Women with limb loss: rationale, design and protocol for a national, exploratory needs assessment to evaluate the unique physical and psychosocial needs of women with limb loss

Abstract

Introduction

There is a growing population of women with limb loss, yet limited research is available to provide evidentiary support for clinical decision-making in this demographic. As such, there is a critical gap in knowledge of evidence-based healthcare practices aimed to maximise the physical and psychosocial needs of women with limb loss. The objective of this study is to develop a comprehensive, survey-based needs assessment to determine the unique impact of limb loss on women, including physical and psychosocial needs.

Methods and analysis

A bank of existing limb loss-specific and non-limb loss-specific surveys were arranged around domains of general health, quality of life, prosthetic use and needs, psychosocial health and behaviours and body image. These surveys were supplemented with written items to ensure coverage of relevant domains. Written items were iteratively refined with a multidisciplinary expert panel. The interpretability of items and relevance to limb loss were then internally tested on a small group of rehabilitation, engineering and research professionals. A diverse sample of 12 individuals with various levels of limb loss piloted the instrument and participated in cognitive interviews. Items from existing surveys were evaluated for relevance and inclusion in the survey, but not solicited for content feedback. Pilot testing resulted in the removal of 13 items from an existing survey due to redundancy. Additionally, 13 written items were deleted, 42 written items were revised and 17 written items were added. The survey-based needs assessment has been crafted to comprehensively assess the wide spectrum of issues facing women with limb loss. The final version of the survey-based needs assessment included 15 subsections.

Ethics and dissemination

This study was approved by the Veterans Affairs Central Institutional Review Board. The results will be disseminated through national and international conferences, as well as through manuscripts in leading peer-reviewed journals.

Trial registration number

No healthcare intervention on human participants was conducted.

Strengths and limitations of this study.

• This protocol is the first to develop a comprehensive, survey-based needs assessment to identify unique physical and psychosocial challenges for women with limb loss.

• The needs assessment will generate a subset of scores across several physical and psychosocial domains, which will help identify salient constructs related to health and well-being of women with limb loss.

• Due to the length of the needs assessment, a significant number of participants may not complete the online survey.

• Some of the existing surveys used in this protocol are not specifically validated for women with limb loss; however, they are considered valid and reliable in the adult population.

• Outcomes may not be specifically generalisable to the upper limb loss population due to the expected small sample size.

Background

After nearly two decades of decreasing amputation rates in the USA, there has been a recent resurgence in the national amputation rate, in part due to the increasing number of women undergoing amputation.¹ This trend is also evident at the Department of Veterans Affairs (VA) where female veterans now account for approximately 10% of the total VA population, with an expected increase to 16% over the next two decades.² Women with limb loss are one of the fastest growing subpopulations in VA²; specifically, from 2015 to 2019 the population of female veterans with limb loss rose from 2% to 3% of the total VA limb loss population.² This increasing rate may be partially due to the historical number of women serving in combat-related positions as more women have experienced amputation of one or more limbs during the most recent conflicts.^{3 4} In response to this rapidly rising demographic, in 2017 VA designated prosthetics for female veterans a national research priority.

Despite the growing population of women with limb loss, little research has been conducted to understand the specific physical and psychosocial needs of this population.⁵ The limited available research in the scientific literature restricts the ability of clinicians to provide evidence-based care for this underserved population.^{6 7} Most commercially available prosthetic devices have been traditionally designed with male anthropometry and biomechanics, which leads to greater dissatisfaction with prosthetic fit and appearance for women with limb loss.⁸ Challenges with prosthetic fit stemming from the lack of female-specific components and heavy prosthetic weight can result in increased movement inside prosthetic sockets, which can lead to skin integrity problems and an increased risk for secondary comorbidities.⁶ As such, women with limb loss are at a significantly higher risk for developing osteoarthritis in the contralateral limb compared with the general population.⁶ Women with limb loss also report significantly greater overall pain intensity compared with men with limb loss, and that pain interferes with function, activities of daily living and social activities to a greater extent.⁷ Overall, many of these physical challenges have led to a higher rate of artificial limb rejection for women with limb loss.^9–11 Furthermore, women with limb loss are less likely to be employed¹² and are more likely to live alone compared with men with limb loss.¹³ The psychosocial adaptation of women with limb loss has not been specifically addressed in existing literature.

