Impact of Recent Job Loss on Sleep, Energy Consumption and Diet

Salma Batool-Anwar, MD, MPH

Candace Mayer

Patricia L. Haynes, PhD

Yilin Liu

Cynthia A. Thomson, PhD, RDN

Stuart F. Quan, MD

Division of Sleep and Circadian Disorders, Brigham and Women’s Hospital, Harvard Medical School, Boston, MA; Mel and Enid Zuckerman College of Public Health, University of Arizona, Tucson, AZ

Abstract 

To examine how sleep quality and sleep duration affect caloric intake among those experiencing involuntary job loss.

Methods

Adequate sleep and self-reported dietary recall data from the Assessing Daily Activity Patterns through Occupational Transitions (ADAPT) study was analyzed. Primary sleep indices used were total sleep time, time spent in bed after final awakening, and sleep quality as measured by the Daily Sleep Diary (DSD). Mean Energy consumption (MEC) was the primary nutritional index. Secondary indices included diet quality using the Health Eating Index 2015 (HEI), and self-reported intake of protein, carbohydrates and fats.

Results

The study participants were comprised mainly of women (61%) and non-Hispanic white. The participants had at least 2 years of college education and mean body mass index of 30.2±8.08 (kg/m ^{2 (}). The average time in bed was 541.8 (9 hrs) ±77.55 minutes and total sleep time was 461.1 (7.7 hrs) ±56.49 minutes. Mean sleep efficiency was 91±6%, self-reported sleep quality was 2.40±0.57 (0-4 scale, 4 = very good), and minutes earlier than planned morning awakening were 14.36±24.15. Mean HEI score was 47.41±10.92. Although the MEC was below national average for both men and women, male sex was associated with higher MEC. In a fully adjusted model sleep quality was positively associated with MEC.

Conclusion

Daily overall assessments of sleep quality among recently unemployed persons were positively associated with mean energy consumption. Additionally, the diet quality of unemployed persons was found to unhealthier than the average American and consistent with the relationship between poor socioeconomic status and lower diet quality.

Abbreviations

• SES: Socioeconomic status
• ADAPT: Assessing Daily Activity Patterns through Occupational Transitions
• UI: unemployment insurance
• MEC: mean energy consumption
• BMI: body mass index
• USDA: United States Department of Agriculture
• HEI: Healthy Eating Index
• DSD: Daily Sleep Diary
• TST: total sleep time
• EMA: earlier than desired morning awakening
• SD: Standard Deviation
• TIB: time in bed

Introduction

Obesity is a major public health concern; 38.3% of women and 34.3% of men in the United States are obese. It is not only the result of low physical activity and overconsumption of high energy yielding foods, socioeconomic status also plays a major role (1,2). Obesity disproportionately affects people of lower socioeconomic status (SES) in part because their limited financial resources result in consumption of calorically dense unhealthy food. This contributes to the risk of developing obesity (3).

A major determinant of SES is employment status. Unemployment is one indicator of a reduced SES and is associated with higher levels of stress. Unemployment includes involuntary job loss which is an important disruptive life event. It can cause additional unanticipated psychological and economic stress with the former afflicting women disproportionately (4). Involuntary job loss is positively associated with greater symptoms of depression, disruptions in daily routine changes and poor sleep quality (5). Unemployment has been shown to be positively associated with obesity (6,7). In one study, women were more likely to be diagnosed with obesity after involuntary job loss (8).

Sleep is another factor that affects obesity risk. Lower sleep quality and reductions in sleep duration have been shown to increase food intake resulting in becoming overweight or obese (9). Sleep deficiency can change the secretory pattern of leptin and ghrelin leading to hunger and a craving for calorically dense food (10,11).

There are no prior studies that have investigated the impact of whether sleep quality or sleep duration influences caloric consumption in those that have experienced involuntary job loss. The Assessing Daily Activity Patterns through Occupational Transitions (ADAPT) Study is an ongoing longitudinal cohort study of individuals who have suffered involuntary job loss. We analyzed cross sectional data from the baseline assessment of the ADAPT study and hypothesized that disrupted, short sleep would be associated with increased energy intake among these individuals.

