Resilience and quality of life in young adults with a 22q11.2 deletion syndrome: a patient’s perspective

Abstract

In the transition period between adolescence and young adulthood, individuals with 22q11.2DS are at an increased risk of developing severe psychiatric disorders. Various studies have focused on detecting risk factors, but until now protective factors are still understudied in 22q11.2DS. The current case–control study focuses on the role of resilience and quality of life (QoL) in young adults with 22q11.2DS and behavioural problems, in comparison with persons with an intellectual disability (ID) without a known genetic disorder. Self-report (and caregiver report) standardized questionnaires were used. Predictive general linear models were constructed to compare the resilience and quality of life across both groups (22q11.2DS vs ID-group) and to analyse the association between personal characteristics in both groups. Young adults with a 22q11.2DS show less resilience compared with both the general population norms and young adults with ID. Only a subscale of resilience (Acceptance of self and life) contributes to the reported level of QoL. Reported health problems are not related to resilience, but have an important effect on QoL. Our results suggest different factors are underlying resilience and the relation with QoL in 22q11.2DS and ID in general. These factors deserve more research and are important to take into account in clinical practice.

Introduction

The 22q11.2 deletion syndrome (22q11.2DS) is the most frequent chromosomal microdeletion syndrome, caused by a microdeletion of the long arm of chromosome 22. It is estimated to occur with a frequency between 1 per 3000 and 1 per 6000 live births [1]. The 22q11.2DS is associated with a highly variable phenotype of multiple physical problems, cognitive delays and psychiatric disorders. The most common physical characteristics are congenital heart disease, palatal abnormalities, immunodeficiency, hypocalcemia due to hypoparathyroidism, severe feeding/gastrointestinal differences, and subtle dysmorphic facial features [1–4]. Intelligence is normally distributed in people with 22q11.2DS with a mean IQ of 70. [1, 4]. In the transition period between adolescence and young adulthood, individuals with 22q11.2DS are at an increased risk of severe psychiatric disorders, such as attention deficit hyperactivity disorder (ADHD), autism spectrum disorder (ASD), anxiety disorders, mood disorders, and schizophrenia spectrum disorders [1, 5].

Various studies have focused on detecting risk factors contributing to the neurodevelopmental outcome and the development of psychiatric disorders in persons with 22q11.2DS. For example, early cognitive decline (verbal decline) is an indicator of the risk of developing a psychotic illness [4, 6]. Protective factors however are still understudied in 22q11.2DS [4]. This study focuses on the role of a possible protective factor, resilience, in young adults with 22q11.2DS, and the relation with Quality of Life (QoL).

The concept ‘resilience’ has its starting point at the finding that some individuals have a better outcome than others who have experienced a comparable level of adversity [7, 8]. Although this concept has been increasingly used across several disciplines, including psychiatry, there is still no consensus about the definition [9, 10]. Resilience can be seen as a fixed, stable personality characteristic with strong genetic influence, or as a dynamic process of personal, interpersonal, and contextual protective mechanisms that results in a positive outcome in the presence of adversity [10, 11]. Resilience is seen as a protective buffer against psychotic illness in the general population [7, 9]. Although several authors mention the concept of resilience in the context of 22q11.2DS [12, 13], no systematic study (to the best of our knowledge) has been done so far.

Another important concept (in the literature) on vulnerability in persons with psychiatric disorders and/or intellectual disabilities is the concept of ‘Quality of Life’ (QoL). The World Health Organization (WHO) defines QoL as “individuals’ perception of their position in life in the context of the culture and value systems in which they live and in relation to their goals, expectations, standards and concerns” [14]. In individuals with 22q11.2DS, the combination of several physical problems, cognitive delays, and psychiatric disorders can have important effects on their QoL [1]. Several researchers investigated the QoL of children and adolescents, aged 2–18 years, with 22q11.2DS using the paired Paediatric Quality of Life Inventory 4.0 questionnaires and the PedsQL^™ 4.0 Generic Core Scales [15, 16]. They all reported a significantly poorer QoL compared with both healthy children and children with chronic disease, such as asthma, diabetes, depression, ADHD, and muscular dystrophy [15, 16]. The QoL of young adults with 22q11.2DS has not been studied yet, although late adolescence and early adulthood are considered critical time periods in the development of psychiatric disorders for persons with 22q11.2DS [17].

