Vol.
-
No.
- -
2014
Journal of Pain and Symptom Management 1
Brief Methodological Report
Psychometric Properties of the Polish Version of MFI-20 in Cancer Patients Tomasz Buss, MD, PhD, Agnieszka Kruk, MSc, Piotr Wisniewski, MD, Aleksandra Modlinska, MD, PhD, Justyna Janiszewska, MA, PhD, and Monika Lichodziejewska-Niemierko, MD, PhD Department of Palliative Medicine (T.B., A.M., J.J., M.L.-N.); Department of Surgical Nursing (A.K.); and Department of Endocrinology and Internal Medicine (P.W.), Medical University of Gda n sk, Poland
Abstract Purpose. The main aims of this study were to evaluate psychometric properties of the Polish version of the Multidimensional Fatigue Inventory-20 (MFI-20) and to deliver to the clinicians a multidimensional tool for cancerrelated fatigue assessment in Polish language-speaking patients with cancer disease. Methods. After forward-backward translation procedures, the Polish version of MFI-20 was applied to 340 cancer patients. The Polish MFI-20 was appraised in terms of acceptability, reliability, and validity. Internal consistency was assessed by calculating Cronbach’s alpha coefficients. Structure validity was evaluated with confirmatory factor analysis. Results. Translated MFI-20 was well accepted because 90% of subjects fully completed the questionnaire. The overall Cronbach’s alpha coefficient was 0.9 ranging from 0.57 to 0.81. All correlation coefficients between Numeric Rating Scale-fatigue, fatigue-related items from Quality of Life Core-30 questionnaire, and MFI-20 were statistically significant (P < 0.001). Confirmatory factor analysis demonstrated good structure validity and revealed only three dimensions in the Polish version of MFI-20dphysical and mental fatigue as well as reduced motivation. Conclusions. The Polish version of MFI-20 is well accepted by the patients, reliable, and a valid instrument to assess cancer-related fatigue in Polish cancer patients. J Pain Symptom Manage 2014;-:-e-. Ó 2014 U.S. Cancer Pain Relief Committee. Published by Elsevier Inc. All rights reserved. Key Words Fatigue, cancer, Multidimensional Fatigue Inventory, reliability, validity
Address correspondence to: Tomasz Buss, MD, PhD, Department of Palliative Medicine, Medical University of Gda nsk, ul. De˛binki 2 80-211 Gda nsk, Poland. E-mail: [email protected] Ó 2014 U.S. Cancer Pain Relief Committee. Published by Elsevier Inc. All rights reserved.
Accepted for publication: December 12, 2013.
0885-3924/$ - see front matter http://dx.doi.org/10.1016/j.jpainsymman.2013.11.015
2
Buss et al.
Introduction Fatigue is a subjective borderline health problem, where there is no clear distinction between physical and mental aspects. It is a universal experience of varying subjective significance. Its occurrence in a previously healthy patient is considered to be an alarm symptom. Persistent fatigue is a detrimental and unproductive side effect that should be treated. It is prevalent both in cancer patients and noncancer patients with progressive life-threatening diseases. Compared with the fatigue experienced by healthy individuals, cancer-related fatigue (CRF) is less likely to be relieved by rest,1 is more distressing, and differs in daily evolution profiles.2 On the other hand, the European Association for Palliative Care (EAPC) expert group stated that CRF does not seem to be qualitatively different from fatigue in healthy humans.3 Inability to relieve fatigue with rest and interference with function seem to be indicators of the intensity of fatigue rather than criteria for the quality of fatigue. EAPC defines fatigue as a subjective feeling of tiredness, weakness, or lack of energy.3 The negative impact of CRF on quality of life has been documented in several studies.4e6 The wide range of prevalence of CRF from 60% to 99% depends on measurement tools and stages of the disease.3,7,8 Assessment of CRF seems to be key to recognizing and managing this symptom. Despite its prevalence and harmful influence on everyday life, patients are reluctant to report fatigue.9 Other barriers concerning CRF management have been extensively discussed by Borneman et al.10 In clinical practice, it is important to assess for CRF systematically rather than waiting until patients spontaneously complain of the symptom. Therefore, appropriate measuring instruments should be used to detect CRF universally. As with the definition of CRF, various approaches to its assessment have been proposed until now. Unidimensional and multidimensional scales scoring fatigue in cancer patients have been developed. The unidimensional scales cover fatigue as a single symptom, whereas multidimensional questionnaires estimate CRF as a symptom cluster or a clinical syndrome. Most of them have been used and validated in an English-speaking population. In accordance with EAPC approach, cultural issues and language peculiarity can affect CRF
Vol.
-
No.
- -
2014
assessment.3 Therefore, we decided to validate the Polish version of Multidimensional Fatigue Inventory-20 (MFI-20) and verify its properties and validity among Polish cancer patients. Among instruments rating CRF, the MFI-20 has the theoretical advantage of encompassing all aspects of fatigue. It was created in 1995 by Smets et al.11 High internal consistency and validity of the MFI-20 have been documented.12 It has been used in cancer and noncancer patients as well as hospital staff.13 To our knowledge, no instrument dedicated to CRF assessment has been validated in Poland till now.
Patients and Methods Study Design This is a qualitative study.
Study Population The study was conducted between February 2008 and December 2011. Included were patients from oncology outpatient clinics and hospital wards in Gda nsk. The eligibility criteria were as follows: 1) cancer diagnosis, 2) 18 years or older, 3) conscious, 4) seems to be able to understand the pattern of answering to the questions contained in the assessment tools after verbal instructions were given, and 5) have enough strength to complete the questionnaires by themselves. Patients were excluded if there was a history of psychotic symptoms, brain metastases, and significant cognitive impairment. The autoorientation and allo-orientation of the patients have been assessed by asking simple questions concerning the name, present date or the year, and place of staying. Patients who met the inclusion and exclusion criteria were identified by the attending physicians. Written informed consent was a prerequisite for inclusion. Consenting patients were given verbal instructions for the completion of questionnaires. After obtaining the information that the instructions were satisfactory, patients were left alone to complete the set of instruments. The questionnaires were taken back a few hours later or the next day. Incompletely filled out questionnaires were excluded from statistical analysis. Thus, a total of 340 adult cancer patients, comprised of 173 women (51%) and 167 men (49%), were included in the study.
Vol.
-
No.
- -
2014
The Polish Version of MFI-20 in Cancer Patients
Instruments Fatigue was assessed using the Numeric Rating Scale (NRS)-fatigue (0e10), questions concerning fatigue from the Polish validated version of Quality of Life Core-30 (QLQ-C30; version 3.0) and the Polish version of MFI-20. Initially, the English version of the MFI-20 was translated into Polish in accordance with the forward-backward procedures recommended by the European Organization for Research and Treatment of Cancer Quality of Life Group.14 The items were first translated into Polish by two independent translators. Then, a native English speaker backward translated the preliminary Polish MFI-20 version into English. Discrepancies between the original English and the translated Polish version of the MFI-20 were analyzed. Finally, the Polish version of the MFI20 was created. The original version of MFI-20 covers general fatigue (GF), physical fatigue (PF), and mental fatigue (MF) as well as reduced motivation (RM) and reduced activity (RA). The QLQ-C30 questionnaire is a widely used tool with good psychometric properties to measure quality of life in cancer patients. It was constructed by the European Organization for Research and Treatment of Cancer Quality of Life Group15 and translated into many languages including Polish. The Polish version of QLQ-C30 was validated by de Walden-Ga1uszko and Majkowicz.16 To reduce the number of items in the study, only three questions from the Polish version of QLQ-C30, directly connected with fatigue, were used (Question 10dDid you need to rest?, Question 12dHave you felt week?, and Question 18dAre you tired?).
