European Journal of Dental Education ISSN 1396-5883
The relationship between sleep habits and academic performance in dental students in Croatia M. Valic1, R. Pecotic1, L. Lusic1, K. Peros2, Z. Pribudic1 and Z. Dogas1 1 2
Department of Neuroscience, University of Split School of Medicine, Split, Croatia, Department of Pharmacology, School of Dental Medicine, University of Zagreb, Zagreb, Croatia
keywords sleep; sleep habits; dental students; academic performance. Correspondence Maja Valic Department of Neuroscience University of Split School of Medicine Soltanska 2 21000 Split, Croatia Tel: +385 21 557 860 Fax: +385 21 557 955 e-mail: [email protected] Accepted: 18 November 2013 doi: 10.1111/eje.12081
Abstract Introduction: It is well accepted that sleep and lifestyle habits affect academic success in students. However, sleep patterns and sleep problems amongst dental students have been insufficiently addressed in the literature. The purpose of this study was to evaluate sleep habits of dental students and the relationship between sleep habits and academic performance. Materials and methods: A self-administered questionnaire on sleep habits, academic performance and lifestyle was administered. The participants were 447 dental students from Split University Dental Medicine School and Zagreb University Dental Medicine School from the six academic years. The subjects were classified into two groups based on academic success (high-performing vs. low-performing students) for comparison of sleep and lifestyle habits. Results: Amongst the whole group of students, average bedtime and wake time during weekday was significantly earlier compared with weekend. Main findings indicate that students with high academic performance had earlier bedtimes during weekdays and weekends, earlier wake times during weekends and shorter sleep latency compared with low academic performing students. Conclusion: Self-reported academic performance of dental students in Croatia is associated with timing of sleep and wakefulness, rather than with total sleep time duration.
Introduction Quality of sleep is essential to an individual’s health and wellbeing and has a substantial role in learning and memory process. Sleep habits are influenced by many internal and external factors such as school and work schedules, often resulting in daytime sleepiness (1). Furthermore, it has been shown that insufficient sleep is the main factor influencing mood and alertness and that sleep deprivation results in impaired neurocognitive and psychomotor performance (2). Recent studies have focused on sleep habits amongst university students. Increased psychological pressure and academic demands can lead to impaired quality and duration of sleep. Numerous studies have shown that students with more regular sleep–wake patterns report higher grade point averages (GPAs), whereas students with lower grades report
ª 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd Eur J Dent Educ 18 (2014) 187–194
increased daytime sleepiness as a consequence of decreased sleep duration (shorter sleeping nights) (3–5). Amongst different student populations, medical students received particular attention in sleep research because of high academic demands and increased workload (6–9). Several studies have determined that medical students’ unique academic commitments and lifestyle can impact their sleep habits and result in sleep deprivation (4, 6, 7, 10). Similarly, the doctor of dental medicine (D.M.D.) curriculum is demanding due to both declarative and procedural learning tasks, with declarative learning being more emphasised during the first years of schooling and procedural learning being added in the later years in dental schools in Croatia. The literature reveals that both declarative and motor procedural learning capabilities are impaired by sleep deprivation (11–13).
187
Sleep habits and academic performance
Valic et al.
Whilst it is well accepted that the role of sleep quality and lifestyle habits affects academic success (3, 4, 7), sleep patterns and sleep problems amongst dental students have been insufficiently addressed in the literature. Therefore, the purpose of this study was to determine sleep and lifestyle habits and to estimate their relationship with academic performance amongst dental students in Croatia.
Materials and methods This cross-sectional study was conducted between June and October 2012 at the University of Split School of Dental Medicine and the University of Zagreb School of Dental Medicine. The curricula of the University of Split School of Dental Medicine and the University of Zagreb School of Dental Medicine are governed by Bologna regulations, recommendations and principles in accordance with the university book of regulations for higher education and with the School of Medicine book of regulations. The protocol for this study was overviewed and approved by the Biomedical Research Ethics Committee of the Split University School of Medicine, Split, Croatia, and by the Ethical Committee of the School of Dental Medicine, University of Zagreb, Croatia. Total population at School of Dental Medicine University of Zagreb is 681 students, 516 women and 165 men. Total population at School of Dental Medicine University of Split is 150 students, 47 men and 103 women. The total number of students who participated in the study was 447 (110 men and 337 women), from six academic levels (Table 1). Participants’ data were collected anonymously and none were paid during the study.
