• France;
  • health-related quality of life;
  • psychometric properties;
  • questionnaire;
  • reference values


  1. Top of page
  3. Introduction
  4. Methods
  5. Results
  6. Conclusions
  7. References

Objectives:  To assess the psychometric properties of the disease-specific Quality of Life Assessment of Growth Hormone Deficiency in Adults (QoL-AGHDA) questionnaire in a general population, and collect French normative data.

Methods:  A postal survey was conducted on 2900 adult panelists representative of the French population. The participants were asked to complete a questionnaire including the QoL-AGHDA and an evaluation of their overall health status (OHS). The QoL-AGHDA score ranges from 0 to 25, a lower score indicating better QoL. Psychometric properties of the QoL-AGHDA were assessed. The mean QoL-AGHDA scores were described by sex and age groups.

Results:  The return rate was 75%. The quality of completion and internal consistency reliability were good: 95% of the respondents completed all 25 QoL-AGHDA items and Cronbach's alpha was 0.86. The QoL-AGHDA score was able to discriminate between the respondents according to their OHS (from 1.5 for excellent to 12.3 for poor OHS, P < 0.001). The mean QoL-AGHDA score was 4.6 for the overall population, 5.1 for females and 4.2 for males, and ranged from 4.8 for the youngest to 6.1 for the oldest respondents.

Conclusions:  The QoL-AGHDA questionnaire showed good psychometric properties when administered in the French population. French reference values were collected, completing the QoL-AGHDA normative database already available in several European countries.


  1. Top of page
  3. Introduction
  4. Methods
  5. Results
  6. Conclusions
  7. References

Growth hormone deficiency (GHD) occurs when the pituitary gland is unable to release or produce adequate amounts of GH. In adults, GHD most frequently results from pituitary tumors and their treatments including surgery or radiotherapy. GHD affects body composition, bone mineral density, muscle strength, and exercise performance as well as the cardiovascular system and metabolism [1]. Studies indicate that patients' health-related quality of life (HRQoL) and social status can also be profoundly affected by GHD [1–4]. The major impacts of GHD on HRQoL and social status include reduced vitality and energy, emotional lability, social isolation, and negative impact on sex life as well as impact on marital and professional status. Therefore, assessing HRQoL in adults with GHD is important, especially when measuring the effect of GH treatment [5–7].

The impairment of HRQoL of GHD patients has often been demonstrated by comparing GHD patients with a control group of healthy subjects [2,4,8,9]. Moreover, the collection of reference values for HRQoL questionnaires in general populations also helps researchers understand and interpret results obtained in clinical studies. Reference data are also very useful to develop realistic hypotheses during the design stage of clinical trials, especially to adequately power the trials. Unfortunately, reference data for HRQoL questionnaires in general populations are often unavailable except for very common questionnaires such as the Short-Form 36 (SF-36) [10], the EuroQoL-5 Dimension (EQ-5D) [11], and the European Organization for Research and Treatment of Cancer QoL Questionnaire (EORTC QLQ-C30) [12,13].

The QoL-Assessment of GHD in Adults (QoL-AGHDA) [14] is a GHD-specific questionnaire that has been widely used to measure the effect of GH treatment on patients' HRQoL [8,15–17]. Age- and sex-specific reference values for the QoL-AGHDA score are available for general populations of England and Wales [18], Spain [19], The Netherlands [15], and Sweden [15,20,21]. However, with the exception of Wiren et al. [21], who examined the scaling properties of the QoL-AGHDA questionnaire in the Swedish general population, none of these studies reported information on the psychometric properties of the QoL-AGHDA questionnaire when administered in a general population. Therefore, to evaluate whether the QoL-AGHDA questionnaire has good psychometric properties in a general population and to complete the QoL-AGHDA normative database, a panel study was conducted in the French general population. The results are reported in this article.