Methods

Participants

Study participants were part of the on-going ADAPT Study, an 18-month longitudinal study examining changes in sleep, social rhythms, and obesity following an involuntary job loss (12). The study protocol and recruitment strategy have been described in detail previously. Briefly, all individuals who applied for unemployment insurance (UI) in the greater Tucson, Arizona and surrounding areas between October 2015 and December 2018 received study recruitment flyers within their UI intake packets. Interested individuals contacted study staff and completed phone screens assessing exclusion criteria; potentially eligible individuals were then scheduled for in-person screening visits. Individuals were eligible if they had experienced an involuntary job loss within 90 days of study enrollment, had been with their employer for at least six months, were currently employed less than 5 hours per week and did not complete any night shift work within the last 30 days. During the in-person screening, participants provided written informed consent, as well as information about their demographics, employment and medical history. They also were screened for homelessness, existing physiological and mental health conditions, substance abuse, and major sleep diagnoses which could interfere with social rhythms and sleep patterns. Those who passed screening completed validated mental health and sleep diagnostic interviews. An overnight at-home screening for sleep apnea was performed utilizing the ApneaLink Plus^TM (ResMed, San Diego, CA) to exclude moderate sleep apnea as a cause of sleep disruption.

Data used in this analysis originated from the study’s baseline visit. Of the 446 adults who provided written consent, 191 participants met eligibility criteria and completed a baseline assessment visit, including an at-home data collection period lasting two weeks. Participants were considered for the current analysis if there was an acceptable assessment of sleep and diet on their sleep diaries and dietary recalls respectively for analysis. However, 8 participants were excluded as outliers because their mean energy consumption (MEC) was significantly less than commonly reported norms (13,14). Descriptive statistics for the study sample are reported in Table 1.

Click here to view Table 1 enlarged in a new window

Measures

Demographic and Anthropometric

Age, ethnicity and biological sex were collected during the initial interview. Height and weight were measured to calculate the body mass index (kg/m²,BMI).

Diet Assessment

During the two-week, at-home baseline data collection period, participants completed up to three 24-hour dietary recalls administered by trained diet assessors at the Behavioral Measurements and Interventions Shared Resource of the University of Arizona Cancer Center utilizing the gold-standard USDA Multi-pass Dietary recall and the Nutrient Database System of the University of Minnesota for nutrient analysis (15,16). These interviews were supported by the Remote Food Photography Method, in which participants took pictures of all food and beverages prior to consumption, as well as after they had finished eating and drinking (17). Photos were used to review recall as a final verification of the multi-pass data. The diet recalls provided information on the types and quantity of food, including energy and nutrient values. At least 3 dietary recalls were completed by 172 participants (95.6% of the cohort). The primary index for this analysis was MEC (kcal). Secondary indices included diet quality estimated using the Health Eating Index 2015 (HEI) using standardized approaches to score, and self-reported intake of protein, carbohydrates and fats (18).

Sleep Measures

Sleep variables of interest were measured using the valid and reliable Daily Sleep Diary (DSD), the recommended subjective sleep assessment instrument of the insomnia research consensus panel (19). Upon wakening from their sleep, participants completed the DSD via mobile application. The DSD was completed at least 15 times by 170 participants with only 3 participants completing less than 4 days of diary data. The primary indices of interest were total sleep time (TST) and minutes earlier than desired morning awakening (EMA) as indices of sleep duration. Additionally, sleep onset latency, sleep efficiency, number of wakes after sleep onset episodes and time in bed were calculated. Self-reported sleep quality was assessed using the 5 point Likert scale incorporated into the DSD (19).

Statistical Analysis

For baseline characteristics, mean (SD) for continuous variables and percentages for categorical variables were calculated. After removing the extreme outliers we fitted linear regression models to predict MEC and HEI with sleep indices as predictors. Finally, five models were performed in multiple regression analysis. These models were adjusted for age, gender, education level, and body mass index. The results from regression analysis are presented as β-coefficients (standardized/unstandardized) with p values. In Model 1 we included energy consumption as predicted by total sleep time. In Model 2 we included energy consumption as predicted by time in bed. In Model 3, we included energy consumption as predicted by early morning awakenings. In Model 4, we included energy consumption as predicted by sleep quality. In our final Model 5, we included energy consumption as predicted by all sleep indices combined from other models. The level of statistical significance for all models was set at 0.05. To test the robustness of the analysis, we conducted a sensitivity analysis by excluding 3 participants who did not completed the DSD at least four times. All statistical analyses were done using STATA version 11 (StataCorp, LLC, College Station, TX, USA).

Results

The demographic and anthropometric characteristics of the study population are described in Table 1 (see above). Most participants were women (61%) and non-Hispanic white with the remainder primarily Hispanic. On average, study participants had at least 2 years of college education and had a mean body mass index of 30.22 (kg/m²).

