| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Lilly Research Laboratories, Eli Lilly & Co. (W.F.B., D.J.E., M.R., M.L.H., F.M., D.V., A.F.A.), Indianapolis, Indiana 46285; Pharma Support, Inc. (E.P.S.), 191119 St. Petersburg, Russia; Department of Internal Medicine, Erasmus University Medical Center (A.J.v.d.L.), 3015 GD Rotterdam, The Netherlands; Medizinische Klinik Innenstadt (C.J.S.), 80336 Munich, Germany; and Institut und Poliklinik für Psychosomatische Medizin, Medizinische Psychologie und Psychotherapie (G.H., P.H.), 81675 Munich, Germany
Address all correspondence and requests for reprints to: Dr. Werner F. Blum, Eli Lilly & Co., Saalburgstrasse 153, D-61350 Bad Homburg, Germany. E-mail: blum_werner{at}lilly.com.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
We have previously described the development and the validation of a new tool, the Questions on Life Satisfaction (QLSM) (13, 14, 15). This questionnaire is concise, easy to complete, and can be applied in a wide variety of settings. The full version of this instrument comprises several modules, one of which, the QLS-H, is purposely designed to address the complaints of patients with hypopituitarism. Psychometric evaluation across various countries has demonstrated the QLS-H to be a valid, reliable, and sensitive tool. It is effective in assessing impaired life satisfaction in adults with GHD and for monitoring the efficacy of GH therapy in improving QoL (14). Another instrument designed for use in GHD patients, the Quality of Life Assessment of GH Deficiency in Adults (QoL-AGHDA), has been developed based on the Nottingham Health Profile and has been validated in several countries (16, 17). Compared with the QoL-AGHDA, the QLS-H has an added advantage of weighting individual items of functioning according to the degree of importance an individual subjectively places upon them. To date, normal reference ranges for the AGHDA have only been established for Spain (18) and Sweden (19) and are not currently available for the QLS-H. These data, however, are essential to determine whether patients have impaired QoL at baseline and particularly whether their QoL normalizes upon GH therapy. Age, gender, and country differences may be confounding factors that would prevent pooling of data from large databases.
The objective of this study was to establish normal reference ranges for the QLS-H in different countries and to examine the main influencing demographic variables that impact upon QoL. In this way we wished to develop a method that would allow the pooling of data from diverse populations. In parallel, we wanted to evaluate whether adults with GHD had an impaired QoL and whether it normalized during replacement therapy.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Samples of adults representative of the general population were studied in various countries between 1995 and 2002. Data collection was performed by experienced social research institutes through trained interviewers who explained and provided the questionnaire. Selection of interviewees and randomization schemes, although different in the various countries, followed accepted standard procedures that are commonly used in market research to ensure that samples were representative. Age ranges were predefined, and subject selection was aimed at mirroring national censuses with respect to geographical region, age, gender, occupation, and marital status. Response rates were good in all countries, ranging from 77% (Netherlands) to 93% (Germany and Italy). Surveys were conducted in the following countries: France (n = 989; 485 males and 504 females), Germany (n = 1867; 865 males and 1002 females), Italy (n = 977; 456 males and 521 females), The Netherlands (n = 1075; 519 males and 556 females), Spain (n = 896; 422 males and 474 females), the United Kingdom (n = 1090; 518 males and 572 females), and the United States (n = 1283; 612 males and 671 females). This represents a total of 8177 subjects (3877 males and 4300 females).