  1. Top of page
  3. Introduction
  4. Methods
  5. Results
  6. Conclusions
  7. References

Study Design

A postal study was conducted by Taylor Neslon Sofrès (TNS) Healthcare Institute using a sample of 2900 adults (≥18 years) belonging to the TNS Access Santé polling panel and selected to be representative of the French population in terms of geographic area, conurbation category, sex, age, and socio-professional category. The Access Santé polling panel is updated monthly to maintain its size and to ensure that it remains representative. Individuals in the polling panel are recruited through several means (face-to-face interviews, mailings, phone calls) to reduce the risk of selection bias possibly associated with a particular methodology. Panelists participate in six to eight health surveys per year and receive incentive services for their participation (no payment). Given the standard rate of response for a survey such as this (70–75%), a target sample size of 2000 participants was expected. The participants gave their consent to participate in the study by returning an eight-page questionnaire to TNS Healthcare. In addition to the QoL-AGHDA questionnaire, the panelists were asked in the eight-page questionnaire to complete the EQ-5D [22], an item on their overall health status, and questions about their general and medical situation.

The QoL-AGHDA Questionnaire

The QoL-AGHDA questionnaire is composed of 25 items assessing the HRQoL of GHD patients. These items measure problems on memory and concentration, tiredness, tenseness, social isolation, and self-confidence. It was developed in 1999 simultaneously in UK English, Swedish, German, Italian, and Spanish based on interviews with adult patients with either adult or childhood onset GHD [14,23]. The five language versions showed good test–retest reliability, internal consistency, and construct validity. The QoL-AGHDA questionnaire was translated into French, and the French version was psychometrically validated in GHD patients in 2003 [24].

Statistical Analyses

Representativeness of the sample.  The sociodemographic characteristics of respondents were compared with those of the nonrespondents and with those of the French general population. If the respondents differed from the general population regarding the sociodemographic characteristics used to establish the quotas of the French general population (i.e., geographic area, conurbation category, sex, age, and socio-professional category), data were weighted to obtain a closer representation of the French general population; weights were calculated to retrieve the proportions of people observed in the French general population regarding geographic area, conurbation category, sex, age and socio-professional category.

Scoring of the QoL-AGHDA questionnaire and EQ-5D.  The QoL-AGHDA questionnaire is composed of 25 items with a “yes/no” answer choice scored as 1/0. The QoL-AGHDA total score is calculated as the sum of item scores ranging from 0 to 25, with a higher score indicating poorer HRQoL. The QoL-AGHDA total score is calculated only if all 25 items are completed.

A utility index ranging from 0 to 1 with 0 representing death and 1 representing full health, was derived from the five EQ-5D items using the European visual analog scale (VAS) valuation set [25]. The EQ-5D also contains a VAS ranging from 0 to 100, with 0 representing the worst imaginable health state and 100 representing the best imaginable health state.

Analysis of the QoL-AGHDA data.  With the exception of the quality of completion that was analyzed on all questionnaires returned to TNS Healthcare, the analysis included all the respondents who entirely completed the 25 items of the QoL-AGHDA questionnaire. The quality of completion was assessed using the percentage of missing data per respondent and per item. The internal consistency reliability of the QoL-AGHDA score was assessed using Cronbach's alpha. A Rasch analysis was conducted on QoL-AGHDA items. The Rasch model enabled the items to be ordered hierarchically in terms of their “difficulty” and the respondents to be ordered hierarchically in terms of their HRQoL impairment. The “difficulty” of an item refers to its probability of being answered “yes,” i.e., of declaring an impairment on this item, with the greater the “difficulty” of an item, the lower its probability of being answered “yes.” The ordering of the items and the respondents were conjointly represented on a person–item map. The QoL-AGHDA mean score was compared between 10-year age groups, sex, and overall health status. The correlation coefficient of the QoL-AGHDA score with EQ-5D scores was calculated.

