Subjective daytime sleepiness in schoolchildren

Family Practice Vol. 17, No. 2 Oxford University Press 2000 Printed in Great Britain Subjective daytime sleepiness in schoolchildren Outi Saarenpää-Heikkilä, Pekka Laippala a and Matti Koivikko Saarenpää-Heikkilä O, Laippala P and Koivikko M. Subjective daytime sleepiness in schoolchildren. Family Practice 2000; 17: 129 133. Objectives. The aim of this study was to find out how the occurrence of subjective daytime sleepiness (SDS) in schoolchildren had changed after 2 years follow-up and which were the predictors associated with persistence and appearance of SDS. Methods. A total of 107 schoolchildren with SDS (Group 1) and an equal number without SDS (Group 2) were chosen to take part in the questionnaire study (age range 12 20 years). The questions concerned sleeping habits, sleep disorders and daytime sleepiness of the child and the family, progress at school and TV/video watching of the child, and social background of the family. A bivariate method (Pearson s chi-square) was used as a statistical tool. Results. We received 68 replies (64%) from Group 1 and 75 (70%) from Group 2. Fifty-four schoolchildren were still sleepy in Group 1 and 28% had become sleepy in Group 2. The persistence of SDS in Group 1 was related to older age, irregular sleeping habits, frequent night waking and the fathers sleep difficulties. The appearance of SDS in Group 2 was related to a high average grade at school. Conclusion. A delayed sleep phase rhythm and problems staying asleep are obvious causes of persistent SDS, but the stress of schoolwork can also cause daytime sleepiness in schoolchildren. Keywords. Average grade, puberty, sleep disorders, sleeping habits. Introduction There are numerous epidemiological studies of daytime sleepiness in schoolchildren and adolescents, but followup studies are rare. Carskadon et al. 1 found that daytime sleepiness appeared during puberty in healthy schoolchildren during a 2 year follow-up. Strauch and Meier, 2 in a study with an 8 year follow-up, found that the need for more sleep was a common but not persistent phenomenon throughout adolescence. In our study, schoolchildren s subjective feelings of daytime sleepiness (SDS) and its development have been followed. The first phase of the study showed that sleeping habits and sleep disorders of the child and sleep problems of the parents have an effect on SDS 3 (Saarenpää-Heikkilä et al., submitted). The aim of the present study was to follow up the consistency of the symptom and variables related to SDS and its chronicity. Received 21 June 1999; Revised 15 October 1999; Accepted 26 October 1999. Medical School, University of Tampere, PO Box 607, SF 33101, Tampere and Department of Paediatrics, University Hospital of Tampere and a School of Public Health, University of Tampere and University Hospital of Tampere, Tampere, Finland. Methods Participants and questionnaire The follow-up was performed by sending two similar questionnaires to the schoolchildren and their parents with an interval of 2 years (1988 and 1990) in two schools (a primary school and a secondary school). The ethical validity of the study protocol was evaluated by the school administrators. The results obtained from the first phase of the study are reported elsewhere 7 (Saarenpää-Heikkilä et al., submitted). The children were asked about waking up times and bedtimes during weekdays and weekends, sleep latency, time spent watching TV and video, dreaming, sleep disorders, daytime sleepiness and progress at school. Their parents answered questions concerning social background, parental sleeping habits, sleep disorders and daytime sleepiness. The design of the whole study is shown in Figure 1. The original subjects consisted of 582 schoolchildren and their parents. In the first phase of the study, the questionnaires were returned by 518 children (89%). Of these, 107 (21%) reported daytime sleepiness (SDS). In the second phase, the follow-up questionnaires (similar to the first) were sent after 2 years to these 107 children (Group 1) and to another 107 children (Group 2) selected randomly from those 411 who in the first questionnaire 129