Patient population
Adults with GHD from five different clinical trials comprising a total of 957 patients were included in an analysis of their QLS-H scores. These studies were all conducted according to International Conference on Harmonization Good Clinical Practice standards. B9R-SB-GDDR and B9R-UT-GDDL were conducted in Germany and The Netherlands, respectively. They had identical protocols designed to investigate the effect of GH replacement on body composition and identify predictors of treatment efficacy. B9R-EW-GDED (20) and B9R-EW-T002 (21) were open-label comparisons of alternate dose regimens and included patients from Italy and the United Kingdom (B9R-EW-GDED) and from Germany, Italy, France, the United Kingdom, and United States (B9R-EW-T002). QLS-H scores were obtained as a secondary clinical end point in these studies. In Spain a separate study was conducted to validate the QLS-H questionnaire in GHD adults prescribed GH for replacement therapy; this study was conducted according to standard clinical trial processes. Although the primary objectives of the trials were different, the overall designs were similar, particularly with respect to the assessment of QoL at baseline and after 6 or 8 months of GH therapy (20). After 6 or 8 months of GH therapy, QLS-H scores were available from 782 patients; 175 patients discontinued from the clinical trials for a variety of reasons, such as adverse events, patient decision, and loss of follow-up. Pooling of the scores from the QLS-H data from the different trials appeared, therefore, to be straightforward. The diagnosis of GHD in these patients was based primarily upon published guidelines provided by the Growth Hormone Research Society (22). Specifically, a low response to a standard GH stimulation test (insulin tolerance test, arginine, GHRH, glucagon, or combined arginine plus GHRH) was required, with a peak serum GH level less than 3.0 µg/liter used for all tests with the exception of the combined arginine plus GHRH test, for which a peak GH level less than 9.0 µg/liter was considered to be a low response based on recommendations in place when the study protocols were designed (23). Replacement of other hormone deficiencies was adequate and stable for at least 3 months before enrollment in the clinical trials. Patients gave written informed consent for their participation, and drug accountability was monitored throughout the trials. The main demographic and clinical characteristics of the patients are listed in Table 1
.
|
A preliminary German language version of the QLS-H was developed from semistructured interviews of adult patients with GHD. The questionnaire was translated into six languages (French, Italian, Dutch, Spanish, British English, and American English), according to a strict protocol. This involved translation from German into each language and back-translation to German by an independent translator. Any language issues (e.g. semantic nonequivalence) that arose were resolved by the first translator and the psychologists who developed the instrument (P. Herschbach and G. Henrich). The questionnaire was then validated in open, noncontrolled, multicenter, longitudinal, prospective studies cross-nationally (14, 15). The final version (QLS-H) comprises a single page of nine items (Fig. 1) focusing upon the specific problems experienced by adults with GHD. The QLS-H questionnaire is self-administered. Subjects must initially indicate how important a certain dimension of QoL is to them and are then questioned as to their degree of satisfaction with that dimension. Effectively, this allows each item to be individually weighted in terms of importance by the patients themselves. Each item is rated on a five-point Likert scale ranging from not important (1) to extremely important (5) and from dissatisfied (1) to very satisfied (5). The weighted score for the degree of satisfaction (weighted satisfaction) with a particular dimension of quality of life is then calculated by the formula (13): weighted satisfaction = (importance - 1) x (2 x satisfaction - 5). The total QLS-H score is subsequently obtained by adding the individual item scores of the nine dimensions and can range from -108 (representing very low satisfaction) to +180 (representing very high satisfaction) (14).
|
Normal ranges of total QLS-H scores were constructed separately for gender in each country, using age as a continuous independent variable. Datasets were tested for normality of distribution using the Shapiro-Wilke W test statistic and visual examination of QQ-normal plots.
Different transformations were used to normalize the data distribution. Logarithmic transformation of the scores (log[score + constant]) was necessary with the British, French, and Spanish samples of either gender and with the German, Dutch, and American male samples before further processing of the data. The dataset of German females was normalized by a power transformation ([score + 150]0.5). Polynomial curves were fitted by backward stepwise regression, starting with a polynomial of degree 4 and reducing the degree until a significant deterioration of the fit occurred (24, 25). The fit of regression models was tested by analysis of residuals. The polynomial curve with the lowest degree and significant polynomial regression coefficients was used for approximation of the underlying trend in the normative data, mean as a function of age [mean(age)].
If the SD of the normative data were approximately constant across ages, it was assessed as the SD of residuals from the regression model. Age-dependent SD values as a function of age [SD(age)] were estimated as linear fits of absolute residuals by age with multiplication of equation coefficients by 1.25 as proposed by Altman (26). Altman’s method uses the properties of the half normal distribution (27). The mean of the absolute residuals multiplied by 
1.25 is an estimate of the SD of the residuals. The age dependence of SDs was observed in French females and Spanish males. Reference ranges were defined as lower/upper limits = mean(age) ± 1.96 x SD(age) of the transformed QLS-H scores comprising 95% of the data points. For curves that were retransformed to absolute QLS-H scores for age, deviation scales (DV) were defined as the difference between mean(age) and lower/upper limit divided by 1.96 for a given age. The z-scores are defined by the following equation: z-score = [QLS-H score - mean(age)]/DV(age) for a given general population. DV = SD for datasets that did not require transformation to obtain normal distribution.
