy (inter-item correlations, r s > 0.8). A correlation network lot was built from those results to grahically illustrate relationshis. A hot-deck multile imutation was ermed to imute missing data subsequent analyses. Construct validity (factor structure) of instrument was assessed through exloratory factor analysis (rincial factor method) to examine underlying constructs characterize dimensionality. Data were first examined with Bartlett s of shericity ( < ) Kaiser Meyer Olkin measure of samling adequacy (0.842), indicating that our samle was suitable conducting exloratory factor analyses (EFAs). As recommended [15,16], we used a combination of various strategies to determine otimal number of factors to retain, including consideration of roortion of variance exlained factor solution retained, using Horn s arallel analysis [17] based on 95th ercentile estimate comuting Velicer s minimum average artial (MAP) criterion [18]. Items were considered deletion if ir factor loadings were <0.4, or/ if ir communalities were <0.3 (uniqueness > 0.7).

4 Nutrients 2018, 10, of 10 Internal consistency reliability (homogeneity of items) was assessed by calculating Cronbach s alha [19]. A coefficient score of >0.8 indicates good internal consistency >0.9 is considered an excellent one. Known-grous validity was investigated by studying wher burden score would differentiate between adjacent severity subgrous. For that urose, differences in burden score were assessed according to New York Heart Association (NYHA) class intensity of rescribed low-sodium diet (normal: >6 g er day; moderately restricted: 3 6 g; highly restricted: <3 g). Scores were comared across grous using nonarametric Kruskal Wallis global comarisons Jonckheere Terstra s. Convergent validity was studied by assessing correlations between global burden score HRQoL Minnesota Living with Heart Failure questionnaire (MLHF), a validated 21-item questionnaire assessing HRQoL in atients with HF [10]. Searman s rank correlation coefficients were comuted between burden score each MLHF domain score, namely, hysical (8 items, score 0 40), emotional (5 items, 0 25), or items (8 items, 0 40). Convergent validity measures how closely related BIRD score is to or s measuring similar but not strictly equivalent constructs. Coefficients less than 0.3 were considered as weak, from 0.3 to 0.5 as moderate, those above 0.5 as strong. Statistical analyses were ermed using Stata v14.1 software (StataCor, College Station, TX, USA) descritive factor analyses R software (R Foundation Statistical Comuting, Vienna, Austria) using hot.deck, aran, sych ackages hot-deck imutation, Horn s arallel analysis, Velicer minimum average artial correlation number of rincial comonents, resectively Translation Cross-Cultural Validation Since original BIRD questionnaire was develoed in French, a well-established methodology was alied to generate an English version. This included ward backward translation while accounting cross-cultural validation [20]. Linguistic cross-cultural adatation was carried out by a secialized institution (Lionbridge, Dublin, Irel) following a nine-ste rocess English language (Sulemental Table S1). 