Article Text

Download PDFPDF

Sleep quality evaluation, chronotype, sleepiness and anxiety of Paralympic Brazilian athletes: Beijing 2008 Paralympic Games
  1. Andressa Silva1,2,4,
  2. Sandra Souza Queiroz1,2,
  3. Ciro Winckler1,4,
  4. Roberto Vital3,4,
  5. Ronnie Andrade Sousa4,
  6. Vander Fagundes5,
  7. Sergio Tufik1,2,6,
  8. Marco Túlio de Mello1,2,6
  1. 1Universidade Federal de São Paulo (UNIFESP), São Paulo, Brazil
  2. 2Centro de Estudos em Psicobiologia e Exercício (CEPE), São Paulo, Brazil
  3. 3Universidade Federal do Rio Grande do Norte, Natal, Brazil
  4. 4Comitê Paraolímpico Brasileiro (CPB), Brasília, Brazil
  5. 5Universidade Federal de Uberlândia (UFU), Uberlândia, Brazil
  6. 6Pesquisador CNPq
  1. Correspondence to Marco Túlio de Mello, Departamento de Psicobiologia, Universidade Federal de São Paulo, Rua Francisco de Castro, 93, Vila Clementino—SP-04020-050, São Paulo, Brazil; tmello{at}


Objective The objective of this study was to evaluate the sleep quality, sleepiness, chronotype and the anxiety level of Brazilian Paralympics athletes before the 2008 Beijing Paralympic Games.

Design Cross-sectional study.

Setting Exercise and Psychobiology Studies Center (CEPE) and Universidade Federal de São Paulo, an urban city in Brazil.

Participants A total of 27 Paralympics athletes of both genders (16 men and 11 women) with an average age of 28±6 years who practised athletics (track and field events) were evaluated.

Main outcome measures Sleep quality was evaluated using the Pittsburgh Scale and the Epworth Sleepiness Scale to evaluate sleepiness. Chronotype was determined by the Horne and Östberg questionnaire and anxiety through the State-Trait Anxiety Inventory. The evaluations were performed in Brazil 10 days before the competition.

Results The study's results demonstrate that 83.3% of the athletes that presented excessive daytime sleepiness also had poor sleep quality. The authors noted that 71.4% were classified into the morning type and 72% of the athletes who presented a medium anxiety level also presented poor sleep quality. Athletes with poor sleep quality showed significantly lower sleep efficiency (p=0.0119) and greater sleep latency (p=0.0068) than athletes with good sleep quality. Athletes who presented excessive daytime sleepiness presented lower sleep efficiency compared to non-sleepy athletes (p=0.0241).

Conclusions The authors conclude that the majority of athletes presented poor sleep quality before the competition. This information should be taken into consideration whenever possible when scheduling rest, training and competition times.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


The Paralympics date back to 1960, the year of the first Paralympic Games in Rome. After the Second World War, the sports designed for athletes with disabilities advanced in the contexts of prevention as well as physical, social and psychic rehabilitation.1 2 Currently, the Paralympics are an evolving process focusing on the participation of disabled athletes at different levels and capabilities. Additionally, the Games focus on scientific development that furthers the understanding of the influence of disabilities on exercise and sports performance.3 In the 2008 Beijing Paralympic Games, Brazilian athletes finished ninth overall with 16 gold medals and a total of 47 medals.4

The Paralympics become more competitive and garner more interest each year. Thus, the study of variables that directly interfere with both team and individual athlete results and performance are the focus of significant attention from researchers and scholars. It is known that these variables are linked to technical/tactical, physical and physiological/psychological aspects that are directly related to performance.

Even though the athletes are well trained technically/tactically and physically, Becker and Samulski5 suggest that athletes respond differently to external stimuli during a competition since pressure from training and competition is transferred to an emotional realm. Therefore, well-prepared athletes can present reductions of physical and emotional performance during competition when they are under strong pressure to perform well.6 However, other athletes can improve their performance if they present a good level of precompetitive anxiety associated with an appropriate emotional and physical balance.6

The effects of excessive anxiety before the competition led to several consequences for the athletes.7 If anxiety levels are within normal levels, these effects can be positive; if they surpass those levels, however, they can cause severe emotional reactions, including a decrease in performance.

Some evidence suggests that regular physical practice produces a variety of physiological benefits. Adaptability to physical exercises can be found in young, adult and older populations as well as populations that are healthy or influenced by some pathology and/or disability. Several factors, such as the initial level of physical fitness, the design and duration of the fitness program, genetic factors, age and gender, should be considered in this long list of adaptabilities.8 9

Samulski and Noce10 carried out a study on the psychological preparation of the Brazilian Paralympic athletes and showed that the most important reasons for becoming involved in a sport were the enjoyment derived from practice and the need for rehabilitation. The majority of athletes mentioned the following stress factors: sleeping problems, pressure to win and interpersonal conflicts.