As shown on Table 1, average time in bed (TIB) and total sleep time (TST) were 541.8 minutes (9.0 hrs) ±77.55 and 461.1 minutes (7.6 hrs) ±56.49 minutes respectively. However, the range was large with minimums of (5.7 hrs for TIB/ 5.2 hrs for TST) and maximums of (15.2 hrs for TIB and 10.1 for TST) respectively. Mean sleep efficiency was within normal limits (91%). There generally were few episodes of wake after sleep onset, and only a small amount of time (14 min) spent awake in the morning (EMA). Self-reported sleep quality was 2.39 using a 5 point Likert scale from 0 to 4 representing very poor, poor, good and very good sleep respectively.

Table 2 shows the MEC as well as the intake of protein, carbohydrates, fats, fruits and vegetables.

Click here to view Table 2 enlarged in a new window

Normative values for men and women in the United States are provided for comparison. The MEC was below averages for national data in both men and women (20-21). Absolute intake of protein and carbohydrates was higher than recommended levels, but was within recommended levels as a percent of energy consumption. Absolute intake of total fat was at the higher limit of recommended levels, but exceeded them as a percent of energy consumption. Consumption of fruit and vegetables were below recommendations. The HEI also was markedly reduced (Men: 43.9±8.9; Women: 49.9±10.9 vs. 59 for average American diet) (21).

Univariate regression models examining the impact of age, gender, education and BMI on MEC are presented in Table 3.

Click here to view Table 3 enlarged in a new window

Male gender was associated with higher MEC but age, education and BMI were not. Univariate regression models assessing the effect of the various sleep variables demonstrated that only TST, TIB, EMA and sleep quality significantly affected MEC (data not shown). Therefore, each of these factors were included by themselves in multivariate regression models that also incorporated age, gender, education and BMI (Table 3). Sleep quality was positively associated with MEC while EMA was negatively associated. There was no significant relationship between MEC and TST. A final model integrating TST, TIB, EMA and sleep quality showed that only sleep quality was associated with higher MEC, but EMA had no significant impact. In a sensitivity analysis, we excluded the participants who had not completed the DSD at least 4 times. However, this did not materially change the results.

Regression models were calculated to examine the impact of sleep on dietary components and the HEI. None were shown to be significant (data not shown).

Discussion

In this study of persons who recently involuntarily became unemployed, we did not find any significant associations between their MEC and various parameters related to sleep duration and sleep fragmentation. However, overall positive subjective sleep quality was associated with greater MEC. Individual dietary components and the HEI also were not related to sleep duration or fragmentation but did indicate that the diet of involuntarily unemployed persons is of lower quality, based on HEI 2015 scoring, than average for US adults (18,21).

In most, but not all studies, sleep duration has been shown to be inversely associated with MEC. In contradistinction, our analyses did not find any significant relationship with respect to either TST or TIB. Although minutes spent awake in the morning before getting out of bed (EMA) was inversely associated with MEC, this association was borderline and not significant in a fully adjusted model. Similarly, sleep efficiency and number of wakes after sleep onset episodes were not related to MEC. Explanations for increased MEC with restricted sleep duration or fragmentation include but are not limited to changes in the relative levels of satiety and hunger hormones, greater available time to eat, altered timing of meals and hedonic feeding (22). Our data suggest that in this population, the impact of these factors is not sufficient enough to alter MEC. Importantly, a large body of evidence suggests under-reporting of dietary intake is associated with obesity, female sex and lower education and may be more common among Hispanics who accounted for 33% of our sample (23,24). Systematic under-reporting of intake may have undermined our ability to capture significant associations between energy intake and sleep in this study.

Subjective sleep quality was positively correlated with MEC; better sleep quality was associated with higher levels of MEC. The direction of this finding is inconsistent with previous studies that have noted better sleep quality is associated with more nutritious diets and less obesity (25,26). The lack of agreement between daily subjective overall sleep quality, and specific individual subjective sleep quality metrics as well as objective sleep quality instruments (e.g., actigraphy) has been reported previously. In a recent study of sleep quality in older adults, the specific measures assessed by the DSD used in this study was compared to the Pittsburgh Sleep Quality Index as well as subjective sleep quality recorded in the diary. Little agreement was observed among all three measures (27). Furthermore, subjective estimates of sleep or alertness have been shown to be a poor predictor of other aspects of human behavior and performance (28,29). Our findings provide a unique perspective on the use of the DSD, an instrument that is considered a gold-standard for the assessment of sleep in persons with insomnia subject to less retrospective recall bias that global estimates of sleep quality.