For both normative and patient data, comparisons among countries and between genders were performed using ANOVA models for raw and rank-transformed data. Multiple comparisons among countries were performed with the Sidak test, and 95% confidence intervals were calculated to detect whether patient z-scores significantly differed from zero. For pairwise comparisons between countries the least squares means (LS-means) were calculated using the general linear model with country effects. LS-means are means appropriately adjusted for the country effect in the model (called "estimated population marginal means"). Unadjusted pairwise comparisons (or repeated t tests) were used for testing differences in LS-means between countries to illustrate that any individual country could be different from any other individual country. The significance of changes from baseline was tested by paired two-tailed t test or Wilcoxon sign test as appropriate. P < 0.05 was considered statistically significant.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
A preliminary analysis of normative QLS-H data obtained from representative samples of the general populations of seven countries (France, Germany, Italy, The Netherlands, Spain, the United Kingdom, and the United States) indicated that QLS-H scores are influenced by age, gender, and country. Therefore, in the final analysis of these data, stratification was performed according to country and gender, and age was considered an independent continuous variable. Figure 2 illustrates the scatter plots of the absolute QLS-H scores plotted against age in males and females in the various countries and provides an impression of the age range of studied subjects. The fitted lines represent the mean together with the upper and lower limits, which contain 95% of the data points. Although there was significant overlap of the reference ranges, patterns of QLS-H scores in the general populations varied among countries and between genders, with females scoring lower than males. The differences between countries became particularly evident on direct comparison of the means, as shown in Fig. 3. In addition, in the majority of countries there was a clear age dependence (Fig. 3), with a decline of QLS-H scores with age, although the pattern varied by country.
|
|
). In addition, when more limited age ranges were examined, significant differences between countries became obvious in countries that otherwise had comparable normal ranges across all age groups. Overall there was up to a 2-fold difference in the absolute QLS-H scores between the lowest and highest scoring countries in subjects under the age of 50 yr. In older individuals, the difference between low scoring countries, which showed a strong age dependence (e.g. Germany or France), and the highest scoring countries (e.g. United States) became even more pronounced. Patients with adult GHD
Absolute QLS-H scores.
Table 2 and Fig. 4 (dark grey bars) illustrate the absolute QLS-H scores of male and female patients with GHD in the various countries at baseline. Male patients had consistently higher scores at baseline than female patients, a difference that was also observed among the general populations. These gender differences were statistically significant in Germany, France, and Italy. After 6 or 8 months of GH treatment, absolute QLS-H scores increased significantly in all groups (Fig. 4, white bars), and gender differences largely disappeared (with the exception of Germany; P = 0.018). Tables 3 and 4 show pairwise comparisons between countries for male patients (Table 3A) and female patients (Table 4A) at baseline. Italian male patients scored significantly higher than the patients in all other countries except the United States. This was mostly the same with Italian female patients, except for Spain and the United States American male patients scored higher than the patients in Germany, The Netherlands, and the United Kingdom, whereas American female patients scored higher than those in France, Germany, and the United Kingdom. In addition, British male patients scored lower than the patients in every other country. British female patients scored lower than the patients in Italy and Spain. For these reasons, it appeared necessary to standardize QLS-H absolute scores into a form that would allow pooling of data across countries and gender.
|
|
|
|
illustrates the z-scores of males and females in the various countries at baseline and after 6 or 8 months of GH treatment. Before GH therapy, z-scores of patients were significantly lower than in the general population (P < 0.001) in all countries and for both males and females, although differences between the genders had been eliminated (Table 2). In addition, differences between countries had largely disappeared with the exception of the UK (Tables 3B and 4B). The British patients still had significantly lower z-scores than the patients in most other countries regardless of gender. After GH therapy, z-scores increased significantly in all countries in both males and females, with mean z-scores scattered around the zero line (Fig. 5). None of the mean z-scores after GH treatment were significantly different from zero, with the exception of British males. This observation indicates that QoL had improved in the patients with adult GHD to levels comparable with their corresponding general populations.
|
Neither absolute QLS-H scores nor QLS-H z-scores showed any correlation with serum IGF-I levels (absolute or SD scores) at baseline or change in IGF-I levels during treatment. Although patients with childhood-onset GHD had higher mean z-scores at baseline than patients with adult-onset GHD, this difference was not significant (-0.89 vs. -1.02). However, the change from baseline in childhood-onset GHD patients was significantly smaller than that in adult-onset GHD patients (0.62 vs. 0.95; P = 0.034).