The Adolescent Resilience Model [18], originally developed in the context of adolescents with cancer is a model that includes both resilience and QoL. This model clarifies the relation between resilience and QoL and correlated factors: resilience and QoL are both correlated with several environmental and personal factors, such as family support and health problems. Haase states that resilience is positively correlated with QoL [18]. This corresponds to other studies that indicate that resilience is positively correlated with QoL [19, 20]. In Fig. 1 an illustration of the adolescent resilience model is provided [18].

Fig. 1.

Relation between resilience, quality of life, environmental and personal factors, based on the adolescent resilience model [18]

The correlation between resilience and QoL has been studied in multiple populations such as patients with schizophrenia and bipolar disorder [19]. To the best of our knowledge, no previous study has investigated this correlation in individuals with a 22q11.2DS and/or in persons with an intellectual disability (ID). In the present study, we aimed to compare the level of resilience of individuals with 22q11.2DS with an ID group of persons with an ID, and with persons in ambulant and residential psychiatric care. Our research questions are the following: is there a difference in resilience, QoL and behavioural problems between individuals with 22q11.2DS and ID groups? Are personal and environmental factors of young adults with 22q11.2DS or an ID correlated with resilience, QoL and behavioural problems? Is there a correlation between resilience, QoL and behavioural problems in young adults with a 22q11.2DS? Our hypothesis is that resilience and QoL are lower for persons with 22q11.2DS compared with the general population. We expect that the resilience and QoL of young adults with 22q11.2DS are similar to the resilience and QoL of young adults with an ID or psychiatric problems.

Methods

Data collection and participants

All young adults with a 22q11.2DS who were known as a patient in Leuven (n = 67) were contacted via the multidisciplinary team of the Center for Human Genetics at the University Hospital Leuven, Belgium. Thirty-seven persons agreed to participate in this case–control study (response rate 55%). For the analysis, we excluded three participants who were older than 35 years and one participant who did not complete the questionnaires. Thirty-three participants (15 males, 18 females) were included in the study, this exact cohort of 33 patients was not included in previous studies and was recruited for the purpose of this study. All 33 participants had a formal diagnosis of a 22q11.2 microdeletion based on FISH or microarray and were between the ages of 18 and 32 years. Controls were recruited at a residential centre for people with intellectual disabilities (OC Clara Fey). Twenty-six individuals (14 males, 12 females) with a mild to moderate intellectual disability agreed to participate. They were all young adults between 18 and 32 years old. All participants received the questionnaires and a cover letter by e-mail, post or after a consultation at the Center for Human Genetics and were included in the study in 2014.

All participants and/or their legal representatives provided written informed consent. Ethical approval for this study was obtained from the Research Medical Ethics Committee UZ / KU Leuven (Belgium).

Measures

First, environmental and personal characteristics, such as the number of health problems and the history of therapy, were questioned. Special education was defined as any other type of education than regular education, for example, a school for students with learning disabilities. To obtain information on resilience, QoL and behavioural problems, we used standardized self-report and caregiver report questionnaires. The level of cognitive functioning was indicated by the caregiver. If an IQ score was found in the medical records of the 22q11.2DS patients, this score was used in the dataset.

Resilience was assessed with the Dutch Resilience Scale ‘RS-nl’ [21]. This is an adaptation of the ‘Resilience Scale’ of Wagnild and Young [22]. The RS-nl is a standardized self-report questionnaire with 25 items rated on a 4-point Likert scale [11]. Portzky and colleagues did a psychometric evaluation of the RS-nl and concluded that the RS-nl is a valid screening instrument. The RS-nl gives a total score and a score on the subscales ‘personal competence’(PC) and ‘acceptance of self and life’(ASL) [9]. Thirty-two of the 33 participants with 22q11.2DS and all participants of the ID group completed the RS-nl. Participants were allowed to get assistance from a caregiver to complete the questionnaire.