TABLE 1. Characteristics of the student sample
Age Average grade University Split Zagreb Year of study 1st 2nd 3rd 4th 5th 6th Gender Male Female Afternoon sleep Yes No Sleep quality satisfaction Yes No Preference Wake up early Staying up late
188
n (%)
n (total)
23.5 (2.09) 3.91 (0.49)
446 432
142 (31.8) 305 (68.2)
447
31 85 135 81 55 60
(6.9) (19) (30.2) (18.1) (12.3) (13.4)
447
110 (24.6) 337 (75.4)
447
220 (49.7) 233 (50.3)
443
249 (55.7) 198 (44.3)
447
284 (64.1) 159 (35.9)
443
The design of the questionnaire used in this study was based on a previously published survey instrument (3) to assess demographics, sleep habits, sleep and sleep-related disturbances, sleep-related behaviour, lifestyle habits and intake of pharmacological substances. The Association of Sleep and Academic Performance questionnaire (3) included 30 questions on sleep habits, dreams, stimulants and general information. We added questions about sleep disturbances and lifestyle habits. The final version of the questionnaire consists of 58 questions. Demographics included questions on age, gender, constitutional parameters and academic success. Academic success was measured by the self-reported average grade. Questions on sleep habits included reports on bedtime and wake time during weekdays and weekends, ideal bedtime and total sleep time estimation. Students were also asked to answer how satisfied they were with the quality and the amount of their sleep. Questions on sleep-related disturbances included questions about nocturnal awakenings, daytime sleepiness, insomnia and breathing problems. The usage of pharmacologically active substances was assessed based on questions regarding smoking and coffee intake and consumption of prescription and over-the-counter medications. All questions were constructed as multiple choices (affirmative/negative and various multiple choices) and as open-ended questions. Students were also asked to answer lifestyle habits questions, such as evening social activities, alcohol consumption, the use of the TV and computer, and exercising and learning habits. Students were asked to report their usual learning time (hours per day and time of the day). Previously trained staff distributed a two-page questionnaire to all students who attended their classes. Staff was informed on research questions, questionnaire structure and a mandatory standardised instruction for students. Considering that students were approached during regular classes, the instruction was necessary to create standardised conditions and accentuate the importance of honest response. Students were informed that participation was anonymous and voluntary. They completed the questionnaire during regular classes at the Schools of Dental Medicine in Split and Zagreb. All students who attended their classes at the time of the survey collection participated in the study and returned completed questionnaires. The time required to complete the questionnaire was approximately 10–12 min. Method of sampling in this study was convenience sampling, meaning that all students who attended the classes at the time survey was conducted had to fulfil the questionnaire. All classes are obligatory for all students. The survey instrument was translated into the Croatian language by a psychologist and a medical doctor (somnologist). The questionnaire was then administered to a pilot group of ten students for feedback on question readability and clarity. The questionnaire was back-translated from Croatian into English by a bilingual professional translator for comparison with the original text. The subjects were classified into two groups based on academic success (high-performing vs. low-performing students). Descriptive results were expressed as frequencies, percentages and mean values with deviation measures. Data were consequently analysed by testing group differences using the chisquare test for categorical and Mann–Whitney nonparametric ª 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd Eur J Dent Educ 18 (2014) 187–194
Valic et al.
test for continuous variables. Data of all subjects were analysed using stepwise regression analysis. All data were entered and analysed using the Statistical Package for Social Sciences program (version 14; SPSS Inc., Chicago, IL, USA). P values less than 0.05 were considered statistically significant.
