Knowledge, experiences, and attitudes of medical students in Rome about tuberculosis.

BACKGROUND: Tuberculosis is the second leading cause of death from infectious disease. Insufficient knowledge among doctors about tuberculosis is one of the reasons for the increased tuberculosis rates in several low-endemic countries. The purpose of this study was to assess knowledge, experience, and attitude about tuberculosis among medical students.
MATERIAL AND METHODS: After a pilot study, a cross-sectional survey was performed on fifth-year medical students at the Catholic University of Rome (Italy), using a self-administered questionnaire on attitude, experience and knowledge about epidemiology, diagnosis, and treatment of tuberculosis. The t test and multivariable linear regression analysis were performed to estimate the association between TB knowledge and investigated variables.
RESULTS: Among 220 fifth-year medical students, the response rate was 83.1%. The mean percentage of correct answers was 56.6% (63.5% for epidemiology and prevention, 54.1% for diagnosis, and 45.7% for treatment). Associations between internships in wards and greater knowledge of tuberculosis diagnosis (55.9% vs. 51.6%, p=0.02), treatment (48.4% vs. 41.8%, p=0.03) and total score (58.1% vs. 54.5%, p=0.04) were found. Students who reported receiving the Mantoux test had higher knowledge of tuberculosis epidemiology and prevention (65.4% vs. 53.3%, p=0.001), diagnosis (55.2% vs. 48.3%, p=0.005), and total score (58.0% vs. 49.1%, p=0.001). Students who had observed at least 1 active pulmonary tuberculosis case had a higher percentage of correct answers about diagnosis (55.5% vs. 51.4%, p=0.03) and total score (57.9% vs. 54.0%, p=0.03). The multivariable linear regression confirmed the association between higher knowledge and receiving the Mantoux test (beta coefficient=7.2; 95% CI 2.6-11.7), as well as having observed at least 1 X-ray of a TB patient (beta coefficient=5.3; 95% CI 1.0-9.7).
CONCLUSIONS: A moderate level of general knowledge about tuberculosis was found, which suggests the need to modify current programs of infectious diseases in the curriculum of medical schools.

Background

Tuberculosis (TB) is one of the most common communicable diseases world and continues to be a major global health problem. It causes disease among millions of people each year, and, after HIV, it ranks as the second leading cause of death from an infectious disease worldwide [1]. In the last 30 years, migration and emigration, negligence of TB control programs, and above all, the emergence of the HIV/AIDS epidemic, have produced an increase of TB rates in several countries [2,3].

Although the principles of TB management are well defined, this infectious disease is often not properly diagnosed and treated. Insufficient knowledge among doctors about TB is one of the reasons for this failure. Despite the abundance of information on all content areas of TB, clinicians still make frequent errors in TB treatment [4].

In this context, there is the need for accurate undergraduate training in TB and of a comprehensive educational strategy is essential to provide medical students with the appropriate knowledge, skills, and attitudes necessary for the effective management of TB to promote effective prevention, early diagnosis, and successful treatment.

In the USA, the National Institutes of Health funded the National Tuberculosis Curriculum Consortium (NTCC) to meet this need. The goal of the NTCC is to improve skills and instill appropriate attitudes in the management and control of TB among medical students in their formative years, and to establish a foundation by which complex issues relating to TB can be continually revisited throughout the span of their careers [5]. Several studies have investigated the knowledge and attitudes of medical students about TB. Emili et al. [6] compared knowledge and practices regarding TB among final-year medical students at schools in endemic and non-endemic areas (Canada, India, and Uganda), revealing significant differences in undergraduate knowledge and practice competency in regards to TB at 3 medical schools.

Kilicaslan et al. [7] conducted a study at the Istanbul Medical School to assess whether TB-related questions asked during chest medicine examinations complied with the World Health Organization’s (WHO) learning objectives for TB training and to investigate students’ skills in interpreting TB radiology and smears. The study showed that the examination questions did not adequately reflect WHO learning objectives, and that the students’ skills suggested that their practical training on TB was insufficient. Jackson et al. [8] performed a self-administered survey among students in NTCC schools to establish a baseline level of knowledge, attitudes, and confidence about TB, and they concluded that there was room for improvement in these 3 areas. In Brazil, Teixeira et al. [9] performed a cross-sectional survey of undergraduate medical students in preclinical, early clinical and late clinical years about previous lectures on TB, knowledge about TB transmission, exposure to patients with active pulmonary TB, and the use of protective respiratory masks. They concluded that many medical students were not aware of the main routes of TB infection and that lectures on TB were not sufficient to change knowledge and practices. Regardless of knowledge about TB transmission, students engaged in risky behaviors: more than two-thirds did not use a protective mask when examining an active TB case.

