Of 138 women recruited, 17 did not return for follow-up at 2 months, and only 121 completed the study. The response rate for the study was 87.7%. Table 1 shows the descriptive analysis of the sociodemographic data.

Table 2 shows the mean scores for KAP at baseline and 2 months after intervention. The mean±standard deviation (SD) scores for KAP were 24.98±4.56, 24.25±4.92, and 3.51±3.51, respectively, before intervention. The mean±SD scores significantly increased to 29.66±2.73, 28.03±3.09, and 6.90±2.81, respectively, after intervention (P<0.001) (Table 2).

Table 3 shows the mean knowledge score for PFME on each questionnaire at pre- and post-intervention. The percentage of correct answers and changes in mean knowledge scores are also shown. A significant change was seen in almost all items except for the importance of PFME in sexual activity.

Table 4 shows the mean attitude score and the change in mean attitude score by item at pre- and post-intervention. A significant change in attitude was seen from this intervention.

The mean practice score and the change in mean practice score is shown in Table 5. Table 6 shows factors associated with the change in practice score after intervention, using multivariate linear regression.

The ethnicity of Malaysia is very diverse but the main population is Malay, Chinese, and Indian. Most Chinese and Indians populate the western part of Peninsular Malaysia, whereas Kelantan, situated on the east part of the peninsula, is populated mainly by Malays. A feature of this study is over-representation by the Malay population (98.3%), as the study was done in Kelantan. Therefore, comparison of data cannot be done and the results could not be generalized to all ethnic groups in Malaysia. Similarly, a study in Australia also showed low numbers of non-Caucasian respondents at only 1.3% [15], and 6% in Canada [17].

Respondent mean±SD age was younger at 31.63±5.26 years, as the study was done during pregnancy. This was similar to another study, in which the mean±SD age of respondents was 27.40±6.0 years [15]. Most studies on PFME were done in a non-pregnancy period, in which the main objective was to determine the effect of PFME on UI or KAP in those with UI. Therefore, the age groups were older. Liao et al. [24] in 2006 reported a mean±SD age of 50.3±10.92 and Geoffrion et al. [17] in 2009 reported a mean±SD age of 61.6±11.0.

Of the 121 respondents, 18.2% were professional workers (doctors, lawyers, teachers accountants, architects, etc.), 45.5%% were non-professional workers (clerks, nurses, medical assistants, paramedics), and 29.9% were housewives and others (waitresses, cleaners, babysitters). The proportions of respondents with primary and secondary school educations and those with education levels beyond secondary school were similar. In Malaysia, education is provided until 17 years of age. The proportions with education up to and beyond secondary school in this study were similar to those reported by Whitford et al. [15] However, Geoffrion [25] and Subak et al. [26] reported that a majority of respondents had an education level more than high school. Monthly income was categorized into five levels, based on the poverty line index for West Malaysia in the 9th Malaysia Plan. Only 9.9% of participants were categorized as poor as they had income less than Ringgit Malaysia (RM) 720.

Most respondents in this study were multiparous or grand multiparous, at 56.25% and 29.8%, respectively. This again was expected, as many Malaysian women have more children, in contrast with the findings by Whitford et al. [15], who reported that the proportion of multiparas and primigravidas was the same. This is important, as pregnancy is a major risk factor for PFMD. Being parous, respondents were assumed to know about or practice PFME to a certain extent, but only about half (49.9%) of the respondents in this study had baseline information on PFME before intervention. Many had obtained information from reading. A higher percentage was found by Whitford et al. [15], who found that 90% of antenatal women had some knowledge of PFME from reading magazines and books.

In this study, the mean KAP score was significantly increased at 2 months after intervention (Table 2), showing that an educational class on PFMD and PFME for antenatal mothers is important. Similar improvement of knowledge after an educational class was observed in other studies discussed below. However, few studies on PFME knowledge are available for comparison, and only a few examined the effect of education on PFME knowledge.

The knowledge score in the present study improved for most respondents in all aspects of the questionnaires (Table 3). The highest correct answer before intervention was on the importance of pelvic muscles on sexual activity (84.3%). After the intervention, all respondents (100%) answered correctly that PFME may treat or prevent UI and that the exercise can be done at any time.

The highest change in knowledge score was in the response to “PFME is important in controlling urinary bladder function.” The second and third highest score changes were on methods of performing PFME, i.e., “PFME should be done at least 3 times a day, in the morning, afternoon, and night,” and “PFME can be done during routine daily activities.” Other high knowledge score changes were in response to “muscles involved in PFME should be contracted for 8 seconds before being release.” From the highest knowledge score changes, we can conclude that the majority of respondents had poor knowledge of the benefits of PFME in preventing or treating UI, even though many had baseline information on PFME. They also had a low level of knowledge on methods of performing PFME before the intervention. Many were not aware that PFME can be done at any time or place, and in any position.

This study result was supported by Geoffrion et al. [17] Knowledge on pelvic floor health and quality of life improved after 2.5 hours in a pelvic floor health workshop [17]. It was noted that knowledge scores were significantly higher immediately after the workshop (P<0.001) and 3 months late (P< 0.001), when compared with scores before the workshop. Baseline symptoms and quality of life scores significantly improved at 3 months.

