Introduction

Osteoporosis is a serious bone disease affecting hundreds of millions of individuals around the world [1]. It is defined by the World Health Organization (WHO) as a skeletal disorder characterized by a decreased bone mass and micro-structural degradation leading to a loss of bone strength and an increased risk of bone fractures mainly involving the spine, hips, and wrists [2]. These fractures are associated with a high morbidity related to pain and disability, mortality, and economic burden [3, 4]. According to a study in which elderly Lebanese individuals aged between 65 and 84 years old from the greater Beirut area were enrolled, osteoporosis was found to be prevalent in 33% of women and 22.7% of men [5].

There are two types of osteoporosis-related risk factors: modifiable and non-modifiable [6, 7]. Non-modifiable risk factors involve age, gender, ethnic origin, the presence or not of osteoporosis in the family along with the history of fractures after a minimal trauma in first-degree relatives, the date of the first menarche and menopause, etc. Modifiable elements mainly include lifestyle behaviors such as smoking, alcohol consumption, poor physical activity, and a diet deficient in adequate calcium and vitamin D levels. Furthermore, men and women are both affected by osteoporosis, but women have a higher risk mainly because of menopause leading to a well-defined entity called “postmenopausal osteoporosis” [8]. In fact, at menopause, the hormonal production of the ovaries is substantially reduced and blood levels of estrogen essential for bone homeostasis markedly decrease [8,9,10,11,12].

There are several treatment options for osteoporosis. They are divided into three different categories: anti-catabolic (anti-resorptive), anabolic drugs, and drugs having both effects. Anti-catabolic drugs reduce the bone turnover which is normally the difference between the formation and the resorption phase, by mainly reducing the latter. Examples include biphosphonates, estrogens, selective estrogen receptor modulators (SERMS), calcitonin, and inhibitors of the receptor activator of the nuclear factor kappa B ligand (RANKL). The anabolic drug category, with teriparatide being the major drug, increases bone remodeling with a greater stimulation of bone formation over resorption and acts at the level of bone microarchitectural parameters most importantly being the cortical and trabecular bone, improving their structure. Deciding on whether to treat or not depends on the results of bone density measured by a dual X-Ray absorptiometry bone mineral density scan (DXA-BMD) and the 10-year risk of developing a fragility fracture using the Fracture Risk Assessment tool (FRAX).

Several research studies have shown a positive relationship between osteoporosis knowledge and application of preventive measures. In fact, in a study in which an informational booklet about osteoporosis was administered to women, with a subsequent follow-up done, has yielded a better knowledge with more favorable attitudes and preventive practices towards the disease [13].

In Lebanon, there is a lack of sufficient evidence and recent data associated to public osteoporosis awareness. For this reason, we developed a study to investigate and identify Lebanese women population groups with the poorest knowledge of osteoporosis definition, risk factors, and preventive measures in order to identify those that should be targeted in future awareness campaigns and educational programs.

Numerous scales assessing the level of knowledge and health belief status associated with osteoporosis were used in previous studies. However, when attempting to compare them, we found that none of these scales consisted of the whole number of questions needed to get a complete idea about these parameters. Subsequently, we used in our study a questionnaire that included questions selected from these well-validated scales. The questions are related to participant’s background characteristics, knowledge about key osteoporosis facts, and health belief status.

Materials and methods

Study and population

A cross-sectional study was conducted between March and June 2018, which enrolled 560 community dwelling women aged 40 years and above using a proportionate random sample from all Lebanese governorates (Beirut, Mount Lebanon, North, South, and Bekaa). The cutoff point of 40 years was chosen in regard to the particular importance of osteoporosis knowledge and health beliefs in this category due to the effects of menopause on bone density and fracture risk. Each governorate is divided into Caza (stratum). Two villages were randomly selected from the list of villages provided by the Central Agency of Statistics in Lebanon (2005) [14]. Patients were randomly selected from each village. Excluded were women with cognitive impairment, dementia, or who refused to fill the questionnaire. Data collection was performed through personal interviews with participants by trained, study independent personnel.

Sample size calculation

A sample of 341 women was targeted to allow for adequate power for bivariable and multivariable analyses to be carried out according to the Epi info sample size calculations with a population size of 5 million in Lebanon, a 51.5% expected frequency of good osteoporosis knowledge among the general population [15], a 95% confidence limits, a power of 80%, and an acceptable margin of error of 5%.

