PLOS ONE RESEARCH ARTICLE Functional social support: A systematic review and standardized comparison of different versions of the DUFSS questionnaire using the EMPRO tool Cristina M. Lozano-Hernández ID1,2,3,4*, Yolanda Pardo Cladellas5,6,7, Mario Gil Conesa8, Olatz Garin5,9, Montserrat Ferrer Forés5,6,9, Isabel del Cura-González2,3,10 1 Interuniversity Doctoral Program in Epidemiology and Public Health, Rey Juan Carlos University, Alcorcon a1111111111 (Madrid), Spain, 2 Research Unit, Primary Health Care Management Madrid, Madrid, Spain, 3 Research a1111111111 Network on Chronicity, Primary Care and Health Promotion -RICAPPS- (RICORS), Spain, 4 Biosanitary a1111111111 Research and Innovation Foundation of Primary Care (FIIBAP), Spain, 5 Networking Group of the Centre for a1111111111 Biomedical Research, Epidemiology and Public Health (CIBERESP), Instituto de Salud Carlos III, Madrid, a1111111111 Spain, 6 Health Services Research Unit, Institut Mar d’Investigacion Mèdiques (IMIM-Hospital del Mar), Barcelona, Spain, 7 Autonomous University of Barcelona, Spain, 8 Preventive Medicine and Public Health Physician, Castellón General Hospital, Castellón, Valencian Community, Castellón, Spain, 9 Pompeu Fabra University, Barcelona, Spain, 10 Department of Medical Specialties and Public Health. Faculty of Health Sciences Rey Juan Carlos University, Madrid, Spain OPEN ACCESS * cristinamaria.lozano@salud.madrid.org Citation: Lozano-Hernández CM, Cladellas YP, Gil Conesa M, Garin O, Forés MF, del Cura-González I (2023) Functional social support: A systematic Abstract review and standardized comparison of different versions of the DUFSS questionnaire using the EMPRO tool. PLoS ONE 18(9): e0291635. https:// Background doi.org/10.1371/journal.pone.0291635 Functional social support is one of the most established predictors of health, and the Duke- Editor: Victor Manuel Mendoza-Nuñez, National UNC Functional Social Support Questionnaire (DUFSS) is one of the most commonly used Autonomous University of Mexico, MEXICO instruments to measure this parameter. The objective of this study is to systematically Received: October 20, 2022 review the available evidence on the psychometric and administration characteristics of the different versions of the DUFSS and perform a standardized assessment though to a specif- Accepted: September 3, 2023 ically designed tool. Published: September 15, 2023 Peer Review History: PLOS recognizes the Methods benefits of transparency in the peer review process; therefore, we enable the publication of A systematic review was performed in the PubMed/MEDLINE, SCOPUS, WOS and all of the content of peer review and author SCIELO databases. All articles that contained information on the development process of responses alongside final, published articles. The the instrument, the psychometric properties and aspects related to its administration were editorial history of this article is available here: https://doi.org/10.1371/journal.pone.0291635 included, without restrictions based on publication date, language, or the version of the questionnaire that was studied. The selection and extraction procedure were carried out by Copyright: © 2023 Lozano-Hernández et al. This is an open access article distributed under the terms two researchers. The articles finally included were peer-reviewed through a standardised of the Creative Commons Attribution License, assessment using the Evaluating the Measurement of Patient-Reported Outcomes which permits unrestricted use, distribution, and (EMPRO) tool. PROSPERO registration number: CRD42022342977. reproduction in any medium, provided the original author and source are credited. Results Data Availability Statement: All relevant data are within the paper and its Supporting Information A total of 54 articles were identified. After eliminating duplicates and screening articles files. The dataset corresponding to the evaluations based on the selection criteria, 15 studies that examined the DUFSS questionnaire resulting PLOS ONE | https://doi.org/10.1371/journal.pone.0291635 September 15, 2023 1 / 17 PLOS ONE Use of the EMPRO tool in different versions of DUFSS of the different versions of the DUFSS in 4 different versions: 3 articles obtained the 8-item version; 11 the 11-item version; and a questionnaire using the EMPRO tool has been single article obtained two versions, the 14-item version and the 5-item version. At least published in Zenodo. The publication is under DOI: https://doi.org/10.5281/zenodo.8211219. 60% of them did so in a young adult population, predominantly female and with a medium- low socio-economic level or with characteristics of social vulnerability. The EMPRO evalua- Funding: The main author CMLH received a grant for the translation of this paper from the tion showed that the 11-item version (54.01 total score) was the only one that had been stud- Foundation for Biosanitary Research and ied on all recommended attributes and had higher total scores than the other versions: 8 Innovation in Primary Care of the Community of items (36.31 total score), 14 items (27.48 total score) and 5 items (23.81 total score). This Madrid (FIIBAP). This study was funded by the difference appears in all attributes studied, with the highest scores in "reliability (internal con- Fondo de Investigaciones Sanitarias ISCIII (Grant Numbers PI15/00276, PI15/00572, PI15/00996), sistency)" and "validity". REDISSEC (Project Numbers RD16/0001/0006, RD16/0001/ 0005 and RD16/0001/0004), and the Conclusions European Regional Development Fund (“A way to build Europe”). Funders had no role in study design Of the 4 versions identified in the DUFSS questionnaire, the 11-item version was found to or in the decision to submit the report for be optimal based on the EMPRO standardized tool. Although, a priori, we could prioritise its publication. The publication of study results was use in epidemiological studies over the other versions, it should be noted that this version not contingent on the sponsor’s approval or censorship of the manuscript. should also be used with caution because there are attributes that have not been studied. Competing interests: The authors have declared that no competing interests exist. Background Research on the relationship between social support and the state of health peaked in the 1970s [1]. Since then, several authors have shown the positive effects of social support on health out- comes [2–4]. Cobb and Cassell [5, 6] argue that the main protective role of social support lies in its moderating effect on life stress. Cohen and Gallant [2, 7] point out that social support impacts on behaviour and the way people manage their health problems and self-care, deter- mining their lifestyles. There are several definitions of social support across different disciplines [8, 9]. Sociologist P. A. Thoits (1982) defines it as the degree to which a person’s basic social needs are satisfied through interaction with others, where basic needs are understood as affiliation, affection, belonging, identity, security and approval [10]. Therefore, there is no consensus regarding the definition of social support [11–13]. Accord- ing to the conceptual model of Barrón A. [14], social support can be understood from three perspectives: structural, contextual and functional. Structural social support studies the so- called support network, which includes all the contacts of the individual; the size, frequency and density of networks