How do surrogates make treatment decisions for patients with dementia: An experimental survey study.

Despite the growing need for surrogate decision-making for older adults, little is known about how surrogates make decisions and whether advance directives would change decision-making. We conducted a nationally representative experimental survey that cross-randomized cognitive impairment, gender, and characteristics of advance care planning among hospitalized older adults through a series of vignettes. Our study yielded three main findings: first, respondents were much less likely to recommend life-sustaining treatments for patients with dementia, especially after personal exposure. Second, respondents were more likely to ignore patient preferences for life-extending treatment when the patient had dementia, and choose unwanted life-extending treatments for patients without dementia. Third, in scenarios where the patient's wishes were unclear, respondents were more likely to choose treatments that matched their own preferences. These findings underscore the need for improved communication and decision-making processes for patients with cognitive impairment and highlight the importance of choosing a surrogate decision-maker with similar treatment preferences.

Exploratory factor analysis and Rasch analysis to assess the structural validity of the Adult Social Care Outcomes Toolkit Proxy version (ASCOT-Proxy) completed by care home staff.

PURPOSE: Rasch analysis and exploratory factor analysis (EFA) were used to evaluate the structural validity of the ASCOT-Proxy measures completed by staff on behalf of older adults resident in care homes, by comparison to the ASCOT-SCT4, the measure of social care-related quality of life (SCRQoL) from which the ASCOT-Proxy was developed. METHODS: EFA was conducted on the ASCOT-SCT4 and the two ASCOT-Proxy measures (Proxy-Proxy, Proxy-Resident), to determine if they retained the single factor of the original ASCOT-SCT4 measure found in samples of older community-dwelling adults. Rasch analysis was also applied to measures with a single factor structure in the EFA. RESULTS: ASCOT-Proxy-Resident had a single factor structure, as did the original ASCOT-SCT4 (also, found in this analysis when completed by care home staff). The ASCOT-Proxy-Proxy had a two factor structure. Rasch analysis of ASCOT-Proxy-Resident and ASCOT-SCT4 had an acceptable model fit, internal consistency and met the assumptions of unidimensionality and local independence. There was evidence of less than optimal distinguishability at some thresholds between responses, and low frequency of rating of the 'high level needs'. CONCLUSION: The ASCOT-Proxy-Resident is a valid instrument of SCRQoL for older adults resident in care homes, completed by staff proxies. Due to the two-factor structure, which differs from the original ASCOT-SCT4, we do not recommend the use of the ASCOT-Proxy-Proxy measure, although collecting data as part of the ASCOT-Proxy questionnaire may support its feasibility and acceptability. Further qualitative study of how care home staff complete and perceive the ASCOT-Proxy is encouraged for future studies.

Measuring health-related quality of life in children with suspected genetic conditions: validation of the PedsQL proxy-report versions.

PURPOSE: Measuring health-related quality of life (HRQoL) of children with suspected genetic conditions is important for understanding the effect of interventions such as genomic sequencing (GS). The Pediatric Quality of Life Inventory (PedsQL) is a widely used generic measure of HRQoL in pediatric patients, but its psychometric properties have not yet been evaluated in children undergoing diagnostic GS. METHODS: In this cross-sectional study, we surveyed caregivers at the time of their child's enrollment into GS research studies as part of the Clinical Sequencing Evidence Generating Research (CSER) consortium. To evaluate structural validity of the PedsQL 4.0 Generic Core Scales and PedsQL Infant Scales parent proxy-report versions, we performed a confirmatory factor analysis of the hypothesized factor structure. To evaluate convergent validity, we examined correlations between caregivers' reports of their child's health, assessed using the EQ VAS, and PedsQL scores by child age. We conducted linear regression analyses to examine whether age moderated the association between caregiver-reported child health and PedsQL scores. We assessed reliability using Cronbach's alpha. RESULTS: We analyzed data for 766 patients across all PedsQL age group versions (1-12 months through 13-18 years). Model fit failed to meet criteria for good fit, even after modification. Neither age group (categorical) nor age (continuous) significantly moderated associations between PedsQL scores and caregiver-reported child health. Cronbach's alphas indicated satisfactory internal consistency for most PedsQL scales. CONCLUSION: The PedsQL Generic Core Scales and Infant Scales may be appropriate to measure HRQoL in pediatric patients with suspected genetic conditions across a wide age range. While we found evidence of acceptable internal consistency and preliminary convergent validity in this sample, there were some potential problems with structural validity and reliability that require further attention.