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Normal ranges
Establishing reference ranges will determine whether the QoL of a patient is impaired relative to that of a general population and whether it normalizes after therapy. Therefore, we set out to establish reference ranges for QLS-H scores.
To establish the influence of possible confounding variables, we investigated representative samples of individuals with balanced numbers of males and females across broad age ranges from a variety of countries. Results confirmed our preliminary findings, which indicated that QoL decreases with increasing age. This trend may not be surprising given that the majority of items in the QLS-H also mirror the changes that occur with aging. However, the extent of this age dependence varied among different countries. For example, there was considerable QoL impairment with increasing age in France, but only a weak trend was observed in the United Kingdom and United States populations. It should be noted, however, that few subjects older than 75 yr were included in these representative samples, so caution should be exercised when extrapolating normal data beyond about 70 yr. A similar age-dependent deterioration in QoL has been demonstrated in a Spanish study that employed the AGHDA tool (18). However, in a Swedish study using the QoL-AGHDA applying the Rasch analysis, no significant differences between age groups were observed (19).
Analysis of the reference ranges revealed that females had lower QLS-H scores than males across all countries, although not across all age groups. This is a more surprising observation, given that the dimensions of the QLS-H were considered a priori to be gender nonspecific. However, it compares favorably with the observation in a large normal Swedish sample using the AGHDA. This study also found poorer QoL in females (19).
Large differences in scores were also observed among the samples of the general populations of the seven countries. For example, the means of QLS-H scores in France, Germany, The Netherlands, and Spain were comparable, particularly in individuals less than 50 yr of age. In contrast, QLS-H scores were significantly higher in the United Kingdom, Italy, and, in particular, the United States, with scores higher by up to 2-fold compared with those in Germany. This effect occurred despite the fact that the development and validation of the QLS-H questionnaire had been conducted using a parallel approach including patient data from all the countries (14). These differences may be attributable to two hypothetical explanations. First, despite the fact that translations followed stringent standardized procedures, focusing on semantic rather than linguistic issues, semantic equivalence of the questionnaires may not have been totally achieved. It is known that even subtle differences in culture-specific connotations of the wording of questions may significantly influence the answers (28, 29). Second, it is possible that true cultural differences exist among countries that may impact upon QLS-H scores. In support of the second hypothesis, other studies using alternative QoL questionnaires have demonstrated such cross-cultural differences (30, 31). However, whatever the reason for the differences among countries, possible semantic nonequivalence of the questionnaires or cultural differences, standardization to the normative data obtained with a given questionnaire should essentially eliminate the differences among patients from the various countries.
The relatively wide variance of the normative data in the United States may reflect cultural heterogeneity compared with the culturally more homogeneous populations in the various European countries. A fair question is, therefore, whether in countries with significantly divergent cultures, such as the United States, culture-specific, rather than country-specific, normal ranges should be used. Minimum denominators of culture could be ethnic origin, commonly used language, or, possibly, geographical regions within a large country such as the United States.
Consequently, normal ranges for QLS-H scores should account for this age dependence and should be stratified according to gender and country. Further stratification may also be required. For example, differences in educational, occupational, or marital status may impact significantly on QLS-H scores. These relationships are likely to be complex. As these questions were not included in the current analysis, a definitive conclusion cannot be drawn. However, additional stratification may have limitations with population sizes becoming too small for meaningful statistical analyses. Furthermore, there are practical limitations in the collection of the data that would be necessary to allow the calculation of a patient’s z-score with more finely stratified normal ranges. Therefore, we decided to limit stratification to gender and country and relate the QLS-H scores to age.