We observed that the diet of recently unemployed persons differed in many categories from recommendations and guidelines made by the Institute of Medicine and the US Department of Agriculture. Additionally, the HEI of the average American is already suboptimal at 59 and mean scores appear to be even lower in our sample of unemployed individuals. Food cost is inversely correlated with diet quality and is one factor that contributes to the higher prevalence of unhealthy diets in those with lower socioeconomic status (30). Our findings extend these previous observations by demonstrating the adverse economic impact of recent job loss is associated with worse diet quality.

In conclusion, in recently unemployed persons, subjective diary assessments of sleep quality were not associated with mean energy consumption. However, the diet quality of unemployed persons was found to unhealthier than the average American and consistent with the relationship between poor socioeconomic status and lower diet quality.

Acknowledgements

The authors would like to thank the staff and participants of the Assessing Daily Activity Patterns Through Occupational Transitions Study (ADAPT). The authors would like to gratefully acknowledge the assistance of the Arizona Department of Economic Security in study recruitment, and the support of the University of Arizona Collaboratory for Metabolic Disease Prevention and Treatment.

The ADAPT study was supported by the US National Heart, Lung, and Blood Institute (HL117995).

References

1. Ogden CL, Carroll MD, Fryar CD, Flegal KM. Prevalence of Obesity Among Adults and Youth: United States, 2011-2014. NCHS Data Brief. 2015 Nov;(219):1-8. [PubMed]
2. Bhupathiraju SN, Hu FB. Epidemiology of Obesity and Diabetes and Their Cardiovascular Complications. Circ Res. 2016 May 27;118(11):1723-35.[CrossRef] [PubMed]
3. Cheon BK, Hong YY. Mere experience of low subjective socioeconomic status stimulates appetite and food intake. Proc Natl Acad Sci U S A. 2017 Jan 3;114(1):72-77. [CrossRef] [PubMed]
4. Baik K, Hosseini M, Priesmeyer HR. Correlates of psychological distress in involuntary job loss. Psychol Rep. 1989 Dec;65(3 Pt 2):1227-33. [CrossRef] [PubMed]
5. Berchick ER, Gallo WT, Maralani V, Kasl SV. Inequality and the association between involuntary job loss and depressive symptoms. Soc Sci Med. 2012 Nov;75(10):1891-4. [CrossRef] [PubMed]
6. Akil L, Ahmad HA. Effects of socioeconomic factors on obesity rates in four southern states and Colorado. Ethn Dis. 2011 Winter;21(1):58-62. [PubMed]
7. Latif E. The impact of macroeconomic conditions on obesity in Canada. Health Econ. 2014 Jun;23(6):751-9. [CrossRef] [PubMed]
8. Schröder M. Jobless now, sick later? Investigating the long-term consequences of involuntary job loss on health. Adv Life Course Res. 2013 Mar;18(1):5-15. [CrossRef] [PubMed]
9. Parvaneh K, Poh, B., Hajifaraji, M., & Ismail, M. Less Sleep Duration and Poor Sleep Quality Lead to Obesity. Proceedings of the Nutrition Society 2016;75(OCE1).
10. Spiegel K, Leproult R, L'hermite-Balériaux M, Copinschi G, Penev PD, Van Cauter E. Leptin levels are dependent on sleep duration: relationships with sympathovagal balance, carbohydrate regulation, cortisol, and thyrotropin. J Clin Endocrinol Metab. 2004 Nov;89(11):5762-71. [CrossRef] [PubMed]
11. Van Cauter E, Leproult R, Plat L. Age-related changes in slow wave sleep and REM sleep and relationship with growth hormone and cortisol levels in healthy men. JAMA. 2000 Aug 16;284(7):861-8. [CrossRef] [PubMed]
12. Haynes PL, Silva GE, Howe GW, et al. Longitudinal assessment of daily activity patterns on weight change after involuntary job loss: the ADAPT study protocol. BMC Public Health. 2017 Oct 10;17(1):793. [CrossRef] [PubMed]
13. Vasquez MM, Goodwin JL, Drescher AA, Smith TW, Quan SF. Associations of dietary intake and physical activity with sleep disordered breathing in the Apnea Positive Pressure Long-Term Efficacy Study (APPLES). J Clin Sleep Med. 2008 Oct 15;4(5):411-8. [PubMed]