There is a critical gap in knowledge of evidence-based healthcare practices aimed to maximise the physical and psychosocial needs of women with limb loss. However, current standards of care for women with limb loss are not well defined and tend to rely on anecdotal evidence in individual clinics. It is important to examine all metrics to develop evidence-based interventions. Healthcare providers for women with limb loss must evolve healthcare delivery and research practices, as well as work jointly with industry to meet the needs of this population. As such, it is critical to understand both the physical and psychosocial needs of women with limb loss to provide evidentiary support for clinical decision-making in this demographic. Hence, the aims for this study are to determine the differences in physical and psychosocial needs of women with limb loss compared with men with limb loss, and to classify these differences. The purpose of this paper is to describe the protocol to develop a survey-based needs assessment designed specifically to understand the wide spectrum of issues facing women with limb loss. Subsequent publications will describe the results of this exploratory needs assessment to help to guide clinical care more appropriately for this underserved population.

Methods and analysis

Study overview

This study represents the development of a comprehensive, survey-based needs assessment to aid in understanding the unique needs of women with limb loss. To help ensure content validity, this assessment primarily consists of existing surveys that are considered reliable and valid for the population of individuals with limb loss,^14–18 as well as a bank of Patient-Reported Outcomes Measurement Information System (PROMIS) scales valid for the adult population.^19–23 Collectively, the valid survey content was arranged around the following domains: general health, quality of life, prosthetic use and needs, psychosocial health and behaviours and body image. Additionally, due to the lack of gender needs assessments in the scientific literature and the limb loss research community, current measures were supplemented with additional questions written by the investigative team and iteratively refined with a multidisciplinary expert panel. The interpretability of items and relevance to limb loss was then internally tested on a small group of rehabilitation, engineering and research professionals, who were not involved in the development of the survey-based needs assessment. To ensure that the survey-based needs assessment addressed significant topics for this demographic, individuals with various levels of limb loss were included in pilot testing of the survey and their responses were used to refine the final survey-based needs assessment. The final survey-based needs assessment consists of 211 items for individuals with unilateral lower limb loss and 198 items for individuals with unilateral upper limb loss. Individuals with loss of multiple limbs answer limb loss specific questions for each limb (table 1). The final survey-based needs assessment was launched in July 2019. Data collection is expected to be completed in the final quarter of 2023 (due to significant delays from COVID-19). Analysis and dissemination of results will occur in 2024.

Table 1.

Total survey questions by limb loss levels

Limb loss level	Total survey questions
Unilateral upper	198
Unilateral lower	211
Bilateral upper	240
Bilateral lower	266
Bilateral upper/lower	253
Trilateral: lower/lower/upper	308
Trilateral: lower/upper/upper	295
Quadruple	350

Patient and public involvement

The survey-based needs assessment was developed and designed by research teams from VA New York Harbor Healthcare System (VANYHHS), James J Peters VA Medical Center (JJPVAMC), New York University Langone Health (NYU), Walter Reed National Military Medical Center (WRNMMC) and the Extremity Trauma and Amputation Center of Excellence (EACE). To help develop the goals and objectives of this study, as well as identify outcomes most meaningful to individuals with limb loss, the study team worked in collaboration with clinicians, limb loss investigators, prosthetists and stakeholders representing the interests of veterans and service members with limb loss and caregivers. Information gained from these discussions was reviewed by the study stakeholders, who used the identified research needs to develop patient-centered research questions and study aims, with an ultimate goal of identifying the unique physical and psychosocial needs of women with limb loss. Veterans and service members with various levels of limb loss were then included in the pilot testing of the survey-based needs assessment. Information from these individuals helped to ensure the comprehensive inclusion of concepts and domains that are important to individuals with limb loss. At the time of publication, participants will be contacted to inform them of the results.

Creation of the expert panel

Panel members were identified from personal contacts within VANYHHS, JJPVAMC, WRNMMC, EACE and NYU. The panel included nine members representing seven rehabilitation and research professions: Occupational therapy, physical therapy, biomedical engineering, prosthetics and orthotics, clinical and rehabilitation psychology, rehabilitation science and kinesiology. All members had previous experience working with the limb loss population.

