Osteoporosis is a disorder linked with ageing. It usually occurs in elderly people, especially postmenopausal females(Reference Delmas and Fraser1). In Asia, osteoporosis is rapidly becoming a major public health problem with increasing life expectancy(Reference Lau and Cooper2). The pathogenesis of osteoporosis is multi-factorial. Several risk factors have been identified and dietary factors play a major role, particularly in the elderly(Reference Bonjour, Schurch and Rizzoli3). It is also reported that low calcium intake is one of the risk factors for osteoporosis amongst Asian women(Reference Lau, Suriwongpaisal and Lee4). Many intervention studies have shown the positive effect of increased calcium intake in reducing the rate of bone loss in postmenopausal women(Reference Reid, Ames and Evans5–Reference Chee, Suriah, Chan, Zaitun and Chan9). However, calcium supplementation is not a sustainable solution in the long term for improving calcium intake, especially in poor countries. A preferred approach to achieve increased calcium intake is to use dietary strategies to enhance the consumption of local calcium-rich foods. It is also proved that nutrition education can improve calcium intake and retard bone loss in women(Reference Peterson, Klesges, Kaufman, Cooper and Vukadinovich10).
Quantitative ultrasound (QUS) is able to distinguish normal persons from persons with low bone density or with fractures due to osteoporosis(Reference Krieg, Thie′baud and Burckhardt11, Reference Reginster, Dethor, Pirenne, Dewe and Albert12). QUS parameters had been decreased over time in subjects of longitudinal studies(Reference Krieg, Thie′baud and Burckhardt11, Reference Schott, Hans, Garnero, Sornay-Rendu, Delmas and Meunier13–Reference Cadossi, Mele, Masci, Ventura, de Aloysio and Bicocchi15). Previous studies that assessed the effect of intervention on QUS indicated that QUS parameters were positively influenced in postmenopausal women in the intervention group, whereas it showed a significant decrease in the control group(Reference Krieg, Jacquet, Bremgartner, Cuttelod, Thie′baud and Burckhardt16–Reference Gonnelli, Cepollaro, Pondrelli, Martini, Rossi and Gennari18).
We have previously reported a relatively high prevalence of osteoporosis in Vietnamese women compared to the neighbouring countries(19). Although local calcium-rich foods are very cheap and available, the mean calcium intake in Vietnamese adult women was still much lower than the recommended dietary allowance (RDA) for calcium(20). Therefore, in the present study we aimed to examine whether a nutrition education intervention during 18 months would have an impact on improving the calcium intake of postmenopausal women, and to explore whether the increase in calcium intake would lead to subsequent effect on retarding bone loss in the women.
Material and methods
Design, settings and subjects
The present study was a controlled evaluation of the impact of an intervention designed to increase calcium intake and retard bone loss. It was approved by the Research and Ethical Committee of the Vietnamese National Institute of Nutrition, and was conducted in Thanh Mien rural district of Hai Duong province located in the Red River Delta. The district is situated about 70 km south-east of Hanoi with a population of nearly 138 000. Both curative and preventive health activities are the responsibility of district and commune health centres. Of the community organisations, which always supported health activities in the district, the Women’s Union is very strong, with a well-organised network from district to communes and villages. Members of the Union are known for their active participation in many primary health-care programmes.
Two communes in the district were selected for the study based on similar socio-economic and calcium-rich food source characteristics (primarily access to small crab, shrimp and fish). Each commune was far from the other, and was assigned randomly to be the intervention place or the control place. In each commune, all women who were aged 55–65 years, with low calcium intake (<400 mg/d) and more than 5 years postmenopausal, were screened for the study. Calcium intake was assessed by the Semi-Quantitative FFQ (SQFFQ). This questionnaire assesses the mean quantity of calcium in gram consumed per day in the previous 6 months. The questionnaire was developed and validated for use in a previous study(Reference Khan, Mai and Hien21). Subjects were excluded if they had the following factors: (i) current cancer or hyperparathyroidism; (ii) a kidney stone in the past 5 years, renal disease or bilateral hip surgery; (iii) therapy with a bisphosphonate, calcitonin, oestrogen, tamoxifen or testosterone in the past 6 months or fluoride in the past 2 years; and (iv) laboratory evidence of kidney or liver disease. After screening, seventy women eligible for participation were recruited for the study in the intervention commune. Then, in the control commune, seventy women who met the criteria and matched with those in the intervention group for age, years of postmenopause, educational level, life-long occupation, current weight-bearing exercise and calcium intake were selected for the study.