Statistical Analysis The psychometric properties of the Polish version of MFI-20 were assessed in terms of acceptability, reliability, validity, reproducibility, and sensitivity to change. Acceptability (as determined by the percentage of patients who answered all items) was evaluated for the MFI-20 as a whole and for each of the five subscales. Internal consistency was assessed using Cronbach’s alpha coefficient, which ranges from 0 to 1. Results exceeding 0.7 are considered good. The Cronbach’s alpha coefficient was calculated both for the entire Polish MFI20 questionnaire and for each of the subscales separately. Convergent validity, as an indicator
3
of how a newly developed tool relates to an already existing tool, was also estimated. The correlation of the Polish MFI-20 with the NRS fatigue and the fatigue concerning questions from QLQ-C30 was assessed using Spearman rank correlation coefficient. The results of the Polish MFI-20 were also regressed against the NRS-fatigue. There was no heteroscedasticity on inspection of the standardized residuals vs. the fitted values. The BlandAltman plot for the Polish MFI-20 and the NRS-fatigue was prepared. Because the proportional bias was noted, the limits of agreement were adjusted with a regression model. Structure validity was assessed with confirmatory factor analysis (CFA), which is superior to the widely using factor analysis. Several models were evaluated: Model Adwith a single underlying latent factor. This model represents the hypothesis that the Polish MFI-20 is a unidimensional questionnaire. Model B with the five factors proposed in the English version of MFI-20 (GF, PF, MF, RA, RM) with all latent factors allowed to covary. Finally, Model Cdthat is the Model B respecified according to post hoc modification indices. Model fit was assessed with the c2 test, comparative fit index, standardized root mean squared residual, and root mean squared error of approximation (RMSEA). All calculations were performed using STATA 12.0 statistical software (StataCorp LP, College Station, TX).
Results The study enrolled 340 adult cancer patients. The mean age was 60.6 years (SD 12.0). The most common diagnoses included lung cancer (51%), breast cancer (10%), and colon cancer (4%). The remaining patients had different far advanced cancers.
Transcultural Translation Process Translators did not have a problem with the translation of the items. Literally, fatigue can be determined in Polish language as zme˛ czenie. However, interviews with patients revealed that cancer patients did not always describe their condition with the word zme˛ czenie. Some of them used the term osłabienie (in Englishdweakness). There is little difference in the meaning of these words in Polish. It has not been established so far which term better
4
Buss et al.
Vol.
-
No.
- -
2014
Table 1 Item-Subscale Correlation Coefficients and Cronbach’s Alpha for Polish MFI-20 Item 1 5 12 16 3 6 10 17 2 8 14 20 7 11 13 19 4 9 15 18
GF
RA
PF
MF
RM
Cronbach’s Alpha
0.81 L0.81 0.78 L0.74 0.64 0.10 0.25 0.56 0.48 0.58 0.64 0.64 0.38 0.41 0.46 0.26 0.44 0.51 0.33 0.44
0.55 0.41 0.38 0.53 0.69 0.57 L0.71 L0.68 0.38 0.56 0.56 0.48 0.21 0.18 0.29 0.23 0.28 0.40 0.24 0.39
0.64 0.59 0.50 0.66 0.63 0.11 0.37 0.57 L0.71 0.81 L0.77 0.81 0.34 0.36 0.41 0.27 0.42 0.49 0.34 0.41
0.40 0.36 0.45 0.32 0.40 0.00 0.18 0.31 0.27 0.39 0.46 0.32 0.71 0.76 L0.77 L0.74 0.29 0.53 0.31 0.43
0.47 0.52 0.46 0.46 0.53 0.03 0.22 0.45 0.38 0.50 0.52 0.42 0.40 0.43 0.45 0.34 0.71 L0.66 0.71 L0.73
0.80
Cronbach’s Alpha if Item Deleted 0.73 0.73 0.76 0.77 0.49 0.61 0.46 0.49 0.78 0.72 0.74 0.71 0.67 0.62 0.66 0.75 0.59 0.61 0.58 0.56
0.59
0.79
0.73
0.65
MFI-20 ¼ Multidimensional Fatigue Inventory-20; GF ¼ general fatigue; RA ¼ reduced activity; PF ¼ physical fatigue; MF ¼ mental fatigue; RM ¼ reduced motivation. Bold text indicates Pearson item-scale corrected correlation coefficient.
describes the condition of cancer patients. The word zme˛ czenie has been chosen because of translator’s indications, the existence of this term in the Polish medical literature, and the conviction of the authors, based on clinical experience, that this word describes well cancer patient’s symptoms.
Acceptability The Polish version was well accepted because 99% of subjects fully completed the questionnaire with no omitted items. The number of missing items was low (less than 1%).
Reliability The reliability of the Polish version of the MFI-20 is high, with the overall Cronbach’s alpha coefficient 0.9. The correlations between MFI-20 subscales and appropriate items are presented in Table 1. The coefficients ranged from 0.57 to 0.81 (absolute values) with the lowest values for the Items 6 and 19. The Cronbach’s alpha for each of the five subscales of MFI-20 ranged from 0.59 to 0.80. The removal of the Item 6 or 19 would result in an increase of alpha coefficient of the respective subscale.
Convergent Validity The correlation coefficients between the Polish MFI-20 (overall and its subscales) and
NRS-fatigue and QLQ-C30 fatigue-related items are shown in Table 2. The ‘‘GF’’ subscale had the highest and ‘‘RA’’ subscale had the lowest correlations with the fatigue scales mentioned previously. The absolute values for the overall score ranged from 0.42 to 0.57, when subscales’ correlations ranged from 0.21 to 0.57. All of them were significant with P < 0.001. The highest correlations were observed for the GF subscale, whereas the lowest correlations were observed for the RA subscale. The Bland-Altman analysis revealed a good conformity between the Polish MFI-20 and the NRS-fatigue (Fig. 1). The adjusted limits of conformity between the methods
Table 2 The Correlation Coefficients Between the Polish MFI (Overall and Its Subscales) and NRS-Fatigue and Selected QLQ-C30 Fatigue-Related Items Polish MFI
NRS-Fatigue
QLQ-C30
Overall MFI GF PF MF RA RM
0.57 0.57 0.49 0.41 0.36 0.40
0.53 0.55 0.37 0.38 0.40 0.36
MFI ¼ Multidimensional Fatigue Inventory; NRS ¼ Numeric Rating Scale; QLQ-C30 ¼ Quality of Life Core 30 questionnaire; GF ¼ general fatigue; PF ¼ physical fatigue; MF ¼ mental fatigue; RA ¼ reduced activity; RM ¼ reduced motivation. All correlation coefficients were significant with P < 0.001.
Vol.
-
No.
- -
2014
The Polish Version of MFI-20 in Cancer Patients
Fig. 1. Bland-Altman plot for Numeric Rating Scale (NRS)-fatigue vs. Polish Multidimensional Fatigue Inventory-20 (MFI-20). The shaded area indicates the adjusted limits of conformity.
were narrow, and 3.57% of measurements were outside the limits. Measurements showed consistent variability around the average.
Structure Validity The structure validity of the Polish translation of the MFI-20 questionnaire was examined with the use of CFA. The goodness-of-fit statistics for all examined models are presented in Table 3. The values for Models A and B indicated a poor fit; these models were therefore rejected. Model C with five latent and mutually correlated factors proposed in the original version of MFI20 resulted in an acceptable fit, and this model was chosen for in-depth analysis. Parameter estimates for Model C are reported in Table 4. The values of standardized factor loadings (equivalent to factor-item correlation coefficients) indicate a high degree of correlation between the five factors and the corresponding items of the Polish MFI-20 questionnaire. For example, for the GF factor, the coefficients were between 0.65 and 0.75 (absolute values). This suggests a high degree of correlation
5
between this factor and Items 1, 5, 12, and 16. The complete listing of factor-item correlation coefficients is shown in Table 4. The lowest observed correlations were for Item 6 (0.14; RA factor), Item 10 (0.37; RA factor), Item 19 (0.47; MF factor), and Item 4 (0.52; RM factor). The between-factor correlation coefficients (for standardized covariance coefficients, see Table 4) ranged from 0.55 to 1.0 (absolute values). High values of mutual correlations for RA, PF, GF factors indicate their poor discriminant validity and suggest that these factors may not be clearly distinct. Correlation coefficients between RM, MF, and the other factors were significantly lower. Modification indices were calculated for Model B. The omitted paths with the highest index values were EItem 7 4 EItem 11 (c2 ¼ 37.9), EItem 2 6 4 EItem 10 (c ¼ 29.0), EItem 4 4 EItem 15 2 (c ¼ 27.5) and PF / Item 16 (c2 ¼ 27.7). On the basis of these indices, Model B was modified into Model C. Because of the almost identical meanings of the Polish translation of Items 7 and 11 as well as Items 6 and 10, Items 6 and 11 have been removed from the model. In addition, most items comprising GF, PF, and RA subscales were included in one combined factor. This new model demonstrated superior fit compared with the previous models (Table 3).