Target population

The target population for the survey includes women and men with major loss of one or more limbs for at least 1 year. Major limb loss was defined as experiencing an amputation (or loss of a limb) through or proximal to the ankle or wrist, due to any aetiology. Twelve participants with limb loss, recruited from VANYHHS, completed pilot testing of the final draft items. After pilot testing was completed, the final survey-based needs assessment was launched in July 2019. For the ongoing data collection, participants are recruited from the seven regional amputation centres in the VA network, three military treatment facilities, private sector medical centres and regional and national conferences targeted towards individuals with limb loss (eg, National and Regional Amputee Coalition conferences).

Outcome measures

The survey-based needs assessment consists of an online instrument based on reliable and valid surveys that collectively measure the physical health, quality of life, prosthetic use and needs and psychosocial experiences of individuals with limb loss. The valid surveys include both non-limb loss-specific surveys and limb loss-specific surveys. After initial consent to participate, the online survey asks a series of screening and demographic questions to determine eligibility. Once deemed eligible, all participants answer the same bank of survey questions modified to the appropriate version of the outcome measures for lower or upper limb loss, or both, depending on responses to the screening and demographic questions. Individuals with loss of more than one limb complete the same limb loss-specific questions targeted for each lost limb.

Non-limb loss-specific surveys

A bank of PROMIS scales were used as part of the survey-based needs assessment. Short forms were used for each of the measures, which assess outcomes over the past 7 days, unless otherwise specified. All eligible participants complete each of the measures below:

PROMIS Global Health: The PROMIS Global Health measure is a general assessment of an individual’s physical, mental and social health. It is not disease specific and is intended to reflect the global assessment of an individual’s health. PROMIS Global Health produces both a physical health and mental health score.¹⁹

PROMIS Emotional Support 8a: The PROMIS Emotional Support survey measures perceived feelings of being cared for and valued as a person. It is not disease specific and a single score is generated.²³

PROMIS Depression 8a: The PROMIS Depression survey is a measure that assesses self-reported views of self, negative mood, decreased positive affect, social cognition and engagement. It is not disease specific and a single score is generated.²²

PROMIS Anxiety 8a: The PROMIS Anxiety survey is a self-reported measure that assesses fear, anxiety, misery, hyperarousal and somatic symptoms related to arousal. The PROMIS anxiety measure is non-disease specific.²²

PROMIS Pain Interference 8a: The PROMIS Pain Interference survey is a universal measure that assesses the extent to which pain interferes with social, cognitive, emotional, physical and recreational activities. It is non-disease specific.²¹

Limb loss-specific surveys

The following surveys (and subdomains) are administered to respondents based on initial responses regarding their limb loss. All participants answer the same bank of items. However, respondents are differentiated based on upper or lower limb loss and branched to the appropriate survey questions for their limb loss. Participants with loss of multiple limbs complete the limb loss-specific surveys for each lost limb.

Trinity Amputation and Prosthesis Experience Scales - Revised (TAPES-R): The TAPES-R survey is designed for use by adults with upper or lower limb loss.^{16 17} This multidimensional assessment evaluates psychosocial adjustment (general adjustment, social adjustment and adjustment to limitation), activity restriction (functional, social and athletic) and satisfaction (weight of prosthesis, functional and aesthetic satisfaction). Additionally, there are separate questions on pain and general health.

Orthotics and Prosthetics Users Survey (OPUS) - Satisfaction with Services: The OPUS Satisfaction with Services module is a self-report questionnaire used to assess quality and satisfaction with prosthetic services.¹⁵

Prosthesis Evaluation Questionnaire (PEQ): The PEQ is comprised of independent domains that can be scored separately. The scales used for the survey-based needs assessment included the appearance scale and the residual limb health scale. Additionally, the PEQ provides individual questions that are not combined into scale scores and are scored individually. For this survey, PEQ questions for pain, transfers and importance were included.¹⁴

Amputee Body Image Scale - Revised (ABIS-R): The ABIS-R survey is a 14-item measure that evaluates body image disturbances in people with limb loss.¹⁸