130 Family Practice an international journal TABLE 1 The dividing points of the non-four-category answers of the child Question Dividing point FIGURE 1 The study design had no daytime sleepiness. Group 2 was matched with Group 1 by age and sex. Data management The SDS of the child was used as a grouping factor. The question Are you sleepy during the daytime? had four possible answers, i.e. categories always/often/sometimes/ never, and the children answering always or often were classified as having SDS. The variables used in the analysis, except for age, were dichotomized. The questions concerning sleep phenomena and disorders were dichotomized as follows: (i) those who answered always or often had the symptom and (ii) those who answered sometimes or never did not have the symptom. The dividing points of the remaining questions answered by the child are presented in Table 1. Parental variables were categorized accordingly. Marital status was dichotomized to (i) marriage or common law marriage; and (ii) divorced, widowed or single. Education was dichotomized to (i) college and academic; and (ii) basic and professional. Social class was divided into the self-employed and the other classes because in the first phase of the study the children whose fathers were selfemployed had more SDS. Statistical analysis The bivariate analyses were carried out separately for Group 1 and Group 2 in order to determine which variables were connected with the change in SDS. A comparison was also made between those in Group 1 who still had SDS (SDS+ in Group 1) and those in Group 2 who still had no SDS (SDS in Group 2). A P-value of 0.1 was chosen as the significant limit because of the small sample size. In these analyses, each variable had four alternatives: either the answers were the same in both phases of the study (two alternatives) or they were changed (two alternatives). Bedtime a 10 p.m. on Fridays 11.30 p.m. on Saturdays 12 p.m. Wake-up time a 7 a.m. on Saturdays 10 a.m. on Sundays 10 a.m. Sleeping time a 8.5 h on Friday nights 10 h on Saturday nights 10 h Sleep latency 30 min TV/video watching 2 h per day Average grade a 7.9 (range 4.0 10.0) Grade a in mathematics 8 (range 4 10) Finnish 8 (range 4 10) a The dividing point was chosen based on the median of the data. A comparison between those who answered and those who did not answer the follow-up questionnaire was also made (Pearson s chi-square). Results A total of 143 replies were received (67%): 68 from Group 1 (64%) and 75 from Group 2 (70%) (Fig. 1). Age in Group 1 varied from 12 to 19 years (mean 15.8 years) and in Group 2 from 12 to 20 years (mean 16.1). There were 28 boys and 40 girls in Group 1, and 31 boys and 44 girls in Group 2. One study participant in Group 1 and three participants in Group 2 were excluded because of insufficient data. SDS was still present in 36/67 pupils (54%) in Group 1; 20/72 pupils (28%) had acquired SDS in Group 2. Parental replies were received from 133 mothers and 113 fathers. In Group 1, the following significant associations were found: short sleeping time and late bedtime and Fridays, late wake-up time on Sundays and frequent night waking were more common in the group with persistent SDS. In Group 2, the children with SDS had higher average grades (AGs) than the children without SDS (Table 2). The persistence of SDS was more frequent in the older age groups in Group 1, but no effect of age was seen in Group 2 (Table 3). When comparing those having persistent SDS with those consistently alert, the following associations were found (Table 2): short sleeping time and late bedtime on

