Development of a questionnaire to measure instrumental activities of daily living (I-ADL) in patients with brain tumors

Traditional outcome measures used in clinical trials in brain tumor patients
(both primary and metastatic) are progression-free and overall survival, next
to tumor response on magnetic resonance imaging (MRI). Apart from these
outcomes, information on the patient*s functioning and wellbeing has become
increasingly important, since in patients who cannot be cured from their
disease, the quality of survival is arguably at least as important as the
duration of survival.

One way to measure patients* functioning and well-being is with the assessment
of a patient*s health-related quality of life (HRQOL). HRQOL is a
multidimensional concept covering physical, psychological and social domains,
as well as symptoms induced by the disease and its treatment. In addition,
measurement of cognitive functioning is nowadays used to objectively assess how
the brain tumor patient is doing. Although HRQOL and cognitive functioning are
relevant outcome measures in brain tumor patients, these do not provide the
complete picture on the patient*s functioning in daily life. HRQOL covers a
wide range of domains, but it does not cover activities of daily living (ADL).
Moreover, cognitive dysfunction as measured with validated tests, is not easily
translated into the patient*s functioning in everyday life. Therefore, an
additional measure specifically aimed at measuring ADL is needed.

ADL are divided into two categories, basic activities of daily living (BADL)
and instrumental activities of daily living (IADL). BADL include basic
skills such as feeding, bathing, dressing, transferring, toileting and
continence. IADL on the other hand, include more complex activities such as
food preparation, ability to handle finances, shopping, doing laundry,
housekeeping, mode of transportation, responsibility for own medication, or
using a telephone. These capacities are required for autonomous function within
in the society. Because IADL are higher order activities, they may therefore be
negatively influenced by a cognitive decline, which is characteristic for brain
tumor patients. Therefore, especially limitations in these IADL are informative
of the brain tumor patient*s functioning in daily life.

Commonly used ADL scales and questionnaires in brain tumor patients such as the
Barthel Index (BI), Karnofsky Performance Score (KPS), Functional Independence
Measure (FIM) and the Functional Independence Measure - Functional Assessment
Measure (FIM-FAM) only cover BADL and not IADL. Even though the FIM assesses
whether there are problems in communication or social cognition, which involve
higher order activities, it is not assessed whether this has an influence on
activities of daily living. Moreover, these questionnaires have several other
limitations, because they were mostly developed in the late 1950s, with more
recent revisions in the early 1990s. First of all, advances in technology have
changed our daily environment dramatically (for example the use of mobile
phones, computers and household appliances). Secondly, quality demands of
self-report questionnaires have changed in the meantime. Although most
psychometric properties, in terms of reliability and validity, of the
questionnaires have shown to be adequate, there are also some limitations. For
example, the BI has marked floor- and ceiling effects, potentially limiting its
responsiveness to change. It thus seems that there is a need for a new
questionnaire, with superior psychometric properties, to measure IADL in
patients with brain tumors.

Although there is consensus that patients are the best source to rate their
functioning and well-being, proxy ratings should be considered as a potentially
appropriate alternative in brain tumor research because proxies might better
judge the patients* functioning in those situations where patients are
cognitively impaired or have a very poor health status. Since proxies are often
involved in the care of a patient, they have a good picture of the patients*
functioning and well-being. Proxy measures could therefore complement or
substitute patient self-assessment. In these cases, differences between patient
and proxy ratings do not necessarily reflect inaccuracy. However, to date it is
unknown if brain tumor patients are able to accurately estimate their own
functioning. Therefore, the aim of this study is to develop parallel versions
of this IADL questionnaire for brain tumor patients; a patient-based version
and a proxy-based version.

Primary Objective:
The overall aim of this study is to develop a reliable and valid IADL
questionnaire for use in brain tumor patients.
Parallel versions of this questionnaire will be developed; a patient-based
version and a proxy-based version.

Secondary Objective:
To determine to what extent patient and proxy ratings on IADL differ. Based on
this result, it can be determined if proxy measures are an appropriate
alternative in cases where the patient is unable to complete the questionnaire
(for example due to cognitive impairments or a poor health status). If
appropriate, questionnaires completed by the partner will be used instead of
the questionnaires completed by the patient.

