Abstract
Introduction
Caregiving, providing regular care or assistance to family members or friends with health problems or disabilities, may affect caregivers’ sleep. This study examined self-reported short sleep duration by caregiving status among US adults.
Methods
Data from 114,496 respondents aged ≥18 years in 19 states, the District of Columbia, and Puerto Rico from the 2016 Behavioral Risk Factor Surveillance System were analyzed. Prevalence of short sleep duration (<7 hours per 24-hour period) by caregiving status was calculated, and adjusted prevalence ratios (PR) and 95% confidence intervals (CI) were derived from a multivariable logistic regression model with adjustment for potential covariates.
Results
Nearly 1 out of 5 adults reported caregiving within the past month. A higher prevalence of short sleep duration was reported among caregivers (39.5%) than among non-caregivers (34.2%, adjusted PR [95% CI]=1.12 [1.06–1.19]). Caregivers who reported prolonged caregiving (≥5 years) reported a higher prevalence of short sleep duration than those with <2 years of caregiving. Similarly, caregivers who provided 20–39 hours of caregiving per week reported a higher prevalence of short sleep duration than those with <20 hours caregiving per week.
Conclusions
Caregivers had a higher prevalence of short sleep duration than non-caregivers. Providing information and community-based resources and supports for caregiving may minimize caregiver stress and improve sleep particularly for those with prolonged or more intense caregiving.
Keywords: Sleep deprivation, family caregivers, population surveillance, Behavioral Risk Factor Surveillance System
Introduction
An estimated 43.5 million adults (18.2% of adult population) in the US have provided unpaid care, usually to a family member or friend in the prior 12 months.1 Caregivers provide substantial economic and non-economic value to the health system, and the need for caregivers is expected to increase as the US population ages.2,3 Sleep is essential for health and wellbeing and can be affected by aging, interpersonal relationships, societal factors, and psychological stressors as well as caregiving.4–6 Sleep disturbance is prevalent among caregivers and may be related to increased anxiety, stress, and nighttime awakenings as part of caregiving duties.7–9 Prior research demonstrated that sleep disturbance10,11 or low sleep quality12 was more common among caregivers than among non-caregivers, and the association of caregiving with sleep disturbance might be impacted by both caregiver and care recipient’s health conditions.13,14 Moreover, sleep quality is only one important aspect of sleep health. Sleep duration is also critical and varies by sociodemographic characteristics such as age, sex, race/ethnicity, education, employment, and marital status; individual behaviors including smoking, exercise, and body weight; and health conditions.15,16 Some studies reported no differences between caregivers and non-caregivers regarding sleep duration; however, the sample sizes in these studies were small.10–12
We assessed the association of short sleep duration with caregiver status by sociodemographic and health-risk factors among adults from 19 states, the District of Columbia (DC), and Puerto Rico in the 2016 Behavioral Risk Factor Surveillance System (BRFSS). Among caregivers, we further assessed the association of short sleep duration with caregiving duration (years providing care), intensity (hours per week providing care), and the caregiver’s own health conditions as well as the care recipient’s health problem.12,17
Methods
Participants
BRFSS, a random-digital-dialed telephone survey of US adults aged ≥18 years, is conducted by state health departments in collaboration with the Centers for Disease Control and Prevention (CDC) in all 50 states, the District of Columbia (DC), and US territories (detailed information available at https://www.cdc.gov/brfss/). Trained interviewers administer standardized questionnaires to all adult respondents. Responses are weighted to the respondent’s probability of selection in order to obtain a study population representative of a given state. The BRFSS questionnaire includes core topics collected yearly, on a rotating basis, and optional modules collected at the discretion of states. The number of states that use a module can vary by year. The 2016 BRFSS core questionnaire included sociodemographic characteristics, health-related behaviors, and sleep duration. The 2016 BRFSS had the largest number of states that collected the optional caregiving module (Arizona, Arkansas, California, Connecticut, Colorado, Georgia, Minnesota, Missouri, Montana, Nevada, New Jersey, New York, North Dakota, Ohio, Oregon, South Dakota, Tennessee, Texas, Utah), DC, and Puerto Rico. The response rate for those 19 states, DC, and Puerto Rico with combined landline and cell phone samples ranged from 31.4% in California to 57.2% in South Dakota (https://www.cdc.gov/brfss/annual_data/2017/pdf/2017-sdqr-508.pdf). The study was approved as exempt research by the Centers for Disease Control and Prevention Institutional Review Board. A total of 134,701 adult respondents residing in 19 states, DC, and Puerto Rico completed the 2016 BRFSS questionnaire including a sleep duration question and the caregiver module.
Measures
Sleep duration
Self-reported sleep duration was based on the question, “On average, how many hours of sleep do you get in a 24-hour period?” The American Academy of Sleep Medicine and the Society for Sleep Research recommends that adults get seven or more hours of sleep per night for optimal health.15 Those who reported <7 hours of sleep were classified as having short sleep duration. Evidence-based studies have shown that sleeping <7 hours in 24 hours on a regular basis is associated with impaired immune function, loss of productivity, increased errors, and greater risk of accidents as well as adverse health conditions.15 In addition, the findings from previous studies on the association of long sleep duration (9 or more hours sleep per night) with health were not consistent and the AASM did not make a recommendation regarding long sleep duration. 15 Therefore, we only compared ≥7 hours vs. <7 hours sleep duration in this study.