QLS-H in patients with adult GHD
From observations made in the general population, adult patients with GHD were stratified according to gender and country. However, for absolute QLS-H scores, no consideration was given to age, as mean ages across the cohorts were similar. The results clearly demonstrate that there are significant differences in QLS-H scores among patients in different countries at baseline and after treatment. The rationale for choosing pairwise comparisons of countries instead of comparing countries across the complete dataset, which would require adjusted P values, was a practical one. To illustrate: in future applications the question will probably be "can QLS-H scores of Italian patients be pooled with the scores of French patients (in a joint study) or is there a systematic difference?" rather than "is there a significant difference between Italian and French patients and the patients from all the other countries?" The highest mean QLS-H scores were found in the same countries that displayed the highest scores in the general population (i.e. Italy and the United States). Lower scores were observed in females in all countries, as would be expected from the stratified normal ranges. This was particularly evident at baseline. These observations show that simple pooling of absolute scores is not adequate. Pooling is only possible when values are adjusted according to age, country, and gender.
One solution to this problem is to calculate z-scores. The z-scores express the difference from the corresponding mean value in the general population by using deviation as a measure. In the simplest case and where scores are normally distributed in the reference population, the average is given by the mean, and the deviation by the SD. However, the distribution of QLS-H scores in some country- and gender-specific cohorts was not normal and therefore required appropriate transformation before further processing. In addition, the mean and the width of the distribution of the scores in some countries varied by age. It should be noted, however, that transformation of raw data into z-scores may introduce an additional source of error. For instance, a normal range may not adequately reflect the situation in the general population if sample selection was biased. Although DV values were similar in most countries, some countries, particularly the United States, showed higher values. This means that the same difference of an absolute QLS-H score from the mean in the United States yields a smaller z-score than, for example, in Germany.
A critical criterion for the suitability of z-score calculation based on the available normal ranges was, therefore, whether transforming absolute QLS-H scores to z-scores eliminates country- and gender-specific differences. Although this was generally the case both at baseline and after GH therapy, there were a few exceptions. Notably significant differences persisted in the z-scores between patients in the United Kingdom and those from the other countries even though baseline patient characteristics did not indicate major differences in study populations. There are several possible explanations for this. First, the sample used for establishing normal ranges may not have been truly representative for the United Kingdom population. However, standard procedures were used for the selection of samples to minimize this potential bias. Second, other confounding factors, such as socio-economic variables in the patient cohorts, may not be balanced. Finally, patient selection in the various countries may differ according to local clinical practice. The significant differences observed between patients’ z-scores in the United Kingdom and other countries may reflect more stringent selection criteria for GH treatment or inclusion in clinical trials. In a British study by Holmes and Shalet (32), it was reported that adults who enter a study of GH replacement therapy exhibit greater distress, which may be less so in other countries. Despite these exceptions, however, differences among countries and between genders were largely eliminated, suggesting that transforming absolute QLS-H scores to z-scores is a viable approach.
At baseline, patient z-scores were significantly below zero in all countries. Therefore, the populations of adults with GHD as a whole in this study have impaired QoL as evaluated by the QLS-H. This is consistent with observations made using generic QoL questionnaires with matched controls (2, 4, 5, 8) or with the AGHDA tool (18, 19, 33, 34). However, in some studies using generic tools, more equivocal results were reported (35, 36). After a 6- to 8-month period of GH therapy, mean z-scores were scattered around zero, indicating that, on the average, the QoL was no longer different from that of the general population in these patients. The change from baseline in all patients was almost 1 DV (comparable to 1 SD), which was statistically highly significant and also appears to be clinically relevant. This improvement in QoL after a period of GH therapy is in accordance with reports from studies in which generic or other questionnaires designed for use in GHD patients were applied (2, 8, 34, 37, 38, 39). Although the present study did not include an untreated control group, several previous studies have documented significant improvements in QoL with GH replacement therapy compared with placebo (3, 6). It is important to note that QoL may not improve with GH therapy in all patients with adult GHD. Using the Psychological General Well-Being Schedule questionnaire, Murray and co-workers (38) demonstrated that patients with greater impairment of QoL at baseline had a greater improvement in QoL with GH therapy. However, patients may experience several metabolic benefits of GH therapy regardless of the patient’s perception of QoL (1).