Data collection

We utilized a questionnaire that included translated items from existing questionnaires to investigate sociodemographic characteristics as well as osteoporosis knowledge and beliefs. A thorough literature review highlighted the presence of 5 osteoporosis-related scales mostly used in research studies: the osteoporosis knowledge assessment tool (OKAT) [16], the osteoporosis knowledge test (OKT) [17], the osteoporosis questionnaire (OPQ) [18], the facts on osteoporosis quiz (FOOQ) [19], designed to collect data related to knowledge of osteoporosis facts and the Osteoporosis Health Belief Scale (OHBS) [20] investigating disease-related beliefs. These tools were the basis for the development of a questionnaire used in this study ( Appendix 1 ). Our questionnaire was in Arabic, the native language of Lebanon, and was divided into three parts. The first part (Part 1: Demographic characteristics) involved 22 questions including the age, the governorate, the family monthly income, the history of personal and familial diseases (including osteoporosis) and surgeries, the presence or not of menopause, the lifestyle (smoking, alcohol consumption, physical activity, calcium, and vitamin D intake), the medications taken, and whether one’s had ever heard about osteoporosis. The educational level (illiterate, primary, secondary, technical, or university level of education) was also included. Technical education involves programs that aim to develop in students specific technical skills in order to directly enter the workforce. The family monthly income was divided into 3 categories based on the monthly salary of the household as follows: low (1000 USD), intermediate (1000–2000 USD), and high (> 2000 USD).

The second part (Part 2: The Knowledge Scale) intended to get information concerning the knowledge of osteoporosis facts was subdivided into 3 sections: “Definition of Osteoporosis,” “Risk Factors of Osteoporosis,” and “Preventive Measures.” Questions making part of the first four scales listed above were distributed between these three categories.

Integrated questions were selected upon consulting osteoporosis experts with the attempt to include the biggest number of questions related to this bone disorder along with a questionnaire that is easy to understand using a simple wording and asking about the basics. Furthermore, we tried to avoid repetition and eliminated similar questions to avoid redundancy (e.g., “If your mother or father have had osteoporosis” (OPQ); “Family history of osteoporosis strongly predisposes a person to osteoporosis” (OKAT); Having a mother or a grandmother who has osteoporosis is” (OKT); “Family history of osteoporosis is not a risk factor for osteoporosis” (FOOQ)) and complicated questions that usually need a more advanced knowledge (e.g.: “Which of the following is the recommended amount of calcium intake for an adult?”(answer in mg-OKT)). Additionally elements not directly associated to our topic were removed (e.g., “Side effects of HRT include” (OPQ); “Osteoporosis and osteoarthritis are” (OPQ)). This yielded a total of 42 questions; for each question, “1” point was given for a correct response and “0” for an incorrect or “I don’t know” response. Hence, the total score ranged between 0 and 42. The “I do not know” answer was added in order to avoid the selection by participants of a “False” response or a random one while completely ignoring the correct answer.

No modifications were made to the original items expect for 6 of them:

  • a question selected from the OPQ “The condition characterized by fragile or brittle bones is commonly known as” was modified to “Osteoporosis is a condition characterized by” with additionally a change of the answers from “(1) Arthritis; (2) Osteoporosis; (3) Spondylitis; (4) Don’t Know” into “(1) Bone pain increased with activity; (2) Fragile or brittle bones; (3) Decreased muscle strength causing bone fragility; (4) Don’t Know.” This was done in order for individuals completing the questionnaire, to choose the right definition of osteoporosis rather than directly giving it in the question and including elements not related to the topic such as “Arthritis” and “Spondylitis.”

  • The item “Osteoporosis is more common in men” from the OKAT was modified to “Osteoporosis is more common in women” to emphasize on the risk of osteoporosis in this category.

  • The question “By age 80, the majority of women have osteoporosis” (OKAT) was changed into “The risk of osteoporosis gradually increases starting the age of 50 years” based on the fact that menopause (developing around 50 years of age) is the principal factor leading to osteoporosis in women. Subsequently, it is essential for us to know if Lebanese women are aware of this process.

