Sex and gender differences in quality of life and related domains for individuals with adult acquired lower-limb amputation: a scoping review

Abstract Purpose To understand what is known about sex and gender differences in quality of life (QoL) and related domains for individuals with an adult acquired lower limb amputation (LLA). Methods A computer-assisted literature search of four online databases was completed. Articles were included if they incorporated sex or gender as part of their data analysis with a focus on QoL-related domains. Data were analyzed using descriptive numerical analysis and thematic analysis. Results One hundred and eleven articles were included in this review. Women were under-represented across studies, with most of the participants being men. No articles described the inclusion of trans or non-binary persons. Differences by sex or gender were reported by 66 articles. Articles reporting on gender seldom provided descriptions of how gender was defined. Overall, women/females seemed to have worse outcomes in terms of prosthesis-related outcomes, mental health, and return to occupations. Conclusion Articles included in this review were not clear with how gender was defined. In order for more targeted interventions that account for sex and gender differences, studies need to be more forthcoming about how they use and define gender. Future research should seek to include gender non-conforming participants to identify additional needs. Implications for rehabilitation Sex and gender are important constructs that influence outcomes following lower limb amputation. Rehabilitation professionals should consider sex and gender-specific outcomes when tailoring programs to ensure ethical clinical care.


Introduction
In the field of disability and rehabilitation, there is growing recognition of the importance of how and to what extent sex and gender may influence outcomes [1]. Sex and gender are terms that are often used interchangeably but are in fact two separate entities. According to the Canadian Institutes of Health Research (CIHR), sex is understood as a set of physical and physiological attributes that are biological in nature [2]. These biological attributes are usually categorized as male or female, but variation is possible in how they are expressed [2]. Gender encompasses the socially constructed roles, behaviours, expressions and identities of girls and boys, women and men, as well as those who identify as gender diverse. Gender influences how people perceive themselves and each other, how they act and interact, and the distribution of power and resources in society [2]. While substantial strides have been made to conduct sex and gender-based analyses in the field of disability and rehabilitation (e.g., traumatic brain injury [3]; Alzheimer's Disease [4]), several gaps still exist for certain disability groups. Based on a preliminary scan of the literature, this is particularly evident in the field of lower limb amputation (LLA).
LLA is a life-changing event that has recently been deemed the leading cause of global disability [5]. Major LLA (ankle and above) may occur due to trauma, cancer, infection and/or complications from diabetes and/or peripheral vascular disease [6][7][8]. In addition to mobility impairments, persons with major LLA experience a variety of health challenges, which include phantom limb pain, low back pain, osteoarthritis, heterotopic ossification, cardiovascular disease, anxiety and depression [9][10][11][12][13]. It is therefore not surprising that undergoing LLA has been shown to negatively influence quality of life (QoL) [14,15].
Broadly, QoL may be conceptualized as an individual's perception of their position in life within the context of which they live (i.e., culture and value systems) as well as in relation to their goals, expectations, standards and concerns [16]. While various measures of QoL have been used in the LLA population to better understand their experiences living with this condition, it is not well understood how sex and gender influence QoL. Further exploration from a sex and gender lens is warranted since there are known differences in risk factors, health outcomes and experiences post-LLA [17,18]. From an epidemiological perspective, males are more likely to be at risk for an amputation [19,20]. While the reasons for this are not well understood [21], preliminary evidence suggests that males (sex) are less likely to seek timely treatment for a diabetic foot [22] and are more likely to have risk factors associated with amputation such as vascular disease and smoking [20].
Differences in experiences and perceptions of overall wellbeing by gender in LLA have also been reported [18,21]. Women are more likely to experience body image issues, are more likely to live alone as well as are less likely to receive a prosthesis than men [18,23]. However, more critical appraisals of how sex and gender are addressed in the LLA literature are needed to improve the health and QoL of this patient population. This is particularly important for the psychosocial factors of QoL (mental/emotional health, community participation, etc.) as the majority of research in this population has focused on function and mobility [14]. Furthermore, identifying variables that interact with sex and gender will help those developing interventions to better understand under which circumstances specific aspects of the intervention will work and for whom it will work [24].
To advance knowledge of how sex and gender influence QoL post-LLA, a scoping review was undertaken to map the literature [25]. Specifically, we aimed to identify studies that reported on sex and/or gender differences in QoL in people with LLA or that contextualized their findings with respect to sex and gender in the case where only one sex or gender is studied. In addition to extracting key findings as it relates to sex and gender, we also mapped which QoL outcome measures or domains have been studied.