are measured, but the availability of this resource is not measured since the support network is not always a source of available social support [15]. Contextual social support addresses the environment and the circumstances that favour or hinder social support [14].. The functional perspective focuses on the subjective assessment that the person makes regarding their own social support based on its availability and accessibility [16]. The meta-analyses performed by Uchino et al. and Dimatteo et al. show how the study of social support in relation to health is carried out from one or several perspectives according to the authors [17, 18]. Functional social support has been described as a stronger predictor of health than the rest of the perspectives of social support [19] Functional social support has three aspects: 1) emo- tional support, focused on the closest and most intimate social relationships, sources of care and empathy and composed of two dimensions (the confidential dimension and the affective dimension); 2) instrumental support, also called tangible or material support, because it refers PLOS ONE | https://doi.org/10.1371/journal.pone.0291635 September 15, 2023 2 / 17 PLOS ONE Use of the EMPRO tool in different versions of DUFSS to practical help in tasks, travel or financial aid; and 3) informational support, referring to sup- port in decision-making or useful advice. In 1983, Broadhead et al. [20] defined the characteristics of the association between func- tional social support and health. Based on this work and on the strategic recommendations issued by House and Kahn in 1985 regarding the study of social support, [11] Broadhead developed the Duke-UNC Fuctional Social Support (DUFSS) questionnaire in 1988 [16]. This questionnaire was validated in the USA in the context of primary care and on a general popu- lation of mostly young adult women with a medium-high socioeconomic status. The original version[16] comprised 14-items across four dimensions: “amount of support”, “confidential support”, “affective support” and “instrumental support”. After examining the test-retest reliability, an 11-item version was obtained. This version included the confidential dimension (CF) and the affective dimension (AF) and measured emotional support but omit- ted measures of the amount of support and instrumental support. The subsequent factor analy- sis revealed that 3 of the 11 items did not correspond to the resulting dimensions in the other versions, indicating that further research was necessary and thus suggesting that the 8-item version be applied. The final 8-item version included a CF composed of items 3, 4, 5, 6 and 7; and an AF composed of items 1, 2 and 8. Later (1991) De la Revilla et al. validated the Broadhead questionnaire, by taking its 11-item version, in Spain in the context of primary care and on a general population of mostly young adult women with a low socio-economic status. As a result, they obtained a version with the same number of items (11-items) with a different distribution in its dimensions: CF composed of items 7,8,6,4,1 and 10; and AF composed of items 11,9,2,3 and 5. This questionnaire has been widely used to study functional social support in National Health Surveys, as is the case in Spain [21], and in European surveys, such as the European Health Interview Survey [22]. Therefore, different versions of the DUFSS questionnaire exist, all of which have been vali- dated across very different populations. Thus, it difficult to choose the most appropriate version. The concept of social support stands out for its subjective character and nature as Patient Reported Outcomes (PRO). According to the recommendations established, a proper valida- tion of a PRO should address certain attributes (FDA, Valderas y Argimon): conceptual and measurement model, reliability, validity, responsiveness and interpretability. In the case of the DUFSS questionnaire, the quality of measurement of social support of each of the different val- idations is unclear, as no work to date has provided information on this. The objective of this study was to systematically review the available evidence on the psy- chometric and administration characteristics of the different versions of the DUFSS question- naire and perform a standardized assessment though to a specifically designed tool. Methods Protocol and registration A systematic review of the literature was carried out, and the results were reported in accor- dance with the guidelines of the Preferred Reporting Items of Systematic Reviews and Meta- Analyses Protocol (PRISMA). The protocol was registered in PROSPERO under registration number CRD42022342977. Eligibility criteria All articles that contained information on the development of the instrument, the psychomet- ric properties and aspects related to the administration of the Duke-UNC Functional Social Support (DUFSS) questionnaire were included. In order to make the search as sensitive as PLOS ONE | https://doi.org/10.1371/journal.pone.0291635 September 15, 2023 3 / 17 PLOS ONE Use of the EMPRO tool in different versions of DUFSS possible, there were no restrictions based on publication date, neither on the format (paper or digital), nor on the language of the article or the version or language of the questionnaire used. Regarding the study population, only studies conducted in a population under 18 years of age were excluded, but there were no restrictions based on other population characteristics or settings. Information sources The search was performed on 21/02/2023 in the PubMed/MEDLINE, SCIELO, SCOPUS and WOS databases with the aim of searching a broad swath of databases. Search strategy To develop the search strategy, the different names used for this questionnaire were taken into account. The search strategy used was adapted to each of the databases, in which the terms that appear in Table 1 were included. The reference lists of the included articles were manually searched, and authors were contacted to obtain additional data if necessary. Table 1 shows the terms that have been used to build the search strategy. Selection and data collection process Initially, two reviewers screened the titles according to the inclusion criteria. Then, the same two reviewers did the same for the abstracts. Once the duplicates had been removed and based on the selection criteria, the eligibility of the full articles was assessed. Discrepancies that arose at each of the selection stages were resolved by discussion and consensus between the two researchers, and a third reviewer was consulted when consensus could not be reached between the two previous reviewers. Data items The selected studies were grouped by version type according to the number of items that made up the version resulting from their study. The unit of analysis through EMPRO was each ver- sion type of the DUFSS. The following data were extracted: author and year; the version of the questionnaire (number of items and language); the characteristics of the population and coun- try; and the results obtained from the factor analysis carried out, whether exploratory and/or confirmatory: the dimensionality of the questionnaire (unifactorial or bifactorial with the items that make up each dimension). The evaluation and synthesis strategy of the selected arti- cles was included stratifying the articles based on the versions of the DUFSS questionnaire. Table 1. Search strategy. • “Social support” AND • “Duke Unc” OR “DUFSS” OR “FSSQ”. AND • Questionnaire* OR instrument* OR scale* OR index* OR survey* OR batter* OR inventor* OR measur* OR rating*. AND • Valid*OR Chronbach* OR "psychometric properties" OR psychometr* OR Factor Analysis,Statistical[MeSH] OR develop* OR valid* OR translat*. https://doi.org/10.1371/journal.pone.0291635.t001 PLOS ONE | https://doi.org/10.1371/journal.pone.0291635 September 15, 2023 4 / 17 PLOS ONE Use of the EMPRO tool in different versions of DUFSS Synthesis methods EMPRO tool. The EMPRO tool [23] was designed to measure the quality of PRO instru- ments. This tool demonstrated excellent reliability in terms of internal consistency (Cron- bach’s alpha = 0.95) and inter-rater concordance (intraclass correlation coefficient: 0.87–0.94). It evaluates quality as a global concept with 39 items across eight attributes (Table 2): “concep- tual and measurement model” (concepts and population to be evaluated); ’’ reliability ’’ (to what extent an instrument is free of random errors); ’’ validity ’’ (to what extent an instrument measures what it intends); “sensitivity to changes” (ability to detect changes over time); “interpretability” (assignment of meanings to instrument scores); “burden” (time, effort and other administration and response requirements); “alternative modes of administration” (self- administered or heteroadministered and route of administration); and ’’ cultural and linguistic adaptations". Responses on each item are given on 4-point Likert scale, where 4 is "totally agree", 1 is "totally disagree". Other response options include "no information" and "not appli- cable". The items answered as ‘‘no information” were assigned a score of 1 (lowest possible score) if at least 50% of all items for one attribute were rated; b) items rated as ‘‘not applicable” (an option that is only available as an answer for 5 items) were not considered as part of the attribute score. Standardized assessment Each instrument was evaluated by two different experts using the EMPRO tool. Three experts in measuring patient reported outcomes (PROs) composed the review group: two were senior researchers who belonged to the EMPRO tool development working group, and the third was a junior researcher who had been previously trained as an EMPRO evaluator. The pairs of reviewers were composed of a senior and a junior researcher. To minimize the likelihood of bias, the experts were not authors, nor had they participated in the process of development or adaptation of the assigned instrument. The EMPRO evaluation process consisted of two consecutive rounds. In the first round, each expert independently evaluated the instrument that had been assigned to them from the full-text articles identified. In the second round, each expert received the results of the rating assigned by their review partner. Discrepancies were resolved by discussion or by consulting a third reviewer. Analytic strategy To carry out the analysis of the information, first, the identified studies were stratified accord- ing to the resulting item version; second, the published recommendations of the EMPRO tool were followed for the calculation of the scores [23, 24]. For this, at least half of the items that made up each attribute had to be rated from 1 to 4 (responses of “no information” were assigned a score of 1). The mean score was transformed into a scale ranging from 0 (the worst possible score) to 100 (the best possible score). The attributes “reliability” and “load” are com- posed of two subattributes each: “internal consistency and reproducibility” and “response load and administration load”, respectively. For reliability, the highest subscore of its components was chosen to represent the attribute. The calculation of the global score was performed with the average of the first five attributes, since the attribute of load and alternative versions are not metric characteristics but management characteristics. The scores of the attribute of alter- native forms were not calculated because there were no validations for different forms of administration (it was self-reported, in some cases via interview), and the scores of the attri- bute of cultural and linguistic adaptations were not calculated because it was beyond the scope of this study. PLOS ONE | https://doi.org/10.1371/journal.pone.0291635 September 15, 2023 5 / 17 PLOS ONE Use of the EMPRO tool in different versions of DUFSS Table 2. Attributes assessed using the evaluating the measurement of patient-reported outcomes (EMPRO) tool. Attribute Definition Items included Conceptual and The rationale for and description of the concept and the populations that a measure is 1. Concept of measurement stated measurement model intended to assess and the relationship between these concepts 2. Obtaining and combining items described 3. Rationality for dimensionality and scales 4. Involvement of target population 5. Scale variability described and adequate 6. Level of measurement described 7. Procedures for deriving scores Reliability The degree to which an instrument is free from random error Internal consistency: 11. Data collection methods described 12. Cronbach’s alpha adequate (QA) 13. IRT estimates provided 14. Testing in different populations Reproducibility: 15. Data collection methods described 16. Test–retest and time interval adequate 17. Reproducibility coefficients adequate (QA) 18. IRT estimates provided Validity The degree to which the instrument measures what it purports to measure. 19. Content validity adequate 20. Construct/criterion validity adequate 21. Sample composition described 22. Prior hypothesis stated (QA) 23. Rational for criterion validity 24. Tested in different populations Responsiveness An instrument’s ability to detect change over time 25. Adequacy of methods (QA) 26. Description of estimated magnitude of change 27. Comparison of stable and unstable groups Interpretability The degree to which one can assign easily understood meaning to an instrument’s 28. Rational of external criteria quantitative scores. 29. Description of interpretation strategies 30. How data should be reported stated Burden The time, effort, and other demands placed on those to whom the instrument is Respondent: administered (respondent burden) or on those who administer the instrument 31. Skills and time needed (administrative burden) 32. Impact on respondents 33. Not suitable circumstances Administrative: 34. Resources required 35. Time required 36. Training and expertise needed 37. Burden of score calculation Alternative modes of Alternative modes of administration used for the administration of the instrument 38. The metric characteristics and use of each administration alternative mode of administration 39. Comparability of alternative modes of administration (Continued ) PLOS ONE | https://doi.org/10.1371/journal.pone.0291635 September 15, 2023 6 / 17 PLOS ONE Use of the EMPRO tool in different versions of DUFSS Table 2. (Continued) Attribute Definition Items included Cultural adaptation Cultural and linguistic adaptation of the instrument. 