Liao et al. performed a study on Taiwanese women with UI. Self-administered questionnaires were completed before and at 8 weeks after a 4-hour educational program that included training on PFME. There was a significant increase in the knowledge score and a decrease in UI symptoms with frequent practice [24].

Another study in Michigan, United States, evaluated the impact of a 2-hour group educational program on the knowledge and practice of PFME among women without PFMD. The participants were followed up after 12 months, and 86% correctly completed the questionnaires [27]. However, as no pre-intervention questionnaires were given in that study, no improvement in the knowledge score can be confirmed and it is unclear whether the high percentage of correct answers was due to the educational class.

Although, the practice of PFME was low in the present study, the majority of participants had a positive attitude toward learning. In pre- and post-intervention studies, the highest score was for the same item, i.e., “PFME should be done by all women, especially pregnant and postnatal women, whether or not they have PFMD,” with mean scores of 3.21 and 3.70, respectively (Table 4). The highest attitude score changes after intervention were in response to “I should practice PFME to prevent uterine prolapse,” followed by “I should practice PFME to prevent or treat UI.”

The majority of them agreed that all women should practice PFME, even in the absence of PFMD. They also agreed that PFME should be taught to all antenatal mothers by healthcare providers. In contrast to the findings of Liao et al. [24], there was no change in attitude after a 4-hour educational program. Education can change attitude towards PFME and is another key factor that can promote patient motivation and improve compliance with PFME practice. Women also need to know that there is no shame in having PFMD and should seek treatment [28].

One study aimed at improving healthcare provider KAP on reproductive health in Romania showed a significant increase in scores on four out of five selected training themes after a 5-day training course. Improvements from pre-test to post-test were observed in the areas of prenatal care, healthy lifestyles, and breast cancer screening [29]. Even though the study was not on PFMD, it showed that KAP in all aspects of medical care can be improved with an intervention program.

Although the majority of respondents were multiparous in this study, only 5.8% practiced PFME regularly before the intervention (defined as at least 3 times a week, i.e., frequently or always). After the intervention, more respondents practiced PFME regularly, at 37.2% (Table 5). Many factors accounted for a low practice score before the intervention, including lack of information or encouragement from medical staff in promoting PFME during antenatal checkups or before discharge from the postnatal ward.

Before the intervention, most respondents never discussed PFME with friends and never sought information regarding PFME. After the intervention, most began to perform PFME and even performed PFME after delivery. The highest practice score change was in response to “I spent time performing PFME,” followed by “I performed PFME when I was not pregnant.”

Our findings are similar to those of previous studies. Less than 17% of antenatal mothers performed PFME at least once a week during pregnancy [14] and only 54% of participants practiced PFME during pregnancy despite the high number with knowledge of PFME [15]. Chiarelli et al. [18] found that 91% of responders had good knowledge of PFME but only 15.7% performed PFME before pregnancy, while 54.5% had done so during pregnancy.

This study showed that respondents without baseline information on PFME before intervention had a significantly higher practice score change than those with baseline information (Table 6). Women who knew about PFME before the intervention were not interested in practicing PFME more, as they were not motivated by the information given or may had tried PFME in the past with no effect [15]. Baseline information was only reported in a few other studies [15,18]. Most of their respondents had a higher percentage with PFME information but with low practice rates. No practice change was observed in those cross-sectional studies that were limited to knowledge and practice.

Other factors including age, race, occupation, income, education level, gravida status, presence of UI, and level of sexual satisfaction were not found to be significant with regard to practice score changes. This again supports the need for healthcare professionals to educate patients in clinical practice. As we do not know which patients have baseline information on PFME, we need to target all groups of women regardless of sociodemographic background. All women need to attend PFME class during antenatal checkups.

In this study, a higher education level was not associated with greater changes in practice score, similar to the findings of Alewijnse et al. [30] This contradicts the findings of Bo et al. [14], who studied the association between practice of PFME and demographic data. The authors found that those with more education were more likely to practice PFME, while those who smoked daily or were delivered by caesarean section were less likely to exercise [14]. In our study, smoking data were not included, because of the low prevalence of female smokers in Kelantan. We also did not assess the type of delivery that could predispose our respondents to a higher practice score change. In another study, women who reported practicing PFME were also significantly more educated, older, and in paid employment [15].

As many of the respondents were parous antenatal mothers, the probability of having UI or any PFMD was expected to be high. However, the results indicated that being parous was not associated with greater practice of PFME [30]. Similarly, Whitford et al. [15] reported that higher parity was not a good indicator of PMFE practice. However, Bo et al. [14] found that having more children is associated with more PFME practice.

The study result was a good indicator of the need for antenatal women to learn and practice PFME. Education is effective in improving KAP of PFME. We recommend greater effort toward increasing awareness of PFME among antenatal women. As healthcare providers, we need to emphasize PFME during antenatal visits as preventive and therapeutic for PFMD. Since a majority of antenatal classes in Malaysia still do not implement PFME programs, we recommend development of a good clinical practice guideline for PFME, to ensure that all healthcare providers are well trained and informed for the benefit of the public.