3. Results Of 152 atients invited, 96 (63%) comleted returned questionnaire. Comarison between non-articiating articiating atients age, gender, NYHA class did not reveal any significant differences (data not shown). Main characteristics of samle, including sociodemograhics, clinical biological features, HF-related diet treatment, Minnesota Living with Heart Failure (MLHF) scores by domains are shown in Sulemental Table S2. The samle included 72% men with a mean age of 62 ± 12 years, of whom 26% had a highly restricted low-sodium diet (<3 g/day) Item Statistics Of 96 articiating atients, 69 (72%) answered all questionnaire items. Among those who did not, missing data ranged from 1 (17%) to 5 (1%) of items. Item statistics related to data quality, distribution, mean scores are shown in Table 2. Mean scores by item varied between 0.6 (items 3, 12, 14) 1.7 (item 1), with floor effects reached in items 2, 3, 5, 10, 13, 14 (>50% of resonses at lowest end of ). There was no evidence of ceiling effects. Missing data ranged from 1% to 10% (item 12, related to rofessional activity), while non-alicable answers varied between 4% 27% (item 12) four items including this answer otion. Searman correlation coefficients between items total score were all above 0.6, with excetion of items 1 (r s 0.47), 12 (r s 0.53), 2 (r s 0.55). Lower inter-item correlations were aarent

5 Nutrients 2018, 10, of 10 items 1, 2,, to a lesser extent, 12 (Figure 1). No inter-item correlation exceeded 0.8, thus ruling out otential redundancy at this stage. Nutrients 2018, 10, x FOR PEER REVIEW 5 of 10 Table 2. Item distribution item-total score correlations. Table 2. Item distribution item-total score correlations. Item Distribution Item-Total Item Distribution Mean ± Missing Score Item-TotalNot at Just a Quite a Very SD Missing Mean ± SD Score Somewhat Not Data Correlation Total All Little Lot Much Data Correlation Not at All 2 Just a Somewhat Quite a Alicable Very Not 0 Total Little 1 2 Lot 3 Much 4 Alicable Item (±1.1) (17%) 27 (29%) 27 (29%) 18 (19%) 6 (6%) 0 (0%) 2 (2%) Item 2 Item 1.0 (±1.1) (±1.1) (43%) (26%) 16 (17%) 19 (20%) 27 (29%) 7 (8%) 27 (29%) 3 (3%) 18 (19%) 06 (0%) (6%) 03 (0%) (3%) 2 (2%) Item 3 Item 0.6 (±1.0) (±1.1) (68%) (16%) 40 (43%) 8 (8%) 24 (26%) 4 (4%) 19 (20%) 3 (3%) 7 (8%) 0 3 (0%) (3%) 0 1 (0%) (1%) 3 (3%) Item (±1.0) (68%) 15 (16%) 8 (8%) 4 (4%) 3 (3%) 0 (0%) 1 (1%) Item (±1.2) (42%) 19 (20%) 15 (16%) 13 (14%) 4 (4%) 4 (4%) 1 (1%) Item (±1.2) (42%) 19 (20%) 15 (16%) 13 (14%) 4 (4%) 4 (4%) 1 (1%) Item (±1.1) (51%) 19 (20%) 7 (7%) 12 (13%) 2 (2%) 6 (6%) 2 (2%) Item (±1.1) (51%) 19 (20%) 7 (7%) 12 (13%) 2 (2%) 6 (6%) 2 (2%) Item (±1.1) (56%) 21 (24%) 8 (9%) 7 (8%) 3 (3%) 0 (0%) 7 (7%) Item (±1.1) (56%) 21 (24%) 8 (9%) 7 (8%) 3 (3%) 0 (0%) 7 (7%) Item 7 Item 0.8 (±1.1) (±1.1) (46%) (20%) 44 (46%) 11 (12%) 19 (20%) 11 (12%) 11 (12%) 1 (1%) 11 (12%) 91 (9%) (1%) 91 (9%) (1%) 1 (1%) Item 8 Item 0.7 (±1.0) (±1.0) (58%) (21%) 52 (58%) 13 (14%) 19 (21%) 3 (3%) 13 (14%) 3 (3%) 3 (3%) 03 (0%) (3%) 06 (0%) (6%) 6 (6%) Item 9 Item 1.1 (±1.5) (±1.5) (47%) (15%) 43 (47%) 8 (9%) 14 (15%) 8 (9%) 8 (9%) 12 (13%) 8 (9%) 612 (7%) (13%) 65 (7%) (5%) 5 (5%) Item 10 Item 0.8 (±1.2) (±1.2) (58%) (18%) 52 (58%) 7 (8%) 16 (18%) 11 (12%) 7 (8%) 4 (4%) 11 (12%) 04 (0%) (4%) 06 (0%) (6%) 6 (6%) Item 11 Item 1.3 (±1.3) (±1.3) (37%) (29%) 35 (37%) 12 (13%) 27 (29%) 12 (13%) 12 (13%) 8 (9%) 12 (13%) 08 (0%) (9%) 02 (0%) (2%) 2 (2%) Item 12 Item 0.6 (±1.1) (±1.1) (44%) 86 9 (10%) 38 (44%) 7 (8%) 9 (10%) 6 (7%) 7 (8%) 