Evidence of Construct Validity for the Modified Mental Fatigue Scale When Used in Persons with Cerebral Palsy

ABSTRACT

Introduction: Fatigue impacts negatively on everyday activities in individuals with cerebral palsy (CP). More knowledge is needed about how mental fatigue is manifested in this target group. The purpose of this study was to gather evidence about the validity of the modified Mental Fatigue Scale (m-MFS) in adults with CP.

Methods: Mixed sequential exploratory design. The respondents were ten persons aged 22–56 with CP (MACS I–II).

Results: The respondents perceived the m-MFS as easy to read and understand. Its structure was characterised as straightforward and the text of the rating options was deemed to assist identification with life situations. Very good agreement was seen between the respondents’ and the instrument designers’ intended meaning for the items in the m-MFS; the weighted kappa was 0.92.

Conclusion: This study showed evidence of construct validity, based on response processes and content, for use of the modified MFS in adults with CP.

KEYWORDS:

• Mental fatigue
• cerebral palsy
• self-assessment
• self-management
• validity

Introduction

Fatigue, described as reduced capacity for physical and/or mental activity causing an imbalance in the resources an individual needs to perform activities¹ has been shown to be more prevalent in individuals with cerebral palsy (CP) than in the general population.^2,3 Studies have found fatigue to exert a major impact on health^2,4–6 and everyday activities in people with CP even though they describe a strong motivation to perform activities by themselves.^7–9 People with brain injury or other neurological impairments are more vulnerable to pathological mental fatigue than others.^10–13 In this study, pathological mental fatigue is defined as mental exhaustion appearing during cognitive activity or sensory stimulation and having a disproportionately long recovery period.^14,15 People with brain injury suffering from pathological mental fatigue have a decreased ability to generate resources to continue an activity.¹⁶ In people with neurological diseases, fatigue is proposed to be a consequence of metabolic and structural lesions in the brain¹⁷ – not an outcome of pain, sleep disturbances or depression.¹⁸ CP is a neurological disorder caused by a lesion in the developing fetal or infant brain which leads to activity limitations. It affects the development of movement and posture and is often accompanied by disturbances of sensation, perception, cognition and communication.¹⁹ Children with CP, even those without intellectual disability, often have impaired cognitive functions compared with peers without CP.^20–23 Adults with mild CP (less severe motor impairment and no intellectual disability) have described how cognitive restrictions impact their ability to plan, initiate and perform tasks in everyday life.⁷ In Sweden, adults with mild disability are expected to cope with integration into society without any special support (be enrolled in mainstream schools, move to their own home as adults and find work on the open market), while those with severe disabilities have a statutory right to special schooling, housing and sheltered employment.²⁴ Young adults with mild CP have identified the combination of high expectations from society and their physical and cognitive restrictions as the reason for their mental fatigue⁷ and highlighted mental fatigue as one of the most hindering factors in everyday life.^7,9 Even though mental fatigue is a well-known symptom in individuals with various neurological diseases^10,11,13, it has rarely been studied in people with CP.²⁵ However, mental fatigue has been reported in people with spastic bilateral CP²⁵, and people with CP have described how mental fatigue interferes with their everyday life.⁷

Fatigue is commonly assessed using rating scales and there are many self-reporting scales for different aspects of fatigue and for different diseases.^26–29 However, to the best of our knowledge, there is only one scale designed for use by individuals with CP: the Fatigue Impact and Severity Self-Assessment (FISSA)^26,30, and its focus is on ‘physical tiredness, muscle soreness, exhaustion of your muscles and body, or any related feelings’ (p. 6)²⁶, not explicitly on mental fatigue. Researchers have stated that there is a need for greater knowledge about how fatigue is manifested in individuals with CP^7,11,26,31, and studies have underlined the need to detect mental fatigue^7,32 and to find appropriate interventions to ensure a balance between activity and rest³³ in individuals with CP. Hence, to provide a broader view of how fatigue is manifested in persons with CP, there is a need to supplement FISSA with a self-reporting scale focusing on mental fatigue and related symptoms. Such a scale could form the basis for early interventions to prevent an unhealthy activity balance in people with CP, and it could also be used for follow-up purposes.^7,8,33

The Mental Fatigue Scale (MFS) is a self-rating scale designed to assess mental fatigue and related symptoms after neurological disease and injury in adulthood. Compared with other self-reported fatigue scales, the MFS includes a broader range of symptoms – all closely linked to mental fatigue and its impact on everyday life.^14,34 The fact that CP is caused by brain injury and that the items of the MFS correlate well with the symptoms described as causing mental fatigue in people with CP⁷ makes the MFS a promising candidate for use in this target group. However, some modification is required in that the assessment should not refer to changes after disease or injury; given that CP is related to prenatal, perinatal or postnatal brain injury.¹⁹ Whenever an instrument is modified or used in a new target group, evidence of validity must be obtained.³⁵ Validity refers to ‘the degree to which evidence and theory support the interpretation of test scores for proposed uses of the test’ (p. 11)³⁵ and is an important aspect of an instrument. According to the recommendation of the American Educational Research Association³⁵, validity should be described under the single overarching framework of construct validity, with five different sources of validity evidence. To evaluate whether the m-MFS is useful in the group of people with CP, two of those sources will be used in this study: evidence based on content and evidence based on response processes. The purpose of this study was to gather such evidence for a slightly modified version of the MFS (m-MFS) by investigating whether it captures experiences of mental fatigue and related symptoms in adults with CP, of less motor severity.