Caregiver status and conditions
Caregiving status was determined by an affirmative response to a single question “During the past 30 days, did you provide regular care or assistance to a friend or family member who has a health problem or disability?”
In addition, caregivers were asked about the duration (length of time providing care to recipient) and intensity of caregiving (hours per week), and their care recipient’s condition. The duration of caregiving was categorized as “<2 years”, “2 to <5 years’, and “≥5 years” based on responses to the question “For how long have you provided care for that person?” Those who responded that they provided care for “<30 days”, “1 month to <6 months”, or “6 months to <2 years” were combined for more statistical power; a similar percent of short sleep duration was observed among those three subgroups. The intensity of caregiving was categorized as “<20 hours”, “20˗39 hours”, and “≥40 hours” per week based on responses to the question “In an average week, how many hours do you provide care or assistance?”
Caregiver’s health status was assessed as the number of chronic conditions, defined as the sum (range: 0–9) of reporting any of the following doctor-diagnosed chronic conditions: diabetes, asthma ever, chronic obstructive pulmonary disease (COPD), coronary heart disease (heart attack, myocardial infarction or angina), stroke, arthritis, any cancer, depression, or chronic kidney disease.
The care recipient’s main condition for which care was needed was based on the caregiver’s response to the question “What is the main health problem, long-term illness, or disability that the person you care for has?” We collapsed some response categories with small sample sizes together based on whether the nature of the condition was similar. The care recipient’s condition was grouped as 1) major chronic condition (arthritis/rheumatism, asthma, chronic respiratory conditions such as emphysema or COPD, diabetes, heart disease, hypertension, stroke, human immunodeficiency virus infection (HIV), other organ failure or disease such as kidney or liver problems); 2) cancer; 3) dementia (or other cognitive impairment disorders); 4) developmental disabilities such as autism, Down’s syndrome, or spina bifida; 5) mental illnesses such as anxiety, depression, or schizophrenia, or substance abuse or addiction disorder; 6) injuries; 7) old age/infirmity/frailty; 8) some other care-needed condition (not specified).
Covariates
The selected sociodemographic characteristics were age group (18–24, 25–34, 35–44, 45–64, or ≥65 years), sex, race/ethnicity (non-Hispanic white, non-Hispanic black, Hispanic, American Indian/Alaska Native, Asian, or non-Hispanic other), education (less than high school, high school diploma or an equivalent general educational development (GED), some college or technical school, or college graduate), employment status (employed, unemployed, unable to work, retired, or homemaker or student), and marital status (married or member of unmarried couple, divorced/separated/widowed, or never married).
Leisure-time physical activity was defined as a ‘yes’ response to the question, “During the past month, other than your regular job, did you participate in any physical activities or exercises such as running, calisthenics, golf, gardening, or walking for exercise?” Smoking status was defined by two questions: “Have you smoked at least 100 cigarettes in your entire life?” and “Do you currently smoke every day, some days, or not at all?” Respondents were categorized as current smokers (smoked at least 100 cigarettes during their lifetime and currently smokes every day or some days), former smokers (smoked at least 100 cigarettes during their lifetime but did not currently smoke), and never smokers (never smoked at least 100 cigarettes during their lifetime). Body mass index (BMI, kg/m2) calculated from self-reported height and weight was categorized as underweight (BMI<18.5), normal weight (BMI=18.5–24.9), overweight (BMI= 25.0–29.9), obese (BMI≥30.0).
Statistical Analysis
Data were analyzed from 114,496 adult respondents with complete information on sleep duration and caregiving status after respondents with missing values on all aforementioned covariates except for BMI were excluded in this study. For BMI, which had the largest number with missing data, we included a “missing BMI” category when adjusting for covariates in order to minimize the impact of excluding participants with valid sleep and caregiving information.
Age-adjusted prevalence of short sleep duration and 95% confidence intervals (CIs) were calculated by caregiving status and selected characteristics. Estimates were age-standardized to the 2000 projected US population aged ≥18 years.18 Multivariable logistic regression analyses were performed to assess the association of short sleep duration with caregiving status after controlling for age, sex, race/ethnicity, education, employment status, marital status, leisure-time physical activity, smoking status, and BMI category. In addition, separate multivariable logistic regression analyses among caregivers were performed to assess relationships between short sleep duration prevalence and the duration and intensity of caregiving after controlling for the aforementioned covariates, number of caregiver chronic conditions, and the care recipient’s condition. Appropriate sampling weights for the caregiver module were applied (https://www.cdc.gov/brfss/annual_data/2016/pdf/2016moduleanalysis.pdf). All analyses were conducted using SAS (Version 9.4, SAS Institute) and SAS–callable SUDAAN (Version 11.0.3, Research Triangle Institute International, Research Triangle Park, North Carolina) to account for the complex sampling design. All significant differences were set at p < 0.05.