Discussion To our knowledge, there is a lack of validated multidimensional CRF assessment tools in Poland. This study established psychometric properties of the Polish version of the MFI-20 questionnaire based on its application in a large sample of cancer patients. Fatigue in cancer patients has so far been mostly investigated in Western societies. Both cultural distinctions and standard of living as
Table 3 Selected Fit Indexes for CFA Models Polish MFI Chi-square CFI SRMR RMSEA 90% CI for RMSEA
Model A
Model B
Model C
747 (df ¼ 170) 0.781 0.073 0.101 0.093e0.108
554 (df ¼ 160) 0.851 0.062 0.086 0.078e0.094
410 (df ¼ 132) 0.882 0.055 0.079 0.063e0.080
CFA ¼ confirmatory factor analysis; df ¼ degrees of freedom; CFI ¼ comparative fix index; SRMR ¼ standardized root mean squared residual; RMSEA ¼ root mean squared error of approximation; 90% CI ¼ 90% confidence interval. Description of the models in the text.
6
Buss et al.
Table 4 Maximum Likelihood Parameter Estimates for Model B Parameter
Unstandardized
Factor loadings GF / Item 1 1a GF / Item 5 0.91 GF / Item 12 0.88 GF / Item 16 0.95 RA/ Item 3 1a RA / Item 6 0.19b RA / Item 10 0.49 RA / Item 17 0.86 PF / Item 2 1a PF / Item 8 1.41 PF / Item 14 1.33 PF / Item 20 1.24 MF / Item 7 1a MF / Item 11 1.04 MF / Item 13 1.11 MF / Item 19 0.87 RM / Item 4 1a RM / Item 9 1.12 RM / Item 15 0.88 RM / Item 18 1.12 Measurement error variances 0.95 EItem 1 1.15 EItem 5 1.36 EItem 12 1.15 EItem 16 1.15 EItem 3 2.33 EItem 6 1.93 EItem 10 1.26 EItem 17 1.50 EItem 2 1.22 EItem 8 1.04 EItem 14 1.01 EItem 20 0.77 EItem 7 0.53 EItem 11 1.17 EItem 13 1.99 EItem 19 1.73 EItem 4 1.18 EItem 9 1.78 EItem 15 1.35 EItem 18 Factor variances and covariance GF 1.26 RA 1.25 PF 0.72 MF 0.76 RM 0.65 GF 4 RA 1.22 GF 4 PF 0.92 GF 4 MF 0.61 GF 4 RM 0.75 RA 4 PF 0.96 RA 4 MF 0.54 RA 4 RM 0.82 PF 4 MF 0.44 PF 4 RM 0.57 MF 4 RM 0.53
SE
Standardized
0.15 0.13 0.15
0.75 0.69 0.65 0.71 0.72 0.14 0.37 0.65 0.57 0.74 0.74 0.73 0.71 0.78 0.67 0.47 0.52 0.63 0.47 0.61
0.09 0.10 0.11 0.10 0.10 0.18 0.15 0.11 0.12 0.11 0.09 0.09 0.08 0.07 0.11 0.16 0.14 0.10 0.14 0.12
0.43 0.52 0.58 0.50 0.47 0.98 0.86 0.57 0.67 0.45 0.44 0.47 0.50 0.39 0.55 0.77 0.72 0.59 0.77 0.62
0.16 0.17 0.13 0.11 0.14 0.14 0.12 0.09 0.11 0.12 0.09 0.12 0.07 0.09 0.08
1 1 1 1 1 0.97 0.96 0.62 0.83 1.00 0.55 0.90 0.59 0.83 0.76
0.07 0.08 0.08 0.08 0.08 0.07 0.14 0.13 0.12 0.08 0.12 0.12
GF ¼ general fatigue; RA ¼ reduced activity; PF ¼ physical fatigue; MF ¼ mental fatigue; RM ¼ reduced motivation. Standardized estimates for measurement errors are proportions of unexplained variance. a Not tested for statistical significance. b P < 0.001 for all other unstandardized estimates except P ¼ 0.015.
Vol.
-
No.
- -
2014
well as gratification of economical concerns may significantly alter the perception and interpretation of fatigue. Our observations show that Polish patients experiencing fatigue or tiredness in the course of cancer generally pay attention to the physical aspects of fatigue such as deterioration in muscle strength to perform normal activities. They seldom report mental aspects of fatigue like RM or concentration and lower mood. In general, it can be said that health deterioration in socially and economically less developed countries is perceived as a physical rather than mental problem. Mental aspects of the disease are ignored more often and taken into consideration unwillingly. From the practical point of view, the management of CRF in those patients depends on whether fatigue or inability to perform daily routines is associated with lack of muscle strength or with the lack of motivation to perform any tasks. This approach corresponds to that presented by Glaus et al.,1 who describes fatigue as a subjective feeling affecting the body (physical), emotions (affective), and mental functioning (mental). This distinction may allow to guide appropriate CRF management. When muscle impairment is observed, factors affecting neuromuscular conduction such as electrolyte imbalances should be corrected first. Exercise may also be encouraged. Benefits of exercise in patients undergoing cancer treatment,17 cancer survivors,18e20 and hospice cancer patients21 were documented. In dealing with the mental or affective aspects of CRF, psychostimulants22 or psychoeducational therapies as well as supportive expressive therapies20,23,24 can be considered. In this study, on the basis of existing Western experience with CRF measurement, we decided to validate MFI-20 in the Polish clinical setting. This questionnaire has good psychometric properties and assesses multiple dimensions of fatigue. The Polish version of MFI-20 was well accepted by patients. The number of missing items was less than 1%. However, the drawback of the study is the lack of information on the number of patients who were disqualified. It was a small number of patients who, despite the lack of significant cognitive impairment, were unable to mark their fatigue level on NRS scale or did not understand the response pattern contained in the MFI-20. In turn, based on the assumption
Vol.
-
No.
- -
2014
The Polish Version of MFI-20 in Cancer Patients
7
Table 5 Items of the New Dimensions of the Polish Version of MFI-20 Item Number Subscales in the in the English Polish Version Version of of MFI-20 MFI-20 PF
MF
1 2 3 5 8 10 12 14 16 17 20 7 13
RM
Removed items
19 4 9 15 18 6 11
Items Translated Into Polish Czuje˛, z_ e jestem w dobrej formie Czuje˛, z_ e fizycznie moge˛ zrobic niewiele Czuje˛ sie˛ bardzo aktywny/a Czuje˛ sie˛ zme˛czony/a Fizycznie potrafie˛ du_zo wzia˛c na siebie Mysle˛, z_ e to co robie˛ w cia˛gu dnia, to bardzo ma1o Jestem wypocze˛ty/a Czuje˛, z_ e jestem w z1ej formie fizycznej qatwo sie˛ me˛cze˛ Niewiele udaje mi sie˛ zrobic Czuje˛, z_ e jestem w swietnej formie fizycznej Kiedy cos robie˛, moge˛ sie˛ na tym skoncentrowac Skoncentrowanie sie˛ na czyms kosztuje mnie du_zo wysi1ku Moje mysli cze˛sto b1a˛dza˛ Mam ochote˛ na wszystkie rodzaje przyjemnosci Przera_za mnie, kiedy musze cos zrobic Mam du_zo plan ow Nie chce mi sie˛ nic robic Mysle˛, z_ e to co robie˛ w cia˛gu dnia, to du_zo Moge˛ sie˛ skoncentrowac
English Items I feel fit Physically I feel only able to do a little I feel very active I feel tired Physically I can take on a lot I think I do very little in a day I am rested Physically I feel I am in a bad condition I tire easily I get little done Physically I feel I am in an excellent condition When I am doing something, I can keep my thoughts on it It takes a lot of effort to concentrate on things My thoughts easily wander I feel like doing all sorts of nice things I I I I I
dread having to do things have a lot of plans do not feel like doing anything think I do a lot in a day can concentrate well
MFI-20 ¼ Multidimensional Fatigue Inventory-20; PF ¼ physical fatigue; MF ¼ mental fatigue; RM ¼ reduced motivation.