Methods

Design

This study had a mixed-method sequential exploratory design^36,37 consisting of two phases where different perspectives and sources were used to collect data aiming to gather evidence of construct validity for a group of people with congenital or early acquired disabilities (cerebral palsy).

In the first, descriptive phase, evidence based on response processes was obtained, i.e. information about how the m-MFS was interpreted and used by the respondents. The respondents were asked to fill in the self-rating form and express their thoughts aloud while they self-rated their own fatigue. Think-aloud interviews and retrospective questions were used to capture the respondents’ perception, interpretation and experiences of the m-MFS. Think-aloud methodology can provide important insights about the functionality of a questionnaire.^38,39 Its purpose is to capture respondents’ thoughts while performing a task^40-43, in the present study while carrying out self-rating with the m-MFS, and to identify differences across respondents.⁴⁰ This method is especially appropriate when the task is language-based and can be divided into short units.⁴⁰ According to Wilson⁴⁴, it is useful to combine think-aloud interviews with retrospective questions, which emphasize or complement the think-aloud method.^44,45 Therefore, the descriptive phase of this study included both think-aloud interviews and retrospective questions. In addition, descriptive statistics were used to study the distribution of the self-ratings made and to identify any differences in self-rated levels across items on a group level.

In the second, comparative phase, evidence based on test content was obtained by comparing the respondents’ ratings with the ratings determined by the designers of the m-MFS on the basis of the respondents’ thoughts about why they had chosen a specific rating option. The qualitative data collected were used to perform a quantitative analysis, in line with claims to the effect that the use of mixed methods is appropriate when searching for evidence of validity.^36,37,44 The purpose was to assess the level of agreement between the respondents’ ratings and the intended meaning of the items of the m-MFS.

Respondents

The respondents were invited to participate in the study via two Swedish rehabilitation centres for adults. Purposive sampling was used. The decisive factor was a willingness to think aloud and share experiences while filling in the self-reporting scale⁴⁰; there was no explicit requirement for experience of fatigue. The inclusion criteria were age 20–65, a diagnosis of mild CP classified as level I or II in the Manual Ability Classification System (i.e. able to handle objects easily or with somewhat reduced quality)^46,47 and at level I or II in the Communication Function Classification System (i.e. able to have a discussion).⁴⁸ The exclusion criterion was intellectual disability. Ten respondents aged 22–56 years participated. Descriptive information regarding each respondent is given in Table 1.

Table 1. Respondent characteristics.

Respondent (R) No	Age	Gender	CP subtype	GMFCS	MACS	CFCS	Employment status (% of full time)	Residential status	Area of residence
R1 (pilot)	38	female	USCP	I	II	I	employed (60%)	living with partner and children	rural
R2	30	male	BSCP	II	II	I	employed (75%)	living alone	urban
R3	24	male	BSCP	I	I	I	student	living with parents	urban
R4	22	male	USCP	II	II	I	unemployed	living with parents	rural
R5	28	female	BSCP	I	I	I	employed (100%)	living with partner	urban
R6	44	female	BSCP	IV	II	I	employed (50%)	living with partner and children	urban
R7	31	male	BSCP	III	II	I	employed (25%)	living with partner and children	rural
R8	42	female	BSCP	I	I	I	employed (75%)	living with partner and child	urban
R9	43	male	BSCP	II	I	I	employed (75%)	living alone	urban
R10	56	female	BSCP	III	II	I	employed (50%)	living alone	urban

Acronyms: GMFCS = Gross Motor Classification Scale, MACS = Manual Ability Classification Scale, CFCS = Communication Function Classification Scale, BSCP = Bilateral Spastic Cerebral Palsy, USCP = Unilateral Spastic Cerebral Palsy.

The Regional Ethical Review Board of Gothenburg, Sweden, has approved the study (Ref. No. 128–17). Written informed consent was given by all respondents.

Instrument

The Mental Fatigue Scale

The original MFS consists of an introductory text, 14 items each consisting of a question and descriptive rating options, including examples, and a final 15th item concerning variations in fatigue. The introductory text briefly describes how to use the scale. The items are based on the symptoms of astheno-emotional disorder, i.e. symptoms occurring after acquired brain injury^15,49, meaning that they relate to affective, cognitive and sensory symptoms, the duration of sleep and time-of-day variation.^14,34 The first 14 items are about fatigue in general, lack of initiative, mental fatigue, mental recovery, concentration difficulties, memory problems, slowness of thinking, sensitivity to stress, increased tendency to become emotional, irritability, sensitivity to light and noise, and decreased or increased sleep. Items 1–14 each have four descriptive rating options (0–3) with half-points in between, i.e. seven options in all. The descriptive rating options are statements describing aspects of a real-life situation that respondents may identify with. Basically, the whole-point options have the following meaning: 0 = no problem, 1 = slight problems, 2 = fairly serious problems and 3 = serious problems.¹⁴ Johansson and Rönnbäck³⁴ have suggested a cut-off score of 10.5 (out of 42; the 15th item are not included as it has a range of 0–2) for the MFS, meaning that a score above 10 indicates the presence of fatigue-related problems. Johansson et al¹⁴ have shown there to be significant correlations among the 14 items, and the items have been found to have adequate internal consistency, with a Cronbach’s alpha of 0.944. Chiu et al⁵⁰ have reported that the MFS can be used in a population culturally different from the one originally targeted, stating that the Chinese version of the MFS has ‘satisfactory measurement properties to quantify mental fatigue in patients with traumatic brain injury‘ (p. 7).