Pilot study
Recently, a new proxy-based questionnaire was developed and validated to
measure problems in IADL in patients with early dementia. This questionnaire
was developed as a proxy-based questionnaire, because it is known that patients
with early dementia are not able to accurately estimate their own functioning.
The Amsterdam IADL Questionnaire® consists of 70 items which are divided into 7
different categories; (1) household activities, (2) household appliances, (3)
finances, (4) work, (5) computer, (6) appliances, and (7) leisure activities.
This questionnaire was developed with input from patients, proxies and
professional experts. It was shown that the questionnaire had favorable
psychometric properties.
Because there is no gold standard to measure IADL in brain tumor patients, and
because of the state-of-the-art properties and the expected similarities in
problems in IADL between both patients with dementia and brain tumors, we will
evaluate in a pilot study if this questionnaire is also appropriate for brain
tumor patients.
First, the items used in the Amsterdam IADL Questionnaire® will be
re-evaluated; it will be established whether these items measure activities
that are relevant for brain tumor patients. In addition, new items will be
generated. The objective of this pilot study is to develop a new set of items
measuring instrumental activities of daily life that are relevant for patients
with a brain tumor.
The methods used in this study will resemble the methods that were originally
used for the development of the Amsterdam IADL Questionnaire in dementia
patients.


Step 1: Evaluation of the activities in the Amsterdam IADL questionnaire®
Ten patients with a primary brain tumor and their proxies (total n = 2 x 10 =
20) as well as professional experts (n=6; 2 neuro-oncologists, 2 specialised
neuro-oncology research nurses and 2 neuropsychologists) will be consulted to
select the relevant and useful items, which are now used in the Amsterdam IADL
Questionnaire. When selecting the items, professional experts should keep in
mind three questions; (1) is the activity considered as IADL using the proposed
definition (=IADL are complex activities with little automated skills for which
multiple cognitive processes are necessary), (2) is the activity likely to be
affected in primary brain tumor patients, and (3) is the item clearly defined
and formulated? Patients and their proxies only need to answer two questions;
(1) is the activity affected during the disease course, and (2) is the activity
clearly formulated?

Step 2: Generating new IADL activities
Next, the same experts (n = 6) and another group of patients and their proxies
(total n = 2 x 5 = 10) will be consulted to explore activities affected in
patients with a primary brain tumor which are not mentioned in the existing
Amsterdam IADL Questionnaire®. To do so, in-depth interviews will be conducted
using the *sampling to redundancy* criterion, which means that persons will be
interviewed until no new themes emerge.

Step 3: Cognitive debriefing
Based on the results obtained in these first two steps, a new set of items will
be created that is believed to measure IADL in patients with brain tumors.
Then, a new group of patients/proxies will cognitively debriefe this new
questionnaire. This group, consisting of 6 brain tumor patients and their
proxies (total n = 2 x 6 = 12) will be requested to complete the questionnaire
while thinking out loud. With this technique it can be tested whether all
questions are interpreted as intended. Ambiguous or incomprehensible questions
will be rephrased or omitted from the questionnaire.

This pilot study results in a provisional item list that will be further
explored in the main study.

Main study
Phase I: Generation of items
Whether the list of issues identified in the pilot study is sufficient and
appropriate for international use remains to be determined. Therefore, the aim
of this phase is to extend the work of the pilot study and to compile an
exhaustive list of items that measure I-ADL relevant to brain tumor patients
(to ensure content validity) in several European countries (Austria, Italy,
United Kingdom). Therefore, four sources of information are used in phase I:
(1) literature, (2) provisional list of items pilot study, (3) patients and
their proxies and (4) health care professionals (HCPs).

Step 1: Literature review
A systematic literature review will be conducted to identify all relevant I-ADL
for brain tumor patients, by searching several databases such as PubMed, Embase
and PsycInfo. In addition, other sources such as the Quality of Life Group
(QLG) item bank (especially role functioning items) and PROQOLID will be
searched to identify existing, relevant issues. This search will result in an
exhaustive list of relevant I-ADL which will be used to add new issues to the
current provisional list of items, or to replace or rephrase inadequate items
that are already in the list.