Gap item development

Due to the lack of women-based needs assessments in the limb loss literature and research community, current measures may not capture all relevant domains for this population. Therefore, as an exploratory analysis, additional questions were developed by the expert panel to ensure supplementary aspects of the needs were represented in the final assessment. These questions included both closed-ended and open-ended items, which enable the research team to gather specific information and more narrative information. Items and instructions were developed in iterative rounds involving all investigators. Questions included demographics (eg, specifics about limb loss, prosthesis and socket use, education, employment, race and ethnicity), interactions with healthcare providers (eg, primary care, prosthetist and physical and occupational therapist) and effect of limb loss on social support and sexual health. All non-demographic questions included 7-point Likert scales of intensity as the response set. Two open-ended questions were included to discern how limb loss has affected the respondent’s employment status and role in his/her residence. An additional open-ended item was included to understand comfort surrounding interactions with the respondent’s healthcare team. Lastly, an open-ended question was included to describe any topics or situations specific to being a woman that have been affected by their limb loss to help identify gaps in item coverage. A ‘prefer not to answer’ option was provided for every question.

Content validation—expert panel review

The expert panel members reviewed each survey item and provided comments on question clarity, relevance and priority in relation to limb loss. Any items that were part of a valid and reliable survey were only reviewed for relevance and priority for inclusion in the survey-based needs assessment. Furthermore, suggestions by the expert panel for improvements and additional domains were solicited to ensure the survey-based needs assessment included critical content areas. All items that were deemed clear, relevant and high priority were retained with no modifications. Any item marked as unclear, low relevance or low priority were marked for further review by the expert panel. Items rated as unclear or somewhat unclear were then modified by the research team to improve clarity. Items that were rated as irrelevant or low priority were removed from the item bank. This first round of revision resulted in the removal of eight items and the revision of 21 questions. In addition, the expert panel identified an additional domain to be included: sexual health. Three sexual health questions were developed and added to the item bank.

The revised item bank was then reviewed again by the expert panel and the content rating process was repeated. No additional items were removed from the item bank. Eight items were revised a second time, resulting in a total of 220 items for individuals with unilateral lower limb loss and 207 for unilateral upper limb loss to be tested in the pre-pilot phase. The pre-pilot assessment was organised into the following sections: Eligibility, demographics, general health, amputation history, prosthesis use/satisfaction, function/mobility, adjustment to prosthesis or limb loss, pain interference, satisfaction with healthcare services, mental health, social support and comfort in intimate situations.

Content validation—pre-pilot testing

Eight staff members from VANYHHS, chosen by the investigative team, pre-piloted the survey-based needs assessment internally to identify issues related to wording, flow and technical problems prior to pilot testing with participants. The group consisted of one prosthetist, two rehabilitation specialists, three engineers and two research coordinators, none of whom had limb loss. Each staff member tested branching of the survey by completing it multiple times under different limb loss scenarios (eg, unilateral transtibial limb loss, bilateral transradial limb loss). Additionally, time to completion was recorded. Pre-pilot participants completed the online survey through a website hosted by the Army Public Health Center using the survey software system Verint (Verint Systems, Melville, New York, USA). Participants were given a template to record any wording, flow or technical challenges that occurred.

Feedback from pre-pilot testing resulted in no items added or removed from the survey-based needs assessment. The most common feedback was framed around wording or tone of an item (n=5 items), technical challenges from the survey software (n=6 items), formatting of the questions (n=10 items) and suggestions for revisions (n=3 items). Feedback was reviewed again by the expert panel, which resulted in the revision of five items. After modifications were completed, the survey-based needs assessment was then considered in final draft form.

Pilot testing

Once all items were in final draft form, participants with major limb loss were recruited to pilot the survey-based needs assessment and undergo cognitive testing of the written gap items. Cognitive testing is a common technique and method used during the development of survey items that centres on the cognitive processes that respondents use while answering survey questions.²⁴ Participants were considered eligible if they were at least 18 years old, had been living with major limb loss for at least 1 year, were able to effectively read or write in English and did not have any cognitive deficits or mental health problems that would limit ability to participate. Pilot participants completed the survey-based needs assessment at VANYHHS via an iPad. Between each section, the survey was paused and feedback was solicited from participants to ensure items and directions were interpreted as intended. For the screening and eligibility section, a study team member repeated specific questions from that section to the respondent and solicited feedback regarding question clarity and any potentially problematic words identified during development. Any modification suggestions were then documented. For the remaining written gap items in the subsequent sections, three main topics were addressed for each written item: (1) How did the participant arrive at her/his answer? (2) Was this question hard or easy to answer? (3) Was the response set sufficient? Additionally, any words or phrases identified as potentially difficult to interpret during item development or content validation were also queried, and clarification or modification suggestions were recorded.

