Comparison of questionnaires used for screening and symptom identification in hypogonadal men

Abstract

Late-onset hypogonadism (LOH) is typically defined as the cluster of symptoms appearing in aging men and accompanied by a decrease in serum testosterone levels. The identification of a simple screening tool with a high level of sensitivity and specificity to predict LOH has remained a challenge. To identify men with LOH, a variety of self-administered questionnaires have been developed including The Saint Louis University Androgen Deficiency in the Aging Male (ADAM) Questionnaire, The Quantitative ADAM (qADAM) Questionnaire, The Aging Male Symptoms (AMS) rating scale, The Massachusetts Male Aging Study (MMAS) questionnaire and The New England Research Institutes (NERI) hypogonadism questionnaire. The applicability of these questionnaires in the clinical setting is debated because some of the symptoms associated with LOH could be attributed to the natural process of aging and comorbidities. The goal of this review is to compare the utility and the validity of the different LOH questionnaires.

• Androgen deficiency
• Androgen deficiency of the aging male
• hypogonadism questionnaires
• late-onset hypogonadism

Introduction

Late-onset hypogonadism (LOH) typically is defined as the complex cluster of symptoms occurring most often in aging men and accompanied by a decrease in serum testosterone levels [1]. LOH is attributed to the 1–2% yearly decline in testosterone that is thought to begin around the age of 30 [2,3]. This decrease in testosterone can lead to a constellation of symptoms including decrease in muscle mass, increase in abdominal and visceral fat, decreased libido and sexual function, osteopenia, and decreased cognitive abilities [4–10]. Because many of these symptoms can be associated with the aging process alone, it can be difficult to determine whether symptoms are occurring as part of normal aging or are a result of LOH [11].

According to the Endocrine Society guidelines, the diagnosis of LOH should involve both the detection of low-serum testosterone and the identification of symptoms [12]. However, the development of a simple screening tool to predict symptoms of androgen deficiency has remained a challenge. The goal of this review is to identify and compare the current self-administered screening questionnaires for the diagnosis of LOH and to review the currently available data for the use of these questionnaires in both symptomatic hypogonadal men and asymptomatic men in the general population [4].

The Saint Louis University androgen deficiency in the aging male questionnaire and quantitative ADAM questionnaire

The Saint Louis University Androgen Deficiency in the Aging Male (ADAM) questionnaire was developed in 2000 and has been widely used to screen for androgen deficiency and identify men at risk for LOH [13]. Because of its widespread use and acceptance as a screening tool, the ADAM questionnaire has been published in a variety of versions and in a number of different languages [14,15]. This questionnaire consists of 10 questions related to androgen deficiency to be answered with “yes” or “no.” Morley et al. initially showed that the ADAM questionnaire had a sensitivity of 88%, suggesting its utility as a screening test. Similarly, others have suggested that the sensitivity of the ADAM questionnaire may be as high as 97% [16]. Despite a high sensitivity, the ADAM questionnaire is reported to have a specificity of only 24–60%, suggesting that its use beyond an initial assessment may be limited [14,17].

When the relationship between the ADAM questionnaire and serum hormone values was examined, a prospective study (n = 230) found that free testosterone (FT) and dehidroepiandrosterone sulfate (DHEA-S) are lower in men with a positive scores on the ADAM questionnaire, i.e. >3 out of 10 symptoms. In this study, only age, FT and diabetes were independently related to a positive ADAM questionnaire [17]. A study by Morley and colleagues (n = 148) found that the mean serum testosterone was different between men who answered “yes” and those who answered “no” to 8 of the questions in the ADAM questionnaire, and found that the specificity for the ADAM was 30% [16]. However, in a prospective study of 587 men with ADAM questionnaires and serum testosterone levels, there was no relationship between total ADAM score and serum testosterone levels regardless of various cut-off levels. Furthermore, testosterone supplementation for a duration of 6 months did not result in a change in average ADAM scores or the number of positive ADAM questionnaires.