In conclusion, QLS-H scores from adults with GHD need to be adjusted for age, country, and gender to enable pooling of data or to compare results from different groups of patients. This can be achieved by establishing reliable, normal ranges from a sufficiently large representative population and deriving equations to calculate z-scores. The z-scores also provide an immediate indication of whether an individual patient’s QoL is compromised and whether it normalizes during GH replacement therapy. This study provides, for the first time, an international approach to resolve many of the issues that surround health-related QoL evaluation in adult patients with GHD. In this context, we can envisage that the QLS-H will become a useful and practical tool in evaluating QoL in these patients and their response to GH therapy.
|
Acknowledgments |
|---|
Data collection for the normative samples from general populations was performed by the following social research institutes: Societé NAXIS (France), Taylor Nelson Sofres EMNID (Germany), DOXA Institute and Mario Negri Institute (Italy), MarketResponse Netherlands (The Netherlands), ECO Consulting S.A. (Spain), Taylor Nelson Sofres (United Kingdom), and Strategic Marketing Associates (United States).
|
Footnotes |
|---|
Abbreviations: AGHDA, Assessment of GH Deficiency in Adults; DV, deviation scale; GHD, GH deficiency; LS, least squares; QLS-H, Questions on Life Satisfaction Hypopituitarism Module; QoL, quality of life.
Received November 14, 2002.
Accepted May 19, 2003.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  O. M. Dekkers, A. A. van der Klaauw, A. M. Pereira, N. R. Biermasz, P. J. Honkoop, F. Roelfsema, J. W. A. Smit, and J. A. Romijn Quality of Life Is Decreased after Treatment for Nonfunctioning Pituitary Macroadenoma J. Clin. Endocrinol. Metab., September 1, 2006; 91(9): 3364 - 3369. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Saller, A. F Mattsson, P. H Kann, H. P Koppeschaar, J. Svensson, M. Pompen, and M. Koltowska-Haggstrom Healthcare utilization, quality of life and patient-reported outcomes during two years of GH replacement therapy in GH-deficient adults - comparison between Sweden, The Netherlands and Germany. Eur. J. Endocrinol., June 1, 2006; 154(6): 843 - 850. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. J. Woodhouse, A. Mukherjee, S. M. Shalet, and S. Ezzat The Influence of Growth Hormone Status on Physical Impairments, Functional Limitations, and Health-Related Quality of Life in Adults Endocr. Rev., May 1, 2006; 27(3): 287 - 317. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. D Veldhuis, J. M Patrie, K. Frick, J. Y Weltman, and A. L Weltman Administration of recombinant human GHRH-1,44-amide for 3 months reduces abdominal visceral fat mass and increases physical performance measures in postmenopausal women Eur. J. Endocrinol., November 1, 2005; 153(5): 669 - 677. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. F. Attanasio, E. P. Shavrikova, W. F. Blum, and S. M. Shalet Quality of Life in Childhood Onset Growth Hormone-Deficient Patients in the Transition Phase from Childhood to Adulthood J. Clin. Endocrinol. Metab., August 1, 2005; 90(8): 4525 - 4529. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. D. Veldhuis, J. Frystyk, A. Iranmanesh, and H. Orskov Testosterone and Estradiol Regulate Free Insulin-Like Growth Factor I (IGF-I), IGF Binding Protein 1 (IGFBP-1), and Dimeric IGF-I/IGFBP-1 Concentrations J. Clin. Endocrinol. Metab., May 1, 2005; 90(5): 2941 - 2947. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Mukherjee, S. Tolhurst-Cleaver, W. D. J. Ryder, L. Smethurst, and S. M. Shalet The Characteristics of Quality of Life Impairment in Adult Growth Hormone (GH)-Deficient Survivors of Cancer and Their Response to GH Replacement Therapy J. Clin. Endocrinol. Metab., March 1, 2005; 90(3): 1542 - 1549. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. R. Hoffman, C. J. Strasburger, A. Zagar, W. F. Blum, A. Kehely, and M. L. Hartman Efficacy and Tolerability of an Individualized Dosing Regimen for Adult Growth Hormone Replacement Therapy in Comparison with Fixed Body Weight-Based Dosing J. Clin. Endocrinol. Metab., July 1, 2004; 89(7): 3224 - 3233. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Rosilio, W. F. Blum, D. J. Edwards, E. P. Shavrikova, D. Valle, S. W. J. Lamberts, E. M. Erfurth, S. M. Webb, R. J. Ross, K. Chihara, et al. Long-Term Improvement of Quality of Life During Growth Hormone (GH) Replacement Therapy in Adults with GH Deficiency, as Measured by Questions on Life Satisfaction-Hypopituitarism (QLS-H) J. Clin. Endocrinol. Metab., April 1, 2004; 89(4): 1684 - 1693. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