Results
In 2016, nearly 1 out of 5 adults (19.8%) in 19 states, DC, and Puerto Rico reported being a caregiver. Compared to non-caregivers, a larger proportion of caregivers were women (59.0% vs. 49.5%), aged 45–64 years (40.9% vs. 32.0%), non-Hispanic white (62.8% vs. 56.7%), and had some college or technical school education (36.9% vs. 30.0%) (p<0.0001. Table 1). Caregivers also had a higher prevalence of obesity (31.8% vs. 27.1%), current smoking (20.0% vs. 14.3%), and short sleep duration (39.0% vs. 34.0%) compared to non-caregivers (p<0.0001, Table 1). Caregivers and non-caregivers showed similar distributions of marital status, employment status, and leisure-time physical activity.
Table 1.
Distribution of selected characteristics among 114,496 respondentsa aged ≥18 years, by caregiving status, 2016 BRFSS
| Total | Caregivers (n=22,993) | Non-caregivers (n=91,503) | ||||
|---|---|---|---|---|---|---|
| Characteristic | Nb | % (95% CI)c | Nb | % (95% CI)c | Nb | % (95% CI)c |
| Sex d | ||||||
| Men | 49,291 | 48.6 (47.7–49.5) | 8,335 | 41.0 (39.0–43.1) | 40,956 | 50.5 (49.5–51.5) |
| Women | 65,193 | 51.4 (50.5–52.3) | 14,655 | 59.0 (56.9–61.0) | 50,538 | 49.5 (48.5–50.5) |
| Age group (years) | ||||||
| 18–24 | 5,776 | 12.6 (11.9–13.4) | 882 | 11.1 ( 9.3–13.3) | 4,894 | 13.0 (12.2–13.8) |
| 25–34 | 10,736 | 17.3 (16.6–18.1) | 1,752 | 15.2 (13.7–16.9) | 8,984 | 17.8 (17.0–18.7) |
| 35–44 | 13,069 | 16.6 (15.9–17.3) | 2,471 | 14.2 (12.8–15.8) | 10,598 | 17.2 (16.4–18.0) |
| 45–64 | 43,418 | 33.8 (33.0–34.6) | 10,568 | 40.9 (39.0–42.9) | 32,850 | 32.0 (31.1–33.0) |
| ≥65 | 41,497 | 19.7 (19.1–20.2) | 7,320 | 18.5 (17.2–19.8) | 34,177 | 20.0 (19.3–20.6) |
| Race/ethnicity | ||||||
| Non-Hispanic white | 87,567 | 57.9 (57.0–58.8) | 17,701 | 62.8 (60.7–64.9) | 69,866 | 56.7 (55.7–57.7) |
| Non-Hispanic black | 7,816 | 9.8 ( 9.3–10.3) | 1,823 | 11.0 ( 9.8–12.5) | 5,993 | 9.5 ( 8.9–10.1) |
| Hispanic | 12,942 | 23.2 (22.3–24.1) | 2,075 | 19.3 (17.4–21.4) | 10,867 | 24.1 (23.2–25.1) |
| American Indian/Alaska Native | 2,142 | 1.0 ( 0.9– 1.2) | 565 | 1.3 ( 1.0– 1.6) | 1,577 | 1.0 ( 0.8– 1.1) |
| Other, non-Hispanic | 4,029 | 8.1 ( 7.4– 8.8) | 829 | 5.5 ( 4.6– 6.6) | 3,200 | 8.7 ( 7.9– 9.5) |
| Education | ||||||
| Less than high school diploma | 8,734 | 14.9 (14.2–15.7) | 1,337 | 11.2 ( 9.9–12.6) | 7,397 | 15.8 (15.0–16.7) |
| High school diploma or GED | 31,278 | 26.5 (25.8–27.3) | 5,964 | 25.9 (24.2–27.6) | 25,314 | 26.7 (25.8–27.5) |
| Some college or technical school | 32,165 | 31.3 (30.5–32.2) | 7,094 | 36.9 (34.7–39.1) | 25,071 | 30.0 (29.0–30.9) |
| College graduate | 42,319 | 27.2 (26.5–28.0) | 8,598 | 26.1 (24.6–27.7) | 33,721 | 27.5 (26.7–28.3) |
| Employment status | ||||||
| Employed | 56,462 | 56.8 (56.0–57.7) | 11,381 | 55.5 (53.5–57.6) | 45,081 | 57.2 (56.2–58.1) |
| Unemployed | 4,655 | 5.7 ( 5.3– 6.2) | 1,179 | 6.9 ( 6.0– 7.9) | 3,476 | 5.5 ( 5.0– 6.0) |
| Retired | 35,982 | 17.7 (17.1–18.2) | 6,814 | 17.7 (16.5–19.0) | 29,168 | 17.7 (17.1–18.3) |
| Unable to work | 7,745 | 6.6 ( 6.2– 7.0) | 1,731 | 7.6 ( 6.7– 8.6) | 6,014 | 6.4 ( 6.0– 6.8) |