Because of variability in sensitivity and specificity, a modified version of the ADAM, the quantitative ADAM questionnaire (qADAM), was proposed to make each question a more objective measure by asking patients to rate their answers for each of 10 questions on a scale from 1 to 5 [18]. Using the qADAM, Mohamed et al. found a strong correlation between the answers on the questionnaires and serum testosterone levels, as well as with scores on the Expanded Prostate Cancer Index Composite (EPIC) sexual and hormonal domain and the Sexual Health Inventory for Men (SHIM) [18]. Table 1 compares the ADAM and qADAM questionnaires to the other self-administered questionnaires in this review.

Table 1. Comparison of self-administered screening questionnaires for hypogonadism.

	Year of development	Number of questions	Sensitivity	Specificity	Advantages	Disadvantages
The Saint Louis University Androgen Deficiency in the Aging Male (ADAM) and Quantitative ADAM (qADAM)	2000	10	83.3–97%	19.7–36.6%	High sensitivity makes it ideal for screening test	Not all studies have shown a direct correlation with the ADAM and decreased serum androgen levels
The Aging Male’s Symptoms (AMS) Rating Scale	1999	17	83–96%	24–39%	Also has a high sensitivity, making it a reasonable choice as a screening questionnaire	Was not developed as a screening tool for diagnosis of LOH, thus may be more of a quality-of-life scale for aging men and not related to serum hormone levels
The Massachusetts Male Aging Study (MMAS) questionnaire	2000	8	59.9–76%	42.9–59%	Shortest of the screening questionnaires for LOH	Clinical data suggest that it is not the most accurate assessment tool
The New England Research Institutes (NERI) Hypogonadism Questionnaire	2009	25	N/A	N/A	Has good test-retest reliability and internal consistency as well as validity	There has yet to be a study looking at the questionnaires’ sensitivity, and thus its ability to act as a screening tool

The Aging Male’s Symptoms Rating Scale

The Aging Male’s Symptoms (AMS) rating scale was developed in 1999 as a measure of three components: to assess the symptoms in aging males, to evaluate symptom severity over time and to measure the changes that occur both before and after the administration of androgen therapy [19,20]. The AMS rating scale was developed in large part because of the lack of a standardized scale not only to measure the presence of age-related symptoms but also to assess the severity of these symptoms and their impact on quality of life [19]. The AMS rating scale consists of 17 items which ask the participant to rank their response on a score from 1 (none) to 5 (extremely severe) for a variety of different variables [21]. Like the ADAM questionnaire, the AMS rating scale has been translated into a variety of languages and accepted as an internationally validated resource [19,22].

Unlike the ADAM questionnaire, the AMS rating scale was not developed as a screening tool for diagnosis of LOH; however, studies have suggested that the AMS rating scale has sensitivity as high as 83–96%, which would in fact make AMS a reasonable choice as a screening questionnaire [16,21]. Furthermore, the AMS rating scale may provide a means of predicting the results of other tools for screening androgen deficiency, such as the ADAM or the Massachusetts Male Aging Study (MMAS) [23].

In the 2006 study by Morley and colleagues, there was a difference in mean testosterone and free testosterone levels between men who answered “yes” and those who answered “no” on 6 of the 16 questions in the AMS (5 of 6 were the same questions). Despite this association, other studies have not found a relationship between the AMS rating scale and hormone levels such as total and free testosterone or DHEA-S [24,25]. The prospective trial by Emmelot-Vonk and colleagues demonstrated no relationship between total AMS score, or AMS sub-category scores, and serum testosterone values. Similar to the ADAM questionnaire in this study, the AMS score showed no improvement with 6 months of testosterone supplementation therapy [26]. This suggests that many of the symptoms that are evaluated by the AMS are not related solely to changes in testosterone, or hypogonadism, and thus it is suggested that the AMS rating scale may be more of a reflection of quality-of-life for aging men than a measure of serum hormone levels [10,25]