that these patients were not able to fill most of the questionnaires, they were not taken into account. Statistical analysis revealed that Polish MFI-20 has both a good reliability with Cronbach’s alpha 0.9 and convergent validity. Structure validity assessed with CFA revealed only three subscales. GF, PF, and RA could not be distinguished in the Polish version of MFI. Similar findings were observed in the French,25 Chinese,26 and Brazilian27 studies where physical and general aspect of fatigue have likewise not been separated. GF is probably too enigmatic to the Polish patients and their doctors. As mentioned previously, our patients’ attitude to fatigue warrants inclusion of GF, PF, and RA into one subscale termed ‘‘physical fatigue.’’ The MF and RM subscales remained similar to the original version. The proposed breakdown of MFI-20 subscales in the Polish version is shown in Table 5. In conclusion, the Polish version of MFI-20 was well accepted by patients. Some dissimilarities were found between the structure of the Polish version of MFI-20 and the original one. Thus, three dimensions of fatigue were proposed in the Polish version of MFI-20. Because of the high number of questions in
the ‘‘PF’’ section of the Polish MFI-20, development of a shorter version of the MFI-20 in Polish can be considered in the future. However, our findings show the usefulness of the Polish version of MFI-20 in the assessment of fatigue in Polish-speaking cancer patients. This instrument can be applied in clinical settings and still takes into account the multidimensionality of fatigue. We hope that the implementation of the Polish version of MFI-20 will enable clinicians both to not overlook the problem and to assess CRF objectively and easily. It has the potential to improve both the detection of CRF and patient care in our country. We hope that this study will inspire further research on the topic of CRF in Poland.
Disclosures The authors declare that they have no conflict of interest.
References 1. Glaus A, Crow R, Hammond S. A qualitative study to explore the concept of fatigue/tiredness
8
Buss et al.
in cancer patients and in healthy individuals. Support Care Cancer 1996;4:82e96. 2. Glaus A. Assessment of fatigue in cancer and non-cancer patients and in healthy individuals. Support Care Cancer 1993;1:305e315. 3. Radbruch L, Strasser F, Elsner F, et al. Fatigue in palliative care patientsdan EAPC approach. Palliat Med 2008;22:13e32. 4. Ferreira KA, Kimura M, Teixeira MJ, et al. Impact of cancer-related symptom synergisms on health-related quality of life and performance status. J Pain Symptom Manage 2008;35:604e616. 5. van den Beuken-van Everdingen MH, de Rijke JM, Kessels AG, Schouten HC, van Kleef M, Patijn J. Quality of life and non-pain symptoms in patients with cancer. J Pain Symptom Manage 2009;38:216e233. 6. Curt GA, Breitbart W, Cella D, et al. Impact of cancer-related fatigue on the lives of patients: new findings from the fatigue coalition. Oncologist 2000;5:353e360. 7. Walsh D, Donnelly S, Rybicki L. The symptoms of advanced cancer: relationship to age, gender, and performance status in 1000 patients. Support Care Cancer 2000;8:175e179. 8. Lawrence DP, Kupelnick B, Miller K, Devine D, Lau J. Evidence report on the occurrence, assessment, and treatment of fatigue in cancer patients. J Natl Cancer Inst Monogr 2004;32:40e50. 9. Stone P, Richardson A, Ream E, Smith AG, Kerr DJ, Kearney N. Cancer-related fatigue: inevitable, unimportant and untreatable? Results of a multi-centre patient survey. Cancer Fatigue Forum. Ann Oncol 2000;11:971e975. 10. Borneman T, Koczywas M, Sun VC, Piper BF, Uman G, Ferrell B. Reducing patient barriers to pain and fatigue management. J Pain Symptom Manage 2010;39:486e501. 11. Smets EM, Garssen B, Bonke B, De Haes JC. The Multidimensional Fatigue Inventory (MFI) psychometric qualities of an instrument to assess fatigue. J Psychosom Res 1995;39:315e325. 12. Smets EM, Garssen B, Cull A, de Haes JC. Application of the multidimensional fatigue inventory (MFI-20) in cancer patients receiving radiotherapy. Br J Cancer 1996;73:241e245. 13. Hagelin CL, Wengstr€ om Y, Runesdotter S, F€ urst CJ. The psychometric properties of the Swedish Multidimensional Fatigue Inventory MFI-20 in four different populations. Acta Oncol 2007;46:97e104. 14. Dewolf L, Koller M, Velikova G, Johnson C, Scott N, Bottomley A. on behalf of the EORTC Quality of Life Group. EORTC Quality of Life Group translation procedure March 2009, 3rd ed. Copyright EORTC Brussels. Available at: http://groups.eortc.be/qol/downloads/translation_ manual_2009.pdf. Accessed June 30, 2011.
Vol.
-
No.
- -
2014
15. Aaronson NK, Ahmedzai S, Bergman B, et al. The European Organization for Research and Treatment of Cancer QLQ-C30: a quality-of-life instrument for use in international clinical trials in oncology. J Natl Cancer Inst 1993;85:365e376. 16. de Walden-Ga1uszko K, Majkowicz M. The assessment of the quality of palliative home care in the theory and practice. (Ocena jakosci opieki paliatywnej w teorii i praktyce). Gda nsk: Medical University of Gda nsk; 2000. 17. Velthuis MJ, Agasi-Idenburg SC, Aufdemkampe G, Wittink HM. The effect of physical exercise on cancer-related fatigue during cancer treatment: a meta-analysis of randomised controlled trials. Clin Oncol (R Coll Radiol) 2010;22:208e221. 18. McNeely ML, Campbell KL, Rowe BH, Klassen TP, Mackey JR, Courneya KS. Effects of exercise on breast cancer patients and survivors: a systematic review and meta-analysis. CMAJ 2006;175:34e41. 19. Oldervoll LM, Kaasa S, Knobel H, Loge JH. Exercise reduces fatigue in chronic fatigued Hodgkins disease survivorsdresults from a pilot study. Eur J Cancer 2003;39:57e63. 20. Duijts SF, Faber MM, Oldenburg HS, van Beurden M, Aaronson NK. Effectiveness of behavioral techniques and physical exercise on psychosocial functioning and health-related quality of life in breast cancer patients and survivorsda meta-analysis. Psychooncology 2011;20:115e126. 21. Buss T, de Walden-Ga1uszko K, Modli nska A, Osowicka M, Lichodziejewska-Niemierko M, Janiszewska J. Kinesitherapy alleviates fatigue in terminal hospice cancer patientsdan experimental, controlled study. Support Care Cancer 2010;18:743e749. 22. Minton O, Richardson A, Sharpe M, Hotopf M, Stone PC. Psychostimulants for the management of cancer-related fatigue: a systematic review and metaanalysis. J Pain Symptom Manage 2011;41:761e767. 23. Boesen EH, Ross L, Frederiksen K, et al. Psychoeducational intervention for patients with cutaneous malignant melanoma: a replication study. J Clin Oncol 2005;23:1270e1277. 24. Dolbeault S, Cayrou S, Br edart A, et al. The effectiveness of a psycho-educational group after earlystage breast cancer treatment: results of a randomized French study. Psychooncology 2009;18:647e656. 25. Gentile S, Delarozi ere JC, Favre F, Sambuc R, San Marco JL. Validation of the French ’multidimensional fatigue inventory’ (MFI 20). Eur J Cancer Care (Engl) 2003;12:58e64. 26. Tian J, Hong JS. Validation of the Chinese version of Multidimensional Fatigue Inventory-20 in Chinese patients with cancer. Support Care Cancer 2012;20:2379e2383. 27. Baptista RL, Biasoli I, Scheliga A, et al. Psychometric properties of the multidimensional fatigue inventory in Brazilian Hodgkin’s lymphoma survivors. J Pain Symptom Manage 2012;44:908e915.
