Background

Tuberculosis (TB) is one of the most common infectious diseases worldwide and continues to be a major public health problem for low and middle-income countries. Several effective strategies have been implemented by the World Health Organization (WHO) to prevent and control the disease, including the directly observed treatment, short-course (DOTS) and stop TB strategies [1, 2]. The Millennium Development Goals (MDGs) [3], which were developed to halt and reverse the TB incidence by 2015, have been achieved, with a total reduction of 18% worldwide, and the death rate is half that reported in 1990 [4]. Despite these gains, the epidemic remains serious. The Global TB Report 2017 [5] published by the WHO reported that, although the fight against TB was paying off, 1.674 million people died from TB in 2016, and the disease ranked as the leading cause of death from a single infectious agent above HIV/AIDS. The report also showed that China, India and Indonesia alone accounted for 45% of global cases in 2016 and that China had a high TB burden even though the country had achieved the MDGs to reduce the prevalence of smear-positive TB by 50% in 2010 [6].

TB trends are influenced by control programmes as well as by biological, social and economic factors [7]. Undoubtedly, a lack of knowledge regarding TB among health care workers may contribute to an increased risk of developing the disease [8, 9]. A study of Chinese TB patients showed that a TB diagnostic delay was related to health facility staff factors, including inadequate TB knowledge, inability to prescribe a smear test for suspected TB cases, inability to refer suspected TB cases to county TB dispensaries or designated hospitals for TB care and misdiagnosis [10]. In this context, TB knowledge among medical students is particularly important; thus, undergraduate training regarding TB should be strengthened, because students may face significant exposures and consequently have the highest risk of infection or disease [11]. Furthermore, because medical students are potential future physicians and leaders, these students need to understand the epidemiology, determinants, screening and management of TB to promote effective prevention, early diagnosis, and successful treatment [12].

However, several studies [13,14,15] have reported that the TB knowledge of undergraduate medical students is insufficient. Medical colleges play an important role in training and shaping the attitudes of the future generations of medical practitioners [14]. Although TB health education in schools was emphasized at all times in China, researchers [16, 17] found that TB knowledge and practices among medical students were generally inadequate. A dearth of data regarding TB knowledge exists among medical students in China. Moreover, Hunan Province reported a notifiable TB incidence rate of 83.0 per 100,000 populations, which was the fourth highest incidence in China [18]. With this background, this study was conducted to survey 1088 preclinical undergraduate students from three medical universities in Hunan Province with the objective of assessing TB knowledge among final-year medical students and analysing the related determinants. Our findings can serve as baseline information for TB knowledge and help improve future TB prevention and control strategies in the region.

Methods

Study population

A cross-sectional study was conducted among three main undergraduate medical universities located in Changsha, which is the capital of Hunan Province, China, from March to June of 2016. Final-year medical students were recruited, because they had taken the infectious disease course prior to this investigation.

The questionnaire

The questionnaire was designed based on the report of Laurenti [19] and adjusted appropriately. The questionnaire focused on two dimensions: attitudes and experiences and knowledge of epidemiology and prevention, diagnosis and treatment. Each question had five possible answers, of which only one was correct. The students were asked to circle the most appropriate answer independently without receiving any help completing the questionnaire. All students participated on a voluntary basis and were not remunerated for their contribution.

Data analysis

The information was recorded in a database using EpiData 3.0 and analysed using SPSS 18.0. The data are reported as percentages. The total score for all questions and separate scores for epidemiology and prevention, diagnosis, and treatment were calculated. In addition, the mean percentage scores in different subpopulations were calculated based on demographic variables, experiences, and attitudes. Descriptive statistics and t-tests were used to analyse differences in different subpopulations. A multivariable linear regression analysis was performed to analyse the determinants associated with epidemiology and prevention, diagnosis, and treatment on TB knowledge.

Results

General characteristics of the participants

A total of 1200 undergraduate medical university students were surveyed, of whom 112 were excluded due to incomplete responses or logical errors. A total of 1088 valid questionnaires were received, for a response rate of 90.7%. Table 1 shows the characteristics of the participants. A total of 435 (40.0%) and 653 (60.0%) of the respondents were male and female, respectively, resulting in a gender ratio of 1:1.5. The mean age was 23 ± 0.92 years, and the ages ranged between 20 and 25 years. Only 160 (14.7%) of the students had received a PPD (purified protein derivative) test, and only 77 (10%) had performed at least one PPD test. More than 80% of the students had observed at least one case of TB and an X-ray of a TB patient (n = 891, 81.9% and n = 929, 85.4%, respectively). Approximately one-fourth (n = 293, 26.9%) of the students were aware of being at risk for TB infection.