To measure Quality of Life, the abbreviated World Health Organization Quality of Life questionnaire ‘WHOQOL-BREF’ and the World Health Organization Quality of Life instrument for people with intellectual and physical disabilities ‘WHOQOL-DIS’ were used. These questionnaires, developed by the WHO, are available in Dutch. The WHOQOL-BREF is a 26-item self-report questionnaire rated on a 5-point Likert scale. It gives scores on five subscales: ‘overall QOL’, ‘physical health’, ‘psychological health’, ‘social relationships’ and ‘environment’ [14]. The WHOQOL-DIS is a modified version for adults with a physical or intellectual disability. The questionnaire is equal to the WHOQOL-BREF, but the questions are formulated in a more concrete language and examples are added. The participant can answer by indicating smileys. The WHOQOL-DIS contains some extra items about disability-specific issues [23]. In this study, we did not include these extra items to be able to combine the scores from the WHOQOL-DIS and the scores of the WHOQOL-BREF. Thirty-one of the 33 participants with 22q11.2DS and all participants of the ID group completed the WHOQOL-BREF or WHOQOL-DIS.

Behavioural problems were measured with the young adult self-report (YASR) and the young adult behaviour checklist (YABCL) [24, 25]. The YASR is a self-report questionnaire while the YABCL is the caregiver-report version. It contains respectively 116 and 120 items rated on a 3-point Likert scale and some open questions about education and work. The questionnaire gives scores on total problems, internalizing and externalizing problems and multiple competence and syndrome scales [24, 25]. For this analysis, we used the total problems score. Thirty-one participants with 22q11.2DS completed the YASR. All participants of the ID group completed the YASR.

Statistical analysis

Statistical analysis was performed using SAS 9.4, SAS Institute Inc., Cary, NC. To describe group differences for the environmental and personal characteristics, parametric or non-parametric tests were used when appropriate: for continuous variables, respectively independent t tests or Wilcoxon Mann–Whitney test were used while for categorical variables, respectively, χ² –tests or Fisher’s exact tests were used. To examine the relationship between resilience, quality of life, separate multiple regression analyses were performed with these variables as a criterion and the environmental and personal factors as predictors. Assumptions of these models (normality, linearity, homoscedasticity) were tested by inspection of the residuals using QQ-plots, histograms, and residual plots of, for example, residuals versus predictors and residuals versus predicted values. Multicollinearity between predictors defined as Pearson correlations > 0.6 was excluded. Missing data occurrences were listed in the Supplementary materials.

To look at the mean difference between groups (ID or 22q11.2DS) regarding the total resilience score as well as the subscales PC and ASL a general linear model was fitted for each of these three dependent variables using a least-squares method. To investigate the role of the environmental and personal characteristics in resilience, three general linear model were fitted with respectively total resilience, PC and ASL as dependent variables. The level of resilience was modelled as a function of group (ID or 22q11), age, gender, ID, reported psychological problems and YASR total score.

To look at the mean difference between groups (ID or 22q11.2DS) regarding QoL, a general linear model was fitted using a least-squares method for each of the five subscales (overall QOL’, ‘QoL physical health’, ‘QoL psychological health’, ‘QoL social relationships’ and ‘QoL environment’). Assumptions of normality, homogeneity of variance and presence of outliers were checked.

To investigate the role of personal characteristics, behavioural problems and the level of resilience in the reported Quality of Life, a general linear model was fitted with respectively overall QoL, QoL physical health and QoL psychological health as dependent variables. All three models used the same personal and behavioural measures as predictors, namely: group (ID or 22q11.2DS), age, gender, cognitive functioning, reported psychological problems, YASR total score, health problems and the subscales of resilience: acceptance of self and life and personal competence. In addition three similar models were made using total resilience instead of the two subscales.

For all general linear models assumptions of normality, homogeneity of variance and presence of outliers were checked.