Results There were 447 dental students who participated in this study. Table 1 summarises demographic characteristics and some other study variables related to sleep. For 432 students responding to the question regarding academic performance, the average GPA was 3.91 0.49. The scholastic performance was stratified as high (highest quintile, GPA > 4.3) or low academic performances (lowest quintile, GPA < 3.5), with 107 students (24%) being classified as high-performing students and 129 (29%) students classified as low-performing students. Amongst the whole group of students, average bedtime during the weekday was significantly earlier compared with the weekend (23:53 0:55 vs. 01:23 1:38; P < 0.001; Fig. 1). Similarly, average wake time during the weekday was earlier compared with the weekend (7:02 0:44 vs. 10:03 1:27; P < 0.001; Fig. 1). For all students, the majority (55.7%) reported satisfaction with their sleep and 64.1% expressed a preference to wake up early in the morning. When the relationship between sleep habits and school performance was analysed, the high-performing group had an earlier bedtime compared with the low-performing group during
Sleep habits and academic performance
weekdays (23:44 0:49 vs. 00:02 1:00; P = 0.004; Fig. 2) and weekends (01:08 1:23 vs. 01:39 1:38; P = 0.016; Fig. 3). Additionally, the low-performing group had a later wake time during weekends (10:18 1:29 vs. 9:53 1:26; P = 0.011; Fig. 3). Amongst sleep habits, sleep latency was shown to be shorter in the group of high-performing students (12:21 9:38; Table 2). However, no significant difference in total sleep time between high and low academic performing groups was observed for both weekdays and weekends (Table 2). High-performing students reported to have the usual amount of sleep the night before an exam, more often than low-performing students (38.8% vs. 21.7%, respectively). Accordingly, low-performing students reported a small amount of sleep more often than high-performing students the night before an examination (39.3% vs. 24%, respectively). Low-performing students reported to stay awake during the night due to computer or TV usage more often than high-performing students (66% vs. 45%; P = 0.001, Table 3). The percentage of students who reported insomnia symptoms was significantly higher in the low-performing students group compared with the high-performing students group (10.4% vs. 3.1%; Table 3). There were significant differences in the gender distribution of the two groups, as low-performing students were more often male compared with high-performing students (40.2% vs. 15.5%; Table 4). Amongst lifestyle habits that could be associated with academic performance, results of our study indicate that low-performing students were more likely to use prescription and
Fig. 1. Sleep–wake cycle of the whole dental student study population, indicating earlier bedtime and wake time during the weekday (23:53 0:55, 7:02 0:44, respectively) compared with the weekend (01:23 1:38, 10:03 1:27, respectively). *P < 0.001 derived from t-test comparison between weekdays and weekends.
Fig. 2. Sleep cycle pattern of low and high academic performing students during the weekday, indicating earlier bedtime and wake time in highperforming students (23:44 0:49; 6:59 0:40, respectively) compared with low-performing students (24:02 1:00; 7:09 0:48, respectively) *P = 0.004 derived from Mann–Whitney U-test comparison between low and high academic performing students.
ª 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd Eur J Dent Educ 18 (2014) 187–194
189
Sleep habits and academic performance
Valic et al.
Fig. 3. Sleep cycle pattern of low and high academic performing students during the weekend, indicating earlier bedtime and wake time in highperforming students (01:08 1:23; 9:53 1:26, respectively) compared with low-performing students (01:39 1:38; 10:18 1:29, respectively) *P = 0.016; **P = 0.011, derived from Mann–Whitney U-test comparison between low and high academic performing students.
TABLE 2. Sleep habits differences of students in the lowest and the highest quintile of success n Weekday Total sleep time Weekend Total sleep time Ideal timing Bedtime Wake time Total sleep time Sleep latency Weekday Earliest bedtime Latest bedtime Afternoon naps (days in the week) Subjective fatigue in the evening start (h)
Lowest quintile (grade point averages, GPA < 3.5)
n
Highest quintile (GPA > 4.3)
P value*
107
6:46 1:10
129
6:55 0:49
0.434
105
8:49 1:32
128
8:52 1:10
0.892
58 58 103 106
22:51 7:49 8:44 18:17
1:07 0:58 1:32 18:41
74 74 127 126
22:33 7:51 8:50 12:21
0:49 0:55 1:08 9:38
0.130 0.937 0.441 0.008**
58 58 59 105
22:55 01:44 3.04 22:50
1:08 1:14 1.47 1:37
75 73 67 129
22:52 01:19 3.04 22:35
0:55 1:04 1.46 1:22
0.490 0.023** 0.930 0.119
Values are expressed as mean standard deviation. *Significance of Mann–Whitney U-test. **Mann–Whitney U-test P < 0.05.
over-the-counter medications and alcohol and that they were more likely to smoke and exercise less compared with high-performing students (Tables 3 and 4). No other variables were associated with dental students’ academic performance. Regression analysis was performed, using the stepwise regression model including all 447 subjects, with average self-reported GPA as a dependent variable. Variables included in the regression were gender, bedtime and wake time during weekdays and weekends, sleep latency, sleep time before an examination, time spent on the computer, smoking and drinking alcohol. After stepwise regression, three variables were included in the final model: gender, sleeping before an examination and sleep latency. With this model, 9.3% of the variability in the selfreported academic achievement was explained (R2 = 0.93; P < 0.001). The beta-coefficients indicate that male gender (b = 0.23; P < 0.001) and less than average sleep before an examination (b = 0.15; P = 0.003) were associated with low academic achievement, whilst decreased sleep latency (b = 0.14; P < 0.001) predicted higher academic achievement. Based on the stepwise model methodological approach, bedtime and wake time variables, and smoking and drinking variables 190
were excluded. Due to low semi-partial correlation with selfreported GPA, they did not add significant predictive value when included in the model.