Agreement between quality of life assessed using family proxy and child self-reports among children with hematologic malignancy.

Evidence-based literature recognizes that the different degrees of agreement between a child self-report and a proxy-report depend on the characteristics of the domains, the child's age and illness, the proxy's own perspective on QoL, and family attendance during the child's hospitalization. This study aims to determine the degree of agreement between proxy-reports and child self-reports on quality of life (QoL) for children with hematologic malignancy ranging in age from 5 to 18 years who are undergoing treatment. We retrieved clinical QoL data from a study titled "Dynamic change in QoL for Vietnamese children with hematologic malignancy" from April 2021 to December 2022. To evaluate the magnitude of agreement between self-reports and proxy-reports, intraclass correlation coefficients (ICCs) for 259 pairs of measurements were quantified. Using independent t tests, the mean differences between self-reports and proxy-reports were tested. Moderate agreement was consistent through all age groups for five subscales, including physical, psychosocial, pain, nausea, and procedural anxiety (ICCs ranged from 0.53 to 0.74). The weakest agreement appeared in two groups, subjects aged 5-7 years and 13-18 years on six domains (school, treatment anxiety, worry, cognitive problems, perceived physical appearance, and communication) (-0.01 to 0.49). Child self-rating was consistently higher than that of proxies for the physical, emotional, and nausea domains among children aged 5-7 years and for procedural anxiety, treatment anxiety, and cognitive problems among children aged 8-12 years.    Conclusion: The agreement level of self-reports and proxy-reports was differently distributed by child age and the PedsQL domains. The proxy children agreement on QoL among children with hematologic malignancy was divergent according to the different age groups, which could potentially be explained by proxy-child bonding at different stages of childhood development. Our recommendation for future studies is to explore children's age as a potential factor influencing proxy agreement on QoL among children with cancer. What is Known: • Children and their proxies may think differently about quality of life (QoL). • Comparing two sources of data (i.e., child and proxy) on aspects of QoL can help identify the discrepancies between children's perceptions of their QoL and their parents' perceptions. This can be useful in terms of identifying potential areas for improvement or concern and may also be helpful in making decisions about treatment and care. What is New: • Our study results demonstrated that proxies who comprised children aged 5-7 years or 13-18 years reported differently among domains that cannot be expressed verbally or with body language, including cognitive problems, perceived physical appearance, and communication. • Children generally perceived their QoL to be better than their proxies. Therefore, a more comprehensive understanding of children's QoL may require the consideration of multiple sources of data from various perspectives.

Parents underestimate fatigue in younger children aged 5-7 years with asthma but not in older children.

AIM: To assess fatigue in children aged 2-17 years with asthma from both child and parent perspectives and describe associated factors. METHODS: Fatigue scores were self-reported by children aged 5-17 years old and proxy-reported by parents or carers for all children. The Pediatric Quality of Life Inventory Multidimensional Fatigue Scale was used. Fatigue scores ranged from 0 to 100, higher scores meant less fatigue. RESULTS: There were 527 children and parents enrolled. The mean overall fatigue score by self-report was 72.7 ± 15.8 and by proxy report was 75.8 ± 16.3. Self-reported fatigue score was lower in children aged 5-7 years (71.5 ± 15.9) compared to proxy-reported score (76.3 ± 15.5). Proxy and self-reported fatigue scores were similar between parents and older children. Fatigue scores were lower in association with poor asthma control and receipt of social support. Lower self-reported, but not proxy-reported, fatigue score was related to asthma severity. Lower proxy-reported, but not self-reported, fatigue score was related to the child being older and having shortness of breath. CONCLUSION: Parents underestimated the fatigue of younger children aged 5-7 years, but fatigue scores were similar between parents and older children. Both clinical and social factors are associated with fatigue in children with asthma.