Results

In total, a group of 59 persons was included in our analysis, 33 persons with 22q11.2DS and 26 persons in the ID group. The two groups did not differ in terms of sex, the presence of current psychological problems, cognitive functioning, history of early intervention at home and history of any type of therapy. They did differ in age and the presence of health problems.

The mean age in the 22q11.2DS group was 23 years old, the mean age in the ID group was 27. Table 1 gives an overview of the frequencies of the environmental and personal characteristics in both groups.

Table 1.

Demographics and personal and environmental characteristics

	All participants (n = 59)	22q11.2DS group (n = 33)	ID group (n = 26)	p value
				t test (df), p value
Age	M = 25.36 (SD = 4.02)	23.61 (3.7)	27.58 (3.4)	t (57) = −4.29, p < 0.0001**
				χ2(df), p value or p value of Fisher’s exact test (FET)*
Sex
Man	29 (49.2%)	15 (45.5%)	14 (53.8%)	χ2(1) = 0.41, p = 0.52
Woman	30 (50.9%)	18 (54.5%)	12 (46.2%)
Health problems
No problems	14 (23.7%)	4 (12.1%)	10 (38.5%)	χ2(2) = 6.15, p = 0.05*
1 or 2 problems	34 (57.6%)	23 (69.7%)	11 (42.3%)
3 or more problems	11 (18.6%)	6 (18.2%)	5 (19.2%)
Current psychological problems
Yes	25 (45.5%)	14 (48.3%)	11 (42.3%)	χ2(1) = 0.20, p = 0.66
No	30 (54.5%)	15 (51.7%)	15 (57.7%)
Level of cognitive functioning
Low to borderline intelligence (IQ > 70)		12 (36.4%)	4(15.4%)	χ2(1) = 3.24, p = 0.07
Mild to moderate intellectual disability (IQ < 70)		21(63.4%)	22(84.6%)
History of early intervention at home
Yes	39 (67.2%)	18 (56.3%)	21 (80.8%)	χ2(1) = 3.91, p = 0.0479*
No	19 (32.8%)	14 (43.8%)	5 (19.2%)
History of any type of therapy (physiotherapy, speech therapy)
Yes	54 (91.5%)	32 (97.0%)	22 (84.6%)
No	5 (8.5%)	1 (3.0%)	4 (15.4%)	χ2(1) = 2.86, p = 0.09

The numbers indicate the absolute values, between brackets the percentages are mentioned. Fisher’s exact test was calculated for cross-tables with 25% of the cells had expected counts less than 5. The last column shows the significance levels of the differences between the 22q11.2DS group and the ID group, for each variable

^*p values significant at the < 0.05 levelr

^**p values significant at the < 0.01 level

Resilience

The total average score on the RS-nl was lower in the 22q11.2DS group (M = 72.09, SE = 1.79) than in the ID group (M = 80.08, SE = 1.99). This difference was significant (F (1.56) = 8.88, p = 0.0043). This means that the individuals with a 22q11.2DS had on average a lower resilience than the individuals of the ID group. The 22q11.2DS group also scored lower than the ID group on the two subscales ‘Personal Competence’ (t (56) = −2.70, p < 0.01) and ‘Acceptance of Self and Life’ (t (56) = −2.17, p = 0.03). An overview of this comparison is given in Fig. 2. In this figure, the norm scores for the RS-nl for the general population, patients in ambulant psychiatric care, and patients in residential psychiatric care are added.

Fig. 2.

Total resilience in persons with 22q11.2DS and adults with ID, score from Dutch Resilience Scale ‘RS-nl’. Note: significant differences at the ≤ 0.05 level are marked with a *, significant differences at the ≤ 0.01 level are marked with **. Data for 22q11.2DS and ID group stem from this data collection. The other scores (general population, ambulant psychiatric care, and residential psychiatric care are norm scores

Quality of life

The overall QoL of individuals with a 22q11.2DS did not differ significantly from the QoL of the ID group (F (1,55) = 0.23, p = 0.63). The mean QoL in the 22s11.2DS group is 7.65 (SE = 0.28) and 7.84 (SE = 0.31) in the ID group. However, there was a significant difference in the domains of ‘physical health’ and ‘psychological health’, where the ID group scored significantly higher on average than the 22q11.2DS group (respectively F (1,55) = 4.87, p = 0.03) and F (1,55) = 4.70, p = 0.03).