Table 1 Demographic characteristics, experiences and attitudes of participants
Full size table

Knowledge about the epidemiology and prevention, diagnosis, and treatment of TB

The total mean percentage of correct answers for TB knowledge was 44.4% (SD 13.5%). The mean percentages of correct answers were 52.5% (SD 16.8%) for epidemiology and prevention, 35.7% (SD 16.1%) for diagnosis, and 47.5% (SD 22.7%) for treatment. In the epidemiology and prevention dimension (Table 2), the mean score for “BCG (bacille Calmette-Guerin) is a vaccine for TB” was relatively high at 90.6%, whereas the mean score for “The Mycobacterium avium complex is the most frequent aetiologic agent of TB” was lowest at 6.4%.

Table 2 Mean percentage of correct answers in epidemiology and prevention
Full size table

In the diagnosis dimension (Table 3), the mean score for “A sputum test is necessary to detect suspected pulmonary TB” was highest at 69.9%, whereas the mean score for “Koch bacillus identification is always performed using a Ziehl-Neelsen stain” was lowest at 8.8%.

Table 3 Mean percentage of correct answers in diagnosis and treatment
Full size table

In the treatment dimension (Table 3), the mean score for “Penicillin G is not useful for TB treatment” was highest at 74.6%, whereas the mean score for “TB patients are supposed to be hospitalized for effective treatment every 7-10 days” was lowest at 19.2%.

Determinants of TB knowledge

Table 4 shows the lack of an association between gender and the mean percentage of correct answers. The older medical students (≥23 years) had more knowledge than the younger medical students (< 23 years) regarding epidemiology and prevention (53.6% vs 49.6%, p = 0.001), diagnosis (37.2% vs 31.7%, p < 0.001) and treatment (49.0% vs 43.5%, p < 0.001) and higher total scores (45.7% vs 40.8%, p < 0.001). Medical students who reported receiving a PPD test had a higher percentage of correct answers in the total score (46.3% vs 44.0%, p = 0.047); conversely, no association was found between whether medical students reported performing at least one PPD test and the mean percentage of correct answers. Medical students who reported observing at least one TB case and one X-ray for a TB case had higher percentages of correct answers for epidemiology and prevention (54.4% vs 43.9%, p < 0.001; 54.3% vs 42.1%, p < 0.001), diagnosis (37.2% vs 29.0%, p < 0.001; 37.1% vs 27.5%, p < 0.001), and treatment (50.0% vs 36.0%, p < 0.001; 49.5% vs 35.7%, p < 0.001) and higher total scores (46.2% vs 36.2%, p < 0.001; 46.0% vs 34.7%, p < 0.001). Medical students who were aware of themselves being at risk for TB reported a similar percentage of correct answers with the other students.

Table 4 Mean percentage of correct answer by demographic characteristics, experiences and attitudes of participants
Full size table

The multivariable linear regression analysis (Table 5) identified an association between observing at least one TB case and an X-ray of a TB case and greater knowledge regarding epidemiology and prevention (β = 5.6, 95% CI: 2.4, 8.9; β = 8.2, 95% CI: 4.6, 11.8), diagnosis (β = 3.9, 95% CI: 0.8, 7.1; β = 5.7, 95% CI: 2.2, 9.2), treatment (β = 10.1, 95% CI: 5.6, 14.5; β = 7.0, 95% CI: 2.2, 11.8) and total scores (β = 5.5, 95% CI: 2.9, 8.1; β = 6.6, 95% CI: 3.8, 9.5). Moreover, older medical students (≥23 years) had more accurate knowledge regarding diagnosis (β = 3.9, 95% CI: 1.8, 6.1) and higher total scores (β = 2.8, 95% CI: 1.1, 4.6). In addition, medical students who considered themselves not at risk for TB were associated with greater knowledge about treatment (β = − 3.5, 95% CI: -6.5, − 0.6).