Several studies have been carried out to evaluate the relationship between physical exercise and sleep quality. The most common change observed is the increase in slow wave sleep.11,,15 Some studies have demonstrated increased rapid eye movement (REM) sleep latency and a decreased percentage of time spent in this stage,11 15 whereas others have noted an increase in total sleeping time and a reduction in sleep latency.16

Very few studies related to Paralympic athletes' sleeping patterns and physical activities are available in the literature. This may be due to the fact that the sleeping patterns of this population do not differ widely from those of Olympic athletes. One of the few studies in this area demonstrates a higher frequency of periodic leg movements in individuals with spinal cord injury; such sleep disturbances are associated with greater sleep fragmentation and less successful physical and mental recovery.17

Vuori and collaborators18 collected epidemiological data from 1600 individuals aged 31–50 years and these individuals were interviewed with the aim of determining the influence of physical activity on sleeping. The authors report that social and physiological factors, conditions of the sleeping location, sleep pattern, lifestyle and conditions of the individuals directly influence sleep quality and physical performance. However, the authors concluded that moderate or vigorous exercise is associated with improvements in sleep quality.

Therefore, the present study aimed to evaluate sleep quality, chronotype and anxiety levels of Brazilian athletes that competed in the Beijing 2008 Paralympic Games.

Materials and Methods


A total of 27 Paralympic athletes of both genders (16 men and 11 women) competing in the track and field were evaluated. They were selected by the Brazilian Paralympics Committee to compete in the Beijing 2008 Paralympic Games.


All evaluations were carried out in periods during which the athletes were not involved in other activities and after obtaining consent from the technical committee of the respective modality. All evaluations were performed during the preparation period in the city of São Paulo, Brazil (August, 2008). The athletes were evaluated with regard to sleep, sleepiness, chronotype and anxiety between 9:00 and 11:00 a.m. as an attempt to avoid collecting data at physically and emotionally stressful moments after practices.

Pittsburgh Questionnaire

The Pittsburgh Sleep Quality Index consists of 21 items that evaluate sleep quality and disturbances through a report card of the last month that includes seven components: sleep subjective quality, sleep latency, sleep duration, sleep efficiency, sleep disturbance, use of sleeping pills and daytime dysfunction.19 It was considered as the cut-off point when classifying the following: ≥5 point was poor quality and <4 good quality sleep.

Epworth Sleepiness Scale

This scale evaluates sleepiness in active and passive situations, including sitting and reading; watching TV; sitting in a public place; sitting on a train, car or bus (non-stop for 1 h); lying down; lying down for an afternoon nap; sitting and talking to someone; sitting quietly for lunch; abstaining from alcohol use; and driving but stuck in traffic for a few minutes. The athlete has to determine the chance of falling sleep in each of the presented situations, scoring likelihood from 0 (no chance) to 3 (high chance). The reference values are: Epworth Sleepiness Scale (SE) normal, from 0 to 6; SE limit, from 7 to 9; SE slight, from 10 to 14; SE moderate, from 15 to 20; SE high, above 20.20

In addition to the above data describing sleep quality, this instrument is important for verifying sleepiness levels in periods when athletes should be awake and attentive.

Horne and Östberg Questionnaire (chronotype)

Chronotype evaluation was performed using the Horne and Östberg21 Questionnaire, which classifies subjects with regard to morningness and eveningness using the following scale: 16–33, evening type; 34–44, moderate evening type; 45–65, indifferent; 66–76, moderate morning type; 77–86, morning type.

State-Trait Anxiety Inventory

Anxiety-E is a provisional emotional state characterised by consciously noticed tension and apprehension of varying intensity over time. People with high levels of Anxiety-E have a tendency to react more frequently to situations as if they were threatening or dangerous.22 The anxiety scale consists of 20 statements in which the subject is instructed to choose the best option that defines his feeling at a determined moment. The answer options are: absolutely not, a little, quite a lot and a lot, with scores ranging from 1 to 4, respectively. The State-Trait Anxiety Inventory (STAI) was designed to be a self-assessed inventory and can be performed individually or in groups, with an average time for completion of 20 min.

Statistical procedures

Data are displayed in tables and graphs. Data normality was verified with the Shapiro–Wilk test.