Behavioural problems

Regarding behavioural problems –measured by the YASR and YABCL- individuals with a 22q11.2DS did not differ significantly from our ID group (t (53) = 1.02, p = 0.31). The parents of the 22q11.2DS group did report on average more attention problems, compared with the caregivers of the ID group (t (55) = 2.39, p = 0.02).

An interesting observation is that both young adults with 22q11.2DS and controls (ID-group) did not report more behavioural problems themselves compared to the norms. This was in contrast to their parents and caregivers who report more internalizing problems and a higher total problem score (than the norm group) of the YABCL. For further analysis the data from the YASR were used because of our interest in the patient’s perspective.

Resilience: model

The level of resilience was modelled as a function of group (ID or 22q11.2DS), age, gender, cognitive functioning, reported psychological problems, and YASR total score. The global null 0-hypothesis that all predictors are zero is rejected, F (7,42) = 4.94, p = 0.0004. The total variance explained by the model is 45%. The results of the analysis indicate that the 22q11.2DS-group has a significantly lower mean resilience (M = 71.35, SE = 2.12) compared to the ID group (M = 81.74, SE = 2.16), F (1,42) = 5.49, p = 0.02. In addition, patients with a higher score on the YASR are characterized by a lower resilience, F (1,42) = 8.58, p = 0.0055. However, the relation between YASR and resilience is not different for both groups, F (1,42) = 2.82, p = 0.10. Results are shown in Table 2.

Table 2.

Model fit to predict total resilience score

R-Square	Coeff var	Root MSE	Total Resilience Score mean
0.451666	11.61715	8.840654	76.10000
Source	Type III SS (mean square)	F Value	Pr > F
Group	429.1746812	5.49	0.0239*
Gender	11.7168371	0.15	0.7006
Cognitive functioning	163.4262073	2.09	0.1556
Age	64.s2279390	0.82	0.3698
Psychological problems	109.6732847	1.40	0.2428
total_YASR	670.6280888	8.58	0.0055**
total_YASR*Group	220.7704899	2.82	0.1002
Group	Total Resilience Score LSMEAN		Standard error
22q11.2DS	71.3475067		2.1153065
ID	81.7420313		2.1571592

^*p values significant at the < 0.05 level

^**p values significant at the < 0.01 level

Similar models were made for the subscales where respectively 32% (PC) and 58% (ASL) of the variance is explained. In the subscale personal competence, the factors 22q11.2DS and YASR are again predictive. For the subscale Acceptance of Self and Life, the presence of psychological problems is an additional predictive factor. (Supplementary materials).

Quality of life: model

We built a model for the five subscales of QoL (overall, perceived physical health, perceived psychological health, social relationships and environment) to investigate the role of resilience and other possible personal and environmental factors. The same factors as in the model for resilience were included, but with the addition of Health problems because of the important effect on QoL (variance improvement of 8.38% for QoL, compared to 0.76% for resilience). The models fit for the five subscales of QoL achieved notable variances explained of respectively 44.9%, 53.5%, 61.3%, 53.7% and 55.7%. These results are shown in Table 3.

Table 3.

