Stress levels in Portuguese 1st year medical students
Ana Filipa Marques; Ana Rita Soares; Carlos Filipe Neves; Emanuel Brilhante; Joana Filipa Tavares; João Pedro Freitas; Mafalda Correia; Mariana Gomes; Pedro Tavares; Vítor Silva.
Prof. Dr. Armando Teixeira-Pinto. Class 21
ABSTRACT
Stress is a problem that affects students, especially medical students who are known for their stressful life. In the case of first year medical students, levels of stress may be in fact high. This study may be considered innovative since it pretends to identify general causes of stress and not specific factors related to school environment or type of learning.Besides that, we consider that focusing on first year students, we may analyse the stress factors of a special population, in which is present an important stress factor - the transition to college.
Therefore, this study pretends to assess stress levels among first year medical students and identify some factors related to those levels.
Accordingly, we have defined the 280 first year medical students of Faculty of Medicine of Porto University as our sample.To find out the incidence of stress in those medical students, they were asked to complete a questionnaire – General Health Questionnaire-12 – and some questions about factors related to stress. Then, the results of this cross-section study were statistically analyzed, using SPSS program.Therefore, wecreated a histogram to evaluate if the variable stress followed a normal distribution. As the results were unclear we proceeded with a Kolmogorov-Smirnov test, and concluded that the data doesn’t follow a normal distribution.But, since Kolmogorov-Sminorv test is very sensible we analyzed variables using t-tests. As we didn’t obtain any statistical significant values, we decided to use non-parametric tests, namely Mann-Whitney U and Kruskal-Wallis. We also didn’t obtain any results, so in our final analyses we categorized the levels of stress in three categories (low, medium and high or very high stress). Finally we used frequencies tables as summary measure and analyzed the results through crosstabs and chi-square tests, considering 0,05 our statistical significance level.
Allthe chi-square tests showed that there is no significant relation between the variables (gender, attending for the first time to an undergraduate course, being away from home, living alone, practicing extracurricular activities, having a boyfriend/girlfriend, being a part-time student, smoking, hours of study, hours of sleep and zodiac sign) and the levels of stress, since p value is always higher that 0,05.
In conclusion, the percentage of students with low and mediumstress levels is higher than the percentage of students with high or very high stress levels of stress and all the variables that were analyzed didn’t show any significant relation with stress levels.
KEY-WORDS: Stress; Medical; Students; University; Porto.
INTRODUCTION
Stress is the body’s specific answer – emotional, physical or cognitive – when it is exposed to an external stimulus, which requires a greater response than the one it is used to giving. Nevertheless, stress can be a positive factor and it is always present, although its intensity changes: it is very low when a person is resting or sleeping and extremely high after a severe accident.
In this manner, society’s concern about stress has been increasing because, nowadays, this problem affects a very large number of people and is felt earlier in life. Therefore, it is in our best interest to analyse this subject. In this way, this project will broach one of the most stressful factors in young people – entering University.
In fact, it is important to study the levels of stress of first year university students because entering and attending a degree at university implies huge changes concerning responsibility, working habits as well as transformations in the youngsters’ social life. Considering all this, some studies(1,2,3,4,5) point out the amount to learn, the time amounts of the course and examinations as factors that increase stress levels. On the other hand,uncertainty about individual study behaviour, fear of failure and inadequacy in social activities are considered important psychological factors. Lastly, inadequate time for relaxation and for practising other activities is also considered by university students as stressful factors. Considering all this, it is easy to understand the intimate relation between high stress levels and mental diseases, such as depression(6,7), as well as with weight changes(8), both loss and gain. Therefore, it is essential to know whether or not there are high stress levels in first year university students.
On the other hand, we chose to focus only on medical students, since this course is, at first glance, extremely stressful, because not only does it require many hours of study, but also because students have to develop interpersonal skills and self-confidence, in order to be “fully”-qualified doctors in the future.
In this way, there have been made, as we have already mentioned, several studies about this subject in other countries. Also in Portugal, it has been made a study to evaluate the stress levels in first year medical students of Faculdade de Medicina da Universidade do Porto(9). So, our work is innovative not because it measures the stress levels of Portuguese students. It will be innovative in the type of stress factors that are analysed. In fact, the study previously made measured not only factors of a more general character, like sleeping hours and the changes associated to the transition to college, concerning autonomy, demands and responsibility, but it also focused very specific factors. Examples of these factors are the evaluation system, FMUP's environment and the fact that the type of education/learning is focused on memorization. This is shown on the graphic below, which is a synthesis of a graphic of this study.
So, our study is innovative because we analyse only general factors, which allow us to broad the results to a larger population. This is very important because the study reaches and is interesting to a large number of people.