Descriptive statistics confirmed the mean calculation and SD for all of the continuous and semicontinuous data with a normal distribution. Data with asymmetrical distribution underwent natural log (base e) transformation and are represented via the geometric mean and 95% CI. Categorical data are expressed in absolute and relative frequencies.

In order to evaluate differences between the sleep efficiency and latency between the good and poor sleep groups, a Student t test was used for independent samples. The same test was used to compare sleep efficiency between the sleepiness and non-sleepiness groups. The variation coefficients were compared with the Levene test, but no significant differences were noted; therefore, the Welch correction was not needed.

For the analyses, SPSS Statistics 17.0 software for Windows (SPSS, Inc, Chicago, IL) and Prism 5.0 for Windows (GraphPad Software, Inc, San Diego, CA) were used. The value of α ≤5% was defined to be statistically significant.


The research was approved by the Ethics Committee of the Federal University of São Paulo (CEP #1148/08). Participants signed informed consent forms authorising collection of data. All of the individuals participated in the tests voluntarily and were aware of their processes and objectives.


Sample characterisation

A total of 27 Paralympic athletes (16 male and 11 female) from track and field sports were included in the current study. The subjects had an average age of 28±6 years, average weight of 62.7 kg (±9.6) and height of 164.6 cm (±30.8). The athletes competed in the different events described below: field (4 athletes) and track (23 athletes).

Table 1 shows that 83.3% of the athletes that presented excessive morning sleepiness also presented poor sleep quality, as evaluated by the Pittsburgh Questionnaire. We also highlight that 71.4% of the athletes presented as morning types. Finally, 72% of the athletes that exhibited a medium level of anxiety presented poor sleep quality.

Table 1

Epworth, chronotype, and Anxiety-E of athletes with good and poor sleep quality

Figure 1 shows that sleep latency was higher in the group with poor sleep quality.

Figure 1

Comparison of sleep latency using the Pittsburgh Questionnaire. α≤5% is defined as statistically significant.

Figure 2 demonstrates that non-sleepy athletes exhibited significantly greater sleep efficiency than sleepy athletes (p=0.0241).

Figure 2

Comparison of sleep efficiency using the Epworth Sleepiness Scale. α≤5% is defined as statistically significant.

Athletes with good sleep quality demonstrated significantly greater sleep efficiency than athletes with poor sleep quality, as shown in figure 3.

Figure 3

Comparison of sleep efficiency using the Pittsburgh Questionnaire. α≤5% is defined as statistically significant.


The results of this sleep evaluation demonstrate that the majority of athletes presented poor sleep quality and excessive daytime sleepiness. These are concerning data because scientific investigations suggest that the non-REM sleep delta (NREM) is fundamental for good physical recuperation. Additionally, it is during the REM sleep phase that memory consolidation and recovery of the cognitive aspects occur. In the current study, the athletes with poor sleep quality showed a sleep efficiency value of only 78.5%; to be considered normal, subjects' sleep efficiency value should be at least 85%.23

Therefore, low sleep efficiency can directly affect physical and cognitive aspects during the practice and competition phases. Hence, monitoring of this aspect is fundamental for optimal practice and competition performance. This finding and hypothesis is confirmed in figures 1 and 2, in which we show that athletes with greater sleep latency presented lower sleep efficiency and greater sleepiness during the wakefulness period. These data demonstrate the low physical and cognitive recuperation achieved during both the resting period and recuperation phase, which occurs during the dark phase and main sleep.

The influence of physical practice on sleep quality in the physical disabled has also been studied by our group. De Mello and collaborators17 24,,26 carried out polysomnography (sleep study) in athletes with spinal cord injury who performed acute (maximum effort test) and chronic (practicing during 44 days at a ventilatory threshold of 1) exercise. When sleeping, these athletes demonstrated a significant decrease in periodic limb movements (PLMs). The effect of physical practice was not significantly different from the results obtained in volunteers treated with l-dopa, which is a standard pharmacologic treatment for PLMs. Therefore, physical practice demonstrates an ability to minimise that disorder in individuals with spinal cord injury. Other researchers have specifically focused on the incidence of PLMs in paraplegics.17 27 28

De Mello and collaborators29 observed that most of the time, the physically disabled individual initiating a fitness program shows characteristics like depression and sleep disturbance. Through a questionnaire, these authors also evaluated sleep patterns and complaints in individuals with full medullar section syndrome. A total of 59 individuals with medullar sectioning were evaluated and divided into athletic and sedentary groups. The results demonstrated that athletes spent more total time sleeping during the week. Significant differences were also observed between athletes and sedentary subjects with regard to snoring incidence (20% and 47%, respectively), night-time waking (64% and 35%, respectively) and lower-body movements (72% and 38%, respectively). These results can be explained by greater excitability of the spinal intrinsic circuits that generate muscle tone and higher automatisms in the athletes that minimise PLM frequency in this specific population.30