The intervention group was given nutrition education to improve calcium intake during 18 months. The control group continued with their usual diet. The control subjects were provided with the education at the end of data collection period. All subjects had given a written informed consent before conducting the study.
Participant retention
Given the all-volunteer nature of the subjects, several efforts were made to minimise the rate of attrition. In addition to indicating that bone mass would be provided for free, we also informed the participants that they would be examined for health status free of charge and given advice if they suffered from any disease.
Method of intervention
Based on calcium RDA for the elderly(Reference Prentice22–25), the overall goal of intervention was to increase the calcium intake of the intervention group up to 800 mg/d.
Exploring resource of calcium-rich foods in the locality
Available calcium-rich foods in the locality were investigated and ranked in order of calcium content. Those food items were also grouped into seasonal food lists. From the food items, a series of seasonal menus were designed for 7 days in a week to meet the calcium intake expected as 800 mg/d.
Approaches to promote participation of subjects
Organising training courses. Subjects were provided with training focused on promoting to eat local available calcium-rich foods with designed menus. Subjects were taught about the importance of osteoporosis, the role of calcium intake in controlling the disease, how to identify calcium-rich foods in the locality and how to prepare meals with local foods based on guided menus. In the training course, subjects were provided with visual education materials such as posters, leaflets, booklet and video tape, which illustrated all essential information with exciting pictures. Practising was also emphasised in the training. All available calcium-rich foods in the locality were bought from the local market, brought to the training site and used as an example for preparing, processing and cooking the meals, with the participation of all subjects and guidance from nutritionists. After the lesson, subjects could understand thoroughly and apply for organising the meals rich in calcium in their household.
Disseminating education message through a loudspeaker. Education messages including ‘Take calcium rich foods every day to enhance your bone health’ and ‘Take guided menu into your meals to reach enough calcium intake’ were given daily to the subjects by a loudspeaker system of the intervention commune. Furthermore, short and clear summary of the lesson given to the subjects in the training course was also repeated every day through a loudspeaker.
Approaches for monitoring, refreshing and maintaining practice of subjects
Monitoring subject’s intake at home. Subjects were provided with recording sheets. Throughout the 18 month period of intervention, subjects recorded all calcium-rich foods that they consumed daily on the sheet. Reliability of those recording sheets was checked and edited by collaborators recruited from the Women’s Union in the locality, who could see and communicate with the subject easily and frequently. They were noted for their creative activities to contribute to the success of previous community mobilisation programmes to control anaemia among reproductive-age women(Reference Khan, Thanh and Berger26).
To participate in the study, collaborators also received a training course with similar contents as the training given to study participants but more detailed and deeply expanded. Otherwise, they were trained on skills of communicating and disseminating knowledge to the subjects. Each collaborator was assigned to guide five to seven subjects. During the intervention period, collaborators visited each participant every day at the household. From the recording sheet, collaborators calculated the calcium intake of the subjects on the previous day and gave advice for adjusting the dietary intake on the next days. Collaborators were also provided with a check list form developed as a material to monitor and note the food consumption of each subject.
Practices of study subjects were also monitored by nutritional experts who visited the households every week to communicate with subjects and collaborators, gather recording sheets and check lists, and give them advice on improving practices at the households.
Group discussion. Group discussions involving study participants, collaborators and nutritional experts were organised every week. In those discussions, the study participants refreshed their knowledge and corrected practices by the efforts of nutritional experts. The experience of organising calcium-rich meals was exchanged between the study participants. They brought calcium-rich foods from their house to the meeting, calculated the quantity of foods needed for meals per day, and prepared standing dish from the foods in the presence of, and revised by, nutritional experts. From the lively discussions, nutritional experts could understand and assess changes in the knowledge and practices of subjects in order to adjust them properly.
Measurements
At screening and baseline survey, subjects were interviewed about their characteristics including age, postmenopausal years, educational level, life-long occupation, number of births and current weight-bearing exercise using a structured questionnaire. At baseline and 6 months thereafter, subjects underwent measurement of weight, height, bone mass and calcium intake. They provided fasting blood for analysing the serum parathyroid hormone (PTH) level at the beginning and at the end of intervention.
Life-long occupation was defined as the occupation that the participant was engaged in most frequently in her life. It was classified as heavy work (farmers, manual workers), office work (office clerks and other sedentary jobs) or domestic work (housewife).
Education level was defined by the number of years of schooling. Weight-bearing exercise was assessed by enquiring about the regular weight-bearing exercise during at least the past 12 months. The subjects reported the number of ≥30min sessions of weight-bearing exercise per week. Active behaviour was defined as more than two sessions per week.