-
No.
- -
2014
Journal of Pain and Symptom Management 1
Brief Methodological Report
Psychometric Properties of the Polish Version of MFI-20 in Cancer Patients Tomasz Buss, MD, PhD, Agnieszka Kruk, MSc, Piotr Wisniewski, MD, Aleksandra Modlinska, MD, PhD, Justyna Janiszewska, MA, PhD, and Monika Lichodziejewska-Niemierko, MD, PhD Department of Palliative Medicine (T.B., A.M., J.J., M.L.-N.); Department of Surgical Nursing (A.K.); and Department of Endocrinology and Internal Medicine (P.W.), Medical University of Gda n sk, Poland
Abstract Purpose. The main aims of this study were to evaluate psychometric properties of the Polish version of the Multidimensional Fatigue Inventory-20 (MFI-20) and to deliver to the clinicians a multidimensional tool for cancerrelated fatigue assessment in Polish language-speaking patients with cancer disease. Methods. After forward-backward translation procedures, the Polish version of MFI-20 was applied to 340 cancer patients. The Polish MFI-20 was appraised in terms of acceptability, reliability, and validity. Internal consistency was assessed by calculating Cronbach’s alpha coefficients. Structure validity was evaluated with confirmatory factor analysis. Results. Translated MFI-20 was well accepted because 90% of subjects fully completed the questionnaire. The overall Cronbach’s alpha coefficient was 0.9 ranging from 0.57 to 0.81. All correlation coefficients between Numeric Rating Scale-fatigue, fatigue-related items from Quality of Life Core-30 questionnaire, and MFI-20 were statistically significant (P < 0.001). Confirmatory factor analysis demonstrated good structure validity and revealed only three dimensions in the Polish version of MFI-20dphysical and mental fatigue as well as reduced motivation. Conclusions. The Polish version of MFI-20 is well accepted by the patients, reliable, and a valid instrument to assess cancer-related fatigue in Polish cancer patients. J Pain Symptom Manage 2014;-:-e-. Ó 2014 U.S. Cancer Pain Relief Committee. Published by Elsevier Inc. All rights reserved. Key Words Fatigue, cancer, Multidimensional Fatigue Inventory, reliability, validity
Address correspondence to: Tomasz Buss, MD, PhD, Department of Palliative Medicine, Medical University of Gda nsk, ul. De˛binki 2 80-211 Gda nsk, Poland. E-mail: [email protected] Ó 2014 U.S. Cancer Pain Relief Committee. Published by Elsevier Inc. All rights reserved.
Accepted for publication: December 12, 2013.
0885-3924/$ - see front matter http://dx.doi.org/10.1016/j.jpainsymman.2013.11.015
2
Buss et al.
Introduction Fatigue is a subjective borderline health problem, where there is no clear distinction between physical and mental aspects. It is a universal experience of varying subjective significance. Its occurrence in a previously healthy patient is considered to be an alarm symptom. Persistent fatigue is a detrimental and unproductive side effect that should be treated. It is prevalent both in cancer patients and noncancer patients with progressive life-threatening diseases. Compared with the fatigue experienced by healthy individuals, cancer-related fatigue (CRF) is less likely to be relieved by rest,1 is more distressing, and differs in daily evolution profiles.2 On the other hand, the European Association for Palliative Care (EAPC) expert group stated that CRF does not seem to be qualitatively different from fatigue in healthy humans.3 Inability to relieve fatigue with rest and interference with function seem to be indicators of the intensity of fatigue rather than criteria for the quality of fatigue. EAPC defines fatigue as a subjective feeling of tiredness, weakness, or lack of energy.3 The negative impact of CRF on quality of life has been documented in several studies.4e6 The wide range of prevalence of CRF from 60% to 99% depends on measurement tools and stages of the disease.3,7,8 Assessment of CRF seems to be key to recognizing and managing this symptom. Despite its prevalence and harmful influence on everyday life, patients are reluctant to report fatigue.9 Other barriers concerning CRF management have been extensively discussed by Borneman et al.10 In clinical practice, it is important to assess for CRF systematically rather than waiting until patients spontaneously complain of the symptom. Therefore, appropriate measuring instruments should be used to detect CRF universally. As with the definition of CRF, various approaches to its assessment have been proposed until now. Unidimensional and multidimensional scales scoring fatigue in cancer patients have been developed. The unidimensional scales cover fatigue as a single symptom, whereas multidimensional questionnaires estimate CRF as a symptom cluster or a clinical syndrome. Most of them have been used and validated in an English-speaking population. In accordance with EAPC approach, cultural issues and language peculiarity can affect CRF
Vol.
-
No.
- -
2014
assessment.3 Therefore, we decided to validate the Polish version of Multidimensional Fatigue Inventory-20 (MFI-20) and verify its properties and validity among Polish cancer patients. Among instruments rating CRF, the MFI-20 has the theoretical advantage of encompassing all aspects of fatigue. It was created in 1995 by Smets et al.11 High internal consistency and validity of the MFI-20 have been documented.12 It has been used in cancer and noncancer patients as well as hospital staff.13 To our knowledge, no instrument dedicated to CRF assessment has been validated in Poland till now.
Patients and Methods Study Design This is a qualitative study.
Study Population The study was conducted between February 2008 and December 2011. Included were patients from oncology outpatient clinics and hospital wards in Gda nsk. The eligibility criteria were as follows: 1) cancer diagnosis, 2) 18 years or older, 3) conscious, 4) seems to be able to understand the pattern of answering to the questions contained in the assessment tools after verbal instructions were given, and 5) have enough strength to complete the questionnaires by themselves. Patients were excluded if there was a history of psychotic symptoms, brain metastases, and significant cognitive impairment. The autoorientation and allo-orientation of the patients have been assessed by asking simple questions concerning the name, present date or the year, and place of staying. Patients who met the inclusion and exclusion criteria were identified by the attending physicians. Written informed consent was a prerequisite for inclusion. Consenting patients were given verbal instructions for the completion of questionnaires. After obtaining the information that the instructions were satisfactory, patients were left alone to complete the set of instruments. The questionnaires were taken back a few hours later or the next day. Incompletely filled out questionnaires were excluded from statistical analysis. Thus, a total of 340 adult cancer patients, comprised of 173 women (51%) and 167 men (49%), were included in the study.
Vol.
-
No.
- -
2014
The Polish Version of MFI-20 in Cancer Patients
Instruments Fatigue was assessed using the Numeric Rating Scale (NRS)-fatigue (0e10), questions concerning fatigue from the Polish validated version of Quality of Life Core-30 (QLQ-C30; version 3.0) and the Polish version of MFI-20. Initially, the English version of the MFI-20 was translated into Polish in accordance with the forward-backward procedures recommended by the European Organization for Research and Treatment of Cancer Quality of Life Group.14 The items were first translated into Polish by two independent translators. Then, a native English speaker backward translated the preliminary Polish MFI-20 version into English. Discrepancies between the original English and the translated Polish version of the MFI-20 were analyzed. Finally, the Polish version of the MFI20 was created. The original version of MFI-20 covers general fatigue (GF), physical fatigue (PF), and mental fatigue (MF) as well as reduced motivation (RM) and reduced activity (RA). The QLQ-C30 questionnaire is a widely used tool with good psychometric properties to measure quality of life in cancer patients. It was constructed by the European Organization for Research and Treatment of Cancer Quality of Life Group15 and translated into many languages including Polish. The Polish version of QLQ-C30 was validated by de Walden-Ga1uszko and Majkowicz.16 To reduce the number of items in the study, only three questions from the Polish version of QLQ-C30, directly connected with fatigue, were used (Question 10dDid you need to rest?, Question 12dHave you felt week?, and Question 18dAre you tired?).