Great alterations in REM sleep have been verified in visually impaired individuals. Congenitally blind individuals present some differences during dreaming; in this sleeping phase, they do not show ocular movements but olfactory, tactile and auditory alterations. In individuals with acquired visual impairments, this condition may not be reduced so significantly and can appear normal in comparison to non-visually impaired individuals.26

Another important aspect to be highlighted is the predominance of the morning chronotype among the Paralympic athletes. The option of practising and competing during that period thus offers a good strategy, as better results may be achieved when the athletes' physical capacity, biological rhythm and chronotype are matched. However, these athletes can show great difficulty adapting to evening practices or transmeridian trips that cross several time zones; such changes may impair performance if the athletes are not well adapted.

With regard to transmeridian trips, evidence suggests that practice can alter the speed of biological clock adjustment. A combination of light exposure, eating habits, outdoor practices and relaxation in the dark represents a method by which to facilitate these adjustments in athletes. Such a procedure was developed for the Brazilian Paralympic athletes that competed in the 2000 Sydney Paralympics Games. The study by De Mello and collaborators on the adjustment of Brazilian Paralympic athletes to the Sydney time zone showed excellent results and contributed to the number of medals achieved by the Brazilian Paralympic team.

Another important aspect observed in some studies is anxiety, which points to an interaction between sleep and anxiety.31,,34 Anxiety is a signal of alertness that is determined by the presence of an internal conflict responsible for alerting an individual to eminent danger and it allows the individual to react to the threat.12 It is probable that everyone has experienced anxiety at some point.35 In the present study, an increase in anxiety was observed in the Paralympic athletes the day before the main competition via STAI-E evaluation. The moment of data collection, however, can be considered a situation of normal anxiety levels. Better observation of this aspect is important in cases demonstrating increased stress and anxiety under non-competitive conditions because this variable can trigger insomnia and reduce the total sleeping period; such alterations completely change the process of recovery and recuperation for these athletes.

In summary, almost all sleep disorders are harmful for the recovery process. Such disorders result in greater sleep fragmentation, reduced sleep quality and efficiency, increased irritability and reduced physical and cognitive performance; these are the classical effects of any sleep disturbance. Almost all sleep disturbances provoke a reduction in the percentage of time spent in Delta sleep stage (stages 3 and 4) and REM sleep, with a consequent increase in stage 2 in NREM sleep. These alterations impair both physical recovery as well as the reduction of the anabolic processes during Delta sleep, which consequently reduces recovery capacity between practices and during competition. Further, REM sleep reduction has consequences for cognitive processes and short-term memory. Thus, REM sleep is a period of great importance that allows athletes to engage in good decision-making processes and concentrate during competitions and games. If the REM sleep phase is perturbed, therefore, consequences can include an increase in irritability, memory loss and injury for high-performance athletes.

Olympic and Paralympic athletes who experience non-recuperative sleep due to the lack of sleep and reduction in the percentage of Delta and REM sleep phases may be more tired at practices because these athletes are not obtaining recuperating benefits from these sleep phases.

We conclude that 72% of the athletes presented precompetition stress. This value may be within the realm of normality if the period during which they were evaluated is considered. However, 70% presented low sleep efficiency, which can cause significant reductions in the physical and mental recovery processes and reduce athletic performance during practices and competitions. Therefore, it is essential to evaluate and determine athletes' sleep quality to determine a good prophylactic regimen. Such prophylaxis is critical and decisive for postexercise recovery and competition performance for athletes that present some kind of sleep disorder.


The authors want to thank the Psychopharmacology Incentive Fund Association (Associação Fundo de Incentivo a Psicofarmacologia—AFIP), the Exercise and Psychobiology Studies Center (Centro de Estudos em Psicobiologia e Exercício—CEPE), the Sleepiness and Accident Multiprofessional Studies Center (Centro de Estudos Multiprofissional em Sonolência e Acidentes—CEMSA), FAPESP (Cepid/Sono 98/143033), CNPq, the Brazilian Paralympic Academy—Brazilian Paralympics Committee and FADA—UNIFESP. The authors would like to thank Leandro Stetner Antonietti for the help with the statistical analysis.


View Abstract


  • Competing interests None.

  • Patient consent Obtained.

  • Ethics approval The research was approved by the Ethics Committee of the Federal University of São Paulo (CEP #1148/08).

  • Provenance and peer review Not commissioned; externally peer reviewed.