Uncertainty in surrogate decision-making about end-of-life care for people with dementia: An integrative review.

AIM: To describe uncertainty in surrogate decision-making regarding end-of-life care for people with dementia using Mishel's reconceptualized uncertainty in illness theory. DESIGN: Integrative literature review using Whittemore and Knafl's approach. DATA SOURCES: PubMed, CINAHL, EMBASE, Scopus and Web of Science were searched using terms such as uncertainty/unpredictability, decision-making/advance care planning/end-of-life care planning, surrogate/family/caregiver/proxy and dementia. The search was initially conducted on 28 September 2021 and updated on 31 July 2023. REVIEW METHODS: Through systematic screening, 20 research articles were included in the analysis. Content related to uncertainty in surrogate decision-making regarding end-of-life care was extracted and analysed, focusing on the reconceptualized uncertainty in illness theory. RESULTS: First, surrogate uncertainty exists in various areas of surrogate decision-making regarding end-of-life care. Second, antecedents of surrogate uncertainty include numerous intrinsic and extrinsic factors. Third, surrogates exhibited some negative psychological responses to uncertainty but continually processed and structured their uncertainty through certain approaches, leading them to grow as decision-makers. Finally, research-based evidence on surrogates' processing of uncertainty and shifts to new life perspectives remains limited. CONCLUSION: Surrogates' uncertainty in decision-making regarding end-of-life care for people with dementia is well characterized using the reconceptualized uncertainty in illness theory. Healthcare providers should help surrogates manage their uncertainty in surrogate decision-making more constructively throughout the dementia trajectory. IMPLICATIONS FOR THE PROFESSION AND/OR PATIENT CARE: The findings highlight the importance of assessing how surrogates process uncertainty and gauging how to help them process uncertainty and transition to new life perspectives. IMPACT: This review contributes to healthcare professionals' understanding of surrogates' uncertainty in end-of-life care planning for people with dementia, especially what they are uncertain about, what influences their uncertainty and how they process it. REPORTING METHOD: This study adheres to the PRISMA reporting guidelines. PATIENT OR PUBLIC CONTRIBUTION: No patient or public contribution.

Reliability of proxy reports on patient reported outcomes measures in stroke: An updated systematic review.

OBJECTIVES: With the rising global burden of stroke-related morbidity, and increased focus on patient-centered healthcare, patient reported outcome measures (PROMs) are increasingly used to inform healthcare decision-making. Some stroke patients with cognitive or motor impairments are unable to respond to PROMs, so proxies may respond on their behalf; the reliability of which remains unclear. The aim of the study is to update a 2010 systematic review to investigate the inter-rater reliability of proxy respondents answering PROMs for stroke patients. MATERIALS AND METHODS: Studies on the reliability of proxy respondents in stroke were searched within CINAHL, Embase, PsycInfo, and WoS databases (01/07/22, 08/07/22). Fifteen studies were included for review. ICC and k-statistic were extracted for PROMs scales and categorized as poor (0.80). Bias was assessed using the CCAT. RESULTS: Five studies reported PROMs with inter-rater reliability scores ranging from 0.80. Two studies reported activities of daily living (ADLs) scores ranging from 0.41 to 0.80 and 8 studies reported quality of life (QoL) measures with scores ranging from 0.80. Subcategories of these scales included physical (ICC/k-statistic 0.41- >0.8), cognitive (ICC/k-statistic 0.40-0.80), communication (ICC/k-statistic <0.4-0.80,) and psychological (ICC/k-statistic <0.40-0.60) measures. CONCLUSIONS: Proxy respondents are reliable sources for PROM reports on physical domains in ADLs, PROMs and QoL scales. Proxy reports for measures of communication and psychological domains had greater variability in reliability scores, ranging from poor to substantial; hence, caution should be applied when interpreting proxy reports for these domains.