Statistical Analysis The psychometric properties of the Polish version of MFI-20 were assessed in terms of acceptability, reliability, validity, reproducibility, and sensitivity to change. Acceptability (as determined by the percentage of patients who answered all items) was evaluated for the MFI-20 as a whole and for each of the five subscales. Internal consistency was assessed using Cronbach’s alpha coefficient, which ranges from 0 to 1. Results exceeding 0.7 are considered good. The Cronbach’s alpha coefficient was calculated both for the entire Polish MFI20 questionnaire and for each of the subscales separately. Convergent validity, as an indicator
3
of how a newly developed tool relates to an already existing tool, was also estimated. The correlation of the Polish MFI-20 with the NRS fatigue and the fatigue concerning questions from QLQ-C30 was assessed using Spearman rank correlation coefficient. The results of the Polish MFI-20 were also regressed against the NRS-fatigue. There was no heteroscedasticity on inspection of the standardized residuals vs. the fitted values. The BlandAltman plot for the Polish MFI-20 and the NRS-fatigue was prepared. Because the proportional bias was noted, the limits of agreement were adjusted with a regression model. Structure validity was assessed with confirmatory factor analysis (CFA), which is superior to the widely using factor analysis. Several models were evaluated: Model Adwith a single underlying latent factor. This model represents the hypothesis that the Polish MFI-20 is a unidimensional questionnaire. Model B with the five factors proposed in the English version of MFI-20 (GF, PF, MF, RA, RM) with all latent factors allowed to covary. Finally, Model Cdthat is the Model B respecified according to post hoc modification indices. Model fit was assessed with the c2 test, comparative fit index, standardized root mean squared residual, and root mean squared error of approximation (RMSEA). All calculations were performed using STATA 12.0 statistical software (StataCorp LP, College Station, TX).
Results The study enrolled 340 adult cancer patients. The mean age was 60.6 years (SD 12.0). The most common diagnoses included lung cancer (51%), breast cancer (10%), and colon cancer (4%). The remaining patients had different far advanced cancers.
Transcultural Translation Process Translators did not have a problem with the translation of the items. Literally, fatigue can be determined in Polish language as zme˛ czenie. However, interviews with patients revealed that cancer patients did not always describe their condition with the word zme˛ czenie. Some of them used the term osłabienie (in Englishdweakness). There is little difference in the meaning of these words in Polish. It has not been established so far which term better
4
Buss et al.
Vol.
-
No.
- -
2014
Table 1 Item-Subscale Correlation Coefficients and Cronbach’s Alpha for Polish MFI-20 Item 1 5 12 16 3 6 10 17 2 8 14 20 7 11 13 19 4 9 15 18
GF
RA
PF
MF
RM
Cronbach’s Alpha
0.81 L0.81 0.78 L0.74 0.64 0.10 0.25 0.56 0.48 0.58 0.64 0.64 0.38 0.41 0.46 0.26 0.44 0.51 0.33 0.44
0.55 0.41 0.38 0.53 0.69 0.57 L0.71 L0.68 0.38 0.56 0.56 0.48 0.21 0.18 0.29 0.23 0.28 0.40 0.24 0.39
0.64 0.59 0.50 0.66 0.63 0.11 0.37 0.57 L0.71 0.81 L0.77 0.81 0.34 0.36 0.41 0.27 0.42 0.49 0.34 0.41
0.40 0.36 0.45 0.32 0.40 0.00 0.18 0.31 0.27 0.39 0.46 0.32 0.71 0.76 L0.77 L0.74 0.29 0.53 0.31 0.43
0.47 0.52 0.46 0.46 0.53 0.03 0.22 0.45 0.38 0.50 0.52 0.42 0.40 0.43 0.45 0.34 0.71 L0.66 0.71 L0.73
0.80
Cronbach’s Alpha if Item Deleted 0.73 0.73 0.76 0.77 0.49 0.61 0.46 0.49 0.78 0.72 0.74 0.71 0.67 0.62 0.66 0.75 0.59 0.61 0.58 0.56
0.59
0.79
0.73
0.65
MFI-20 ¼ Multidimensional Fatigue Inventory-20; GF ¼ general fatigue; RA ¼ reduced activity; PF ¼ physical fatigue; MF ¼ mental fatigue; RM ¼ reduced motivation. Bold text indicates Pearson item-scale corrected correlation coefficient.
describes the condition of cancer patients. The word zme˛ czenie has been chosen because of translator’s indications, the existence of this term in the Polish medical literature, and the conviction of the authors, based on clinical experience, that this word describes well cancer patient’s symptoms.
Acceptability The Polish version was well accepted because 99% of subjects fully completed the questionnaire with no omitted items. The number of missing items was low (less than 1%).
Reliability The reliability of the Polish version of the MFI-20 is high, with the overall Cronbach’s alpha coefficient 0.9. The correlations between MFI-20 subscales and appropriate items are presented in Table 1. The coefficients ranged from 0.57 to 0.81 (absolute values) with the lowest values for the Items 6 and 19. The Cronbach’s alpha for each of the five subscales of MFI-20 ranged from 0.59 to 0.80. The removal of the Item 6 or 19 would result in an increase of alpha coefficient of the respective subscale.
Convergent Validity The correlation coefficients between the Polish MFI-20 (overall and its subscales) and
NRS-fatigue and QLQ-C30 fatigue-related items are shown in Table 2. The ‘‘GF’’ subscale had the highest and ‘‘RA’’ subscale had the lowest correlations with the fatigue scales mentioned previously. The absolute values for the overall score ranged from 0.42 to 0.57, when subscales’ correlations ranged from 0.21 to 0.57. All of them were significant with P < 0.001. The highest correlations were observed for the GF subscale, whereas the lowest correlations were observed for the RA subscale. The Bland-Altman analysis revealed a good conformity between the Polish MFI-20 and the NRS-fatigue (Fig. 1). The adjusted limits of conformity between the methods
Table 2 The Correlation Coefficients Between the Polish MFI (Overall and Its Subscales) and NRS-Fatigue and Selected QLQ-C30 Fatigue-Related Items Polish MFI
NRS-Fatigue
QLQ-C30
Overall MFI GF PF MF RA RM
0.57 0.57 0.49 0.41 0.36 0.40
0.53 0.55 0.37 0.38 0.40 0.36
MFI ¼ Multidimensional Fatigue Inventory; NRS ¼ Numeric Rating Scale; QLQ-C30 ¼ Quality of Life Core 30 questionnaire; GF ¼ general fatigue; PF ¼ physical fatigue; MF ¼ mental fatigue; RA ¼ reduced activity; RM ¼ reduced motivation. All correlation coefficients were significant with P < 0.001.
Vol.
-
No.
- -
2014
The Polish Version of MFI-20 in Cancer Patients
Fig. 1. Bland-Altman plot for Numeric Rating Scale (NRS)-fatigue vs. Polish Multidimensional Fatigue Inventory-20 (MFI-20). The shaded area indicates the adjusted limits of conformity.
were narrow, and 3.57% of measurements were outside the limits. Measurements showed consistent variability around the average.
Structure Validity The structure validity of the Polish translation of the MFI-20 questionnaire was examined with the use of CFA. The goodness-of-fit statistics for all examined models are presented in Table 3. The values for Models A and B indicated a poor fit; these models were therefore rejected. Model C with five latent and mutually correlated factors proposed in the original version of MFI20 resulted in an acceptable fit, and this model was chosen for in-depth analysis. Parameter estimates for Model C are reported in Table 4. The values of standardized factor loadings (equivalent to factor-item correlation coefficients) indicate a high degree of correlation between the five factors and the corresponding items of the Polish MFI-20 questionnaire. For example, for the GF factor, the coefficients were between 0.65 and 0.75 (absolute values). This suggests a high degree of correlation
5
between this factor and Items 1, 5, 12, and 16. The complete listing of factor-item correlation coefficients is shown in Table 4. The lowest observed correlations were for Item 6 (0.14; RA factor), Item 10 (0.37; RA factor), Item 19 (0.47; MF factor), and Item 4 (0.52; RM factor). The between-factor correlation coefficients (for standardized covariance coefficients, see Table 4) ranged from 0.55 to 1.0 (absolute values). High values of mutual correlations for RA, PF, GF factors indicate their poor discriminant validity and suggest that these factors may not be clearly distinct. Correlation coefficients between RM, MF, and the other factors were significantly lower. Modification indices were calculated for Model B. The omitted paths with the highest index values were EItem 7 4 EItem 11 (c2 ¼ 37.9), EItem 2 6 4 EItem 10 (c ¼ 29.0), EItem 4 4 EItem 15 2 (c ¼ 27.5) and PF / Item 16 (c2 ¼ 27.7). On the basis of these indices, Model B was modified into Model C. Because of the almost identical meanings of the Polish translation of Items 7 and 11 as well as Items 6 and 10, Items 6 and 11 have been removed from the model. In addition, most items comprising GF, PF, and RA subscales were included in one combined factor. This new model demonstrated superior fit compared with the previous models (Table 3).