Table 5 Multiple linear regression analysis of determinants of the percentage of correct answers
Full size table

Discussion

We found that medical students had generally poor knowledge about TB. More than 50% of the total questions regarding TB knowledge were incorrect, which was consistent with another survey conducted in southwest China in 2011 [16] that reported that fewer than half the students had knowledge of TB symptoms of cough/blood-tinged sputum, local TB dispensaries and the national free TB treatment policy. Compared with the results reported by Laurenti [19], the total mean percentage of correct answers in our study was lower (44.4% vs 56.5%), as were those for epidemiology and prevention (52.5% vs 63.5%) and diagnosis (35.7% vs 54.1%); however, our results regarding treatment were slightly higher (47.5% vs 45.7%). Moreover, the general knowledge level for TB in the present study was inferior to other surveys conducted among medical students in Italy [11], Brazil [13], and National TB Curriculum Consortium (NTCC) schools [20], which suggested a considerable need for improvement in TB knowledge among medical students. Because this lack of knowledge may be an influencing factor for a higher TB prevalence, improved instruction in TB for medical students is necessary.

Medical students’ attitudes towards TB may play an integral role in how they manage TB patients during their careers [21]. However, our study indicated that the majority of medical students lacked an accurate TB attitude. Approximately one-quarter of the medical students considered themselves at risk of contracting TB, which was consistent with the results reported by Bhandari [22]. Nevertheless, in the subsequent multiple linear regression analysis, we observed a negative association between attitude and the level of knowledge of TB treatment. In other words, the medical students who believed they were not at risk of contracting TB had a higher level of knowledge of TB treatment. This finding disclosed that the medical students in our study lacked prevention awareness, although they had comparatively more TB knowledge. Therefore, we could infer that a relationship did not necessarily exist between knowledge and attitudes for TB and that higher levels of knowledge did not result in positive attitudes. However, this result might have been influenced by the sample size and questions asked. Admittedly, correct knowledge and positive perceptions towards TB were prerequisites to seek early prevention and control measures.

In our study, more than 80% of the medical students had observed at least one TB case and an X-ray of a TB patient; these students reflected a better grasp of TB knowledge. Furthermore, the multivariable linear regression analysis showed a significant association between observing at least one TB case and an X-ray of a TB patient and greater knowledge of epidemiology and prevention, diagnosis, and treatment and the total score. This finding suggested that clinical experience was important for promoting TB knowledge among medical students. This result might be related to the amount of curriculum time devoted to TB and the number of TB patients seen by the students [20, 23]. However, Hoffman [24] reported in 2016 that even among health professionals, a significant gap remained for improvement of knowledge and practices regarding TB.

We also found that older medical students (≥23 years) had more accurate TB knowledge about the diagnosis and total scores, leading to a positive association between age and the level of TB knowledge. Similarly, a study by Montagna [11] reported that correct answers to questions concerning TB and its vaccine were associated with increasing age. TB knowledge increased with age, probably because older medical students developed more correct attitudes and behaviours regarding the disease. The TB course schedule might also be responsible for this finding.

To appropriately address the specialized problems of TB prevention and control, education and competence need to be strengthened among health care providers. The U.S. National Institutes of Health (NIH) funded the NTCC in 2003 to develop educational programmes to improve TB knowledge and practices among health professions students [20]. China also published the National Guideline for TB control, which emphasized TB health education [25]. Compared to traditional teaching, modern training methods, such as student-based teaching methods (problem-based learning), lead to improved student satisfaction and provide additional learning opportunities [26]. Hence, the results of the present study highlight a demand to innovate modern educational methods to promote knowledge and social responsibility among medical students.

Limitations

Our study has several limitations. First, the questionnaire from Laurenti may not be comprehensive and appropriately adapted to the specific situation in China. Second, the representativeness of the sample needs to be improved further by increasing the sample size or adopting randomized sampling methods. Third, gauging a person’s attitude towards TB based solely on the question “I am at risk for TB” is insufficient. This issue should be improved in a future study. Fourth, due to the cross-sectional study design, the causal relationship between the TB knowledge level and determinants could not be explored; however, the findings provide a basis for acquiring and testing a causal hypothesis. Finally, some reporting bias may have been inherent in this study design.

Conclusion

The present study found poor knowledge of TB among final-year medical students and considerable room for improvement in terms of TB knowledge, attitudes and practices. Moreover, practical clinical experience, attitude and age were significantly associated with more TB knowledge through a multiple linear regression analysis. We present the current status of TB knowledge levels among medical students and promote the need for active-learning methodologies.