The use of proxy measurement of internal states in people with intellectual disability: A scoping review.

BACKGROUND: The measurement of internal states of people with an intellectual disability is challenging. Given these difficulties, proxy respondents are sometimes used. However, the literature provides some caution about the validity of the use of proxy respondents. AIMS: This review aims to identify the available evidence regarding the use of proxy respondents in the measurement of internal states of people with intellectual disabilities. METHOD: The Arksey and O'Malley scoping review framework was used for this review, refined with the Joanna Briggs Institute Methodology. RESULTS: Results indicate there are some differences in the findings reported with respect to the efficacy of the use of proxy respondents. These may be due to a number of inconsistencies and weaknesses in examinations of their utility. CONCLUSIONS: If proxy-reports are to be useful in understanding the internal states of people with intellectual disability, researchers need to address the shortcomings revealed by this review.

Proxy Variables and the Generalizability of Study Results.

When individuals self-select (or are selected) into a study based on factors that influence the outcome, conclusions may not generalize to the full population. To compensate for this, results may be adjusted, for example, by standardization on the set of common causes of participation and outcome. Although such standardization is useful in some contexts, the common causes of participation and outcome may in practice not be fully observed. Instead, the researcher may have access to one or several variables related to the common causes, that is, to proxies for the common causes. This article defines and examines different types of proxy variables and shows how these can be used to obtain generalizable study results. First of all, the researcher may exploit proxies that influence only participation or outcome but which still allow for perfect generalizability by rendering participation and outcome conditionally independent. Further, generalizability can be achieved by leveraging 2 proxies, one of which is allowed to influence participation and one of which is allowed to influence the outcome, even if participation and outcome do not become independent conditional on these. Finally, approximate generalizability may be obtained by exploiting a single proxy that does not itself influence participation or outcome.

Proxy ratings of psychological well-being in patients with primary brain tumors: A systematic review.

OBJECTIVE: This systematic review examined the agreement of proxy ratings of depression and anxiety in neuro-oncology patients. METHODS: Searches were conducted across 4 databases (MEDLINE, Embase, PsycINFO, CINAHL, and Web of Science) to identify studies that compared proxy ratings (non-health care providers) of anxiety and depression in patients with brain cancer. Methodological quality and potential risk of bias were evaluated using the Joanna Briggs Institute Critical Appraisal Checklist. RESULTS: Out of the 936 studies that were screened for inclusion, 6 were included for review. The findings were mixed in terms of whether patient and proxy ratings were accurate (e.g., deemed equivalent), with many of the selected studies suggesting moderate level of agreement for several of the selected studies and, when both depression and anxiety were included, depression ratings from proxy raters were more accurate than for anxiety. We identified important limitations across the selected articles, such as low sample size, clarity on defining proxy raters and the different instructions that proxy raters are given when asked to assess patients' mood symptoms. CONCLUSIONS: Our findings suggest that proxy ratings of depression and anxiety should be interpreted with caution. While there is some agreement in proxy and patients with brain cancer ratings of depression and anxiety (greater agreement for depression), future work should recruit larger samples, while also remaining mindful of defining proxy raters and the instructions given in collecting these ratings.

Projection of Care Partners' Preferences in the Proxy Assessments of Everyday Preferences for Persons With Cognitive Impairment.