  • The first element of the FOOQ “Physical activity increases the risk of osteoporosis” was modified to “Regular physical activity increases the risk of osteoporosis” to highlight the importance of the regular pattern of exercising.

  • A question selected from the OKAT “There are no effective treatments for osteoporosis available in Australia” was also changed into “There are no effective treatments for osteoporosis available in Lebanon” since the study is conducted among Lebanese women.

  • The question “An adequate calcium intake helps reducing the development of osteoporosis” was added because calcium is one of the main supplements that should be prescribed, as recommended, for postmenopausal women and it is critical to have an idea whether Lebanese women perceive it as essential for the prevention and reduction of the onset of osteoporosis or not.

The last part “Part 3: Health Belief Scale” used The osteoporosis Health Belief scale (OHBS). In its original version, the OHBS includes 42 statements divided into 7 categories. Every question has 5 possible answers: “Strongly Disagree,” “Disagree,” “Neutral,” “Agree,” and “Strongly Agree” scored respectively “1, 2, 3, 4, 5.” Subsequently, every subscale has a scoring varying between 6 and 30. Each category is analyzed separately with higher scores related to a greater health belief status.

The first category asks about the perceived risk of developing osteoporosis “Perceived susceptibility” (items 1 to 6) and the second about how severe it is considered “Perceived seriousness” (items 7 to 12). The third and the fourth respectively investigate the degree of advantage that women find from practicing a regular physical activity “Exercise benefits” (items 13 to 18) and from consuming calcium-rich products “Calcium benefits” (items 19 to 24). Moreover, the fifth and sixth categories included questions about perceived boundaries against regular exercising “Exercise barriers” (items 25 to 30) and adequate calcium intake “Calcium barriers” (items 31 to 36). Finally, the “Health Motivation” category (items 37 to 42) measures the degree of willingness to be screened and treated for osteoporosis when it is present.

The questionnaire used in our study involved 28 selected questions from the original scales. Multiple questions were not included because their translation to Arabic would have resulted in the same meaning (e.g., “It is extremely likely that you will get osteoporosis” and “There is a good chance that you will get osteoporosis” ♦ “If you had osteoporosis you would be crippled” and “When you think about osteoporosis you get depressed”). Moreover, other questions were removed because of similarities with other ones, e.g., in the “Exercise barriers” subscale, the question “Exercising regularly makes you uncomfortable” is a general question, whereas the rest of the questions making part of this scale explain why it would be uncomfortable.

In order for the questionnaire to be accessible to all Lebanese women, its English version was translated to Arabic by a certified translator, then a second certified translator translated the Arabic version back to English. The translators had a good understanding of the content and concept of the different scales. It is noteworthy that the English versions did not significantly differ; therefore, the translated Arabic version was used as is. A committee including specialists in the field (internists, rheumatologists, and endocrinologists), a language expert, and our translators examined the Arabic questionnaire to make sure it has an idiomatic equivalence [21, 22]. A pilot study was conducted on 20 women to ensure the complete understanding and clarity of the questions. No changes were made thereafter.

Statistical analysis

Data analysis was conducted using SPSS software version 23. The independent sample t test was used when comparing two means, whereas the ANOVA and Kruskal-Wallis tests were used to compare between three groups or more, and Pearson correlation coefficient was used to correlate between continuous variables. A stepwise linear regression was conducted, taking the knowledge score as the dependent variable. All variables that showed a p < 0.1 in the bivariate analysis were taken as independent variables in each model to decrease the possibility of confounding factors. Besides, Cronbach’s alpha was recorded for reliability analysis for all the scales. A p value less than 0.05 was considered significant. In the absence of a cutoff point for the “Knowledge Scale” and “Health Belief Scale,” we chose the medians as cutoff points.