8. Linguistic equivalence (QA) 9. Conceptual equivalence 10 Differences between the original and the adapted versions QA: quality assessed The overall score of the tool gives each item values from 0 (the worst possible score) to 100 (the best possible score), resulting in an overall average score based on the attributes. The result is considered adequate if it reaches at least 50 points. https://doi.org/10.1371/journal.pone.0291635.t002 Results The search strategy (Table 1) yielded 54 studies; 52 studies were obtained from the databases used, and the remaining 2 articles were obtained via manual search. After eliminating dupli- cates, 30 potentially eligible studies remained. After screening the titles and abstracts, 23 stud- ies remained for full text review. Eight studies were then excluded, including 2 meeting abstracts and 6 studies that examined a different tool that also used the abbreviation DUFFS. Finally, a total of 15 articles were included in this review. The flow diagram is detailed in Fig 1. Table 3 shows the main characteristics of each of the published validations of the DUFSS ques- tionnaire. Of the 15 studies included, 73.3% of them examined the 11-item version while the rest start from the 8-item and 14-item version in equal parts (13.3% each). The studies that used the 11-item version as a reference were carried out in Spanish-speaking countries [15, 25–32] and in European countries, Italy [33] and Portugal [34], while those that used the 8-item and 14-item versions were carried out in English-speaking countries [16, 35–37], mostly in the USA. The original version of the questionnaire was designed as a tool to be used in the primary care setting. Thirty-three percent of the localized validations were performed in the field of pri- mary care, and the rest were performed in educational or specific care settings, such as mental health care centres or among pregnant mothers. Regarding the sociodemographic characteris- tics of the population in which they have been validated, 66.7% have been performed on a pre- dominantly female population; 60% in young adults, whose average age did not exceed 40 years, and 60% in population with medium-low socioeconomic status or who met characteris- tics of social vulnerability. From a methodological point of view, only 20% of the included studies used a confirmatory factor analysis to examine the factorial structure of the questionnaire and the dimensions in which its items are grouped. In the exploratory factor analysis, a total of 73.3% of studies reported a two-dimension structure, consisting of the dimensions obtained by Broadhead (confidential and affective). In contrast, 20% of the studies reported a one-dimensional struc- ture, and 6% of studies, all of which examined the modified 11-item version, reported a three- dimensional structure (confidential, affective and instrumental). Across the 15 studies, the characteristics of 4 different versions of the DUFSS questionnaire were evaluated: the 14-item, 11-item, 8-item and 5-item version. All versions use a Likert scale with 5 response options, except for the 5-item scale, which used both a 5-point Likert scale and a 3-point Likert scale. Table 4 shows the items that make up each of the versions. The elimination of some items in successive versions has resulted in the reestablishment of the numbering values. Therefore, the 8-item and 5-item versions do not share any numbering with the original version. A study performed on the 11-item version revealed that 2 new items were included but were not part of the original 14-item version. PLOS ONE | https://doi.org/10.1371/journal.pone.0291635 September 15, 2023 7 / 17 PLOS ONE Use of the EMPRO tool in different versions of DUFSS Fig 1. PRISMA flow chart—systematic literature search. https://doi.org/10.1371/journal.pone.0291635.g001 The results of the EMPRO evaluation about psychometric values of each of each version are shown in Table 5. The highest score was obtained by the 11-item version (54.01 points), fol- lowed by the 8-item version (36.31 points), the 14-item version (27.48 points) and the 5-item version (23.81 points). Conceptual and measurement model The 11-item version obtained the highest score for this attribute (54.76 points), while the rest of the versions obtained a score of 35.71 points. The aspects that were least addressed by the included studies were the description of the measurement scale (including its scores) and the participation of the sample in a previous pilot. Reliability Only studies of the 11-item version evaluated both aspects; the other studies just examined the data collection method and calculated Cronbach’s alpha coefficient without assessing internal consistency or reproducibility. The highest internal consistency score was obtained by the PLOS ONE | https://doi.org/10.1371/journal.pone.0291635 September 15, 2023 8 / 17 PLOS ONE Use of the EMPRO tool in different versions of DUFSS Table 3. Main characteristics of the DUFSS validations. Resulting Author/year Language Version Setting & Country Sample Results version version used 8-item version Broadhead et al. English 14-item Primary Care General Population n = 401 [1988] [16] USA Women: 78% AFE Age: 35.7 (± *) CF: 3, 4, 5, 6 and 7 Medium-high socioeconomic AF: 1, 2 and 8 status Kathy B Isaacs et al. English 8-items Specialized centre for pregnant women n = 186 AFE [2011] [35] USA Women: 100% Unifactorial - Low socioeconomic status H. M. Epino et al. English 8-item Rural Primary Care n = 603 AFE [2012] [36] HIV-positive Rwanda Women: 62% Unifactorial Age: 38 ± 10 Low socioeconomic status 11-item version De la Revilla et al. Spanish 11-item Primary Care n = 139 AFE [1991] [25] General Population Spain Women: 82% CF: 7,8,6,4,1 and 10 Age: 46 ± 17.6 AF: 11,9,2,3 and 5 Low socioeconomic status Bellón S. JA. et al. Spanish 11-item Primary Care n = 656 AFE [1996 [15] General Population Spain Women: 72% CF: 1, 2, 6, 7,8,9 and 10 Age: 50.6 ± 18.9 AF: 3, 4, 5 and 11 Low socioeconomic status Alvarado BE. et al. Spanish 11-item Municipal Population Register n = 193 AFE [2005] [26] Mothers of children between Women: 100% CF: 4,5,10 and 11 6–18 m Colombia - AF: 6,7,8 Piña L. A. et al. Spanish 11-item Rural specialized centre for HIV-positive n = 67 AFE [2007] [27] individuals Mexico Women: 32.40% Unifactorial Age: 36.4 ± 10.6 Low socioeconomic status Ayala A. et al. [2012] Spanish 11-item Municipal Population Register n = 1012 AFE [28] Noninstitutionalized seniors Spain Women: 56.30% CF: 7, 8, 6, 5, 4, 11 and 10; AF: Age: 72.1 ± 7.8 2, 1, 9 and 3 - Cuellar-Flores L. et al. Spanish 11-item Primary Care Caregivers Spain n = 128 AFE [2012] [29] Women: 85.90% CF: 2, 6, 7, 8, 9, 10 and 11; AF: Age: 54.9 ± 15.1 1, 3, 4 and 5 - Mas-Exposito L. et al. Spanish 11-item Specialized centre for people with mental illness n = 241 AFE [2013] [30] Spain Women: 32.40% CF: 4, 6,7,8, 10 and 11; Age: 41.7 ± 11.6 AF: 1, 2, 9, 3 and 5 Low socioeconomic status Rivas Diez [2013] [31] Spanish 11-item Educational centres Chile n = 371 AFE/AFC Women: 100% CF: 4, 5, 6, 7, 8, 10 and 11 Age: 37.6 ± 13.1 AF: 1, 2, 3 and 9 Medium-high socioeconomic status Specialized centre for victims of abuse Chile n = 97 AFE/AFC Women: 100% CF: 3, 5, 6, 7, 8, 9 and 10 Age: 41.9 ± 10 AF: 1, 2, 4 and 11 Medium-low socioeconomic status Caycho R. T. et al. Italian 11-item Specialized Centre for Peruvian Migrants Italy n = 150 AFC [2014] [33] Women: 58% CF: 1,4,6,7,8 y10; Age: 34.6 ± 10.3 AF: 2, 3,5, 9 and 11 Medium-low socioeconomic status Mónica Aguilar-Sizer Spanish 11-item Educational centre for the general population n = 535Women: 75.5%Age: AFE et al. [2021] [32] Ecuador 22± 5 CF: 6, 7 and 8 AF: 1, 2, 3, 4, 5, 9, 10 and 11 Martins, S. et al. Portuguese 11-item Mother and fathers of young children educational n = 1.058 AFC CF: 1, 6, 7,8,9 and 10 AF: [2022] [34] centres