Statistical tests, level of significance and software.  A chi-square test was used when comparing two categorical variables, t test or analysis of variance when comparing means, and Pearson's correlation coefficient when analyzing the linear relationship between variables. The threshold for statistical significance was fixed at 5% for each test. All analyses were performed with SAS software for Windows (version 9.1.3, SAS Institute, Inc., Cary, NC) and RUMM software (Rasch Unidimensional Model Measurement, version 3.1, RUMM laboratory, Perth, WA, Australia).


  1. Top of page
  3. Introduction
  4. Methods
  5. Results
  6. Conclusions
  7. References

Demographic and Clinical Characteristics of the Study Sample

Among the 2900 panelists who were asked to participate in this study, 2181 (75%) returned the questionnaire to TNS Healthcare. The mean age was 50, with 54% females (Table 1). Less than a quarter (21%) of the respondents had a long-standing illness (Table 2) and 76% had good to excellent overall health status. The majority of the respondents had no problems regarding mobility, self-care, usual activities (≥89%), and anxiety/depression (69%), but 49% reported some problems with pain or discomfort.

Table 1.  Representativeness of the study sample
 Respondents N = 2181Nonrespondents N = 719French general population
  • *

    Chi-square test.

  • Source: INSEE, Enquête Emploi 2002 redressée 2007.

  • NA, not applicable; NS, not significant.

Age in years   
 Mean (SD)50 (18)41 (17)NA
Age groups (%)   
 Difference with respondents P < 0.0001*P < 0.0001*
Sex (%)   
 Difference with respondents P < 0.001*NS*
Socio-professional category (%)   
 Employed (self-employed, managerial staff, other employment, etc.)546657
 Unemployed (job seeker, student, housewife, other)182119
 Difference with respondents P < 0.0001*P < 0.05*
Marital status (%)   
 Significant other1117
 Difference with respondents P < 0.0001*
Education level (%)   
 ≤5 years of education10722
 6–9 years of education353036
 10–12 years of education252716
 ≥13 years of education313626
 Difference with respondents P < 0.05*P < 0.0001*
Table 2.  Health characteristics of respondents
 Respondent population N = 2181
  • *

    No problem/some problems/extreme problems.

  • Percentages include missing data.

  • EQ-5D, EuroQoL-5 Dimension; VAS, Visual Analogue Scale.

Long-standing illness (%)21
Overall health status (%) 
 Very good31
EQ-5D index 
 Mean (SD)0.8 (0.2)
 Mean (SD)79 (16)
EQ-5D items (%) 
 Usual activities89/10/0

The respondents were significantly older than the nonrespondents (mean age 50 vs. 41) (Table 1). More often than the nonrespondents, the respondents were females (54% vs. 46%), married (51% vs. 44%) or widowed (9% vs. 4%), retired (28% vs. 13%), and less educated (45% with less than 10 years of education vs. 37%). The respondents were significantly older than the French general population (51% older than 50 years vs. 45%) (Table 1) and more often, they were retired (28% vs. 24%). They also had a higher level of education than the French general population (56% with more than 10 years of education vs. 42%).

French Normative Reference Values for the QoL-AGHDA Score

As the respondent sample appeared to be slightly different from the French general population in terms of sociodemographic parameters, the QoL-AGHDA score was calculated using weighted data to get descriptive statistics for a representative sample of the French population. Among the 2181 respondents, 2065 completed all 25 QoL-AGHDA items.

The mean QoL-AGHDA score was relatively low (4.6), with 20% of the respondents having the lowest possible score (i.e., 0), indicating good HRQoL (Table 3 and Fig. 1). A slightly higher mean QoL-AGHDA score was observed for the females than for the males, and for the older respondents than for the younger respondents, indicating poorer HRQoL for the females and the older respondents.

Table 3.  French normative reference values for the QoL-AGHDA score (N = 2065)
 NMean (SD)Median (Q1–Q3)Min–Max% at floor% at ceiling
  1. Scores calculated using weighted data.

