Adaptation of the German language version of the QUAlity of life assessment of spina bifida for adults (QUALAS-A-G)

Abstract

Introduction

Spina bifida (SB) affects almost all activities in daily life and therefore also health-related quality of life (HRQOL). To assess the HRQOL of adults with SB, a self-reported QUAlity of Life Assessment of Spina bifida in Adults (QUALAS-A) was validated in English. The purpose of this study was to develop and validate a German version of QUALAS-A.

Methods

German-speaking adults > = 18 years were recruited at a tertiary center and through an SB association. Two urologists translated the QUALAS-A into German. Face and content validity were assessed by adults with SB. Back-translation into English was performed. QUALAS-A-G was administered online as part of a larger study of sexuality in SB. Internal consistency was verified using Cronbach’s alpha. Factor analysis, convergent validity, and differences between groups based on continence were established.

Results

A total of 45 adults with SB participated (median age of 29 years old). Domain scores demonstrated no floor and minimal ceiling effects. Cronbach’s alpha determined good internal consistency (0.58–0.70). The factor analysis converged to a somewhat different three-factor structure compared to the original version, but otherwise had similar properties to the original. Construct validity revealed moderate associations (r = 0.36–0.65) between QUALAS-A-G domains, indicating they measure different aspects of HRQOL. Adults who were continent of both urine and stool reported higher QUALAS-A-G scores than those who were incontinent.

Conclusion

We demonstrated QUALAS-A-G to be reliable and valid in German-speaking adults with SB. The QUALAS-A-G could be a useful tool for communication between adults with SB and medical staff.

Introduction

With the establishment of pharmacotherapy (usually anticholinergic therapy), clean intermittent catheterization (CIC), in selected patients, antibacterial prophylaxis as well as interdisciplinary collaboration, the prognosis of patients with spina bifida (SB) has significantly improved (1). One of the main goals in the treatment of patients with SB is to prevent deterioration of the urinary tract function and achieve continence at an appropriate age. Urinary and stool incontinence are affecting almost all aspects of daily life and with increasing age, quality of life (QOL) gains more and more importance. In recent years, health-related quality of life (HRQOL) questionnaires have been used to measure the psychosocial status of children with and without chronic conditions such as SB (2–4). Although generic, or overall, HRQOL has been frequently compared between healthy children and those with SB (2, 5–8), such a generic comparison lacks the resolution required to be of clinical value. This is because people with SB live with multiple condition-specific concerns that people without SB do not typically experience and are not captured by generic HRQOL questionnaires. Therefore, capturing SB-specific HRQOL requires an SB-specific questionnaire. Measuring health status in patients with SB is difficult and has been attempted in several studies (9–15). Additionally to the “normal” changes of puberty, SB patients are affected by challenges resulting from the need to become more independent and sometimes also newly emerging health problems related to renal function or continence (16, 17).

While care management shifts from the parent to the patient with SB during adolescence, this process frequently continues into early adulthood. Given the dynamic nature of changing medical and social issues, assessing HRQOL in this patient group is of great importance.

In this study, we focused on validating the Quality of Life Assessment in Spina Bifida for adults (QUALAS-A) questionnaire. These QUALAS questionnaires were developed by Szymanski et al., for children (8–12 years), teenagers (13–17 years), and adults (>18 years) with SB, which is already available in English, Japanese, and Korean (4, 12, 14, 18). QUALAS-A captures self-reported SB-specific HRQOL in three domains, including health and relationships, esteem and sexuality, as well as bladder and bowel function. The aim of the present study was to develop and validate a German language version of QUALAS-A.

Methods

Participants and research setting

German-speaking adults aged 18 years old or older were recruited at our tertiary center for neurogenic bladders and through the “Arbeitsgemeinschaft Spina Bifida und Hydrocephalus (ASBH)”, the largest German-speaking patient and family association for SB. Adults who had undergone spinal repair at birth due to spinal dystrophy with a documented ICD-10 code Q06.x were included in this study. Exclusion criteria were no surgery for spinal dysraphism at birth, no confirmed diagnosis of SB, a severe diagnosis other than SB (e.g. cloacal exstrophy, VACTERL-association, and anorectal malformations), surgery in the last month, or inability to complete a self-reported questionnaire. This study questionnaire was administered as part of a larger study of sexuality in SB, described elsewhere.