| Homemaker or student | 9,652 | 13.1 (12.5–13.8) | 1,888 | 12.3 (10.9–13.8) | 7,764 | 13.3 (12.6–14.1) |
| Marital status | ||||||
| Married | 64,396 | 56.1 (55.2–57.0) | 14,151 | 57.3 (55.2–59.4) | 50,245 | 55.9 (54.9–56.8) |
| Divorced, separated, widowed | 32,467 | 19.7 (19.1–20.3) | 5,601 | 19.8 (18.4–21.3) | 26,866 | 19.7 (19.0–20.4) |
| Never married | 17,633 | 24.2 (23.3–25.0) | 3,241 | 22.9 (21.0–25.0) | 14,392 | 24.5 (23.6–25.4) |
| Body mass index (kg/m2) | ||||||
| Underweight (<18.5) | 1,706 | 1.9 ( 1.7– 2.2) | 316 | 1.8 ( 1.3– 2.4) | 1,390 | 1.9 ( 1.7– 2.2) |
| Normal weigh (18.5–24.9) | 34,220 | 31.3 (30.5–32.2) | 6,493 | 29.5 (27.7–31.5) | 27,727 | 31.8 (30.8–32.7) |
| Overweight (25.0–29.9) | 39,257 | 32.7 (31.8–33.5) | 7,742 | 31.4 (29.5–33.4) | 31,515 | 33.0 (32.0–33.9) |
| Obese (≥30.0) | 32,870 | 28.0 (27.2–28.9) | 7,185 | 31.8 (29.9–33.7) | 25,685 | 27.1 (26.2–28.1) |
| Missing | 6,443 | 6.1 ( 5.7– 6.5) | 1,257 | 5.5 ( 4.6– 6.7) | 5,186 | 6.2 ( 5.8– 6.7) |
| Cigarette smoking status | ||||||
| Current smoker | 16,499 | 15.4 (14.8–16.1) | 4,165 | 20.0 (18.4–21.7) | 12,334 | 14.3 (13.6–15.0) |
| Former smoker | 32,616 | 23.9 (23.1–24.6) | 6,378 | 23.5 (22.0–25.1) | 26,238 | 23.9 (23.1–24.8) |
| Never smoker | 65,381 | 60.7 (59.9–61.6) | 12,450 | 56.5 (54.4–58.5) | 52,931 | 61.8 (60.8–62.7) |
| Leisure-time physical activity | ||||||
| Yes | 85,671 | 75.8 (75.1–76.6) | 17,834 | 77.7 (75.9–79.3) | 67,837 | 75.4 (74.5–76.2) |
| No | 28,825 | 24.2 (23.4–24.9) | 5,159 | 22.3 (20.7–24.1) | 23,666 | 24.6 (23.8–25.5) |
| Short sleep duration (<7 hours/24hours) | ||||||
| Yes | 35,733 | 35.0 (34.1–35.8) | 8,313 | 39.0 (37.0–41.0) | 27,420 | 34.0 (33.0–34.9) |
| No | 78,763 | 65.0 (64.2–65.9) | 14,680 | 61.0 (59.0–63.0) | 64,083 | 66.0 (65.1–67.0) |
Respondents were from the District of Columbia, Puerto Rico, and 19 states (Arizona, Arkansas, California, Connecticut, Colorado, Georgia, Minnesota, Missouri, Montana, Nevada, New Jersey, New York, North Dakota, Ohio, Oregon, South Dakota, Tennessee, Texas, Utah) that administered the caregiving module in 2016.
Unweighted sample size.
Weighted distribution of selected characteristics overall and by caregiving status. The sum of percentage of multiple-level variable may not be equal to 100% due to rounding.
12 respondents who reported “refused to answer” for sex variable were categorized as “other”, which was not showed due to unreliable estimates in the table but were included in the analyses.
The significant association of short sleep duration with caregiving status overall persisted after controlling for sociodemographic characteristics and adverse risk behaviors (overall adjusted PR [95% CI] =1.12 [1.06–1.19]). Among subgroups, a higher adjusted prevalence ratio of short sleep duration among caregivers versus non-caregivers was observed among both men and women, adults aged ≥35 years, non-Hispanic whites, those with ≤12 years education, employed persons, those who were married or divorced/separated/widowed, those who were underweight, former smokers or never smokers, and those with and without leisure-time physical activity (Table 2).
Table 2.
