Measurement properties of assessment instruments of quality of life in people with spinal cord injury: A systematic review

Abstract

Context

A spinal cord injury or disease (SCI/D) is a devastating condition that affects all areas of a person’s life, including quality of life (QoL). Assessing this construct using clinical instruments with adequate measurement properties is fundamental for an effective multi-professional treatment.

Objective

To identify the clinical instruments for assessing the QoL that present the best recommendation for use in people with SCI based on their measurement properties.

Methods

The overall methodology was conducted according to Consensus-based Standards for the selection of health Measurement Instruments (COSMIN) systematic review guidance and reported according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement. A literature search was conducted up to March 2023 on MEDLINE/PubMed, Web of Science, Scopus, CINAHL, and Embase databases.

Results

Seventy-seven studies were included in this review, and 49 instruments were identified. The overall methodological quality of all studies ranged from insufficient to sufficient, and the level of evidence ranged from very low to high confidence. Twenty-six instruments may have the potential to be recommended, and the other 23 should not be recommended.

Conclusion

None of the QoL instruments can be highly recommended as the most suitable instrument for the construct and population of interest. The generic instruments SF-36, SF-12, QWB, WHOQOL-DIS, WHOQOL-BREF, QLI-SCI, QOLP-PD, LS Questions, Lisat-9, and BRFSS are the clinical instruments that have the best measurement properties tested and have the potential to be the current best recommendation for assessing QoL in individuals with SCI.

Introduction

Spinal cord injury or disease (SCI/D) is a devastating lifelong condition that affects all areas of a person’s life (1–7). Among other aims, rehabilitation seeks to improve the quality of life (QoL) of individuals with SCI/D (8–10). The QoL construct is vast and quite diverse among different authors in the literature due to the individual’s subjectiveness, encompassing aspects related to life satisfaction and general well-being, and because of that, challenging to measure. For multi-professional rehabilitation teams, assessing the QoL in individuals with SCI/D is essential (11).

Increasing attention is being given to assessing changes in overall well-being or QoL in individuals with SCI/D. However, assessing this specific outcome can be challenging due to the complexity of capturing QoL-related constructs, the lack of consensus regarding the definition of general QoL, and the need to distinguish between objective and subjective measures (12). Moreover, how people adapt to their condition may interfere with their perception of QoL. One of the most significant challenges in defining this construct remains the difficulties in creating a clinical instrument sufficiently sensitive for use with the varied clinical spectrum of individuals with SCI/D (13).

The increasing interest in assessing QoL has led to the development of new SCI/D-specific instruments and validation studies of generic QoL instruments for use with the same group. Selecting the appropriate clinical instrument to assess QoL should be based on the type of measurement, personal and clinical factors, feasibility and its measurement properties (14). To our knowledge, four previous systematic reviews have sought to identify the best instruments for assessing QoL in the SCI/D population (13, 15–17). However, these previous reviews were published more than ten years ago. As a result, many different instruments were developed or had their measurement properties tested after this period. Additionally, none of these reviews had applied a standardised methodology to assess the quality of the measurement properties and to conduct systematic reviews on measurement properties. The Consensus-based Standards for the selection of health Measurement Instruments (COSMIN) developed a taxonomy to guide the measurement properties’ definitions and a methodology guideline to conduct systematic reviews of patient-reported outcome measures (PROMs) that use systematic criteria to standardise this type of study by using the GRADE system (18,19).

The COSMIN initiative was founded in 2005 by an international multidisciplinary team of researchers with a background in epidemiology, psychometrics, medicine, qualitative research, and health care, who have expertise in developing and evaluating outcome measurement instruments (20,21). Thus, a systematic review of measurement instruments following the COSMIN methodology follows strict and transparent recommendations for processing evidence.

This systematic review aims to identify the QoL instruments available in the literature, critically analyse their methodological quality of measurement properties and determine which instrument can be recommended for use based on their measurement properties.

Methods

This systematic review was conducted according to the COSMIN guidelines (18), and its findings are reported according to the PRISMA statement (22). In addition, the protocol was registered in the international prospective register of systematic reviews (http://www.crd.york.ac.uk/PROSPERO/), registration number: CRD42020192674.

Data sources and searches

A literature search was conducted up to July 2020 (updated in 25th March 2023) in the following electronic databases: MEDLINE/PubMed, Web of Science, Scopus, CINAHL, and Embase. The complete search strategy for each electronic database can be found in Supplemental Appendix A. The general search strategy consisted of 4 groups of MESH search terms combined with the Boolean operator “AND” 1) SCI, 2) quality of life OR life satisfaction, 3) outcome assessment, and 4) measurement properties. No date restrictions were applied to the search. The search protocol was not published a priori. Google Scholar was also searched by combining the Boolean operator “AND” with the following terms: 1) SCI, 2) quality of life OR life satisfaction, 3) reliability OR validity, OR measurement properties. Finally, citation tracking of the eligible studies was carried out by checking their reference lists for additional studies.

Study selection

References from the database search were entered into the Rayyan online platform (23). Two independent reviewers initially screened the titles (L.A., G.S.) to exclude prominent non-fitting titles and duplicate references. After, the remaining titles, abstracts, and full texts were then independently reviewed for eligibility by two reviewers (L.A., G.S.). Any reviewer disagreements were discussed with a third reviewer (J.I.) until a consensus was reached. Studies were eligible if: 1) the study sample contained individuals with SCI/D (>18 years of age); 2) they clearly described the quality of life or life satisfaction-related outcomes; 3) they assessed at least one measurement property or developed an instrument intended to evaluate a QoL-related outcome. Studies were excluded when they were systematic reviews, study proposals or protocols; case studies or series; books or book chapters; poster abstracts; conference abstracts, correspondence, or commentaries (Supplemental Appendix B). For the purpose of this systematic review, we included studies that clearly describe that the instrument was developed and/or is being tested to assess the construct QoL in people with SCI.

Data extraction

Using a standard extraction form, one reviewer (G.S) extracted data, which was then checked by the second reviewer (L.A.). The data initially extracted from the included studies were: 1) author; 2) country/language; 3) aim of the study; 4) characteristics of the study sample; 5) studied measurement properties; 6) the number of items in the instrument, and; 7) instrument type and administration.

Methodological quality assessment and formulating recommendations

As recommended by Mokkink et al. (19), two reviewers (L.A., A.L.) independently assessed the methodological quality of the included studies using the COSMIN checklist risk of bias (24). The checklist consists of 12 boxes: 10 can be used to assess the methodological quality of studies on measurement properties, and two boxes contain general requirements. After retrieving the available evidence, the COSMIN guidance for systematic reviews of PROMs recommends assessing the measurement properties in the following order: 1) content validity, 2) internal structure (structural validity, internal consistency and cross-cultural validity), and 3) reliability, measurement error, criterion validity, construct validity, responsiveness. The same reviewers perform the three assessment phases for each measurement property.

In the first phase, the risk of bias assessment of the methodological quality of the included studies regarding each measurement property was assessed in each study by using the COSMIN Risk of Bias checklist(24). Each study was then rated as either very good (V), adequate (A), doubtful (D) or inadequate (I) quality according to the instructions in the user manual. When rating the study’s methodological quality, the lowest rating of any item in a box was taken (the ‘worst score counts’ method) according to the COSMIN guidelines. That is, a poor assessment of any item in the assessment box is therefore considered a fatal flaw and impacts the quality of that instrument. In case of disagreement, a third evaluator was consulted (J.I.).

Afterwards (second phase), the results for each measurement property in each study are rated against the quality criteria for determining the sufficiency or insufficiency of each measurement property based on Prinsen et al. (25). Then, the results of all available studies on a measurement property are qualitatively summarised and compared against the criteria for good measurement properties to determine whether the overall – the measurement property of the instrument is sufficient (+), insufficient (−), inconsistent (±), or indeterminate (?). From this step, the main focus of the analysis becomes the instrument and no longer each study. If all studies show the same result (all-sufficient or insufficient), the summarisation is done based on the criteria of good measurement properties. But, if the results obtained are inconsistent, the explanations for these inconsistencies must be presented by the authors of the studies, and general classifications by subgroups according to the characteristics must be analysed (e.g. elderly versus children). However, if the study authors do not provide explanations, the overall rating of the instrument will be inconsistent (±). And finally, when not enough information is available, the overall rating of the instrument will be indeterminate (?). The criteria for evaluating the measurement properties of the QoL instruments assessed in each study were analysed and rated according to the consensus-based criteria proposed by Prisen et al. (25) (Supplemental Appendix C). To evaluate construct validity (hypothesis testing and known-groups validity) and responsiveness, the review team formulated a set of a priori hypotheses against which to test these properties based on previous studies (26–28) (Supplemental Appendix C).

The overall rating for each measurement property of each instrument was obtained by combining all the ratings from each study in which that measurement property was assessed. Then, the overall rating for the quality criteria based on its measurement properties is defined as either sufficient (+), insufficient (–), inconsistent (+/−), or indeterminate (?). Within studies, construct validity and responsiveness were considered sufficient (+) if >75% of the hypotheses were met; otherwise, they were deemed inconsistent (+/−) or insufficient (−) if > 75% of the hypotheses were not met. Between studies, results were considered inconsistent (+/−) if they did not display the same results (i.e. they did not all report sufficient, insufficient, or inconsistent findings).

The third phase consisted of summarising the quality of the evidence (confidence level), in which each instrument received a 4-point score using the GRADE approach (high, moderate, low, or very low quality of evidence). The GRADE approach for systematic reviews is applied depending on the presence of four factors: risk of bias, indirectness, inconsistency and imprecision, and information about how to downgrade is described in detail in the COSMIN user manual (19)

Finally, each assessed instrument was assigned to a recommendation category regarding potential clinical and research use according to the criteria proposed by COSMIN guideline (18): (A) instruments that have the potential to be recommended as the most suitable PROM for the construct and population of interest (i.e. instruments with evidence for sufficient content validity (any level) and at least low evidence for sufficient internal consistency). In addition, the results obtained using these instruments can be considered trustworthy; (B) instruments that may have the potential to be recommended, but further validation studies are needed (i.e. instruments categorised not in A or C); and (C) instruments that should not be recommended for use (i.e. instruments with high quality of evidence for an insufficient measurement property). A step-by-step method to conduct the systematic review according to the COSMIN guidelines is presented in Fig. 1.

Figure 1.

The systematic review method summarised.

Results

A total of 367 records were identified in the electronic searches. After screening for duplicates, 139 were removed. The titles and abstracts of 228 studies were read, which led to the exclusion of 140 studies. The 88 remaining studies were read in full, and eight did not meet the inclusion criteria. However, a further 19 were identified through the citations and references in the remaining studies (Fig. 2). Of these 99 studies, 22 full-text studies were excluded because they either: involved structured phone interviews to monitor discharged patients in a follow-up study (none measure properties were assessed) and/or did not clearly defines that the instrument’s purpose was to measure the construct QoL; were related to other outcomes or other populations (non-SCI/D subjects or health professionals); described rather than assessed a QoL instrument and its related measurement properties; were an event abstract, reviews or observational studies; there was an error with the DOI registration or lacked author contact details.

Figure 2.

Flow diagram.

Among the 77 studies included in this review, 49 instruments were identified, as shown in Table 1. A brief description of each instrument is provided in Supplemental Appendix D.

Table 1.

List of included QoL instruments for people with SCI.