Instrument

The QUALAS-A is a validated self-reported SB-specific HRQOL questionnaire. It consists of 15 items and three domains: Health and Relationships (five items), Esteem and Sexuality (five items), and Bladder and Bowel (five items). Responses to each item, or question, are captured on a Likert scale with five options to capture the level of concern in the past month: never, almost never, sometimes, almost always, and always. Certain items also included the response “not applicable”, which is given the maximum score. This was done since a complaint that does not apply to someone (such as not having incontinence) does not affect their HRQOL and ensures everyone had the same potential maximum score (denominator). Higher scores for each domain represent a higher level of HRQOL (range: 0–100 points). Cronbach’s alpha in the original version of QUALAS-A was 0.68–0.77 (18).

Developmental procedure

The validation approach was similar to that previously applied in the Japanese and Korean validations of QUALAS. Briefly, two native German-speaking pediatric urologists (R.S. and N.Y.) translated the 15 items of QUALAS-A into German. This version was back-translated and compared to the original English version to ensure accuracy by a professional bilingual translator unfamiliar with the original version. We obtained and incorporated feedback from semi-structured interviews with 10 German-speaking adults with SB to ensure clarity and understanding to develop the final QUALAS-A-G version.

QUALAS-A-G was distributed in person at clinics or via mail from March 2018 to January 2019. As in previous, similar studies, the recruitment goal was 50 participants (19), relying on the ratio of 5 participants per each of the 15 QUALAS items analyzed (20). Adults provided data on sex, age, household arrangements, school type, mobility, placement of ventriculoperitoneal shunt and use of clean intermittent catheterization (CIC), urinary and fecal continence, and use of catheterized channel and/or bladder augmentation. Urinary and fecal continence were defined as no urinary or stool leaking in the past 4 weeks, respectively.

Validation analysis

We assessed several types of validity to determine if QUALAS-A-G measures what it was intended to measure (21). Face and content validity were established through a review and semi-structured interviews by German-speaking adults with SB and experts. Interviews were conducted by the study authors. Construct validity was assessed through factor analysis and comparing QUALAS-A-G scores between groups that have been previously reported to have different HRQOL. We calculated the domain means, standard deviations (SDs), medians, ranges, and percentages of subjects scoring the minimum (floor) and maximum (ceiling).

Reproducibility (reliability) was captured by measuring internal consistency with Cronbach’s alpha: values 0.7–0.9 signify robust consistency without redundancy.

For the factor analysis of the QUALAS-A-G, exploratory factor analysis was conducted with varimax rotation. The original QUALAS-A had three loading factors which accounted for 64% of variance (19). For exploratory factor analysis, a sample fit score of ≥0.60 on a Kaiser-Meyer-Olkin scale was considered suitable (22). Factors with loadings >40 or < −40 (range −100–100) were retained.

Convergent validity evaluates the degree to which QUALAS-A-G domains converge with each other using Pearson’s correlation coefficients (r). Pearson’s r values range from −1 to 1, with 0 indicating no correlation. Values < = 0.3 represent weak correlation, 0.4–0.6 represent moderate correlation, and values > = 0.7 strong correlation. In the original QUALAS-A paper, correlations between domains were fair (r< = 0.48) (19).

Statistical analysis

Study data were managed using REDCap. Due to skewed score distributions, the non-parametric Kruskal- Wallis rank test was used to compare differences in scores between groups. A critical P = 0.05 was used (software: Stata, StataCorp, College Station, TX).

Ethics

The study was approved by the local ethic committee.

Results

Sample characteristics

A total of 45 adults with SB participated at a median age of 29 years old (IQR 23-36) (Table 1). Overall, 56% had shunted hydrocephalus and 22% were community ambulators. Most participants lived in Germany (73%). Most adults (82%) use clean intermittent catheterizations for bladder management (60% reported urinary incontinence in the last 4 weeks). Most adults (84%) did not have colostomy or Malone antegrade continence enema (38% reported fecal incontinence).

Table 1.

Population characteristics.