Subjective daytime sleepiness in schoolchildren 131 TABLE 2 SDS and the significant variables in cross-tabulations with P-values for: (a) Group 1; (b) Group 2; and (c) SDS+ in Group 1 versus SDS in Group 2 (a) Group 1 Sleeping time on SDS SDS+ Total Bed time on SDS SDS+ Total weekdays n (%) n (%) n weekdays n (%) n (%) n 8.5 h 8.5 h 10 (33) 21 (60) 31 10 p.m. 10 p.m. 5 (17) 23 (66) 28 8.5 h 8.5 h 6 (20) 5 (14) 11 10 p.m. 10 p.m. 6 (20) 6 (17) 12 8.5 h 8.5 h 1 (4) 4 (12) 5 10 p.m. 10 p.m. 3 (10) 0 (0) 3 8.5 h 8.5 h 13 (43) 5 (14) 18 10 p.m. 10 p.m. 16 (53) 6 (17) 22 Total 30 (100) 35 (100) 65 Total 30 (100) 35 (100) 65 P = 0.0290 P = 0.0003 Bedtime on Fridays SDS SDS+ Total Wake-up time SDS SDS+ Total 1988 1990 n (%) n (%) n on Sundays n (%) n (%) n 1988 1990 11.30 p.m. 11.30 p.m. 9 (29) 18 (50) 27 10 a.m. 10 a.m. 10 (32) 23 (64) 33 11.30 p.m. 11.30 p.m. 4 (13) 10 (28) 14 10 a.m. 10 a.m. 10 (32) 3 (8) 13 11.30 p.m. 11.30 p.m. 5 (16) 2 (5) 7 10 a.m. 10 a.m. 2 (7) 3 (8) 5 11.30 p.m. 11.30 p.m. 13 (42) 6 (17) 19 10 a.m. 10 a.m. 9 (29) 7 (20) 16 Total 31 (100) 36 (100) 67 Total 31 (100) 36 (100) 67 P = 0.0278 P = 0.0291 Night waking SDS SDS+ Total 1988 1990 n (%) n (%) n Yes yes 1 (3) 3 (8) 4 No yes 0 (0) 5 (14) 5 Yes no 3 (10) 5 (14) 8 No no 27 (87) 23 (64) 50 Total 31 (100) 36 (100) 67 P = 0.093 (b) Group 2 Average grade SDS SDS+ Total 1988 1990 n (%) n (%) n 8.0 8.0 23 (53) 4 (25) 27 8.0 8.0 3 (7) 1 (6) 4 8.0 8.0 6 (14) 1 (6) 7 8.0 8.0 11 (26) 10 (63) 21 Total 43 (100) 16 (100) 59 P = 0.068

132 Family Practice an international journal TABLE 2 Continued (c) SDS+ in Group 1/SDS in Group 2 Sleeping time on SDS SDS+ Total Bedtime on SDS SDS+ Total weekdays n (%) n (%) n weekdays n (%) n (%) n 8.5 h 8.5 h 17 (33) 21 (60) 38 10 p.m. 10 p.m. 13 (25) 23 (66) 36 8.5 h 8.5 h 8 (16) 5 (14) 13 10 p.m. 10 p.m. 12 (24) 6 (17) 18 8.5 h 8.5 h 6 (12) 4 (12) 10 10 p.m. 10 p.m. 4 (8) 0 (0) 4 8.5 h 8.5 h 20 (39) 5 (14) 25 10 p.m. 10 p.m. 22 (43) 6 (17) 28 Total 51 (100) 35 (100) 86 Total 51 (100) 35 (100) 86 P = 0.0502 P = 0.0014 Bedtime on SDS SDS+ Total Wake-up time on SDS SDS+ Total Fridays n (%) n (%) n Saturdays n (%) n (%) n 11.30 p.m. 11.30 p.m. 13 (25) 18 (50) 31 10 a.m. 10 a.m. 13 (25) 24 (67) 37 11.30 p.m. 11.30 p.m. 7 (13) 10 (28) 17 10 a.m. 10 a.m. 7 (14) 3 (8) 10 11.30 p.m. 11.30 p.m. 6 (12) 2 (5) 8 10 a.m. 10 a.m. 8 (16) 3 (8) 11 11.30 p.m. 11.30 p.m. 26 (50) 6 (17) 32 10 a.m. 10 a.m. 23 (45) 6 (17) 29 Total 52 (100) 36 (100) 88 Total 51 (100) 36 (100) 87 P = 0.0039 P = 0.0018 Wake-up time on SDS SDS+ Total Night waking SDS SDS+ Total Sundays n (%) n (%) n 1988 1990 n (%) n (%) n 1988 1990 10 a.m. 10 a.m. 18 (35) 23 (64) 41 Yes yes 1 (2) 3 (8) 4 10 a.m. 10 a.m. 7 (13) 3 (8) 10 No yes 6 (12) 5 (14) 11 10 a.m. 10 a.m. 6 (12) 3 (8) 9 Yes no 1 (2) 5 (14) 6 10 a.m. 10 a.m. 21 (40) 7 (20) 28 No no 43 (84) 23 (64) 66 Total 52 (100) 36 (100) 88 Total 51 (100) 36 (100) 87 P = 0.0563 P = 0.0588 Father s SDS SDS+ Total Father s poor SDS SDS+ Total insomnia n (%) n (%) n sleep quality n (%) n (%) n Yes yes 0 (0) 4 (16) 4 Yes yes 0 (0) 4 (17) 4 No yes 3 (8) 0 (0) 3 No yes 3 (8) 5 (21) 8 Yes no 0 (0) 1 (4) 1 Yes no 0 (0) 0 (0) 0 No no 35 (92) 20 (80) 55 No no 35 (92) 15 (62) 50 Total 38 (100) 25 (100) 63 Total 38 (100) 24 (100) 62 P = 0.0201 P = 0.0073