Results

Sociodemographic and other characteristics of the participants

The sociodemographic and other characteristics of the participants are summarized in Table 1. The results of the Chi-square test showed that a significantly higher percentage of participants with poor knowledge about osteoporosis was found in participants who live in Bekaa (11.3% vs 2.4%) and were illiterate (20.8% vs 7.1%). Moreover, a significantly higher percentage of participants with poor knowledge were unemployed (62.3% vs 45.4%) and had a low family monthly income (56.6% vs 43.1%). Finally, the results of the Student t test showed that participants with a poor knowledge had a higher mean age (52.42 vs 50.34 years). The osteoporosis knowledge score mean and standard deviation for the whole sample were 24.13 ± 6.12 (median = 25; minimum = 0 and maximum = 42). The percentage of patients who had adequate knowledge (scores ≥ 25) about osteoporosis was 52.7% (295 patients). The Cronbach alpha for this scale was high (0.825).

Table 1 Sociodemographic and other characteristics of the participants
Full size table

Bivariate analysis

To assess factors associated with the knowledge score, the Student t test was used for dichotomous/binary variables, whereas the ANOVA test was used for variables that had 3 categories or more. The results showed that a higher mean knowledge score (25.90 ± 5.38) was found in participants from North Lebanon compared to all other districts, with a university level of education (26.51 ± 5.67) compared to all other levels, in women taking calcium and vitamin D supplements (25.47 ± 6.40) compared to those taking either supplement alone or no supplement, and in those not taking corticosteroids, aromatase inhibitors, heparin, or chemotherapy compared to those with current or previous usage of these medications. Finally, the Pearson test showed that increased age was significantly correlated with decreased knowledge (r = − 0.097) (Supplementary Table 1).

Multivariable analysis

The results of a linear stepwise regression, taking the knowledge score as the dependent variable, showed that technical (Beta = 1.36; CI 0.013–2.714) and university levels of education (Beta = 2.95; CI 1.956–3.951) compared to illiteracy, those who previously heard about osteoporosis (Beta = 5.70; CI 3.764–7.653), those who take calcium and vitamin D supplements compared to none (Beta = 1.61; CI 0.381–2.850), and those exercising 10–20 min per day (Beta = 2.73; CI 0.891–4.572) and more than 30 min per day (Beta = 1.84; CI 0.311–3.377) compared to no exercising were significantly associated with higher knowledge about osteoporosis. On another hand, women with a previous fracture (Beta = − 2.03; CI − 3.147–− 0.917), taking calcium supplements only (Beta = − 1.63; CI − 2.942–0.334), exercising for less than 10 min daily (Beta = −5.49; CI − 8.386–− 2.605) compared to no exercising, having diabetes (Beta = − 1.65; CI − 2.906–− 0.398), hyperparathyroidism (Beta = − 4.94; CI − 9.016–− 0.883), and taking current chemotherapy treatment (Beta = − 4.07; CI − 7.358–− 0.783) compared to no treatment were significantly associated with lower knowledge about osteoporosis (Table 2).

Table 2 Multivariable analysis: linear regression taking the knowledge score as the dependent variable
Full size table

Health belief model

The description of the health belief model scores subscales, with the respective Cronbach alpha values for each of these subscales, can be found in Supplementary Table 2.

Bivariate analysis of factors associated with each OHBS subscale

The bivariate analysis results of factors associated with each of the OHBS subscales are summarized in Supplementary Table 3. A significantly higher susceptibility score was found in illiterate people compared to all other educational levels (8.88). In addition, a higher age (r = 0.204) and a higher number of kids (r = 0.140) were associated with an increased susceptibility score. A higher mean seriousness score was found in illiterate women compared to all other educational levels.

An increased age (r = − 0.101) and an increased number of kids (r = − 0.041) were significantly associated with decreased exercise benefits score. Whereas an increased knowledge score (r = 0.295) and (r = 0.265) was respectively associated with increased exercise benefits score and mean calcium benefits scores.

A higher exercise barriers score was found in illiterate women compared to all other educational levels and in women with higher age (r = 0.225), whereas a higher knowledge score was correlated with a lower exercise barriers score (r = − 0.142). In addition, a higher calcium barriers score was found with increased age (r = 0.132), whereas a higher knowledge score was correlated with a lower calcium barriers score (r = − 0.198).Finally, a higher health motivation score was found in women with higher knowledge about osteoporosis (r = 0.275).

Multivariable analysis

The results of a linear regression, taking the susceptibility score as the dependent variable, showed that age (Beta = 0.051; CI 0.031–0.072) was significantly associated with a higher susceptibility score. Being employed (Beta = − 0.514; CI − 0.944–− 0.083) was significantly associated with a higher seriousness score.