Other than that, this study was about students of all years, while our study analyse only first year students. We believe that focusing our study on this population, we will be able to analyse the stress factors of a very unique population in which its present a very important stressor factor – the transition to college.
RESEARCH QUESTIONS AND AIMS
The study that is being developed has two main objectives. The first one is to assess the stress levels of first year medical students in a Portuguese medical school. The second objective is to identify some factors that might be related to academic stress, namely: gender, being away from his normal residency, attending for the first time to an undergraduate course, having extracurricular activities, having a partner, being a part-time student, amount of study hours, smoking status, living alone and sleeping hours.
Our study is of great interest because, although it has been made at least one study about stress levels in Portuguese students, our project has specific characteristics that make it unique and important to analyse a population that, according to our research, has never been properly studied.
PARTICIPANTS AND METHODS
Study participants
Regarding the participants, the target population is first year medical students. Accordingly, we have defined the first year medical students of Faculty of Medicine of Porto University as our sample.
In what concerns sampling methods we chose a non-randomized method, specifically convenience sampling. In our study the inclusion criteria was being a first year medical student. No exclusion criteria was defined.
Study design
Concerning the study design, our study is a cross-sectional study.
Data collection methods
In order to collect the data, we used the General Health Questionnaire-12 (GHQ-12)(10). This questionnaire is a resumed version of the General Health Questionnaire-28 and accesses the general state of the health of an individual, being the majority of the questions about the psychological state. The English version of this questionnaire was chosen since it is validatedand scored. The translation was made separately by two translators and then the two versions were confronted with each other. The final translation was also compared with a translation that had already been made in another study(9). Finally, we added to the questionnaire the general questions related to the variables in study. Before the application of the questionnaire to the sample, it was made a pilot-application in a sample of ten individuals that are not part of the sample.
Regarding the score of the questionnaire, the answers of every question are scored from 0 to 3. So, to obtain the level of stress we some them up. Than, the value obtained is put into one of the three categories defined from the questionnaire: less or equal to 10 means low stress; from 11 to 15 means medium and more than 15 means high or very high stress.
Variables description
In relation to variables, the dependent variable is the stress level. In what concerns the independent variables, our aim is to find out if they have a significative relation with stress levels. Therefore, they can be considered either causes or consequences of stress.In fact, these independent variables are factors as gender, being away from his normal residency, living alone, being a part-time student, amount of study hours, attending for the first time to an undergraduate course, smoking status and sleeping hours. Finally, having a partner and extracurricular activities are factors that give us these two perspectives because, on one hand, they can be causes of stress by demanding time and dedication and, on the other hand, they can be factors that minimize stress by being a distraction, allowing the abstraction from questions related to the course. We also added to the questionnaire a question about the zodiac sign that will work as a control factor.
We chose to apply the questionnaire during the class of Introduction to Medicine. In order to do this, we asked the collaboration of the professors to deliver the questionnaire in the first three minutes of the class. This application was made in the first two weeks of March.
Statistical Analysis
The statistical analysis was performed using SPSS for Windows. Descriptive methods were used to summarize the data. Since all our variables are categorical, we used absolute and relative frequencies.
Firstly, we wanted to find out if the variable stress could follow a Normal distribution. So, we built a histogram and the result wasn’t conclusive enough. Thus, we applied to the Kolmogorov-Smirnov test, which lead us to reject H0 (H0: the variable stress follows a Normal distribution), because the value of p (significance level) was lesser than 0,05 (p=0,001). But due to the very high sensibility of the Kolmogorov-Smirnov and the presence of some stress values that had great deviation to the mean we analysed the stress level as a variable with a normal distribution. This analysis didn’t show us any significant relation between the stress levels and the variables in study. We also consider the variable as a categorical one, which was divided into four categories accordingly to the questionnaire (low stress levels, medium stress levels, high stress values and very high stress values). This way, we analysed the results through crosstabs and a chi-square. This analysis showed that there was a significant relation between the stress levels and the variables “Having a boyfriend/girlfriend” and “Hours of study”. But, due to the low frequencies of students in the high and in the very high categories we decided to collapse these two categories in one category (high or very high stress levels) so that the statistical analysis could be more solid and to prevent statistical errors. Consequently we performed our analysis of the results through crosstabs and chi-square tests, using only the final three categories.
Besides considering stress as a categorical variable, we decided to analyse it as a continuous one, using non-parametric tests. Related to the variables gender, attending for the first time to an undergraduate course, being away from his normal residency, living alone, having extracurricular activities, having a partner, being a part-time student and smoking status, we used Mann-Whitney U test to verify if there was any relation between these variables and the stress levels. To verify if there was any relation between the variables amount of study hours, sleeping hours and zodiac sign and the stress levels, we used Kruskal-Wallis test.