Materials & methods
For the present review, the scoping review framework described by Levac et al. [25] was used. This framework builds on the original guidelines developed by Arksey et al. [26] by making specific recommendations and clarifications for each of the five stages [25]. The use of a scoping review is appropriate for examining issues of sex and gender as it relates to QoL and related domains post-LLA since scoping reviews are designed to examine the extent, range and nature of research activity as well identify gaps in the existing literature [26]. This review followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews guidelines [27]. Finally, the protocol for this review was registered with OSF Registries (https://osf.io/89czj).

Stage 1 -identifying research questions
The guiding research question for this review was: What is known about sex and gender differences in QoL and related domains for individuals living with an adult acquired LLA? The sub-questions of interest were: 1) How are sex and gender defined/used in the available LLA literature?; 2) What domains and/or tools have been used to measure QoL and related domains with a sex and gender lens?; and 3) What are the major findings with respect to QoL and related domains as it pertains to sex and gender? By addressing these objectives, a better understanding of what gaps exist in the literature with respect to sex and gender analysis will be obtained and can be used to inform future research in the area.

Stage 2 -identifying relevant studies
A computer-assisted search of the literature on articles published from 1 January 2000 to 13 July 2020 was conducted. The date of 1 January 2000 was selected as care for individuals with LLA has advanced greatly over time with respect to surgical interventions, rehabilitation, and prosthetic care [28]. Of particular importance is that both rehabilitation [29] and prosthetic care [30] have been shown to impact QoL. Therefore, including studies from the past 20 years would ensure that the most up-to-date information is included in this scoping review. Four online databases were used to search the literature: Medline (OVID), EMBASE (OVID), CINAHL (EBSCO) and PsycINFO (OVID). In addition, key journals (e.g., Journal of Prosthetics & Orthotics, International Journal of Prosthetics & Orthotics, Disability and Rehabilitation) were manually searched to ensure that all relevant articles were identified. For knowledge syntheses (e.g., systematic reviews) that met the inclusion criteria, their reference sections were hand searched for articles that met the inclusion criteria and were not captured by the database searches. Grey literature (e.g., theses and dissertations, clinical trial protocols, key agency reports) was included to avoid publication bias and ensure thoroughness of the review. After a discussion with an expert librarian at the University of Toronto, two key concepts were used to search for relevant studies. First, sex and gender and the associated key terms were included. MESH terms for this concept included: sex characteristics; sex factors; sex distribution or sex ratio; gender identity and gender. The second concept was lower limb loss and associated key terms. These included but were not limited to: lower extremity; amputees; amputation or disarticulation; and prosthetics. For a sample of the search strategy for Medline (OVID) please refer to Appendix A. This search strategy was adopted for the remaining databases.

Stage 3 -selection of relevant studies
This study sought to include articles that covered overall QoL and related domains. From previous work done in similar reviews on QoL in this population, potential domains relevant to QoL included psychosocial wellbeing, physical functioning, environmental factors and community participation [14,30]. While these are the most commonly reported domains, other areas were included as they arose (e.g., discharge destination, sexual health). Included studies must have met the following eligibility criteria: 1) any study design (e.g., cross-sectional, qualitative, knowledge syntheses), except for clinical case studies, that included sex and/or gender as part of their analysis of the data (e.g., females have higher QoL than men); 2) had at least one individual with a major LLA (i.e., ankle disarticulation and above) who were older than 18 years of age at the time of their amputation, and their amputation was due to trauma (i.e., accident, burns), dysvascular causes (i.e., diabetes complications or peripheral vascular disease), cancer or infection; 3) articles published in any language since 1 January 2000; and 4) not an opinion piece (e.g., narrative review article) or conference publication.
The resulting list of articles (n ¼ 3,110) was de-duplicated with Endnote using the technique described by Bramer and colleagues [34]. This systematic guideline for removing duplicates in EndNote helped to streamline the article review process. To facilitate the screening process, the systematic reviews software production tool, Covidence, was used. Two members of the study team (SRC and AV) were responsible for the title and abstract screening. A pilot test was conducted where each reviewer independently screened 10 articles after which the agreement was compared. Interrater agreement was 90% which was sufficient to continue the title and abstract screening [35]. All titles and abstracts were screened by both reviewers.
Once the title and abstract screening was completed, full-text screening began. An initial pilot of 10 articles was conducted by the same two reviewers (SRC and AV) and an interrater agreement of 80% was achieved. Each of the full-text articles was screened by both reviewers. Data extraction began once the fulltext screening was completed. Two knowledge syntheses articles that met all the inclusion criteria were not included in data extraction. Instead, their references were searched for articles that met the inclusion criteria.