Portugal Women: 90.5% 3, 4 and 5 IF: 2, 11, 12 and 13 Age: 35.7± 5.2 Medium-high socioeconomic status (Continued ) PLOS ONE | https://doi.org/10.1371/journal.pone.0291635 September 15, 2023 9 / 17 PLOS ONE Use of the EMPRO tool in different versions of DUFSS Table 3. (Continued) Resulting Author/year Language Version Setting & Country Sample Results version version used 14-item version Rebecca Saracino English 14-item Specialized centres for patients with incurable n = 253 AFE Unifactorial 5-item/ et al. [2014] [37] and advanced diseases (AIDS and cancer) USA Women: 69.6% AFE Unifactorial 3-response Age: 58.2± 11 - version * Data not available. Factor analysis performed: Confirmatory Factor Analysis (CFA), Exploratory Factor Analysis (EFA) CF (Confidential Factor), AF (Affective Factor) and FI (Instrumental Factor) https://doi.org/10.1371/journal.pone.0291635.t003 11-item version (70.83 points), because Cronbach’s alpha in all studies was higher than 0.9. In addition, most of the internal consistency quality criteria, with the exception of the IRT crite- rion, also had high scores (three or four crosses). Although most studies reported an adequate Cronbach’s alpha coefficient (�0.70), they did not comprehensively measure the reliability of the instrument. Validity Only one study, from the 11-item version, used the EFA and CFA to examine content validity. The versions with the highest scores for this attribute were the 11-item version (66.67 points) and the 8-item version (62.50 points). In the included studies, differences were been observed between the dimensions of the DUFSS both in the distribution of its items and in the number of dimensions included in the tool. The relationship with previous hypotheses and related Table 4. Different versions of the DUFSS questionnaire by number of items. 14-item version* 11-item 11-item version modified* 8-item 5-item version** version* version* Item 1: visits with friends and relatives. Item 1 Item 1 Item 5 Item 5 Item 2: help around the house. Item 2 Item 2 Item 6 Item 6 Item 3: help with money in an emergency. Item 4 Item 4. Item 8 Item 9 Item 4: praise for a good job. Item 5 Item 5 Item 9 Item 12 Item 5: people who care what happens to me. Item 6 Item 6 Item 10 help when I need transportation. Item 6: love and affection. Item 8 Item 8 Item 11 Item 7: telephone calls from people I Know. Item 9 Item 11 Item 12 Item 8: chances to talk to someone about problems at work or with Item 10 Item 9 Item 14 my housework. Item 9: chances to talk to someone I trust about my personal and Item 11 Item 12 family problems. Item 10: chances to talk about money matters. Item 12 Item 10 Item 11: invitations to go out and do things with other people. Item 14 Item 14 Item 12: I get useful advice about important things in life. help with transportation and move Item 13: help when I need transportation. help with my children’s care Item 14: help when I’m sick in bed. *5-response Likert-type **3-response Likert-type https://doi.org/10.1371/journal.pone.0291635.t004 PLOS ONE | https://doi.org/10.1371/journal.pone.0291635 September 15, 2023 10 / 17 PLOS ONE Use of the EMPRO tool in different versions of DUFSS Table 5. EMPRO results: psychometric values of each version of the DUFSS questionnaire. Attributes 14-Items 11-Items 8-Items 5-Items Conceptual and measurement model 35.71 54.76 35.71 35.71 1. Concept of measurement stated +++I +++I +++I +++I 2. Obtaining and combining items described ++ +++ ++ ++ 3. Rationality for dimensionality and scales +++ +++ ++I +++ 4. Involvement of target population + ++ + + 5. Scale variability described and adequate + +++ +I + 6. Level of measurement described ++ ++ ++ ++ 7. Procedures for deriving scores ++ ++ ++ ++ Reliability 41.66 70.83 44.44 25 Internal consistency 41.66 70.83 44.44 25 8. Data collection methods described +++ +++I ++I ++ 9. Cronbach’s alpha adequate ++++ +++ +++I +++ 10. IRT estimates provided NI +++ NI NI 11. Testing in different populations NI +++ NI NI Reproducibility - 45.83 - - 12. Data collection methods described NI +++I NI NI 13. Test–retest and time interval adequate NI ++ NI NI 14. Reproducibility coefficients adequate + +++ NI NI 15. IRT* estimates provided NI NI NI NI Validity 26.67 66.67 62.50 25 16. Content validity adequate + ++ +I + 17. Construct/criterion validity adequate ++ +++I ++I ++ 18. Sample composition described +++ +++I ++++ +++I 19. Prior hypothesis stated ++ +++ +++I NI 20. Rational for criterion validity NA NA NA ++ 21. Tested in different populations + +++ NA + Responsiveness - 33.33 - - 22. Adequacy of methods NI +++ NI NI 23. Description of estimated magnitude of change NI ++ NI NI 24. Comparison of stable and unstable groups NI NI NI NI Interpretability 33.33 44.44 38.89 33.33 25. Rational of external criteria ++ +++ ++I ++ 26. Description of interpretation strategies NI ++ NI NI 27. How data should be reported stated +++ ++ +++ +++ Overall score 27.48 54.01 36.31 23.81 Burden Burden: respondent 33.33 66.67 38.89 33.33 28. Skills and time needed +I +++I ++ +I 29. Impact on respondents +I ++I ++ +I 30. Not suitable circumstances +++ +++ ++I +++ Burden: administrative 33.33 50.00 66.67 66.67 31. Resources required NI +++ ++I +++ 32. Time required NA NA NA NA 33. Training and expertise needed NA NA NA NA 34. Burden of score calculation +++ ++ +++I +++ Ítems Score: + 1 point, I ½ point, NI no information, NA. not applicable. *IRT (Item Response Theory) https://doi.org/10.1371/journal.pone.0291635.t005 PLOS ONE | https://doi.org/10.1371/journal.pone.0291635 September 15, 2023 11 / 17 PLOS ONE Use of the EMPRO tool in different versions of DUFSS variables (convergent validity) was analysed in the 14-, 11- and 8-item versions, which improved their scores on the validity attribute. Responsiveness Regarding the sensitivity to change, only the 11-item version detects the sensitivity to change compared with of other validated scales (33.33 points); however, the 11-item version does not offer comparative results between groups (longitudinal validity), because most of the included studies used a cross-sectional design. In the rest of the versions, not enough information was found to calculate scores. Interpretability The interpretability of the questionnaire was similar across all versions (with a range of 33.33 to 44.44 points); the 11-item version had the highest score, followed by the 8-item version. None of the 4 versions studied herein offer sufficient information on the measurement and interpretation strategies of the DUFSS questionnaire. Only Bellón et al. (11-item version) pro- vided information by stating that the 15th percentile (score �32) of their sample was the cut- off point for differentiating "good" from "low" social support. Burden Regarding response load and administration load, both attributes had scores ranging from 33.33 to 66.67. The 11-item version obtained a higher score for response load, whereas the 8- and 5-item versions obtained the highest scores for administration load. Discussion Main results This systematic review has identified 15 studies on the validation of the DUFSS that have obtained different versions of the questionnaire according to the number of resulting items: 14-items, 11-items, 8-items and 5-items. Validations carried out in Spanish-speaking countries predominate in which, as in Italy and Portugal, the 11-item version validated by De La Revilla et al. was used, while in English-speaking countries, mostly in the