Variable	Number (%)
Age (median, IQR)	29 (23-36)
Age (mean, SD)	31.8 (11.1)
Males (%)	18 (40%)
Shunted hydrocephalus (%)	25 (56%)
Community ambulators (%)	10 (22%)
Household arrangement (%)*
Living with parents	17 (38%)
Living with spouse/partner	12 (27%)
Living alone/single	15 (33%)
Living with room mates	2 (4%)
Living in an assisted living facility	5 (11%)
Living with other family members/caregivers	1 (2%)
Living with siblings	3 (7%)
Relationship status (%)
Single without dating	12 (27%)
Single, but looking for a partner/not in a committed partnership	11 (24%)
Committed partnership with separate households	9 (20%)
Committed partnership, living together	4 (9%)
Married	7 (16%)
Divorced or separated	1 (2%)
Prefer not to Answer	1 (2%)
Education level for those 25 years or older (n = 31)
High school degree	7 (23%)
Some college, Associate degree (Abitur, Fachabitur)	4 (13)
Apprenticeship (Lehre)	11 (35%)
Bachelor degree (Studium)	2 (6%)
Graduate degree (Hochschulabschluss)	6 (19%)
Prefer not to Answer	1 (3%)
Country
Germany	33 (73%)
Switzerland	8 (18%)
Netherlands	2 (4%)
Austria	1 (2%)
Luxembourg	1 (2%)
Bladder management
Clean intermittent catheterizations (%)	37 (82%)
Incontinent urinary diversion (%)	3 (7%)
Neither (%)	17 (2%)
Unknown (%)	1 (2%)
Continent of urine in the last 4 weeks
Yes (%)	16 (36%)
No (%)	27 (60%)
Unknown (%)	2 (4%)
Bowel management
Malone antegrade continence enema (%)	5 (11%)
Colostomy (%)	0 (0%)
Neither (%)	38 (84%)
Unknown (%)	2 (4%)
Continent of stool in the last 4 weeks
Yes (%)	17 (38%)
No (%)	25 (56%)
Unknown (%)	3 (7%)

*May add up to more than 100% since multiple options could be chosen

Distribution, reliability, and factor analysis

The median scores for QUALAS-A-G for Health and Relationships, Esteem and Sexuality, and Bladder and Bowel domains were 75 (IQR 65–85), 70 (IQR 50–85), and 67.5 (IQR 40–85), respectively (Table 2). Missing rates were low (range 0–9%). There was no floor effect (there were no scores of 0) and the ceiling effect was low (0–12% of adults reported maximum scores of 100).

Table 2.

Domain characteristics of the self-reported QUALAS-A-G.

Domain	N	% missing	Mean (SD)	Median (IQR, range)	% scoring minimum (0)	% scoring maximum (100)	Cronbach’s alpha
Health and Relationships	45	0%	74.2 (18.4)	75 (65-85, 20-100)	0%	11%	0.70
Esteem and Sexuality	41	9%	69.2 (21.0)	70 (50-85, 20-100)	0%	12%	0.58
Bladder and Bowel	42	7%	63. 8 (23.9)	67.5 (40-85, 20-95)	0%	0%	0.68

Internal consistency was high in two domains (Health and Relationships: 0.70 and Bladder and Bowel: 0.68), and lower in Esteem and Sexuality (0.58).

On factor analysis, all items had factor loadings of >40, with a satisfactory Kaiser-Meyer-Olkin score of 0.65, according to the approach outlined in Methods (Table 3). While three factors explained 56% of the variance, the items did not load in a fashion like the ordinal QUALAS-A. Rather, 12 of the 15 items were primarily loaded on a single factor. Given satisfactory face and content validity established by adults with SB and experts during the translation process, the domain structure of QUALAS-A was maintained in the German version.

Table 3.

Factor analysis of responses from 45 adults with spina bifida.

Item	Factor 1	Factor 2	Factor 3
Overall health	70*	−11	27
People saw you as more than health problems	61*	−5	−58
Able to do fun things	65*	−38	21
Time with friends	42	−63	36
Close friendships outside family	62*	−56	−24
Embarrassment about your look	62*	−14	50
Being treated differently because of health problems	71*	−13	−42
Having future children	10	56*	18
Satisfaction with sex life	43*	10	−34
Future sexual satisfaction impacted by health	76*	28	−19
Worry about pads being noticed	59*	20	−6
Bother by urine leak	54*	42	30
Urine problems stop you from fun things	75*	10	16
Bother by bowel leak	40	41*	8
Bother by waiting for bowel movement	52*	48	2

Note: Kaiser-Meyer-Olkin score: 0.65. Cumulative contribution ratio of the three factors: 56%.

*The exploratory factor analysis we employed does not provide p-values for individual factor loadings.

Convergent validity

Health and Relationships domain had a correlation of 0.60 (95% CI 0.37–0.76) with the Esteem and Sexuality domains (P < 0.0001) and a correlation of 0.36 (95% CI 0.06–0.60) with the Bladder and Bowel domain (P = 0.02). The Esteem and Sexuality and Bladder and Bowel domains had a correlation of 0.65 (95% CI 0.43–0.80, P < 0.0001).