The Massachusetts Male Aging Study questionnaire

The Massachusetts Male Aging Study (MMAS) questionnaire, or Smith’s screener questionnaire, was developed in 2000 as an eight-item, self-administered questionnaire with the goal of serving as a testosterone deficiency screener [27]. When initially administered in a clinical setting, the MMAS had a sensitivity of 76% and a specificity of 49% [27]. Similarly, when the MMAS questionnaire was tested as a possible surrogate for testosterone in the screening of androgen deficiency, it was found to have a sensitivity of 66.4% and a specificity of 46.1% when total testosterone was used as a reference parameter [28]. Interestingly, a study that subsequently compared the ADAM, AMS and MMAS questionnaires found that the ADAM was the best screening tool with a sensitivity of 97% compared with the AMS rating scale, which had a sensitivity of 83% and the MMAS with a sensitivity of 60% [16], which differs from the sensitivity observed when the MMAS questionnaire initially was studied. To date, no studies have examined the independent relationship between serum testosterone values and individual MMAS questions, or total overall score. While the MMAS questionnaire was designed as a tool to screen for LOH and may be attractive to use because it is the shortest questionnaire with only eight items, the clinical data suggest that it may not be the most accurate assessment tool.

The New England Research Institutes hypogonadism questionnaire

The New England Research Institutes (NERI) hypogonadism questionnaire was developed in 2009 specifically to screen for hypogonadism [29]. The questionnaire originally consisted of 67 items, but was modified to include 25 items that were considered to be of the most effective in identifying hypogonadal men [30]. While the initial work on the NERI hypogonadism questionnaire suggests that it has good test–retest reliability and internal consistency, as well as validity [30], there has yet to be a study of the questionnaires’ sensitivity, which would allow it to be applied as a screening tool for clinical hypogonadism. Until a study is performed defining the sensitivity and specificity of the NERI hypogonadism questionnaire, it is unlikely to be applied widely in clinical settings.

Structured interview hypogonadism screening questionnaires

In addition to the self-reported questionnaires, there are several other interview-based questionnaires for the screening of men with hypogonadism. While these are also valuable tools, they do require administration by a medical practitioner, and thus resources of a higher level. The main structured interview screening questionnaires for hypogonadism include the ANDROTEST [31,32] and the Canadian Society for the Study of the Aging Male (CSAM-Q) questionnaire [33]. Testing on these questionnaires has revealed a sensitivity of approximately 71% for the ANDROTEST [31] and sensitivity less than that of the AMS rating scale or the ADAM questionnaire for the CSAM-Q [33]. Given these facts, the structured interview hypogonadism screening questionnaires may not be as economical and efficient as a first-line screening test for hypogonadism.

Hypogonadism screening in young men (<40 years)

Screening for hypogonadism in young men has not been widely studied, thus there is little known about the use of these screening tools in this population. What has been seen is that the most common cause of symptomatic hypogonadism in young men is from anabolic steroid use [34]. Because the etiology of hypogonadism and symptom profile in young men is usually different from older men, controlled studies in this population using these questionnaires is warranted prior to determining the most appropriate tool for screening and diagnosis in young hypogonadal men.

Clinical application of screening questionnaires

While there are no data widely available on how long these questionnaires take to fill out for patients, each questionnaire would likely take no more than a few minutes for a majority of patients. Because of this, patients could be asked to fill out questionnaires while in the waiting room of the office, or they could be screened by age prior to their appointments and asked to fill them out at home before their appointments, allowing them to be integrated into the electronic medical record. Either of these methods of integration would allow a provider to quickly see the results for their patient, and determine if exploration for a possible diagnosis of LOH should be investigated.

Conclusions

The quest for a simple test to screen and identify men with LOH has led to the development of a variety of self-administered questionnaires. While several of these questionnaires have a high sensitivity in identifying hypogonadal men, they cannot be used reliably in an outpatient setting to diagnose hypogonadism because of their poor specificity. In addition, these questionnaires likely do not provide a good measure with which to follow patients with LOH after they have begun testosterone supplementation therapy because of the lack of follow-up data. Because of the complex variety of symptoms that occur with both hypogonadism and aging, a thorough history and measurement of serum testosterone levels remain equally important in diagnosing LOH. Additionally, the importance of following resolution of symptoms after initiation of testosterone supplementation therapy in these men suggests the need for additional questionnaires that can be validated and used for this purpose.

Declaration of interest

RR is an NIH K12 Scholar supported by a Male Reproductive Health Research Career (MHRH) Development Physician-Scientist Award (HD073917-01) from the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD).