Abstract
Background. The aims of this study are to present the results of a new general population normative study of the quality of life questionnaire EORTC QLQ-C30 and to give European reference values averaged across six studies.
Methods. The empirical study was based on a representative sample of the German adult population (N = 2448). The subjects were asked to fill in several questionnaires, one of them being the EORTC QLQ-C30.
Results. EORTC QLQ-C30 mean scores of this sample indicated slightly better quality of life (QoL) than in previous European studies. QoL decreased with age, but there were only small gender differences. The mean scores were compared with the age and gender adjusted scores of five other European normative studies from Sweden, the Netherlands, Norway, and Germany (N between 1731 and 4910). Finally, the data of these five studies and the new study were combined to arrive at averaged European normative values for the scales and the symptom items of the questionnaire.
Conclusion. The reference values of the scales pooled across six European studies (N = 16 151) can be used as general population references for QoL scores of cancer patients.
Quality of life (QoL) has become an important outcome criterion in oncology. Multiple questionnaires measuring QoL have been developed. One of the most frequently used questionnaires designed especially for cancer patients is the European Organization for Research and Treatment of Cancer Quality of Life Questionnaire EORTC QLQ-C30 [Citation1]. It has been used in many oncological studies [Citation2] and shows good psychometric properties [Citation3]. Normative values are necessary both for assessing the values of individual patients and for making adjusted comparisons among samples of patients differing in age and gender distributions. Several normative examinations of the EORTC QLQ-C30 have already been performed in Norway [Citation4], Germany [Citation5], Sweden [Citation6,Citation7], the Netherlands [Citation8], South Korea [Citation9], and Colombia [Citation10]. Of these, mean QoL scores were lowest in South Korea. However, there are also several differences among the European studies. The mean scores of the general two-item scale global health/QoL range between 70.4 (Germany [Citation5]) and 78.0 (Netherlands [Citation8]). Some studies yielded strong age differences with decreasing QoL in older subjects [Citation4,Citation5], while in other studies the age effects were weak. Women reported more symptoms and worse QoL than men in most of the studies although this effect is not generally observed. People from Norway [Citation4] reported markedly more symptoms than the populations of the other countries even though their global QoL assessment (mean = 73.7) was not bad.
The two Swedish normative studies [Citation6,Citation7], with a 12-year time interval, yielded similar results. One aim of this paper is to test whether the German normative values can also be replicated after 12 years, and to compare the findings of this new study with those of the other international investigations.
Moreover, we believe that it is time to give a summarizing report on the normative examinations already published. Two papers [Citation8,Citation9] already compared their study results with mean scores obtained in the other examinations, but they neither considered the different age distributions nor aggregated mean scores across these studies. Researchers from countries in which no normative studies have been conducted might profit from normative scores that integrate across the available data sets from several countries. Providing data for these comparisons is one central aim of this paper. Since there are cultural differences between Europe, Asia (Korea) and Latin America (Colombia), we restrict the combined analysis to the studies performed in Europe.
In summary, the aims of this paper are: 1) to present the results of a new German normative study; 2) to compare the results with those of the German normative study from 2001 and with the other international normative examinations; and 3) to provide reference values averaged across six European normative studies.
Methods
Sample
A representative sample of the German general population was selected in 2012 with the assistance of a demographic consulting company (USUMA, Berlin, Germany). The entire country was separated into 320 sample areas representing different regions. Once a sample area was selected, street, house, and household were selected randomly. The person within the household was also selected on a random basis using the Kish-selection-grid technique. The resulting sample is roughly representative of the German population living in private houses in terms of age, gender, and education. A first attempt was made for 4480 addresses, of which 4436 were valid. If the selected person was not at home, a maximum of three attempts was made to contact them. In all, 1195 subjects (27.0%) refused participation, 700 subjects (15.8%) could not be reached after three attempts, and 23 subjects (0.5%) refused participation because of severe health problems. All subjects were visited by a study assistant, gave written informed consent, and filled in several questionnaires. A total of 2510 people between 14 and 92 years old agreed to participate and completed the self-rating questionnaires (participation rate: 56.6% of valid addresses). Subjects younger than 18 years were excluded from the analysis (N = 62). Therefore, the final sample consisted of 2448 subjects. The study was approved by the ethical committee of the Leipzig University.