Known group differences

Health and Relationships domain scores were not statically different between adults who were continent of both urine and stool, continent in only one, and not continent for either one (P = 0.12, Table 4). On the other hand, scores for the Esteem and Sexuality domain and the Bladder and Bowel domains differed between these three continence subgroups (P< = 0.03). For these domains, HRQOL was highest in the subgroup which was the continent of urine and stool.

Table 4.

Differences in QUALAS-A-G scores between adults with spina bifida based on urine and stool continence.

Domain	Continent of both urine and stool		Not continent of either urine or stool		Not continent of both stool and urine
	n	Median (IQR)	n	Median (IQR)	n	Median (IQR)	P-value
Health and Relationships	9	85 (75-95)	13	70 (60-75)	19	75 (70-85)	0.12
Esteem and Sexuality	9	85 (80-100)	13	60 (50-75)	18	70 (55-75)	0.03
Bladder and Bowel	9	90 (80-90)	13	75 (40-85)	19	55 (35-70)	0.02

Discussion

Assisting children with SB into adulthood was a great problem in the past, and it continues to be a challenge in most parts of the world. Today, we have more and more transition clinics for SB patients, which can at least help manage problems related to the urinary tract (1). However, the change of responsibility and the decision-making process regarding treatments in aging patients is not easy for the patients, their parents, and even the doctors (23). In the past, some self-reported HRQOLs have often used a visual analog scale (VAS) (24–26). Such an approach may give an overall impression of HRQOL, but cannot measure multiple psychosocial aspects needed for clinical decision making or research. The QUALAS-A has shown to be a suitable self-reported HRQOL instrument for adolescents with SB in English, Japanese, and Korean versions (4, 12, 14, 18).

Our analysis indicates that the QUALAS-A-G is reliable and valid. While it retains many aspects of validity and reliability of the original, including domains capturing independent aspects of HRQOL and producing different scores between known subgroups, the underlying factor structure appears to be different. A similar difference has been noted during the cross-cultural adaption of QUALAS-A in Korea (19). While the Korean team opted to drop some items and significantly revise the questionnaire, we opted for the alternative of retaining the original questionnaire structure. This decision was based on high face validity and continent validity, assessed particularly by adults with SB, clinical expertise, and other satisfactory aspects of the valuation process. The fact that QUALAS-A-G was able to detect differences in HRQOL between subgroups of adults with different continence also indicated its clinical utility. The differences in the underlying factor structure may be due to chance, the relatively small number of participants, or real differences between German-speaking and the original international cohort, potentially because they enrolled in a study primarily geared at aspects of sexuality, rather than HRQOL. The fact that internal consistency was lowest for the Esteem and Sexuality domain is likely due to the complex and multifaceted nature of these constructs, which can vary considerably among individuals with SB. This lower value has been observed in the original English language version of QUALAS-A (18).

Admittedly, while not a psychometrically ideal questionnaire, the QUALAS-A-G represents the only valid German-language questionnaire capturing HRQOL among adults with adults with SB. For that reason, we believe, the QUALAS-A-G can help better assess regarding medical and surgical treatments for bladder and bowel management in this unique group of patients. It also could support adolescents with SB as well as medical staff members as the patients get older and it may be useful for communication between adolescents.

Our study has several limitations. SB is a rare congenital anomaly and the number of participants in this study was smaller than usually needed for factor analysis (27). Recruiting participants by email through the Spina Bifida Association of Germany could lead to selection bias, as the patients responding have a higher awareness of the need to have such investigations and may also have more interest in answering these questions. Since content validity was established during semi-structured interviews conducted by the study authors, this could predispose to potential bias. Nonetheless, the same approach was used during English, Japanese, and Korean-language validations (4, 12, 14, 18). Since the validation study was embedded in a larger, longer anonymous study of sexuality in SB, the number of concurrent HRQOL questionnaires administered was significantly limited. For that reason, we were unable to evaluate convergent validity with other HRQOL instruments. Since the study was cross-sectional, we did not evaluate retest validity. This will need to be established in the future.

In conclusion, the reliability and validity of the QUALAS-A were verified in German-speaking adults with SB. The QUALAS-A-G could be a useful tool for communication between adults with SB and medical staff.

Disclaimer statements

Contributors None.

Funding None.

Declaration of interest None.

Conflicts of interest Authors have no conflict of interests to declare.