Total20654.6 (4.5)4 (1–7)0–25200
 Female11075.1 (4.5)4 (1–7)0–25150
 Male9584.2 (4.4)3 (1–6)0–22250
Age group      
 18–19514.8 (4.2)5 (3–7)0–13390
 20–292584.3 (4.6)3 (1–6)0–22290
 30–393954.3 (4.5)3 (1–6)0–24240
 40–493454.4 (4.7)3 (1–7)0–25310
 50–593654.8 (4.3)4 (1–7)0–21220
 60–693104.5 (4.6)3 (1–6)0–23250
 70–792385.4 (4.3)5 (1–8)0–22120
 ≥801036.1 (4.4)5 (2–9)0–21120

Figure 1. Cumulative distribution curves for the QoL-AGHDA score (N = 2065).

Download figure to PowerPoint

Analysis of the QoL-AGHDA Data

The mean percentage of missing items per respondent was 0.6%. Less than 1.0% of missing data was observed per item. The QoL-AGHDA score had good internal consistency reliability with a Cronbach's alpha of 0.86. The Rasch analysis showed that items 6 (problems with doing simple tasks), 18 (feeling burdensome to people), and 24 (problems staying awake) were found to be the most “difficult” items (Fig. 2). As the probability of declaring an impairment on these items was low compared with the other items, the respondents who answered “yes” to these three items may have an impaired HRQoL and probably answered “yes” to all the other QoL-AGHDA items considered less “difficult.” Conversely, items 9 (lacking confidence), 11 (feeling very tense), and 15 (feeling very low) were found to be the least “difficult” items, indicating that the respondents frequently answered “yes” to these items regardless of their level of HRQoL impairment. The person–item map resulting from the Rasch analysis also showed that the majority of the respondents were not impaired in terms of HRQoL because a large majority of them was located on the map at a lower level than the least “difficult” items.


Figure 2. Person–item map resulting from Rasch analysis (N = 2065). aAn “X” represents 11 respondents. bAn item is considered “difficult” if it is “difficult” to answer positively to this item, i.e., if a few respondents answered “yes” to this item.

Download figure to PowerPoint

The females and the males reported a significantly different QoL-AGHDA mean score (P < 0.001), with poorer HRQoL reported for the females (5.1 ± SD = 4.7) than for the males (4.3 ± 4.4). The respondents of the different age groups also reported a significantly different QoL-AGHDA mean score (P < 0.001), with poorer HRQoL for the older respondents (5.3 ± 4.7 for 70–79 years and 6.1 ± 4.8 for ≥80 years) than the younger respondents (between 4.3 ± 4.2 for 20–29 years and 4.9 ± 3.6 for 18–19 years). A significant difference in QoL-AGHDA mean score was observed in relation to the respondents' overall health status (Fig. 3), with HRQoL decreasing when overall health status deteriorated (P < 0.0001). The QoL-AGHDA score was negatively correlated with both the EQ-5D index and VAS (−0.59 and −0.55, respectively), i.e., the better the health state, the lower the QoL-AGHDA score.


Figure 3. Mean QoL-AGHDA score (SD) according to overall health status (N = 2065). aAnalysis of variance. Note: scores calculated using nonweighted data.

Download figure to PowerPoint


  1. Top of page
  3. Introduction
  4. Methods
  5. Results
  6. Conclusions
  7. References

This study collected normative reference data for the QoL-AGHDA score in the French population. For all the French respondents, a mean QoL-AGHDA score of 4.6 was observed, with 20% of the respondents reporting the lowest possible score (i.e., 0). This result was similar to the normative reference data collected in the general populations of England and Wales, Spain, The Netherlands, and Sweden that reported mean QoL-AGHDA scores of 6.7, 5.0, 5.0 and 3.9, respectively [15]. The French female respondents reported a slightly higher QoL-AGHDA score than the males, i.e., a slightly poorer level of HRQoL, as observed for the other countries, with the exception of Spain [15]. As expected, the youngest French respondents reported a slightly better level of HRQoL than the oldest respondents, as HRQoL can be impacted by various age-related conditions.