Name	Acronym/Initials	Number of studies	Reference
23-item Spinal Cord Injury – specific Health Related Quality of Life questionnaire	SCIQL-23	2	Ebrahimzadeh et al., 2014; Jain et al., 2007
5 satisfaction items from the World Health Organization Quality of Life Assessment	WHOQOL-5	1	Geyh et al., 2010
Assistive Technology Device Predisposition Assessment	ATD PA	1	Scherer & Cushman, 2001
Behavioral Risk Factor Surveillance Survey	BRFSS	1	Andresen et al., 1999
Incontinence Quality of Life Questionnaire	I-QOL	1	Schurch et al., 2007
Instrumental activities of daily living	IADL	1	Andresen et al., 1999
International Spinal Cord Injury Quality of Life Basic Data Set	SCI QoL-BDS	6	New et al., 2019; Post et al., 2016; Post et al., 2019; Pattanakuhar et al., 2020; Konstantinidis et al., 2021; Nijhawan et al., 2022
International Spinal Cord Injury Quality of Life Basic Data Set version 2.0	SCI QoL-BDS version 2.0	1	Post et al., 2022
Item Bank to Measure Systems, Services, and Policies: Environmental Factors Affecting People With Disabilities	IBMSSP – EFAPD	1	Lai et al., 2016
King’s Health questionnaire	KHQ	3	Karapolat et al., 2018; Liu et al., 2010; Romero-Cullerés et al., 2011
Life Satisfaction Questionnaire	LiSat-9	2	Geyh et al., 2010; Post et al., 2012;
Life Satisfaction questions	LS Questions	1	Post et al., 2012
Medical Outcomes Short-Form Health Survey	SF-36	7	Andresen et al., 1999; Forchheimer et al., 2004; Lee et al., 2008; Lin et al., 2007; Schurch et al., 2007; Liu et al., 2010; Versteeg et al., 2018
Medical Outcomes Short-Form Health Survey – 12-Item Short Form	SF-12	2	Andresen et al., 1999; Vishwanathan et al., 2020
Medical Outcomes Short-Form Health Survey – 6 dimensions	SF-6D	1	Engel, 2014; Lee, 2008
Medical Outcomes Short-Form Health Survey – mental health	SF – 36 (MH-5)	1	Van Leeuwen et al., 2012
Medical Outcomes Short-Form Health Survey – Physical Functioning Scale for Use With Veterans With Spinal Cord Injury	SF-36V	1	Luther et al., 2006
Neurogenic Bowel Dysfunction	NBD	1	Erdem et al., 2017
Neurogenic Bladder Symptom Score Questionnaire	NBSS	4	Cintra et al., 2019; Guler et al., 2020; Przydacz et al., 2020; Welk et al., 2014
Modified PRISM	M-PRISM	1	Sweatman et al., 2020
Patient Reported Impact of Spasticity Measure	PRISM	1	Cook et al., 2007
Personal Well-Being Index	PWI	1	Geyh et al., 2010
Quality of Life Index-Spinal Cord Injury version	QLI-SCI	3	Kovacs et al., 2015; May & Waren, 2001; May & Waren, 2002
Quality of Life Profile for Adults with Physical Disabilities	QOLP-PD	1	Renwick et al., 2003
Quality of Well-Being scale	QWB	1	Andresen et al., 1999
Qualiveen	Qualiveen	5	Bonniaud et al., 2011; D’Ancona, 2009; Nikfallah et al., 2015; Pannek et al., 2007; Reuvers et al., 2017
Qualiveen Short Form	Qualiveen SF	4	Reuvers et al., 2017; Przydocz et al., 2021; Konstantinidis et al., 2020; Krebs et al., 2021
Satisfaction with Life Scale	SWLS	4	Geyh, 2010; Post, 2012; Renwick, 2003; Scherer & Cushman, 2001
Sense of Well-Being Inventory	SWBI	1	Chapin et al., 2004
Spinal Cord Injury – Quality of Life – Pressure Ulcers	SCI-QOL Pressure Ulcers	1	Kisala et al., 2015
Spinal Cord Injury – Quality of Life 7-item short form (SF) Pressure Ulcers	SCI-QOL Pressure Ulcers SF	1	Kisala et al., 2015
Spinal Cord Injury – Quality of Life Ability to Participate and Satisfactions with Social Roles and Activities item banks	SCI-QOL APSSRA	1	Heinemann et al., 2015
Spinal Cord Injury – Quality of Life Anxiety	SCI-QOL Anxiety	1	Kisala et al., 2015
Spinal Cord Injury – Quality of Life Bladder Complications item bank	SCI-QOL Bladder Complications	1	Tulsky et al., 2015
Spinal Cord Injury – Quality of Life Bladder Management Difficulties item bank	SCI-QOL Bladder Management Difficulties	1	Tulsky et al., 2015
Spinal Cord Injury – Quality of Life Bowel Management Dificulties item bank	SCI-QOL Bowel Management Dificulties	1	Tulsky et al., 2015
Spinal Cord Injury – Quality of Life Depression	SCI-QOL Depression	1	Tulsky et al., 2015
Spinal Cord Injury – Quality of Life Economic	SCI-QOL Economic	1	Tulsky et al., 2015
Spinal Cord Injury – Quality of Life Grief and Loss item bank	SCI-QOL Grief and Loss	1	Kalpakjian et al., 2015
Spinal Cord Injury – Quality of Life item Bank	SCI-QOL item bank	1	Tulsky et al., 2011
Spinal Cord Injury – Quality of Life Positive Affect and Well-being bank	SCI-QOL PAWB	1	Bertisch et al., 2015
Spinal Cord Injury – Quality of Life Psychological Trauma	SCI-QOL Psychological Trauma	1	Kisala et al., 2015
Spinal Cord Injury – Quality of Life Resilience item bank	SCI-QOL Resilience	2	Victorson et al., 2015; Kuzu et al., 2022
Spinal Cord Injury – Quality of Life Self-esteem	SCI-QOL Self-esteem	1	Kalpakjian et al., 2015
Spinal Cord Injury – Quality of Life – Stigma	SCI-QOL – Stigma	1	Kisala et al., 2015
Spinal Cord Injury – Quality of Life – Stigma Short Form	SCI-QOL – Stigma SF	1	Kisala et al., 2015
Spine Oncology Study Group Outcomes 2.0	SOSGOQ 2.0	3	Versteeg et al., 2018; Gal et al., 2022; Yin et al., 2022
World Health Organization Quality of Life – Bref version	WHOQOL-BREF	6	Chapin, 2004; Lee, 2017; Lin, 2007; New, 2019; Odole, 2018; Salvador-De La Barrera, 2018
World Health Organization Quality of Life – Disabilities module	WHOQOL-DIS	1	Lee, 2017

The overall methodological quality of all the studies ranged from inadequate to very good, and the quality of evidence ranged from very low to high. As shown in Fig. 2 and Tables 2–5, all 77 studies reported at least one measurement property. Among them seventeen studies (37%) reported the content validity, allowing the review team to analyse this property (Neurogenic Bladder Symptom Score Questionnaire – NBSS, Patient Reported Impact of Spasticity Measure – PRISM, Quality of Life Profile for Adults with Physical Disabilities – QOLP-PD, Spinal Cord Injury – Quality of Life – SCI-QOL Pressure Ulcers, Spinal Cord Injury – Quality of Life – SCI-QOL Pressure Ulcers SF, Spinal Cord Injury – Quality of Life – SCI-QOL Anxiety, Spinal Cord Injury – Quality of Life – SCI-QOL Bank, Spinal Cord Injury – Quality of Life – SCI-QOL Economic, Spinal Cord Injury – Quality of Life – SCI-QOL Depression, Spinal Cord Injury – Quality of Life – SCI-QOL Positive Affect and Well-being, Spinal Cord Injury – Quality of Life – SCI-QOL Self-esteem, Spinal Cord Injury – Quality of Life – SCI-QOL Grief and Loss, Spinal Cord Injury – Quality of Life – SCI-QOL Psychological Trauma, Spinal Cord Injury – Quality of Life – SCI-QOL Bladder Management Difficulties, Spinal Cord Injury – Quality of Life – SCI-QOL Bowel Management Difficulties, Spinal Cord Injury – Quality of Life – SCI-QOL Bladder Complications, Spinal Cord Injury – Quality of Life Ability to Participate and Satisfactions with Social Roles and Activities item banks – SCI-QOL APSSRA) as can be seen in Table 2. All these instruments, for which the content validity was reported, were rated as having very low quality of evidence, except the PRISM, which had high quality. Although the study’s methodological quality in assessing this instrument was sufficient for relevance and comprehensiveness, the comprehensibility analysis was not clearly described.

Table 2.

Content validity.

					Content validity
					Relevance		Comprehensiveness		Comprehensibility
Instrument	Reference	Country (language) in which the questionnaire was evaluated	Number of items	Mode of administration	Meth. Quality	Rating	Meth. Quality	Rating	Meth. Quality	Rating
NBSS	[Guler et al., 2020]	Turkey	24	interview	I	–	I	–	I	–
	[Przydacz et al., 2020]	Poland	24	interview	I	–	I	–	I	–
PRISM	[Cook, 2007]	USA	55	interview	V	+	V	+	V	?
QOLP-PD	[Renwick, 2003]	Canada	102	interview	D	–	D	–	D	–
SCI-QOL Pressure ulcers item bank	[Kisala, 2015a]	USA	12	interview	A	±			A	±
SCI-QOL Pressure ulcers item bank SF	[Kisala, 2015a]	USA	12	interview	A	±			A	±
SCI-QOL Anxiety item bank	[Kisala, 2015b]	USA	25	phone interview and CAT program	A	±			A	±
SCI-QOL Economic item bank	[Tulsky, 2015a]	USA	28	face-to-face or telephone interview	A	±			A	±
SCI-QOL item banks (Methodology for the development and calibration)	[Tulsky, 2015b]	USA	14 unidimensional calibrated item banks and 3 calibrated scales across	phone interview and computer adaptive test (CAT)	A	±			A	±
SCI-QOL Depression Item Bank	[Tulsky, 2015c]	USA	28	phone interview and computer adaptive test (CAT)	A	±			A	±
SCI-QOL PAWB	[Bertisch et al., 2015]	USA	28	computer adaptive test (CAT)	A	±			A	±
SCI-QOL APSSRA	[Heinemann et al., 2015]	USA	27	computer adaptive test (CAT)	A	±			A	±
SCI-QOL Self-esteem	[Kalpakjian et al., 2015]	USA	23	computer adaptive test (CAT)	A	±			A	±
SCI-QOL Grief and Loss	[Kalpakjian et al., 2015]	USA	17	computer adaptive test (CAT)	A	±			A	±
SCI-QOL Psychological Trauma	[Kisala, 2015d]	USA	19	computer adaptive test (CAT)	A	±			A	±
SCI-QOL Bladder Management Difficulties	[Tulsky et al., 2015]	USA	15	computer adaptive test (CAT)	A	±			A	±
SCI-QOL Bowel Management Difficulties	[Tulsky et al., 2015]	USA	26	computer adaptive test (CAT)	A	±			A	±
SCI-QOL Bladder Complications	[Tulsky et al., 2015]	USA	5	computer adaptive test (CAT)	A	±			A	±

Overall rating: + = sufficient; − = insufficient; ? = indeterminate; +/− = inconsistent. ; V = very good; A = adequate; D = doubtfull; I = inadequate.

PRISM, Patient Reported Impact of Spasticity Measure; SCI-QOL Pressure Ulcers, Spinal Cord Injury – Quality of Life Pressure Ulcers; SCI-QOL Pressure Ulcers SF, SCI-QOL 7-item short form (SF) Pressure Ulcers; SCI-QOL Anxiety, Spinal Cord Injury – Quality of Life Anxiety; SCI-QOL Economic, Spinal Cord Injury – Quality of Life Economic; SCI-QOL Depression, Spinal Cord Injury – Quality of Life Depression; SCI-QOL PAWB, Spinal Cord Injury – Quality of Life Positive Affect and Well-being; SCI-QOL Self-esteem, Spinal Cord Injury – Quality of Life Self-esteem; SCI-QOL Grief and Loss, Spinal Cord Injury – Quality of Life Grief and Loss; SCI-QOL Psychological Trauma, Spinal Cord Injury – Quality of Life Psychological Trauma; SCI-QOL Resilience, Spinal Cord Injury – Quality of Life Resilience; SCI-QOL Bladder Management Difficulties, Spinal Cord Injury – Quality of Life Bladder Management Difficulties; SCI-QOL Bowel Management Dificulties, Spinal Cord Injury – Quality of Life Bowel Management Dificulties; SCI-QOL Bladder Complications, Spinal Cord Injury – Quality of Life Bladder Complications; SCI-QOL APSSRA, Spinal Cord Injury – Quality of Life Ability to Participate and Satisfactions with Social Roles and Activities(APSSRA) ; SCI-QOL – Stigma, Spinal Cord Injury – Quality of Life – Stigma; SCI-QOL – Stigma SF, Spinal Cord Injury – Quality of Life – Stigma Short Form.

