This study is the first to provide detailed subjective and objective data on sleep-wake circadian activity rhythm parameters and fatigue in a sample of FCs of oncology patients prior to the initiation of the patient’s RT. Based on these FCs’ subjective responses to the PSQI and the GSDS, approximately 40% to 60% experienced clinically significant levels of sleep disturbance. This percentage is consistent with findings from two studies,15,16
but lower than findings from three studies of FCs of oncology patients.12,14,17
Using an actigraphy-sleep efficiency cutoff of < 80%,3
29.5% of the FCs had a “bad night’s sleep”. Both the subjective and objective data suggest that a significant percentage of these FCs experience clinically meaningful levels of sleep disturbance. An evaluation of both the subjective and the objective data suggests that many of these FCs had difficulties with sleep maintenance.
Consistent with previous studies of oncology patients48,53,54
and elderly persons,55
only a limited number of significant correlations were found between the nocturnal sleep/rest, daytime wake/activity, and circadian activity rhythm parameters obtained with actigraphy and the subjective ratings of various aspects of sleep disturbance using the PSQI and the GSDS. As shown in and , the strength of most of these correlations was small. As noted in a recent review,35
numerous methodological challenges exist with actigraphy that may affect the strength of the correlations between objective and subjective measures of sleep disturbance. While the agreement between actigraphy and polysomnography is high in normal sleepers,37,56
it can be lower in persons with poor sleep quality57
because these persons tend to lie in bed motionless but awake for long periods of time. In this situation, while actigraphy would overestimate sleep duration, FCs would report poor sleep quality and less sleep time. This disparity is confirmed in this study because using previously established cutoff scores for the subjective measures, approximately 40% to 60% of the FCs had significant sleep disturbance at the initiation of RT. However, using a sleep efficiency cutoff of < 80% for actigraphy, only 29.5% of the FCs were classified as having a significant level of sleep disturbance. In addition, in this study the FCs’ sleep period time ranged from 294 to 669 minutes (i.e., 5 hours to 11 hours) and based on the actigraphy data over 55% of FCs spent more than 8 hours in bed each night.
In several studies,35,48,54,55
recommendations were made to use both subjective and objective measures to evaluate sleep because these different approaches capture different aspects of disturbed sleep. For example, subjective measures capture the physical and mental aspects of sleep and the impact of sleep on an individual’s ability to function. The use of multiple measures to evaluate sleep disturbance in FCs warrants investigation in future studies, particularly in terms of which measures are most sensitive to changes in various sleep parameters over time. This information is critical to the evaluation of the efficacy of pharmacologic and nonpharmacologic interventions to reduce sleep disturbance in FCs.
The PSQI was the most frequently used self-report measure to assess sleep quality in previous studies of FCs of oncology patients12,14,17
and in FCs of patients with Alzheimer’s disease,58
and Parkinson’s disease.60
The PSQI global score for this sample (5.7 ± 3.2) was higher than values reported for healthy controls (range 1.9 to 3.1);31,61
comparable to some studies of FCs of patients with cancer14
and Parkinson’s disease60
(range 5.5 to 5.8), but lower than studies of FCs of patients with advanced cancer (11.3),12
or advanced Alzheimer’s disease58
(range 9.1 to 10.4). Differences among these studies, in the various PSQI subscale scores followed a similar pattern. Of note, based on the PSQI data, over 24% of the FCs in this study had problems with the initiation (sleep latency subscale) and 21% had problems with maintenance (sleep duration subscale) of sleep on two to three nights per week.
This study is the first to report data on subjective sleep disturbance using the GSDS in FCs of oncology patients. Using this scale, approximately 40% of the sample reported clinically significant levels of sleep disturbance. Similar to the PSQI, 19.8% of the FCs had a problem with the initiation of sleep (i.e., sleep onset latency). However, a higher percentage of FCs (i.e., 37.1% (early awakenings) to 67.7% (mid-sleep awakenings)) reported problems with sleep maintenance. To put the total GDSD score of the FCs in this study into context (i.e., 39.1 ± 16.0), these FCs had lower scores than patients with a variety of cancer diagnoses (i.e., 54.7 and 52.1).44,62
However, FCs in this study had GSDS total scores comparable to mothers in their third trimester of pregnancy (43.9)42
and women before and after hysterectomy (42.3 and 45.7),63
but lower than nurses who worked nights (60.5) or rotated shifts (56.6).32
Because the GSDS items are rated on a scale of 0 (never) to 7 (everyday), the subscale scores provide an estimation of the number of days per week that FCs experienced a particular problem. As shown in , 95% of the FCs in this study reported an insufficient amount of sleep on 4 or more days per week. In addition, almost 68% of the FCs experienced a clinically significant number of mid-sleep awakenings on almost 4 nights per week. Actigraphy data showed that FCs averaged 17.6 brief awakenings per night which is well above healthy adult values of 2 to 6 times per night.3
This high number of mid-sleep awakenings may be related to the fact that the majority of the FCs in this study was the spouses of patients with prostate cancer who awakened multiple times during the night to urinate. Taken together, data from the PSQI and the GSDS suggest that FCs in this study had problems with the maintenance of sleep.
Compared to healthy adult values,49
all of the nocturnal sleep/rest actigraphy parameters, except sleep onset latency, were outside the normal ranges. When these FCs actigrapy data were compared to previous studies of FCs of patients with advanced cancer,8,16
sleep onset latency was shorter in this study (12.97 minutes) compared to Carter’s study (40.0 to 45.0 minutes). In addition, total sleep time was longer in this study (406.5 minutes) compared to Carter’s study (290 to 332 minues). However, sleep efficiency (84%) was similar to Carter’s study (73% to 80%), but worse than the study by Gibbin and colleagues (90% to 92%). The number of awakenings identified in this study using actigraphy (17.6) was significantly higher than that reported in studies of FCs of patients with Alzheimers Disease (6.5)64
or dementia (4.1).59
Additional studies need to evaluate for differences in sleep-wake parameters in FCs of oncology patients at different points in the patient’s disease and treatment trajectory.
Only two studies were found that reported on daytime wake/activity in FCs. In a study of FCs of patients with advanced cancer,16
based on actigraphy data, these FCs were inactive for 29% to 32% of the day and took 7 to 10 naps per day of approximately 8 to 10 minutes duration. In a study of FCs of patients with Alzheimer’s Disease,64
these FCs slept approximately 30 minutes per day which is slightly less than was found in the current study (43 minutes/day). The percent of time asleep per day in the current study (6%) was relatively low and is supported by the relatively small percentage of FCs who reported significant daytime dysfunction using both the PSQI (6.0%) and the GSDS (14.9%). This finding may be partially explained by the relatively high percentage of FCs (47.5%) who were employed. Additional research is warranted on when, how often, as well as the duration of naps in FCs of oncology patients because findings from several studies suggest that “power naps” (i.e., naps of less than 30 minutes duration that occur at around 3:00 PM) improve an individual’s ability to function throughout the rest of the day.65
While findings from several studies of primarily patients with breast cancer,48,49,66–70
suggest that circadian rhythm parameters are significantly disrupted in oncology patients, no studies were found that evaluated circadian activity rhythm parameters in FCs of oncology patients. In this study, acrophase values were similar to the general population. However, all of the other circadian rhythm parameters were below healthy adult values.49
This finding suggests that these FCs had dampened circadian rhythms with low daytime activity and higher nighttime activity. However, because circadian rhythm data were collected for only 36 hours, additional research is warranted to confirm the findings from this study.
FCs in this study reported moderate levels of morning and evening fatigue. In addition, approximately 50% reported moderate levels of attentional fatigue. An important finding is that over 35% of the FCs in this study reported low levels of morning and evening energy. As expected, increased levels of morning fatigue, as well as attentional fatigue, and decreased levels of morning energy were associated with the majority of the subscale and total scores on the PSQI and the GSDS. However, fatigue and energy scores were not correlated with the majority of the actigraphy measures.
A number of study limitations need to be acknowledged. The sample was primarily female, White, and well educated with a mean age of 62 years. Therefore, these findings may not generalize to all FCs of patients with cancer. While the sample size was relatively small, this study provides important information on sleep-wake circadian activity rhythm parameters that can be used for comparative purposes. Ideally, actigraphy data should be collected for longer periods of time to study circadian rhythms. However, data were collected only on weekdays which eliminated changes in the various objective parameters that might occur on weekends. Therefore, these results should be interpreted with caution and not generalized to both weekdays and weekends. Finally, data were not available on the specific sleep medications that these FCs used or on other medications that could contribute to sleep disturbance.
Despite these limitations, findings from this study suggest that a significant percentage of FCs of oncology patients experience clinically meaningful disturbances in sleep-wake circadian activity rhythms. Additional research is warranted on how both subjective and objective parameters change over time; on which FCs are at greatest risk for these disturbances; on what factors contribute to sleep disturbance in these FCs; and on the efficacy of various pharmacologic and nonpharmacologic interventions to decrease the levels of disturbance in these FCs. In addition, the relationships between changes in patients’ and FCs’ subjective and objective sleep disturbance parameters need to be evaluated during and following the patients’ cancer treatment.
As noted in a number of reviews,1,4,5
clinicians need to perform systematic assessments of FCs’ sleep quality. These sleep assessments should focus on difficulty falling asleep and staying asleep and excessive daytime sleepiness. In addition, the impact of sleep disturbance on FCs’ ability to function needs to be assessed. Oncology clinicians need to assist FCs and patients to follow basic sleep hygiene principles. For example, FCs and patients should be encouraged to establish regular sleep and wake times and to engage in regular exercise. FCs and patients need to be reminded to avoid excessive fluid intake and restrict the consumption of caffeinated beverages in the evening.1
Implementation of these simple measures may improve FCs’ and patients’ sleep quality during and following cancer treatment.