Osteoporosis knowledge and beliefs in diabetic patients: a cross sectional study from Palestine

This was a cross sectional study. Participants in this study were diabetic patients attending Al-Makhfiah primary healthcare unit in Nablus, north of Palestine. The study was carried out for four months, from September to December 2016. Al-Makhfiah clinic offers primary healthcare services and medications for individuals with chronic diseases who have governmental medical insurance. The number of diabetic individuals who attend Al-Makhfiah clinic on regular basis is less than 700. Recruited participants for this study were given full information regarding the purpose of the study and were asked to give informed consent before the interview. Participants were asked to complete a questionnaire through a face-to-face interview. The participants were interviewed by two co-authors who are senior medical students (G.I and Y.A). Participants in this study were limited to diabetic patients regardless of type. Participants who reported being diagnosed with osteoporosis or had previous bone fracture or currently taking calcium/vitamin D were excluded.

To estimate the sample size, a margin of error less of than 5% and a confidence level of 95% were entered in Raosoft calculator [37]. The total number of target population was entered as 700 while the he response rate was entered as 70%. A sample size of at least 222 participants was needed to perform the study. Assuming a response rate of 50% will yield a sample size of 249.

Tools used to collect the required data were the “Osteoporosis Health Belief Scale” (OHBS) and the “Osteoporosis Knowledge Test (OKT)”. Both were obtained from the developer with permission to use in Palestine [38]. The OKT and OHBS had been translated into different languages and were found reliable and valid in different communities [32, 39‐42]. The Arabic version of OKT and OHBS had been previously published and found to be reliable and acceptable [43]. We used the Arabic version of published OKT and OHBS after introduction of certain minor language changes to suite the Arabic accent of Palestinian culture. Permission to use the original scales and scoring methodology were obtained from the developer [44]. The translated scales are attached (Additional file 1).

The OKT is a 32-multiple choice instrument and has two subscales, a nutrition, and an exercise - related subscale. The two subscales had 14 items in common. In scoring the OKT, correct answers were coded as 1 while incorrect answers were coded as 0. Total score for OKT scale ranges from 0 to 32. The range for nutrition subscale was from 0 to 26 while that for exercise subscale ranges from 0 to 20. For each interviewed patient, scores of the OKT and the two subscales were analyzed as continuous variables and no categorization of scores was used. Examples of questions in the nutrition and exercise subscales include: “best source of calcium?: apple, cheese, cucumber”, “best way to reduce osteoporosis: swim, walk, stretching?”.

The OHBS consists of 42 items divided into seven subscales: susceptibility (items 1–6), seriousness (items 7–12), benefits of exercise (items 13–18), benefits of calcium intake (items 19–24), barriers of exercise (items 25–30), barriers of calcium intake (items 31–36), and health motivation (items 37–42). The internal reliability testing for the OHBS yielded a Cronbach’s alpha value of 0.720. The internal reliability of the seven OHBS subscales were 0.83, 0.71, 0.73, 0.72, 0.68, 0.69, and 0.66 for susceptibility, seriousness, benefits of exercise, benefits of calcium intake, barriers of exercise, barriers of calcium intake and health motivation subscales respectively. The OHBS is scored by awarding 5 for responses of “Strongly Agree” to 1 for “Strongly Disagree” for each item. Since there are 6 items in each subscale, the possible score for each subscale ranges from 6 to 30 with higher scores indicating higher belief in the tested subscale. The total score for the whole OHBS scale ranges from 42 to 210. Total score for each OHBS subscale was analyzed separately and as a continuous variable with no categorization of scores. The Arabic version of the OHBS subscales is shown in the additional file.

Examples of questions in the OHBS subscales include: “chances of getting osteoporosis high”; “when think about osteoporosis, get depressed”; “regular exercise helps build strong bones”; “taking enough calcium cuts chance of broken bones”; “have no place where can exercise”; “to eat more calcium foods have to give up other I like”; “look for new information related to health”.

To maintain the rights of participants in the study an Institutional Review Board (IRB) application was submitted to An-Najah National University. The IRB approved the study and asked for verbal consent from the participants and not a written consent because the study does not involve any invasive procedure. Furthermore, the Palestinian Ministry of Health approved the study and gave permission to authors to conduct the study after obtaining verbal approval from the participants. Before the collection of data, the participants were handed an informed consent that included information about all aspects of the study. Developers of the questionnaire gave permission to use the tool for the study.