Discussion The present study indicates that academic performance of dental students is associated with their sleep habits. Students with better academic performance went to bed earlier on weekdays and weekends and woke up earlier in the morning on weekends compared with low academic performing students. Furthermore, students with high academic performance had shorter sleep latency than students with low academic performance. However, total sleep time did not differ amongst high- and low-performing students. These findings are in accordance with several other studies conducted on college students (3, 14). Timing of sleep and wakefulness is an important contributor to academic performance, in a manner that students with later bedtime and wake time had lower academic performance. In addition, besides circadian factors that play an important role in academic ª 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd Eur J Dent Educ 18 (2014) 187–194
Valic et al.
Sleep habits and academic performance
TABLE 3. Sleep habits differences of students in the lowest and the highest quintile of success Lowest quintile (grade point averages, GPA < 3.5) n (%) Awakening Alone 10 (9.5) Alarm clock 95 (90.5) During weekday, waking up earlier than ideal time Yes 79 (73.8) No 28 (26.2) Subjective feeling after weekday awakening Rested 20 (19) Drowsy 72 (68.6) Very drowsy 13 (12.4) Subjective feeling after weekend awakening Rested 55 (51.9) Drowsy 40 (37.7) Very drowsy 11 (10.4) Preference Early awakenings 35 (33.3) Staying up late 70 (66.7) Sleep quality satisfaction Yes 60 (56.1) No 47 (43.9) Satisfaction with the amount of sleep Yes 39 (36.4) No 68 (63.6) Frequent night awakening Yes 18 (16.8) No 89 (83.2) Afternoon naps Yes 58 (54.7) No 48 (45.3) Subjective feeling after napping Rested 34 (58.6) Sleepy 24 (41.4) Chronically tired Yes 42 (40) No 63 (60) Sleepiness during classes Yes 77 (73.3) No 28 (26.7) Insomnia Yes 11 (10.4) No 95 (89.6) Intake of sleep pharmacies Yes 22 (21) No 83 (79) Awake during the night due to computer or TV Yes 70 (66) No 36 (34) Hours of sleep before an exam Not at all 7 (6.6) Small amount of sleep 42 (39.3) Less than average 30 (28.3) Usual 23 (21.7) Longer than average 4 (3.8)
Highest quintile (GPA > 4.3) n (%)
P value *
4 (3.1) 124 (96.9)
0.053
103 (79.8) 26 (20.2)
0.274
18 (14.1) 94 (73.4) 16 (12.5)
0.585
83 (64.8) 40 (31.3) 5 (3.9)
0.052
57 (44.5) 71 (55.5)
0.082
76 (58.9) 53 (41.1)
0.660
32 (25.2) 95 (74.8)
0.062
15 (11.6) 114 (88.4)
0.252
68 (52.7) 61 (47.3)
0.759
40 (59.7) 27 (40.3)
0.902
44 (34.1) 85 (65.9)
0.353
91 (70.5) 38 (29.5)
0.637
4 (3.1) 125 (96.9)
0.022**
14 (10.9) 114 (89.1)
0.035**
58 (45) 71 (55)
0.001**
2 31 44 50 2
0.004**
(1.6) (24) (34.1) (38.8) (1.6)
*Significance of chi-square test. **Chi-square test P < 0.05.
ª 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd Eur J Dent Educ 18 (2014) 187–194
191
Sleep habits and academic performance
Valic et al.
