Relevance of the semi-quantitative short Food Frequency Questionnaire in assessment of calcium consumption by female inhabitants of Zabrze over the age of 55 years (the Silesia Osteo Active Study)

Osteoporosis is a disease that is characterized by low bone mass, deterioration of bone tissue, and disruption of bone microarchitecture: it can lead to compromised bone strength and an increase in the risk of fractures [1]. Osteoporosis is one of the civilisation diseases, the onset and course of which depend on the diet with an appropriate supplementation of nutritional components, i.e. calcium, phosphorus, magnesium and vitamin D3 or protein [2‐5].

Among the methods of its prevention and slowing down, the course of this illness is physical activity, healthy lifestyle and proper diet.

Calcium is an essential element in the human body and is necessary to various cell functions. Calcium is not only important to bone health, but it is also essential for neuromuscular activity, blood coagulation and normal cardiac function. It is a vital component of skeleton where it is deposited by osteoblasts on a bone matrix throughout life. Food-derived calcium is absorbed in small intestine to blood plasma where its level is controlled by parathyroid glands. In case of low calcium level, parathyroids stimulate increased resorption of calcium in the kidneys and intestines, accelerating by that bone resorption. Therefore, an adequate intake of calcium is necessary to maintain this balance and healthy bones. [6, 7]

To assess the way of nutrition, it is necessary to use reliable, repeatable and simple diagnostic tools.

Researchers are making attempts to evaluate the relationships which seem to occur between diet components and the development of given disease entities. Reference methods are usually employed to assess the kind of diet, such as the method of keeping records from one or more days, a 24-h history method, analytical methods or biomarkers [8]. Furthermore, the validated Food Frequency Questionnaire begins to be more and more often used in nutritional epidemiological studies [9].

The aim of this reported study was a comparison of the 3-day food record method (3DFR) with the short, semi-quantitative Food Frequency Questionnaire (sFFQ). Then, the suitability for use of either method was analysed in medical outpatient environment.

A definition of a valuable, short method for patients’ diet assessment may allow to design preliminary, nutritional, medical screening programmes, complemented by properly selected prophylactic measures and patient’s reference to a clinical dietician.

A proper nutrition scheme, perceived as a necessary prophylactic element, is of particular importance for patients with the risk of osteoporosis [19, 20]. A multitude of analyses, both at national and international level, demonstrates too little calcium consumption in various populations [21, 22]. A very good example confirming this thesis is given in a study by Australian researchers, indicating a correlation between the consumption of calcium in food stuffs and a decreased bone fracture risk [23, 24]. Our research group also revealed too little calcium supplementation in relation to the standards recommended by the Polish Institute of Food and Nutrition in 2017. The Institute of Food and Nutrition recommends the Estimated Average Requirement (EAR) to be 1000 mg/day for that group of women [25].

Therefore, it is very important to find a reliable method, enabling a quick assessment of the consumption levels of all the components which are important for healthy bones. An analysis, carried out by Magarey et al., was a review of 36 assessment tools, which are applied to evaluate the consumption levels of calcium and dairy products (including the Food Frequency Questionnaire—FFQ—for dairy products, FFQ for calcium-rich products, an on-line FFQ for 15 and 25 products or a shortened FFQ). The analysis demonstrated that the used methods were not reliable and should thus be used with due care. The kappa statistics characterised some of the methods at the level of 0.8, while the value was lower in case of our study. Nevertheless, the authors of the mentioned report suggest that when selecting the method of evaluation, it makes sense to take into account the type, size, age, sex and physiological condition of a population. It may then be worthwhile to classify nutrition evaluation methods in the context of various studied groups [26]. A Croatian study of 333 postmenopausal women has demonstrated that FFQ may not be suitable for determination of either low, i.e. < 500 mg or very high, i.e. > 1200 mg, calcium supplementation. Similarly, as in the study of Dickson et al., we may observe from the Bland-Altman plot that FFQ is not suitable for individual calcium consumption assessment for its rather broad range of distribution. The cited study used data from a 27-component sFFQ and from a 24-h history. A comparison of the methods gave the kappa coefficient of 0.43 [27].