Stage 4 -charting the data
Relevant data (i.e., study information, participant characteristics, outcome measures used, and results pertaining to sex and gender) were extracted using Microsoft Excel. Two members of the research team (SRC and AV) were responsible for extracting the relevant data. First, a pilot extraction was conducted on 10 randomly selected studies. The two reviewers independently collected the data from these 10 studies and met to compare findings. An acceptable level of agreement was obtained between the two reviewers (approximately 90%) [35]. Disagreements and clarifications were discussed in person until consensus was reached. Following the pilot, the remaining articles were split between the two reviewers and full data extraction was completed. For the purposes of this review, a quality assessment of the articles was not conducted. This is consistent with scoping review methodology as the purpose of this type of review is to provide an overview of existing literature regardless of quality [36]. Thus, the articles findings were not classified based on quality and reported results reflect the terminology used by the original authors of the included studies.

Stage 5 -collating, summarizing and reporting the results
Results pertaining only to sex and gender were compared across studies. To analyze the extracted data two approaches were used: a descriptive numerical analysis and thematic analysis. In addition to these analyses, the extracted data were used to look for gaps in the literature and develop suggestions for future research.

Overall QoL
Overall QoL differences by gender were described by six studies utilizing the Short-Form-36 (SF-36) [37,57,74] and the World Health Organization Quality of Life Questionnaire (WHO QOL-BREF) [74,121,123,144]. Reported findings were mixed with respect to whether men or women had higher scores on either measure. In one study, SF-36 sub-scores for role physical and role emotional were reported to be higher in women than men in a dysvascular population [74] however, vitality scores were reported to be higher for men than women (reason for amputation not reported) in a second study [37]. On the WHO QOL-BREF, Cox et al. [74] reported that women had better overall QoL scores. However, two others reported that men had higher scores on the overall health, physical health, social relationship domains in a mostly traumatic LLA population [121] and better psychological QoL in persons with dysvascular LLA [144]. An additional study with a mostly dysvascular population highlighted that only men had lower scores on physical health and social relationship domains compared to normative data [123]. No studies reported differences in QoLusing subjective measures.

Physical health
Based on the included articles, various aspects of physical health such as pain, comorbidities, length of stay, and level of amputation were investigated. For articles that reported on sex, Bosmans et al. [79] noted that there was a protective effect of being male on the incidence of phantom limb pain in a mostly dysvascular LLA population and Ebrahimzadeh et al. [107] found rates of phantom limb pain and sensation to be 87% and 45.1% respectively in a traumatic LLA cohort. This article also reported rates of residual limb pain, chronic back pain and contralateral knee pain (64.5%, 61.2% and 54.8% respectively) [107]. However, articles that reported on gender suggested mixed findings with respect to phantom limb pain. While both studies had all trauma or mostly trauma participants, one study reported men had significantly higher odds of persistent/recurrent pain compared to women [63], while another suggested that the prevalence of phantom limb pain was higher in women than men [43].  � Men also were more likely to be fit than women for TT and TF (OR ¼  With respect to comorbidities, three articles reported on sex [47, 117,130] and one reported on gender [97]. Two studies with predominantly dysvascular cohorts claimed females had a higher risk for wound complications [47] as well as lumbago, rheumatoid arthritis or osteoarthritis [117]. One study claimed that men with amputation had greater weight gain if they had a transtibial or transfemoral amputation compared to partial foot [130]. For the study reporting on gender, skin problems (e.g., pain, stinging) were claimed to have a larger impact on day-to-day life in women compared to men in a mostly traumatic LLA population [97].
Level of amputation was investigated by two articles, where one reported on gender [131] and the other on sex [68]. Both of these studies did not report the reason for amputation but reported that female sex and women made up a significantly larger proportion of participants with an above-knee amputation [68,131]. These studies also claimed that being a woman was associated with a higher risk of having a transfemoral amputation [68,131].
Two studies claimed that length of stay in acute inpatient care and inpatient rehabilitation differed by sex [71] and gender [55]. Kurichi et al. [71] reported that females had a lower mean inpatient length of stay compared to males while Davie-Smith et al.
[55] noted that women had a significantly longer rehabilitation length of stay compared to men. Both articles reported on samples comprised mostly or entirely of dysvascular participants.

