J Sleep Res. (2015) 24, 40–46
Daytime functioning and nightime sleep
The relationship between a night’s sleep and subsequent daytime functioning in older poor and good sleepers R A S H E L L E A . S M I T H , L E O N C . L A C K , N I C O L E L O V A T O and H E L E N W R I G H T School of Psychology, Flinders University, Adelaide, SA, Australia
Keywords insomnia treatment, night-to-day relationships Correspondence Professor Leon C. Lack, School of Psychology, Flinders University, GPO Box 2100, Adelaide, SA 5001, Australia. Tel.: 61-8-8201-2391; fax: 61-8-8201-3877; e-mail: leon.lack@flinders.edu.au Accepted in revised form 28 July 2014; received 8 April 2014 DOI: 10.1111/jsr.12237
SUMMARY
Those suffering insomnia symptoms generally report daytime impairments. However, research has not assessed whether this relationship holds on a nightly basis, despite the strongly held belief that a night of poor sleep impairs mood and functioning the following day. The objective of this study was to test this relationship in a group of older poor sleepers with insomnia symptoms compared with good sleepers. This study utilized a within-subjects design to investigate day-to-day subjective daytime functioning and its relation to the previous night’s sleep. Seventeen older individuals (mean age: 67.5 years) were identified with a retrospective questionnaire and 2 weeks of sleep–wake diary to have poor sleep consistent with insomnia. Seventeen good sleepers (mean age: 67.8 years) were selected using the same measures. Participants reported their beliefs about sleep and daytime functioning on the Dysfunctional Beliefs and Attitudes about Sleep Scale (DBAS-16). One week later they commenced a 14-day period of sleep–wake diaries and concurrent responses to a modified Daytime Insomnia Symptom Scale (DISS). Results showed significant night-to-day covariation between sleep efficiency and daytime functioning for individuals with poor sleep (r = 0.34), but not for good sleepers (r = 0.08). Those poor sleepers who held this covariation belief most strongly were those who subsequently showed this night-to-day relationship the most strongly (r = 0.56). This was not true for good sleepers. For those suffering insomnia, these findings demonstrate their belief that a poor sleep is followed by an impaired daytime, consistent with their experience.
INTRODUCTION Prevoius research indicates that insomnia sufferers hold strong beliefs concerning the importance of sleep (Carney and Edinger, 2006). For example, those with insomnia strongly endorse the beliefs that they ‘cannot function without a good night of sleep’ (Harvey et al., 2005, 2007), and that ‘feeling tired, having no energy, or not functioning well is due to poor sleep’ (Morin et al., 2007). In general, the literature shows subjective deficits in daytime functioning associated with having insomnia, such as self-reported difficulties in physical, psychiatric, cognitive and interpersonal domains (Espie et al., 2012; Orff et al., 2007; Riedel and Lichstein, 2000).. However, research has generally focused on a global evaluation of daytime impairment, rather than on the individual’s perceptions of the day-to-day variability of daytime
40
functioning related to the previous night’s sleep. Consequently, it is unclear whether there is a paired relationship between sleep on any given night and the following day’s feelings and performance as expressed in these strongly held beliefs. The extent to which these beliefs are, indeed, an accurate assessment of the relationship between a specific night of sleep and subsequent daytime functioning is unknown. Sleep, daytime functioning and mood are dynamic states that show day-to-day fluctuations (Buysse et al., 2010; McCrae et al., 2008; Perlis et al., 2010; Sanchez-Ortuno and Edinger, 2012; Vallieres et al., 2011). Most studies, however, have only documented the broad consequences of sleep disturbances on daytime functioning (Baglioni, 2009). Interindividual correlations show that those individuals with lower sleep quality in general have lowered daytime mood ª 2014 European Sleep Research Society