Table 3.

Measurement properties – internal consistency, reliability and cross-cultural validity.

					Internal consistency			Reliability			Cross-cultural validity/measurement invariance
Instrument	Reference	Country (language) in which the questionnaire was evaluated	Number of items	Mode of administration	n	Meth. quality	Result (rating)	n	Meth. quality	Result (rating)	n	Meth. quality	Result (rating)
ATD PA	[Scherer, 2001]	USA	11 (ATD PA sections B and C)	not clearly reported	20	V	α = 0.80 (+)
I-QOL	[Schurch et al., 2007]	Belgium, France, and Switzerland	22	quastionnaires/interview	59 (SCI and Multiple sclerosis)	V	α = 0.93 (+)
KHQ	[Romero-Cullerés, 2011]	Spain	21	not clearly reported	120	V	α = 0.91 (?)
	[Liu, 2010]	United Kingdom	21	not clearly reported	142	V	α = 0.77 to 0.91 (?)
	[Karapolat et al., 2018]	Turkey	21	face-to-face interviews	92	V	α = 0.68-0.93 (?)	92	A	ICC = 0.69-0.94 (−)
LiSat-9	[Post, 2012]	Netherlands	9	interviews	225	V	α = 0.75 (?)
	[Geyh, 2010]	Australia, Brazil, Canada, Israel, South Africa, and the USA	9	not clearly reported				243	D	r = 0.86 (?)	243	V	Merged item 4 and 5 F(5) = 1.19; p = 0.095 (+)
LS Questions	[Post, 2012]	Netherlands	2	interview	225	V	α = 0.60 (?)
NBD	[Erdem et al., 2017]	Turkey	10	not clearly reported	42	V	α = 0.547 (−)	42	A	κ =  1.000, (p < 0.001)
NBSS	[Welk et al., 2014]	USA	22	interview	80 (SCI), 136 (MS) and 14 congenital neurogenic bladder	V	α = 0.89 (?)	131 individuals with SCI, MS and congenital neurogenic bladder	A	ICC (2,1) of the total score = 0.91 (+)
	[Guler et al., 2020]	Turkey	24	interview	84 (SCI) and 18 (MS)	V	α  = 0.893–0.910 (?)	84 (SCI) and 18 (MS)	D	ICC = 0.91 (95% CI 0.87–0.94, p < 0.001) (+)
	[Przydacz et al., 2020]	Poland	24	interview	101	V	α = 0.83 (?)	101	D	ICC (2,1) = 0.87 (0.80–0.92) (+)
	[Cintra et al., 2019]	Brazil	24	interview	66 (SCI), 1 (MS) and 1 non indentified	A	α = 0.81 (?)	46	A	ICC = 0.86 [0.76–0.92] (p < 0.0001) (+)
PRISM	[Cook, 2007]	EUA	55	interview	180	V	α = 0.74-0.96 (?)	180	D	ICC = 0.82-0.91 (+)
PWI	[Geyh, 2010]	Australia, Brazil, Canada, Israel, South Africa, and USA	7	not clearly reported				243	D	r = 0.92 (?)	243	V	Merged item 3 and 8 F(5) = 9.05 ; p = 0.000 Merged item 5 and 6 F(5) = 0.39; p = 0.858 (+)
QLI-SCI	[Kovacs, 2015]	Spain, United Kingdom	37	interview				77 (43 with SCI).	A	QLI/SCI (SCI) ICC =  0.830; Entire sample ICC = 0.801 (−)
QOLP-PD	[Renwick, 2003]	Canada	102	interview	cross-disabilities n = 27 and only SCI n = 40	V	α = 0,67–0,97 (cross-disability) and 0,84–0,98 (SCI) (?)
Qualiveen	[D’Ancona, 2009]	Brazil	39	self-administered	51 cross-disabilities	V	α = 0.75–0.90 (?)	51 cross-disabilities	A	ICC = 0.62 to 0.86 (−)
	[Bonniaud, 2011]	Italy	30	interview	128	V	α = 0.94 (?)	128	D	ICC = 0.77–0.90 (+)
	[Nikfallah et al., 2015]	Iran	30	self-administered	154 cross-disability	V	α = 0.82 to 0.95.(+)	154 cross-disability	I	ICC = 0.94 to .97(+)
	[Przydocz, 2020a]	Poland	30	interview	126	V	α = 0.87 (?)	126	D	ICC = 0,92 (+)
	[Costa, 2001]	France	30	questionnaire	268	V	α = 0.80 (?)	46	A	ICC = 0.85 and 0.92 (+)
	[Pannek, 2007]	Germany	30	interview	439	V	α = 0.80 to 0.91 (?)
Qualiveen SF	[Reuvers, 2017]	Netherlands	8	interview	57 SCI and 50 controls	V	α = 0.89–0.92 (?)	57 SCI and 50 controls	A	ICC = 0.94 LOA = –0.72 to 0.70 (+)
	[Przydocz, 2020a]	Poland	8	interview	126	V	α = 0.84 (?)	126	D	ICC = 0,93 (+)
	[Krebs et al, 2021]	Switzerland	8	self-report questionnaire	50(test) and 35 (retest)	V	α = 0.84–0.90 (?)	35 to 37	I	ICC = 0.91 to 0.94 (+)
	[Konstantinidis et al., 2020]	Greece	8	interview	71 SCI and 53 Multiple Sclerosis	V	α = 0.71 to 0.90 (+)	71 SCI and 53 Multiple Sclerosis	A	r = 0.26–0.88 (p < 0.05); p < 0.01; p < 0.001 (?)
SCI QoL-BDS	[Post, 2016]	Netherlands	3	self-report questionnaire	261	V	α = 0.81 (?)
	[Post, 2019]	Australia, Brazil, Netherlands, USA	3	self-report questionnaire	79	V	α = 0.84 to 0.86 (?)	79	A	ICC total = 0.66 to 0.80 (?)
	[New, 2019]	Australia, Brazil, India, Netherlands, and USA	3	survey or face-face interview	959	V	α = 0.84 (?)
	[Kunz et al., 2021]	Switzerland	3	interview	218	V	α = 0.82–0.90 (?)
	[Nijhawan et al., 2022]	India	3	not clearly reported				50	V	ICC = 0.91 to 0.98 (+)
	Pattanakuhar, 2020]	Thailand	3	interview	130	I	α  = 0.89 (?)	130	A	ICC = 0,73-0,86 (?)
SCI QoL-BDS version 2.0	[Post et al., 2022]	Australia, Brazil Netherlands, and USA	4	interview	565	V	α = 0.81 (+)
SCI-QOL Pressure ulcers item bank	[Kisala, 2015a]	USA	12	interview	189	D	α = 0.927 (−)
SCI-QOL Pressure ulcers SF	[Kisala, 2015a]	USA	7	interview	189	V	α = 0.874 (+)	245	V	ICC (2,1) = 0.79 (95% CI: 0.74 to 0.84) (+)
SCI-QOL Anxiety item bank	[Kisala, 2015b ]	USA	25	phone interview and CAT program	716	V	α = 0.946 (−)	245	V	ICC =  0.80 (95% CI, 0.75 to 0.84) (+)
SCI-QOL item banks (Methodology for the development and calibration)	[Tulsky, 2015b]	USA	14 unidimensional calibrated item banks and 3 calibrated scales across	phone interview and computer adaptive test (CAT)	825	V	α = 0.79–0,99 (?)	245	V	ICC = 0.74–0,96 (+)
SCI-QOL Depression Item Bank	[Tulsky, 2015c]	USA	28	phone interview and computer adaptive test (CAT)	716	V	α = 0.964 (+)	245	V	r = 0.80 (P < .001) and ICC (2,1) = 0.80 (95% CI: 0.75 to 0.84). (+)
SCI-QOL Stigma item bank	[Kisala, 2015c]	USA	23	computer adaptive test (CAT)	611	V	α = 0.936 (−)	245	V	r = 0.80 (p < 0.001) ICC = 0.79 (95% CI: 0.74 to 0.84) (−)
SCI-QOL Stigma item bank SF	[Kisala, 2015c]	USA	10	computer adaptive test (CAT)	611	V	α = 0.895 (−)	168	A	r = 0.84 (−)
SCI-QOL PAWB	[Bertisch et al., 2015]	USA	28	computer adaptive test (CAT)	717	V	α = 0.970 (−)	245	V	r = 0.78 and ICC (2,1) = 0.78 (95% CI = 0.72 to 0.82) (+)
SCI-QOL APSSRA	[Heinemann et al., 2015]	USA	27	computer adaptive test (CAT)	641	V	α = 0.95 (−)	245	V	ICC (2,1) for Ability to Participate = 0.74 (95% CI =  0.67, 0.79) and Satisfaction the ICC (2,1) = 0.77; 95% (CI = 0.72, 0.82) (+).
SCI-QOL Self-esteem	[Kalpakjian et al., 2015]	USA	23	computer adaptive test (CAT)	717	V	α = 0.950 (−)	245	V	ICC (2,1) = 0.84 (95% CI =  0.80 to 0.88) (+)
SCI-QOL Grief and Loss	[Kalpakjian et al., 2015]	USA	17	computer adaptive test (CAT)	717	V	α = 0.947 (−)	245	V	ICC (2,1) = 0,83 (95% CI = 0,78 to 0,87) (+)
SCI-QOL Psychological Trauma	[Kisala, 2015d]	USA	19	computer adaptive test (CAT)	717	V	α = 0.947 (−)	245	V	ICC (2,1) = 0,84 (95% CI = 0,80 to 0,88) (+)
SCI-QOL Resilience	[Kuzu et al., 2022]	USA	21	computer adaptive test (CAT)	133	D	α = 0.89 (−)
	[Victorson et al., 2015]	USA	21	computer adaptive test (CAT)	717	V	α = 0.95 (−)	245	V	ICC (2,1) = 0.83 (95%CI = 0.78 to 0.87) (+)
SCI-QOL Bladder Management Difficulties	[Tulsky et al., 2015]	USA	15	computer adaptive test (CAT)	757	V	α = 0.91 (−)	245	V	(ICC) (2,1) = 0.76 ( 95% CI =  0.70 to 0.81) (+)
SCI-QOL Bowel Management Difficulties	[Tulsky et al., 2015]	USA	26	computer adaptive test (CAT)	757	V	α = 0.95 (−)	245	V	ICC (2,1) = 0.74 (95% CI = 0.68 to 0.79) (+)
SCI-QOL Bladder Complications	[Tulsky et al., 2015]	USA	5	computer adaptive test (CAT)	757	V	α = 0.72 (−)	245	V	ICC (2,1) = 0.69 (95% CI =  0.61 to 0.76) (−)
Sense of Well-Being Inventory (SWBI)	[Chapin, 2004]	Canada	36	not clearly reported	132	V	α = 0.79 to 0.88 (?)
SF-36	[Forchheimer, 2004	USA	36	telephone	215	V	α = 0.76 to 0.90 (?)
	[Lin, 2007]	Taiwan	26	telephone interviews	187	V	α = 0.72-0.9 (?)	187	I	ICC intra-evaluator = 0.71-0.99 ; ICC inter = 0.41-0.98 (+)
	[Liu, 2010]	United Kingdom	36	not clearly reported	142	V	α = 0.76to 0.89 (?)
SF-36V	[Luther, 2006]	EUA	8	self-administered	359	I	α = 0.90 (−)
SOSGOQ	[Yin et al., 2022]	China	20	self-administered	120	V	α = 0.87 (+)	120	A	ICC = 0.72 to 0.81 (+)
	[Gal et al., 2022]	Netherlands	20	self-administered	147	V	α = 0.68 (?)	43	V	ICC  = 0.78 to 0.93 (+)
	[Versteeg, 2018]	Canada and USA	20	not clearly reported				243	D	r = 0.88 (−).	243	V	2 Items deleted: Got things I want F(5) = 4.01; p = 0.002; and Change nothing in life F(5) = 11.20; P = 0.000 (+)
SWLS	[Post, 2012]	Netherlands	5	self-report	225	V	α = 0.83 (?)
	[Geyh, 2010]	Australia, Brazil, Canada, Israel, South Africa, and USA	5	not clearly reported				243	D	r =  0.78 (?)	243	V	Merged item 1 and 5 F(5) = 1.47; p = 0.200 (+)
WHOQOL-5	[Geyh, 2010]	Australia, Brazil, Canada, Israel, South Africa, and USA	5	not clearly reported				243	D	r = 0.78 (?)	243	V	Merged item 1 and 5 F(5) = 1.47; p = 0.200 (+)
WHOQOL-BREF	[Lin, 2007]	Taiwan	28	telephone interviews	187	V	α = 0.75-0.87 (?)	187	I	ICC intra-evaluator =  0.84-0.98; ICC inter =  0.56-0.95 (+)
	[Salvador-De La Barrera, 2018]	Spain	26	interview	54	V	α = 0.887 (?)	54	A	ICC = 0.85 (+)
	[Lee, 2017]				85 patients (SCI = 58, stroke = 27).	V	α = 0.896 (?)	85 patients (SCI = 58, stroke = 27).	A	r = 0.896; p < 0.01 (?)
	[Odole, 2018]	Nigeria	26	self-administered				49 individuals with SCI and 49 controls	A	r = 0.322 to 0.766; p < 0.01 (?)
WHOQOL-DIS	[Lee, 2017]	Korea	39	interview	85 patients (SCI = 58, stroke = 27).	V	α = 0.737 (+)	85 patients (SCI = 58, stroke = 27).	A	r = 0.759; p < 0.01 (?)	85 patients (SCI = 58, stroke = 27).	V	no data reported (?)