Age-adjusted prevalence of short sleep duration (<7 hours/24 hours) and adjusted prevalence ratio by caregiver status within selected characteristics among 114,496 adult respondents ≥18 yearsa, 2016
| Characteristic | Caregivers % (95% CI)b | Non-caregivers % (95% CI)b | Adjusted prevalence ratio (95% CI)c |
|---|---|---|---|
| Age-adjusted prevalence | 39.5 (37.4–41.7) | 34.2 (33.2–35.3) | 1.12 (1.06–1.19) |
| Sex d | |||
| Men | 39.7 (36.4–43.1) | 34.8 (33.3–36.2) | 1.11 (1.02–1.22) |
| Women | 39.2 (36.6–41.9) | 33.7 (32.3–35.1) | 1.13 (1.05–1.22) |
| Age groups (years) | |||
| 18–24 | 29.2 (22.3–37.3) | 31.8 (28.6–35.2) | 0.91 (0.69–1.20) |
| 25–34 | 43.7 (38.2–49.4) | 36.4 (34.0–38.9) | 1.13 (0.98–1.30) |
| 35–44 | 47.1 (41.6–52.6) | 36.5 (34.0–39.1) | 1.21 (1.06–1.38) |
| 45–64 | 40.6 (37.8–43.5) | 36.3 (34.6–38.0) | 1.13 (1.04–1.22) |
| ≥65 | 31.3 (27.9–35.0) | 27.2 (25.8–28.7) | 1.17 (1.04–1.31) |
| Race/ethnicity | |||
| Non-Hispanic white | 38.5 (35.6–41.4) | 32.4 (31.3–33.6) | 1.14 (1.07–1.22) |
| Non-Hispanic black | 48.3 (42.0–54.6) | 43.8 (40.6–46.9) | 1.11 (0.96–1.28) |
| Hispanic | 41.0 (36.1–46.0) | 34.0 (31.8–36.4) | 1.10 (0.94–1.29) |
| American Indian/Alaska Native | 44.8 (35.0–55.2) | 34.5 (28.1–41.6) | 1.31 (0.99–1.73) |
| Other, Non-Hispanic | 42.6 (34.9–50.7) | 41.2 (36.6–46.1) | 0.95 (0.75–1.21) |
| Education | |||
| Less than high school diploma | 45.8 (39.3–52.4) | 32.0 (29.4–34.8) | 1.30 (1.11–1.52) |
| High school diploma or GED | 42.2 (38.7–45.7) | 36.9 (35.0–38.7) | 1.12 (1.01–1.24) |
| Some college or technical school | 42.9 (39.1–46.8) | 37.9 (35.9–39.9) | 1.11 (1.00–1.23) |
| College graduate | 32.7 (29.4–36.2) | 29.3 (27.7–30.9) | 1.05 (0.95–1.17) |
| Employment status | |||
| Employed | 40.8 (37.7–44.0) | 35.4 (33.9–36.9) | 1.16 (1.07–1.25) |
| Unemployed | 49.0 (41.2–56.9) | 35.5 (31.5–39.8) | 1.20 (1.00–1.44) |
| Retired | 54.6 (42.7–66.1) | 44.4 (34.8–54.5) | 1.13 (1.00–1.28) |
| Unable to work | 48.8 (41.8–55.9) | 47.5 (43.3–51.8) | 0.97 (0.85–1.10) |
| Homemaker or student | 32.4 (27.6–37.7) | 28.4 (25.2–31.8) | 1.01 (0.83–1.23) |
| Marital status | |||
| Married | 37.2 (33.5–41.1) | 32.5 (30.9–34.1) | 1.12 (1.03–1.21) |
| Divorced, separated, widowed | 50.1 (43.9–56.2) | 40.3 (37.1–43.6) | 1.26 (1.15–1.39) |
| Never married | 38.1 (33.9–42.5) | 37.7 (35.4–39.9) | 1.00 (0.87–1.15) |
| Body mass index (kg/m2) | |||
| Underweight (<18.5) | 45.9 (35.9–56.2) | 33.3 (26.4–41.0) | 1.61 (1.13–2.30) |
| Normal weigh (18.5–24.9) | 35.3 (32.0–38.7) | 31.5 (29.8–33.3) | 1.10 (0.98–1.22) |
| Overweight (25.0–29.9) | 37.1 (33.1–41.2) | 34.8 (33.0–36.7) | 1.07 (0.97–1.19) |
| Obese (≥30.0) | 42.5 (39.1–45.9) | 38.5 (36.4–40.7) | 1.10 (1.00–1.21) |
| Missing | 48.5 (41.0–56.0) | 30.0 (26.9–33.4) | 1.50 (1.24–1.80) |
| Cigarette smoking status | |||
| Current smoker | 46.3 (41.9–50.8) | 42.3 (39.8–44.8) | 1.08 (0.97–1.20) |
| Former smoker | 41.1 (35.9–46.6) | 34.6 (32.2–37.1) | 1.13 (1.02–1.26) |
| Never smoker | 36.8 (34.0–39.6) | 32.2 (30.9–33.5) | 1.12 (1.03–1.22) |
| Leisure-time physical activity | |||
| Yes | 36.8 (34.5–39.2) | 32.6 (31.5–33.8) | 1.09 (1.02–1.17) |
| No | 48.7 (43.3–54.2) | 38.8 (36.7–41.0) | 1.20 (1.08–1.33) |
Respondents were from the District of Columbia, Puerto Rico, and 19 states (Arizona, Arkansas, California, Connecticut, Colorado, Georgia, Minnesota, Missouri, Montana, Nevada, New Jersey, New York, North Dakota, Ohio, Oregon, South Dakota, Tennessee, Texas, Utah) that administered the caregiving module in 2016.
Except for age groups, estimates were age-adjusted to the 2000 projected U.S. population aged ≥18 years. https://www.cdc.gov/nchs/data/statnt/statnt20.pdf.
Adjusted prevalence ratio and 95% confidence interval (CI) were derived from multivariable logistic regression model with sex, age group, race/ethnicity, education, employment status, marital status, BMI category, cigarette smoking status, and leisure-time physical activity as covariates.
12 respondents who reported “refused to answer” for sex variable were categorized as “other”, which was not showed due to unreliable estimates in the table but were included in the analyses.
Among caregivers, 38.0% reported no chronic conditions, 30.0% reported one, 17.3% reported two, 12.1% reported three to four, and 2.7% reported five or more chronic conditions (Table 3). Caregivers who had at least one chronic condition were more likely to report short sleep duration than those with none of the nine chronic conditions. Additionally, the prevalence ratio of short sleep duration increased with the number of co-occurring conditions.