At the end, we verified that there was no relation in both situations. Thus, we decided to consider the stress levels a categorical variable, with the high and the very high categories collapsed, so that we could achieve more consistence in the applied method.
RESULTS, TABLES AND GRAPHICS
After the questionnaire delivery, there were obtained 236 answered questionnaires, out of the 280 first year medical students.
As mentioned before, the stress levels were divided in three categories. The distribution in these categories(Table 2 and Graph 2) is: 38,6%(N=91) of the students have low levels of stress, 37,3%(N=88) have medium levels and 23,3%(N=55) have high or very high levels of stress. Besides this there were two missing values (0,8%).
Frequency / PercentageValid
values / Low levels of stress / 91 / 38,6
Medium levels of stress / 88 / 37,3
High or very high levels of stress / 55 / 23,3
Total / 234 / 99,2
Missing / 2 / 0,8
Total / 236 / 100,0
Table 1 – Frequency of stress categories
Gender
Starting by the variable gender, the frequencies table (Table 2) showed that 64.4% of the students are female (N=152) and that 35.6% are male (N=84).
Frequency / PercentageValid
values / Male / 84 / 35,6
Female / 152 / 64,4
Total / 236 / 100,0
According to the crosstab (Table 13) we verify that the percentage of male and female per each stress category is similar. Consistent with, the chi-square test showed that there is no significative relation between the gender of an individual and the levels of stress, since p>0,05 (p=0,972). Therefore we accept the H0: Levels of stress on male are equal to levels of stress on female.
Attending for the first time to an undergraduate course
Related to the variable attending for the first time to an undergraduate course, the frequencies table (Table 3) showed that 86.4% of the students are away from home (N=204) and 13.6% are not (N=32).
Frequency / PercentageValid
values / Yes / 204 / 86,4
No / 32 / 13,6
Total / 236 / 100,0
The crosstab (Table 13) allow us to verify that the percentage of students who aren´t for the first time at an undergraduate course is higher in the low level of stress category. On the other hand, the percentage of students who are for the first time at an undergraduate course is higher in the category of medium levels of stress. In the high or very high category we observed that the percentage of students who are for the first time at an undergraduate course or not is approximately the same. The chi-square test shows that there aren’t significative differences, since p>0,05 (p=0,565). Therefore, we accept H0:Levels of stress on students that are attending for the first time to an undergraduate course or not are equal.
Being away from home
Related to the variable being away from home, the frequencies table (Table 4) showed that 51,3% of the students are away from home (N=121) and 48,7% are not (N=115).
Frequency / PercentageValid
values / Yes / 121 / 51,3
No / 115 / 48,7
Total / 236 / 100,0
The crosstab (Table 13) show us that there is a higher percentage of students who aren’t away from home in low stress levels, however there is a higher percentage of students who are away from home with high or very high levels of stress. In the medium category of stress the percentage of students who are away from home or not is similar. The chi-square test allows us to verify that this differences aren’t statistically significative, since p>0,05 (p=0,492). Therefore, we accept H0:Levels of stress on students that are away from home or not are equal.
Living alone
In what concerns to this variable, the frequencies table (Table 5) showed that 10.6% of the students live alone (N=25), while 89.4% don’t (N=211).
Frequency / PercentageValid
values / Yes / 25 / 10,6
No / 211 / 89,4
Total / 236 / 100,0
The crosstab (Table 13) allow us to verify that there are some differences between students who live alone and not. There is a higher percentage of students who don’t live alone in the low level of stress category. In the high or very high stress category there is a higher percentage of students who live alone. In the medium stress level category the percentage of students who live alone and those who don’t are approximately the same. The chi-square test showed us that those differences are not statistically significant, because p>0.05 (p=0.543). So, we accept H0: the levels of stress on students who live alone are equal to the levels of those who don’t.
Having an extracurricular activity
Regarding the variable having an extracurricular activity, the frequencies table (Table 6) showed that 59,3% of the students practice an extracurricular activity(N=140), while 40,7% don’t (N=96).
Frequency / PercentageValid
values / Yes / 140 / 59,3
No / 96 / 40,7
Total / 236 / 100,0
Looking at the crosstab (Table 13), we see thatthe percentage of students who practice an extracurricular activity and those who don’t per each stress category is similar. Consistent with, the chi-square test showed that there is no significative relation between this variable and the levels of stress, since p>0,05 (p=0,994). Therefore we accept the H0: Levels of stress on students who have an extracurricular activity and on those who don’t are equal.