Discussion To our knowledge, there is a lack of validated multidimensional CRF assessment tools in Poland. This study established psychometric properties of the Polish version of the MFI-20 questionnaire based on its application in a large sample of cancer patients. Fatigue in cancer patients has so far been mostly investigated in Western societies. Both cultural distinctions and standard of living as
Table 3 Selected Fit Indexes for CFA Models Polish MFI Chi-square CFI SRMR RMSEA 90% CI for RMSEA
Model A
Model B
Model C
747 (df ¼ 170) 0.781 0.073 0.101 0.093e0.108
554 (df ¼ 160) 0.851 0.062 0.086 0.078e0.094
410 (df ¼ 132) 0.882 0.055 0.079 0.063e0.080
CFA ¼ confirmatory factor analysis; df ¼ degrees of freedom; CFI ¼ comparative fix index; SRMR ¼ standardized root mean squared residual; RMSEA ¼ root mean squared error of approximation; 90% CI ¼ 90% confidence interval. Description of the models in the text.
6
Buss et al.
Table 4 Maximum Likelihood Parameter Estimates for Model B Parameter
Unstandardized
Factor loadings GF / Item 1 1a GF / Item 5 0.91 GF / Item 12 0.88 GF / Item 16 0.95 RA/ Item 3 1a RA / Item 6 0.19b RA / Item 10 0.49 RA / Item 17 0.86 PF / Item 2 1a PF / Item 8 1.41 PF / Item 14 1.33 PF / Item 20 1.24 MF / Item 7 1a MF / Item 11 1.04 MF / Item 13 1.11 MF / Item 19 0.87 RM / Item 4 1a RM / Item 9 1.12 RM / Item 15 0.88 RM / Item 18 1.12 Measurement error variances 0.95 EItem 1 1.15 EItem 5 1.36 EItem 12 1.15 EItem 16 1.15 EItem 3 2.33 EItem 6 1.93 EItem 10 1.26 EItem 17 1.50 EItem 2 1.22 EItem 8 1.04 EItem 14 1.01 EItem 20 0.77 EItem 7 0.53 EItem 11 1.17 EItem 13 1.99 EItem 19 1.73 EItem 4 1.18 EItem 9 1.78 EItem 15 1.35 EItem 18 Factor variances and covariance GF 1.26 RA 1.25 PF 0.72 MF 0.76 RM 0.65 GF 4 RA 1.22 GF 4 PF 0.92 GF 4 MF 0.61 GF 4 RM 0.75 RA 4 PF 0.96 RA 4 MF 0.54 RA 4 RM 0.82 PF 4 MF 0.44 PF 4 RM 0.57 MF 4 RM 0.53
SE
Standardized
0.15 0.13 0.15
0.75 0.69 0.65 0.71 0.72 0.14 0.37 0.65 0.57 0.74 0.74 0.73 0.71 0.78 0.67 0.47 0.52 0.63 0.47 0.61
0.09 0.10 0.11 0.10 0.10 0.18 0.15 0.11 0.12 0.11 0.09 0.09 0.08 0.07 0.11 0.16 0.14 0.10 0.14 0.12
0.43 0.52 0.58 0.50 0.47 0.98 0.86 0.57 0.67 0.45 0.44 0.47 0.50 0.39 0.55 0.77 0.72 0.59 0.77 0.62
0.16 0.17 0.13 0.11 0.14 0.14 0.12 0.09 0.11 0.12 0.09 0.12 0.07 0.09 0.08
1 1 1 1 1 0.97 0.96 0.62 0.83 1.00 0.55 0.90 0.59 0.83 0.76
0.07 0.08 0.08 0.08 0.08 0.07 0.14 0.13 0.12 0.08 0.12 0.12
GF ¼ general fatigue; RA ¼ reduced activity; PF ¼ physical fatigue; MF ¼ mental fatigue; RM ¼ reduced motivation. Standardized estimates for measurement errors are proportions of unexplained variance. a Not tested for statistical significance. b P < 0.001 for all other unstandardized estimates except P ¼ 0.015.
Vol.
-
No.
- -
2014
well as gratification of economical concerns may significantly alter the perception and interpretation of fatigue. Our observations show that Polish patients experiencing fatigue or tiredness in the course of cancer generally pay attention to the physical aspects of fatigue such as deterioration in muscle strength to perform normal activities. They seldom report mental aspects of fatigue like RM or concentration and lower mood. In general, it can be said that health deterioration in socially and economically less developed countries is perceived as a physical rather than mental problem. Mental aspects of the disease are ignored more often and taken into consideration unwillingly. From the practical point of view, the management of CRF in those patients depends on whether fatigue or inability to perform daily routines is associated with lack of muscle strength or with the lack of motivation to perform any tasks. This approach corresponds to that presented by Glaus et al.,1 who describes fatigue as a subjective feeling affecting the body (physical), emotions (affective), and mental functioning (mental). This distinction may allow to guide appropriate CRF management. When muscle impairment is observed, factors affecting neuromuscular conduction such as electrolyte imbalances should be corrected first. Exercise may also be encouraged. Benefits of exercise in patients undergoing cancer treatment,17 cancer survivors,18e20 and hospice cancer patients21 were documented. In dealing with the mental or affective aspects of CRF, psychostimulants22 or psychoeducational therapies as well as supportive expressive therapies20,23,24 can be considered. In this study, on the basis of existing Western experience with CRF measurement, we decided to validate MFI-20 in the Polish clinical setting. This questionnaire has good psychometric properties and assesses multiple dimensions of fatigue. The Polish version of MFI-20 was well accepted by patients. The number of missing items was less than 1%. However, the drawback of the study is the lack of information on the number of patients who were disqualified. It was a small number of patients who, despite the lack of significant cognitive impairment, were unable to mark their fatigue level on NRS scale or did not understand the response pattern contained in the MFI-20. In turn, based on the assumption
Vol.
-
No.
- -
2014
The Polish Version of MFI-20 in Cancer Patients
7
Table 5 Items of the New Dimensions of the Polish Version of MFI-20 Item Number Subscales in the in the English Polish Version Version of of MFI-20 MFI-20 PF
MF
1 2 3 5 8 10 12 14 16 17 20 7 13
RM
Removed items
19 4 9 15 18 6 11
Items Translated Into Polish Czuje˛, z_ e jestem w dobrej formie Czuje˛, z_ e fizycznie moge˛ zrobic niewiele Czuje˛ sie˛ bardzo aktywny/a Czuje˛ sie˛ zme˛czony/a Fizycznie potrafie˛ du_zo wzia˛c na siebie Mysle˛, z_ e to co robie˛ w cia˛gu dnia, to bardzo ma1o Jestem wypocze˛ty/a Czuje˛, z_ e jestem w z1ej formie fizycznej qatwo sie˛ me˛cze˛ Niewiele udaje mi sie˛ zrobic Czuje˛, z_ e jestem w swietnej formie fizycznej Kiedy cos robie˛, moge˛ sie˛ na tym skoncentrowac Skoncentrowanie sie˛ na czyms kosztuje mnie du_zo wysi1ku Moje mysli cze˛sto b1a˛dza˛ Mam ochote˛ na wszystkie rodzaje przyjemnosci Przera_za mnie, kiedy musze cos zrobic Mam du_zo plan ow Nie chce mi sie˛ nic robic Mysle˛, z_ e to co robie˛ w cia˛gu dnia, to du_zo Moge˛ sie˛ skoncentrowac
English Items I feel fit Physically I feel only able to do a little I feel very active I feel tired Physically I can take on a lot I think I do very little in a day I am rested Physically I feel I am in a bad condition I tire easily I get little done Physically I feel I am in an excellent condition When I am doing something, I can keep my thoughts on it It takes a lot of effort to concentrate on things My thoughts easily wander I feel like doing all sorts of nice things I I I I I
dread having to do things have a lot of plans do not feel like doing anything think I do a lot in a day can concentrate well
MFI-20 ¼ Multidimensional Fatigue Inventory-20; PF ¼ physical fatigue; MF ¼ mental fatigue; RM ¼ reduced motivation.