3 (3%) 6 (7%) 233 (27%) (3%) (27%) (10%) 10 (10%) Item 13 Item 0.7 (±1.2) (±1.2) (65%) (13%) 61 (65%) 12 (13%) 12 (13%) 3 (3%) 12 (13%) 6 (6%) 3 (3%) 06 (0%) (6%) 02 (0%) (2%) 2 (2%) Item 14 Item 0.6 (±1.0) (±1.0) (66%) (18%) 61 (66%) 7 (8%) 17 (18%) 5 (5%) 7 (8%) 3 (3%) 5 (5%) 03 (0%) (3%) 03 (0%) (3%) 3 (3%) SD: stard SD: stard deviation. deviation. Figure 1. Correlation between items: (A) Searman s rank correlation coefficients matrix Figure (B) 1. correlation Correlation network. between The items: matrix (A) Searman s contains rank Searman s correlation rankcoefficients correlationmatrix coefficients (B) between correlation 14 items network. of cidate The matrix questionnaire. contains Searman s Colors indicate rank correlation direction coefficients strength between of 14 correlation, items with of ositive cidate correlations questionnaire. being Colors dislayed indicate as green direction tones negative strength of ones as correlation, red tones. Bolded with ositive results indicate correlations statistical being significance dislayed as at green < tones 0.05 level. negative The correlation ones as network red tones. is Bolded constructed from results all indicate airwise statistical correlations significance between at items < 0.05 in (A). level. Items The correlation are reresented network byis nodes constructed arefrom connected by all airwise edges. correlations Red green between lines items reresent in (A). negative Items are reresented ositive correlations, by nodes resectively. are connected Lineby width color edges. saturation Red green is roortional lines reresent to negative strength of ositive correlation. correlations, resectively. Line width color saturation is roortional to strength of correlation Factor Structure 3.2. Factor Structure A one-factor solution exlained 79.8% of total variance, with both MAP criterion arallel A one-factor analysis solution suorting exlained this79.8% solution. of Thetotal unadjusted variance, with adjusted both eigenvalues MAP criterion (arallel analysis) arallel analysis first three suorting factors were this solution. 5.93/0.78/0.53 The unadjusted 4.87/ 0.04/ 0.12, adjusted resectively. eigenvalues Items (arallel 1 2 were analysis) left out because first three of ir factors insufficient were 5.93/0.78/0.53 loading (<0.4) /or 4.87/ 0.04/ 0.12, communality resectively. (<0.3) onitems first 1 axis. Factor 2 were analysis left out rerun because after of exclusion ir insufficient of seloading two items (<0.4) confirmed /or communality unidimensionality, (<0.3) first axis. suorted by Factor analysis MAP criterion, rerun after arallel exclusion analysis, of se two large items roortion confirmed of total variance unidimensionality, (88.0%) being exlained suorted by by first factor. MAP These criterion, findings arallel indicate analysis, construct validity large roortion thus validate of total variance calculation (88.0%) of a single being BIRD exlained score, by which first brings factor. toger These all findings items. indicate A global construct 12-item validity burden score thus validate was n comuted calculation of a single BIRD score, which brings toger all items. A global 12-item burden score was n comuted (maximum score: 48; mean: 9.8 ± 10.0; median: 6.5 (IQR 2 14) (Sulemental Table S3). The comlete distribution is shown in Sulemental Figure S1.