Modified Version of the Mental Fatigue Scale

The work to produce a modified version of the MFS was a joint effort by the first and last authors and the third author, who was one of the designers of the original MFS. This set-up enabled a deeper understanding of the decision-making process during the design of the original MFS. Changes were made to the original MFS based on knowledge about why the MFS has its specific design and also considering that it would be used by people with congenital or early acquired brain injury. As few changes as possible were made. Basically, the questions and rating options were modified so that, instead of referring to perceived differences in fatigue before and after an acquired injury, they related to the situation in the past four weeks. Further, the examples for two of the rating options were changed to better suit the target group. In items 13 and 14, the rating options were changed so that respondents would choose among specific numbers of sleep hours instead of comparing their present situation with their earlier one. Finally, item 15 was removed because it was a supplementary question to the first 14 items. The new content in the m-MFS was discussed during a new session and a good level of agreement was found between the first and last authors and the MFS designer (and third author). Hence the m-MFS tested in this study consisted of 14 items intended to be answered by self-assessment based on the individual’s life situation in the past four weeks (the m-MFS as tested is included in Appendix).

Data Collection

Descriptive Phase

The think-aloud interviews were conducted by the first author. The pilot interview (with respondent R1) was rich in content, and it was decided to use the material from that interview in the study. The interviews took place at a location convenient for each respondent and lasted between 35 and 85 minutes. Before the interview, the respondent received information about the topic of the interview and his or her rights, and demographic information was collected (Table 1). Then the respondent was asked to fill in the m-MFS self-rating scale while verbalising his or her thoughts and feelings about the self-rating scale. No information was given about how to complete the form (besides that provided in the introductory text). To avoid interference with the respondent’s own thoughts, the interviewer was silent (while showing interest using means other than speech). Field notes were taken about the respondent’s non-verbal communication and about verbalised thoughts that required additional questions. The interviewer used prompts such as ‘keep talking’ or ‘keep thinking aloud’ if the respondent was quiet for a long time. Since thinking aloud for more than a few minutes at a time can be taxing, the think-aloud interview was conducted one item at a time. After each item, the interviewer asked retrospective questions to clarify or expand the respondent’s thoughts. The interviews were digitally recorded and transcribed verbatim by the interviewer. All transcriptions were double-checked against the recording and corrections were made. The self-ratings made by the respondents during the think-aloud interviews were used for descriptive statistics.

Comparative Phase

To be able to compare the respondents’ ratings with the intended meanings of the items, the two designers of the original MFS chose, for each respondent and each item, rating options based on the respondent’s thoughts. For this purpose, the designers used a compilation of each respondent’s own thoughts and arguments about why he or she had chosen a certain rating option. A session was held where the designers informed the first author about their chosen ratings and these ratings were then compared with the respondents’ ratings.

Analysis

Descriptive Phase

As a preparatory step, all text read out by the respondents was checked against the original text of the m-MFS and all reading errors made by respondents were marked. The analysis of the transcribed interviews was then carried out in accordance with Fonteyn.⁴³ During the first part of the analysis, called ‘Referring Phrase Analysis’, the first author read all the interviews to obtain an overview of the material and identify the most significant elements in the interviews. The transcribed interviews were imported into the NVivo 10 software⁵¹ to facilitate coding and analysis. Nodes were created in NVivo on the basis of the structure of the m-MFS (introduction and 14 items) with the addition of ‘general thoughts about the self-rating scale’, yielding a total of 16 nodes. The content of each interview was allocated line-by-line to the different nodes. Then the material in each node was read, and phrases in each respondent’s verbal data which had the same central meaning and were related to each other were identified and coded. The codes and their definitions were revised several times. During coding, it became clear that the respondents’ reasoning mainly concerned three areas: thoughts about the text, thoughts about the design of the self-rating scale and thoughts about their self-rating (including why they had chosen a specific rating option). Finally, the last author analysed a random subset of the transcripts, and some revisions had to be made to the categories before the first and last authors could reach a consensus. During the second part of the analysis, called ‘Assertional Analysis’, the first author studied in what way the respondents expressed their thoughts verbally and with body language. If their thoughts were said with strong emotions, or if it was just mentioned in passing. The respondents’ expressions during statements were identified by means of a second analysis of the transcribed meaning units, the field notes and the interview recordings. During the third and final part of the analysis, called ‘Script Analysis’, the first author used the material from the Referring Phrase Analysis and the Assertional Analysis to produce an overall description of how the respondents structured their use of the m-MFS, their rationales for the decisions they made and their final plan for self-rating. The respondents’ most frequent reasoning processes were identified and defined. Finally, the categories describing the reasoning process emerged in a discussion involving all six authors.

Descriptive statistics were used to study the distribution of the self-rating options selected by the respondents, and the median of the ratings for each item was calculated.