Overall rating: + = sufficient; − = insufficient; ? = indeterminate; V = very good; A = adequate; D = doubtful; I = inadequate r = Pearson correlation; LOA = Limits of Agreement; ICC = intraclass correlation coefficient; α = Cronbach alpha; CI = Confidence Interval. ATD PA, Assistive Technology Device Predisposition Assessment; BRFSS, Behavioral Risk Factor Surveillance Survey; I-QOL, Incontinence Quality of Life Questionnaire; IADL, Instrumental activities of daily living; IBMSSP – EFAPD; Item Bank to Measure Systems, Services, and Policies: Environmental Factors Affecting People With Disabilities; KHQ, King’s Health questionnaire; LiSat-9, Life Satisfaction Questionnaire; LS Questions, Life Satisfaction questions; SF-36, Medical Outcomes Short-Form Health Survey; SF-12, Medical Outcomes Short-Form Health Survey – 12-Item Short Form; SF-36V, Medical Outcomes Short-Form Health Survey – Physical Functioning Scale for Use With Veterans With Spinal Cord Injury; SF-6D, Medical Outcomes Short-Form Health Survey – 6 dimensions; NBD – Neurogenic Bowel Dysfunction; NBSS – Neurogenic Bladder Symptom Score Questionnaire; PRISM, Patient Reported Impact of Spasticity Measure; PWI, Personal Well-Being Index; QLI-SCI, Quality of Life Index-Spinal Cord Injury version; QOLP-PD, Quality of Life Profile for Adults with Physical Disabilities; QWB, Quality of Well-Being scale; Qualiveen – SF, Qualiveen Short Form; SWLS, Satisfaction with Life Scale; SWBI, Sense of Well-Being Inventory; SCI QoL version 2.0 – International Spinal Cord Injury Quality of Life Basic Data Set version 2.0; SCIQL-23, 23-item Spinal Cord Injury – specific Health Related Quality of Life questionnaire; SCI-QOL Pressure Ulcers, Spinal Cord Injury – Quality of Life Pressure Ulcers; SCI-QOL Pressure Ulcers SF, SCI-QOL 7-item short form (SF) Pressure Ulcers; SCI-QOL Anxiety, Spinal Cord Injury – Quality of Life Anxiety; SCI-QOL Economic, Spinal Cord Injury – Quality of Life Economic; SCI-QOL Depression, Spinal Cord Injury – Quality of Life Depression; SCI-QOL PAWB, Spinal Cord Injury – Quality of Life Positive Affect and Well-being; SCI-QOL Self-esteem, Spinal Cord Injury – Quality of Life Self-esteem; SCI-QOL Grief and Loss, Spinal Cord Injury – Quality of Life Grief and Loss; SCI-QOL Psychological Trauma, Spinal Cord Injury – Quality of Life Psychological Trauma; SCI-QOL Resilience, Spinal Cord Injury – Quality of Life Resilience; SCI-QOL Bladder Management Difficulties, Spinal Cord Injury – Quality of Life Bladder Management Difficulties; SCI-QOL Bowel Management Dificulties, Spinal Cord Injury – Quality of Life Bowel Management Dificulties; SCI-QOL Bladder Complications, Spinal Cord Injury – Quality of Life Bladder Complications; SCI-QOL APSSRA, Spinal Cord Injury – Quality of Life Ability to Participate and Satisfactions with Social Roles and Activities(APSSRA); SCI-QOL – Stigma, Spinal Cord Injury – Quality of Life – Stigma; SCI-QOL – Stigma SF, Spinal Cord Injury – Quality of Life – Stigma Short Form; SOSGOQ, Spine Oncology Study Group Outcomes; WHOQOL-5, 5 satisfaction items from the World Health Organization Quality of Life; Assessment; WHOQOL-BREF, World Health Organization Quality of Life – Bref version; WHOQOL-DIS, World Health Organization Quality of Life – Disabilities module.

Table 4.

Measurement properties – validity.

