Quality of life in children with cochlear implants in Kazakhstan

The sample involved parents/caregivers of children (child age at study mean = 7.33, SD = 3.28, range = 2.8-15) who attended CI follow-up care and consulted an audiologist at a time of this study. Altogether, 53 families with their children (73.6% males, 62.3% congenital hearing loss, 92.5% unilateral CI) who had used cochlear implants for at least one year participated in this study. The study was conducted between July 20th 2019 and February 20th 2020 in the Audiological Сenter of Almaty, Republic of Kazakhstan. Parents/caregivers independently (without any help of the audiologist or an interviewer) completed the questionnaire. Written informed consent was obtained from all parents/caregivers who agreed to participate in this study. Furthermore, parents/caregivers were asked some additional information related to their family and child, see Tables 1, and 2 and Fig. 1. The variables were categorized in the following way: age at assessment, age at CI activation and duration of CI was treated as continuous variables. Gender was coded; 1 for ‘male’ and 2 for ‘female’. Parental educational level was categorized into two groups; ‘high school’ for 2 and ‘university’ for 1. Attendance at school for the deaf was coded; 0 for ‘no’ and 1 for ‘yes’. Attendance at special educational institutions was coded; 0 for ‘no’ and 1 for ‘yes’. CI surgery was coded; 1 for ‘unilateral’ and 2 for ‘bilateral’. Participants, who completed the questionnaire were coded; 1 for ‘father’, 2 for ‘mother’, 3 for ‘grandmother’ and 4 for ‘caregiver’. Etiology of deafness was categorized; 1 for ‘congenital’ and 2 for ‘acquired’. Language (language used by the parent who completed the questionnaire) coded; 1 for ‘Kazakh’ and 2 for ‘Russian’.

Taking into account that some families might use both Kazakh and Russian languages equally in daily life, the language in which the questionnaire was filled out was considered to be the language of the family. All methods were carried out in accordance with relevant guidelines and regulations.

This research involved participants from the all regions of Republic of Kazakhstan. The exclusion criteria were: children using an implant for less than one year, and non-citizens of Republic of Kazakhstan. In Kazakhstan, children are considered as persons under the age of 18. Al-Farabi Kazakh National University Review Board approved this study (IRB – A060). Participants of this research received no compensation for their input.

The HRQoL questionnaire, Children with Cochlear Implants: Parental Perspectives was filled out by the parents/caregivers and had been widely used in the previous studies [14, 19‐23, 28, 29]. The Children with Cochlear Implants: Parental Perspectives is a validated, approved, CI-specific, closed-set questionnaire based on parental responses. It includes 74 statements covering two domains of the process of CI: decision-making (26 items) and outcomes of implantation (48 items). The outcomes of implantation domain consist of eight subdomains, including six child-related subdomains (‘communication’, ‘general functioning’, ‘well-being and happiness’, ‘self-reliance’, ‘social relations’, ‘education’) and two family-related subdomains (‘effects of implantation’ and ‘supporting the child’) (see Table 3 for subdomain descriptions and sample items). The CCIPP questionnaire is intended for children using the cochlear implant more than one year. This questionnaire was translated and adapted into Kazakh and Russian. The translation procedure consisted of direct translation by two independent translation agencies and a reverse translation by another two independent translation agencies. An audiologist then generated the final version of translated questionnaire. Direct translation from Kazakh to English and reverse translation were performed without intermediary translation to Russian.

Mean ratings were greater than three for all eight HRQoL subdomains on a 5-point Likert scale (mean = 3.65, SD = 0.4, range = 2.31–4.52), indicating that parents considered their child’s HRQoL as either average or being more positive than negative. ‘Well-being and happiness’ received the highest ratings (mean = 3.89, SD = 0.5, range = 2.8–5.0), followed by ‘supporting of the child’ (mean = 3.87, SD = 0.48, range = 2.67–5), ‘social relations’ (mean = 3.85, SD = 0.52, range = 2.29–5.0) and ‘self-reliance’ (mean = 3.80, SD = 0.64, range = 2.0–5.0) as shown in Fig. 2. ‘Communication’ (mean = 3.58, SD = 0.73, range = 1.17–4.67), ‘general functioning’ (mean = 3.57, SD = 0.51, range = 1.67–4.67) and ‘education’ (mean = 3.49, SD = 0.45, range = 2.57–4.43) also rated positive. The ‘effects of implantation’ received the least positive ratings (mean = 3.15, SD = 0.6, range = 1.43–4.33).