Comparative Phase

The respondents’ self-ratings were compared with the ratings made by the designers based on the intended meaning of the items. The level of agreement between the respondents’ and the designers’ ratings was analysed using Excel for Microsoft Office 365 to calculate percentage agreement (PA) and weighted kappa (kw).⁵² In line with Katzin⁵³, a PA above 80% was deemed satisfactory. For kw, the interpretation suggested by Altman⁵² was used: <0.20 = poor; 0.21–0.40 = fair; 0.41–0.60 = moderate; 0.61–0.80 = good; 0.81–1.00 = very good.

Results

The results of the present study reflect the extent to which the respondents’ thoughts about the m-MFS and their self-ratings fit the intended construct of the self-rating scale.

Respondents’ Thoughts and Perceptions

The analysis of the respondents’ views on the m-MFS yielded five categories: (i) ‘Understanding of the text’ – the extent to which the respondents understood the text of the m-MFS; (ii) ‘Rating time span’ – the respondents’ perception of the period addressed by the m-MFS; (iii) ‘Identification with content’ – the respondents’ views on how well the rating options corresponded to themselves and their life situation; (iv) ‘Design of rating options’ – the respondents’ views on the design of the rating options; and (v) ‘Reasons for selecting options’ – the respondents’ own thoughts and arguments about why they selected specific rating options. Below are more detailed descriptions with illustrative quotations which have been translated from Swedish.

Understanding of the Text

All respondents read the introductory text of the m-MFS without difficulty and without making any reading errors. They found the introductory text to be clear, easy to understand and unambiguous. Further, they pointed out that all questions were relevant, that the purpose of the questions and rating options was obvious and that the explicitness of the rating options facilitated understanding. They also noted that the links between headings and questions were excellent overall.

I think the questions form a coherent whole, that they sort of latch on to each other. The headings are good and it all kind of sticks together, the questions don’t sprawl out in different directions. (R8)

The structure of the questions was perceived as obvious and the headings were mainly perceived as suitable. When reading the text aloud, a few respondents had difficulties pronouncing four (fairly infrequent) words in different headings. Some respondents claimed that the word ’irritability‘ (irritabilitet) (found in the heading of question 10) was new and unknown to them, but all respondents understood that heading after having read the actual question. However, even though all respondents eventually understood the word ‘irritability’ (irritabilitet), four of them suggested that it should be replaced with ’easily irritated‘ (lättirriterad).

I may understand that it’s to do with being irritated, but I’ve never heard that word before. … Because those who have a little more problems than I do with written language, they wouldn’t have been able to handle it. (R6)

Further, one respondent (who was not a native speaker of Swedish) claimed not to understand two passages including idiomatic expressions but was still able to understand the meaning of the questions concerned. In cases where the descriptions of rating options included several examples, the respondents found it difficult to pay sufficient attention to the conjunctions ‘or’ (eller) and ’and‘ (och) between the examples. They suggested that those words should be underlined to facilitate understanding.

I’m probably a bit thick, but I get it. It’s either–or here, isn’t it? I have slight problems falling asleep OR my sleep is light, restless or short. One of those things? (R5)

The changes suggested by the respondents to make the text easier to understand are listed in Table 2.

Table 2. Changes to the modified version of the Mental Fatigue Scale suggested by the respondents.

Category	Question (Q) number and, where relevant, response-option (A) number	Respondent (R) who suggested the change	Present wording	By the respondents proposed wording (changes underlined)
Understanding	Q2, A3	R1	I can’t do the simplest of everyday tasks (eating, getting dressed). I need help with everything.	I can’t do the simplest of everyday tasks, e.g. eating, getting dressed. I need help with everything.
Understanding	Q4	R8	If you have to take a break, how long do you need to recover after you have worked ‘until you drop’ or are no longer able to concentrate on what you are doing?	If you have to take a break, how long do you need to recover when you have worked ‘until you drop’ or are no longer able to concentrate on what you are doing?
Understanding	Q10, heading	R7, R8, R9, R10	Irritability or ‘a short fuse’	Easily irritated or ‘a short fuse’
Understanding	Q13, A1	R10	I have slight problems falling asleep or my sleep is light, restless or short.	I have slight problems falling asleep or my sleep is light, restless, or I just sleep for short periods.
Identification	Q2, A2	R8	It takes a great effort to start things. This applies to everyday activities such as getting out of bed, washing myself and eating.	It takes a great effort to start things. This applies to everyday activities such as getting out of bed, going to bed, washing myself and eating.
Identification	Q5, A2	R7	I find it so difficult to concentrate that I have problems, for example reading a newspaper or taking part in a conversation with a group of people.	I find it so difficult to concentrate that I have problems, for example reading a newspaper, taking part in a conversation with a group of people, or concentrate when hearing a dripping water tap.
Identification	Q6 Q6, A1	R1, R8, R3	Do you forget things often, do you need to make [paper] notes or do you have to search for things at home or at work?	Do you forget things often, do you need to make notes or do you have to search for things at home or at work?
Identification	Q8	R4	Do you find it difficult to cope with stress, that is doing several things at the same time while under time pressure?	Do you find it difficult to cope with stress, that is doing several things at the same time and/or under time pressure?
Rating options	Q14, A0	R6	I sleep between 6 and 8 hours per night.	I sleep between 6 and 8 hours per night or less.