					Structural			Hypotesing Testing						Criterion
								Convergent and Divergent			Known groups
Instrument	Reference	Country (language) in which the questionnaire was evaluated	Number of items	Mode of administration	n	Meth. quality	Result (rating)	n	Meth. quality	Result (rating)	n	Meth. quality	Result (rating)	n	Meth. quality	Result (rating)
ATD PA	[Scherer, 2001]	USA	11 (ATD PA sections B and C)	not clearly reported				20	V	Result in line with hypos (38+)
BRFSS	[Andresen, 1999]	USA	14	Survey				183	I	Result in line with all hypos	183	V	Result in line with 4 hypo (4+)
IADL	[Andresen, 1999]	USA	8	Survey				183	I	Result in line with all hypos	183	V	Result in line with 4 hypo (4+)
IBMSSP – EFAPD	[Lai, 2016]	USA	41	research platform	571	V	Unidimentional scale (CFI = .964; RMSEA = .058; all R2>0.3; all residual correlations < 0.2) (+)				571	V	Result in line with hypo (1+)
I-QOL	[Schurch et al., 2007]	Belgium, France, and Switzerland	22	quastionnaires/interview				59 (SCI and Multiple sclerosis)	I	Result in line with hypos (24+)	59 (SCI and Multiple sclerosis)	V	Result in line with all hypos
KHQ	[Romero-Cullerés, 2011]	Spain	21	not clearly reported							120	V	Result in line with hypo (3+) and not in line with 2 hypos (2-)
	[Liu, 2010]	United Kingdom	21	not clearly reported				142	V	Result in line with hypo (1+)	142	A	Result in line with hypo (1+)
	[Karapolat et al., 2018]	Turkey	21	face-to-face interviews				92	V	Result in line with 2 hypo (2+)	92	A	Result in line with hypo (1+)
LiSat-9	[Post, 2012]	Netherlands	9	interviews				225	V	Result in line with hypo (1+)	225	V	Result not in line with hypo (1-)	225	V	Result in line with hypo (1+)
	[Geyh, 2010]	Australia, Brazil, Canada, Israel, South Africa, and USA	9	not clearly reported	243	V	Unidimentionality rejected (Chi2 = 50.60, df = 18, p = .000) (−)
LS Questions	[Post, 2012]	Netherlands	2	interview				225	V	Result in line with all 2 hypos (2+)	225	V	Result not in line with hypo (3-)	225	V	Result in line with all 2 hypos (2+)
NBD	[Erdem et al., 2017]	Turkey	10	not clearly reported				42	D	Result in line with hypo (+)
NBSS	[Welk et al., 2014]	USA	22	interview	80 (SCI), 136 (MS) and 14 congenital neurogenic bladder	D	factors were greater than 0.60 (range 0.73 to 0.90) (?)				131 individuals with SCI, MS and congenital neurogenic bladder	I	Result in line with hypo (+)	131 individuals with SCI, MS and congenital neurogenic bladder	V	Result in line with hypo (+)
	[Guler et al., 2020]	Turkey	24	interview										84 (SCI) and 18 (MS)	V	Result not in line with hypo (−)
	[Przydacz et al., 2020]	Poland	24	interview				101	D	Result in line with all 3 hypos (3+)
	[Cintra et al., 2019]	Brazil	24	interview										46	I	Result not in line with hypo (−)
PRISM	[Cook, 2007]	USA	55	interview	180	I	unidimentionality not clear (−)				180	V	Result in line with 2 hypo (2+) Results not in line with 1 hypo’s (1-)
M-PRISM	[Sweatman et al., 2020]	USA	37	survey	1080	very good	Residual variance > 10% (11.6 to 21.9 among subscales) Person logit between −5.39 to 6.13 among subscales(−)				1080	very good	Result in line with 1 hypo (1+)
PWI	[Geyh, 2010]	Australia, Brazil, Canada, Israel, South Africa, and USA	7	not clearly reported	243	V	Unidimentionality partially supported (Chi2 = 15.62, df = 16, p = .480)
QLI-SCI	[Kovacs, 2015]	Spain; United Kingdom	37	interview				77 (43 with SCI).	V	Result in line with 2 hypo (2+)
	[May, 2001]	Canada	36	interview	11	I	Not satifctory, need futher studies (−)
	[May & Waren, 2002]	Canada	37	interview	92	I	Factor analysis resulted in a five-factor structure that difered from the four-domain model of the original QLI, need further studies (?)	92	V	Result in line with 4 hypo (4+) Results not in line with 1 hypo (1-)	92	V	Results not in line with 2 hypo (2-)
QOLP-PD	[Renwick, 2003]	Canada	102	interview				cross-disabilities n = 27	V	Results in line with 2 hypos (2+)
Qualiveen	[D’Ancona, 2009]	Brazil	39	self-administered				51 cross-disabilities	I	Result in line with 1 hypo (1+)
	[Bonniaud, 2011]	Italy	30	interview				128	V	Result in line with all hypos (+)
	[Nikfallah et al., 2015]	Iran	30	self-administered	154 cross-disability	V	Unidimentional scale χ2  = 815.27, df = 399, (RMSEA = 0.078, CFI = 0.95, NNFI = 0.94).(+)	154 cross-disability	V	Result in line with all hypos (+)	154 cross-disability	V	Result in line with all hypos
	[Przydocz, 2020a]	Poland	30	interview				178	A	r = 0.693 and P < 0.001 (+)
	[Costa, 2001]	France	30	quastionnaires				268	V	Result in line with 3 hypos (−)	268	V	Result in line with 3 hypos (−)
	[Pannek, 2007]	Germany	30	interview							439	V	Result in line with all hypos
Qualiveen SF	[Reuvers, 2017]	Netherlands	30	interview				57 SCI and 50 controls	A	Result in line with 2 hypo (2+)	57 SCI and 50 controls	V	Result in line with 1 hypo (1+)
	[Przydocz, 2020a]	Poland	8	interview										178	V	r = 0.611 (p < 0.001) (+)
	[Krebs et al, 2021]	Switzerland	8	self-report questionnaire				50	D	Result in line with hypo (+)				50(test) and 35 (retest)	V	Result in line with hypo (+)
	[Konstantinidis, 2020]	Greece	8	interview	71 SCI and 53 Multiple sclerosis.	V	[χ2 (14)  = 19.133, GFI = 0.964, NFI = 0.954, RMSEA = 0.055, CFI = 0.987] (+)							71 SCI and 53 Multiple sclerosis.	V	(r =  0.36-0.76, p < 0.001) (−)
QWB	[Andresen, 1999]	USA	71	survey				183	I	Result in line with 3 hypo (3+) Results not in line with 10 hypo (10-)	183	V	Result in line with 2 hypo (2+) Results not in line with 2 hypo (2-)
SCI QoL-BDS	[Post, 2016]	Netherlands	3	self-report questionnaire				261	I	Result in line with 4 hypo (4+)
	[New, 2019]	Australia, Brazil, India, Netherlands, and USA	3	survey or face-face interview				959	I	Result in line with 4 hypo (4+)	959	V	Result in line with 3 hypo (3+)
	[Pattanakuhar, 2020]	Thailand	3	interview				130	V	Result in line with 1 hypo (1+)
SCI QoL-BDS version 2.0	[Post et al., 2022]	Australia, Brazil Netherlands, and USA	4	interview	565	V	Infit and outfit ranged 0.62-0.91 (+)
SCIQL-23	[Jain, 2007]	Iran	23	self-complete and phone administration							356	V	hypo not clearly described (−)
	[Ebrahimzadeh, 2014]	USA	23	not clearly reported				52	A	Result in line with 3 hypo (3+) Results not in line with 11 hypo’s ( 11 -)
SCI-QOL Pressure ulcers item bank	[Kisala, 2015a]	USA	12	interview	189	V	Unidimentional scale (CFI = 0.961, TLI = 0.952, RMSEA = 0.124) (−)
SCI-QOL Anxiety item bank	[Kisala, 2015b ]	USA	25	phone interview and CAT program	716	V	Unidimentional scale (CFI = 0.953; RMSEA =  0.069) (−)	465	V	Result in line with hypo (+)
SCI-QOL Economic item bank	[Tulsky, 2015a]	USA	28	face-to-face or telephone interview	305	V	Unidimentional sclae (CFI = 0.939, RMSEA = 0.089) (−)
SCI-QOL item banks (Methodology for the development and calibration)	[Tulsky, 2015b]	USA	14 unidimensional calibrated item banks and 3 calibrated scales across	phone interview and computer adaptive test (CAT)	825	V	(CFI = 0.914–0.999, RMSEA =  0.039–0.124) (?)	825	V	Result in line with hypo (?)
SCI-QOL Depression Item Bank	[Tulsky, 2015c]	USA	28	phone interview and computer adaptive test (CAT)	716	V	Unidimentional scale (CFI = 0.968; RMSEA =  0.066) (−)	716	V	Result in line with hypo (+)
SCI-QOL Stigma item bank	[Kisala, 2015c]	USA	23	computer adaptive test (CAT)	611	V	Unidimentional scale (CFI = 0.941, TLI = 0.935, and RMSEA = 0.088) (−)
SCI-QOL PAWB	[Bertisch et al., 2015]	USA	28	computer adaptive test (CAT)	717	V	Unidimensional scale (CFI = 0.947; RMSEA = 0.094) (−)
SCI-QOL APSSRA	[Heinemann et al., 2015]	USA	27	computer adaptive test (CAT)	641	V	Unidimensional scale (CFI = 0.968; RMSEA =  0.074) (−)
SCI-QOL Self-esteem	[Kalpakjian et al., 2015]	USA	23	computer adaptive test (CAT)	717	V	Unidimensional scale (CFI = 0.946; RMSEA = 0.087) (−)
SCI-QOL Grief and Loss	[Kalpakjian et al., 2015]	USA	17	computer adaptive test (CAT)	717	V	Unidimensional scale (CFI = 0.976; RMSEA = 0.078) (−)
SCI-QOL Psychological Trauma	[Kisala, 2015d]	USA	19	computer adaptive test (CAT)	717	V	Unidimensional scale (CFI = 0.976; RMSEA = 0.078) (−)
SCI-QOL Resilience	[Kuzu et al., 2022]	USA	21	computer adaptive test (CAT)							133	V	Result in line with 3 hypos (3+)	133	V	Result in line with 3 hypos (3+)
	[Victorson et al., 2015]	USA	21	computer adaptive test (CAT)	717	V	Unidimensional scale (CFI = 0.976; RMSEA = 0.078) (−)
SCI-QOL Bladder Management Difficulties	[Tulsky et al., 2015]	USA	15	computer adaptive test (CAT)	757	V	Unidimensional scale (CFI = 0.965; RMSEA = 0.093) (−)
SCI-QOL Bowel Management Difficulties	[Tulsky et al., 2015]	USA	26	computer adaptive test (CAT)	757	V	Unidimensional scale (CFI = 0.955; RMSEA = 0.078) (−)
SCI-QOL Bladder Complications	[Tulsky et al., 2015]	USA	5	computer adaptive test (CAT)	757	V	Unidimensional scale (CFI = 0.992, RMSEA = 0.050) (−)
SWBI	[Chapin, 2004]	Canada	36	not clearly reported	132	I	Factors ranged from 0.43 to 0.99 (−)	132	V	Result in line with 4 hypos (4+)
SF-12	[Andresen,1999]	USA	12	survey				183	I	Result in line with all hypos (−)	183	V	Result in line with 4 hypo (4+)
SF-36	[Forchheimer, 2004	USA	36	telephone				215	V	Result in line with hypo (+)	215	V	Result in line with hypo (+)
	[Lin, 2007]	Taiwan	26	telephone interviews				187	V	Result in line with hypo (+)	187	V	Result in line with 4 hypos (4+) Results not in line with 1 hypo (1-)
	[Liu, 2010]	United Kingdom	36	not clearly reported				142	V	Result in line with hypo (+)	142	V	Result in line with hypo (+)
	[Andresen,1999]	USA	36	survey				183	I	Result in line with all hypos (+)	183	V	Result in line with 4 hypo (4+)
SF-36V	[Luther, 2006]	USA	8	self-administered	359	A	8 items both individual item and total model fit statistics supported model fit based on likelihood-ratio chi-square statistics, with all of the p > .05	359	I	Result in line with all 3 hypos (3+)	359	V	Result in line with hypo (+)
SF-6D	[Engel, 2014]	Canada	11	telephone interview, self-completed postal questionnaire or in-person				358	V	Result in line with all 5 hypos (5+)	358	V	Result in line with 3 hypos (3+) Result not in line with 1 hypo (1-)
	[Lee, 2008]	Australia	11	self-report							305	V	Result in line with 3 hypos (3+)
SOS-GOQ2.0	[Yin et al., 2022]	China	20					120	D	Result in line with hypos (2+)	30	V	Result in line with hypo (+)	120	V	Result in line with hypo (+)
	[Gal et al., 2022]	Netherlands	20	self-administered				147	inadequate	Result in line with 5 hypos (5+)	86	very good	Result in line with hypo (+)
	147	I	Result in line with 5 hypos (5+)	86	V	Result in line with hypo (+)
	[Versteeg, 2018]	Canada and USA	20	not clearly reported	133	V	A model fit (RMSEA, 0.074 [90% CI, 0.055- 0.092]; CFI, 0.928; and SRMR, 0.06). (+)	153	V	Result in line with all 2 hypos (2+)	162	V	Result in line with hypo (+)	162	V	Result in line with hypo (1+)
SWLS	[Geyh, 2010]	Australia, Brazil, Canada, Israel, South Africa, and USA	5	not clearly reported	243	V	Unidimentionality rejected (Chi2 = 78.54, df = 10, p = .000) (−)
	[Post, 2012]	Netherlands	5	self-report				225	V	Result in line with 3 hypos (3+) Result not in line with 1 hypo (1-)	225	V	Result not in line with hypo (3-)	225	V	Result not in line with hypo (−)
WHOQOL-5	[Geyh, 2010]	Australia, Brazil, Canada, Israel, South Africa, and USA	5	not clearly reported	243	V	Unidimensional scale (Chi2 = 16.43, df = 10, p = .088) One of the items misfitted the model.
WHOQOL-BREF	[Lin, 2007]	Taiwan	28	telephone interviews				187	V	Result in line with almost all hypos (+)	187	V	Result in line with hypos (3+) Results not in line with 3 hypos (3-)
	[Salvador-De La Barrera, 2018]	Spain	26	interview				54	I	Result in line with 3 hypos (3+)	54	V	Result in line with 2 hypos (2+) Result in line with 3 hypos (3+)
	[Odole,2018]	Nigeria	26	self-administered							49 individuals with SCI and 49 controls	A	Result in line with hypo (+)	49 individuals with SCI and 49 controls	V	Result in line with hypo (+)
WHOQOL-DIS	[Lee, 2017]	Korea	39	interview	85 patients (SCI = 58, stroke = 27).	I	The three-domain structure model with a higher-order factor fit well (CFI = 0.89, NFI = 0.83, RMSEA = 0.13, x2 = 152.46, df = 61, p < 0.001) (+)	85 patients (SCI = 58, stroke = 27).	V	Result in line with hypo (+)

Overall rating: + = sufficient; − = insufficient; ? = indeterminate; V = very good; A = adequate; D = doubtfull; I = inadequate CFI, Confirmatory Fit Index; RMSEA, Root Mean Square Error ofApproximation; ATD PA, Assistive Technology Device Predisposition Assessment; BRFSS, Behavioral Risk Factor Surveillance Survey; I-QOL, Incontinence Quality of Life Questionnaire; IADL, Instrumental activities of daily living; IBMSSP – EFAPD; Item Bank to Measure Systems, Services, and Policies: Environmental Factors Affecting People With Disabilities; KHQ, King’s Health questionnaire; LiSat-9, Life Satisfaction Questionnaire; LS Questions, Life Satisfaction questions; SF-36, Medical Outcomes Short-Form Health Survey; SF-12, Medical Outcomes Short-Form Health Survey – 12-Item Short Form; SF-36V, Medical Outcomes Short-Form Health Survey – Physical Functioning Scale for Use With Veterans With Spinal Cord Injury; SF-6D, Medical Outcomes Short-Form Health Survey – 6 dimensions; M-PRISM, Modified Patient Reported Impact of Spasticity; PRISM, Patient Reported Impact of Spasticity Measure; PWI, Personal Well-Being Index; QLI-SCI, Quality of Life Index-Spinal Cord Injury version; QOLP-PD, Quality of Life Profile for Adults with Physical Disabilities; QWB, Quality of Well-Being scale; Qualiveen – SF, Qualiveen Short Form; SWLS, Satisfaction with Life Scale; SWBI, Sense of Well-Being Inventory; SCIQL-23, 23-item Spinal Cord Injury – specific Health Related Quality of Life questionnaire; SCI-QOL Pressure Ulcers, Spinal Cord Injury – Quality of Life Pressure Ulcers; SCI-QOL Pressure Ulcers SF, SCI QoL version 2.0 – International Spinal Cord Injury Quality of Life Basic Data Set version 2.0SCI-QOL 7-item short form (SF) Pressure Ulcers; SCI-QOL Anxiety, Spinal Cord Injury – Quality of Life Anxiety; SCI-QOL Economic, Spinal Cord Injury – Quality of Life Economic; SCI-QOL Depression, Spinal Cord Injury – Quality of Life Depression; SCI-QOL PAWB, Spinal Cord Injury – Quality of Life Positive Affect and Well-being; SCI-QOL Self-esteem, Spinal Cord Injury – Quality of Life Self-esteem; SCI-QOL Grief and Loss, Spinal Cord Injury – Quality of Life Grief and Loss; SCI-QOL Psychological Trauma, Spinal Cord Injury – Quality of Life Psychological Trauma; SCI-QOL Resilience, Spinal Cord Injury – Quality of Life Resilience; SCI-QOL Bladder Management Difficulties, Spinal Cord Injury – Quality of Life Bladder Management Difficulties; SCI-QOL Bowel Management Dificulties, Spinal Cord Injury – Quality of Life Bowel Management Dificulties; SCI-QOL Bladder Complications, Spinal Cord Injury – Quality of Life Bladder Complications; SCI-QOL APSSRA, Spinal Cord Injury – Quality of Life Ability to Participate and Satisfactions with Social Roles and Activities(APSSRA); SCI-QOL – Stigma, Spinal Cord Injury – Quality of Life – Stigma; SCI-QOL – Stigma SF, Spinal Cord Injury – Quality of Life – Stigma Short Form; SOSGOQ, Spine Oncology Study Group Outcomes; WHOQOL-5, 5 satisfaction items from the World Health Organization Quality of Life; Assessment; WHOQOL-BREF, World Health Organization Quality of Life – Bref version; WHOQOL-DIS, World Health Organization Quality of Life – Disabilities module.