Indicators for models fit to predict the quality of life. Total resilience (upper model) compared to resilience’s subscales personal competence and acceptance of self (lower model)

	QoL_overall	Phys. health	Psych. health	Social relationships	Environment	VIF
R-Square	0.363027	0.462385	0.590912	0.464693	0.475690
Group	0.3562	0.6097	0.7661	0.2790	0.9414	2.36
Gender	0.7780	0.9583	0.1749	0.0026*	0.0377*	1.08
Cognitive functioning	0.9252	0.8337	0.3043	0.4194	0.4901	1.39
Age	0.4457	0.8649	0.2412	0.9753	0.6878	1.54
Psychological problems	0.9992	0.3969	0.1887	0.7365	0.5528	1.45
Total YASR	0.0032**	0.0006**	0.0001**	0.0046**	0.0010**	1.73
Total resilience	0.8496	0.4774	0.6079	0.3435	0.9948	1.97
Health problems	0.0271*	0.0097**	0.0855	0.2726	0.0337*	1.26
R-Square	0.449482	0.535377	0.613326	0.537081	0.556689
Group	0.1368	0.9840	0.5308	0.0995	0.5044	2.49
Gender	0.8850	0.8339	0.1994	0.0024*	0.0390*	1.08
Cognitive functioning	0.8991	0.6543	0.2410	0.2824	0.3281	1.40
Age	0.4102	0.8454	0.2310	0.9625	0.6551	1.54
Psychological problems	0.3430	0.9046	0.5072	0.2051	0.6611	1.70
Total YASR	0.0469*	0.0126*	0.0018**	0.0619	0.0207*	2.06
Personal competence	0.1000	0.0246*	0.4569	0.2657	0.0571	2.10
Acceptance of self and life	0.0197*	0.0367*	0.1198	0.0114*	0.0143*	3.86
Health problems	0.0048**	0.0015**	0.0406*	0.0831	0.0051**	1.35

The p values for all factors are presented, as well as the model R-square value (explained variance, in bold). Variance inflation factor (VIF) shows moderate-low collinearity across factors

^*p values significant at the < 0.05 level

^**p values significant at the < 0.01 level

The subscale of resilience Acceptance of Self and Life (ASL) has a predictive value for four of the five QoL domains (p values between 0.0114 and 0.0367), whereas the total resilience score didn’t affect the domains of QoL (p values between 0.3435 and 0.9948). For all models where subscales of resilience are used instead of total resilience score, the variance explained improved notably (between 2.24% and 8.65%). Note that some limited collinearity effects exist, mainly a moderate correlation between ASL and Total YASR (Pearson correlation of −0.67). The Variance Inflation Factor (VIF) of 3.86 for the former variable is within common thresholds of 4, 5 or even 10 for indication of strong multicollinearity [26].

The total score of YASR and health problems are the most important factors in the prediction of QoL, and statistically significant (p values between 0.0018 and 0.0469, and between 0.0015 and 0.0406) for all subscales except for social relationships (where p values are respectively 0.0619 and 0.0831). Additionally, for interpersonal domains (subscales for social relationships and environment), the female gender correlated with higher scores (p values of respectively 0.0024 and 0.0390). Other factors (including 22q11.2DS) did not show any significant predictive power.

Discussion

In the transition period between adolescence and young adulthood, individuals with 22q11.2DS are at an increased risk for developing severe psychiatric disorders [1, 5]. Various studies focused on detecting risk factors contributing to the development of psychiatric disorders in 22q11.2DS [4], but protective factors are still understudied. This study focuses on resilience and QoL in young adults with 22q11.2DS in comparison to a control group of young adults with an ID and on the association between resilience, QoL and behavioural problems.

Resilience

Several authors mentioned resilience as a contributing factor in 22q11.2DS [12, 13], but this is the first study that specifically investigated resilience in a group of young adults with 22q11.2DS. Comparing our results with the norm groups showed that the level of resilience in young adults with a 22q11.2DS takes an intermediate position between the level of resilience of persons with an ID and of persons in residential psychiatric care. The level of resilience of young adults with a 22q11.2DS is similar to the level of resilience of persons in ambulant psychiatric care. This could suggest that a lack of resilience is more related to the psychiatric vulnerability in 22q11.2DS than to cognitive disability. Our models for resilience support this hypothesis finding that both self-reported behavioural problems and the presence of 22q11.2DS were significant predictors for the level of resilience. This suggests young adults with 22q11.2DS seem to have an inherent vulnerability regarding their resilience, independent from the behavioural problems. Furthermore, other factors including health problems and cognitive function were not associated with resilience, emphasizing the need for a more detailed look at the possible personal and environmental factors that could be relevant.