14. Rock CL, Thornquist MD, Kristal AR, et al. Demographic, dietary and lifestyle factors differentially explain variability in serum carotenoids and fat-soluble vitamins: baseline results from the sentinel site of the Olestra Post-Marketing Surveillance Study. J Nutr. 1999 Apr;129(4):855-64. [CrossRef] [PubMed]
15. Subar AF, Thompson FE, Potischman N, Forsyth BH, Buday R, Richards D, McNutt S, Hull SG, Guenther PM, Schatzkin A, Baranowski T. Formative research of a quick list for an automated self-administered 24-hour dietary recall. J Am Diet Assoc. 2007 Jun;107(6):1002-7. [CrossRef] [PubMed]
16. Feskanich D, Sielaff BH, Chong K, Buzzard IM. Computerized collection and analysis of dietary intake information. Comput Methods Programs Biomed. 1989 Sep;30(1):47-57. [CrossRef] [PubMed]
17. Martin CK, Han H, Coulon SM, Allen HR, Champagne CM, Anton SD. A novel method to remotely measure food intake of free-living individuals in real time: the remote food photography method. Br J Nutr. 2009 Feb;101(3):446-56. [CrossRef] [PubMed]
18. U.S. Department of Agriculture How the HEI Is Scored. 03/16/2018. Available at https://www.fns.usda.gov/how-hei-scored (accessed 11-11-21).
19. Buysse DJ, Ancoli-Israel S, Edinger JD, Lichstein KL, Morin CM. Recommendations for a standard research assessment of insomnia. Sleep. 2006 Sep;29(9):1155-73. [CrossRef] [PubMed]Erratum in: Sleep. 2006 Nov 1;29(11):1380.
20. U.S. Department of Agriculture and U.S. Department of Health and Human Services. Dietary Guidelines for Americans, 2020-2025. 9th Edition. December 2020. Available at https://dietaryguidelines.gov/ (accessed 11-11-21).
21. Department of Health and Human Services and U.S. Department of Agriculture. 2015–2020 Dietary Guidelines for Americans. 2015–2020 Dietary Guidelines for Americans. 8th Edition. December 2015. Available at https://health.gov/our-work/food-nutrition/previous-dietary-guidelines/2015 (accessed 11-11-21).
22. Dashti HS, Scheer FA, Jacques PF, Lamon-Fava S, Ordovás JM. Short sleep duration and dietary intake: epidemiologic evidence, mechanisms, and health implications. Adv Nutr. 2015 Nov 13;6(6):648-59. [CrossRef] [PubMed]
23. Kye S, Kwon SO, Lee SY, Lee J, Kim BH, Suh HJ, Moon HK. Under-reporting of Energy Intake from 24-hour Dietary Recalls in the Korean National Health and Nutrition Examination Survey. Osong Public Health Res Perspect. 2014 Apr;5(2):85-91. [CrossRef] [PubMed]
24. Bothwell EK, Ayala GX, Conway TL, Rock CL, Gallo LC, Elder JP. Underreporting of food intake among Mexican/Mexican-American Women: rates and correlates. J Am Diet Assoc. 2009 Apr;109(4):624-32. [CrossRef] [PubMed]
25. St-Onge MP, Mikic A, Pietrolungo CE. Effects of Diet on Sleep Quality. Adv Nutr. 2016 Sep 15;7(5):938-49. [CrossRef] [PubMed]
26. Ding C, Lim LL, Xu L, Kong APS. Sleep and Obesity. J Obes Metab Syndr. 2018 Mar 30;27(1):4-24. [CrossRef] [PubMed]
27. Landry GJ, Best JR, Liu-Ambrose T. Measuring sleep quality in older adults: a comparison using subjective and objective methods. Front Aging Neurosci. 2015 Sep 7;7:166. [CrossRef] [PubMed]
28. Van Dongen HP, Maislin G, Mullington JM, Dinges DF. The cumulative cost of additional wakefulness: dose-response effects on neurobehavioral functions and sleep physiology from chronic sleep restriction and total sleep deprivation. Sleep. 2003 Mar 15;26(2):117-26. [CrossRef] [PubMed] Erratum in: Sleep. 2004 Jun 15;27(4):600.
29. Zavecz Z, Nagy T, Galkó A, Nemeth D, Janacsek K. The relationship between subjective sleep quality and cognitive performance in healthy young adults: Evidence from three empirical studies. Sci Rep. 2020 Mar 17;10(1):4855. [CrossRef] [PubMed]
30. Rehm CD, Monsivais P, Drewnowski A. Relation between diet cost and Healthy Eating Index 2010 scores among adults in the United States 2007-2010. Prev Med. 2015 Apr;73:70-5. [CrossRef] [PubMed]

Cite as: Batool-Anwar S, Mayer C, Haynes PL, Liu Y, Thomson CA, Quan SF. Impact of Recent Job Loss on Sleep, Energy Consumption and Diet. Southwest J Pulm Crit Care. 2021;23(5):129-37. doi: https://doi.org/10.13175/swjpcc045-21 PDF

The Impact of Sleep-Disordered Breathing on Body Mass Index (BMI): The Sleep Heart Health Study (SHHS)