Step 2: Review of current provisional list of items
Twelve patients and their proxies (4 patient-proxy dyads per country; 2 primary
and 2 metastatic brain tumor patients), as well as six HCPs (2 per country),
will be recruited to review the provisional list of 38 items measuring I-ADL in
brain tumor patients. Participants are asked to answer two questions: (1) is
the activity likely to be affected in brain tumor patients? and (2) is the item
clearly defined and formulated? Moreover, HCPs will answer one additional
question: is the activity considered as I-ADL using the proposed?

Step 3: Patient and proxy interviews
Semi-structured interviews will be conducted with 30 (5 interviews * 2 tumor
types (primary brain tumor or metastatic brain tumor) * 3 geographical regions)
brain tumor patients attending hospital outpatient clinics, and their proxies,
to explore all I-ADL affected in brain tumor patients. First, participants will
be asked to answer an open question about problems with activities in daily
living. Next, participants will be asked to review all activities revealed in
the literature search and from the pilot study, and to identify activities that
are currently missing. The interview will continue until no new activities
arise. Lastly, patients and proxies will be asked to indicate the relevance and
importance to their own situation of all these activities on a 4-point Likert
scale.

Step 4: Health care professional interviews
Semi-structured interviews will also be conducted with 18 HCPs (2
neuro-oncologists, 2 specialised neuro-oncology research nurses and 2
neuropsychologists per region * 3 geographical regions) in order to explore
activities affected in brain tumor patients. HCPs will be asked to indicate if
(1) the activities included are considered relevant for this patient group and
(2) if there are missing activities that are considered to be relevant. When
HCPs mention that there are irrelevant or missing activities, they should
provide a reason. Moreover, HCPs will be asked to identify the 10 activities
that mostly determine patients* functioning in daily living and should
definitely be included in the final questionnaire.

Step 5: Synthesis
The list of issues from all sources will be reviewed to produce a single,
comprehensive list of issues for formulation into the draft questionnaire (in
all languages) in phase II. Prior to item selection, specific decision rules
will be selected from the available EORTC QLG decision rules.

Step 6: Cognitive debriefing
In another 12 patients and their proxies (4 patient-proxy dyads per country; 2
primary and 2 metastatic brain tumor patients), a debriefing interview will be
conducted: the list of issues will be discussed while the participants is asked
to think aloud. This may be useful to refine items and avoid ambiguity or other
difficulties in the final provisional list.

Phase II: Construction of the item list
The aim of this phase is to construct a draft I*ADL questionnaire resulting
from issues identified in phase I. To do so, the list of relevant I*ADL issues
will be converted into questions which should be clear, brief and unambiguous.
Also, it will be explored if forming multiple item scales is feasible by
grouping several items with similar constructs, based on both empirical and
conceptual research data. Two versions of the draft questionnaire will be
developed: a patient*based version and a proxy*based version. Translations of
all questions will be carried out following the European Organistaion of
Research and Treatment (EORTC) Quality of Life Group Guidelines for Developing
Questionnaire Modules.
Similar to the EORTC QLQ*C30 questionnaire, a 4*point Likert scale will be
applied (ranging from *not at all* to *very much*) to assess difficulties in
performing the specified activities. In addition to this traditional scoring
system, we will add one option (*not applicable*) to each question to account
for the fact that some activities are never performed by patients (for example,
working). We will do this instead of using conditional questions to lower the
response burden. Moreover, questions will refer to the patient*s experience
during the past month, because many problems are unlikely to be captured within
a one week timeframe.
Lastly, two HCPs who were not involved in the first phase of this study will be
consulted to review all items for overlap and clarity.