Little is known about the effectiveness of training programs for the treatment of TB in Europe. With this background, we performed a survey to assess knowledge, attitudes, and competencies about TB among fifth-year medical students at the School of Medicine of the Catholic University in Rome, Italy.

The aim of this study was 2-fold: 1) to determine knowledge, experiences, and attitudes of medical students regarding TB in a University Teaching Hospital and 2) to establish which individual factors are associated with a greater knowledge about TB.

Material and Methods

Study design

A cross-sectional survey was conducted at the School of Medicine of the Catholic University in Rome among fifth-year medical students. This group of students was selected because the course on infectious diseases takes place in the fourth year.

The survey was carried out with a self-administered questionnaire designed on the basis of NTCC contents. The questionnaire consisted of 39 multiple-choice questions divided by area: attitudes and experiences (7 questions), knowledge about epidemiology and prevention (13 questions), diagnosis (14 questions), and treatment (5 questions). Each question about knowledge had 5 possible answers, of which only 1 was correct. To test the questionnaire, a pilot study was conducted from April to June 2012 among second- and third-year midwifery students and first-, second-, and third-year medical residents in hygiene, occupational medicine, and forensic medicine. Thirty-seven students and residents were enrolled in this pilot phase and all completed the questionnaire. The pilot study allowed us to modify the wording of some questions.

The survey was conducted from July to October 2012 on fifth-year medical students. To increase the number of respondents, students were enrolled before being examined in Public Health. Students were introduced to the aims, objectives, and the methodology of the study, and they were invited to participate on a voluntary basis. After obtaining informed verbal consent, the students were asked to complete the questionnaire anonymously. They were given a maximum of 30 minutes to complete it.

Statistical analyses

An overall score was computed for all questions related to knowledge (total score). Separate scores were calculated for questions related to epidemiology and prevention, diagnosis, and treatment.

Mean percentage scores were computed for different subgroups on the basis of demographic variables, experiences, and attitudes.

Differences between these subgroups in mean percentage scores were tested using the t test.

Multivariable linear regression analysis was performed to estimate the effect on TB knowledge of each variable, adjusting for the effect of the other variables in the model. Multivariable regression models were built for epidemiology and prevention, diagnosis, and treatment, respectively.

Results of these models are shown using β coefficients with 95% confidence intervals (95% CI), which describe the expected change in the percentage of correct answers associated with each variable.

Results

Among 220 fifth-year medical students enrolled in the University, 186 students were present at the time of data collection. Of these, 183 (83.1% of fifth-year students) completed the questionnaire.

Table 1 shows the characteristics of the sample. The mean age was 24 years (SD 0.12, range 22–30). About half were female (52%, n=95), and more than half (58.5%, n=107) reported attending a clinical unit.

Table 1

Characteristics of the sample.

	N	%
Demographic
Gender
Male	88	48
Female	95	52
Age
Mean (Std. Dev.)	24 (0.12)
Internship
Yes	107	58.5
No	76	41.5
Experiences and attitudes
I received Mantoux test
Yes	154	84.2
No, I don’t know	29	15.8
I performed at least one Mantoux test
Yes	40	21.9
No, I don’t know	143	78.1
I observed at least one TB case
Yes	121	66.1
No, I don’t know	62	33.9
I observed at least one x-ray of TB patient
Yes	152	83.1
No, I don’t know	31	16.9
I think I am at risk for TB
Yes	122	66.7
No, I don’t know	61	33.3

Most of the sample (84.2%, n=154) reported receiving the Mantoux test. About one-fifth (21.9%, n=40) had performed at least 1 Mantoux test. Two-thirds of the students (66.1%, n=121) observed at least 1 TB case during their medical education and 83.1% (n=152) had observed at least 1 TB patient X-ray. Only two-thirds (66.7%, n=122) were aware of being at risk for TB infection.

Knowledge

Among the 32 multiple-choice questions designed to evaluate student knowledge of TB, the mean percentage of correct answers was 56.6% (SD 11.6%).

The mean percentage of correct answers was 63.5% (SD 16.3%) for epidemiology and prevention, 54.1% (SD 12.4%) for diagnosis, and 45.7% (SD 20.4%) for treatment. Figure 1 shows the distribution of the mean percentage of correct answers by area of knowledge.

Figure 1

Distribution of correct answers.

In the “Epidemiology and Prevention” area (Figure 2), the highest and the lowest mean scores were reported for the following questions, respectively: “the etiologic agent of TB is transmitted by ____” (98.9% correct answers); and “currently the prevalence of TB is highest in the following areas of the world: ____” (8.2% correct answers).