Prosthesis-related outcomes
Studies reporting on gender and prosthesis use claimed mixed results. These articles had all or mostly dysvascular participants. Three found that women use their prosthesis less often [67, 94, 119] while one suggested that more prosthesis use was associated with being a woman [123]. One additional article reported on prosthesis use and sex [137]. In a cohort of mostly trauma participants, the authors suggested that number of hours of prosthesis use was correlated with functional satisfaction for males, but all levels of satisfaction (total, functional, aesthetic and weight) for females [137]. The number of hours per day of use was also potentially connected with positive body image for males but not for females in the same study by Murray et al. [137].
Prosthesis prescription was posited to be associated with gender in three studies [18,55,82]. These articles reported that men, regardless of the level of amputation, were more often fitted with a prosthesis compared to women in mostly dysvascular cohorts [18,55,82]. However, in a study by Highsmith et al. [59], the authors reported prosthesis cover prescription tended to favour women, regardless of the level of amputation. An additional study suggested that the durability of cosmesis was also impacted by gender, where women were less satisfied with durability compared to men [128]. Further, the included articles suggested that satisfaction with prosthesis differed by sex and gender. In terms of sex, a study with mostly non-traumatic participants claimed that lower levels of body image disturbance were correlated with higher levels of functional satisfaction for males but females also required aesthetic and weight satisfaction in addition to function [137]. In another study looking at mostly trauma participants, gender appeared to have an influence on satisfaction with prosthesis fit, comfort and appearance with levels being higher in men than women [89].

Mobility/function
The included articles that described differences in sex and gender on mobility and/or function used a variety of different outcome measures. The most common tools included the Two-minute walk test (2MWT) [147], and the Activities-specific balance confidence (ABC) scale [148]. Few studies used population-specific measures of mobility such as the Locomotor Capability Index (LCI) [149,150], K-classification levels [151], Trinity Amputee and Prosthesis Experience Scale (TAPES) [152], Prosthetic evaluation questionnaire (PEQ) [153] and Prosthetic limb users survey of mobility (PLUS-M) [154].
Generic measures. Gender differences were identified when results of generic measures were captured based on the included articles. The functional level at the end of rehabilitation was found to be higher for men than women for persons with dysvascular LLA in a study by Knezevic et al. [56]. Two studies reported the opposite in the dysvascular population, with women having better mobility according to the Functional Independence Measure [74] and a 45-meter walk-test [46].
Sex differences were also reported by the included articles in relation to these parameters. Based on the results reported by several studies, being male was associated with better 2MWT scores [40,114], better ambulatory status [49,54,87], mobility [66, 116,143] and activities of daily living scores [68] in persons with LLA due to trauma, osteoarthritis and dysvascular causes. One article reported conflicting findings, with males having longer Timed-Up-And-Go scores compared to females, suggesting greater mobility impairment [72]. With respect to falls, being female was potentially associated with a higher risk of falls and fall-related injury in persons with vascular LLA in two studies [60,129]. Balance was also measured using the ABC scale in two studies, with results claiming that males have higher balance confidence compared to females [100,136].

Lower-limb amputation specific measures.
For gender, five studies used population-specific measures. For these studies, men reported better K-levels than women at discharge in persons with dyvascular LLA [139], while a study looking at only trauma LLA found better physical capacity for men according to the TAPES [43]. Similar results were reported with respect to the Locomotor Capability Index in three studies. Specifically, one study claimed that men tended to report less worsening in this score over time [119] and had higher scores than women; suggesting better mobility in both trauma and dysvascular cohorts [43,139]. With respect to sex, Kelly et al. [100] claimed that scores on the PLUS-M were lower for women than men in a mostly trauma cohort. However, when using the PEQ, Wong et al. [40] posited that male sex was associated with lower perceived functional ability.

