Clinical assessment and validation of an Arabic Aging Male Symptoms questionnaire in patients with androgen deficiency

Abstract

Aim. To develop and to validate an Arabic Aging Male Symptoms (AMS) tool and to clinically assess patients with hypoganadism after hormonal treatment.

Methods. The tool was translated into Arabic and tested on 15 Saudi men. During a period of 9 months all males presented to the andrology clinic of the main University Hospital, King Saud University, Saudi Arabia with signs and symptoms of hypogonadism, were included in the study. Arabic AMS scale was applied in the base line visit, then 12 weeks after treatment. Testosterone was monitored before treatment, 4 weeks and after 12 weeks.

Results. Ninety-two subjects were included, Cronbach's α of 0.91 showed a very good internal consistency of the Arabic AMS questionnaire. The corresponding α for the subscales were 0.83, 0.84 and 0.73. There was a significant improvement in the mean level of TT after hormonal therapy (HT), this was reflected on the mean differences of improvement in the total Arabic AMS scores and subscales scores after HT, ranged from 31 to 35%.

Conclusion. The present study revealed a significant association between testosterone levels and AMS tool manifested by a its good ability to measure the effect of treatment on quality of life for patients with hypogonadism.

Keywords:

• AMS scale
• Arabic
• hypogonadism

Abbreviations
BT	=	bioavailable testosterone
HT	=	hormonal treatment
TT	=	serum total testosterone
QoL	=	quality of life

Introduction

With the increase in the life expectancy worldwide, there is a higher demand for a good quality of life (QoL) despite advancing age. Aging is a complex process, influenced by numerous factors. In men, age-associated testosterone decline is an important factor that impairs the QoL [1]. After the age of 40 years, the total serum testosterone (TT) levels decrease in men, especially the so called serum bioavailable testosterone (BT), whereas the serum levels of sex hormone binding globulin increase [2–4]. While these age-related changes do not affect fertility, these are responsible for various changes, collectively known as the Aging Male Symptoms (AMS). In individuals showing signs of androgen deficiency or hypogonadism, testosterone replacement therapy can significantly improve the QoL [5] in terms of physiological and psychological health. Currently, different forms of hormone therapy (HT) are available, such as the testosterone gel used here, which is easy to use, causes no irritation or side effects but is more expensive than the others [6].

Several questionnaires have been developed for assessing the effects of hypogonadism on QoL in men. One of these is the AMS questionnaire, which serves to evaluate the aging symptoms and for follow-up of hormonal therapy. It was developed in response to the lack of fully standardised scales to measure the severity of aging symptoms and their impact on QoL in men [7,8]. The AMS questionnaire consists of 17 questions, for which the responses are rated from 1 to 5 in terms of severity. Thus, a range of total scores between 17 and 85 is possible. The questions can be classified into psychological, somatic and sexual assessment, and the subscale scores to each of these can be analysed separately. A total score ≥27 is considered positive for age-related effects of androgen deficiency. The AMS scale was internationally accepted and translated from the original German version into 33 different languages, but not in Arabic. Hence, the purpose of the current study was to develop a reliable Arabic version of this tool to be used in Arabic-speaking populations and to clinically assess its efficacy to measure health-related complaints due to hypogonadism.

Methods and subjects

Development of the Arabic version of the AMS questionnaire

The AMS questionnaire was translated from English into Arabic by a bilingual professional translator. The Arabic version's content validity was assessed and amendments were made as recommended. This draft of the Arabic AMS questionnaire was then back-translated to English independently by three bilingual professionals. The three back-translated English versions were then assessed and compared with the original English version. The final version of the Arabic AMS questionnaire was administered to 15 Saudi Arabian men (mean age 51 ± 4.6 years) twice within an interval of about 10 days to assess the test–retest reliability.

Clinical application of the Arabic AMS

During a period of 9 months (January through September 2009), all males (age range 28–69 years) presenting to the andrology clinic of the main University Hospital, King Saud University, Riyadh, Saudi Arabia with signs and symptoms of hypogonadism, confirmed by low testosterone (TT) levels (less than 10.4 nmol/l or 300 ng/dl) were invited to participate in the study. The exclusion criteria included cancer, untreated DM, untreated hyperprolactinemia, high PSA levels (above 4 ng/dl). Apart from hypogonadism, the subjects were in good health as evidenced by medical history, physical examination and complete blood count. All participants who satisfied the selection criteria were enrolled in the study.