6 Nutrients 2018, 10, of 10 (maximum score: 48; mean: 9.8 ± 10.0; median: 6.5 (IQR 2 14) (Sulemental Table S3). The comlete distribution is shown in Sulemental Figure S1. Nutrients 2018, 10, x FOR PEER REVIEW 6 of Internal Internal Consistency, Consistency, Convergent Convergent Known-Grous Known-Grous Validity Validity The The develoed develoed showed showed excellent excellent internal internal consistency consistency (Cronbach (Cronbach alha alha 0.912) ). Results Results known-grous known-grous validity validity are shown are shown in Figure in Figure 2. As 2. exected, As exected, atients atients with an with increased an increased NYHA NYHA class scored class scored significantly significantly higher higher on on global global score (global score (global 0.055; 0.055; 0.003) ). This suorts This suorts s s ability ability to discriminate to discriminate according according to clinical to clinical disease disease severity. severity. A A statistical statistical higher higher scores scores was was also also found found in in atients atients with with most most restricted restricted diet diet (global (global 0.30; 0.30; 0.06; 0.06; (<3 (<3 g) g) vs. vs. (3 6 (3 6 g g or or >6 >6 g) g) 0.06), 0.06), suorting suorting ability ability of of to to identify identify those those who who suffer suffer heaviest heaviest burden burden of of a a low-sodium low-sodium diet. diet. Table Table 3 3 exhibits exhibits convergent convergent validity validity results. results. Significant Significant ositive ositive correlations correlations were were found found between between global global burden burden score score scores scores from from MLHF MLHF domains domains global global score: rs score: r s (hysical (hysical domain), domain), (mental (mental domain), domain), (global (global score); score); all all < < Figure 2. Global 12-item BIRD score according to (A) rescribed low-sodium diet (B) NYHA class. Figure Error margins 2. Global indicate 12-item stard BIRD score erroraccording of mean to (SEM). (A) rescribed NYHA: New low-sodium York Heart diet Association. (B) NYHA class. Error margins indicate stard error of mean (SEM). NYHA: New York Heart Association. Table 3. Results convergent validity: Searman correlation coefficients between global burden score Minnesota Living with Heart Failure questionnaire domains. Table 3. Results convergent validity: Searman correlation coefficients between global burden score Minnesota Living with Heart Failure questionnaire Minnesota Living domains. with Heart Failure Physical Minnesota Emotional Living with Heart Or Failure Global Domain Domain Items Score Physical Emotional Or Global Minnesota Living with Heart Failure Domain Domain Items Score Physical Domain Minnesota Living Emotional with Heart Domain Failure 0.80 Physical Domain Or Items Emotional Domain Global Score Or Global Items Burden Score Global Score All results are statistically 0.94 significant at 0.92 < 0.05 level Global Burden Score Discussion All results are statistically significant at < 0.05 level. 4. Discussion Health-related quality of life (HRQoL) burden s have roved to be useful in estimating imact of diseases treatments from atient s ersective on ir hysical, mental, social Health-related health status. quality Several of life generic (HRQoL) or disease-secific burden s atient-reorted have roved to outcomes be useful are in available estimating to assess imact of self-rated diseases health treatments status of atients from with atient s heart ersective failure. However, on ir none hysical, cature mental, secific social domains health related status. to Several sodium generic restriction or disease-secific often rescribed atient-reorted in HF. To best outcomes of our knowledge, are available BIRD to assess self-rated health status of atients with heart failure. However, none cature secific domains related to sodium restriction often rescribed in HF. To best of our knowledge, BIRD is first validated questionnaire assessing secifically burden of a low-sodium diet in atients with HF. BIRD was found to be sychometrically robust, with excellent internal consistency good