Table 5.

Measurement properties – other properties.

					Measurement error			Interpretability		Responsiveness
Instrument	Reference	Country (language) in which the questionnaire was evaluated	Number of items	Mode of administration	n	Meth. quality	Result (rating)	n	Result (rating)	n	Meth. quality	Result (rating)
BRFSS	[Andresen, 1999]	USA	14	Survey				183	8 of the BRFSS "days" questions exhibited ceiling effects, and 4 also showed floor effects.
IADL	[Andresen, 1999]	USA	8	Survey				183	no floor or ceiling effect were observed
I-QOL	[Schurch et al., 2007]	Belgium, France, and Switzerland	22	quastionnaires/interview				59 (SCI and Multiple sclerosis)	MID was approximately 4 to 11 points when defined as that corresponding to a small and medium effect size respectively . Small floor effects were observed for the social embarrassment and the psychosocial impact domain	59 (SCI and Multiple sclerosis)	V	Decrease, >25% fewer incontinence episodes (p < 0.05)
NBD	[Erdem et al., 2017]	Turkey	10	not clearly reported						42	I	Significant positive correlation between Global Rating of Change scale and the change in total NBD score at the end of 2 months (r =  0.821, P =  0.007) (+)
QLI-SCI	[Kovacs, 2015]	Spain, UK	37	interview				77 (43 with SCI).	Floor and ceeling effects total score range = 12.75 - 28.87 (2.3% both)
Qualiveen	[Bonniaud, 2011]	Italy	30	interview				128	Minimally important difference values in the four domains varied from 0.34 to 0.47.	128	V	The SRMs were 1.81 for bothers with limitations, 2.23 for frequency of limitations, 1.94 for fears, 1.23 for feelings, and 2.17 for the overall score (?)
	[Nikfallah et al., 2015]	Iran	30	self-administered				154	Floor and ceiling effects were not found
QWB	[Andresen, 1999]	USA	71	survey				183	No floor or ceiling effect were observed
SCI QoL-BDS	[Post, 2016]	Netherlands	3	self-report questionnaire				261	No floor or ceiling effects were found (0.4 to 1.5%).
	[New, 2019]	Australia, Brazil, India, Netherlands, and USA	3	survey or face-face interview				959	There were few notable floor or ceiling effects, 19% of the Indian inpatients gave their physical health the lowest score, and 30% of the Brazilian sample gave their mental health the best positive score
	[Kunz et al., 2021]	Switzerland	3	interview				218	Longitudinal measurement invariance: scaled Δχ2(3) = 6.77, p = .080 and ΔCFI = − .004.
	[Pattanakuhar, 2020]	Thailand	3	interview				130	Floor and the ceiling effects on the Thai version of QoL-BSD (3.8% and 14.6%, respectively)
SCIQL-23	[Ebrahimzadeh, 2014]	USA	23	not clearly reported				52	No floor or ceiling effect were observed
Sense of Well-Being Inventory (SWBI)	[Chapin, 2004]	Canada	36	not clearly reported				183	No floor or ceiling effect were observed
SF-36	[Lin, 2007]	USA	36	telephone interviews				187	SF-36 had percentages of ceiling (0.4%-63.8%) and floor (0.4%- 28.1%) effects	30	V	5 domains ranged from 0.00 to 0.44, and 2 domains ranged from 0.60 to 0.92 (+)
	[Liu, 2010]	United Kingdom	36	not clearly reported				183	No floor or ceiling effect were observed
SF-36V	[Luther, 2006]	EUA	8	self-administered						358	V	improvement in their health status compared with 1 year ago also showed improvement in SF-6D scores (ES = 0.27) (−)
SF-12	[Andresen, 1999]	USA	12	survey				177	No floor or ceiling effect were observed for physical or mental health summary
SF-6D	[Engel, 2014]	Canada	11	telephone interview, self-completed postal questionnaire or in-person				305	MID ranged from 0.03 to 0.10. An apparent floor effect (37%) in the Physical Functioning dimension in the overall sample by the tetraplegia group (63%). There were no notable floor effects in the Australian sample	138	V	For participants who developed UTI, the overall age- and sex-adjusted utility score was 0.68 (SD 0.20) before and 0.57 (SD 0.15) after UTI (−)
WHOQOL-BREF	[Lin, 2007]	Taiwan	28	telephone interviews				187	WHOQOL-BREF had lower percentages of ceiling (0.0%-0.4%) and floor effects (0.0%-1.3%)
	[Salvador-De La Barrera, 2018]	Spain	26	interview				54	Floor effect percentage was <2.0% (score = 0) in all domains, as well as the ceiling effect percentage (score = 100).

Overall rating: + = sufficient; − = insufficient; ? = indeterminate; V = very good; A = adequate; D = doubtfull; I = inA ATD PA, Assistive Technology Device Predisposition Assessment; BRFSS, Behavioral Risk Factor Surveillance Survey; I-QOL, Incontinence Quality of Life Questionnaire; IADL, Instrumental activities of daily living; SCI QoL-BDS, International Spinal Cord Injury Quality of Life Basic Data Set; IBMSSP – EFAPD; Item Bank to Measure Systems, Services, and Policies: Environmental Factors Affecting People With Disabilities; KHQ, King’s Health questionnaire; LiSat-9, Life Satisfaction Questionnaire; LS Questions, Life Satisfaction questions; SF-36, Medical Outcomes Short-Form Health Survey; SF-12, Medical Outcomes Short-Form Health Survey – 12-Item Short Form; SF-36V, Medical Outcomes Short-Form Health Survey – Physical Functioning Scale for Use With Veterans With Spinal Cord Injury; SF-6D, Medical Outcomes Short-Form Health Survey – 6 dimensions; PRISM, Patient Reported Impact of Spasticity Measure; PWI, Personal Well-Being Index; QLI-SCI, Quality of Life Index-Spinal Cord Injury version; QOLP-PD, Quality of Life Profile for Adults with Physical Disabilities; QWB, Quality of Well-Being scale; Qualiveen – SF, Qualiveen Short Form; SWLS, Satisfaction with Life Scale; SWBI, Sense of Well-Being Inventory; SCIQL-23, 23-item Spinal Cord Injury – specific Health Related Quality of Life questionnaire; SCI-QOL Pressure Ulcers, Spinal Cord Injury – Quality of Life Pressure Ulcers; SCI-QOL Pressure Ulcers SF, SCI-QOL 7-item short form (SF) Pressure Ulcers; SCI-QOL Anxiety, Spinal Cord Injury – Quality of Life Anxiety; SCI-QOL Economic, Spinal Cord Injury – Quality of Life Economic; SCI-QOL Depression, Spinal Cord Injury – Quality of Life Depression; SCI-QOL PAWB, Spinal Cord Injury – Quality of Life Positive Affect and Well-being; SCI-QOL Self-esteem, Spinal Cord Injury – Quality of Life Self-esteem; SCI-QOL Grief and Loss, Spinal Cord Injury – Quality of Life Grief and Loss; SCI-QOL Psychological Trauma, Spinal Cord Injury – Quality of Life Psychological Trauma; SCI-QOL Resilience, Spinal Cord Injury – Quality of Life Resilience; SCI-QOL Bladder Management Difficulties, Spinal Cord Injury – Quality of Life Bladder Management Difficulties; SCI-QOL Bowel Management Dificulties, Spinal Cord Injury – Quality of Life Bowel Management Dificulties; SCI-QOL Bladder Complications, Spinal Cord Injury – Quality of Life Bladder Complications; SCI-QOL APSSRA, Spinal Cord Injury – Quality of Life Ability to Participate and Satisfactions with Social Roles and Activities(APSSRA); SCI-QOL – Stigma, Spinal Cord Injury – Quality of Life – Stigma; SCI-QOL – Stigma SF, Spinal Cord Injury – Quality of Life – Stigma Short Form; SOSGOQ, Spine Oncology Study Group Outcomes; WHOQOL-5, 5 satisfaction items from the World Health Organization Quality of Life; Assessment; WHOQOL-BREF, World Health Organization Quality of Life – Bref version; WHOQOL-DIS, World Health Organization Quality of Life – Disabilities module.

The measurement properties most frequently assessed by the studies were internal consistency, reliability, and structural validity, followed by construct validity (Tables 3 and 4). Of the 49 identified instruments, internal consistency was evaluated in 38 (77%), reliability was tested in 32 (65%) instruments, structural validity in 31 (63%) and construct validity in 31(63%) instruments. Using the GRADE analysis, three instruments (International Spinal Cord Injury Quality of Life Basic Data Set version 2.0 – SCI QoL-BDS version 2.0, Spine Oncology Study Group Outcomes – SOSGOQ, World Health Organization Quality of Life – Disabilities module – WHOQOL – DIS) were rated as having high-quality evidence for sufficient internal consistency; six instruments (PRISM, Qualiveen, SCI-QOL Pressure Ulcers, SCI-QOL Depression, SCI-QOL Positive Affect and Well-being and SCI-QOL Resilience) were rated as having high-quality evidence for sufficient reliability; five instruments (Item Bank to Measure Systems, Services, and Policies: Environmental Factors Affecting People With Disabilities – IBMSSP – EFAPD, Modified Patient Reported Impact of Spasticity Measure – M PRISM, Qualiveen, SOSGOQ, WHOQOL – DIS) were rated as having high-quality evidence for sufficient structural validity; fourteen instruments (ATD PA, Life Satisfaction Questionnaire – LiSat-9, Life Satisfaction questions – LS Questions, Neurogenic Bowel Dysfunction – NBD, PRISM, M-PRISM, Quality of Life Index-Spinal Cord Injury version – QLI-SCI, Quality of Life Profile for Adults with Physical Disabilities – QOLP-PD, Qualiveen SF, Qualiveen, SCI-QOL BDS, SCI-QOL Anxiety, SCI-QOL Depression, World Health Organization Quality of Life – Bref version – WHOQOL-BREF, WHOQOL – DIS) were rated as having high-quality evidence for sufficient construct validity, and four instruments (LiSat-9, Personal Well-Being Index – PWI, Satisfaction with Life Scale – SWLS and 5 satisfaction items from the World Health Organization Quality of Life Assessment – WHOQoL-5) were rated as having high-quality evidence for sufficient cross-cultural validity/measurement invariance.

The best evidence for each instrument is synthesised and shown in Table 6. Finally, the synthesised best evidence was used to make final recommendations according to the COSMIN guidelines (18) for all PROMs, as reported in Table 7.

Table 6.

Summary of the confidence level of evidence and overall rating according to COSMIN taxonomy.