Even though the study questionnaire was quite long (eight pages), it was well accepted by French panelists, with a return rate of 75%. The respondents were different from the nonrespondents because they tended to be older and more often, they were female, retired, married or widowed, and less educated. The respondents also differed from the French general population as they tended to be older, and more often, they were retired. This result may be explained by the length and purpose of the study questionnaire. Indeed, the older panelists, who are more likely to be retired than the younger panelists, might have more time to complete the questionnaire and be happier to be asked about their HRQoL than the younger panelists. Therefore the older panelists may have been more interested in completing the questionnaire. One could thus question the generalizability of the results and their interpretation as the nonresponse may lead to a bias in estimates of QoL-AGHDA reference values. In the present study, however, data were weighted for the establishment of the reference values to better reflect the French general population and to control for potential nonresponse bias by giving a lower weight for categories that were more represented in the study than in the general population and a higher weight for categories that were less represented. This method is often used in survey research and has proved to be generally beneficial. Our study also reports QoL-AGHDA scores analyzed across sex and age groups using unweighted data, and similar results as those reported with weighted data were observed.

One could argue that QoL-AGHDA reference data collected in general populations are irrelevant because the content of the questionnaire is adapted to patients but not to healthy people. According to this argument, it might be more appropriate to use reference data collected in patients with other chronic diseases such as diabetes [26,27]. This would enable the confounding factor of coping with a chronic condition to be taken into account [27]. Other chronic diseases and their treatments, however, do not necessarily have a similar impact to GHD and GH treatment on patients' HRQoL. Problems measured by QoL-AGHDA items may not be necessarily the same as those experienced by patients with other chronic diseases, and therefore, no ideal control population exists [28].

In the field of GHD, there are several advantages to using reference values collected in general populations. First, general population normative data are collected in large representative samples, allowing reference data to be established for different subgroups such as sex or age, and therefore, permitting the comparisons between patients and normative data to be refined. Sex is of particular interest when measuring the HRQoL of GHD patients because significantly different QoL-AGHDA scores were reported between males and females patients with GHD [15,17]. Second, reference data collected in general populations can be used for economic evaluation. For example, general population-based utilities were derived for the QoL-AGHDA questionnaire [20,29] to calculate quality-adjusted life years and for use in the pharmacoeconomic evaluation of GH replacement in adults with GHD from Sweden and England and Wales.

It is important to note that it is preferable to collect normative data at a country level to take into account the differences between cultures. Indeed, culture can be a major source of variability in the measurement of nonobservable concepts such as HRQoL [30,31] because cross-cultural differences are not only due to the language but also to many other aspects of the social and cultural context of each country. This highlights the importance of completing the collection of the QoL-AGHDA normative database at the country level.

This study also provided evidence of the good psychometric properties of the QoL-AGHDA questionnaire in the French general population. As expected, the QoL-AGHDA questionnaire produced skewed scores toward higher levels of HRQoL because it was not developed for use in general populations. This was shown by the Rasch analysis as well as by the high percentage of respondents at the lowest possible score. Nevertheless, the quality of completion of the QoL-AGHDA questionnaire, as well as the internal consistency reliability, was very good in this French sample. In addition, the clinical validity of the QoL-AGHDA questionnaire was demonstrated as its score decreased, reflecting improving HRQoL, when overall health status improved. These results showed that even though the questionnaire was not intended for use in general populations, the psychometric properties of the QoL-AGHDA questionnaire were satisfactory when administered in the French general population. This is not surprising as some QoL-AGHDA items refer to problems that are not necessarily GHD-related (e.g., lacking confidence, feeling tense, avoiding responsibilities). By contrast, the most “difficult” items (problems with doing simple tasks, feeling burdensome to people, and difficulties staying awake) were not frequently reported in the general population. These results are similar to those reported in the Swedish population [21], with the exception of lacking confidence, which was not characterized as an “easy” item by the respondents in the Swedish general population.