OBJECTIVE: Bias in surrogate decision-making can occur when proxy decision-makers overestimate the degree to which their preferences are shared by others, resulting in a projection of their beliefs onto others. The purpose of this study is to assess projection of care partners' preferences onto surrogate assessments of everyday preferences for persons with cognitive impairment (CI) and to address clinical and demographic factors as predictors of projection. METHODS: The sample included 116 dyads of persons with CI (Clinical Dementia Rating Scale score ≥ 0.5) and their care partners. The Preferences for Everyday Living Inventory (PELI) was used to assess importance of preferences among persons with CI. Care partners completed two separate PELI assessments: one from the perspective of the persons with CI (i.e., acting as a surrogate decision-maker) and one from their own perspective. To assess for projection of care partners' preferences onto surrogate assessments of preferences for persons with CI, two-step regression with multivariable-adjusted general linear models was used. RESULTS: Significant projection was noted within the PELI domains of autonomous choice, personal growth, and keeping a routine (p < 0.005). More significant cognitive impairment was associated with increased projection within the PELI domains of autonomous choice and personal growth (p < 0.05). CONCLUSION: The results of this study suggest that projection of care partners' own preferences may be a significant source of bias in proxy decision-making regarding everyday preferences for persons with CI, particularly for those with more significant CI.

Validation of the parent-proxy version of the pediatric Charcot-Marie-Tooth disease quality of life instrument for children aged 0-7 years.

OBJECTIVE: To evaluate the parent-proxy version of the pediatric Charcot Marie Tooth specific quality of life (pCMT-QOL) outcome instrument for children aged 7 or younger with CMT. We have previously developed and validated the direct-report pCMT-QOL for children aged 8-18 years and a parent proxy version of the instrument for children 8-18 years old. There is currently no CMT-QOL outcome measure for children aged 0-7 years old. METHODS: Testing was conducted in parents or caregivers of children aged 0-7 years old with CMT evaluated at participating INC sites from the USA, United Kingdom, and Australia. The development of the instrument was iterative, involving identification of relevant domains, item pool generation, prospective pilot testing and clinical assessments, structured focus group interviews, and psychometric testing. The parent-proxy instrument was validated rigorously by examining previously identified domains and undergoing psychometric tests for children aged 0-7. RESULTS: The parent-proxy pCMT-QOL working versions were administered to 128 parents/caregivers of children aged 0-7 years old between 2010 and 2016. The resulting data underwent rigorous psychometric analysis, including factor analysis, internal consistency, and convergent validity, and longitudinal analysis to develop the final parent-proxy version of the pCMT-QOL outcome measure for children aged 0-7 years old. CONCLUSIONS: The parent-proxy version of the pCMT-QOL outcome measure, known as the pCMT-QOL (0-7 years parent-proxy) is a valid and sensitive proxy measure of health-related QOL for children aged 0-7 years with CMT.

Differences in health-related quality of life in children with chronic kidney disease as reported by children and parent proxies.

BACKGROUND: Correlation between reports of children and parent for health-related quality of life (HRQOL) is not well studied. This study aims to assess the degree of agreement between child self- and parent proxy-rated HRQOL and to identify factors associated with discordance at baseline and during follow-up in Taiwanese children with chronic kidney disease (CKD). METHODS: This study includes pediatric patients aged 5-18 years with confirmed CKD. Participants completed the generic version of the Pediatric Quality of Life Inventory (PedsQL) at baseline and every 6 months during follow-up. Child-parent agreement on HRQOL reports was assessed using intraclass correlation coefficient (ICC). Multivariate regression models were used to determine factors associated with child-parent discordance. RESULTS: Of the 112 child-parent dyads included in the analysis, 97 dyads with 640 patient visits were assessed in 4.5 years. Children reported higher total scores on the physical and psychosocial domains as compared to their parent proxies. ICC was low (< 0.5) for the psychosocial domain and moderate for the physical health domain at initial assessment and slightly increased for the physical health (0.62) and for school functioning (0.51) during follow-up. Development of mineral bone disorder/anemia (ß, 11.75 [3.77-19.72]) and proteinuria (ß, 8.48 [1.15-15.81]) in the follow-up were associated with increased discordance in school functioning, and fathers with chronic disease were associated with increased discordance in social functioning (ß, 4.21 [0.68-7.74]). CONCLUSIONS: Parent proxy consistently estimated lower PedsQL score compared to self-reports of children. Child self-rated psychosocial health domains should be evaluated whenever possible to better elucidate treatment outcome over time. A higher resolution version of the Graphical abstract is available as Supplementary information.