TABLE 4. Lifestyle habits differences in high- and low-performing students Lowest quintile (grade point averages, GPA < 3.5) n (%) School Split Zagreb Gender Male Female Exercise Yes No Coffee Yes No Smoking Yes No Alcohol Never Maximum seven drinks per week Maximum 20 drinks per week More than 20 drinks per week Watching TV (h/day) Time on the computer (h/day) Studying (h/day) Cell phone use (min/day) Medicina
Highest quintile (GPA > 4.3)
M SD
n (%)
M SD
P value
41 (38.3) 66 (61.7)
–
45 (34.9) 84 (65.1)
–
0.585
43 (40.2) 64 (59.8)
–
20 (15.5) 109 (84.5)
–
The relationship between sleep habits and academic performance in dental students in Croatia M. Valic1, R. Pecotic1, L. Lusic1, K. Peros2, Z. Pribudic1 and Z. Dogas1 1 2
Department of Neuroscience, University of Split School of Medicine, Split, Croatia, Department of Pharmacology, School of Dental Medicine, University of Zagreb, Zagreb, Croatia
keywords sleep; sleep habits; dental students; academic performance. Correspondence Maja Valic Department of Neuroscience University of Split School of Medicine Soltanska 2 21000 Split, Croatia Tel: +385 21 557 860 Fax: +385 21 557 955 e-mail: [email protected] Accepted: 18 November 2013 doi: 10.1111/eje.12081
Abstract Introduction: It is well accepted that sleep and lifestyle habits affect academic success in students. However, sleep patterns and sleep problems amongst dental students have been insufficiently addressed in the literature. The purpose of this study was to evaluate sleep habits of dental students and the relationship between sleep habits and academic performance. Materials and methods: A self-administered questionnaire on sleep habits, academic performance and lifestyle was administered. The participants were 447 dental students from Split University Dental Medicine School and Zagreb University Dental Medicine School from the six academic years. The subjects were classified into two groups based on academic success (high-performing vs. low-performing students) for comparison of sleep and lifestyle habits. Results: Amongst the whole group of students, average bedtime and wake time during weekday was significantly earlier compared with weekend. Main findings indicate that students with high academic performance had earlier bedtimes during weekdays and weekends, earlier wake times during weekends and shorter sleep latency compared with low academic performing students. Conclusion: Self-reported academic performance of dental students in Croatia is associated with timing of sleep and wakefulness, rather than with total sleep time duration.
Introduction Quality of sleep is essential to an individual’s health and wellbeing and has a substantial role in learning and memory process. Sleep habits are influenced by many internal and external factors such as school and work schedules, often resulting in daytime sleepiness (1). Furthermore, it has been shown that insufficient sleep is the main factor influencing mood and alertness and that sleep deprivation results in impaired neurocognitive and psychomotor performance (2). Recent studies have focused on sleep habits amongst university students. Increased psychological pressure and academic demands can lead to impaired quality and duration of sleep. Numerous studies have shown that students with more regular sleep–wake patterns report higher grade point averages (GPAs), whereas students with lower grades report
ª 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd Eur J Dent Educ 18 (2014) 187–194
increased daytime sleepiness as a consequence of decreased sleep duration (shorter sleeping nights) (3–5). Amongst different student populations, medical students received particular attention in sleep research because of high academic demands and increased workload (6–9). Several studies have determined that medical students’ unique academic commitments and lifestyle can impact their sleep habits and result in sleep deprivation (4, 6, 7, 10). Similarly, the doctor of dental medicine (D.M.D.) curriculum is demanding due to both declarative and procedural learning tasks, with declarative learning being more emphasised during the first years of schooling and procedural learning being added in the later years in dental schools in Croatia. The literature reveals that both declarative and motor procedural learning capabilities are impaired by sleep deprivation (11–13).
187
Sleep habits and academic performance
Valic et al.
Whilst it is well accepted that the role of sleep quality and lifestyle habits affects academic success (3, 4, 7), sleep patterns and sleep problems amongst dental students have been insufficiently addressed in the literature. Therefore, the purpose of this study was to determine sleep and lifestyle habits and to estimate their relationship with academic performance amongst dental students in Croatia.
Materials and methods This cross-sectional study was conducted between June and October 2012 at the University of Split School of Dental Medicine and the University of Zagreb School of Dental Medicine. The curricula of the University of Split School of Dental Medicine and the University of Zagreb School of Dental Medicine are governed by Bologna regulations, recommendations and principles in accordance with the university book of regulations for higher education and with the School of Medicine book of regulations. The protocol for this study was overviewed and approved by the Biomedical Research Ethics Committee of the Split University School of Medicine, Split, Croatia, and by the Ethical Committee of the School of Dental Medicine, University of Zagreb, Croatia. Total population at School of Dental Medicine University of Zagreb is 681 students, 516 women and 165 men. Total population at School of Dental Medicine University of Split is 150 students, 47 men and 103 women. The total number of students who participated in the study was 447 (110 men and 337 women), from six academic levels (Table 1). Participants’ data were collected anonymously and none were paid during the study.