Descriptive statistics including mean ± standard deviation (SD) or frequency (percentage) were presented for each variable, stratified by gender. Scores for all subscales were tested for normality using Kolmogrov - Smirnov test and were found to be not normally distributed. In Bivariate analysis, Mann-Whitney U test was used for ordinal variables while correlation between total knowledge scores and various belief subscales were carried out using Spearman correlation.

Three hundred diabetic patients were recruited; 192 (64%) were females. The mean ± SD age of the participants was 58.5 ± 9.3 years. The majority of participants were living in urban areas (215; 71.7%). The majority also reported being currently married (251; 83.7%). For the level of education, the majority (236; 78.7%) did not have a college degree. Regarding occupation, the majority (217; 72.3%) of participants were unemployed. Participants had mean ± SD duration of DM of 11.3 ± 8.0 years. More than half of the participants (174; 58%) were insulin users and two thirds of the participants were hypertensive. Regarding co-morbidities, 229 (76.3%) of studied patients have at least one co-morbidity other than DM and 91 patients (30.3%) had retinopathy.

Comparison of the demographic and clinical characteristics between female and male participants (Table 1) showed that females were significantly younger than males (mean ± SD age for females was 57.1 ± 8.9 while that for males was 61 ± 9.6 year; p <  0.01). Furthermore, females had significantly shorter duration of diabetes than males (mean ± SD duration of DM in females was 10.6 ± 7.6 while that for males was 12.6 ± 8.6 years; p = 0.036). Significant differences between males and females were also shown in place of residence (p = 0.022), marital status (p <  0.01), level of education (p <  0.01), employment status (p <  0.01), use of insulin (p = 0.012), and presence of hypertension (p <  0.01).

The mean ± SD of OKT total score was 13.5 ± 4.2. The mean ± SD of exercise knowledge score was 8.8 ± 2.8 while that for nutrition knowledge score was 11 ± 3.6. Most of the participants correctly answered question 5 (OK5; 81.7%) and question 10 (OK10; 78%) (Table 2). Females had a significantly higher overall OKT score (p = 0.021), higher exercise knowledge score (p = 0.043), and higher nutrition knowledge score (p = 0.037) (Table 3).

The mean ± standard deviation for osteoporosis belief subscales are shown in Table 4. Diabetic patients had the highest score (mean ± SD = 23.1 ± 2.5) in belief subscale pertaining to benefit of exercise as protective against osteoporosis. On the other hand, diabetic patients had the lowest score in belief subscale pertaining to the presence of barriers to calcium intake. Examples of frequency of answers for certain questions in OHBS include the followings. (1) more than half (54%) of participants did not believe that they have high chances of getting osteoporosis (item #1); (2) the majority (78.4%) of participants incorrectly answered that family history makes them more likely to have osteoporosis (item # 6); (3) 46% agreed that it would be scary to get osteoporosis (item # 7); (4) 75% correctly answered that exercise is protective toward osteoporosis (item # 14); (5) 85% correctly answered that calcium intake is protective toward osteoporosis (item # 19); (6) 68% agreed that it would be serious to get osteoporosis (item # 12). The majority of participants disagreed with the five statements of perceived barriers to exercise (items number 26–30), but the majority agreed that their physical weakness is a perceived barrier to exercise (item # 25). Similarly, the majority of participants disagreed with the five statements of the perceived barriers to take calcium (items number 31–35). However, the majority agreed that food rich in calcium might be rich in cholesterol and therefore was a perceived barrier to intake of calcium rich food (item # 36). The majority of participants agreed to all statements that assess health motivation (items number 37–42).

When belief scores were examined by gender, female participants showed significantly higher scores in perceived susceptibility (17.4 ± 4.1 for females versus 15.8 ± 3.8 for males; p <  0.01) and perceived seriousness (19.8 ± 3.6 for females versus 18.7 ± 3.6 for males; p < 0.01) of osteoporosis when compared to male participants. No significant difference was found between males and females in other belief subscales (Table 4).