Table 4 presents all correlations between HRQoL subdomains.

Analysis of the sample (N=53) indicated age at assessment was not significantly correlated with any of the eight HRQoL subdomains. Age at CI activation and duration of CI use were also not significantly correlated with any of the eight HRQoL subdomains. In order to examine onset of hearing loss, the groups were divided into those with congenital deafness (n=33) and those children with acquired (n=20) deafness. For the group of children with acquired deafness, only age at assessment was significantly positively correlated with ‘general functioning’, ‘effects of implantation’ and ‘supporting the child’ (p < 0.01). Table 5 presents all correlations between above-mentioned variables and HRQoL subdomains. Duration of CI use was significantly correlated with age at assessment, so was not included in Table 5 and excluded from further analysis.

We identified correlations between language used by the parent who completed the questionnaire and three HRQoL subdomains including ‘well-being and happiness’ (Mann-Whitney test, p = 0.004), ‘supporting the child’ (Mann-Whitney U test, p = 0.008), and ‘social relations’ (Mann-Whitney U test, p = 0.039). Parents/caregivers of children with CIs who spoke Kazakh (mean = 3.99, SD = 0.54) showed a higher rating in ‘social relations’ than those who spoke Russian (mean = 3.72, SD = 0.47). Parents/caregivers of children with CIs who spoke Russian (mean = 4.08, SD = 0.55) had higher ratings of ‘well-being and happiness’, than those who spoke Kazakh (mean = 3.69, SD = 0.42). Parents/caregivers of children with CIs who spoke Russian (mean = 4.04, SD = 0.37) showed a higher rating in ‘supporting the child’, than those who spoke Kazakh (mean = 3.68, SD = 0.53). There were no statistically significant correlations between other socio-demographic variables, such as gender, parental education level and attendance at special educational institutions at the time of assessment and any of the eight HRQoL subdomains.

In all studies, using the CCIPP questionnaire, the subdomains achieved more than three points and the result was rated as positive [7, 14, 19‐23, 28, 29]. The present study hypothesis that HRQoL subdomains were generally positive was supported (mean ratings for all eight HRQoL = 3.65). The highest three subdomains were ‘well-being and happiness’, ‘supporting the child’, and ‘social relations’ which obtained ratings of 3.89, 3.87 and 3.85 respectively. In contrast, Kumar and colleagues (2015) obtained 3.65 for ‘well-being and happiness’; their highest rating was obtained for ‘communication’ with 3.93 [23]. Zhao et al. [22] and Byčkova et al. [20] reported the higher rating in ‘social relations’ with 3.72 and 4.05 respectively [20, 22].

Moreover, Kumar et al. (2015) noted that the high ratings in ‘communication’, ‘general functioning’ and ‘social relations’ subdomains were due to the filling out of questionnaires by parents/caregivers in the summer camp, where the children with implants were gathered together and professionals worked with them [23]. The same subdomains had a high level in the study of Huttunen et al. (2009), but children after CI with concomitant pathology were included in the sample [21] making direct comparison difficult. Zhao et al. [22] reported a low rating (3.45) in the ‘communication’ subdomain, consistent with the present study (3.59) [22]. Zhao et al. [22] speculated that CIs were used for a shorter period than in the study by Kumar et al. (2015) [22, 23]. In the present study the duration of CI use was 4.21 years, while in the study by Kumar et al. (2015) it was 7.47 years and 3.7 years in Byčkova et al. (2018) study [20, 23].

The lowest rating in the present study was in the ‘effects of implantation’ subdomain (3.15), which was consistent with: Kumar et al. (2015) (3.11); Byčkova et al. (2018) (3.16) [20] and Huttunen et al. (2009) (3.7) [20, 21, 23]. Low ratings of this subdomain suggested that parents’ observation of their child post-CI did not reach their pre-implant expectations for communication outcomes. Differences in the study samples and the age range at CI and age range at survey make it challenging to account for differences in findings. Of note, ‘communication’ and ‘social relations’ also obtained high ratings in past studies assessing HRQoL for children with CIs [6, 8, 17, 30]. Further direct comparisons of the present study HRQoL subdomains findings with Brewis et al. [29], Byčkova et al. [20], Zhao et al. [22] and Kumar et al. (2015) are in Table 6 [20, 22, 23, 29].