In conclusion, the disease-specific QoL-AGHDA questionnaire showed good psychometric results when completed by the respondents of the French general population, supporting its use in a general population. Reference values for the QoL-AGHDA questionnaire are now available for the French general population, completing the QoL-AGHDA normative database already available in a number of European countries. This database has previously been used to assess the impact of GHD on patients' HRQoL (mean QoL-AGHDA score in general population vs. GHD patients was 6.7 vs. 15.8 in England and Wales, 5.0 vs. 12.1 in Spain, 5.0 vs. 9.9 in the Netherlands and 3.9 vs. 8.1 in Sweden [15]), and our study makes reference values available for future such comparison with QoL-AGHDA data collected in French GHD patients.

Source of financial support: This work was funded by Pfizer. This work was funded by Pfizer. Hélène Gilet, Muriel Viala-Danten and Sébastien Auzière are paid consultants to Pfizer. Maria Koltowska-Häggström and Anne Chachuat are employees of Pfizer. The manuscript has been reviewed for its English by Sara Strzok who received a fee from Pfizer.


  1. Top of page
  3. Introduction
  4. Methods
  5. Results
  6. Conclusions
  7. References
  • 1
    Carroll PV, Christ ER, Bengtsson BA, et al. Growth hormone deficiency in adulthood and the effects of growth hormone replacement: a review. Growth hormone research society scientific committee. J Clin Endocrinol Metab 1998;83:38295.
  • 2
    Bjork S, Jonsson B, Westphal O, et al. Quality of life of adults with growth hormone deficiency: a controlled study. Acta Paediatr Scand 1989;356(Suppl.):S559.
  • 3
    Rikken B, Van Busschbach J, Le Cessie S, et al. Impaired social status of growth hormone deficient adults as compared to controls with short or normal stature. Dutch growth hormone working group. Clin Endocrinol (Oxf) 1995;43:20511.
  • 4
    Rosen T, Wiren L, Wilhelmsen L, et al. Decreased psychological well-being in adult patients with growth hormone deficiency. Clin Endocrinol (Oxf) 1994;40:11116.
  • 5
    Deijen JB, Arwert LI, Witlox J, et al. Differential effect sizes of growth hormone replacement on quality of life, well-being and health status in growth hormone deficient patients: a meta-analysis. Health Qual Life Outcomes 2005;3:639.
  • 6
    Radcliffe DJ, Pliskin JS, Silvers JB, et al. Growth hormone therapy and quality of life in adults and children. Pharmacoeconomics 2004;22:499524.
  • 7
    Woodhouse LJ, Mukherjee A, Shalet SM, et al. The influence of growth hormone status on physical impairments, functional limitations, and health-related quality of life in adults. Endocr Rev 2006;27:287317.
  • 8
    Malik IA, Foy P, Wallymahmed M, et al. Assessment of quality of life in adults receiving long-term growth hormone replacement compared to control subjects. Clin Endocrinol (Oxf) 2003;59:7581.
  • 9
    McGauley GA. Quality of life assessment before and after growth hormone treatment in adults with growth hormone deficiency. Acta Paediatr Scand 1989;356(Suppl.):S702.
  • 10
    Jenkinson C, Coulter A, Wright L. Short form 36 (SF36) health survey questionnaire: normative data for adults of working age. BMJ 1993;306:143740.
  • 11
    Badia X, Schiaffino A, Alonso J, et al. Using the EuroQoI 5-D in the Catalan general population: feasibility and construct validity. Qual Life Res 1998;7:31122.
  • 12
    Michelson H, Bolund C, Nilsson B, et al. Health-related quality of life measured by the EORTC QLQ-C30––reference values from a large sample of Swedish population. Acta Oncol 2000;39:47784.
  • 13
    Schwarz R, Hinz A. Reference data for the quality of life questionnaire EORTC QLQ-C30 in the general German population. Eur J Cancer 2001;37:134551.