that these patients were not able to fill most of the questionnaires, they were not taken into account. Statistical analysis revealed that Polish MFI-20 has both a good reliability with Cronbach’s alpha 0.9 and convergent validity. Structure validity assessed with CFA revealed only three subscales. GF, PF, and RA could not be distinguished in the Polish version of MFI. Similar findings were observed in the French,25 Chinese,26 and Brazilian27 studies where physical and general aspect of fatigue have likewise not been separated. GF is probably too enigmatic to the Polish patients and their doctors. As mentioned previously, our patients’ attitude to fatigue warrants inclusion of GF, PF, and RA into one subscale termed ‘‘physical fatigue.’’ The MF and RM subscales remained similar to the original version. The proposed breakdown of MFI-20 subscales in the Polish version is shown in Table 5. In conclusion, the Polish version of MFI-20 was well accepted by patients. Some dissimilarities were found between the structure of the Polish version of MFI-20 and the original one. Thus, three dimensions of fatigue were proposed in the Polish version of MFI-20. Because of the high number of questions in
the ‘‘PF’’ section of the Polish MFI-20, development of a shorter version of the MFI-20 in Polish can be considered in the future. However, our findings show the usefulness of the Polish version of MFI-20 in the assessment of fatigue in Polish-speaking cancer patients. This instrument can be applied in clinical settings and still takes into account the multidimensionality of fatigue. We hope that the implementation of the Polish version of MFI-20 will enable clinicians both to not overlook the problem and to assess CRF objectively and easily. It has the potential to improve both the detection of CRF and patient care in our country. We hope that this study will inspire further research on the topic of CRF in Poland.
Disclosures The authors declare that they have no conflict of interest.
References 1. Glaus A, Crow R, Hammond S. A qualitative study to explore the concept of fatigue/tiredness
8
Buss et al.
in cancer patients and in healthy individuals. Support Care Cancer 1996;4:82e96. 2. Glaus A. Assessment of fatigue in cancer and non-cancer patients and in healthy individuals. Support Care Cancer 1993;1:305e315. 3. Radbruch L, Strasser F, Elsner F, et al. Fatigue in palliative care patientsdan EAPC approach. Palliat Med 2008;22:13e32. 4. Ferreira KA, Kimura M, Teixeira MJ, et al. Impact of cancer-related symptom synergisms on health-related quality of life and performance status. J Pain Symptom Manage 2008;35:604e616. 5. van den Beuken-van Everdingen MH, de Rijke JM, Kessels AG, Schouten HC, van Kleef M, Patijn J. Quality of life and non-pain symptoms in patients with cancer. J Pain Symptom Manage 2009;38:216e233. 6. Curt GA, Breitbart W, Cella D, et al. Impact of cancer-related fatigue on the lives of patients: new findings from the fatigue coalition. Oncologist 2000;5:353e360. 7. Walsh D, Donnelly S, Rybicki L. The symptoms of advanced cancer: relationship to age, gender, and performance status in 1000 patients. Support Care Cancer 2000;8:175e179. 8. Lawrence DP, Kupelnick B, Miller K, Devine D, Lau J. Evidence report on the occurrence, assessment, and treatment of fatigue in cancer patients. J Natl Cancer Inst Monogr 2004;32:40e50. 9. Stone P, Richardson A, Ream E, Smith AG, Kerr DJ, Kearney N. Cancer-related fatigue: inevitable, unimportant and untreatable? Results of a multi-centre patient survey. Cancer Fatigue Forum. Ann Oncol 2000;11:971e975. 10. Borneman T, Koczywas M, Sun VC, Piper BF, Uman G, Ferrell B. Reducing patient barriers to pain and fatigue management. J Pain Symptom Manage 2010;39:486e501. 11. Smets EM, Garssen B, Bonke B, De Haes JC. The Multidimensional Fatigue Inventory (MFI) psychometric qualities of an instrument to assess fatigue. J Psychosom Res 1995;39:315e325. 12. Smets EM, Garssen B, Cull A, de Haes JC. Application of the multidimensional fatigue inventory (MFI-20) in cancer patients receiving radiotherapy. Br J Cancer 1996;73:241e245. 13. Hagelin CL, Wengstr€ om Y, Runesdotter S, F€ urst CJ. The psychometric properties of the Swedish Multidimensional Fatigue Inventory MFI-20 in four different populations. Acta Oncol 2007;46:97e104. 14. Dewolf L, Koller M, Velikova G, Johnson C, Scott N, Bottomley A. on behalf of the EORTC Quality of Life Group. EORTC Quality of Life Group translation procedure March 2009, 3rd ed. Copyright EORTC Brussels. Available at: http://groups.eortc.be/qol/downloads/translation_ manual_2009.pdf. Accessed June 30, 2011.
Vol.
-
No.
- -
2014
15. Aaronson NK, Ahmedzai S, Bergman B, et al. The European Organization for Research and Treatment of Cancer QLQ-C30: a quality-of-life instrument for use in international clinical trials in oncology. J Natl Cancer Inst 1993;85:365e376. 16. de Walden-Ga1uszko K, Majkowicz M. The assessment of the quality of palliative home care in the theory and practice. (Ocena jakosci opieki paliatywnej w teorii i praktyce). Gda nsk: Medical University of Gda nsk; 2000. 17. Velthuis MJ, Agasi-Idenburg SC, Aufdemkampe G, Wittink HM. The effect of physical exercise on cancer-related fatigue during cancer treatment: a meta-analysis of randomised controlled trials. Clin Oncol (R Coll Radiol) 2010;22:208e221. 18. McNeely ML, Campbell KL, Rowe BH, Klassen TP, Mackey JR, Courneya KS. Effects of exercise on breast cancer patients and survivors: a systematic review and meta-analysis. CMAJ 2006;175:34e41. 19. Oldervoll LM, Kaasa S, Knobel H, Loge JH. Exercise reduces fatigue in chronic fatigued Hodgkins disease survivorsdresults from a pilot study. Eur J Cancer 2003;39:57e63. 20. Duijts SF, Faber MM, Oldenburg HS, van Beurden M, Aaronson NK. Effectiveness of behavioral techniques and physical exercise on psychosocial functioning and health-related quality of life in breast cancer patients and survivorsda meta-analysis. Psychooncology 2011;20:115e126. 21. Buss T, de Walden-Ga1uszko K, Modli nska A, Osowicka M, Lichodziejewska-Niemierko M, Janiszewska J. Kinesitherapy alleviates fatigue in terminal hospice cancer patientsdan experimental, controlled study. Support Care Cancer 2010;18:743e749. 22. Minton O, Richardson A, Sharpe M, Hotopf M, Stone PC. Psychostimulants for the management of cancer-related fatigue: a systematic review and metaanalysis. J Pain Symptom Manage 2011;41:761e767. 23. Boesen EH, Ross L, Frederiksen K, et al. Psychoeducational intervention for patients with cutaneous malignant melanoma: a replication study. J Clin Oncol 2005;23:1270e1277. 24. Dolbeault S, Cayrou S, Br edart A, et al. The effectiveness of a psycho-educational group after earlystage breast cancer treatment: results of a randomized French study. Psychooncology 2009;18:647e656. 25. Gentile S, Delarozi ere JC, Favre F, Sambuc R, San Marco JL. Validation of the French ’multidimensional fatigue inventory’ (MFI 20). Eur J Cancer Care (Engl) 2003;12:58e64. 26. Tian J, Hong JS. Validation of the Chinese version of Multidimensional Fatigue Inventory-20 in Chinese patients with cancer. Support Care Cancer 2012;20:2379e2383. 27. Baptista RL, Biasoli I, Scheliga A, et al. Psychometric properties of the multidimensional fatigue inventory in Brazilian Hodgkin’s lymphoma survivors. J Pain Symptom Manage 2012;44:908e915.