The associations between socio-demographic variables such as age at the time of the study, age at CI activation, duration of CI experience and psychosocial subdomains were not statistically significant in total sample (N=53). With regard to children with congenital deafness (n=33) there were no significant correlations between age at assessment, age at CI activation, duration of CI experience and any of eight HRQoL subdomains. Only age at assessment showed positive significant correlations with three psychosocial subdomains, including ‘general functioning’, ‘effects of implantation’, ‘supporting the child’ in children with acquired deafness (n=20). Dividing our sample into two smaller groups reduced statistical power and thus precludes making firm conclusions about HRQoL subdomains in our cohort. Due to the wide range in the age of child participants, presence/absence of child concomitant pathologies, differences in family socioeconomic status, greater/lesser parental participation in child development post-CI, and higher/lower parental education, past study findings regarding associations between child/family variables and child outcomes and quality of life vary [8, 10, 14, 21‐23, 29, 30].

With respect to the family’s socio-demographic variables, the present study demonstrated a positive relationship between the HRQoL subdomains and the language used by the parent who completed the questionnaire. Mann-Whitney U test indicated a positive association between the language used by the parent who completed the questionnaire and ‘well-being and happiness’ (p = 0.004), ‘social relations’ (p = 0.039) and ‘supporting the child’ (p = 0.008). The ‘social relations’ subdomain received a high rating when the language used by the parent who completed the questionnaire was Kazakh. In contrast, ‘well-being and happiness’, and ‘supporting the child’ subdomains received a high rating when the language used by the parent who completed the questionnaire was Russian. This can be explained by the history and traditions of nations of the Central Asian region, who used to live in large families (including not only mother, father, children, but also grandfathers, grandmothers, cousins, uncles, aunts) [31]. This way of life is preserved to this day among the majority of the population of this territory. This mentality, intrinsic to the Kazakh people, includes the significance of relations between relatives and neighbors; also, at school, university and at work. In Kazakh-speaking families parents (and often other close relatives) instill in children the need to establish relationships with people around them [31]. Another aspect of this is that in Kazakh-speaking families there are often three or more children, so parents probably cannot devote more time to the child with implant (work, life, welfare of other children). The differences in family life for Kazakh-speaking versus Russian-speaking families may account for some of the differences in parent ratings of HRQoL subdomains observed in the present study. As past studies have not compared Kazakh-speaking versus Russian-speaking participants, this is a finding unique to the present study that requires replication in a larger cohort. Kumar et al. (2015) surveyed their participants in English and Brewis et al. [29] also had English as an inclusion criterion [23, 29]. Zhao et al. [22] measured the quality of life of children within a Mandarin language environment, whilst Huttunen et al. (2009) carried out their study in Finnish [21, 22]. As indicated above, Kazakhstan is a bilingual country where Kazakh and Russian languages are used equally, so understanding any potential differences in HRQoL subdomains is important.

In the present study, there were no statistically significant associations between socio-demographic variables such as gender, level of parents’ education, or attendance at specialized educational institutions by the child post-CI and the HRQoL subdomains. Huttunen et al. (2009) and Zhao et al. [22] also showed no statistically significant associations between the number of children in the family and HRQoL subdomains [21, 22].

This study is not without limitations. Firstly, the study was carried out in one center, and although participants were from all over the Republic of Kazakhstan, patients who visited the other two centers did not enroll in the study, thereby, the sample size was limited. Secondly, parents/caregivers filled out the questionnaire by themselves, which could affect the study. We could not verify our data with departmental records, because we could not gain access to them, which is another limitation of our study. A further limitation of the present study was the inability to distinguish between various causes of congenital anomalies, such as CMV, genetic, hypoxia, jaundice, given that congenital abnormalities is a very heterogenous group. Additional needs of children were not considered in this study, which is another limitation of our report. We also chose not to use other questionnaires, such as CAP, SIR, MAIS and MUSS which could also be acknowledged as a limitation. Additionally, the fact that private tutoring with children after CI, plays a large role in their development, it would be correct to integrate the socioeconomic status of parents/caregivers, which our study did not consider. It is also necessary to notice that the audiological screening of newborns in Kazakhstan started in 2010, hence, data on the etiology of deafness prior to 2010 may have been less accurate for 12 children (out of 53) who were ten years and older at the time of this study. The age range is too wide, which did not allow conclusions to be drawn by age groups and excluded the possibility of identifying significant confounding factors.

The primary strength of this study is that for the first time in Kazakhstan an assessment was conducted of the HRQoL of children after CI. The assessment instrument was a translated and culturally adapted questionnaire, CCIPP, which is widely and universally used to indicate the HRQoL of children after CI throughout the world.