7 Nutrients 2018, 10, of 10 is first validated questionnaire assessing secifically burden of a low-sodium diet in atients with HF. BIRD was found to be sychometrically robust, with excellent internal consistency good item-, convergent validity, construct validity. BIRD also correlated significantly with comonents of MHF, confirming its concurrent validity Dietary Sodium Restriction Quality of Life in HF Heart failure is associated with activation of neurohormonal system, leading to sodium water retention. All national international guidelines currently recommend that HF management should include dietary sodium restriction [21]. Based on athohysiological exerimental studies, it has been assumed that a low-sodium diet would relieve symtoms imrove quality of life outcomes in HF atients. However, observational clinical studies have shown mixed results [6]. Although adverse neurohormonal activation related to sodium restriction in HF remains a concern, benefit-imact balance atients HRQoL with regard to sodium restriction in this context is unknown, with a substantial ga in ability to evaluate imact of this regimen. The low-sodium diet effects on quality of life have only so far been considered a otential favorable outcome, desite fact that burden related to restricted diets has extensively been described characterized in several or diseases. It has thus been reorted in children to a lesser extent in adults chronic diseases, including celiac disease, irritable bowel syndrome, diabetes [22 27]. Data regarding sodium restriction in HF are missing, essentially because of lack of a validated accurate measurement tool Burden in Chronic Diseases HF The notion of burden has been introduced by World Health Organization to hel quantify health of a oulation to determine riorities ublic health interventions [28]. The burden of disease concet has been extended to now distinguish between (i) overall burden, by measuring economic imact on society, (ii) individual burden, relating to atients ir families. The latter assesses disability in a broad sense, i.e., sychological, social, economic, hysical, also accounting everyday-life organization use of medical resources. The concet of burden has increasingly been used in medical research since it rovides useful insight evaluating care of chronic diseases. In context of HF, such burden may relate to a variety of accumulating constraints [8]. Patients are required to follow comlex medication regimens to imlement imortant lifestyle changes, such as sodium restriction. Combined with direct consequences of disease, treatment burden may contribute to increasing atient s stress decreasing ir adherence to treatment Imlications Practice BIRD is a quick easy-to-use tool allowing a single diet-related burden score to be calculated. It has advantage of covering multile relevant asects of burden that can result from a low-sodium diet in a atient s daily life. Its widesread administration will be facilitated by atness of questionnaire, as demonstrated by satisfying sychometric roerties of instrument in resent study, i.e., understability, reliability, validity, as well as by its easy-to-use arsimonious structure, with only 12 items to collect. As such, it may serve as a strategic tool screening atients suffering highest burden from ir diet who might benefit from secific interventions, such as sychological social hel, close follow-u, etc. It may also foster negotiation cooeration between atients, hysicians, health authorities, agro-alimentary industries by roviding a common tool quantifying diet-related constraints. Finally, it is noteworthy that items constituting BIRD questionnaire relate to organizational, social, general health status asects which would most likely also be affected in or conditions requiring restricted diets. Thus, it would be of great interest to study validity clinical utility of instrument in or settings