Subjective daytime sleepiness in schoolchildren 133 TABLE 3 weekdays, late bed time on Fridays, late wake-up times on Saturdays and Sundays and frequent night waking were more common in the group with persistent SDS. Their fathers had more insomnia and poorer sleep quality than the fathers of the consistently alert children. In both Groups 1 and 2, the non-participants were more often boys, were older and had later bedtimes and shorter sleeping times than the participants. In Group 1, the non-participants also more often had later bedtimes on Saturdays, lower AG, watched more TV and more often had fathers with their own business than the participants. Discussion Change of SDS with age during the follow-up SDS in Group 1 SDS in Group 2 Age Still sleepy Alert Total Still alert Sleepy Total in years n n n n n n 12 0 7 7 4 1 5 13 1 3 4 5 0 5 14 1 2 3 5 1 6 15 0 3 3 2 1 3 16 14 10 24 19 6 25 17 9 4 13 5 7 12 18 10 2 12 9 4 13 19 1 0 1 2 0 2 20 0 0 0 1 0 1 Total 36 31 67 52 20 72 Group 1 chi-square = 19.995, P = 0.0056; Group 2 chi-square = 9.464, P = 0.3047. The variability of SDS in our study was expected on the basis of earlier findings. In the study by Strauch and Meier, 3 14.5% of the adolescents consistently expressed a wish for more sleep, but in our study consistent SDS was found in only 6.9% (36/518). However, the question in their study ( Would you like to sleep more? ) differed from ours ( Are you sleepy during the daytime? ). The sleeping habits of the children who had persistent SDS differed markedly in both phases of the study from the habits of the other children. Night waking was also significantly more common among those who were still sleepy. According to these results, persistent lack of sleep and poor sleep quality seem to affect the chronicity of daytime sleepiness. Higher AGs of those who had SDS in Group 2 shows that daytime sleepiness may also depend on demands to succeed at school. The effect of age on daytime sleepiness has been explained, on one hand, by a change of sleeping habits, and on the other hand by the effect of puberty. 1,2,4 7 In our study, the persistence of SDS was more common in older age groups. In contrast, all seven 12-year-old children in Group 1 who had been sleepy 2 years earlier had become alert. Thus their SDS was not a consequence of puberty. Our explanation is that these children had been in the third grade 2 years previously, which is more demanding than the first and second grades in the Finnish school system. Insomnia and poor sleep quality in the fathers of the persistently sleepy children may indicate a genetic predisposition to sleep disturbances or social stress in the family. In conclusion, from our results, SDS is not very consistent. It seems to disappear from the younger age group, perhaps when schoolchildren have become used to the longer schooldays. SDS seems to be most persistent in late puberty when undesirable sleeping habits are more common than in the younger age group. Also, poor sleep quality has an obvious effect on chronic daytime sleepiness which may be due partly to genetic or social factors. Demands for success at school also seem to increase daytime sleepiness. Acknowledgements This study was supported financially by the Emil Aaltonen Foundation and the Yrjö Jahnsson Foundation. References 1 Carskadon M, Harvey K, Duke P, Anders T, Litt I, Dement W. Pubertal changes in daytime sleepiness. Sleep 1980; 2: 453 460. 2 Strauch I, Meier B. Sleep need in adolescents: a longitudinal approach. Sleep 1988; 11: 378 386. 3 Saarenpää-Heikkilä OA, Rintahaka PJ, Laippala PJ, Koivikko MJ. Sleep habits and disorders in Finnish schoolchildren. J Sleep Res 1995; 4: 173 182. 4 Billiard M, Alperovitch A, Perot C, Jammes A. Excessive daytime somnolence in young men: prevalence and contributing factors. Sleep 1987; 10: 297 305. 5 Lack LC. Delayed sleep and sleep loss in university students. College Health 1986; 35: 105 110. 6 Kirmil-Gray K, Eagleston JR, Gibson E, Thoresen CE. Sleep disturbance in adolescents: sleep quality, sleep habits, beliefs about sleep, and daytime functioning. J Youth Adolesc 1984; 13: 375 384. 7 Price VA, Coates TJ, Thoresen CE, Grinstead OA. Prevalence and correlates of poor sleep among adolescents. Am J Dis Child 1978; 132: 583 586.