Figure 2

Mean percentage of correct answers in epidemiology and prevention area.

In the “Diagnosis” area (Figure 3), the questions with the highest and the lowest mean scores, respectively, were: “which organ could be affected by M. tuberculosis” (95.6%); and “The tine test is ____” (10.4%).

Figure 3

Mean percentage of correct answers in diagnosis area.

In the “Treatment” area (Figure 4), the questions with the highest and the lowest mean scores, respectively, were: “which of these antibiotics are not useful for TB treatment: ____” (82.0%); and “prophylactic treatment with isoniazid is implemented to ____” (13.1%).

Figure 4

Mean percentage of correct answers in treatment area.

Factors associated with knowledge

Table 2 shows that younger students (age <24) had on average greater TB knowledge (58.1% of correct answers vs. 54.7%, p=0.05), whereas no association was found between gender and the mean percentage of correct answers. There was a significant association between internship in units and departments and greater knowledge about TB diagnosis (55.9% vs. 51.6%, p=0.02), treatment (48.4% vs. 41.8%, p=0.03), and total score (58.1% vs. 54.5%, p=0.04).

Table 2

Mean percentage of correct answers by characteristics of the sample.

	Total		Epidemiology and prevention		Diagnosis		Treatment
	Mean score	p-value	Mean score	p-value	Mean score	p-value	Mean score	p-value
Gender
M	56.5	0.934	62.9	0.669	55.0	0.348	44.1	0.311
F	56.7		64.0		53.3		47.2
Age
<24	58.1	0.053	65.3	0.079	55.2	0.175	47.1	0.277
≥24	54.7		61.1		52.7		43.8
Internship
Yes	58.1	0.036	64.2	0.475	55.9	0.019	48.4	0.031
No	54.5		62.4		51.6		41.8
I received Mantoux test
Yes	58.0	0.001	65.4	0.001	55.2	0.005	46.6	0.151
No, I don’t know	49.1		53.3		48.3		40.7
I performed at least one Mantoux test
Yes	55.9	0.680	60.6	0.203	55.0	0.619	46.5	0.775
No, I don’t know	56.8		64.3		53.9		45.5
I observed at least one TB case
Yes	57.9	0.031	64.8	0.111	55.5	0.031	46.6	0.391
No, I don’t know	54.0		60.8		51.4		43.9
I observed at least one x-ray of TB patient
Yes	57.7	0.004	64.7	0.020	55.1	0.023	46.8	0.089
No, I don’t know	51.2		57.3		49.5		40.0
I think I am at risk for TB
Yes	57.6	0.109	64.7	0.152	54.6	0.452	47.4	0.112
No, I don’t know	54.7		61.0		53.2		42.3

Students who reported receiving the Mantoux test had greater knowledge about TB epidemiology and prevention (65.4% vs. 53.3%, p=0.001), diagnosis (55.2% vs. 48.3%, p=0.005), and total score (58.0% vs. 49.1%, p=0.001).

Students who reported observing at least 1 active pulmonary TB case had a higher percentage of correct answers about diagnosis (55.5% vs. 51.4%, p=0.03) and total score (57.9% vs. 54.0%, p=0.03).

Students who reported observing at least 1 X-ray of a TB patient had a higher percentage of correct answers about epidemiology and prevention (64.7% vs. 57.3%, p=0.02), diagnosis (55.1% vs. 49.5%, p=0.02), and total score (57.7% vs. 51.2%, p=0.004). Students who considered themselves to be at risk for TB reported a similar percentage of correct answers as other students.

The multivariable linear regression analysis (Table 3) confirmed the association between receiving the Mantoux test and greater knowledge about TB epidemiology and prevention (β coefficient=10.8; 95% CI 4.2–17.3), diagnosis (β coefficient=5.0; 95% CI 0.1–10.0), and total score (β coefficient=7.2; 95% CI 2.6–11.7). In addition, having observed at least 1 X-ray of a TB patient was associated with a higher total score (β coefficient=5.3; 95% CI 1.0–9.7). The analysis did not reveal any association between greater knowledge and gender, age, and feeling at risk.

Table 3

Results of the multivariable linear regression of the percentage of correct answers.