Rating Time Span

The results show that not all respondents complied with the instruction to the effect that they should think about the situation in the past four weeks when self-rating. Instead, some thought of their situation in general, except in two questions where the time span was explicitly mentioned in the question. However, it did seem that the introduction influenced the respondents’ thinking such that their self-assessment did not refer to their present-day status.

Well, I think it’s good that it says four weeks, in fact. Because if it said ‘right now’ that could be taken to mean, actually, what it’s like today … So it’s good anyway that it specifies a time period. (R4)

Identification with Content

The respondents described how, in most cases, they could easily identify with one of the rating options under each question and that this feeling of identification with an option facilitated their self-assessment. Further, they pointed out that the detailed questions and rating options gave rise to many thoughts about everyday-life situations and mental fatigue.

This test [self-report form] really affects me, actually. It really, really makes you think about fatigue … It’s the strongest self-report form I’ve ever met [experienced] or had. So detailed. (R7)

The content of question 4, about mental recovery, was easy to identify with for most of the respondents, but three of them had difficulties imagining a situation like the one described in the question. Those three therefore found it somewhat difficult to make a self-assessment, but even so they all succeeded in doing so.

I can’t really feel that I’ve ever worked until I lost the ability to concentrate, or it’s probably happened, but I can’t really say … [pause for thinking] … say why that was and that sort of thing. … So I think that one [the question] is fairly hard to answer. (R3)

One respondent was not sure whether his/her fatigue was mental or physical, which made it more difficult to select a rating option for question 3 (mental fatigue). Two respondents perceived some difficulties with question 8 (sensitivity to stress), because they experienced stress when doing several things at the same time even without time pressure, and the question included ‘while under time pressure’ in its definition of ‘stress’.

The examples included in the descriptive rating options were perceived as appropriate and as helping the respondents to identify the rating option that best described their own life situation. However, it was proposed that a few new examples should be added. Specifically, it was suggested that option 2 under question 2 (lack of initiative) should be supplemented with ‘going to bed’, as the respondents found it harder to summon the energy to go to bed than to get up in the morning. Further, it was pointed out that sensory impressions from the environment influenced the respondents’ ability to concentrate and it was suggested that an example of such disturbing stimuli that could affect concentration should be added to option 2 under question 5 (concentration difficulties). Finally, the respondents also recommended a change in question 6 (memory problems) and in option 1 under that question, whose Swedish versions refer to pieces of paper used to make notes on, because they generally used their mobile phones rather than paper to make notes. (The English version of both the MFS and m-MFS reads ‘make/making notes’, which is neutral as to the technology used.)

The changes suggested by the respondents to improve the text of the m-MFS with regard to identification are listed in Table 2.

Design of Rating Options

The respondents emphasised that the descriptive rating options, which included examples of life situations, made things clearer than what would have been the case if there had only been a rating-scale consisting of No problem (0), Slight problems (l), Fairly serious problems (2) and Serious problems (3).

It’s precisely having these detailed statements, actually. … So it’s very good that there’s a clear statement … for each number. Instead of one statement and five different answers [response options]. (R)

The half-points without descriptive text were perceived as helpful in that they gave the respondents an opportunity to combine two descriptive rating options that they agreed in part with.

All respondents completed question 13 (decreased sleep at night) and 14 (increased sleep), but some perceived those questions demanding because of the need to calculate sleep hours. It was suggested that the design of the rating options could be improved by adding the words ‘or less’ to option 1 under question 14 so that respondents who slept less than 6 to 8 hours would have an obvious option to select. This proposed change to improve the text of the m-MFS with regard to the design of the rating options is shown in Table 2.

Further, the respondents suggested that a supplementary question should be added to the m-MFS to reflect the fact that it is to be filled in by people who have lived their whole life with disability and are likely to have adapted their lives. The proposed question would assess whether respondents have adapted their everyday life to avoid fatigue. According to the respondents, such a question might yield information indicating whether the overall score was lower because of such adaptations.

If you’ve changed your life, adjusted your life to change fatigue levels, or whatever. Then you’ll probably score lower because you’ll have a little more energy (R4)

Reasons for Selecting Options

The respondents’ descriptions of their thoughts while assessing their mental fatigue and related symptoms were rich in content. Their thoughts about why they selected a specific level among the rating options available included extensive descriptions of the rating options along with several examples of activities and situations where mental fatigue affected their lives.

Well, I think I can put 1.5 there, actually. I sort of forget things I need to take with me sometimes. And I have to go back and check. Right, did I bring that one along or didn’t I? (R9)

Descriptions of the Respondents’ Ratings

The fatigue problems were self-rated highest for the items concerning mental fatigue, mental recovery, concentration difficulties, slowness of thinking and sensitivity to stress. The range of rating options selected in each item was wide. Eight of the respondents scored above the cut-off of 10. Table 3 shows each respondent’s self-assessment.

Table 3. Self-assessment by each respondent.