The purpose of the pilot testing was to fine-tune the wording of the gap question item bank and to ensure an appropriate flow of all questions. The performance of the gap items was assessed using qualitative approaches. Responses to these questions were evaluated for item coverage. The goal of assessing these items is to ultimately develop a women-specific, patient-reported outcome measure of the most important concepts and domains applicable to women with limb loss. Psychometric characteristics, such as factor structure, were not evaluated. This study was approved by the VA Central Institutional Review Board (IRB) (#18–20). All participants provided consent to participate in cognitive interviews and survey data collection.

Sample characteristics

A total of 12 individuals with limb loss participated in the pilot study (table 2), of which 10 participated in cognitive interviews. The final two participants completed the survey to obtain an accurate representation of the survey timing and were queried on length and flow of the survey, as well as any overall comments. Participants were, on average, 61.8±13.6 years old with an average time since initial limb loss of 27.9±22.4 years. The popluation included participants with various levels of lower and upper limb loss (table 2).

Table 2.

Pilot participant demographics

Participant	Age	Sex	Level(s) of limb Loss	Years since initial limb loss	Race	Ethnicity
1	72	M	Forequarter (L)	59	White	Not Hispanic or Latino
2	67	M	Unilateral transtibial (R)	24	Black or African American	Not Hispanic or Latino
3	61	M	Bilateral transtibial	12 (L); 11 (R)	Black or African American	Not Hispanic or Latino
4	66	F	Unilateral transfemoral (L)	9	Black or African American	Not Hispanic or Latino
5	59	F	Unilateral transradial (L)	13	White	Hispanic or Latino
6	59	M	Unilateral transfemoral (L)	9	Black or African American	Not Hispanic or Latino
7	69	M	Unilateral transtibial (R)	3	Black or African American	Not Hispanic or Latino
8	72	M	Wrist disarticulation (L)	52 (both)	White	Not Hispanic or Latino
			Transradial (R)
9	72	M	Unilateral transtibial (R)	53	White	Hispanic or Latino
10	73	M	Transtibial (R)	48 (both)	White	Hispanic or Latino
			Ankle disarticulation (L)
11	45	M	Unilateral transfemoral (L)	7	Black or African American	Not Hispanic or Latino
12	27	F	Unilateral transfemoral (L)	3	White	Not Hispanic or Latino

F, female; L, left; M, male; R, right.

Cognitive interview feedback

A total of 220 items for individuals with unilateral lower limb loss, and 207 items for individuals with unilateral upper limb loss were included for testing in the pilot study of the 12 participants (table 3). A total of 140 items for individuals with lower limb loss and 127 items for individuals with upper limb loss were from valid and reliable surveys. These items were only evaluated for inclusion in the survey, but not solicited for content feedback. The remaining 80 items were tested in one round of cognitive interviews. No items were skipped by any participants. All items were felt to be important to individuals with limb loss. Following participant feedback, 13 items from the PEQ pain subscale were removed due to redundancy. An additional 13 gap items were removed, 42 items were revised and 17 items were added. The remaining items were retained without any modifications or revision. The most common feedback was wording or tone of the item (n=16 items), suggestions for revisions (n=14 items), unclear reference or lack of understanding (n=10 items) and limited applicability (n=5 items). In some cases, respondents felt that certain items did not specifically apply to them, but could be relevant to other people with limb loss. Respondents also felt some items were redundant, but these items were part of valid surveys. Additionally, feedback was solicited on timing and survey formatting. Three participants said the survey was too long and three suggested better spacing of the items. Time to survey completion was extracted from Verint for the final two participants, both with unilateral transfemoral limb loss, which was 41 and 44 min.

Table 3.