Questionnaires
The EORTC QLQ-C30 is comprised of 30 items arranged into nine scales and six single items. There are five functioning scales (Physical, Role, Cognitive, Emotional, and Social Functioning), three symptom scales (Fatigue, Pain, and Nausea/Vomiting) and one two-item global health/QoL scale. Each item is to be answered with a four-point scale, from 1 (not at all) to 4 (very much), except for the two global health/QoL items. These have seven response options ranging from 1 (very poor) to 7 (excellent).
In addition to the EORTC QLQ-C30, the subjects of this study also completed the Patients Health Questionnaire PHQ-9, measuring depression with nine items [Citation11], the Generalized Anxiety Disorder Questionnaire GAD-2, measuring anxiety with two items [Citation12], and the Patients Health Questionnaire – Somatic Symptom Short Form (PHQ-SSS) with eight items measuring bodily complaints [Citation13].
Statistical analyses
According to the manual of the EORTC QLQ-C30, all scales and items were transformed to a 0–100 range. Cronbach's alpha was used to assess internal consistency of multi-item scales, and Pearson product-moment-correlations were calculated to describe the association between the scales of the questionnaires.
One problem for the comparison of mean scores across several normative studies is that, there are differences in the age and gender distributions of the samples. However, fortunately, all EORTC QLQ-C30 normative studies used identical age decades for the description of the normative data. To compare the mean scores of the studies, it is useful to relate the results to a standard population. We used the European population (EU-27; 501 million people) reported from EUROSTAT [Citation14] data from 2011. The percentages of subjects per age and gender group are as follows: Males: 19.6% (18–29 years), 18.9% (30–39 years), 19.0% (40–49 years), 16.6% (50–59 years), 13.0% (60–69 years), 13.0 (≥ 70 years); Females: 18.2% (18–29 years), 17.8% (30–39 years), 17.6% (40–49 years), 15.6% (50–59 years), 12.0% (60–69 years), and 18.7% (≥ 70 years). Using these percentages as weighting factors, we calculated weighted EORTC QLQ-C30 mean scores for each country. This procedure makes the mean scores independent from the distribution of the study populations. The combined analysis comprises six European normative studies, based on these weighting factors. A further normative study from Denmark [Citation15] was not included in this comparison because the sample was comprised solely of women, and because this study used version 1.0 of the EORTC QLQ-C30.
Results
Age and gender differences
presents characteristics of the subject sample. Concerning age and gender distribution, the sample can be assumed to be fairly representative of the German general population. The corresponding percentages for the age groups (18–29 years, 30–39 years, 40–49 years, 50–59 years, 60–69 years and ≥ 70 years), taken from the census [Citation16], are as follows: Males: 18.1%, 15.2%, 21.4%, 17.3%, 13.5%, and 14.6%, and Females: 16.4%, 14.0%, 19.4%, 16.3%, 13.4%, and 20.4%, respectively.
Table I. Sociodemographic characteristics of the sample.
presents mean values of the EORTC QLQ-C30 scales, stratified by age group and gender. While there are clear age effects in most scales, the gender effects are very small in this study.
Table II. Mean scores (standard deviations in parentheses) of the scales and items broken down by sex and age groups.
Psychometric properties
Internal consistency (Cronbach's alpha) coefficients are given in the right part of for all scales. With two exceptions (Cognitive functioning and Nausea/Vomiting), all alpha values are greater than 0.80. Correlation coefficients with the three other questionnaires are also presented in . The correlations are highest for the PHQ-SSS, measuring bodily complaints, but Emotional functioning is correlated highest with the depression scale PHQ-9.
Table III. Pearson's correlations with depression (PHQ-9), anxiety (GAD-2) and complaints (PHQ-SSS), and reliability (Cronbach's alpha).
International comparisons
Since the samples of the normative studies reported in the literature have different age and gender distributions, a fair comparison should take into account these differences. reports the mean scores of the studies that would have been resulted if the age and gender distributions were identical to the European standard distribution (see Methods section). Therefore, the mean scores given in differ slightly from those reported in the original studies. Next to the column of the new normative study, the scores from the German 2001 examination are presented in . The global health/QoL mean score was worse (M = 71.5) in 2001 compared with the new examination (M = 75.9). The column ‘Europe’ of presents the means across the six European examinations.