Height and weight were measured while standing, wearing light clothing and no shoes. BMI was calculated as the ratio of weight (kg) to height squared (m2).
Bone mass was assessed by speed of sound (SOS (m/s)) using a QUS device (CM-100; ELK Corporation, Tokyo, Japan) as described in our previous publication(19). CV for the device were measured short-term in vivo and in vitro. The precision error (per cent CV) using a phantom was 0·15 % and in vivo was 0·27 %(Reference Kishimoto27). All subjects had SOS measured at the right calcaneus. The measurement was taken in a temperature-controlled environment, and performed only by a trained medical technician. Standardisation and calibration with standards were performed before the first measurement of each survey day.
Calcium intake was determined by the SQFFQ(Reference Khan, Mai and Hien21) and was based on Vietnamese Food Composition Table(28).
Intact PTH was measured by immunoradiometric assay (IRMA) using the DiaSorin N-tact PTH SP IRMA kit (DiaSorin Inc., Minnesota, MN, USA).
Statistical analysis
Differences in baseline characteristics between intervention and control groups were compared by the t-test and χ 2 test. Differences between the groups at each time of follow-up were compared by repeated measures analysis of covariance with baseline values as a covariate and time as the repeated measure. Within-group differences were compared by repeated measures ANOVA with time as the repeated measure. All P values are two tailed.
All statistical analyses were performed with the SPSS software for Windows version 10.0 (SPSS, Chicago, IL, USA).
Results
Reasons of dropouts
Thirty-two of 140 subjects dropped out from the study; thus 108 women (fifty-seven in interventions and fifty-one in controls) completed the follow-up, yielding a total dropout rate of 22·8 %. In the control group, nine women withdrew because they lost interest in the study, four subjects moved out of the study site, two subjects had serious illnesses and four others cited personal reasons for dropping out. In the intervention group, only two subjects were unhappy to continue follow-up. Other reasons were moving away from the study site (five subjects), serious illnesses (two subjects) and personal reasons (four subjects).
Baseline characteristics
Only results of 108 women who completed the follow-up were analysed. Of the subjects, 105 (97 %) were farmers and the others were office clerks. The subjects had an average age, time since menopause, years of schooling, number of births and BMI, respectively, of 57·6 (sd 3·0) years, 7·9 (sd 3·2) years, 5·4 (sd 2·7) years, 3·4 (sd 1·4) and 20·3 (sd 2·5) kg/m2. As shown in Table 1, no significant difference between interventions and controls existed in all considered variables. The prevalence of subjects defined as active weight-bearing exercise was 57·9 % and 60·8 % in interventions and controls, respectively.
Table 1 Baseline characteristics of subjects who had completed the study
SOS, speed of sound; PTH, parathyroid hormone.
Values are mean and sd or percentage. Intervention group was compared to control group by independent t-test or χ 2 test.
Anthropometry and PTH level changes
The results showed that changes in weight and height in both interventions and controls were not significant from the beginning of study to 18 months thereafter (Table 2). The intervention led to a decrease in PTH by 12 % (P < 0·01). In the controls, there was an increase in PTH by 32 % (P < 0·001) after 18 months.
Table 2 Mean changes in anthropometric indices and serum parathyroid hormone (PTH) level of subjects after 18 months
Values were compared between data of baseline and 18 months thereafter in both intervention and control groups by t-test.
Calcium intake and bone mass changes
Figure 1 shows the changes in the mean calcium intake at baseline and every 6 months thereafter. Calcium intake in the control group did not change significantly over time. In interventions, calcium intake increased significantly with time from 345 (sd 54) mg/d at baseline to 657 (sd 64) mg/d after 18 months (P < 0·01). The mean calcium intake of interventions was also significantly higher than that in controls at 18 months (P < 0·01).
Fig. 1 Mean change in calcium intake from baseline to the end of study (
, intervention group; , control group)
The percentage changes in bone mass over the study period of 18 months are presented in Figure 2. Bone mass (SOS value) was reduced significantly in controls by 0·5 % at 18 months (P < 0·01), while there was no significant difference for SOS value in the interventions between baseline and every assessment. After 18 months, the change in SOS value differed by 0·4 % between the two groups (P < 0·05).