Illness-related parental stress and quality of life in children with kidney diseases.

BACKGROUND: This cross-sectional study investigated quality of life (QoL) and illness-related parental stress in children with kidney diseases by (1) comparing mean levels of these two variables between several kidney disease categories; (2) exploring correlations between QoL and parental stress; and (3) describing which disease category reports lowest QoL and highest parental stress. METHODS: We included 295 patients with a kidney disease (0-18 years) and their parents, followed at 6 reference centers for pediatric nephrology. Children's QoL was assessed by the PedsQL™ 4.0 Generic Core Scales, and illness-related stress by the Pediatric Inventory for Parents. All patients were divided into 5 kidney disease categories according to the multidisciplinary care program criteria prescribed by the Belgian authorities: (1) structural kidney diseases, (2) tubulopathies and metabolic diseases, (3) nephrotic syndrome, (4) acquired diseases with proteinuria and hypertension, and (5) kidney transplantation. RESULTS: Child self-reports showed no differences in QoL between kidney disease categories, in contrast to parent proxy reports. Parents of transplant patients reported lower QoL in their child and more parental stress compared with the 4 non-transplant categories. QoL and parental stress were negatively correlated. Lowest QoL and highest parental stress scores were mainly found in transplant patients. CONCLUSIONS: This study showed lower QoL and higher parental stress in pediatric transplant patients compared with non-transplants, based on parent reports. Higher parental stress is associated with worse QoL in the child. These results highlight the importance of multidisciplinary care for children with kidney diseases, with special attention to transplant patients and their parents. A higher resolution version of the Graphical abstract is available as Supplementary information.

Alzheimer's, Advance Directives, and Interpretive Authority.

Philosophers have debated whether the advance directives of Alzheimer's patients should be enforced, even if patients seem content in their demented state. The debate raises deep questions about the nature of human autonomy and personal identity. But it tends to proceed on the assumption that the advance directive's terms are clear, whereas in practice they are often vague or ambiguous, requiring the patient's healthcare proxy to make difficult judgment calls. This practical wrinkle raises its own, distinct but related, philosophical question: what criteria may the proxy bring to bear when making such interpretive judgments on which the patient's life may depend? After defending a general policy of enforcing advance directives on normative (rather than metaphysical) grounds, I argue that when advance directives are vague, a patient's proxy may permissibly make her own fresh evaluation of the patient's life as a whole and, in so doing, consider how the patient's character as a demented person contributes or fails to contribute to that life.

Against an Inflexible, Prioritized List for Default Surrogate Selection.

AbstractSurrogate selection can be extremely consequential for patients. Most surrogates are selected by default, so we should care about whether legal provisions for default surrogate selections are ethically justified. Most U.S. states use an inflexible, prioritized list of relationships, that is, a hierarchical list where eligible classes of higher-ranked individuals must be selected before lower-ranked individuals. I argue that while some inflexible, prioritized lists may roughly reflect the order that many patients would select, there is a significant minority that inflexible lists systematically disempower. This is morally unacceptable given the availability of less morally problematic alternatives. One alternative is a flexible, prioritized list, which provides conditions for lower-ranked individuals to be selected ahead of higher-ranked ones. I argue that since all the U.S. states that currently have an inflexible, prioritized list systematically disempower a significant proportion of their residents, they have good reason to adopt a flexible, prioritized list instead. Furthermore, the Universal Law Commission currently recommends that states adopt an inflexible, prioritized list, so they have good reason to change their recommendation.