TABLE 1. Characteristics of the student sample
Age Average grade University Split Zagreb Year of study 1st 2nd 3rd 4th 5th 6th Gender Male Female Afternoon sleep Yes No Sleep quality satisfaction Yes No Preference Wake up early Staying up late
188
n (%)
n (total)
23.5 (2.09) 3.91 (0.49)
446 432
142 (31.8) 305 (68.2)
447
31 85 135 81 55 60
(6.9) (19) (30.2) (18.1) (12.3) (13.4)
447
110 (24.6) 337 (75.4)
447
220 (49.7) 233 (50.3)
443
249 (55.7) 198 (44.3)
447
284 (64.1) 159 (35.9)
443
The design of the questionnaire used in this study was based on a previously published survey instrument (3) to assess demographics, sleep habits, sleep and sleep-related disturbances, sleep-related behaviour, lifestyle habits and intake of pharmacological substances. The Association of Sleep and Academic Performance questionnaire (3) included 30 questions on sleep habits, dreams, stimulants and general information. We added questions about sleep disturbances and lifestyle habits. The final version of the questionnaire consists of 58 questions. Demographics included questions on age, gender, constitutional parameters and academic success. Academic success was measured by the self-reported average grade. Questions on sleep habits included reports on bedtime and wake time during weekdays and weekends, ideal bedtime and total sleep time estimation. Students were also asked to answer how satisfied they were with the quality and the amount of their sleep. Questions on sleep-related disturbances included questions about nocturnal awakenings, daytime sleepiness, insomnia and breathing problems. The usage of pharmacologically active substances was assessed based on questions regarding smoking and coffee intake and consumption of prescription and over-the-counter medications. All questions were constructed as multiple choices (affirmative/negative and various multiple choices) and as open-ended questions. Students were also asked to answer lifestyle habits questions, such as evening social activities, alcohol consumption, the use of the TV and computer, and exercising and learning habits. Students were asked to report their usual learning time (hours per day and time of the day). Previously trained staff distributed a two-page questionnaire to all students who attended their classes. Staff was informed on research questions, questionnaire structure and a mandatory standardised instruction for students. Considering that students were approached during regular classes, the instruction was necessary to create standardised conditions and accentuate the importance of honest response. Students were informed that participation was anonymous and voluntary. They completed the questionnaire during regular classes at the Schools of Dental Medicine in Split and Zagreb. All students who attended their classes at the time of the survey collection participated in the study and returned completed questionnaires. The time required to complete the questionnaire was approximately 10–12 min. Method of sampling in this study was convenience sampling, meaning that all students who attended the classes at the time survey was conducted had to fulfil the questionnaire. All classes are obligatory for all students. The survey instrument was translated into the Croatian language by a psychologist and a medical doctor (somnologist). The questionnaire was then administered to a pilot group of ten students for feedback on question readability and clarity. The questionnaire was back-translated from Croatian into English by a bilingual professional translator for comparison with the original text. The subjects were classified into two groups based on academic success (high-performing vs. low-performing students). Descriptive results were expressed as frequencies, percentages and mean values with deviation measures. Data were consequently analysed by testing group differences using the chisquare test for categorical and Mann–Whitney nonparametric ª 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd Eur J Dent Educ 18 (2014) 187–194
Valic et al.
test for continuous variables. Data of all subjects were analysed using stepwise regression analysis. All data were entered and analysed using the Statistical Package for Social Sciences program (version 14; SPSS Inc., Chicago, IL, USA). P values less than 0.05 were considered statistically significant.