Relation between sleep and subsequent daytime functioning and functioning in general (Riedel and Lichstein, 2000). Such methods do not assess intra-individual day-to-day variability in daytime function in relation to the preceding night’s sleep. A small number of studies have prospectively measured day-to-day performance and nightly sleep in the insomnia sufferers’ naturalistic setting. Buysse et al. (2007) had both insomnia and good sleepers keep daily sleep diaries and record their mood, alertness, energy and concentration in a daytime symptom diary at four intervals during the day. While the studies provided useful information on morning, afternoon and evening temporal variability of daytime symptoms over the course of 1 day, the analysis did not organize sleep variables and daytime symptoms in night-to-day pairings. Therefore, the direct relationship experienced for an individual between their night sleep and subsequent daytime functioning was not identified. To the authors’ knowledge, only four previous studies have assessed intra-individual night-to-day associations. McCrae et al. (2008), in a sample (n = 103) of healthy, aged individuals, kept sleep diaries and completed the positive and negative affect schedule each morning for 14 days. Subjective sleep reports correlated with affect in the expected direction; nights with greater reported wake time or lower sleep quality were followed by mornings characterized by more negative affect and less positive affect. However, possible demand characteristics of evaluating affect immediately after the sleep diary in the morning, and possible sleep inertia effects, limit conclusions about the relation of sleep quality to subsequent evaluation of the following day as a whole. Recently, Tang et al. (2012) evaluated the relationship in chronic pain sufferers between sleep and subsequent experience of daytime pain over a 7-day period. Although subjectively rated lower sleep quality predicted greater experience of morning pain but not later pain, sleep efficiency was not related to experienced daytime pain. Another recent study over a 5-day period found longer reported sleep and higher subjective sleep quality associated with higher positive affect and lower negative affect (de Wild-Hartmann et al., 2013). Although the associations were significant, the effect sizes were small in their very large sample (n > 500). In a recent study, Gordon and Chen (2013) found somewhat more (P < 0.06) interpersonal conflict between romantically related individuals following nights of poorer sleep than after good sleep. In a second study they found that a night of poor sleep tended to bias towards negative affect, less empathetic accuracy in a conflict situation and reduced conflict resolution ability. Although interpersonal relations are important, a greater range of cognitive and mood variables need to be assessed in relation to the impact of the previous night’s sleep. Furthermore, these previous four studies did not specifically include those suffering insomnia symptoms. Due to the increased concern that poor sleepers have about their sleep and its impact on functioning, we hypothesized a positive relationship between their sleep and subsequent daytime functioning. Because good sleepers ª 2014 European Sleep Research Society
41
subscribe less strongly to the effect of sleep on daytime functioning, we hypothesized a weaker relationship between their sleep and daytime functioning. We utilized a sample of older individuals due to the increased prevalence of insomnia symptoms in late life (Foley et al., 1995) and the attempt to keep our small sample size relatively homogeneous. METHODS Participants Participants in an older age group were recruited from the general community in metropolitan Adelaide and a more rural area of South Australia. Participants were individuals aged between 55 and 86 years who responded to advertisements placed on public notice boards inviting screening for a ‘Sleep and Well-Being’ study. Fifty-five participants completed the Pittsburgh Sleep Quality Index (PSQI; Buysse et al., 1989). Thirty-four individuals (17 with elevated sleep disturbance and 17 good sleepers) were invited back to participate in the current study, having indicated that they either had disturbed sleep (PSQI score ≥9) or that they were good sleepers (PSQI score 18 for stress, >9 for anxiety and >13 for depression) on the Depression, Anxiety, Stress Scales (DASS21; Lovibond and Lovibond, 1995), as these comorbid conditions may influence daytime functioning (Philip et al., 2006). Also excluded were individuals taking hypnotic medications regularly (n = 2) and self-reporting another sleep disorder (i.e. sleep apnea and restless leg syndrome) evident from item 10 on the PSQI (n = 3). Good sleepers showing no variation of night sleep over the 2 weeks (where correlations could not be calculated) were also excluded (n = 2). Materials Baseline questionnaire Beliefs about the relationship between sleep and daytime functioning were assessed through the Dysfunctional Beliefs and Attitudes about Sleep Scale (DBAS16; Morin et al., 2007). The DBAS comprises four subscales of which the scale of primary interest in the present study was the effects scale, as it relates directly to the current research question:
42
R. A. Smith et al.
Table 1 Age, PSQI, gender and employment status for good sleepers and sleep disturbed group comparing mean age and PSQI with t-tests and gender and employment proportions with v2 tests Good sleepers (n = 17)
Sleep disturbed (n = 17)
Characteristics
Mean (SD)
Mean (SD)
v2/ t(32)
P
Age PSQI Gender Employed
67.82 (9.33) 3.41 (1.50) 8 male, 9 female 9
67.53 (8.73) 11.24 (1.56) 6 male, 11, female 10
0.095 14.88 0.40 0.115
0.925
Daytime functioning and nightime sleep
The relationship between a night’s sleep and subsequent daytime functioning in older poor and good sleepers R A S H E L L E A . S M I T H , L E O N C . L A C K , N I C O L E L O V A T O and H E L E N W R I G H T School of Psychology, Flinders University, Adelaide, SA, Australia
Keywords insomnia treatment, night-to-day relationships Correspondence Professor Leon C. Lack, School of Psychology, Flinders University, GPO Box 2100, Adelaide, SA 5001, Australia. Tel.: 61-8-8201-2391; fax: 61-8-8201-3877; e-mail: leon.lack@flinders.edu.au Accepted in revised form 28 July 2014; received 8 April 2014 DOI: 10.1111/jsr.12237
SUMMARY
Those suffering insomnia symptoms generally report daytime impairments. However, research has not assessed whether this relationship holds on a nightly basis, despite the strongly held belief that a night of poor sleep impairs mood and functioning the following day. The objective of this study was to test this relationship in a group of older poor sleepers with insomnia symptoms compared with good sleepers. This study utilized a within-subjects design to investigate day-to-day subjective daytime functioning and its relation to the previous night’s sleep. Seventeen older individuals (mean age: 67.5 years) were identified with a retrospective questionnaire and 2 weeks of sleep–wake diary to have poor sleep consistent with insomnia. Seventeen good sleepers (mean age: 67.8 years) were selected using the same measures. Participants reported their beliefs about sleep and daytime functioning on the Dysfunctional Beliefs and Attitudes about Sleep Scale (DBAS-16). One week later they commenced a 14-day period of sleep–wake diaries and concurrent responses to a modified Daytime Insomnia Symptom Scale (DISS). Results showed significant night-to-day covariation between sleep efficiency and daytime functioning for individuals with poor sleep (r = 0.34), but not for good sleepers (r = 0.08). Those poor sleepers who held this covariation belief most strongly were those who subsequently showed this night-to-day relationship the most strongly (r = 0.56). This was not true for good sleepers. For those suffering insomnia, these findings demonstrate their belief that a poor sleep is followed by an impaired daytime, consistent with their experience.
INTRODUCTION Prevoius research indicates that insomnia sufferers hold strong beliefs concerning the importance of sleep (Carney and Edinger, 2006). For example, those with insomnia strongly endorse the beliefs that they ‘cannot function without a good night of sleep’ (Harvey et al., 2005, 2007), and that ‘feeling tired, having no energy, or not functioning well is due to poor sleep’ (Morin et al., 2007). In general, the literature shows subjective deficits in daytime functioning associated with having insomnia, such as self-reported difficulties in physical, psychiatric, cognitive and interpersonal domains (Espie et al., 2012; Orff et al., 2007; Riedel and Lichstein, 2000).. However, research has generally focused on a global evaluation of daytime impairment, rather than on the individual’s perceptions of the day-to-day variability of daytime
40
functioning related to the previous night’s sleep. Consequently, it is unclear whether there is a paired relationship between sleep on any given night and the following day’s feelings and performance as expressed in these strongly held beliefs. The extent to which these beliefs are, indeed, an accurate assessment of the relationship between a specific night of sleep and subsequent daytime functioning is unknown. Sleep, daytime functioning and mood are dynamic states that show day-to-day fluctuations (Buysse et al., 2010; McCrae et al., 2008; Perlis et al., 2010; Sanchez-Ortuno and Edinger, 2012; Vallieres et al., 2011). Most studies, however, have only documented the broad consequences of sleep disturbances on daytime functioning (Baglioni, 2009). Interindividual correlations show that those individuals with lower sleep quality in general have lowered daytime mood ª 2014 European Sleep Research Society