When Should Providers Defer versus Impose Their Views?

AbstractThis piece discusses perhaps the most agonizing ethical decision ethics consultants and other providers encounter. This is the extent to which providers should defer decisions to patients or to their proxy decision makers as opposed to imposing their own views as to what they think is ethically right. It discusses the most difficult issues these providers may encounter, especially when they wish to depart from authoritative bodies' standards or guidelines, and it presents initial steps providers may take to help patients and their families work together to resolve these dilemmas more harmoniously. It highlights how providers may inadvertently impose flawed biases on patients and families. Finally, it discusses how providers should take initiative with both parties to offer to help appeal when these avenues already exist and seek to establish the appellate procedures when they are absent.

Reconsidering Capacity to Appoint a Healthcare Proxy.

Clinicians are often called upon to assess the capacity of a patient to appoint a healthcare agent. Although a consensus has emerged that the standard for such assessment should differ from that for capacity to render specific healthcare decisions, exactly what standard should be employed remains unsettled and differs by jurisdiction. The current models in use draw heavily upon analogous methods used in clinical assessment, such as the "four skills" approach. This essay proposes an alternative model that relies upon categorization and sliding scale risk assessment that can be used to determine to how much scrutiny the proxy appointment should be subjected and how much certainty of accuracy should be required in order to maximize the patient's autonomy and ensure that her underlying wishes are met.

Exploring the social care-related quality of life outcomes of adults with intellectual disabilities through the use of Talking Mats®.

BACKGROUND: Views on social care related quality of life (SCRQol) are typically gathered directly from individuals or from proxy informants. Easy-read options are available, although such methods may not be accessible to those with greater communication challenges. METHODS: Fifteen people with intellectual disabilities were interviewed about their SCRQoL using Talking Mats® (TM). A descriptive approach was taken to exploring what people had to say. RESULTS: Participants reported high levels of satisfaction with living conditions and personal care routines. They were mostly satisfied with the level of choice and control they had and with their carers. CONCLUSIONS: People with intellectual disabilities who have greater communication challenges can make meaningful contributions to the assessment of the care they receive. Talking Mats is one tool which can be used to support people with intellectual disabilities to give their views.

Role of Proxy Respondents in International Stroke Research: Experience of the INTERSTROKE Study.

INTRODUCTION: Measuring patient-reported information in stroke research is challenging. To overcome this, use of proxy respondents is often a necessary strategy. In this study, we report on use and effect of proxy respondents on patient case-mix in a large international epidemiologic stroke study (INTERSTROKE). METHODS: This was a cross-sectional study of 13,458 cases of acute first stroke in 32 countries. A standardized study questionnaire recording behavioural cardiovascular risk factors was administered to the patient, and if unable to communicate adequately, a valid proxy, or both. We used logistic regression to evaluate the association of age, sex, education, occupation, stroke severity, and region with need for proxy respondent, and report odds ratio (OR) with 95% confidence interval (CI). RESULTS: Among 13,458 participants with acute stroke, questionnaires were completed by patients alone in 41.4% (n = 5,573), combination of patient and proxy together in 21.7% (n = 2,918), and proxy alone in 36.9% (n = 4,967). Use of proxy alone was greater in participants with severe stroke (4.7% with modified-Rankin score of 0 vs. 80.5% in those with score 5; OR 187.13; 95% CI: 119.61-308.22), older persons (43.8% of those aged 80 years and over vs. 33.2% of those aged less than 40 years; age per decade OR 1.09; 95% CI: 1.06-1.12), women (40.7% vs. 34.3% of men; OR 1.32 95% CI: 1.22-1.43), and those less educated (58.9% of those never educated vs. 25.7% of those who attended third level education; OR 7.84; 95% CI: 6.78-9.08). CONCLUSION: Use of proxy respondents enhances the generalizability of international research studies of stroke, by increasing representation of women, patients with severe stroke, older age, and lower education.