Results There were 447 dental students who participated in this study. Table 1 summarises demographic characteristics and some other study variables related to sleep. For 432 students responding to the question regarding academic performance, the average GPA was 3.91 0.49. The scholastic performance was stratified as high (highest quintile, GPA > 4.3) or low academic performances (lowest quintile, GPA < 3.5), with 107 students (24%) being classified as high-performing students and 129 (29%) students classified as low-performing students. Amongst the whole group of students, average bedtime during the weekday was significantly earlier compared with the weekend (23:53 0:55 vs. 01:23 1:38; P < 0.001; Fig. 1). Similarly, average wake time during the weekday was earlier compared with the weekend (7:02 0:44 vs. 10:03 1:27; P < 0.001; Fig. 1). For all students, the majority (55.7%) reported satisfaction with their sleep and 64.1% expressed a preference to wake up early in the morning. When the relationship between sleep habits and school performance was analysed, the high-performing group had an earlier bedtime compared with the low-performing group during
Sleep habits and academic performance
weekdays (23:44 0:49 vs. 00:02 1:00; P = 0.004; Fig. 2) and weekends (01:08 1:23 vs. 01:39 1:38; P = 0.016; Fig. 3). Additionally, the low-performing group had a later wake time during weekends (10:18 1:29 vs. 9:53 1:26; P = 0.011; Fig. 3). Amongst sleep habits, sleep latency was shown to be shorter in the group of high-performing students (12:21 9:38; Table 2). However, no significant difference in total sleep time between high and low academic performing groups was observed for both weekdays and weekends (Table 2). High-performing students reported to have the usual amount of sleep the night before an exam, more often than low-performing students (38.8% vs. 21.7%, respectively). Accordingly, low-performing students reported a small amount of sleep more often than high-performing students the night before an examination (39.3% vs. 24%, respectively). Low-performing students reported to stay awake during the night due to computer or TV usage more often than high-performing students (66% vs. 45%; P = 0.001, Table 3). The percentage of students who reported insomnia symptoms was significantly higher in the low-performing students group compared with the high-performing students group (10.4% vs. 3.1%; Table 3). There were significant differences in the gender distribution of the two groups, as low-performing students were more often male compared with high-performing students (40.2% vs. 15.5%; Table 4). Amongst lifestyle habits that could be associated with academic performance, results of our study indicate that low-performing students were more likely to use prescription and
Fig. 1. Sleep–wake cycle of the whole dental student study population, indicating earlier bedtime and wake time during the weekday (23:53 0:55, 7:02 0:44, respectively) compared with the weekend (01:23 1:38, 10:03 1:27, respectively). *P < 0.001 derived from t-test comparison between weekdays and weekends.
Fig. 2. Sleep cycle pattern of low and high academic performing students during the weekday, indicating earlier bedtime and wake time in highperforming students (23:44 0:49; 6:59 0:40, respectively) compared with low-performing students (24:02 1:00; 7:09 0:48, respectively) *P = 0.004 derived from Mann–Whitney U-test comparison between low and high academic performing students.
ª 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd Eur J Dent Educ 18 (2014) 187–194
189
Sleep habits and academic performance
Valic et al.
Fig. 3. Sleep cycle pattern of low and high academic performing students during the weekend, indicating earlier bedtime and wake time in highperforming students (01:08 1:23; 9:53 1:26, respectively) compared with low-performing students (01:39 1:38; 10:18 1:29, respectively) *P = 0.016; **P = 0.011, derived from Mann–Whitney U-test comparison between low and high academic performing students.
TABLE 2. Sleep habits differences of students in the lowest and the highest quintile of success n Weekday Total sleep time Weekend Total sleep time Ideal timing Bedtime Wake time Total sleep time Sleep latency Weekday Earliest bedtime Latest bedtime Afternoon naps (days in the week) Subjective fatigue in the evening start (h)
Lowest quintile (grade point averages, GPA < 3.5)
n
Highest quintile (GPA > 4.3)
P value*
107
6:46 1:10
129
6:55 0:49
0.434
105
8:49 1:32
128
8:52 1:10
0.892
58 58 103 106
22:51 7:49 8:44 18:17
1:07 0:58 1:32 18:41
74 74 127 126
22:33 7:51 8:50 12:21
0:49 0:55 1:08 9:38
0.130 0.937 0.441 0.008**
58 58 59 105
22:55 01:44 3.04 22:50
1:08 1:14 1.47 1:37
75 73 67 129
22:52 01:19 3.04 22:35
0:55 1:04 1.46 1:22
0.490 0.023** 0.930 0.119
Values are expressed as mean standard deviation. *Significance of Mann–Whitney U-test. **Mann–Whitney U-test P < 0.05.
over-the-counter medications and alcohol and that they were more likely to smoke and exercise less compared with high-performing students (Tables 3 and 4). No other variables were associated with dental students’ academic performance. Regression analysis was performed, using the stepwise regression model including all 447 subjects, with average self-reported GPA as a dependent variable. Variables included in the regression were gender, bedtime and wake time during weekdays and weekends, sleep latency, sleep time before an examination, time spent on the computer, smoking and drinking alcohol. After stepwise regression, three variables were included in the final model: gender, sleeping before an examination and sleep latency. With this model, 9.3% of the variability in the selfreported academic achievement was explained (R2 = 0.93; P < 0.001). The beta-coefficients indicate that male gender (b = 0.23; P < 0.001) and less than average sleep before an examination (b = 0.15; P = 0.003) were associated with low academic achievement, whilst decreased sleep latency (b = 0.14; P < 0.001) predicted higher academic achievement. Based on the stepwise model methodological approach, bedtime and wake time variables, and smoking and drinking variables 190
were excluded. Due to low semi-partial correlation with selfreported GPA, they did not add significant predictive value when included in the model.