This intermediate position of the resilience of young adults with a 22q11.2DS and their inherent vulnerability has important clinical implications that we will discuss further.

Quality of life

Several researchers investigated the QoL of children and adolescents with 22q11.2DS, and they reported a significantly poorer QoL compared with both healthy children and children with chronic disease, such as asthma, diabetes, depression, ADHD and muscular dystrophy [15, 16]. On average individuals with a 22q11.2DS reported less satisfaction with their physical and psychological health than the individuals with ID. However, the presence of a 22q11.2DS was not predictive for quality of life in any of the domains.

In addition, when comparing the scores of the 22q11.2DS group in the domains ‘social relationships’ and ‘environment’, these appeared to be equal to or even higher than the norm scores for the general population. These results do not seem to fit with problems and concerns reported by individuals with 22q11.2DS and their environment in a clinical setting.

Individuals with a 22q11.2DS seem to be aware of their health and psychological problems, but on the questions about social relationships and environment, they (possibly) respond socially desirable. Individuals with 22q11.2DS often want to please other people and do their very best in any circumstances. It is possible that they don’t want to bother anyone with their difficulties in social relationships and interaction with their environment. Another possible explanation could be that they are not aware of their difficulties in social situations. A reduced awareness of social difficulties could fit with an increased prevalence of autism spectrum disorder (ASD) in 22q11.2DS, with an estimated prevalence of 21.5% [27].

As with resilience, higher overall levels of behavioural problems appear to be related to lower levels of QoL in almost all domains. This does suggest that although these individuals might not report a lack of quality of life in their social relationships, the quality of social relationships does seem to be influenced by their behavioural problems as well as the level of resilience.

Gender was a significant predictor of perceived social relationships and perceived QoL of the environment with higher scores in women. This fits with lower scores seen in men in norm scores [28]. Explanations could be an increased prevalence of ASD which is more prevalent in men in the general population, though in 22q11.2DS it appears there is no gender difference [5].

Interestingly a higher amount of reported health problems was related to QoL in several domains, including not only as could be expected in QoL physical health but also overall QoL, QoL related to the environment and to a lesser extent QoL psychological health. As the reported physical health problems were on average higher, it seems to be an important contributing factor to QoL in patients with 22q11.2DS. This is in accordance with the study of Vergaelen et al. who reported a high level of fatigue in adults with 22q11.2DS and the effect on their daily life [29]. It is therefore crucial in daily life and care for young adults with 22q11.2DS to pay attention to both physical and psychological problems.

Correlation between resilience and quality of life

The Adolescent Resilience Model states that resilience and QoL are both correlated with several environmental and personal factors [18]. Different studies indicate that resilience is positively correlated with QoL [19, 20]. This is partly confirmed in our study. Higher levels of resilience for the domain ‘Acceptance of Self and Life were associated with a better QoL in most domains. The exception was the perceived quality of psychological health which showed no association with resilience and appeared to be most directly related to the level of self-reported behavioural problems and in a minor way to reported health problems. This suggests that while higher levels of resilience can positively impact the QoL in most domains, counter to the negative impact of behavioural problems, this is not (or not significantly) the case for the perceived quality of psychological health. As the subscale personal competence was not associated with QoL, it suggests that a specific domain of resilience is impacting QoL. As described by the original developers of the Resilience Scale Wagnild and Young, ASL is more about the adaptability and flexibility of a person and being able to find a balance in life in spite of adversity. [30] Future research should focus on investigating the factors that can help these young adults finding this balance. An interesting question could be if this could be related to the growing difference between the increasing environmental demands and the abilities of the young adult as is seen often in neurodevelopmental disorders and more specifically 22q11.2DS. [31]

Clinical implications

The level of resilience of young adults with a 22q11.2DS and their inherent psychological vulnerability have important clinical implications. They have a lower level of resilience than other young adults with an ID, and it is, therefore, crucial -from childhood on- to pay attention to this limited resilience. Parents and care providers should be made aware of this limited resilience so that they can adjust -in time-their expectations and follow up the resilience in their children. At the same time, targeted interventions to foster and increase the resilience in individuals with 22q11.2DS should be provided: for example, the book “Resilience Interventions for Youth in Various Populations” provides an overview of interventions that seek to increase resilience in both clinical and non-clinical groups [32].