USA, the 8-item and 14-item version validated by Broadhead et al. was used for their study (citation). Despite the fact that the origin of the questionnaire is American and that more than half of the studies were carried out in countries with different languages (Spanish, Portuguese, Italian and English) and in very spe- cific populations (young adults, women and with a medium-low socio-economic level or with characteristics of social vulnerability), none of the studies analysed describe the process used to translate and culturally adapt the instrument to their study populations. This aspect could call into question the content validity of the different versions of the questionnaire in relation to the original. Comparing this reality with that of similar studies, we found that the 12-item Multidi- mensional Scale of Perceived Social Support (MSPSS) [28] was assessed for the psychometric properties of its existing translations using the COSMIN tool [29] and concluded that the trans- lated versions provided little evidence for content validity. This absence may explain the differ- ences in the distribution of its items as a result of the EFA and ACE (Table 4). The quality assessment of the DUFSS questionnaire through the EMPRO tool showed that the 11-item version is the only one that has been studied through all the recommended attri- butes and the one that has obtained the highest score in its evaluation. This version scored higher than the rest in all the attributes studied, with the best scores for "reliability (internal consistency)" and "validity"; its lowest score was for the attribute "responsiveness", where the PLOS ONE | https://doi.org/10.1371/journal.pone.0291635 September 15, 2023 12 / 17 PLOS ONE Use of the EMPRO tool in different versions of DUFSS rest of the versions had no information (N/A). This result is probably due to the fact that the 11-item version has been studied by 73.3% of the studies analysed, while the other versions have less validations. The 8-item version obtained its best score in the attribute "validity", and attributes such as "responsiveness" and "reliability (reproducibility)"; the 14-item version obtained its highest score in the section "reliability (internal consistency)", despite the lack of information in "reliability (reproducibility)"; the 8-item version did not present information in the attributes "responsiveness" and "reliability (reproducibility)", hence the resulting low score. The complex nature of PRO instruments often raises important questions about how to inter- pret and communicate the results in a way that is not misleading, so it is essential that the vali- dation of the instrument makes clear how the results are to be interpreted. None of the 4 versions studied herein offer sufficient information on the measurement and interpretation strategies of the DUFSS questionnaire. One of the most frequent strategies for interpreting PROs is to calculate percentiles from population values [38]. This method is only valid if the data come from a representative sample of the general population; however, this was not the case in the included studies. Bellón et al. found that the 15th percentile was the cut-off point to differentiate “good” from “low” social support. This percentile corresponded to a score �32, a score that later Fernández et al. [39] and Ruiz et al. [40] applied in their studies. On the other hand, authors such as Slade et al. [41] and Harley et al. [42] and the National Health Survey of Spain [43] expressed the results of social support in the form of a quantitative variable. The lat- ter option seems to be the most appropriate approach, since it does not require percentiles to be extracted from the general population. However, the most appropriate approach would be to use population values or norms, reference samples, or standardized mean responses. Limitations and strengths This work constitutes the first review of the social support construct using a standardized methodology, the EMPRO tool, for the evaluation of one of the most well-known instruments. A limitation of this work is that the systematic reviews depend on the information retrieved through the search strategy, so it is possible that we have not identified all the articles published on the questionnaire to be studied. However, given that social support is a construct of interest in multiple disciplines and that some validations might be indexed in non-health science data- bases, in addition to including Pubmed, we also searched multidisciplinary databases such as SCOPUS and WOS. This aspect added to the delicate search strategy designed, the additional manual search of references, as well as the double independent review process followed, may have minimised this problem [16]. The most important aspect to note as a limitation is that the results shown here do not sug- gest a single, unequivocal ranking or recommendation, as the EMPRO assessment includes several attributes and they have different relevance according to the purpose of the application of the questionnaire. For example, if the purpose is monitoring patients, responsiveness is the key attribute. Although this tool has shown excellent reliability in terms of internal consistency and inter-rater concordance the EMPRO scores depend on the quantity and quality of infor- mation provided for the assessment. Thus, information on psychometric properties that are not available, they are not taken into account and therefore penalise the assessment. On the other hand, EMPRO offers the possibility to answer with "no information" and if the half of items has not information, the attribute score is not calculated. In this regard, it is important to highlight that the work carried out by Ayala et al. [28] on the 11-item version addresses most of the attributes proposed by EMPRO, which increases the score of this version. EMPRO ratings may be biased by the individual expertise of the evaluators, although the double and independent review conducted, as well as a comprehensive description of each PLOS ONE | https://doi.org/10.1371/journal.pone.0291635 September 15, 2023 13 / 17 PLOS ONE Use of the EMPRO tool in different versions of DUFSS item, may have attenuated this concern. Studies on metric properties from different country versions were considered in our EMPRO evaluation. Although these country versions can add noise in one sense, they also provide valuable information about the generalizability of the psy- chometric data to these measures. The coexistence of different nomenclatures has been a challenge in the process of reviewing and selecting articles. The name of the questionnaire has been modified since the original author called it the Duke-UNC Functional Support Questionnaire (DUFSS). Subsequently, various researchers have modified the original acronym, thus yielding a variety of names: DUKE-UNC, DUKE-UNC-11 DUKE-UNK, FSSQ and DUFSSQ. This increases the confusion among researchers who want to use some version of the DUFSS; additionally, these alternate acronyms may be similar to different tools that measure separate concepts, thus creating the potential for more confusion–e.g., the Duke Social Support Index (DSSI) [44] or the Duke- UNC Health Profile (DUHP) [45]. In order to ensure that all validation articles on the DUFSS were located, we included the possible names it may acquire in the search strategy (this part is explained in the methodology); and once we had the search