Table 3.
Age-adjusted prevalence of short sleep duration (<7 hours/24 hours) associated with duration and intensity of caregiving among 22,640 caregivers aged ≥18 years, 2016 BRFSS
| Caregiving characteristic | na (%) | Age-adjusted short sleep duration, % (95% CI)b | Adjusted prevalence ratioc |
|---|---|---|---|
| Number of caregiver’s chronic conditions d | |||
| None | 7,118 (38.0) | 34.5 (31.4–37.7) | 1.00 (referent) |
| 1 | 6,746 (30.0) | 41.7 (37.6–45.8) | 1.16 (1.02–1.32) |
| 2 | 4,569 (17.3) | 39.3 (34.8–43.9) | 1.18 (1.03–1.35) |
| 3 | 2,369 ( 8.0) | 49.4 (42.5–56.4) | 1.24 (1.06–1.45) |
| 4 | 1,138 ( 4.1) | 52.2 (42.3–61.9) | 1.29 (1.06–1.55) |
| ≥5 | 699 ( 2.7) | 69.6 (61.5–76.6) | 1.60 (1.31–1.96) |
| Care recipient’s condition | |||
| Major chronic conditione | 5,285 (25.2) | 42.6 (38.7–46.7) | 1.15 (1.01–1.31) |
| Cancer | 1,782 ( 7.2) | 42.6 (36.7–48.8) | 1.21 (1.02–1.43) |
| Dementia | 2,147 ( 9.1) | 36.8 (31.1–42.8) | 1.10 (0.91–1.33) |
| Developmental disabilities such as autism, Down’s syndrome, spina bifida | 846 ( 4.0) | 43.2 (34.8–52.0) | 1.18 (0.94–1.49) |
| Mental illnesses, substance abuse | 1,121 ( 6.2) | 34.0 (27.1–41.7) | 1.00 (0.79–1.28) |
| Injuries | 2,517 (15.2) | 37.0 (31.2–43.2) | 1.07 (0.89–1.27) |
| Old age/infirmity/frailty | 2,543 ( 8.8) | 39.9 (32.8–47.5) | 1.06 (0.89–1.25) |
| Other unspecified care-needed condition | 5,414 (20.6) | 35.6 (31.6–39.8) | 1.00 (referent) |
| Duration of caregiving (year) | |||
| <2 | 11,444 (50.2) | 36.3 (33.6–39.2) | 1.00 (referent) |
| 2–<5 | 4,521 (19.4) | 40.1 (36.0–44.4) | 1.08 (0.96–1.21) |
| ≥5 | 6,675 (30.4) | 44.9 (40.8–49.2) | 1.19 (1.06–1.32) |
| Intensity of caregiving (hours/week) | |||
| <20 | 15,248 (69.7) | 37.0 (34.5–39.7) | 1.00 (referent) |
| 20–39 | 2,063 (11.0) | 47.2 (41.2–53.2) | 1.20 (1.04–1.38) |
| ≥40 | 4,136 (19.3) | 43.2 (38.6–48.0) | 1.13 (1.00–1.27) |
Unweighted sample size. Due to missing responses, the total sample size for analysis by duration of caregiving does not equal that for analysis by intensity of caregiving and percentage may not be added up to 100%.
Age-adjusted to the 2000 projected U.S. population aged ≥18 years. https://www.cdc.gov/nchs/data/statnt/statnt20.pdf
Adjusted prevalence ratio and 95% confidence interval (CI) were derived from multivariable logistic regression model among caregivers with sex, age group, race/ethnicity, education, employment status, marital status, BMI category, cigarette smoking status, leisure-time physical activity, number of caregiver’s chronic condition, and care recipient’s condition.
Caregiver’s chronic condition including diabetes, asthma ever, chronic obstructive pulmonary disease (COPD), coronary heart disease (heart attack, myocardial infarction or angina), stroke, arthritis, any cancer, depression, or chronic kidney disease.
Major chronic condition including arthritis/rheumatism, asthma, chronic respiratory conditions such as emphysema or COPD, diabetes, heart disease, hypertension, stroke, human immunodeficiency virus infection (HIV), other organ failure or disease such as kidney or liver problems.
Of the care recipients, 25.2% had major chronic conditions, 15.2% had injury, 9.1% had dementia, 8.8% were reported with old age/infirmity/frailty, 7.2% had cancer, 6.2% had mental illness or substance abuse, 4.0% had developmental disability, and 20.6% had other care-needed conditions (Table 3). Caregivers caring for those with a major chronic condition or cancer were more likely to report short sleep duration than caregivers caring for those with other unspecified care-needed conditions after multivariable adjustment.
Almost half (49.8%) of caregivers had ≥2 years caregiving (i.e., duration) and nearly one third (30.3%) provided ≥20 hours of caregiving per week (i.e., intensity) (Table 3). Caregivers who had ≥5 years caregiving reported a higher prevalence of short sleep duration than those who had <2 years caregiving, after controlling for all covariates including caregiver’s and care recipient’s chronic conditions. Similarly, caregivers who provided 20–39 hours of caregiving per week reported a higher prevalence of short sleep duration than those who provided <20 hours of caregiving per week.