Mental health
Mental health outcome measures (e.g., social adjustment, positive/negative affect, depression, post-traumatic stress disorder) varied between studies. As with mobility/function, most of the measures used were not specific to LLA. These included the Positive Negative Affect Schedule (PANAS) [155], Beck Depression Inventory [156,157], State-trait anxiety inventory [158], Functional comorbidity index [159], Self-reported questionnaire-20 [160], Psychological adjustment scale [161], the Brief COPE scale [162], and the Post-traumatic growth inventory [163]. The only population-specific measures used that determined sex and gender differences were the TAPES. Three additional articles conducted qualitative analyses with two using interpretive phenomenological analysis [39,118] and one using thematic analysis [126].
Based on the included articles, gender differences were reported when using either generic measures or population-specific (limb loss) measures. In particular, four studies suggested, that compared to men, women in mostly dyvascular cohorts had higher negative affect [102,144], poorer general adjustment [144] and a higher rate of mental health problems [73], including greater symptoms of depression [144] and moderate-to-high posttraumatic growth scores [118]. However, there is conflicting evidence when it comes to social adjustment. In two articles that utilized the TAPES, one study in a mostly dysvascular population reported that compared to men, women tended to have poorer social adjustment [144], while another with mostly traumatic LLA found women had higher psychosocial adjustment scores when Locomotor Capability Index and phantom limb pain scores were held constant [43]. For the qualitative studies, the notion of body image was most often described by women from mostly dysvascular LLA cohorts [43,118,126]. Although two studies were all women, one indicated that appearance is considered to be a "women specific challenge" [118].
When looking at sex differences, one study of females in a mostly vascular LLA cohort tended to have a higher risk of anxiety or panic disorder [117]. In two studies with all males with traumatic LLA, almost half of participants had symptoms of post-traumatic stress disorder [105] and those with psychological problems tended to have more phantom limb pain [107]. In terms of adjustment, differences depended on the type of adjustment and the measurement tool used. For example, in two studies looking at mostly traumatic LLA, one found that females had higher scores than males on psychological adjustment [51] while another (that used the TAPES) found that males were most socially adjusted [69].

Occupations
The articles included in this review reported that engagement in physical activity was potentially influenced by gender. Of the four studies that reported differences, only one reported a reason for amputation (mostly non-traumatic) [38]. Rates of physical activity were suggested to be higher in men than in women in two studies [38,91] with men reporting higher levels of high energy activities in an additional study [125]. However, in two studies, women were more likely to complete more moderate levels of activities [125] and reported higher daily activity levels than men [138]. Employment was also potentially associated with gender in two studies, with men having a higher rate of employment following LLA [91,112]. Only one article reported sex differences in return to driving in a mostly dyvascular cohort [85]. In this study, the authors claimed that males were more likely to return to driving following their amputation [85].

Sexual health
The two articles that reported differences in sexual health reported on sex rather than gender [48,113]. Sexually active females reported greater self-consciousness than their male counterparts (38% versus 25% respectively) in a study by Woods et al. [48]. In the same article, non-sensuality was reported to be a concern for both males and females, but females were reported to have higher rates compared to males (63% versus 45%) [48]. One additional study focused on males only and claimed that there were no correlations between having a prosthesis, experiencing pain or depression and sexual functioning [113]. This article also suggested that only half of the men in their study were moderatel to extremely satisfied with their ability and intensity of orgasm [113]. This was reflected in the study by Woods et al. [48] where the authors claimed that high rates of male sexual dysfunction were associated with impotence (70%) and premature ejaculation (63%) [48].

Discharge destination
Only two articles reported on sex and gender differences in discharge destination. Both studies focused on dysvascular populations, with one reporting on sex [134] and the other on gender [133]. Both articles found that female sex or being a woman was potentially associated with being discharged to institutional care [133,134].