Discussion The present study indicates that academic performance of dental students is associated with their sleep habits. Students with better academic performance went to bed earlier on weekdays and weekends and woke up earlier in the morning on weekends compared with low academic performing students. Furthermore, students with high academic performance had shorter sleep latency than students with low academic performance. However, total sleep time did not differ amongst high- and low-performing students. These findings are in accordance with several other studies conducted on college students (3, 14). Timing of sleep and wakefulness is an important contributor to academic performance, in a manner that students with later bedtime and wake time had lower academic performance. In addition, besides circadian factors that play an important role in academic ª 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd Eur J Dent Educ 18 (2014) 187–194
Valic et al.
Sleep habits and academic performance
TABLE 3. Sleep habits differences of students in the lowest and the highest quintile of success Lowest quintile (grade point averages, GPA < 3.5) n (%) Awakening Alone 10 (9.5) Alarm clock 95 (90.5) During weekday, waking up earlier than ideal time Yes 79 (73.8) No 28 (26.2) Subjective feeling after weekday awakening Rested 20 (19) Drowsy 72 (68.6) Very drowsy 13 (12.4) Subjective feeling after weekend awakening Rested 55 (51.9) Drowsy 40 (37.7) Very drowsy 11 (10.4) Preference Early awakenings 35 (33.3) Staying up late 70 (66.7) Sleep quality satisfaction Yes 60 (56.1) No 47 (43.9) Satisfaction with the amount of sleep Yes 39 (36.4) No 68 (63.6) Frequent night awakening Yes 18 (16.8) No 89 (83.2) Afternoon naps Yes 58 (54.7) No 48 (45.3) Subjective feeling after napping Rested 34 (58.6) Sleepy 24 (41.4) Chronically tired Yes 42 (40) No 63 (60) Sleepiness during classes Yes 77 (73.3) No 28 (26.7) Insomnia Yes 11 (10.4) No 95 (89.6) Intake of sleep pharmacies Yes 22 (21) No 83 (79) Awake during the night due to computer or TV Yes 70 (66) No 36 (34) Hours of sleep before an exam Not at all 7 (6.6) Small amount of sleep 42 (39.3) Less than average 30 (28.3) Usual 23 (21.7) Longer than average 4 (3.8)
Highest quintile (GPA > 4.3) n (%)
P value *
4 (3.1) 124 (96.9)
0.053
103 (79.8) 26 (20.2)
0.274
18 (14.1) 94 (73.4) 16 (12.5)
0.585
83 (64.8) 40 (31.3) 5 (3.9)
0.052
57 (44.5) 71 (55.5)
0.082
76 (58.9) 53 (41.1)
0.660
32 (25.2) 95 (74.8)
0.062
15 (11.6) 114 (88.4)
0.252
68 (52.7) 61 (47.3)
0.759
40 (59.7) 27 (40.3)
0.902
44 (34.1) 85 (65.9)
0.353
91 (70.5) 38 (29.5)
0.637
4 (3.1) 125 (96.9)
0.022**
14 (10.9) 114 (89.1)
0.035**
58 (45) 71 (55)
0.001**
2 31 44 50 2
0.004**
(1.6) (24) (34.1) (38.8) (1.6)
*Significance of chi-square test. **Chi-square test P < 0.05.
ª 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd Eur J Dent Educ 18 (2014) 187–194
191
Sleep habits and academic performance
Valic et al.
TABLE 4. Lifestyle habits differences in high- and low-performing students Lowest quintile (grade point averages, GPA < 3.5) n (%) School Split Zagreb Gender Male Female Exercise Yes No Coffee Yes No Smoking Yes No Alcohol Never Maximum seven drinks per week Maximum 20 drinks per week More than 20 drinks per week Watching TV (h/day) Time on the computer (h/day) Studying (h/day) Cell phone use (min/day) Medicina
Highest quintile (GPA > 4.3)
M SD
n (%)
M SD
P value
41 (38.3) 66 (61.7)
–
45 (34.9) 84 (65.1)
–
0.585
43 (40.2) 64 (59.8)
–
20 (15.5) 109 (84.5)
–