Discussion
Our results showed that caregiving status was independently associated with a higher prevalence of short sleep duration, both overall and within particular demographic subgroups (e.g. both men and women, those aged 45–64 years, less educated, and both employed and unemployed adults). Furthermore, short sleep duration was associated with extended duration of providing care as well as the number of hours of caregiving provided per week. To the best of our knowledge, this is the first report to demonstrate an association between self-reported short sleep duration and caregiving status in a large dataset of community dwelling adults. Short sleep duration may lead to chronic sleep loss19 and insomnia20 especially when caregiving is prolonged. Without intervention, caregivers with sleep loss or insomnia are at a higher risk for developing anxiety or other risk behaviors such as smoking and alcohol use, which may further disturb sleep.21,22 Short sleep duration is associated with chronic conditions and can lead to motor vehicle crashes and accidents at work (https://www.cdc.gov/sleep/index.html). Therefore, getting enough sleep is essential to a caregiver’s health and well-being.
Our results suggested that short sleep duration among caregivers versus non-caregivers might be greater among persons with adverse health status (more chronic conditions, underweight) and limited resources (e.g. lower education). For example, caregivers who were underweight were 61% more likely to have short sleep duration than non-caregivers who were underweight. Our data indicated that underweight caregivers were more likely to have 3 or more common chronic conditions such as cancer and heart disease than did underweight non-caregivers (19.1% vs. 9.5%, data not shown). In addition, caregivers with <12 years education were 21% more likely to have short sleep duration than non-caregivers with <12 years education. These results are in line with prior studies suggesting that factors such as chronic conditions23 and living conditions among those with low socioeconomic status24 affect short sleep duration and might be considered among caregivers with short sleep duration as well.
This study is based on a survey representing a large sample size of adults in 19 states, DC, and Puerto Rico. Our results on the overall caregiver percentage (19.8%) are comparable to that from the AARP national caregiving study in 2015 (18.2%).1 The BRFSS was designed as a general health survey, and more detailed information about caregiving, including actual activities, perceived burden or stress, or even positive affect such as fulfillment in caring for a loved one, which may offer further insight into the relationship of caregiving to sleep health, are not collected. Furthermore, our findings are subject to the following limitations. First, our results were obtained from a cross-sectional study so we cannot draw causal inference about whether caregiving leads to short sleep duration. Second, self-reported responses to all questions in the BRFSS may be subject to recall bias, which could influence our results. Although previous studies using Actigraphy validated subjective sleep duration as a reliable measure for sleep duration,25 further studies using Polysomnography measures of sleep may reveal more insights of the impact of caregiving on sleep. Third, the low response rate in BRFSS may result in selection bias and influence our results. Lastly, this study was conducted in 19 states, DC, and Puerto Rico so our findings are not generalizable to the whole US.
Conclusion
Caregivers of friends or family members had a higher prevalence of short sleep duration than non-caregivers. The overall effect size of caregiving on short sleep duration is small. However, informal caregivers are a major support to the health system. Our data indicated that about 22.4 million persons provided care in 19 states, DC, and Puerto Rico in 2016, which is comparable to 43.5 million caregivers in the U.S. overall according to the 2015 AARP national caregiver report.1 Therefore, it is of public health importance to explore sleep duration among caregivers because it affects the health of both caregivers and care recipients. Caregivers may benefit from employer- and community-based programs designed for caregivers. There are several caregiver support programs or toolkits available to those caring for people with specific conditions, for example, REACH OUT for dementia (https://www.cdc.gov/aging/caregiving/activities.htm); The Alzheimer’s Association caregiving programs for Alzheimer’s disease and dementia (https://www.alz.org/help-support/caregiving); the COPD caregiver programs from the COPD Foundation (https://www.copdfoundation.org/Learn-More/I-am-a-Caregiver/The-COPD-Caregiver.aspx); and heart failure caregiving programs from the American Heart Association (https://www.heart.org/en/health-topics/heart-failure/living-with-heart-failure-and-managing-advanced-hf/help-for-heart-failure-caregivers).
In conclusion, maintaining and improving the health status of caregivers, including their sleep health, is critical as the need for caregivers can be expected to increase with the aging population, while at the same time the availability of caregivers is expected to decrease.26 Community supports can benefit caregivers. In addition, primary health care providers can provide information about sleep hygiene to family caregivers in order to promote the caregiver’s awareness of sleep problems and improve caregiver’s sleep and health.