A higher knowledge (Beta = 0.129; CI 0.094–0.164), divorced status compared to single (Beta = 1.004; CI 0.076–1.933), and smoking waterpipe compared to non-smokers (Beta = 0.511; CI 0.007–1.015) were associated with a higher exercise benefits score. Furthermore, an increase in the knowledge score (Beta = 0.093; CI 0.064–0.122) was significantly associated with a higher calcium benefits score, whereas cigarette smoking compared to non-smokers (Beta = −0.581; CI − 0.971–− 0.192) was significantly associated with a lower calcium benefits score.

Increased age (Beta = 0.081; CI 0.047–0.115) and a technical educational level compared to illiteracy (Beta = 1.083; CI 0.133–2.033) were associated with a higher exercise barriers score, whereas a higher knowledge score (Beta = − 0.068; CI − 0.123–− 0.013) and being employed vs unemployed (Beta = − 0.846; CI − 1.540–− 0.151) were associated with a lower exercise barriers score.

Increased age (Beta = 0.034; CI 0.011–0.056) was significantly associated with a higher calcium barriers score, whereas a higher knowledge score (Beta = −− 0.086; CI − 0.122–− 0.049) and a primary level of education compared to illiteracy (Beta = −0.7; CI − 1.317–− 0.083) were significantly associated with a lower calcium barriers score.

Finally, a higher knowledge score (Beta = 0.187; CI 0.132–0.241) was significantly associated with a higher health motivation score (Table 3).

Table 3 Multivariable analysis
Full size table

Discussion

This study assessed the level of knowledge about osteoporosis in women aged 40 years or more in a representative sample of the Lebanese population. The results showed that women who received no education, compared to a higher education and have not previously heard about the disease, had lower knowledge scores. In addition, women not taking calcium and vitamin D supplements and not exercising at all or exercising less than 20 min per day compared to those having these characteristics had lower osteoporosis knowledge scores. A lower knowledge score was also associated with less recognized benefits of adequate calcium intake and regular physical activity, more perceived barriers towards their practice, and a less important health motivation. A lower level of education correlated to the same results.

Multiple studies conducted in the USA, Canada, Poland, India, and the Middle East have aimed to evaluate the knowledge of osteoporosis main causes and consequences, attitudes essential to limit its occurrence along with personal health beliefs in a subject [23,24,25,26,27,28,29]. The majority of previous studies pointed at the need for further education. Our study was conducted among Lebanese women aged 40 years and above in order to investigate the characteristics of Lebanese women population groups with the poorest knowledge of osteoporosis definition, risk factors, and preventive measures in order to identify categories that should be targeted in future campaigns and educational programs.

The mean osteoporosis knowledge score was lower in participants who never went to school as compared to those who attended a technical college or a university. This is in accordance with other research projects in which the greater osteoporosis awareness was found in women with higher education levels and was attributed to increased availability and easier access to information [28, 30,31,32,33]. Our results could be similarly explained. In fact, the greater degree of education could have helped women find and appraise health information and subsequently apply these skills to osteoporosis knowledge throughout life. This higher degree of education is usually associated to a higher socio-economic status and subsequent employment with better lifestyle conditions and improved states of health and wellbeing [34].

Lower knowledge scores were found in women who never previously heard about osteoporosis. It is also the case of a survey carried out in Singapore in which women with this characteristic were less aware of osteoporosis facts regarding risk factors and complications [35]. Additionally, it has been established that one’s exposure to another individual with osteoporosis, and not necessarily a relative, led to an improved understanding of the issue [36]. This finding supports the development of peer-education programs in order to improve osteoporosis knowledge.

In contrast to a study conducted in the USA, concluding that there is an absence of any association between the degree of knowledge and the practice of preventive measures, particularly diet supplementation as recommended and regular physical activity [37], women enrolled in this study having less healthier lifestyles reached less important knowledge scores. In fact, those not taking any calcium and vitamin D supplements had a less recognition of osteoporosis key points relatively to those taking calcium supplements. Women with a self-reported regular physical activity lasting less than 10 min compared to those exercising between 10 and 20 min had reached lower knowledge scores. This could be directly interpreted by two different theories that we advance. The first one considers that women with healthier lifestyles could be enough motivated to seek additional information about the different medical disorders. Whereas the second theory states that women with higher health literacy had the required abilities to find and understand health information and then also the resources necessary to put that information into practice. As a result, improved health literacy has the potential to increase access to health care [38].