Table IV. EORTC QLQ-C30 mean scores of international studies, adjusted for age and gender.
presents the European mean scores for each age and gender category. The standard deviations (SDs) given in are the pooled SDs of all studies that reported the corresponding SDs.
Table V. Pooled mean scores of the scales and items of six European normative studies.
Discussion
Compared with the German normative scores [Citation5] of 2001 (data from 1998), the mean scores of this study are higher in most scales, indicating better QoL in the present study. The global health/QoL mean score of the new study (M = 75.9) is nearly identical with the European mean (M = 75.7), while the corresponding value of 2001 was lower (M = 71.5) [Citation5].
As in the previous German study [Citation5], more pronounced age differences were observed in the new study than in the other European examinations. The global health/QoL mean score difference (averaged across males and females) between the oldest and the youngest age group was d = 22.3 in the new study () and d = 23.0 in the German 2001 study [Citation5], while the corresponding difference in the European data set () was only d = 11.7. There were however only small gender effects in the new study. The difference between males and females with regard to global health/QoL was d = 1.0 (new study), compared with d = 3.5 (German 2001 study) [Citation5], and d = 3.2 in Europe (). The small gender differences in the new study are in contrast to most other studies, which generally found better global QoL, better functioning, and fewer symptoms (especially fatigue and pain) in males than in females.
Compared with the differences between the German studies, the results of the two Swedish investigations with a 12 year interval [Citation6,Citation7] were very similar. Since the sampling procedures of the two German studies were identical, the differences in mean scores and gender dependency cannot be traced back to the sampling method. Other possible reasons are: real time trends, peculiarities of the samples of participants, seasonal effects, effects of the position of the questionnaire within the set of questionnaires, or just chance. We do not think that it is justified to conclude a historical trend toward better QoL from the comparison of the two German studies.
Two studies already presented international comparisons [Citation8,Citation9], both done on the basis of the originally reported values. Since we know that there are age and gender differences in the data, a fair comparison should take into account age and gender distributions. This was done in the calculation of the scores given in . Moreover, presents mean scores averaged across six European studies. Each of the six studies was included with the same weight. It would also be possible to give the studies different weighting factors, according to the sample sizes or the nationality (two countries with two studies each). However, as long as we do not have explicit criteria for weighting the examinations, it seems best to treat all studies equally.
All six European studies were performed in Northern or Central European countries. The non-European studies were not included because of the cultural differences between Europe and Korea and Colombia. While the Colombian data indicated higher QoL in the functioning scales and mean values in the symptom scales, the Korean data set indicated worse QoL compared with Europe.
There are several cross-cultural comparisons with respect to QoL in the literature. An international study [Citation17] reported population-based WHOQOL-BREF mean values of 23 countries, including three countries (Germany, Norway, and the Netherlands) that were also considered in the calculation of the EORTC QLQ-C30 means (). The mean values of these three countries were higher than the world mean scores in nine of 12 (three countries * four domains) comparisons [Citation17]. Averaged across the four dimensions of the WHOQOL-BREF and the three countries, the effect size for the mean score difference between the three countries and the world means was 0.27. However, industrialized countries are not generally characterized by better QoL than low and medium income countries. International population mean scores of the SF-36 were compared in another study [Citation18], including Brazil, Turkey, UK, USA, and Croatia. The mean score comparison between Brazil and Turkey, on the one hand, and UK and USA, on the other hand, did not provide systematic differences: in 18 of 32 comparisons (four pairs of countries * eight dimensions) Brazil and Turkey showed worse QoL. The Croatian values were lowest across the board, though the Croatian WHOQOL-BREF mean scores were above the WHOQOL-BREF total mean score in the study mentioned above [Citation17] in all four domains. SF-36 population mean scores in Mexico were higher (better QoL) in five of the eight dimensions compared to Canada and the USA [Citation19].
There are several studies examining population-based mean scores of anxiety and depression with the Hospital Anxiety and Depression Scale (HADS). Among the European countries, depression mean scores were between 3.3 (Norway) [Citation20] and 4.7 (Germany) [Citation21], and there were even greater differences in the anxiety subscale, with values ranging from 3.9 (Netherlands) [Citation22] to 6.4 (UK) [Citation23].