List of abbreviations:
- BRFSS
Behavioral Risk Factor Surveillance System
- APR
adjusted prevalence ratio
- CI
confidence interval
- GED
General Educational Development
Footnotes
Disclaimer: The findings and conclusions in this report are those of the authors and do not necessarily reflect the official position of the Centers for Disease Control and Prevention
References
- 1.The National Alliance for Caregiving and AARP Public Policy Institute. Caregiving in the U.S. 2015 2015. [Google Scholar]
- 2.Rabarison KM, Bouldin ED, Bish CL, McGuire LC, Taylor CA, Greenlund KJ. The economic value of informal caregiving for persons with dementia: results from 38 states, the District of Columbia, and Puerto Rico, 2015 and 2016 BRFSS. Am J Public Health. 2018; 108 (10): 1370–1377. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Talley RC, Crews JE. Framing the public health of caregiving. Am J Public Health. 2007; 97 (2): 224–228. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Kim EJ, Dimsdale JE. The effect of psychosocial stress on sleep: a review of polysomnographic evidence. Behav Sleep Med. 2007; 5 (4): 256–278. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Byun E, Lerdal A, Gay CL, Lee KA. How adult caregiving impacts sleep: a systematic review. Curr Sleep Med Rep. 2016; 2 (4): 191–205. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Grandner MA. Sleep, health, and society. Sleep Med Clin. 2017; 12 (1): 1–22. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Byun E, Riegel B, Sommers M, Tkacs N, Evans L. Effects of uncertainty on perceived and physiological stress in caregivers of stroke survivors: A 6-week longitudinal study. J Gerontol Nurs. 2017; 43 (11): 30–40. [DOI] [PubMed] [Google Scholar]
- 8.Joling KJ, van Marwijk HW, Veldhuijzen AE, et al. The two-year incidence of depression and anxiety disorders in spousal caregivers of persons with dementia: who is at the greatest risk? Am J Geriatr Psychiatry. 2015; 23 (3): 293–303. [DOI] [PubMed] [Google Scholar]
- 9.McCurry SM, Logsdon RG, Teri L, Vitiello MV. Sleep disturbances in caregivers of persons with dementia: contributing factors and treatment implications. Sleep Med Rev. 2007; 11 (2): 143–153. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.McKibbin CL, Ancoli-Israel S, Dimsdale J, et al. Sleep in spousal caregivers of people with Alzheimer’s disease. Sleep. 2005; 28 (10): 1245–1250. [DOI] [PubMed] [Google Scholar]
- 11.Castro CM, Lee A, Bliwise DC, Urizar GG. Sleep patterns and sleep-related factors between caregiving and non-caregiving women. Behav Sleep Med 2009; 7 (3): 164–179. [DOI] [PubMed] [Google Scholar]
- 12.von Kanel R, Ancoli-Israel S, Dimsdale JE, et al. Sleep and biomarkers of atherosclerosis in elderly Alzheimer caregivers and controls. Gerontology. 2010; 56 (1): 41–50. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Kotronoulas G, Wengstrom Y, Kearney N. Sleep and sleep-wake disturbances in care recipient-caregiver dyads in the context of a chronic illness: a critical review of the literature. J Pain Symptom Manage. 2013; 45 (3): 579–594. [DOI] [PubMed] [Google Scholar]
- 14.McCurry SM, Vitiello MV, Gibbons LE, Logsdon RG, Teri L. Factors associated with caregiver reports of sleep disturbances in persons with dementia. Am J Geriatr Psychiatry. 2006; 14 (2): 112–120. [DOI] [PubMed] [Google Scholar]
- 15.Watson NF, Badr MS, Belenky G, et al. Recommended amount of sleep for a healthy adult: a joint consensus statement of the American Academy of Sleep Medicine and Sleep Research Society. Sleep. 2015; 38 (6): 843–844. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Grandner MA, Hale L, Moore M, Patel NP. Mortality associated with short sleep duration: The evidence, the possible mechanisms, and the future. Sleep Med Rev. 2010; 14 (3): 191–203. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Simpson C, Carter P. Dementia behavioral and psychiatric symptoms: Effects on caregiver’s sleep. J Clin Nurs. 2013; 22: 3042–3052. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Klein RJ, Schoenborn CA. Age adjustment using the 2000 projected U.S. population. Healthy People 2000 Stat Notes. 2001; (20): 1–9. [PubMed] [Google Scholar]
- 19.Van Dongen HP, Rogers NL, Dinges DF. Sleep debt: Theoretical and empirical issues. Sleep Bio Rhyth. 2003; 1: 5–13. [Google Scholar]
- 20.Bonnet MH, Arand DL. Hyperarousal and insomnia: state of the science. Sleep Med Rev. 2010; 14 (1): 9–15. [DOI] [PubMed] [Google Scholar]
- 21.Salgado-Garcia FI, Zuber JK, Graney MJ, Nichols LO, Martindale-Adams JL, Andrasik F. Smoking and smoking increase in caregivers of Alzheimer’s patients. Gerontologist. 2015; 55 (5): 780–792. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Rospenda KM, Minich LM, Milner LA, Richman JA. Caregiver burden and alcohol use in a community sample. J Addict Dis. 2010; 29 (3): 314–324. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Liu Y WA, Chapman DP, Croft JB. Sleep duration and chronic diseases among U.S. adults age 45 years and older: evidence from the 2010 Behavioral Risk Factor Surveillance System. sleep. 2013; 36 (10): 8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Stamatakis KA, Kaplan GA, Roberts RE. Short sleep duration across income, education, and race/ethnic groups: population prevalence and growing disparities during 34 years of follow-up. Ann Epidemiol. 2007; 17 (12): 948–955. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Lauderdale DS, Knutson KL, Yan LL, Liu K, Rathouz PJ. Self-reported and measured sleep duration: how similar are they? Epidemiology. 2008; 19 (6): 838–845. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Redfoot D, Feinberg L, Houser A. The aging of the Baby Boom and the growing care gap: A look at future declines in the aviability of family caregivers. Washington, DC.: AARP Public Health Institute;2013. [Google Scholar]