  • 14
    McKenna SP, Doward LC, Alonso J, et al. The QoL-AGHDA: an instrument for the assessment of quality of life in adults with growth hormone deficiency. Qual Life Res 1999;8:37383.
  • 15
    Koltowska-Häggström M, Mattsson AF, Monson JP, et al. Does long-term GH replacement therapy in hypopituitary adults with GH deficiency normalise quality of life? Eur J Endocrinol 2006;155:10919.
  • 16
    Murray RD, Skillicorn CJ, Howell SJ, et al. Influences on quality of life in GH deficient adults and their effect on response to treatment. Clin Endocrinol (Oxf) 1999;51:56573.
  • 17
    Svensson J, Mattsson A, Rosen T, et al. Three-years of growth hormone (GH) replacement therapy in GH-deficient adults: effects on quality of life, patient-reported outcomes and healthcare consumption. Growth Horm IGF Res 2004;14:20715.
  • 18
    Koltowska-Häggström M, Hennessy S, Mattsson AF, et al. Quality of life assessment of growth hormone deficiency in adults (QoL-AGHDA): comparison of normative reference data for the general population of England and Wales with results for adult hypopituitary patients with growth hormone deficiency. Horm Res 2005;64:4654.
  • 19
    Badia X, Lucast A, Sanmarti A, et al. One-year follow-up of quality of life in adults with untreated growth hormone deficiency. Clin Endocrinol (Oxf) 1998;49:76571.
  • 20
    Koltowska-Häggström M, Jonsson B, Isacson D, et al. Using EQ-5D to derive general population-based utilities for the quality of life assessment of growth hormone deficiency in adults (QoL-AGHDA). Value Health 2007;10:7381.
  • 21
    Wiren L, Whalley D, McKenna S, et al. Application of a disease-specific, quality-of-life measure (QoL-AGHDA) in growth hormone-deficient adults and a random population sample in Sweden: validation of the measure by Rasch analysis. Clin Endocrinol (Oxf) 2000;52:14352.
  • 22
    The EuroQol Group. EuroQol––a new facility for the measurement of health-related quality of life. Health Policy 1990;16:199208.
  • 23
    Swaine-Verdier A, Doward LC, Hagell P, et al. Adapting quality of life instruments. Value Health 2004;7(Suppl. 1):S2730.
  • 24
    Leplege A, Ecosse E, Archambau F, et al. Pilot study and preliminary validation of the French version of a quality-of-life questionnaire specific for subjects with growth hormone deficiency. Ann Endocrinol (Paris) 2003;64:1917.
  • 25
    Greiner W, Weijnen T, Nieuwenhuizen M, et al. A single European currency for EQ-5D health states. Results from a six-country study. Eur J Health Econ 2003;4:22231.
  • 26
    Lynch S, Merson S, Beshyah SA, et al. Psychiatric morbidity in adults with hypopituitarism. J R Soc Med 1994;87:4457.
  • 27
    Wallymahmed ME, Foy P, MacFarlane IA. The quality of life of adults with growth hormone deficiency: comparison with diabetic patients and control subjects. Clin Endocrinol (Oxf) 1999;51:3338.
  • 28
    Murray RD, Shalet SM. The use of self-rating questionnaires as a quantitative measure of quality of life in adult growth hormone deficiency. J Endocrinol Invest 1999;22(Suppl. 5):S11826.
  • 29
    Koltowska-Häggström M, Kind P, Monson JP, et al. Growth hormone (GH) replacement in hypopituitary adults with GH deficiency evaluated by a utility-weighted quality of life index: a precursor to cost–utility analysis. Clin Endocrinol (Oxf) 2008;68:1229.
  • 30
    Berry JW, Poortinga YH, Segall MH, et al. Cross-Cultural Psychology––Research and Applications. Cambridge: The Press Syndicate of the University of Cambridge, 1992.
  • 31
    Ember CR, Ember M. Cross-Cultural Research Methods. Lanham, MD: Rowman & Littlefield Publishers, 2001.