Discussion
The aim of this scoping review was to highlight the differences that exist between sex and gender for QoL and related domains for persons with an adult acquired LLA. Overall, there were limited details on how these constructs were measured. Thus, while identified studies showed differences across QoL and related domains, it is difficult to parse out whether the differences were by sex or gender. Due to the uncertainty of how these constructs were measured in the included studies, the main findings of differences identified in this review should be interpreted with caution. Although there were mixed findings for many of the domains, it appears that female sex and individuals who identified as a woman experience a greater impact of their LLA on QoL related domains compared to male sex and men. Most notably, potential areas of differences were reported for prothesis-related outcomes, mental health, return to work and driving.
Articles that reported on prosthesis-related outcomes reported findings by sex and gender. The results of the included studies suggest that less prosthesis prescription and lower prosthesis satisfaction for both female sex and being a woman (gender) compared to male sex and men (gender). It is possible that this discrepancy in prosthesis prescription may be due to the increased risk of a higher level of amputation for female sex and women [68,131]. In general, above knee prostheses tend to be heavy and difficult to use. From a biological perspective, females tend to be smaller and may not have sufficient strength to effectively use this type of device. Concerns with weight were also identified from the patient perspective, where the weight of the prosthesis was a more important factor for prosthesis satisfaction in women than in men [137]. In addition, comorbidities may also play a role in prosthesis prescription as it was suggested that women tended to have higher rates of comorbidities that may impact prosthesis prescription (e.g., osteoarthritis [117]). These biological differences may result in clinician hesitancy to prescribe women this type of prosthesis due to safety concerns (e.g., increased risk of fatigue resulting in falls). More work is needed to understand why prescription rates between sexes and genders are different.
Issues of mental health and its associations with sex and gender were described. Overall, the findings from the included studies suggested that compared to men and males, women and females experienced a higher rate of mental health problems including depression and anxiety [73,117,144]. However, none of the studies that reported these differences provided potential explanations for why women are experiencing increased symptoms of mental health challenges. One potential explanation may be related to issues of body image. Three qualitative studies included in this review reported that challenges with body image were most commonly discussed by women [43,118,126]. It has been shown that body image in women impacts levels of depression and anxiety [164]. It is important that these sex and gender considerations are taken into account moving forward if rehabilitation aims to address these mental health concerns in women.
Finally, we identified in this review that studies on return to work and return to driving accounted for sex and gender; albeit there is minimal work in this domain. The included studies claimed that men (gender) were more likely to return to work [91,112] and males (sex) were also more likely to return to driving following their LLA [85]. Unfortunately, the authors of these studies did not provide reasoning for why sex and gender disparities exist for returning to work and driving. With respect to employment, potential reasons include differences in prosthesis prescription and level of amputation. Those with a prosthesis [165] and those with a lower level of amputation (e.g., below the knee) [166,167] were found to be potentially more likely to return to work. There may be a relationship between return to work and return to driving. Those who return to work may also be more likely to want to return to driving in order to continue working. For example, in persons with a spinal cord injury, those who had the ability to drive were also more likely to be employed [168]. Since women are less likely to return to work, it is possible that they do not see the need to return to driving and therefore, choose not to undergo the relicensing process. However, research is needed to explore this potential connection between gender, return to work and driving following LLA as none of the included studies in this review described such a phenomenon.
In addition to these domains, more work is needed in other areas such as sexual health and sexuality [169,170]. Only two studies that were included in this review suggested differences by sex for sexual health [48,113]. The limited findings on sexual health are in line with a recent systematic review on sexuality and sexual health in persons with limb loss [171]. The authors of this review highlight that more targeted efforts are needed to better understand sexuality and sexual health in women since most of the literature was focused on males/men [171]. It is important for researchers who are exploring sexuality and sexual health in LLA to take into account sex and gender when identifying challenges faced in this population in order to develop interventions that are sex and gender-specific. Not only do biological differences exist by sex, but there are also more social aspects of sexuality and sexual health that may come into play. For example, the more physical aspects of sexual health (e.g., sexual performance) may be more difficult for men, as links have been shown between sexual functioning and men's identities [172]. On the other hand, there are societal implications that women must face that are not the same as in men, including issues of sexual pleasure [173]. These aspects need to be explored in order to provide more nuanced interventions for improving sexuality and sexual health in persons with LLA by sex and gender.
Interestingly, mixed findings for objective QoL measures were reported by the included articles. While none of the studies reported differences by gender in overall QoL scores on the SF-36, specific sub-domains (e.g., role physical or role emotional), were posited to be significantly different by gender. Furthermore, more differences were identified for sub-domains on the WHOQOL-BREF, rather than the overall score (three studies compared to one). This is important as it highlights how the subscales of these generic QoL measures appear to be sensitive to differences in the LLA population, but not the overall scores. However, it is a well-known issue with generic measures that they fail to capture population-specific issues such as satisfaction with mobility aids, accessibility and transportation which have been shown to influence community participation and thus the quality of life [88,174]. This may account for the discrepancy of gender findings and the lack of significant findings for overall scores on these generic measures.
In this review, we identified an important limitation in the existing research in the reporting of sex and gender. Only four of 69 articles made clear how they defined the term gender. All four of these articles identified that gender was collected by indicating if the participant was a man or woman. While the rest of the 69 articles did not provide a definition, results were only presented using the same man/woman binary. Furthermore, no information was presented regarding the inclusion of trans or non-binary individuals. Taken together, these findings are indicative of measurement issues when it comes to collecting gender. It is possible that the literature included in this review may be using the terms sex and gender interchangeably. However, this should be understood with the caveat that most of the included articles were not clear on how gender was collected. For example, whether participants were asked more specific questions of their gender, but no responses were received or if participants were not provided the opportunity to give a response outside of the man/woman binary was not made explicit. It is also unclear whether binary trans participants (i.e., trans individuals who identify with binary gender identities) were included.
To ensure a clear understanding of sex and gender differences in LLA QoL research, it is pertinent to understand how participants are being asked about their gender. If participants were asked to provide their gender with the options "male or female", there is no way to understand how someone who identifies as non-binary would answer this question [175]. It is possible that these participants may indicate their sex assigned at birth, choose the option that they find the least offensive or choose to abandon the survey altogether [175]. Phrasing questions about gender this way may even be difficult for binary trans participants since it is unclear whether researchers are asking about their gender identity or sex assigned at birth [175]. Currently, there are no validated and reliable measures that exist to collect gender identities that are inclusive of trans and nonbinary participants [176]. However, this should not excuse research from providing participants with a broader range of gender identity options or allowing participants to self-identify their gender to ensure ethically responsible research [175]. In order to ensure that gender is captured properly, future work should seek to make a broad range of gender options available to participants to choose from. This way, participants will be able to self-report their gender identity with the most accuracy. Furthermore, researchers should be explicit with how participants are asked about their gender, in order for those reviewing the literature to have a clear understanding of who is being included and what conclusions can be drawn.
There are also clinical implications regarding a lack of information on trans and non-binary individuals. In a recent study by Baldwin et al. [177] patients who identified as transgender or genderqueer highlighted positive and negative interactions with their healthcare provider. Among the key themes of this qualitative study, participants who had negative experiences noted that they experienced misgendering,the clinician had a lack of knowledge about gender diverse individuals and/or experienced transphobia [177]. The lack of information on trans and non-binary participants may stem from the paucity of research that is inclusive of gender-diverse participants. While this study may not be focused on persons with LLA, it is important that future research in LLA that aims to inform clinical care make explicit the inclusion of gender diverse participants, in order to generate ethical best practices for this population [175]. Furthermore, it is necessary to establish the prevalence of LLA in trans and non-binary persons in order to identify population-specific needs and inform genderspecific interventions.