results, we went on to read the full text of any paper that might raise doubts, in order to ensure that it was the right questionnaire. Applicability This paper can shed light on the study of social support as a PRO in different domains and help to unravel the current complexity that exists around this questionnaire. To know the different versions of the DUFSS questionnaire and providing relevant infor- mation about each one will allow researchers who wish to study this subject to choose the ver- sion that best suits their interests and to be aware of the evaluation of its quality. In addition, improving knowledge about this PRO will allow progress to be made and give greater strength to the work in the field of epidemiology and public health on person-centred care. Furthermore, in the educational field, this study has two applications: on the one hand, to train specific tools in the study of PROs; on the other hand, it focuses on the importance of consulting the original sources and investigating the work that has been done previously on the research question. Conclusions There are 4 versions of the EMPRO questionnaire with different numbers of items: 14, 11, 8 and 5 items. All of them have been validated in very specific populations and not in the general population. Among the 4 versions the DUFSS questionnaire, the 11-item version has been the most studied, especially in Spanish-speaking countries. This version scored higher than the others because it was the version with the largest number of studies and therefore more likely to address all the attributes taken into account by the EMPRO tool. In order to be able to state with certainty that the 11-item version is more appropriate than the other versions, more studies are needed to evaluate each of the other versions. Although, a priori, we could prioritise its use in epidemiological studies over the other versions, it should be noted that this version should also be used with caution because there are attributes that have not been studied. All versions of the DUFSS questionnaire should be used with caution, since many of the attributes studied have not shown sufficient quality in any of the versions analysed herein. It is necessary to conduct future studies on the DUFSS questionnaire to evaluate aspects such as its reproducibility and to perform complete factor analysis in the general population. PLOS ONE | https://doi.org/10.1371/journal.pone.0291635 September 15, 2023 14 / 17 PLOS ONE Use of the EMPRO tool in different versions of DUFSS Supporting information S1 Checklist. PRISMA 2020 checklist. (DOCX) S1 Dataset. (XLSX) Acknowledgments We thank JA López-Rodrı́guez for his help in reviewing this work. Author Contributions Conceptualization: Cristina M. Lozano-Hernández, Yolanda Pardo Cladellas, Isabel del Cura-González. Data curation: Cristina M. Lozano-Hernández, Yolanda Pardo Cladellas, Mario Gil Conesa, Olatz Garin, Isabel del Cura-González. Formal analysis: Cristina M. Lozano-Hernández, Yolanda Pardo Cladellas, Mario Gil Conesa, Olatz Garin, Isabel del Cura-González. Funding acquisition: Cristina M. Lozano-Hernández, Isabel del Cura-González. Investigation: Cristina M. Lozano-Hernández, Isabel del Cura-González. Methodology: Cristina M. Lozano-Hernández, Yolanda Pardo Cladellas, Mario Gil Conesa, Olatz Garin, Montserrat Ferrer Forés, Isabel del Cura-González. Resources: Cristina M. Lozano-Hernández. Supervision: Montserrat Ferrer Forés, Isabel del Cura-González. Validation: Cristina M. Lozano-Hernández, Yolanda Pardo Cladellas, Isabel del Cura- González. Visualization: Cristina M. Lozano-Hernández. Writing – original draft: Cristina M. Lozano-Hernández. Writing – review & editing: Cristina M. Lozano-Hernández, Yolanda Pardo Cladellas, Mont- serrat Ferrer Forés, Isabel del Cura-González. References 1. Kaplan BH, Cassel JC, Gore S. Social Support and Health. Med Care. 1977; 15: 47–58. https://doi.org/ 10.1097/00005650-197705001-00006 PMID: 853781 2. Gallant MP. The influence of social support on chronic illness self-management: A review and directions for research. Health Education and Behavior. 2003; 30: 170–195. https://doi.org/10.1177/ 1090198102251030 PMID: 12693522 3. Lozano-Hernández CM, López-Rodrı́guez JA, Leiva-Fernández F, Calderón-Larrañaga A, Barrio-Cor- tes J, Gimeno-Feliu LA, et al. Social support, social context and nonadherence to treatment in young senior patients with multimorbidity and polypharmacy followed-up in primary care. MULTIPAP study. PLoS One. 2020; 15: 1–15. https://doi.org/10.1371/journal.pone.0235148 PMID: 32579616 4. Blazer DG. Social support and mortality in an elderly community population. Am J Epidemiol. 1982; 115: 684–694. Available: https://academic.oup.com/aje/article/115/5/684/153301 https://doi.org/10. 1093/oxfordjournals.aje.a113351 PMID: 7081200 5. Sidney Cobb MD. Social Support as a Moderator of Life Stress. Psychosomatic Medicine. 1976;38. PLOS ONE | https://doi.org/10.1371/journal.pone.0291635 September 15, 2023 15 / 17 PLOS ONE Use of the EMPRO tool in different versions of DUFSS 6. Cassell A, Edwards D, Harshfield A, Rhodes K, Brimicombe J, Payne R, et al. The epidemiology of mul- timorbidity in primary care: A retrospective cohort study. British Journal of General Practice. 2018; 68: e245–e251. https://doi.org/10.3399/bjgp18X695465 PMID: 29530918 7. Cohen S. Psychosocial models of the role of social support in the etiology of physical disease. Health Psychol. 1988; 7: 269–97. https://doi.org/10.1037//0278-6133.7.3.269 PMID: 3289916 8. Kahn Rober L., Antonucci Toni C. Convoys Over the Life Course: Attachment Roles and Social Support. Life-Span Development and Behavior. 1980; 3: 253–283. Available: https://www.researchgate.net/ publication/259253271 9. To T, Ly ANA, Winter TIONS, Wong ST, Yoo GJ, Stewart AL, et al. Conceptual, Methodological, and Theoretical Problems in Studying Social Support as a Buffer Against Life Stress Author (s): Peggy A. Thoits Source: Journal of Health and Social Behavior, Vol. 23, No. 2 (Jun., 1982), pp. 145–159 Pub- lished by. Nurs Res. 1998;61: 2–5. https://doi.org/10.1002/(SICI)1097-4679(199810)54:6<845::AID- JCLP13>3.0.CO;2-L 10. Thoits PA. Conceptual, methodological, and theoretical problems in studying social support as a buffer against life stress. J Health Soc Behav. 1982; 23: 145–159. https://doi.org/10.2307/2136511 PMID: 7108180 11. House JS, Kahn RL. Measures and concept of social support. Social support and health. 1985. pp. 83– 108. Available: http://www.isr.umich.edu/williams/All Publications/DRW pubs 1985/measures and con- cept of social support.pdf 12. Gottlieb BH, Bergen AE. Social support concepts and measures. J Psychosom Res. 2010; 69: 511– 520. https://doi.org/10.1016/j.jpsychores.2009.10.001 PMID: 20955871 13. Aranda B. C, Pando M. M. Conceptualización del apoyo social y las redes de apoyo social. Revista de Investigación en Psicologı́a. 2014; 16: 233. https://doi.org/10.15381/rinvp.v16i1.3929 14. Barrón Ana. Apoyo social: aspectos teóricos y aplicaciones. Siglo XXI de España General, editor. Madrid: Siglo XXI; 1996. 15. Bellón JA, Delgado A, Luna J LP. Validez y fiabilidad del cuestionario de apoyo social funcional Duke- UNC-11. Aten Primaria. 1996; 18: 153–63. 16. Broadhead W.E., Gehlbach H Stephen, de Gruy Frank V, Kaplan Berton H. The Duke-UNC Funtional Social Support Questionnaire. Measurement of Social Support in Family Medicine Patients. Med Care. 1988; 26: 709–723. 