Intrument	Reliability						Validity
				Content validity				Construct validity
	Internal consistency	Reliability	Measurement error	Relevance	Comprehensiveness	Comprehensibility	Content validity	Structural validity	Criterion validity	Construct validity (Hypothesis testing)	Cross-cultural/ Measurement invariance	Responsiveness
ATD PA	high (?)									high (+)
BRFSS										very low (+)
IADL										very low (−)
IBMSSP – EFAPD								high (+)		moderate (−)
I-QOL	high (?)									low (+)		low (+)
KHQ	high (?)	low (−)								moderate (+)
LiSat-9	high (?)	low (?)						high (−)	high (?)	high (+)	high (+)
NBD	low (−)	low (+)								low (−)		very low (+)
LS Questions	high (?)								high (+)	high (+)
NBSS	moderate (?)	moderate (+)		very low (−)	very low (−)	very low (−)	very low (−)	very low (?)	moderate (−)	very low (+)
PRISM	high (?)	high (+)		high (+)	high (+)	high (?)	high (?)	very low (−)		high (+)
M-PRISM								high (−)		high (+)
PWI		low (?)						high (−)			high (+)
QLI-SCI		moderate (−)						low (−)		high (+)
QOLP-PD	high (?)			very low (±)	very low (±)	very low (−)	very low (−)			high (+)
Qualiveen SF	moderate (?)	Moderate (+)						moderate (+)	moderate (+)	high (+)
Qualiveen	high (?)	high (+)						high (+)	high (?)	high (+)		high (?)
QWB										very low (−)
SCIQL-23	very low (?)									high (−)
SCI-QOL BDS	very low (?)	high (+)								high (+)
SCI-QOL BDS version 2.0	high (+)	high (+)
SCI-QOL Pressure Ulcers	low (−)			high (±)	very low (±)	high (±)	very low (±)	high (−)
SCI-QOL Pressure Ulcers SF	high (−)	high (+)		high (±)	very low (±)	high (±)	very low (±)
SCI-QOL Anxiety	high (−)	moderate (+)		low (−)	very low (±)	high (−)	very low (−)			high (+)
SCI-QOL Bank	high (?)	moderate (+)		high (±)	very low (±)	low (−)	very low (±)	high (?)
SCI-QOL Economic				high (±)	very low (±)	high (−)	very low (±)	high (−)
SCI-QOL Depression	high (−)	high (+)		high (±)	very low (±)	high (−)	very low (±)	high (−)		high (+)
SCI-QOL Positive Affect and Well-being	high (−)	high (+)		high (±)	very low (−)	high (−)	very low (−)	high (−)
SCI-QOL Self-esteem	high (−)	moderate (+)		high (±)	very low (−)	high (−)	very low (−)	high (−)
SCI-QOL Grief and Loss	high (−)	moderate (+)		high (±)	very low (−)	high (−)	very low (−)	high (−)
SCI-QOL Psychological Trauma	high (−)	moderate (+)		high (±)	very low (−)	high (−)	very low (−)	high (−)
SCI-QOL Resilience	high (−)	high (+)					high (+)	high (−)
SCI-QOL Bladder Management Difficulties	high (−)	moderate (+)		high (±)	very low (−)	high (−)	very low (−)	high (−)
SCI-QOL Bowel Management Dificulties	high (−)	moderate (+)		high (±)	very low (−)	high (−)	very low (−)	high (−)
SCI-QOL Bladder Complications	high (−)	moderate (+)		high (±)	very low (−)	high (−)	very low (−)	high (−)
SCI-QOL APSSRA	high (−)	moderate (+)		high (±)	very low (−)	high (−)	very low (−)	high (−)
SCI-QOL – Stigma	high (−)	moderate (+)						high (−)
SCI-QOL – Stigma SF	high (−)	moderate (+)
SF-6D										high (−)		high (−)
SF-12										low (±)		moderate (+)
SF-36	high (?)	very low (+)						moderate (?)		very low (−)		high (+)
SF-36 – MHI-5 (mental health)	moderate (?)								high (?)	very low (?)
SF-36V	very low (−)							moderate (?)		very low (−)
SOSGOQ	high (+)	moderate (+)						high (+)	high (+)	moderate (+)
SWBI	high (?)							very low (−)		high (−)
SWLS	high (?)	moderate (?)						high (−)	high (?)	moderate (+)	high (+)
WHOQoL-5		moderate (?)						high (−)			high (+)
WHOQOL-BREF	high (?)	high (?)						high (?)		high (+)		high (+)
WHOQOL – DIS	high (+)	low (?)						high (+)		high (+)	high (?)

Overall rating: + = sufficient; − = insufficient; ? = indeterminate; +/− = inconsistent.

Quality of evidence (GRADE aproach): high; moderate; low; very low.

ATD PA, Assistive Technology Device Predisposition Assessment; BRFSS, Behavioral Risk Factor Surveillance Survey; I-QOL, Incontinence Quality of Life Questionnaire; IADL, Instrumental activities of daily living; IBMSSP – EFAPD; Item Bank to Measure Systems, Services, and Policies: Environmental Factors Affecting People With Disabilities; KHQ, King’s Health questionnaire; LiSat-9, Life Satisfaction Questionnaire; LS Questions, Life Satisfaction questions; SF-36, Medical Outcomes Short-Form Health Survey; SF-12, Medical Outcomes Short-Form Health Survey – 12-Item Short Form; SF-36V, Medical Outcomes Short-Form Health Survey – Physical Functioning Scale for Use With Veterans With Spinal Cord Injury; SF-6D, Medical Outcomes Short-Form Health Survey – 6 dimensions; NBD – Neurogenic Bowel Dysfunction; NBSS – Neurogenic Bladder Symptom Score Questionnaire; PRISM, Patient Reported Impact of Spasticity Measure; M-PRISM, Modified Patient Reported Impact of Spasticity; PWI, Personal Well-Being Index; QLI-SCI, Quality of Life Index-Spinal Cord Injury version; QOLP-PD, Quality of Life Profile for Adults with Physical Disabilities; QWB, Quality of Well-Being scale; Qualiveen – SF, Qualiveen Short Form; SWLS, Satisfaction with Life Scale; SWBI, Sense of Well-Being Inventory; SCI QoL BDS – International Spinal Cord Injury Quality of Life Basic Data Set; SCI QoL version 2.0 – International Spinal Cord Injury Quality of Life Basic Data Set version 2.0; SCIQL-23, 23-item Spinal Cord Injury – specific Health Related Quality of Life questionnaire; SCI-QOL Pressure Ulcers, Spinal Cord Injury – Quality of Life Pressure Ulcers; SCI-QOL Pressure Ulcers SF, SCI-QOL 7-item short form (SF) Pressure Ulcers; SCI-QOL Anxiety, Spinal Cord Injury – Quality of Life Anxiety; SCI-QOL Economic, Spinal Cord Injury – Quality of Life Economic; SCI-QOL Depression, Spinal Cord Injury – Quality of Life Depression; SCI-QOL PAWB, Spinal Cord Injury – Quality of Life Positive Affect and Well-being; SCI-QOL Self-esteem, Spinal Cord Injury – Quality of Life Self-esteem; SCI-QOL Grief and Loss, Spinal Cord Injury – Quality of Life Grief and Loss; SCI-QOL Psychological Trauma, Spinal Cord Injury – Quality of Life Psychological Trauma; SCI-QOL Resilience, Spinal Cord Injury – Quality of Life Resilience; SCI-QOL Bladder Management Difficulties, Spinal Cord Injury – Quality of Life Bladder Management Difficulties; SCI-QOL Bowel Management Dificulties, Spinal Cord Injury – Quality of Life Bowel Management Dificulties; SCI-QOL Bladder Complications, Spinal Cord Injury – Quality of Life Bladder Complications; SCI-QOL APSSRA, Spinal Cord Injury – Quality of Life Ability to Participate and Satisfactions with Social Roles and Activities(APSSRA); SCI-QOL – Stigma, Spinal Cord Injury – Quality of Life – Stigma; SCI-QOL – Stigma SF, Spinal Cord Injury – Quality of Life – Stigma Short Form; SOSGOQ 2.0, Spine Oncology Study Group Outcomes; WHOQOL-5, 5 satisfaction items from the World Health Organization Quality of Life; Assessment; WHOQOL-BREF, World Health Organization Quality of Life – Bref version; WHOQOL-DIS, World Health Organization Quality of Life – Disabilities module.

Table 7.

Recommendations for use in future trials.

	Category A		Category C
Instrument	Sufficient content validity (any level)	At least low-quality evidence for sufficient internal consistency	High quality evidence for an insufficient measurement property	Recommendation
ATD PA				B
BRFSS				B
IADL				B
IBMSSP – EFAPD				B
I-QOL				B
KHQ			✓	C
LiSat-9				B
LS Questions				B
NBD				B
NBSS				B
PRISM				B
M-PRISM				B
PWI			✓	C
QLI-SCI				B
QOLP-PD				B
Qualiveen SF				B
Qualiveen				B
QWB				B
SCIQL-23			✓	C
SCI-QOL BDS				B
SCI QoL-BDS version 2.0		✓		B
SCI-QOL Pressure Ulcers			✓	C
SCI-QOL Pressure Ulcers SF			✓	C
SCI-QOL Anxiety			✓	C
SCI-QOL Bank				B
SCI-QOL Economic			✓	C
SCI-QOL Depression			✓	C
SCI-QOL Positive Affect and Well-being			✓	C
SCI-QOL Self-esteem			✓	C
SCI-QOL Grief and Loss			✓	C
SCI-QOL Psychological Trauma			✓	C
SCI-QOL Resilience		✓	✓	C
SCI-QOL Bladder Management Difficulties			✓	C
SCI-QOL Bowel Management Dificulties			✓	C
SCI-QOL Bladder Complications			✓	C
SCI-QOL (APSSRA)			✓	C
SCI-QOL – Stigma			✓	C
SCI-QOL – Stigma SF			✓	C
SF-6D			✓	C
SF-12				B
SF-36				B
SF-36 (mental health)				B
SF-36V				B
SOSGOQ		✓		B
SWBI			✓	C
SWLS			✓	C
WHOQoL-5			✓	C
WHOQOL-BREF				B
WHOQOL-DIS		✓		B

ATD PA, Assistive Technology Device Predisposition Assessment; BRFSS, Behavioral Risk Factor Surveillance Survey; I-QOL, Incontinence Quality of Life Questionnaire; IADL, Instrumental activities of daily living; IBMSSP – EFAPD; Item Bank to Measure Systems, Services, and Policies: Environmental Factors Affecting People With Disabilities; KHQ, King’s Health questionnaire; LiSat-9, Life Satisfaction Questionnaire; LS Questions, Life Satisfaction questions; SF-36, Medical Outcomes Short-Form Health Survey; SF-12, Medical Outcomes Short-Form Health Survey – 12-Item Short Form; SF-36V, Medical Outcomes Short-Form Health Survey – Physical Functioning Scale for Use With Veterans With Spinal Cord Injury; SF-6D, Medical Outcomes Short-Form Health Survey – 6 dimensions; NBD – Neurogenic Bowel Dysfunction; NBSS – Neurogenic Bladder Symptom Score Questionnaire; PRISM, Patient Reported Impact of Spasticity Measure; M-PRISM, Modified Patient Reported Impact of Spasticity; PWI, Personal Well-Being Index; QLI-SCI, Quality of Life Index-Spinal Cord Injury version; QOLP-PD, Quality of Life Profile for Adults with Physical Disabilities; QWB, Quality of Well-Being scale; Qualiveen – SF, Qualiveen Short Form; SWLS, Satisfaction with Life Scale; SWBI, Sense of Well-Being Inventory; SCI QoL – International Spinal Cord Injury Quality of Life Basic Data Set; SCI QoL version 2.0 – International Spinal Cord Injury Quality of Life Basic Data Set version 2.0; SCIQL-23, 23-item Spinal Cord Injury – specific Health Related Quality of Life questionnaire; SCI-QOL Pressure Ulcers, Spinal Cord Injury – Quality of Life Pressure Ulcers; SCI-QOL Pressure Ulcers SF, SCI-QOL 7-item short form (SF) Pressure Ulcers; SCI-QOL Anxiety, Spinal Cord Injury – Quality of Life Anxiety; SCI-QOL Economic, Spinal Cord Injury – Quality of Life Economic; SCI-QOL Depression, Spinal Cord Injury – Quality of Life Depression; SCI-QOL PAWB, Spinal Cord Injury – Quality of Life Positive Affect and Well-being; SCI-QOL Self-esteem, Spinal Cord Injury – Quality of Life Self-esteem; SCI-QOL Grief and Loss, Spinal Cord Injury – Quality of Life Grief and Loss; SCI-QOL Psychological Trauma, Spinal Cord Injury – Quality of Life Psychological Trauma; SCI-QOL Resilience, Spinal Cord Injury – Quality of Life Resilience; SCI-QOL Bladder Management Difficulties, Spinal Cord Injury – Quality of Life Bladder Management Difficulties; SCI-QOL Bowel Management Dificulties, Spinal Cord Injury – Quality of Life Bowel Management Dificulties; SCI-QOL Bladder Complications, Spinal Cord Injury – Quality of Life Bladder Complications; SCI-QOL APSSRA, Spinal Cord Injury – Quality of Life Ability to Participate and Satisfactions with Social Roles and Activities(APSSRA) ; SCI-QOL – Stigma, Spinal Cord Injury – Quality of Life – Stigma; SCI-QOL – Stigma SF, Spinal Cord Injury – Quality of Life – Stigma Short Form; SOSGOQ 2.0, Spine Oncology Study Group Outcomes; WHOQOL-5, 5 satisfaction items from the World Health Organization Quality of Life; Assessment; WHOQOL-BREF, World Health.