An analysis, performed by Angel M. Ong et al. in a group of 108 postmenopausal women, showed FFQ as a simple tool for calcium consumption assessment within 600–1000 mg, as well as for calcium assays in epidemiological studies [28]. A study similar to our internal research was conducted by Jensen in 162 Asians, Spaniards and white population subjects, all of them being the inhabitants of the USA. That group was requested to fill an FFQ questionnaire, based on 80 products, to reveal the amount of consumed calcium. In addition, FFQ was completed in the 2nd and the 3rd week with a 24-h history. The dependence of calcium supplementation, determined by FFQ and obtained from a twice-repeated 24-h history, attained 0.54 in Pearson’s coefficient (r). It demonstrates a strong correlation of the results, obtained by means of the two methods [29]. While making a review of the collected studies, one may perceive a high similarity, while there is no specific research management algorithm which would allow for unification of the research procedure, followed by a more precise evaluation. The nutrition evaluation methods, used by other researchers in the above-mentioned publications, differ from our apparatus with the following: the selection of method type, accounted food products in FFQ, the number of questions, the repetition or not of the 24-h food record, or the method of questionnaire and food record concepts (together with patient or on-line).

All the mentioned factors could possibly have had some impact on the final results and conclusions. An example of analysis with a methodology concept, different from those in the above presented studies, is a study of a group of women in the Northern Africa. The researchers made an attempt to estimate the volumes of consumed vegetables and fruit (as an element of balanced diet and a kind of prophylactics against non-infectious diseases). In this regard, a three times repeated 24-h nutrition record and a short FFQ (8-components), also repeated twice, were employed (which was not the case for other authors). As the outcome of the analysis demonstrated, that specific questionnaire of food consumption frequency turned out to be a reliable tool in the assessment of consumed vegetables and fruit by a population of 100 women, inhabiting the areas of the Northern Africa [30]. Another research project, this time of Australian researchers was based on a comparison of a 9-part FFQ (marked as MFQ—containing products rich in omega-3, plus a 74-part semi-quantitative FFQ). Both questionnaires were validated. As it was demonstrated by obtained results, MFQ may be a quick tool to identify consumed products, rich in omega-3. It is not recommended in terms of an individual analysis but may be an extremely useful tool at population level [31].

FFQ is a tool which can qualitatively estimate nutritional habits, and, when it is completed with questions, concerning the quantity and size of food portions, it becomes useful to determine (semi-quantitatively) the supplementation of vitamins or mineral components.

A nutrition diary is a cheap and simple method for evaluation (the record method). Such a record book usually covers, at least, 3 days. It provides information on nutrition methods and, by conducted actual recording, specifies the consumed quantities of nutrients. This method may, however, introduce a number of errors into the study. The character of errors depends mainly on the accuracy of the surveyed respondent. Among others, a tendency is often observed to record what is, in the respondent’s understanding, optimal in a given study (thus, the notes can be underestimated and inadequate vs. the actual, consumed portions, e.g. by disregarding unhealthy snacks). In addition, keeping food consumption records demands the patient’s writing and reading skills, enforces his/her ability to assess consumed portions and is, eventually, labour-intensive and tiresome [32, 33].

It should be emphasised that the results of nutrition analyses may, in effect, be more or less reliable, which, in turn, indicates the necessity of proper evaluation method selection, depending on a given situation. A validation process of the selected method, which determines its repeatability and reliability, is also an important element.

From the point of view of nutrition, a dietician has in his/her practice a possibility (tools) to run a full nutritional analysis and diagnosis; he/she may also repeat them during follow-up visits. However, it is a long-term and experience-demanding process, taking into account the fact that the usually limited time at a doctor’s office does not allow for a full nutrition analysis.