Mark A. Brown, M.D. ¹

James L. Goodwin, Ph.D.²

Graciela E. Silva, Ph.D, MPH.³

Ajay Behari, M.D.⁴

Anne B. Newman, M.D., M.P.H^5,6

Naresh M. Punjabi, M.D., Ph.D.⁷

Helaine E. Resnick, Ph.D., M.P.H.⁸

John A. Robbins, M.D., M.S.H.⁹

Stuart F. Quan, M.D.^2,10

¹Department of Psychiatry, Kaiser Permanente, Portland, OR (markbrownmd@gmail.com); 

²Sleep and Arizona Respiratory Centers, University of Arizona College of Medicine, Tucson, AZ(jamieg@arc.arizona.edu);

³College of Nursing & Health Innovation, Arizona State University, Tempe, AZ (Graciela.Silva@asu.edu); 

⁴Pulmonary and Critical Care Associates of Baltimore, Baltimore, MD (ajaybehari@yahoo.com);

⁵Graduate School of Public Health, Department of Epidemiology, University of Pittsburgh, Pittsburgh, PA

⁶Division of Geriatric Medicine, University of Pittsburgh School of Medicine, Pittsburgh, PA (NewmanA@edc.pitt.edu);

⁷Department of Medicine, Johns Hopkins University School of Medicine, Baltimore, MD (npunjabi@jhmi.edu);

⁸American Association of Homes and Services for the Aging, Washington, DC (heresnick@gmail.com); 

⁹Center for HealthCare Policy and Research, University of California, Davis, Sacramento, CA (jarobbins@ucdavis.edu);

¹⁰Division of Sleep Medicine, Harvard Medical School, Boston, MA (squan@arc.arizona.edu)

Address for correspondence and reprint requests: Stuart F. Quan, M.D., Division of Sleep Medicine, Harvard Medical School, 401 Park Dr., 2^nd Floor East, Boston, MA 02215, Tel (617) 998-8842, Fax (617) 998-8823, Email: squan@arc.arizona.edu

Conflict of Interest Statement: None of the authors have conflicts of interest pertinent to the subject matter of this manuscript. 

Reference as: Brown MA, Goodwin JL, Silva GE, Behari A, Newman AB, Punjabi NM, Resnick HE, Robbins JA, Quan SF. The impact of sleep-disordered breathing on body mass index (BMI): the sleep heart health study (SHHS). Southwest J Pulm Crit Care 2011;3:159-68. (Click here for PDF version of the manuscript)

Abstract

Introduction: It is well known that obesity is a risk factor for sleep-disordered breathing (SDB). However, whether SDB predicts increase in BMI is not well defined. Data from the Sleep Heart Health Study (SHHS) were analyzed to determine whether SDB predicts longitudinal increase in BMI, adjusted for confounding factors.

Methods: A full-montage unattended home polysomnogram (PSG) and body anthropometric measurements were obtained approximately five years apart in 3001 participants. Apnea-hypopnea index (AHI) was categorized using clinical thresholds: < 5 (normal), ≥ 5 to <15 (mild sleep apnea), and ³ 15 (moderate to severe sleep apnea). Linear regression was used to examine the association between the three AHI groups and increased BMI. The model included age, gender, race, baseline BMI, and change in AHI as covariates.

Results: Mean (SD) age was 62.2 years (10.14), 55.2% were female and 76.1% were Caucasian. Five-year increase in BMI was modest with a mean (SD) change of 0.53 (2.62) kg/m² (p=0.071). A multivariate regression model showed that subjects with a baseline AHI between 5-15 had a mean increase in BMI of 0.22 kg/m² (p=0.055) and those with baseline AHI ≥ 15 had a BMI increase of 0.51 kg/m² (p<0.001) compared to those with baseline AHI of <5.

Conclusion: Our findings suggest that there is a positive association between severity of SDB and subsequent increased BMI over approximately 5 years. This observation may help explain why persons with SDB have difficulty losing weight.

Key Words: Sleep Apnea, Weight Gain, Obesity

Abbreviation List: PSG-polysomnogram, SDB-sleep disordered breathing, AHI-apnea hypopnea index, SHHS-Sleep Heart Health Study, BMI-body mass index, SD-standard deviation, SEM-standard error of the mean, ANOVA-analysis of variance

Introduction

There is overwhelming epidemiological and clinical data indicating that obesity is a risk factor for sleep disordered breathing (SDB).^1-8 The association between obesity and SDB is substantial, with high body mass index (BMI) contributing to moderate to severe SDB in 58% of affected persons.⁹ The effect of obesity is greater in men than women^1,10-12 although it decreases with increasing age.^6,7 In addition, weight loss has been demonstrated to decrease the severity of SDB.^10,13,14 Longitudinal data from population studies including the Sleep Heart Health Study (SHHS),¹⁰ the Wisconsin Sleep Cohort,¹⁵ and the Cleveland Family Study⁶ have initially focused on the impact of increased weight on SDB severity. However, examination of the opposite causal pathway has yet to be prospectively addressed.