Discussion

We have sought to identify and recommend instruments for assessing QoL in subjects with SCI/D based on the best available evidence regarding the measurement properties of those instruments. However, this systematic review has revealed the need for instruments rated as having a high quality of evidence for sufficient measurement properties to support a strong recommendation for use. Internal consistency and construct validity were the most assessed properties, while content validity, responsiveness, cross-cultural validity/measurement invariance and criterion validity were the least assessed. No study has evaluated the measurement error.

All four previous reviews that proposed to analyse the measurement properties of QoL measurement instruments were conducted more than ten years ago. Approximately 60% of the studies included in our review were published after this period. Furthermore, these reviews were developed before the COSMIN guidelines for developing systematic reviews of PROMs (18). As a result, they did not follow a unified and standardised approach, making difficult comparisons among their findings. For example, adopting different criteria according to correlation or ICC values, just describing the instrument, or even only classifying instruments as a subjective or objective measure. Moreover, their recommendations for the use of those instruments were based on different criteria rather than using a modified GRADE approach, as the COSMIN’s recommendations followed in our review.

According to experts, content validity is the first measurement property to consider when choosing an instrument (25). The instrument’s other measurement properties become irrelevant if it is unclear which construct is being measured in a specific population by the instrument (25, 29). Therefore, testing content validity should establish whether the content of an instrument reflects the construct it intends to measure (25, 29, 30). In our review, seventeen instruments have this property assessed for use in people with SCI/D. However, none of those instruments was rated as having a sufficient quality of evidence to support their content validity.

The second most crucial measurement property to consider when selecting an instrument is its internal structure, which can be divided into structural validity and internal consistency (25). Together, these two aspects of internal structure are related to the degree of interrelatedness among instrument items and the degree to which the scores of an instrument reflect the dimensionality of the construct to be measured. Because these two aspects are closely related, several instruments in our review were classified as having high-quality evidence for indeterminate internal consistency, namely ATD-PA, I-QOL, King’s Health questionnaire – KHQ, LiSat-9, LS Questions, PRISM, QOLP-PD, Qualiveen, SCI QOL Bank, Medical Outcomes Short-Form Health Survey – SF-36, Sense of Well-Being Inventory – SWBI, SWLS and WHOQOL-BREF. These instruments showed no structural validity or were rated as having very low or high-quality evidence for insufficient structural validity, which means further studies are needed. Only IBMSSP – EFAPD, Qualiveen, SOSGOQ and WHOQOL-DIS were rated as having high-quality evidence for sufficient structural validity and internal consistency. However, the IBMSSP – EFAPD, M-PRIM, and Qualiveen were developed to measure aspects related to QoL common in individuals with SCI/D, such as environmental barriers, spasticity, or neurogenic bladder symptoms. It might mean that these instruments are developed to assess how these specific constructs affect the QoL. In the case of the SOSGOQ, it was developed for the oncologic SCI/D population, which may limit its use among individuals with injuries from other aetiologies, such as traumatic injuries.

An instrument’s internal consistency refers to the interrelatedness among the items it contains. It defines whether all the items have the same construct and is also associated with the reliability domain. Although the statistical method used to assess this property is relatively simple and widely evaluated, the resulting Cronbach’s Alpha is often misinterpreted. For example, some authors misconstrue the definition of excellent internal consistency based only on a higher Cronbach’s alpha value just based on the information that Cronbach’s alpha between 0.70 and 0.90 indicates good internal consistency, and values above 0.91 suggest a redundancy of items (31). However, due to the experiences of the COSMIN team where many good instruments showed higher Cronbach’s alpha without the necessity of removing items. In those cases, a positive rating for internal consistency is given only if a factor analysis has been applied and Cronbach’s alpha is between 0.70 and 0.95. Based on that, we can state that SCI-QOL BDS version 2.0, SOSGOQ and WHOQOL-DIS had high-quality evidence for sufficient internal consistency.

Only eight instruments were rated as having high-quality evidence for sufficient reliability (PRISM, Qualiveen, SCI-QOL BDS, SCI-QOL BDS version 2.0, SCI-QOL Pressure Ulcers SF, SCI-QOL Depression, and SCI-QOL Positive Affect and Well-being and SCI-QOL Resilience). Reliability refers to the degree to which an instrument is free from measurement error for individuals whose clinical state has not changed and who have similar scores under several conditions, i.e. no changes at specific periods (test-retest), no changes when scored by different raters on the same occasion (inter-rater) or by the same raters on different occasions (intra-rater) (21). Within the reliability domain, measurement error refers to a systematic and random error in an individual’s score that is not attributed to real changes in its construct (21,30). Reliability and measurement error are critical in clinical practice and research to avoid misinterpretation of an individual’s progress when the change has occurred due to different raters or circumstances involving the evaluation.

Regarding construct validity (hypothesis testing, cross-cultural and measurement invariance) and criterion validity, several instruments were rated as having a high quality of evidence for sufficient hypothesis testing, namely: ATD PA, LiSat-9, LS Questions, PRISM, M-PRISM, QLI-SCI, QOLP-DP, Qualiveen, Qualiveen SF, SCI QOL-BDS, SCI QOL-BDS version 2.0, SCI-QOL Anxiety, SCI-QOL Depression, WHOQOL-BREF and WHOQOL-DIS. Nonetheless, only the LS Questions and SOSGOQ were rated as having a high quality of evidence for sufficient criterion validity. According to Lankhorst et al. (26, 24), criterion validity is often considered more powerful than construct validity, but in the COSMIN taxonomy, there is no hierarchy between those properties (21). Criterion validity represents the degree to which an instrument reflects a gold standard measure. Still, it is not commonly assessed in Patient-Reported Outcome Measures (PROMs) because it is too challenging to determine what can be usefully deemed a gold standard measure of a PROM. In such situations, the COSMIN panel suggests that criterion validity can only be assessed when comparing an instrument’s long and short versions by adopting the extended version as the gold standard measure. However, when that is impossible, the review team must define the gold standard measure before analysing the methodological quality (19).

Our review team found it challenging to identify the gold standards measures in the included studies because many authors used terms related to criterion validity, such as predictive or concurrent validity, but applied methods more closely associated with construct validity (convergent, divergent, or discriminative). The same situation occurred concerning construct validity. Many authors used terms related to criterion validity when attempting to assess hypothesis testing or used terms not recommended by the COSMIN panel, such as longitudinal construct validity, as in Bonniaud et al. (32). Most of these situations occurred in studies published before the creation of COSMIN in 2010 or in studies that, even after the publication of COSMIN, did not adopt the terminology COSMIN recommends. The non-use of standardised language tends to generate communication noise in the literature and the analysis of these measurement properties.

In the five instruments in which cross-cultural validity/measurement invariance was assessed, the high quality of evidence for sufficient measurement properties in the SCI/D population was rated, except the WHOQOL-DIS, which was rated as having a high quality of evidence for indeterminate use. This measurement property is essential when checking whether the items in a translated or a culturally adapted version of an instrument perform in line with those of the original version and considering the different populations (21,30). Given that QoL is a multifactorial outcome related to many social and cultural aspects, this review team believes it is essential to assess the cross-cultural validity/measurement invariance of the instruments used to measure it to ensure the validity of the responses obtained among the SCI/D population. Additionally, only two instruments, the SF-36 and WHOQOL-BREF – both general instruments commonly used in literature – were rated as having a high quality of evidence for sufficient responsiveness. This property refers to the ability of an instrument to detect change over time in the construct to be measured. It is an essential piece of information for clinicians and researchers in this field.

Forty-nine instruments related to QoL were identified in this systematic review, among which the SCI QoL-BDS, SWLS, SF-36, WHOQOL-BREF, and Qualiveen seem the most widely applied in studies involving the SCI/D population, cited in at least four or more studies in this review. However, that does not mean they are also the instruments of choice among healthcare professionals in clinical practice or have evidence for use. Our review reveals that all the identified instruments lacked the full range of properly assessed measurement properties concerning people with SCI/D.

According to the COSMIN guidelines followed in this review, which sought to recommend the most suitable PROMs for assessing QoL in people with SCI/D, no instrument was fit to be rated as category A. Instead, we found 22 instruments rateable as category B, which means that they have the potential to be recommended for use, but further measurement analysis studies are necessary for this population. In addition, many of the instruments classified here as category B assess particular aspects of subgroups of individuals with SCI/D, such as I-QOL, NBSS, Qualiveen (long and short version), Medical Outcomes Short-Form Health Survey – Physical Functioning Scale for Use with Veterans with Spinal Cord Injury – SF-36 V, and PRISM, which limit their more comprehensive application. Therefore, the authors recommend that the clinician or researcher pay particular attention to the QoL-related construct they want to assess when choosing a PROM. To obtain a broader assessment of QoL, generic instruments such as SF-36, Medical Outcomes Short-Form Health Survey – 12-Item Short Form – SF-12, WHOQOL-DIS, WHOQOL-BREF, QLI-SCI, QOLP-PD, LS Questions, Lisat-9, and Behavioral Risk Factor Surveillance Survey – BRFSS should be used, while further measurement property studies are being conducted.

One methodological barrier the review team faced concerns the non-specification of the sample size used by the authors for each measurement property (i.e. when more than one property is analysed in the same study). In those cases, the review team considered the general sample size cited by the authors for all the analysed measurement properties. This may have affected the assessment of the methodological quality in some studies. Another difficulty encountered by the review team was that the authors did not clearly define the hypotheses tested in their validity studies, making it challenging to analyse the methodological quality and, consequently, the evidence in some cases. Additionally, several instruments have subscales, but the individual analysis of each subscale was not the author’s focus in this review.

Conclusion

Despite the variety of instruments used to assess QoL in the SCI/D population, there are significant limitations to the current evidence on the measurement properties. The review team has identified a few instruments that could be used in future studies and pointed out gaps in knowledge that will need to be filled by such studies. With few exceptions, there is a need for a higher quality of evidence for all the measurement properties of the instruments included in this review.

There are no instruments categorised as highly recommended for use (A). Whereas, based on the available evidence, 26 instruments have shown the potential for use (B), but they need more high-quality measurement properties studies to improve their recommendations. Among these, we might suggest generic instruments such as SF-36, SF-12, QWB, WHOQOL-DIS, WHOQOL-BREF, QLI-SCI, QOLP-PD, LS Questions, Lisat-9, and BRFSS could have the greatest potential to be recommended to assess general aspects of QoL in people with SCI/D. In particular, the WHOQOL-DIS was rated as having high-quality evidence for sufficient internal consistency and structural and construct validity. However, caution is suggested regarding this recommendation since further analyses of its measurement properties are required.

Author contributions

All authors contributed to the study’s conception and design. G.S.S. assisted with the study selection and data extraction. L.O.A., J.I., and A.L. performed all the material preparation, data collection, and analysis. L.O.A. wrote the first draft of the manuscript and all the authors commented on previous versions. All the authors read and approved the final manuscript.

Conflict of interest No potential conflict of interest was reported by the author(s).

Funding This work was supported by Coordenação de Aperfeiçoamento de Pessoal de Nível Superior: [Grant Number Finance Code 001]; Fundação de Amparo à Pesquisa e Inovação do Estado de Santa Catarina: [Grant Number TO 2019TR767].

Registration The protocol for this systematic review was registered on PROSPERO: CRD42020192674.