8 Nutrients 2018, 10, of 10 or conditions, e.g., diabetes celiac disease,, ultimately, to comare resulting burden scores using this common reference Limitations Limitations associated with this study include monocentric recruitment relatively limited samle size that restrained ossibility of subgrou analyses. Larger confirmatory studies using BIRD questionnaire are warranted to furr exlore otential effects of age, disease onset, culinary culture habits, comorbidities, coexistence of or diet restrictions, etc. on burden of sodium restriction in HF. Second, 37% of invited atients did not comlete return questionnaire. Yet, our resonse rate aears consistent with, even somewhat higher than, those generally observed in surveys where articiants are required to send questionnaires by ostal mail, as in resent study. Relatedly, a selection bias resulting from such non-articiation cannot be ruled out, though likely of limited magnitude, as re was no significant difference in demograhics NYHA classes between articiating non-articiating atients. Finally, due to cross-sectional design of study, we also lacked ability to investigate redictive value of tool future treatment comliance /or subsequent cardiologic events. Furr research efts will include conduct of rosective studies to address se elements. 5. Conclusions In conclusion, re is a need hysicians to recognize individual burden of atients with HF under sodium restriction. Quantification of this burden by BIRD questionnaire may hel otimize dietary interventions imrove overall management of atients with HF. Sulementary Materials: The following are available online at htt:// s1, Figure S1: Global 12-item BIRD score distribution in validation study oulation, Table S1: Stes linguistic cross-cultural adatation, Table S2: General characteristics of study oulation, Table S3: One-factor solution of BIRD questionnaire: results initial 14-items after deletion of items 1 2. Author Contributions: Concetualization, E.A., C.T., A.G., T.D.; Methodology, E.A., F.C.-P., E.G.; Validation, V.B., H.D. T.D.; Formal Analysis, E.A., E.G., F.C.-P.; Writing Original Draft Prearation, A.G. E.A. Writing Review & Editing, E.A., A.G., E.G., F.C.-P., V.B., H.D., C.T., T.D.; Suervision, T.D. Funding: This research received no external funding. Conflicts of Interest: The authors declare no conflict of interest. Abbreviations HF BIRD HRQoL EFA MAP NYHA MLHF chronic heart failure Burden In Restricted Diet health-related quality of life exloratory factor analyses minimum average artial New York Heart Association Minnesota Living with Heart Failure questionnaire References 1. van Riet, E.E.; Hoes, A.W.; Wagenaar, K.P.; Limburg, A.; Lman, M.A.; Rutten, F.H. Eidemiology of heart failure: The revalence of heart failure ventricular dysfunction in older adults over time. A systematic review. Eur. J. Heart Fail. 2016, 18, [CrossRef] [PubMed] 2. Mosterd, A.; Hoes, A.W. Clinical eidemiology of heart failure. Heart 2007, 93, [CrossRef] [PubMed] 3. Roger, V.L.; Go, A.S.; Lloyd-Jones, D.M.; Adams, R.J.; Berry, J.D.; Brown, T.M.; Carnethon, M.R.; Dai, S.; de Simone, G.; Ford, E.S.; et al. Heart disease stroke statistics 2011 udate: A reort from american heart association. Circulation 2011, 123, e18 e209. [CrossRef] [PubMed]

9 Nutrients 2018, 10, of Yancy, C.W.; Jessu, M.; Bozkurt, B.; Butler, J.; Casey, D.E., Jr.; Drazner, M.H.; Fonarow, G.C.; Geraci, S.A.; Horwich, T.; Januzzi, J.L.; et al ACCF/AHA guideline management of heart failure: A reort of american college of cardiology foundation/american heart association task ce on ractice guidelines. Circulation 2013, 128, e [PubMed] 5. McMurray, J.J.; Adamooulos, S.; Anker, S.D.; Auricchio, A.; Bohm, M.; Dickstein, K.; Falk, V.; Filiatos, G.; Fonseca, C.; Gomez-Sanchez, M.A.; et al. ESC guidelines diagnosis treatment of acute chronic heart failure 2012: The task ce diagnosis treatment of acute chronic heart failure 2012 of euroean society of cardiology. Develoed in collaboration with heart failure association (HFA) of ESC. Eur. Heart J. 2012, 33, [PubMed] 6. Guta, D.; Georgiooulou, V.V.; Kalogerooulos, A.P.; Dunbar, S.B.; Reilly, C.M.; Ss, J.M.; Fonarow, G.C.; Jessu, M.; Gheorghiade, M.; Yancy, C.; et al. Dietary sodium intake in heart failure. Circulation 2012, 126, [CrossRef] [PubMed] 7. Chung, M.L.; Lennie, T.A.; Mudd-Martin, G.; Moser, D.K. Adherence to a low-sodium diet in atients with heart failure is best when family members also follow diet: A multicenter observational study. J. Cardiovasc. Nurs. 2015, 30, [CrossRef] [PubMed] 8. Gallacher, K.; May, C.R.; Montori, V.M.; Mair, F.S. Understing