These comparisons show that Northern and Central European countries are often (though not always) characterized by high levels of QoL. But there are also many studies reporting high QoL levels in low and medium income countries. Furthermore, there are unsystematic differences among the studies performed in European countries with similar social systems. Reasons for these differences are difficult to establish. Putative factors are setting (written or telephone), presence or absence of an interviewer, inviting company (governmental institution or private institute), time trends, or seasonal trends. Since it is impossible to identify the reasons for the differences here, we believe that it is best to aggregate across these different studies.
The European data set seems to be well suited for utilization in Northern and Central Europe, but countries that have no own normative examinations might profit from the aggregated data set more than from the selection of a single normative study.
In contrast to most other QoL questionnaires, the EORTC QLQ-C30 comprises a broad spectrum of functioning scores and single symptoms. The comparison among the normative studies could be facilitated if summarizing scales or scales of higher order were calculated. Three proposals for the calculation of dimensions of higher order have been described in the literature [Citation24–26]. However, until now there is no recommendation of the EORTC for the use of a certain way to derive dimensions of higher order, and the reliability and validity of the summarizing scores offered in these publications are not sufficiently tested. Therefore, we restricted our analyses to the scales and items of first order, but we encourage the development and the use of summarizing scales.
Some limitations of the study should be mentioned. One disadvantage of the new study is the relatively low response rate of 57%, compared to percentages between 68% and 78% in the other European investigations. We cannot infer whether the non-responders led to an overestimation or underestimation of QoL. In the compilation of the six European studies, Southern and Eastern European countries are not represented. Each of the six studies was included in the summarizing European analysis with the same weight. While all studies reported mean scores for the age groups, standard deviations were not available for all of them. Therefore, we only report pooled standard deviations across the examinations that reported these figures. As there are too few normative studies from low and middle income countries, we cannot derive worldwide normative data. Regression analyses can be helpful in calculating expected mean scores for each age and gender distribution. This procedure is described in two studies [Citation5,Citation27]. It would be useful to derive such regression coefficients for the compiled European data.
In summary, the European normative scores can be used as reference values when no better references are available. Physicians can use the columns of when they evaluate the EORTC QLQ-C30 scores of a single patient. In addition, in clinical studies samples of patients with different age and gender distributions can be compared using the figures of , based on a sound data basis of 16 151 subjects.
Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.
References
- Aaronson NK, Ahmedzai S, Bergman B, Bullinger M, Cull A, Duez NJ, 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;85:365–76.
- Martinelli F, Quinten C, Maringwa JT, Coens C, Vercauteren J, Cleeland CS, et al. Examining the relationships among health-related quality-of-life indicators in cancer patients participating in clinical trials: A pooled study of baseline EORTC QLQ-C30 data. Expert Rev Pharmacoeconom Outcomes Res 2011;11:587–99.
- Bjordal K, de Graeff A, Fayers PM, Hammerlid E, van Pottelsberghe C, Curran D, et al. A 12 country field study of the EORTC QLQ-C30 (version 3.0) and the head and neck cancer specific module (EORTC QLQ-H & N35) in head and neck patients. Eur J Cancer 2000;36: 1796–807.
- Hjermstad MJ, Fayers PM, Bjordal K, Kaasa S. Health- related quality of life in the general Norwegian population assessed by the European Organization for Research and Treatment of Cancer Core Quality-of-Life Questionnaire: The QLQ = C30 (+ 3). J Clin Oncol 1998;16:1188–96.
- Schwarz R, Hinz A. Reference data for the quality of life questionnaire EORTC QLQ-C30 in the general German population. Eur J Cancer 2001;37:1345–51.
- Derogar M, van der Schaaf M, Lagergren P. Reference values for the EORTC QLQ-C30 quality of life questionnaire in a random sample of the Swedish population. Acta Oncol 2012;51:10–6.
- Michelson H, Bolund C, Nilsson B, Brandberg Y. Health-related quality of life measured by the EORTC QLQ-C30 – Reference values from a large sample of the Swedish population. Acta Oncol 2000;39:477–84.
- van de Poll-Franse L, Mols F, Gundy CM, Creutzberg CL, Nout RA, Verdonck-de Leeuw IM, et al. Normative data for the EORTC QLQ-C30 and EORTC-sexuality items in the general Dutch population. Eur J Cancer 2011;47:667–75.
