Elsevier

Maturitas

Volume 29, Issue 1, 20 May 1998, Pages 25-31
Maturitas

Constructing a standard climacteric scale

https://doi.org/10.1016/S0378-5122(98)00025-5Get rights and content

Abstract

Issues relating to the design of scales and their psychometric properties are discussed in the context of constructing a standard measure of core climacteric symptoms. Seven factor analytic studies of climacteric symptoms are examined to determine whether or not there is sufficient consensus across studies to permit agreement on the symptom content and the structure of such a scale. It is argued that these factor analytic studies do indeed contain sufficient consensus on the basis of which a standard climacteric scale can be constructed. Such a scale is described.

Introduction

Climacteric research has long suffered from the lack of a standard instrument to measure the range of symptoms most commonly experienced by women at that time of their lives. It is a long recognised problem, highlighted by Jaszmann [1]at the Fourth and by Utian [2]at the Sixth International Congress on the Menopause.

In the absence of a standard instrument, individual researchers have tended to draw up their own ad hoc measures. These usually consist of lists of `menopausal' symptoms, subjectively selected, arbitrarily classified and rated merely present or absent, the resulting measure being poorly scaled and of unknown psychometric properties. The oft cited and seminal study by Neugarten and Kraines [3]of `menopausal symptoms in women of various ages', published over 30 years ago, is an early example of this practice—a practice which still persists even in more recent times 4, 5.

Alternatively, others have relied on the use of existing standard scales. While, in the context of the research objectives, such scales may be quite appropriate, they must be used with care since they have been designed for use with other types of populations, such as psychiatric patients. In this respect the Hamilton Rating Scales of Clinical Anxiety and Depression 6, 7and the Beck Depression Index 8, 9have proved to be the most widely used. Another popular measure is the General Health Questionnaire 8, 9, an instrument designed to measure psychiatric morbidity in general population samples.

This scale contains one item on hot flushes, one on sweating and six relating to sleep difficulties, a complaint which among climacteric women is generally thought by clinicians to be due to night sweats. It is little wonder therefore that the indiscriminate use of the GHQ results in a spuriously high psychiatric morbidity rate (47%) among perimenopausal women in the general population [10].

The only `menopause specific' symptom measure in current use is the outdated Blatt/Kupperman Index, which because of its inadequate psychometric properties and archaic wording, has now been seriously discredited [11]. The major criticism of this scale is that the summation of diverse symptoms to yield a total `menopause' symptoms index is without meaning. At this point it is appropriate to distinguish between the terms climacteric and menopause. Climacteric refers to a transitional phase in the life span during which a woman gradually moves from being reproductive to non-reproductive. As well as biological implications it also has social and psychological ones [12]. The menopause, the final cessation of menses, is a more circumscribed event occurring within the longer climacteric phase and is a sign that a particular biological point has been reached in the transitional climacteric process. The climacteric being a multi-faceted phenomenon, it follows that symptoms occurring during that time may come from different domains, have differing aetiologies and should consequently be categorised and measured separately from each other and not totalled to yield a single score as they are in the Blatt/Kupperman Index. This issue will be developed later in this paper. In the meantime let us turn to the broader issues of test construction.

Section snippets

Principles of test construction

To be regarded as having adequate psychometric properties, a measure should fulfil, to some extent, the four major principles of test construction listed in Table 1.

At the outset, it is essential that an instrument be initially constructed according to the principles of scaling. This means that the items included in the measure and the way they are categorised should be determined in some objective way and that the scoring system be both objective and sensitive. One way of doing this is by the

Factor analysis

As indicated above, the first step in any endeavour to develop a standard measuring instrument in a field where practice varies, is to achieve a consensus as to which symptoms should be included in the scale and how they should be categorised.

One way of doing this, in an objective way, is to use the mathematical technique of factor analysis to more clearly delineate those symptoms thought to constitute the syndrome or condition.

Factor analysis is a multi-variant mathematical technique

Factor analytic studies of climacteric symptoms

The first factor analytic study of symptoms presented by women during the climacteric was carried out by Greene [14], for the purpose of constructing a comprehensive measure of the multi-faceted and wide ranging symptom picture presented by climacteric women. The resulting scale consisted of three separate sub-scales measuring vasomotor, somatic and psychological complaints. This study was quickly followed by six other similar factor analytic studies of similar types of symptoms 15, 16, 17, 18,

Construction of a standard climacteric scale

It has been argued in this paper that it is possible to construct a measure of climacteric symptoms based on objective scientific principles. On the basis of the foregoing examination of factorial studies it follows that such a measure should consist of at least three separate sub-scales—one to measure vasomotor symptoms, one somatic symptoms and one psychological symptoms, although the last named may be further subdivided to measure anxiety and depressed mood. It should be noted that

Conclusion

The scale described in this paper is intended specifically to be a brief and standard measure of core climacteric symptoms or complaints to be used for comparative and replicative purposes across different types of studies whether they are medical, psychological, sociological or epidemiological in nature. Depending on the purpose of the research and the research questions, which in theory driven research would consist of a number of hypotheses, this scale would be supplemented by other measures

References (28)

  • Jaszmann L. A plea for a uniform menopausal index. Fourth International Congress on the Menopause, Orlando, Abstract...
  • Utian W. Menopause, sex hormones and quality of life. Sixth International Congress on the Menopause, Bangkok, Abstract...
  • BL Neugarten et al.

    Menopausal symptoms in women of various ages

    Psychosom Med

    (1965)
  • G Bungay et al.

    Study of symptoms in middle life with special reference to the menopause

    Br Med J

    (1980)
  • Cited by (468)

    View all citing articles on Scopus
    View full text