Relation between sleep and subsequent daytime functioning and functioning in general (Riedel and Lichstein, 2000). Such methods do not assess intra-individual day-to-day variability in daytime function in relation to the preceding night’s sleep. A small number of studies have prospectively measured day-to-day performance and nightly sleep in the insomnia sufferers’ naturalistic setting. Buysse et al. (2007) had both insomnia and good sleepers keep daily sleep diaries and record their mood, alertness, energy and concentration in a daytime symptom diary at four intervals during the day. While the studies provided useful information on morning, afternoon and evening temporal variability of daytime symptoms over the course of 1 day, the analysis did not organize sleep variables and daytime symptoms in night-to-day pairings. Therefore, the direct relationship experienced for an individual between their night sleep and subsequent daytime functioning was not identified. To the authors’ knowledge, only four previous studies have assessed intra-individual night-to-day associations. McCrae et al. (2008), in a sample (n = 103) of healthy, aged individuals, kept sleep diaries and completed the positive and negative affect schedule each morning for 14 days. Subjective sleep reports correlated with affect in the expected direction; nights with greater reported wake time or lower sleep quality were followed by mornings characterized by more negative affect and less positive affect. However, possible demand characteristics of evaluating affect immediately after the sleep diary in the morning, and possible sleep inertia effects, limit conclusions about the relation of sleep quality to subsequent evaluation of the following day as a whole. Recently, Tang et al. (2012) evaluated the relationship in chronic pain sufferers between sleep and subsequent experience of daytime pain over a 7-day period. Although subjectively rated lower sleep quality predicted greater experience of morning pain but not later pain, sleep efficiency was not related to experienced daytime pain. Another recent study over a 5-day period found longer reported sleep and higher subjective sleep quality associated with higher positive affect and lower negative affect (de Wild-Hartmann et al., 2013). Although the associations were significant, the effect sizes were small in their very large sample (n > 500). In a recent study, Gordon and Chen (2013) found somewhat more (P < 0.06) interpersonal conflict between romantically related individuals following nights of poorer sleep than after good sleep. In a second study they found that a night of poor sleep tended to bias towards negative affect, less empathetic accuracy in a conflict situation and reduced conflict resolution ability. Although interpersonal relations are important, a greater range of cognitive and mood variables need to be assessed in relation to the impact of the previous night’s sleep. Furthermore, these previous four studies did not specifically include those suffering insomnia symptoms. Due to the increased concern that poor sleepers have about their sleep and its impact on functioning, we hypothesized a positive relationship between their sleep and subsequent daytime functioning. Because good sleepers ª 2014 European Sleep Research Society
41
subscribe less strongly to the effect of sleep on daytime functioning, we hypothesized a weaker relationship between their sleep and daytime functioning. We utilized a sample of older individuals due to the increased prevalence of insomnia symptoms in late life (Foley et al., 1995) and the attempt to keep our small sample size relatively homogeneous. METHODS Participants Participants in an older age group were recruited from the general community in metropolitan Adelaide and a more rural area of South Australia. Participants were individuals aged between 55 and 86 years who responded to advertisements placed on public notice boards inviting screening for a ‘Sleep and Well-Being’ study. Fifty-five participants completed the Pittsburgh Sleep Quality Index (PSQI; Buysse et al., 1989). Thirty-four individuals (17 with elevated sleep disturbance and 17 good sleepers) were invited back to participate in the current study, having indicated that they either had disturbed sleep (PSQI score ≥9) or that they were good sleepers (PSQI score 18 for stress, >9 for anxiety and >13 for depression) on the Depression, Anxiety, Stress Scales (DASS21; Lovibond and Lovibond, 1995), as these comorbid conditions may influence daytime functioning (Philip et al., 2006). Also excluded were individuals taking hypnotic medications regularly (n = 2) and self-reporting another sleep disorder (i.e. sleep apnea and restless leg syndrome) evident from item 10 on the PSQI (n = 3). Good sleepers showing no variation of night sleep over the 2 weeks (where correlations could not be calculated) were also excluded (n = 2). Materials Baseline questionnaire Beliefs about the relationship between sleep and daytime functioning were assessed through the Dysfunctional Beliefs and Attitudes about Sleep Scale (DBAS16; Morin et al., 2007). The DBAS comprises four subscales of which the scale of primary interest in the present study was the effects scale, as it relates directly to the current research question:
42
R. A. Smith et al.
Table 1 Age, PSQI, gender and employment status for good sleepers and sleep disturbed group comparing mean age and PSQI with t-tests and gender and employment proportions with v2 tests Good sleepers (n = 17)
Sleep disturbed (n = 17)
Characteristics
Mean (SD)
Mean (SD)
v2/ t(32)
P
Age PSQI Gender Employed
67.82 (9.33) 3.41 (1.50) 8 male, 9 female 9
67.53 (8.73) 11.24 (1.56) 6 male, 11, female 10
0.095 14.88 0.40 0.115
0.925