Phase III: Pre-testing
In phase III, the resulting draft I*ADL questionnaire will be pre*tested in at
least 6 countries in 4 main European regions and non-European country (United
Kingdom, The Netherlands, Austria, Italy, Poland and Japan. The primary aim of
pretesting the questionnaire is to identify and solve potential problems
(phrasing, sequence questions) and to identify missing or redundant questions.
Second, preliminary psychometric testing will be performed.
The questionnaire will be administered to brain tumor patients (patient*based
version) and their proxies (proxy*based version) who were not involved in the
first phase of the project. In addition to completing the questionnaire,
respondents will be asked to rate each item on relevance and importance on a 4*
point Likert*type scale. Moreover, structured interviews with patients and
proxies will be conducted after completion of the questionnaire to determine if
the questionnaire is complete and if the questions are acceptable (not too
difficult, confusing or upsetting). Finally, both patients and proxies have to
indicate 10 items they deem the most important to get insight into which items
should be maintained and which might be redundant. Approximately 144 patients
(and their proxies) will be recruited in total (24 patients per country).
Patients will be stratified on two important parameters: tumor type (primary or
metastatic) and grade (grade II or grade III/IV), and presence of cognitive
deficits (present/not present). These cognitive deficits need to be established
by objective neuropsychological testing (patients only) and scores will be
compared to norm population.

Phase IV: Field testing
The questionnaire and its scale structure will be field-tested in a large,
international group of patients in order to determine its acceptability,
reliability, validity, responsiveness and cross-cultural applicability. The
same 6 countries as in phase III and possibly a 7th country (Croatia) will
participate in this phase.
The questionnaire will be filled in by the patients (patient-based version)
and the proxies (proxy-based version), which will be followed by the completion
of a debriefing questionnaire by each participant to determine the
acceptability of the questionnaire. The debriefing questionnaire comprises the
following questions: (1) How long did it take you to complete the
questionnaire?, (2) Did anyone help you to complete the questionnaire and, if
so, what kind of help and how much help was provided?, (3) Were there questions
that you found confusing or difficult to answer?, (4) Were there questions that
you found upsetting? and (5) Please use the space below if you have other
comments about the questionnaire.In addition, patients will also be asked to
fill in the Barthel Index (BI) (10 items) and the Medical Outcomes Study
Cognitive Functioning Scale (MOS Cog-R) (6 items), necessary for the known*
group comparison analyses.
To determine several aspects of reliability and responsiveness, both patients
and proxies are requested to complete the all three questionnaires again after
approximately two weeks and 3 months.
Lastly, the congruence between patient and proxy rating will be determined. The
number of patients and proxies recruited in this phase will be based on the
number of items in the questionnaire. We aim to recruit a minimum of 10
patients and their proxies per item in the IADL questionnaire, resulting in a
few hundred patients and proxies. Selection of patients should represent the
target population. We therefore aim to include patients with primary (both low
and high grade gliomas) and metastatic (both 1-3 and >3 metastases) brain
tumors, with and without cognitive deficits and in different stages of their
disease and in different treatment stages. This will ensure generalizability of
the questionnaire.

There are no direct benefits for patients and proxies participating in this
study. Nevertheless, their participation will contribute to the development of
a valid and reliable questionnaire to measure functioning in daily living of
patients with primary brain tumors. Eventually, this questionnaire can be used
in clinical trials and also in daily clinical practice. Results of the
questionnaire may facilitate treatment decision-making by informing physicians
about the value and impact of a specific treatment strategy and it may provide
the physician with additional information on the patients* well-being which
facilitates communication between the physician and the patient.

On the other hand, participation in this study has possible risks for the
patients and their proxies. Patients and proxies are confronted with all
problems that the brain tumor patients have with functioning in daily living,
which may pose a little psychological burden on them. Moreover, it will cost
the patients and proxies time to attend the interviews and complete the
questionnaires, although this is expected not to be substantial. Previous
experience with the Amsterdam IADL Questionnaire showed limited respondent
burden.
The exact burden for patients/proxies depends on the phase of the study. In
phase I (generation of items) interviews will be conducted at sereval steps
each with new participants. A session will last for approximately 60-90
minutes. Participants in phase III (pre-testing) are requested to complete the
questionnaire, they will be interviewed and undergo a neuropsychological
assessment. In total this will take approximately 120 minutes. Participants in
phase IV (field testing) are requested to complete the questionnaire three
times over a period of 3 months (baseline, after 2 weeks and after 3 months).
Completing the questionnaire will probably take about 25 minutes. No session
will take more than 120 minutes, the respondent burden seems therefore
relatively low.