17. DiMatteo MR. Social Support and Patient Adherence to Medical Treatment: A Meta-Analysis. Health Psychology. 2004; 23: 207–218. https://doi.org/10.1037/0278-6133.23.2.207 PMID: 15008666 18. Uchino BN, Cacioppo JT, Kiecolt-Glaser JK. The relationship between social support and physiological processes: A review with emphasis on underlying mechanisms and implications for health. Psychol Bull. 1996; 119: 488–531. https://doi.org/10.1037/0033-2909.119.3.488 PMID: 8668748 19. Broadhead W.E., Gehlbach H Stephen, de Gruy Frank V, Kaplan Berton H. Functional versus Structural Social Support and Health Care Utilization in a Family Medicine Outpatient Practice. Med Care. 1989; 27: 221–233. https://doi.org/10.1097/00005650-198903000-00001 PMID: 2784523 20. Broadhead WE, Kaplan BH, James SA, Wagner EH, Schoenbach VJ, Grimson R, et al. The epidemio- logic evidence for a relationship between social support and health. J Epidemiol. 1983; 117: 521–537. Available: https://academic.oup.com/aje/article/117/5/521/102512 https://doi.org/10.1093/ oxfordjournals.aje.a113575 PMID: 6342368 21. Instituto Nacional de Estadı́stica, Ministerio de Sanidad SS e I. Encuesta Nacional de Salud 2017 ENSE 2017 Metodologı́a. 2017. 22. European Health Interview Survey—Access to microdata—Eurostat. [cited 25 Jul 2022]. Available: https://ec.europa.eu/eurostat/web/microdata/european-health-interview-survey 23. Valderas JM, Ferrer M, Mendı́vil J, Garin O, Rajmil L, Herdman M, et al. Development of EMPRO: A tool for the standardized assessment of patient-reported outcome measures. Value in Health. 2008; 11: 700–708. https://doi.org/10.1111/j.1524-4733.2007.00309.x PMID: 18194398 24. Schmidt S, Garin O, Pardo Y, Valderas JM, Alonso J, Rebollo P, et al. Assessing quality of life in patients with prostate cancer: A systematic and standardized comparison of available instruments. Quality of Life Research. Springer International Publishing; 2014. pp. 2169–2181. https://doi.org/10. 1007/s11136-014-0678-8 PMID: 24748557 25. de la Revilla Ahumada L, Bailón E, de Dios Luna J, Delgado A, Prados MA, Fleitas L. Validation of a functional social support scale for use in the family doctor’s office. Aten Primaria. 1991; 8: 688–92. 26. Alvarado BE, Zunzunegui MV, Delisle H. Validación de escalas de seguridad alimentaria y de apoyo social en una población afro-colombiana: aplicación en el estudio de prevalencia del estado nutricional en niños de 6 a 18 meses. Cad Saude Publica. 2005; 21: 724–736. https://doi.org/10.1590/S0102- 311X2005000300006 PMID: 15868030 PLOS ONE | https://doi.org/10.1371/journal.pone.0291635 September 15, 2023 16 / 17 PLOS ONE Use of the EMPRO tool in different versions of DUFSS 27. Piña LJ, Rivera IB. Validación del Cuestionario de Apoyo Social Funcional en personas seropositivas al VIH del noroeste de México. Ciencia y Enfermerı́a. 2007; XIII: 53–63. https://doi.org/10.4067/S0717- 95532007000200007 28. Ayala A, Rodrı́guez-Blázquez C, Frades-Payo B, Forjaz MJ, Martı́nez-Martı́n P, Fernández-Mayoralas G, et al. Psychometric properties of the Functional Social Support Questionnaire and the Loneliness Scale in non-institutionalized older adults in Spain. Gac Sanit. 2012; 26: 317–324. https://doi.org/10. 1016/j.gaceta.2011.08.009 PMID: 22265651 29. Cuéllar-Flores I, Dresch V. Validación del cuestionario de Apoyo Social Funcional Duke-UNK-11 en personas cuidadoras. Ridep � No. 2012; 34: 89–101. 30. Mas-Expósito L, Amador-Campos JA, Gómez-Benito J, Lalucat-Jo L. Validation of the modified DUKE- UNC Functional Social Support Questionnaire in patients with schizophrenia. Soc Psychiatry Psychiatr Epidemiol. 2013; 48: 1675–1685. https://doi.org/10.1007/s00127-012-0633-3 PMID: 23229203 31. Rivas Diez Raquel. Apoyo Social Funcional en mujeres de la población general y en mujeres maltrata- das chilenas. Propiedades psicométricas del Duke-UNC-11 Functional social support in the general population and Chilean battered women. Psychometric properties of the. 2013. 32. Mónica Aguilar-Sizer Sandra Lima-Castro, Paúl Arias Medina Eva Karina Peña Contreras, Marcela Cabrera-Vélez Alexandra Bueno-Pacheco. Propiedades Psicométricas del Cuestionario de Apoyo Social Funcional Duke-UNK-11 en una Muestra de Adultos Ecuatorianos. “EUREKA” Revista de Inves- tigación Cientı́fica en Psicologı́a. 2021; 18: 55–71. Available: www.psicoeureka.com.py 33. Caycho Rodrı́guez T, Domı́nguez Lara S, Villegas G, Sotelo N, Carbajal León C. Análisis psicométrico del Cuestionario de Apoyo Social Funcional DUKE-UNK-11 en inmigrantes peruanos en Italia. Pensa- miento Psicológico. 2014; 12: 25–35. https://doi.org/10.11144/Javerianacali.PPSI12-2.apca 34. Martins S, Martins C, Almeida A, Ayala-Nunes L, Gonçalves A, Nunes C. The Adapted DUKE-UNC Functional Social Support Questionnaire in a Community Sample of Portuguese Parents. Res Soc Work Pract. 2022; 104973152210760. https://doi.org/10.1177/10497315221076039 35. Isaacs KB, Hall LA. A psychometric analysis of the functional social support questionnaire in low- income pregnant women. Issues Ment Health Nurs. 2011; 32: 766–773. https://doi.org/10.3109/ 01612840.2011.610561 PMID: 22077749 36. Epino HM, Rich ML, Kaigamba F, Hakizamungu M, Socci AR, Bagiruwigize E, et al. Reliability and con- struct validity of three health-related self-report scales in HIV-positive adults in rural Rwanda. AIDS Care—Psychological and Socio-Medical Aspects of AIDS/HIV. 2012; 24: 1576–1583. https://doi.org/ 10.1080/09540121.2012.661840 PMID: 22428702 37. Saracino R, Kolva E, Rosenfeld B, Breitbart W, R. S, E. K, et al. Measuring social support in patients with advanced medical illnesses: An analysis of the Duke–UNC Functional Social Support Question- naire. Palliat Support Care. 2015; 13: 1153–1163. https://doi.org/10.1017/S1478951514000996 PMID: 25201170 38. Argimon Pallás JMa. Métodos de investigación clı́nica y epidemiológica. Elsevier España. Elsevier España SLU, editor. 2019. 39. Fernández Vargas AM BZMLFMF. Salud autopercibida, apoyo social y familiar de los pacientes con enfermedad pulmonar obstructiva crónica. Medifarm. 2001; 11: 530–539. 40. Ruiz I, Olry A, Delgado CJ, Herrero MM, Muñoz N, Pasquau J, et al. Impacto del apoyo social y la mor- bilidad psı́quica en la calidad de vida en pacientes tratados con antirretrovirales. Psicothema. 2005; 17: 245–249. 41. Slade P., C.O Neill, Simpson A.J., Lashen H. The relationship between perceived stigma,disclosure patterns, support and distress in new attendees an infertitlty clinic. Human Reproduction. 2007; 22: 2309–2317. 42. Harley K, Eskenazi B. Time in the United States, social support and health behaviors during pregnancy among women of Mexican descent. Soc Sci Med. 2006; 62: 3048–3061. https://doi.org/10.1016/j. socscimed.2005.11.036 PMID: 16403596 43. Instituto Nacional de Estadistica. Nota Técnica Encuesta Nacional de Salud. España 2017 Principales resultados. 2017. Available: https://www.mscbs.gob.es/estadEstudios/estadisticas/encuestaNacional/ encuestaNac2017/ENSE2017_notatecnica.pdf 44. Koenig HG, Westlund RE, Linda George MK, Hughes DC, Blazer DG, Hybels C. Abbreviating the Duke Social Support Index for Use in Chronically mElderly Individuals. 1993. 45. Parkerson GR, Gehlbach SH, Wagner EH, James SA, Clapp NE, Muhlbaier LH. The Duke-UNC Health Profile: An Adult Health Status Instrument for Primary Care. Med Care. 1981; 19: 806–28. https://doi. org/10.1097/00005650-198108000-00002 PMID: 7278417 PLOS ONE | https://doi.org/10.1371/journal.pone.0291635 September 15, 2023 17 / 17