Age, anthropometric measurements, blood pressure, haemoglobin, total cholesterol, HDL and LDL cholesterol were recorded at the baseline visit. An interview questionnaire to record the socio-demographic data (age, education, and marital status) was administered by trained interviewers. The Arabic AMS questionnaire was filled by the participants at the baseline visit, and again after 12 weeks of HT.

Hormonal therapy (testosterone gel)

Testosterone treatment comprised of 1 g of 1% gel, equivalent to 10 mg testosterone, applied in the morning to the skin of the upper arm. Subjects were instructed to avoid showering for 5 h after application. Patients were monitored for treatment compliance which was estimated by counting the number of packets of the gel dispensed and those returned at the clinics during the patients' regular visits.

Serum testosterone measurements

An early morning non-fasting blood sample was collected from all participants to avoid diurnal variation in hormone levels. Total testosterone was measured in the initial visit to confirm hypogonadism, and again after 4 weeks and 12 weeks of HT.

Ethical aspects

The study was approved by the Ethics Committee in the main University Hospital. All participants signed a written informed consent.

Statistical analysis

Cronbach's scale reliability analyses was used to assess the internal consistency of the Arabic AMS questionnaire on the whole sample. The test–retest reliability of the Arabic AMS questionnaire was analysed by using the Pearson correlation coefficient of the two scores on five Saudi men, and the differences between the two scores were compared by the paired t test. The difference between the AMS questionnaire scores before and after the HRT was evaluated by paired t test. ANOVA was used to assess the difference between the mean levels of the three TT readings.

Results

Reliability of the Arabic AMS questionnaire

The test–retest reliability of the Arabic AMS questionnaire (n = 15) was good as indicated by the two-tailed Pearson correlation coefficient, r = 0.86, p < 0.001,) with no significant difference between the test and retest AMS scores (mean ± standard error of the mean 1.34 ± 0.34 vs. 1.61 ± 0.45; p = .33, paired t test, two-tailed). Similarly, an excellent internal consistency was indicated by the Cronbach's α value for the total AMS score was 0.91. The Cronbach's α values for the psychological, somatic and sexual subscales were 0.83, 0.84 and 0.73, respectively.

Clinical application of the Arabic AMS

Out of the 103 patients enrolled, 92 completed the study satisfactorily. The age range was from 28 to 69 years (34 years ± 5.3), all were married and most of them (89%) had a high level of education. The mean body mass index (BMI) was 29.4 ± 4.73.

Table I illustrates the significant improvement in the mean TT levels after HT during the course of the study. This was reflected on the mean differences in the total Arabic AMS scores as well as subscale scores, as shown in Table II. The mean AMS scores decreased significantly after HT. Notably, the BMI was not significantly correlated with the Arabic AMS questionnaire (data not shown). On the other hand, when age was stratified into four categories, it was observed that the severity of symptoms increased with advancing age as indicated by the higher total Arabic AMS scores. Thus, change in severity of impairment (difference between pre- and post- HT total score/ pre-HT score) was 0, 47, 52.2 and 87.2% in the corresponding age categories of <30 years, 30–45 years, 46–60 years and >60 years, respectively. When stratified age was tabulated against categorical improvement (Table III), the improvement in the patients' condition was more evident for those at an advanced age. It was observed that the per cent increase in improvement increased with age. Thus, among those who showed >23% improvement, the distribution was 2.9, 20.6, 27.9 and 48.5% for the age categories of <30 years, 31–45 years, 46–60 years and >60 years, respectively.

Table I.  Mean testosterone levels in the three readings.

Visit	Testosterone (nmol/l)	F = 29.5, significance = 0.000
Base Line	6.17 + 3.9
4 weeks	14.26 + 5.8
12 weeks	21.33 + 8.9

Table II.  Mean total AMS and sub-scores before and after hormonal therapy.