Development of final item bank by section

Item bank—pilot testing 220 items: lower limb loss 207 items: upper limb loss
Section	Validated survey items	Gap items	Gap items added	Items removed	Gap items revised	Final number of items
Eligibility	–	7	2	0	4	9
Demographics	–	21	4	1	9	24
General health	10: PROMIS Global Health	12	0	0	12	22
Amputation history	–	9	1	2	2	8
Prosthesis history	20: PEQ (lower extremity); 17: PEQ (upper extremity)	0	0	0	0	20 lower; 17 upper
Function/mobility	12: PEQ	0	0	7	0	5
Adjustment to prosthesis/limb loss	34: TAPES-R (lower extremity); 24: TAPES-R (upper extremity)	0	0	0	0	34 lower 24 upper
Pain	13: PEQ; 8: PROMIS Pain Interference	3	2	13 from PEQ; 3 from gap items	0	10
Interaction with healthcare providers	5: OPUS	8	0	0	5	13
Mental health	8: PROMIS Anxiety; 8: PROMIS Depression; 14: ABIS-R	0	0	0	0	30
Social support and situations	8: PROMIS Social Isolation	18	8	0	10	34
Final comments	–	1	0	0	0	1
Women’s specific question	–	1	0	0	0	1
Totals	140 lower; 127 upper	80	17	26	42	211 lower 198 upper

Final item bank: 211 items for unilateral lower limb loss and 198 items for unilateral upper limb loss.

ABIS-R, Amputee Body Image Scale - Revised; OPUS, Orthotics and Prosthetics Users Survey; PEQ, Prosthesis Evaluation Questionnaire; PROMIS, Patient-Reported Outcomes Measurement Information System; TAPES-R, Trinity Amputation and Prosthesis Experience Scales - Revised.

Common themes

Common themes were reviewed following cognitive interviews to help refine items and ensure adequate coverage of important topics.

Post-limb loss changes in employment status: Many participants commented on how their employment opportunities were limited by their limb loss. Most participants reflected that physical limitations and pain affected their ability to perform many job duties.

Post-limb loss changes in role in their residence: Participants commented on how they were more reliant on others to complete tasks in the household. When they can perform tasks, it often takes more planning and time to complete them.

General health: For older participants and individuals with non-traumatic limb loss, respondents thought about health issues that caused the limb loss or could lead to future limb loss (eg, diabetes, dysvascular disease) and the importance of managing their health. Many respondents also discussed the need for knowledgeable, multidisciplinary healthcare providers to help manage their overall healthcare.

Mental health and support of others: Most participants thought that all questions asked in this section were important, even if not specifically applicable to them. Importance was framed around holistic treatment and the need for social and mental health support.

Women-specific challenges: Common themes regarding the specific needs of women with limb loss included challenges of hormonal changes, weight management, social isolation and lack of participation in physical activities (eg, sports, exercise). Women also expressed the need for mutual decision-making with their healthcare providers, specifically their prosthetist.

Final survey content

The final version of the survey-based needs assessment includes 15 subsections. Table 4 outlines the description of each subsection for respondents. One open-ended item on the specific needs of women with limb loss was also included. Data collection is currently ongoing and all participants complete the online assessment anonymously. The final items, data dictionary and cognitive interview feedback are given in supplemental materials (online supplemental additional files 1–3, respectively).

Table 4.

Survey-based needs assessment subsections

Subsections	Instructions for respondent
Eligibility	To begin, we have a series of questions to determine your eligibility to participate in this survey. Please note: For the survey, amputation and limb loss refer to missing a limb for any reason, including since birth.
Demographics	Next, we would like to know a little more about you. These questions will help provide context for experiences you will be asked about later in the survey.
General health	In this section we will be asking questions about your overall health and any other medical conditions that you may have.
Amputation history and prosthesis	In this section we will be asking questions to better understand your amputation and your experiences with your prosthesis.
About your prosthesis	Now, we are going to ask you to answer a series of questions from standardized surveys. Some questions may appear to be repetitive but are important to help determine the optimal outcomes for individuals living with amputation.
Function and mobility	In this section, we are going to ask questions regarding your ability to move around or questions regarding your mobility.
Adjustment to prosthesis/adjustment to limb loss	In this section, we are going to ask about your general and social adjustment to your prosthesis; In this section, we are going to ask about your general and social adjustment to your limb loss.
Pain interference	In this section, we are going to ask about how pain hinders your engagement with social, cognitive, emotional, physical and recreational activities.
Healthcare	Now, we are going to ask questions about your overall experience with your healthcare providers. When answering the questions please consider your general experience with your prosthetist and healthcare providers.
Mental health	In this section, we are asking questions regarding your mental health and body image. This is a research study that involves questions related to sensitive topics. As researchers, we do not provide mental health services. However, we want to provide you with contact information for available resources, should you decide you need assistance at any time. National resources for veterans and civilians are included below. In the event that you feel overwhelming emotional distress or the intent to harm yourself, please use these resources.
Social support and situations	In this section, we are going to ask you questions about your overall comfort levels in different situations.
Final comments	In this last section, we would like to know any comments you would like to make about your experience with your amputation.