	Total		Epidemiology and prevention		Diagnosis		Treatment
	β coef.	95% CI	β coef.	95% CI	β coef.	95% CI	β coef.	95% CI
Gender
F	−0.1	(−3.3; 3.1)	1.0	(−3.6; 5.6)	−2.1	(−5.6; 1.4)	2.6	(−3.3; 8.5)
Age
≥24	−1.5	(−4.8; 1.8)	−2.4	(−7.1; 2.4)	−0.9	(−4.5; 2.7)	−0.9	(−7.0; 5.2)
Internship
Yes	2.8	(−0.5; 6.1)	0.9	(−3.9; 5.7)	3.5	(−0.1; 7.1)	5.8	(−0.4; 11.9)
I received Mantoux test
Yes	7.2	(2.6; 11.7)	10.8	(4.2; 17.3)	5.0	(0.1; 10.0)	3.8	(−4.6; 12.1)
I performed at least one Mantoux test
Yes	2.0	(−1.3; 5.3)	0.0	(−4.8; 4.8)	3.2	(−0.4; 6.8)	3.7	(−2.4; 9.8)
I observed at least one TB case
Yes	1.8	(−1.7; 5.3)	2.0	(−3.0; 7.0)	2.4	(−1.4; 6.2)	−0.3	(−6.7; 6.2)
I observed at least one x-ray of TB patient
Yes	5.3	(1.0; 9.7)	5.9	(−0.4; 12.1)	4.4	(−0.3; 9.2)	6.5	(−1.5; 14.5)
I think I am at risk for TB
Yes	2.4	(−1.1; 5.8)	2.5	(−2.4; 7.5)	1.3	(−2.5; 5.0)	5.1	(−1.2; 11.5)
Intercept	41.5	(34.8; 48.1)	46.5	(36.9; 56.1)	41.6	(34.4; 48.9)	27.8	(15.4; 40.1)

Discussion

This study showed that more than half of the questions about knowledge of TB treatment were wrong. This is in line with other surveys carried out among medical students [7-9]. One-sixth of the students did not remember receiving the Mantoux test, which was compulsory for the investigated medical students. Furthermore, medical students who remembered receiving the Mantoux test demonstrated a better grasp of knowledge about TB.

Specifically, we found an association between memory receiving the Mantoux test and greater knowledge of epidemiology and prevention, diagnosis, and the total score, which strengthens the role of the memory effect.

The relationship between recall of preventive medicine activities and greater knowledge about TB is consistent with previous studies that reported an association between HPV and influenza vaccination and a greater knowledge of those diseases [10-12].

We also found evidence of a significant association between greater clinical experience, indicated by those who had seen cases and a chest X-ray of a TB patient and/or those who had attended an internship in clinical wards and had greater knowledge of TB diagnosis and treatment. This may be related to the time devoted to personal insights about TB and the number of patients with TB seen by medical students [6].

In recent years, the role of Problem-Based Learning in medical education was highlighted by empirical studies that found strong evidence that this method had positive effects on student knowledge, stressing the benefits of Problem-Based Learning integrated within a traditional curriculum [13,14].

Appropriate practice and attitudes in the treatment of TB are among the main goals that can be achieved through the adoption of competency-based programs that are responsive to rapidly changing needs, as well as promoting a new instrument of learning that includes the capacity to integrate knowledge and good practices [15]. Studies like ours, which are designed to evaluate the training of medical students and to identify inadequacies in their education, are needed to implement changes because failure to address these issues can affect both patient and public health.

According to the WHO, medical professionals should know about national and international expansion of the TB burden, national TB prevention policies such as control strategies, and the BCG vaccination recommendations [16]. Knowledge of these topics was investigated in the present study and a poor level of knowledge was reported. In recent years, the Centers for Disease Control and Prevention and NTCC have proposed curricula offering the basics of training undergraduate students and doctors about the fundamentals of TB prevention, diagnosis, and treatment [5,17]. Our survey results highlight the need to develop educational tools using active learning strategies to improve knowledge and ensure social accountability of medical schools.

This study has some limitations as well as some strengths. The questionnaire was administered prior to the public health course examination; consequently, some medical students might have answered haphazardly because they were more concerned about the upcoming examination. However, this timing of data collection allowed us to obtain a very large sample and high compliance. On the other hand, the preparation for the public health exam may have increased the students’ general level of knowledge about TB.

We investigated not only comprehensive knowledge, but also experience and attitudes about TB among students and related these to knowledge. The cross-sectional design of this survey does not allow the evaluation of causal relationships, but it does make associations between several factors that represent, in any case, the basis for testing causal hypothesis in further studies.

Our survey represents a good starting point for future longitudinal studies aimed at assessing the impact of changes in curricula on student knowledge.

Conclusions

Our findings show a moderate knowledge about TB among fifth-year medical students. Internship was found to be strongly associated with a greater knowledge of TB diagnosis, epidemiology, and prevention.

We provide a clear snapshot of the differences in TB learning between the more sensitized and experienced students who apply what they have learned, and others who are more dedicated to theoretical learning.