atients exeriences of treatment burden in chronic heart failure using normalization rocess ory. Ann. Fam. Med. 2011, 9, [CrossRef] [PubMed] 9. McHorney, C.A.; Ware, J.E., Jr.; Raczek, A.E. The MOS 36-item short-m health survey (SF-36): Ii. Psychometric clinical s of validity in measuring hysical mental health constructs. Med. Care 1993, 31, [CrossRef] [PubMed] 10. Rector, T.S.; Cohn, J.N. Assessment of atient outcome with minnesota living with heart failure questionnaire: Reliability validity during a romized, double-blind, lacebo-controlled trial of imobendan. Pimobendan multicenter research grou. Am. Heart J. 1992, 124, [CrossRef] 11. Honda, M.; Wakita, T.; Onishi, Y.; Nunobe, S.; Miura, A.; Nishigori, T.; Kusanagi, H.; Yamamoto, T.; Boddy, A.; Fukuhara, S. Develoment validation of a disease-secific instrument to measure diet-targeted quality of life ostoerative atients with esohagogastric cancer. Ann. Surg. Oncol. 2015, 22 (Sul. 3), [CrossRef] 12. Sato, E.; Suzukamo, Y.; Miyashita, M.; Kazuma, K. Develoment of a diabetes diet-related quality-of-life. Diabetes Care 2004, 27, [CrossRef] [PubMed] 13. Marquis, P. Déveloement et validation d un questionnaire: État de l art et niveau d évidence requis. In Qualité de vie liée à l état de santé: Critère d évaluation; Chassagny, O., Caulin, C., Eds.; Sringer-Verlag: Paris, France, 2002; Leidy, N.K.; Revicki, D.A.; Geneste, B. Recommendations evaluating validity of quality of life claims labeling romotion. Value Health ISPOR 1999, 2, [CrossRef] [PubMed] 15. Fabrigar, L.R.; Wegener, D.T.; MacCallum, R.C.; Strahan, E.J. Evaluating use of exloratory factor analysis in sychological research. Psychol. Methods 1999, 4, [CrossRef] 16. Izquierdo, I.; Olea, J.; Abad, F.J. Exloratory factor analysis in validation studies: Uses recommendations. Psicoma 2014, 26, [PubMed] 17. Horn, J.L. A rationale number of factors in factor analysis. Psychometrika 1965, 30, [CrossRef] [PubMed] 18. Velicer, W.F. Determining number of comonents from matrix of artial correlations. Psychometrika 1976, 41, [CrossRef] 19. Cronbach, L. Coefficient alha internal structure of s. Psychometrica 1951, 16, [CrossRef] 20. Wild, D.; Grove, A.; Martin, M.; Eremenco, S.; McElroy, S.; Verjee-Lorenz, A.; Erikson, P.; Translation, I.T.F.f.; Cultural, A. Princiles of good ractice translation cultural adatation rocess atient-reorted outcomes (PRO) measures: Reort of isor task ce translation cultural adatation. Value Health ISPOR 2005, 8, [CrossRef] [PubMed] 21. Beich, K.R.; Yancy, C. The heart failure sodium restriction controversy: Challenging conventional ractice. Nutr. Clin. Pract. 2008, 23, [CrossRef] [PubMed] 22. Simsek, S.; Baysoy, G.; Gencoglan, S.; Uluca, U. Effects of gluten-free diet on quality of life deression in children with celiac disease. J. Pediatr. Gastroenterol. Nutr. 2015, 61, [CrossRef] [PubMed]

10 Nutrients 2018, 10, of Santolaria-Piedrafita, S.; Montoro-Huguet, M. Celiac disease, gluten-free diet health-related quality of life. Rev. Es. Enferm. Dig. 2015, 107, [PubMed] 24. Samasca, G.; Sur, G.; Luan, I.; Deleanu, D. Gluten-free diet quality of life in celiac disease. Gastroenterol. Heatol. Bed Bench 2014, 7, [PubMed] 25. Ostgaard, H.; Hausken, T.; Gundersen, D.; El-Salhy, M. Diet effects of diet management on quality of life symtoms in atients with irritable bowel syndrome. Mol. Med. Re. 2012, 5, [PubMed] 26. Cotugno, G.; Nicolo, R.; Caelletti, S.; Goffredo, B.M.; Dionisi Vici, C.; Di Ciommo, V. Adherence to diet quality of life in atients with henylketonuria. Acta Paediatr. 2011, 100, [PubMed] 27. Barratt, S.M.; Leeds, J.S.; Sers, D.S. Quality of life in coeliac disease is determined by erceived degree of difficulty adhering to a gluten-free diet, not level of dietary adherence ultimately achieved. J. Gastroinin. Liver Dis. 2011, 20, [PubMed] 28. WHO. About Global Burden of Disease (GBD) Project. Available online: htt:// healthinfo/global_burden_disease/en/ (accessed on 5 October 2018) by authors. Licensee MDPI, Basel, Switzerl. This article is an oen access article distributed under terms conditions of Creative Commons Attribution (CC BY) license (htt://creativecommons.org/licenses/by/4.0/).