Anecdotally, patients with SDB appear to have more difficulty losing weight than obese patients without SDB. They also report marked weight gain prior to confirmation of their diagnosis. Two small studies support these empiric observations.^5,16  Given this limited information on the impact of SDB on BMI, data from SHHS was analyzed to examine the impact of SDB on BMI after controlling for change in AHI and severity of SDB.

Methods

Study Design and Population. The SHHS is a multi-center, community-based prospective cohort study of the natural history and cardiovascular consequences of SDB. Details of the study design, sampling, and procedures have been reported.¹⁷ Briefly, between November 1995 and January 1998 participants were recruited from several ongoing prospective cohort studies--the Framingham Offspring and Omni Studies, the Atherosclerosis Risk in Communities Study, the Cardiovascular Health Study, the Strong Heart Study, and the cohort studies of respiratory disease in Tucson and of hypertension in New York. Participants were eligible if they were ≥ 40 years of age and were not being treated for sleep apnea with positive pressure therapy, an oral appliance, oxygen, or a tracheostomy. Habitual snorers < 65 years were over sampled to increase the prevalence of obstructive sleep apnea. Subjects were required to provide written consent and the protocol was approved by the institutional review boards of each of the eight investigative sites. 

Data Collection. A total of 6,441 subjects completed the baseline polysomnogram (PSG), and 4,586 consented to have a second evaluation approximately five years later. This analysis focuses on the 3,040 participants who had PSG and BMI data at both time points. Data from all 215 participants who had a follow-up PSG from the New York center were excluded because they did not meet quality standards for the follow-up examination. The remaining participants died, were too ill to participate, refused to participate, were lost to follow-up or had incomplete covariate data such as weight. This latter group had a higher percentage of Whites (85%) compared with the study group (75.5%) (p-value <0.001). There also were statistically significant differences in baseline BMI, baseline AHI, and age between the study group compared with the excluded group, however, these differences were very small and were not clinically significant. There was no gender difference between the two groups.

Weight was measured on the night of the PSG examination with the participant in light clothes on a calibrated portable scale. Height was obtained at the baseline home visit if not already measured within + 3 months of the parent study. BMI was calculated as weight in kilograms divided by the square of height in meters. Baseline height was used for baseline and follow-up BMI calculations. Age, sex, and ethnicity were self-reported. 

The PSG was conducted using a portable monitor (PS-2 System; Compumedics Limited, Abbotsford, Victoria, Australia), using methods previously described.¹⁸ Apnea was present if there was an absence or near absence of airflow or thoracoabdominal movement (at least < 25% of baseline) for > 10 seconds. Hypopnea was defined as a decrease in the amplitude of the airflow or thoracoabdominal movement below 70% of baseline for > 10 seconds. The apnea-hypopnea index (AHI) was calculated as the number of apnea and hypopnea events, each associated with at least a 4% decrease in oxygen saturation, divided by total sleep time in hours.

Results

Participant characteristics are provided in Table 1. 

Table 1: Characteristics of participants of the Sleep Heart Health Study cohort with complete baseline and follow-up polysomnography and weight measurements as a function of sleep apnea severity.

As expected, women were over-represented in the baseline AHI < 5 group (64.9%) and men were over-represented in the AHI ³ 15 group (63.5%) (p<0.001). Baseline BMI increased as baseline AHI severity increased.  Overall unadjusted five-year increase in BMI was modest with a mean (SD) BMI change of 0.53 (2.61) kg/m². The unadjusted five-year increase in BMI was 0.63 (2.54) kg/m² for those with baseline AHI < 5, 0.43 (2.48) kg/m² among those with AHI ≥ 5 to < 15 and 0.37 (3.05) kg/m² for the AHI group ≥ 15. These values were not statistically different from each other.

A multivariate regression model was constructed predicting five-year change in BMI by baseline AHI category adjusted for age, gender, race, baseline BMI, and AHI change (Table 2).

Table 2: Adjusted β coefficients of BMI change according to AHI and continuous variables in the Sleep Heart Health Study*.