Question	R1	R2	R3	R4	R5	R6	R7	R8	R9	R10	Median/item (max 3)
1 Fatigue	1	0.5	0	1	3	0.5	1	1.5	1.5	1	1
2 Lack of initiative	0	1	1	1	2.5	0.5	2	1	1.5	1	1
3 Mental fatigue	1.5	1	1	0.5	2	1	1	1.5	1.5	1.5	1.25
4 Mental recovery	2	1	1	1	1	3	1.5	1.5	1.5	0	1.25
5 Concentration difficulties	1	0.5	1	0.5	2	1	1.5	2	2	1.5	1.25
6 Memory problems	0	0.5	0.5	0	0	1	1	1	1.5	1.5	0.75
7 Slowness of thinking	0.5	0	1	1	2.5	0	2	1.5	2	2	1.25
8 Sensitivity to stress	1	1	1	1	1.5	2	1	1	2.5	2.5	1.25
9 Increased tendency to become emotional	1	1	0.5	0	0	1	2	1	2.5	0.5	1
10 Irritability or ‘a short fuse’	1	0.5	1	0	0	1	1	0.5	3	1	1
11 Sensitivity to light	0	0	0	0	1.5	0.5	0	1	0.5	0	0
12 Sensitivity to noise	1.5	1	1	0	1.5	1	2	1	1	1	1
13 Decreased sleep at night	0	0	1.5	1.5	0	2	0.5	0.5	1	1	0.75
14 Increased sleep	0	0	0	0	0	0	2	0	0	1	0
Sum/respondent (max 42)	10.5	8	10.5	7.5	17.5	14.5	18.5	15	22	15.5

Each respondent’s self-assessment, the sum of each respondent’s ratings (bold figures horizontally), median per item (bold figures vertically)

Agreement between Ratings

The comparison between the respondents’ self-ratings and the designers’ ratings showed very good agreement: k_w = 0.92. PA was 89%, meaning that in most cases the respondent and the designers selected the same rating. Further, where there was disagreement between the ratings, this amounted in most cases only to a difference of one level (0.5 points); see Table 4.

Table 4. Ratings by respondents and designers.

Percentage agreement (PA) = 89% (124/140). The figures in the diagonal indicate the number of instances where the respondent and the designers selected the same rating-scale level.

Discussion

This study showed that the m-MFS has the potential to be used as a mental-fatigue self-rating scale for persons with CP. It yielded evidence of construct validity based on both response processes and test content.³⁵ The respondents perceived the m-MFS as relevant and easy to understand. They felt that it was easy to identify with the questions and the rating options. The descriptive self-rating options with examples from real life were highlighted as helpful when self-assessing. This showed that the design of the m-MFS – both the questions and the rating options – corresponds well with the perception of fatigue experienced by adults with CP, yielding evidence based on test content. Further, very good agreement was found between the respondents’ reasoning about why they selected a specific rating and the designers’ intended meaning for the rating options, yielding evidence based on test content. However, the validation of a new instrument is an ongoing process. Therefore, further research is needed into other sources of construct validity for the m-MFS.

Previous studies have shown that cognitive dysfunction complicates everyday life and causes mental fatigue in individuals with CP.^7,20,21 The MFS was chosen as the basis for the instrument to be used in the present study because it focuses on a broad spectrum of symptoms linked to mental fatigue^14,34 rather than on physical or general fatigue. The results show that the respondents could easily identify with the questions and rating options. Further, there was good agreement between the respondents’ own ratings and the ratings chosen by the designers on the basis of the respondents’ reasoning about their choice of ratings. According to Wilson⁴⁴, the likelihood of obtaining an answer which is consistent with the real-life situation increases when respondents are able to identify with the rating options. Hence there is reason to believe that the descriptive content of the rating options increased the ability of the individuals with CP to understand the intended meaning of the self-rating scale.

Research has shown both how extremely important it is for individuals with mild CP to perform activities themselves^7,54,55 and how physical deterioration, pain, stress, depression and fatigue negatively affect everyday activities in individuals with CP.^6,25,32,56 This illustrates the challenging life situation often faced by adults with CP, where difficulties finding an activity balance in life are constantly present. Finding the right balance between activity and rest⁵⁷ while at the same time having opportunities for doing, being, becoming and belonging is a key predictor of an individual’s sustainable development.⁵⁸ Sienko et al³² have stated that all persons with CP should be offered assessments of pain and fatigue early in life. Using the m-MFS in early adulthood to obtain a fatigue baseline could enable identification of increased fatigue through later re-assessments. The assessment results could then be used as a basis for discussions with the individual about his or her performance in everyday activities and its relationship to fatigue. If the person is at risk for pathological mental fatigue the person could be offered treatment to reduce unhealthy activity balance, e.g. the Swedish version of the course ‘Managing fatigue: A six-week course for energy conservation’.⁵⁹ There is an interactional relationship between fatigue, pain and depression, so it is essential to focus on all three parameters when assessing problems related to them.²⁵ Judging from the results of this study, the m-MFS may be a good tool for predicting some parts of the need for interventions to prevent imbalances in daily life.

Further, a dialogue with the individual about the outcome of the self-assessment using the m-MFS may also increase his or her self-awareness, since the instrument highlights how cognitive, affective and sensory symptoms are linked to fatigue. With greater knowledge about the reasons for their difficulties, individuals with CP may be better able to explain the background to those difficulties and so may avoid being misinterpreted to a greater extent.⁷ According to Brunton et al.⁸, health-care professionals have a responsibility to create opportunities for people to gain access to the process of self-awareness about their disability and abilities. In individuals with CP, the m-MFS could be one of the instruments used to increase awareness of what causes their fatigue.