Women who had a bone fracture and those who are affected by diseases like diabetes and hyperparathyroidism or are exposed to treatments as current chemotherapy and past heparin therapy, all at risk of accelerated bone loss, were less informed about osteoporosis. A study conducted in Canada among patients of both genders, who were managed by orthopedic surgeons for any type of fragility fracture, demonstrated that there is a major lack of awareness linked to osteoporosis even in those at an increased risk of developing or already being affected by this bone disorder [39]. In the same study, individuals who were able to adequately define osteoporosis were those who have been diagnosed to have it and thus, additional information might have been provided by their physicians regarding the different aspects of the disease. Furthermore, it has been demonstrated by multiple research projects that medical doctors were not the main source of information [24, 25]. In contrast, media such as television and internet corresponded to the leading origin. The findings of our study might be attributed to the lack of information given by healthcare professionals to patients with or at increased risk for the disease.

As previously discussed, health behaviors are highly related to the degree of knowledge related to osteoporosis. Furthermore, personal beliefs are also major factors that influence attitudes. In the current study, a higher susceptibility score was reached in women with a more advanced age. This correlates well with other studies that found a positive relationship between older age and a greater noted osteoporosis risk [40, 41]. This could be explained by the fact that osteoporosis is considered a disease of the elderly by younger age groups with no short-term risk of developing it. Furthermore, an increasing age was associated with increased calcium and exercise barriers. The main reason could be that people of older ages have multiple comorbidities not allowing them, in certain cases, to be able to readily perform physical activities (e.g., osteoarthritis, heart failure) and include calcium-rich products like milk, cheese, and ice cream in their diet (e.g., diabetes, dyslipidemia).

Women with a lower knowledge, less healthy lifestyles, and a less important educational status had lower calcium and exercise benefits scores, higher calcium and exercise barriers scores, and were much less motivated to engage in practices and behaviors aiming to reach a better health status. Besides, unemployed women had particularly more recognized exercise barriers and perceived a lower seriousness of the disease as compared to employed ones. In light of these findings, better knowledge helps people develop suitable preventive behaviors. Thereby, presenting information in ways that reach socially disadvantaged populations and with lower levels of education may help improve health behaviors in this group.

Limitations

This research project has multiple limitations. It followed a cross-sectional design being not capable of establishing a cause and effect relationship. The reached sample may not have been representative of the whole Lebanese population as the biggest number of respondents lived in North Lebanon. In view of a questionnaire administered via an interview, anonymity may have not been completely respected and respondent and information biases are possible since participants may have misreported or underreported some of their exposures and characteristics. The questionnaire was translated into Arabic, but this has implications for non-Arabic speaking women in the area possibly excluding those with the poorest osteoporosis knowledge. The cutoff point that was used to determine adequate knowledge was based on the median score and is not representative of the level of knowledge that is necessary for the management/prevention of osteoporosis. The cutoff points of the questionnaire have not been validated and therefore estimates of “poor” knowledge scores may not reflect the level of knowledge required to engage in osteoporosis behaviors. It is possible that women with greater knowledge were more likely to self-report their health behaviors, which could lead to a result associated to better application of preventive measures in the case when it could be not. Moreover, the Arabic translation of these existing questionnaires has not been tested for reliability or validity in this population. A future study will be targeted towards validating this developed scale. Furthermore, the most complicate questions were excluded from this questionnaire with subsequently a less accurate estimate of “poor” knowledge.

Conclusion

This study showed that certain Lebanese women population groups have poorer knowledge than others and osteoporosis health beliefs also differed between them. Thus, the Lebanese community should aim at developing osteoporosis education programs tailored to meet the needs of the population groups identified in the study as having poor osteoporosis knowledge. In addition, medical practitioners carry the responsibility for providing patients at an increased risk of the disease with adequate information. Furthermore, as a next step, the translation of the previously developed osteoporosis questionnaires utilized in this study should the tested and validated for appropriateness in this population group.