A priori sample size estimation

This study proposes to enroll a convenience sample of 200 participants (100 men and 100 women) who have limb loss. This proposed sample size is supported by previous investigations, which have evaluated gender differences on limb loss outcomes.^{7 12 13} In particular, Singh et al¹³ included gender differences in a regression analysis of factors that influenced the success of limb-fitting after limb loss. The regression analysis used by Singh et al¹³ most closely resembles the analyses proposed in this investigation and can serve as a reasonable estimate of effect size, which was reported to be 0.05. This effect size is considered between a small (0.02) and medium effect (0.15) based on Cohen’s recommended sizes.²⁵ To adequately power this study, assuming an alpha error rate of 5% with an effect size of 0.05, the total required sample size to achieve 80% power is 192 participants. Therefore, we propose to collect a convenience sample of 200 individuals (100 per group), which should provide adequate power to detect differences in outcomes between women and men with limb loss. Furthermore, the study was designed as a needs assessment to identify preliminary physical and psychosocial factors unique to women with limb loss. As such, data generated from this study can serve as pilot information to further investigate the specific needs of women with limb loss.

Planned statistical analysis plan

The primary goal of the analysis will be to compare outcomes between male-identifying and female-identifying individuals with limb loss. Scales for upper and lower limb loss will be analysed separately, where applicable. Tables of descriptive statistics, including minimum, maximum, mean, SD and median will be generated for each scale and demographic variable, stratified by age, identified gender, education, employment status, living situation and other socio-demographic variables. Multiple linear and generalised linear regression models will be used to test the association of identified gender with scale outcomes, while adjusting (controlling) for additional covariates. Potential covariates include: Age, education, living situation, general health, reason for limb loss and years since limb loss. Choice of model, as well as the covariates used, will be dependent on the outcome (see table 5 for further details). For example, scales transformed into T-scores will be evaluated using linear regression, whereas scales that are averaged or summed across items will be analysed using generalised linear models. All statistical analyses will be conducted in R (V.4.1.0).²⁶

Table 5.

Planned statistical models by outcome measure

Scale	Proposed model
Difference in physical needs
Appearance	Generalised linear model
Residual limb health	Generalised linear model
PEQ importance	Generalised linear model
Aesthetic satisfaction	Generalised linear model
Functional satisfaction	Generalised linear model
Overall satisfaction	Generalised linear model
Mobility	Generalised linear model
Activity restriction	Generalised linear model
Usefulness/reliability	Generalised linear model
Difference in psychosocial needs
Healthcare	Linear regression; generalised linear model
PEQ importance	Generalised linear model
Overall satisfaction	Generalised linear model
Comfort in situations	Generalised linear model
Change in comfort	Generalised linear model
General adjustment	Generalised linear model
Emotional distress/anxiety	Linear regression
Depression	Linear regression
Social support	Linear regression
Body image	Generalised linear model

PEQ, Prosthesis Evaluation Questionnaire.

Ethics and dissemination

This study was approved by VA Central IRB (#18–20). The US Army Medical Research and Development Command Office of Human Research Oversight also provides oversight of the investigation (log number E04083). Final results of the survey-based needs assessment will be submitted for publication in peer-reviewed scientific journals. Preliminary and final results will also be presented at national and international conferences and in-service presentations for limb loss providers.

Ethics statements

Patient consent for publication

Not applicable.

Ethics approval

This study was approved by VA Central IRB (#18–20). The U.S. Army Medical Research and Development Command Office of Human Research Oversight also provides oversight of the investigation (Log Number E04083).