When the respondents in this study thought aloud about the self-rating options, they talked about the more or less extensive adaptations that they had made to their life situation in order to avoid fatigue. Many of them described how their need for rest reduced the time available for everyday activities and social contacts, and how their reduced working hours resulted in poorer finances. Against this background, they suggested that the m-MFS should be supplemented with a question eliciting information about whether a respondent scored lower because of such adaptations. Moreover, the self-assessments made by the respondents in the present study demonstrated a higher level of fatigue than what is normal in the general population.³⁴ Eight out of ten respondents scored above the cut-off of 10 – even though the presence of fatigue was not an inclusion criterion. Hence, although the respondents had adapted their life situation to their propensity for fatigue, their scores still suggested problems with fatigue. While it should be kept in mind that the results of the present study are based on rather few respondents, it needs to be taken into consideration in clinical work that patients with CP may be in a similar situation.

Based on the study findings, a few changes (relating to aspects which yielded frequent comments) should be made to the text of the m-MFS to ensure an even better fit with the target group of adults with mild CP. Two words, those for ‘irritability’ and ‘paper notes’, need to be replaced with a more common word and a more appropriate one, respectively. In order to reduce the respondents’ perceived difficulties paying attention to the conjunctions corresponding to ‘and’ and ‘or’ between examples in the rating options, those words should be underlined. Finally, additional text is needed in questions 13 (decreased sleep at night) and 14 (increased sleep) to facilitate self-assessment.

The respondents in this study expressed a high level of satisfaction with the instrument and very good agreement was found between the respondents’ reasoning behind their choice of ratings and the designers’ intended meaning of the rating options. However, as this is the first study to obtain evidence for the validity of the m-MFS, further research is required to evaluate the m-MFS from other perspectives.

Study Limitations and Strengths

One limitation of this study is that rather a small sample was used as the basis for the descriptive statistics and comparative phase. However, the use of mixed methods is likely to improve the quality of the study. Another limitation is that this study evaluated only parts of the construct validity of the m-MFS. In addition, the reliability has not been tested and the English version of the m-MFS has not yet been tested for its psychometric properties.

One strength of the present study is that the respondents read the instrument text aloud, which enabled the interviewer to identify reading errors. This yielded useful data about the response process.³⁹ To avoid overloading the respondents’ working memory, the interviews were so organised that the respondents thought aloud about one question at a time and were then asked retrospective questions by the interviewer before moving on to the next question.^39,40 Moreover, even though the study used purposive sampling, the respondent group recruited represented a good mix with regard to age, gender, employment status, residential status and residential area (urban or rural); it is to be hoped that this strengthened the richness of the interview material.

Conclusion

This study showed evidence of construct validity based on response processes and content for the use of the modified Mental Fatigue Scale (m-MFS) in adults with cerebral palsy (CP). The respondents perceived the m-MFS as easy to read and understand. Its design was deemed to help the respondents recognise different life situations. Consequently, only minor changes are recommended in order to clarify the text of the m-MFS. The study also showed very good agreement between the respondents’ self-ratings and the designers’ intended meaning for the items. Although further psychometric properties of the m-MFS has to be tested, this study shows that the m-MFS has the potential to be a useful instrument for self-rating of mental fatigue in individuals with CP.

Declaration of interest

The authors have no conflicts of interest to report. The authors alone are responsible for the content and writing of the paper.

Acknowledgments

The authors would like to send a special thank you to the ten adults with CP who participated in this study.

Additional information

Funding

The study was supported financially by the Norrbacka Eugenia Foundation, Sweden. The authors gratefully acknowledge this support.

Appendix

Modified version of the Mental Fatigue Scale

Self-rating of mental fatigue for persons with an innate or early acquired disability.

Please self-rate your present condition, that is how you have felt during the past four weeks. Each question below is followed by four statements that describe: No (0), Slight (1), Fairly serious (2) and Serious (3) problems. We would like you to place a circle around the figure before the statement that best describes your problems. Should you find that your problem falls between two statements, there are also figures to indicate this.

_________________________________________________________________________________________________________________________

1. Fatigue

Have you felt fatigued during the past month? It does not matter if the fatigue is physical (muscular) or mental. If you recently experienced something unusual (for example an accident or short illness) you should try to disregard it when assessing your fatigue.

Table

0	I do not feel fatigued at all. (No abnormal fatigue, do not need to rest more than usual).
0.5
1	I feel fatigued several times every day but I feel more alert after a rest.
1.5
2	I feel fatigued for most of the day and taking a rest has little or no effect.
2.5
3	I feel fatigued all the time and taking a rest makes no difference.

2. Lack of initiative

Do you find it difficult to start things? Do you experience resistance or a lack of initiative when you have to start something, no matter whether it is a new task or part of your everyday activities?

Table

0	I have no difficulty starting things.
0.5
1	I find it difficult starting things. I’d rather do it some other time.
1.5
2	It takes a great effort to start things. This applies to everyday activities such as getting out of bed, washing myself and eating.
2.5
3	I can’t do the simplest of everyday tasks (eating, getting dressed). I need help with everything.

3. Mental fatigue

Does your brain become fatigued quickly when you have to think hard? For example when reading, watching TV or taking part in a conversation with several people? Do you have to take breaks or change to another activity?

Table

0	My ability for sustained mental effort has not been reduced during the past 4 weeks.
0.5
1	I become fatigued quickly but am able to make the same mental effort as normal.
1.5
2	I become fatigued quickly and often I have to take a break or do something else.
2.5
3	I become fatigued so quickly that I can do nothing or have to abandon everything after a short period (5 minutes).