Regarding QoL, behavioural problems and reported health problems appear to be the most important factors influencing almost all domains of QoL. Young adults with 22q11.2DS report more health problems and more reported health problems appeared to be negatively associated with most domains of QoL, with the exception of the perceived quality of social relationships. As expected behavioural problems were associated with a lower quality of psychological health. However, we also see a trend towards an association with the quality of physical health in the 22Q11.2DS group. This confirms the importance of looking at both psychological and physical health when a child or adult with 22Q11.2DS is seen in a clinical setting. Though health problems are less prevalent in the young adults with ID, it appears looking at both psychological and physical health is also important in this group. This finding implicates the importance of careful psychological/psychiatric follow-up in 22q11.2DS and of providing guidance and intervention in time.

QoL was in most domains related to a specific subscale of resilience: the ASL subscale. This suggests interventions focused on improving acceptance of life and a balanced perspective in individuals with 22q11.2DS as well as young adults with ID might be helpful in improving QoL.

Finally, both for resilience interventions, psychological counselling, and social skills training, it is important to take into account the specific characteristics of the 22q11.2DS and their inherent vulnerability for severe psychiatric disorders.

Strengths and limitations

This is the first study that examined resilience and QoL of young adults with a 22q11.2DS and young adults with ID. The size of the 22q11.2DS group and the presence of a matched control group (young adults with ID) are two strengths of this study. After data collection, it appeared that the 22q11.2DS group was more varied in terms of age than the ID group. People with a 22q11.2DS also had significantly more (medical) health problems than the ID group. Overall our ID group was a good match to compare the resilience and QoL of persons with a 22q11.2DS. A third strength was the comparison with norms of persons in psychiatric care. This enabled us to demonstrate the intermediate position of the resilience of persons with 22q11.2DS.

A possible limitation may be that self-report questionnaires were used to assess resilience and QoL. In the future, it may be advisable to use a multi-informant method. On the other hand, the patient’s perspective in this study is also an added value, because this study provides an insight into the perception of people with a 22q11.2DS and ID on resilience and QoL. It is also important to mention that for the Dutch Resilience Scale norm groups are available for men, women, and patients with psychiatric symptoms, but not for ID. However, if the patient needed help or explanation for the completion of the questionnaires, a caregiver was allowed to assist the patient. A possible limitation is the way cognitive functioning was collected. We used the exact IQ-score when available in the medical records, but when unavailable, the caregiver filled in the level of cognitive functioning. However, we believe this only has a small effect on the quality of the data, since our analysis is performed on two groups of IQ: > 70 and < 70. We aimed to describe the personal and environmental characters of the participants as completely as possible. However, unmeasured variables, such as socioeconomic status, might have an effect on our resilience and QoL. We chose our control group based on a trait characteristic (cognitive impairment) of 22q11.2DS, but some differences between the two groups, such as age, might have an effect on the results.

Conclusions

The main conclusion of this study is the inherent vulnerability of resilience in young adults with a 22q11.2DS. Independent from several personal and environmental characteristics, young adults with a 22q11.2DS have a limited resilience in comparison to both the general population and to young adults with an ID. Regarding the level of resilience, they have an intermediate position between young adults with an ID and young adults in residential psychiatric care. Our results suggest different factors are underlying resilience in 22q11.2DS and ID in general. While resilience and QoL are correlated in both young adults with 22q11.2DS and in young adults with an ID, this seems only to be the case for the domain of acceptance of self and life. In addition while health problems do not appear to be related to the level of resilience, they do appear to be an important factor in most domains of QoL including perceived psychological health and quality of the environment.