Fig. 2 Percentage change in bone mass, as assessed by speed of sound at the calcaneus on quantitative ultrasound, from baseline to the end of study (▪, intervention group; ▒, control group)
Discussion
Osteoporosis is presenting in epidemic proportions in Asia(Reference Lau and Cooper2). Because consumption of milk is poor, the traditional Asian diet is considered to be low in calcium content(Reference Lau, Woo, Lam and Hong8). In Vietnam, there are many local calcium-rich foods which have high calcium content, especially small crab. But people do not know the importance of these food. Thus, the mean calcium intake of Vietnamese adult women was still much lower than the RDA for calcium, while such calcium-rich foods are cheap and available in the locality.
The current study evaluated whether or not nutrition education intervention would affect the calcium intake of postmenopausal women. The results showed that women in the intervention group had a greater increase in calcium intake than women in the control group. This finding is similar to a previous nutrition education study in young women with low calcium intake(Reference Peterson, Klesges, Kaufman, Cooper and Vukadinovich10). Consequently, women in the intervention group did not experience significant changes in SOS value, while women in the control group had significantly decreased SOS value by 0·5 % at 18 months. These results of our study generally are consistent with a previously published study, which showed that in postmenopausal women whose calcium intake was lower than 400 mg/d, calcium supplementation of 500 mg/d prevented bone loss(Reference Dawson-Hughes, Dallal and Krall29).
The current study also indicated that the difference between SOS value in interventions and controls after 18 months was 0·43 %. This finding is higher than the results of Krieg et al.(Reference Krieg, Jacquet, Bremgartner, Cuttelod, Thie′baud and Burckhardt16) and Giorgino et al. (Reference Giorgino, Paparella, Lorusso and Mancuso17), but still worse than the others where the differences were 0·8 % and 0·7 %, respectively(Reference Gonnelli, Cepollaro, Pondrelli, Martini, Rossi and Gennari18, Reference Giogino, Lorusso and Paparella30).
In the present study, we found that the mechanism by which increase of calcium intake retarded bone loss was associated with the suppression of PTH levels. We observed a higher level of PTH in controls than in interventions at the end of 18 months. Other calcium supplementation studies have also shown an age-related increase in PTH levels in the controls with time while interventions reduced the serum PTH levels(Reference Lau, Woo, Lam and Hong8, Reference Riggs, O Fallon and Muhs31).
A limitation of the present study was that we could not analyse other bone turnovers to determine the effect of improving calcium intake in bone mechanism. However, substantial effects of the intervention on PTH levels and SOS value make it very likely that there is also an effect on turnover(Reference Chee, Suriah, Chan, Zaitun and Chan9).
Perhaps the most remarkable finding was that the nutrition education promoted successfully participation of the community in the studied area, so that the calcium intake of women in the intervention group was improved. Increased consumption of calcium-rich foods may have resulted in improving the intake of other nutrients, which also positively contributed to bone health. Such improvements finally led to retarding bone loss in interventions. Ideally, in view of an overall dietary improvement, our study revealed that there was no significant consequence of weight gain. This is similar to results of milk supplementation trial in postmenopausal Chinese women conducted by Chee et al. (Reference Chee, Suriah, Chan, Zaitun and Chan9). It is worth noting that the current intervention targeted women who were consuming low calcium intake while local calcium-rich food sources are abundant. Further, the study was based on the close cooperation between researchers and collaborators as well as on the active participation of subjects. Those efforts strongly contributed to the success of our intervention approach. The effect of this education model should be confirmed in future studies for further expanded application to a larger population.
In conclusion, the results of this study indicated that community-based nutrition education intervention was beneficial in improving the calcium intake of postmenopausal Vietnamese women. Such dietary change had an effect on retarding bone loss in postmenopausal women as measured by QUS.
Acknowledgements
Conflicts of interest: The authors declare no conflict of interest. We declare that all authors have approved of the manuscript to submit to Public Health Nutrition. We also declare that contents of the manuscript have not been previously published and are not currently submitted elsewhere.
Source of funding: Funding for the study was supported by Shigeru Yamamoto.
Author contribution: V.T.T.H., N.C.K. and L.B.M. contributed to the design protocol and implement activities in the field. In addition, V.T.T.H. interpreted all results of the study. N.T.L. participated in developing the protocol and monitoring activities in the field. T.M.P. and N.V.N. participated in collecting data. B.T.N. and M.N. helped in analysing data. S.Y. gave advice to design and develop the protocol, monitoring all activities of the study and interpreting results.
Acknowledgements: We deeply thank all the subjects for participating in the study. We are greatly indebted to the staff of Vietnam National Institute of Nutrition and locality officials, especially collaborators for their efforts in implementing the study. We also thank Afework Kassu for help with the manuscript.