4. Mental recovery

If you have to take a break, how long do you need to recover after you have worked ‘until you drop’ or are no longer able to concentrate on what you are doing?

Table

0	I need to rest for less than an hour before continuing whatever I am doing.
0.5
1	I need to rest for more than an hour but do not require a night’s sleep.
1.5
2	I need a night’s sleep before I can continue whatever I am doing.
2.5
3	I need several days rest in order to recover.

5. Concentration difficulties

Do you find it difficult to gather your thoughts and concentrate?

Table

0	I can concentrate.
0.5
1	I sometimes lose concentration, for example when reading or watching TV.
1.5
2	I find it so difficult to concentrate that I have problems, for example reading a newspaper or taking part in a conversation with a group of people.
2.5
3	I always have such difficulty concentrating that it is almost impossible to do anything.

6. Memory problems

Do you forget things often, do you need to make notes or do you have to search for things at home or at work?

Table

0	I have no memory problems.
0.5
1	I forget things slightly more often than I should, but I am able to manage by making notes.
1.5
2	My poor memory causes frequent problems (for example forgetting important meetings or things I have to take with me).
2.5
3	I can hardly remember anything at all.

7. Slowness of thinking

Do you feel slow or sluggish when you think about something? Do you feel that it takes an unusually long time to conclude a train of thought or solve a task that requires mental effort?

Table

0	My thoughts are neither slow nor sluggish when it comes to work involving mental effort.
0.5
1	My thoughts are a bit slow one or a few times each day when I have to do something that requires serious mental effort.
1.5
2	My thoughts often feel slow and sluggish, even when carrying out everyday activities, for example a conversation with a person or when reading the newspaper
2.5
3	My thoughts always feel very slow and sluggish.

8. Sensitivity to stress

Do you find it difficult to cope with stress, that is doing several things at the same time while under time pressure?

Table

0	I am able to cope with stress.
0.5
1	I become stressed, but only in demanding situations.
1.5
2	I easily become stressed in less demanding situations.
2.5
3	I become stressed very easily. I feel stressed in unfamiliar or trying situations.

9. Increased tendency to become emotional

Do you find that you cry more easily than others? Do you often burst into tears when, for example, you watch a sad film or talk about sensitive topics? If you recently experienced something unusual (for example an accident or short illness) you should try to disregard it in your assessment.

Table

0	I am not emotional.
0.5
1	I am more emotional than other people but it is something that is natural for me. I start to cry or my eyes fill with tears easily, but only in relation to things that affect me deeply.
1.5
2	My emotions are problematic or embarrassing. I sometimes even start to cry about things that mean nothing to me. I try to avoid certain situations because of this.
2.5
3	My emotions cause me great problems. They disturb my day-to-day relationship with members of my immediate family and make it difficult for me to cope outside the home.

10. Irritability or ‘a short fuse’

Are you unusually short-tempered or irritable about small things?

Table

0	I am not short-tempered or irritable.
0.5
1	I become easily irritated, but it does not last very long.
1.5
2	I become irritated very quickly about small things or things that do not bother other people.
2.5
3	I react with extreme anger or rage, which I find very difficult to control.

11. Sensitivity to light

Are you sensitive to strong light?

Table

0	I have no particularly strong sensitivity to light.
0.5
1	I sometimes experience problems with strong light such as sunlight reflected by snow, water or glass or strong lights at home, but I am able to cope with it, for example by wearing sunglasses. My sensitivity to light does not disturb my everyday life.
1.5
2	I am so sensitive to light that I prefer to carry out my daily activities in dim light. I find it difficult to leave the house without sunglasses.
2.5
3	My sensitivity to light is so strong that I am unable to leave the house without sunglasses. I keep the blinds (or equivalent) drawn at all times.

12. Sensitivity to noise

Are you sensitive to noise?

Table

0	I do not suffer from a particularly strong sensitivity to noise.
0.5
1	I sometimes have difficulty with loud noise (for example music, noise from the TV or radio or sudden unexpected sounds), but I can deal with it easily by turning down the volume. My sensitivity to noise does not disturb my everyday life.
1.5
2	I have a marked over-sensitivity to noise. I have to avoid loud noise or reduce it (for example by means of ear plugs) in order to cope with everyday life.
2.5
3	My sensitivity to noise is so great that I find it difficult to manage at home despite sound insulation.

13. Decreased sleep at night

Do you sleep badly at night? If you are taking sleeping tablets and sleep normally, please place a circle around the ‘0ʹ.

Table

0	I sleep between 6 and 8 hours per night.
0.5
1	I have slight problems falling asleep or my sleep is light, restless or short.
1.5
2	I sleep between 4 and 5 hours per night and wake up frequently during the night without anything disturbing me.
2.5
3	I sleep less than 2–3 hours per night.

14. Increased sleep

Do you sleep long and/or deep? N.B. Please take account of time spent sleeping during the day.

Table

0	I sleep between 6 and 8 hours per night
0.5
1	I sleep long or deep, but no more than 9 hours per day and night, including naps during the day.
1.5
2	I sleep long or deep. At least 10 hours per day and night, including naps.
2.5
3	I sleep long or deep. At least 12 hours per day and night and in addition I need to take a nap during the day.