AMS scores	Before HRT	After HRT	% improvement	Paired t test	Significance
Total score	45.61 + 12.35	31.07 + 8.90	31.87	6.65	0.000
Somatic subscore	18.62 + 5.5	12.95 + 3.7	30.45	6.0	0.000
Psychological subscore	12.65 + 4.7	8.68 + 2.9	31.38	6.10	0.000
Sexual subscore	14.8 + 3.7	9.6 + 3.01	35.13	7.1	0.000

Table III.  Per cent improvement of symptoms severity by age.

Percentage range for improvement in AMS scores	Age range (years)
	<30	30–45	46–60	>60	Total
<15%	2 (15.4%)	2 (15.4%)	4 (30.7%)	5 (38.5%)	13
15–22%	0 (0%)	4 (36.3%)	2 (18.1%)	5 (45.6%)	11
>23%	2 (2.9%)	14 (20.6%)	19 (27.9%)	33 (48.5%)	68
Total	4 (4.3%)	20 (21.7%)	25 (27.1%)	43 (47.8%)	92

Discussion

Aging is a complex process and in men, is accompanied by lower levels of testosterone which affects their sexual functions and in turn their QoL. Several tools have been developed to screen for signs of androgen deficiency. Amongst these is the Androgen Deficiency in the Aging Male (ADAM) questionnaire which we recently translated into Arabic language and validated [9]. This study revealed a prevalence of hypogonadism in our community at 18.2% as detected by low serum BT levels and at 77.6% based on a positive ADAM questionnaire. Moreover, this prevalence was found to increase with age [9]. Results of this study highlighted the need for an Arabic tool for assessment and follow-up of HT in Arabic speaking populations.

The AMS scale is another popular tool, of interest to clinical researchers in measuring the health-related QoL (HRQoL) and associated symptoms in hypogonadal patients. In the present study, we have described the development of the Arabic version of this tool. The Arabic AMS was found to have very good content validity, test–retest reliability as well as an excellent internal consistency, reflected by a Cronbach's α values of 0.91 for the total score and 0.73–0.84 for the sub-scores. The analysis of the scores revealed a significant association between the TT levels and the AMS scores. There was a significant reduction in the total and the subscale scores after HT, reflected in noticeable improvement in the patients' condition, particularly in the sexual sub-scale. Previously, a Japanese researcher group [10,11] and a Polish researcher group [12] have reported a correlation between AMS scores and the TT levels, while some other studies did not find such an association in their study populations [13,14], likely due to cultural differences and/or differences in study design. An important outcome of the present study is that the Arabic AMS scale was able to mirror changes in the HRQoL. The results of Heinemann et al. [15,16] revealed that relative to the scores at baseline, the total scores and the psychological, somatic and sexual sub-scores improved by 30.7, 27.3, 30.5 and 30.7%, respectively, values comparable with our data (Table II).

An important finding reported by Kratzik et al. [17] in a population-based study in Vienna, was the impressive association between AMS subscales or free TT levels and age and BMI. In our study on an Arabic population, we did not find any correlation between the AMS scores and BMI, but age was an important correlate. Age was significantly associated with the severity of AMS impairment and with higher degrees of per cent improvement (more than 23%) in the symptoms after HT. In other words, older men, with higher initial AMS scores (and low TT) appeared to respond more favourably to the HT compared to younger men.

The AMS scale is not designed or standardised as a screening instrument for androgen deficiency. However, in conjunction with other validated tools and biochemical measurements, it may be possible to extend its use for mass screening in a population-based approach.

Conclusion

The present study shows an acceptable consistency, reliability and test re-test stability of the Arabic AMS tool. It showed a convincing ability to measure treatment effects on QoL across different levels of symptom severities. Nevertheless, for testing its validity (sensitivity and specificity) as a screening tool, mass screening of a community-based sample will be required.

Acknowledgments

The research has been funded by Princess Al-Johara Al-Ibrahim Center for Cancer Research, King Saud University, Riyadh, Saudi Arabia. We thank Dr. Anuradha Alahari of Accent Medical and Scientific Writing for help in editing the manuscript.