Table 2 Quality ratings of development studies following the COSMIN methodology

DISABKIDS

V

V

V

V

V

D

D

Focus groups and individual interviews were conducted; however, information on experience and skills of group moderators/interviewers is missing [69].

V

D

N/A

D

Difficulties of children noticed; however, these were not considered in item reduction/reformulation process, which was only guided by quantitative data and expert consensus [52, 53, 69].

D

KIDSCREEN

V

V

V

V

V

V

V

V

I

N/A

I

Pilot-testing of 185 preliminary items resulted in adaptations; however, items were not tested in their final form and analysis and results were only described vaguely [48, 73].

I

KINDL-R Generic

V

V

V

V

V

D

D

Item pool developed from literature and discussions with children, but methods not clearly described [75, 76].

Missing information about development of revised version (KINDL-R); studies cited do not contain further information ([76] referred to in [49]) or are not accessible (habilitation thesis referred to in [77]).

V

D

N/A

D

Pre-test on comprehensibility of preliminary version conducted, but only healthy children (N = 28) and teachers, only written information, and revisions re-evaluated with only three healthy children [75].

D

KINDL-R Oncology

D

D

V

D

I

N/A

I

No published development study available. When approached for literature on this module, the responsible office of QOL measures at the university clinic Hamburg-Eppendorf (UKE) referred to the study by Ergin et al. [60] on developing a Turkish translation (mail, 04/16/2020). However, this study does not report anything about the development of the original module.

N/A

N/A

N/A

I

No CI study available.

I

PAC-QoL Child

V

V

V

V

I

N/A

I

No children were included in concept elicitation. For ethical reasons, Cataudella et al. [61] decided to only interview bereaved parents and HCPs at this stage.

V

D

N/A

D

CIs indicated that younger children had considerable difficulties in understanding, which has not been adequately addressed so far [62].

I

PedsQL Generic

V

V

V

V

A

D

D

No clear information is available on how the PedsQL was derived from the previous PCQL [see also 31]. Brief descriptions of PCQL development do not provide sufficient information on sample and methods [79, 80, 82].

N/A

N/A

N/A

I

No CIs conducted for the latest version(s) of the PedsQL Generic Core Scale.

I

PedsQL Cancer

V

V

V

V

A

D

D

Again, missing information on how the PedsQL was derived from PCQL [see also 31].

N/A

N/A

N/A

I

NO CI study available.

I

PedsQL Brain Tumor

V

V

V

V

A

D

D

Palmer et al. [63] state that it was “developed through focus groups with healthcare providers, children and parents, cognitive interviews, pre-testing, and field testing protocols.” (p. 288). Thus, it is assumable that children of the target population were involved. However, as no further information is given, the quality remains ‘doubtful’.

D

D

D

D

Missing information results in ‘doubtful’ ratings, see statement by Palmer et al. [63] provided as justification for rating of development studies.

D

PROMIS Ped Profile

per scale

V

V

V

V

V

V

V

A

A

A

Even though very good methods applied, rating off CIs can only be ‘adequate’, because 93% of the items have been tested in 5 or 6 patients only [85], which is insufficient to obtain a ‘very good’ rating in the COSMIN manual (7 required). Additional items developed by Quinn et al. [89] have been tested in at least 5 children per item.

Further item reduction relied on quantitative methods only (methods described in [103]; results per domain: [87, 90,91,92,93].

A

Anxiety and depression²

V

“

“

“

“

“

“

“

“

“

“

“

Fatigue

V

“

“

“

“

“

“

“

“

“

“

“

Mobility and upper extremity²

V

“

“

“

“

“

“

“

“

“

“

“

Pain intensity

V

“

“

“

“

“

“

“

“

“

“

“

Pain interference

V

“

“

“

“

“

“

“

“

“

“

“

Peer relationships

V

“

“

“

“

“

“

“

“

“

“

“

QOLCC-7-12

V

V

V

V

V

I

I

The qualitative methods applied were mostly ‘adequate’ to ‘very good’ [101]. ‘Inadequate’ rating because data were not rated by two independent investigators (Yeh was the sole author). However, it must be acknowledged that the results were discussed several times with faculty members, nursing students and participants, i.e., pediatric cancer patients.

D

D

D

D

Only three children were included in CIs and no further information is available on methods or results [64].

I

SQOLPOP

I

V

V

V

I

N/A

I

Kudubes and Bektas [67] state that after a “literature review, dimensions were formed to determine the quality of life from all aspects and item pools were developed for these dimensions” (p. 524), without naming these dimensions at any point. Thus, the definition of the construct to be measured was rated ‘inadequate’. No children were involved in the concept elicitation.

D

D

D

D

Pilot-study conducted with 25 children and their parents, who apparently had no negative feedback to give [67]. However, missing information on sample and methods results in ‘doubtful’ rating.

I

TACQOL

V

V

V

V

I

N/A

I

No children involved in concept elicitation and issue generation.

N/A

N/A

N/A

I

Pilot-testing of provisional item list and relied on quantitative methods only [58].

I