- Yun YH, Kim SH, Lee KM, Park SM, Kim YM. Age, sex, and comorbidities were considered in comparing reference data for health-related quality of life in the general and cancer populations. J Clin Epidemiol 2007;60:1164–75.
- Finck C, Barradas S, Singer S, Zenger M, Hinz A. Health related quality of life in Colombia: Reference values of the EORTC QLQ-C30. Eur J Cancer Care 2012;21:829–36.
- Kroenke K, Spitzer RL, Williams JBW. The PHQ-9 – Validity of a brief depression severity measure. J Gen Intern Med 2001;16:606–13.
- Lowe B, Decker O, Muller S, Brahler E, Schellberg D, Herzog W, et al. Validation and standardization of the generalized anxiety disorder screener (GAD-7) in the general population. Med Care 2008;46:266–74.
- Jasper F, Hiller W, Rist F, Bailer J, Witthoft M. Somatic symptom reporting has a dimensional latent structure: Results from taxometric analyses. J Abnorm Psychol 2012; 121:725–38.
- EUROSTAT. Available from: http://appsso.eurostat.ec.europa.eu/nui/show.do?dataset = demo_pjangroup&lang = en.
- Klee M, Groenvold M, Machin D. Quality of life of Danish women: Population-based norms for the EORTC QLQ-C30. Qual Life Res 1997;6:27–34.
- Destatis. Available from: “https://www.destatis.de/DE/Publikationen/StatistischesJahrbuch/Bevoelkerung.pdf?__blob = publicationFile.
- Skevington SM, Lotfy M, O’Connell KA. The World Health Organization’s WHOQOL-BREF quality of life assessment: Psychometric properties and results of the international field trial – A report from the WHOQOL group. Qual Life Res 2004;13:299–310.
- Cruz LN, Fleck MPD, Oliveira MR, Camey SA, Hoffmann JF, Bagattini AM, et al. Health-related quality of life in Brazil: Normative data for the SF-36 in a general population sample in the south of the country. Ciencia Saude Coletiva 2013;18:1911–21.
- Duran-Arenas L, Gallegos-Carrillo K, Salinas-Escudero G, Martinez-Salgado H. Towards a Mexican normative standard for measurement of the Short Format 36 health-related quality of life instrument. Salud Publica de Mexico 2004;46:306–15.
- Nortvedt MW, Riise T, Sanne B. Are men more depressed than women in Norway? Validity of the Hospital Anxiety and Depression Scale. J Psychosom Res 2006;60:195–8.
- Hinz A, Brähler E. Normative values for the Hospital Anxiety and Depression Scale (HADS) in the general German population. J Psychosom Res 2011;71:74–8.
- Spinhoven P, Ormel J, Sloekers PPA, Kempen GIJM, Speckens AEM, VanHemert AM. A validation study of the Hospital Anxiety and Depression Scale (HADS) in different groups of Dutch subjects. Psychol Med 1997;27:363–70.
- Crawford JR, Garthwaite PH, Lawrie CJ, Henry JD, MacDonald MA, Sutherland J, et al. A convenient method of obtaining percentile norms and accompanying interval estimates for self-report mood scales (DASS, DASS-21, HADS, PANAS, and sAD). Br J Clin Psychol 2009;48: 163–80.
- Hinz A, Einenkel J, Briest S, Stolzenburg J, Papsdorf K, Singer S. Is it useful to calculate sum scores of the quality of life questionnaire EORTC QLQ-C30?Eur J Cancer Care 2012;21:677–83.
- Gundy CM, Fayers PM, Groenvold M, Petersen MA, Scott NW, Sprangers MAG, et al. Comparing higher order models for the EORTC QLQ-C30. Qual Life Res 2012;21: 1607–17.
- Tian J, Hong JS. Validation of the Chinese version of Multidimensional Fatigue Inventory-20 in Chinese patients with cancer. Support Care Cancer 2012;20: 2379–83.
- Hjermstad MJ, Fayers PM, Bjordal K, Kaasa S. Using reference data on quality of life – the importance of adjusting for age and gender, exemplified by the EORTC QLQ-C30 (+ 3). Eur J Cancer 1998;34:1381–9.