Limitations
As with all scoping reviews, there is the possibility that not all relevant literature was included in the review due to human error. However, every effort was taken to ensure that a rigorous selection process was followed to ensure that unintended omissions were kept to a minimum. This included double screening at each stage of the screening process, grey literature searching and reference searches. A second limitation of the current review involves the scope of the results. While it was not the purpose of this review, it is important to consider the intersectional nature of QoL and its related domains. Future research should seek to include intersectional analyses to highlight the differences by sex and gender in conjunction with other sociodemographic information such as race in order to get a better understanding of specific population group needs. Finally, the included studies were not assessed for quality. A lack of quality assessment, although not required for scoping reviews [36], leaves room for questions about methodological implications. It is possible that some of the findings from this review stem from articles that have methodological issues. Therefore, it is important that the quality of these articles is assessed in the future in order to draw stronger conclusions from the currently available literature.

Conclusion
Sex and gender differences are suggested for QoL and related domains for individuals with adult acquired LLA. Overall, findings tended to be more positive for men/males than women/females based on findings of the included studies. Articles included in this review were not clear with how gender was defined. In order for more targeted interventions that account for sex and gender differences, studies need to be more forthcoming about how they use and define gender. This will ensure that we are capturing relevant data and develop more definitive understandings of how sex and gender impact QoL. Furthermore, information regarding genderdiverse persons should be made explicit in future research. The needs of this population are currently not well understood and are therefore necessary in order to ensure we are providing equitable care to improve long-term health and well-being.