Compared to baseline AHI group of < 5, those with AHI between ≥ 5 to < 15 had a mean adjusted increase in BMI of 0.21 that approached statistical significance (p=0.055). However, those with AHI ≥ 15 had a statistically significant adjusted BMI increase of 0.51 (p<0.001).  Younger age, lower baseline BMI and greater AHI change also were associated with a larger BMI increase. There was a trend for women to have a greater increase in BMI, but no effect of race was observed. However, the model only accounted for 7% of the total variance. Adjusted means by baseline AHI group are displayed graphically in Figure 1.

Figure 1: Estimated Adjusted Means of BMI increase according to AHI in the Sleep Heart Health Study. Data are adjusted for baseline age (continuous), race (categorical), gender (categorical), baseline BMI (continuous), change in AHI (continuous). Covariates fixed at: baseline BMI = 28.7, baseline age 62.1, change in RDI = 2.7. Bars represent 95% confidence intervals.

Discussion

Our findings indicate that there is a positive association between severity of SDB and five-year increase in BMI. The finding was demonstrated after controlling for key covariates including age, gender, race, baseline BMI, and AHI change. This observation may help explain the difficulty patients with SDB have in trying to lose weight.

Two previous small studies have demonstrated a positive association between newly diagnosed SDB and weight gain. A retrospective study by Phillips et al. compared one-year weight histories of 53 men and women patients who were recently diagnosed with SDB with 24 control subjects matched for gender, age, BMI and percent body fat.⁵ Subjects in that study were somewhat younger than the SHHS cohort with an age difference of approximately 10 years. The SDB among subjects in the previous study tended to be moderate to severe with mean ± SEM AHI 33 ± 5 /h for men and 37 ± 10 /h for women. Mean ± SEM of BMI at time of diagnosis was somewhat higher than in the SHHS with 35 ± 1 kg/m² for men and 44 ± 2 kg/m² for women.  Men and women patients with SDB had reported a recent weight gain of 7.4 ± 1.5 kg compared with a weight loss of 0.5 ± 1.7 kg (p=0.001) in obese controls without SDB. However, given the design of this study it is not possible to determine whether weight gain contributed to the onset of SDB or was a result of SDB. The study was also limited by reliance on self-report of weight gain history.

Another study by Traviss et al. prospectively evaluated 49 obese patients with newly diagnosed SDB.¹⁶ Mean ± SD of AHI at diagnosis was severe at 45 ± 27 /h.  BMI at diagnosis was elevated at 36.5 ± 6.2 kg/m². Of the 49 subjects, 43 could estimate the duration of their symptoms with 84% reporting weight gain since becoming symptomatic.  Weight gain was relatively large, with a reported 17 ± 15 kg over 5.3 ± 4.8 years. However, this study was limited by the lack of a control group and reliance on self-report of weight history.  These two small studies, in addition to our findings, suggest that there is an association between SDB and increased BMI.

Interestingly, unadjusted BMI change in our study was quite modest and not statistically different as a function of SDB severity. However, BMI change over time is a complex phenomenon influenced by several variables. A large (29,799 subjects) prospective study examining 5-year change in weight in a multi-ethnic cohort of men and women explored several of these relationships.¹⁹ In that study, younger men and to a greater degree, younger women were at greater risk for weight gain compared to older adults. This is consistent with our initial findings. In addition, there was a trend for women in the higher baseline BMI categories of ‘overweight’ (BMI >25– 30 kg/m²) and ‘obese’ (BMI >30 kg/m²) in the aforementioned cohort to gain more weight than men in the higher baseline BMI categories. In order to more precisely examine the effect of AHI on weight change, we controlled for these confounders in our final multivariate model thus resulting in the finding of an increase in BMI as a function of SDB severity in this study. 

Several mechanisms could explain why SDB contributes to increased BMI. First, persons with SDB may have a reduction in the quantity and quality of their sleep. Recent data indicate that insufficient sleep may be a risk factor for obesity.²⁰ Experimental sleep restriction increases ghrelin and reduces leptin production favoring appetite enhancement,²¹ a finding that also has been observed in a large population cohort.²² Second, those with SDB may eat a diet that favors weight gain. In support of this hypothesis, sleep restriction has been shown to increase craving for calorie dense food with high carbohydrate content. The Apnea Positive Pressure Long-Term Efficacy Study (APPLES) demonstrated that those with severe SDB consumed a diet higher in cholesterol, protein, total fat and total saturated fatty acids, even after adjusting for BMI, age, and daytime sleepiness.²³ Third, a cardinal symptom of SDB is excessive daytime sleepiness. Thus, it is possible that persons with SDB engage in less physical activity because they are too fatigued to exercise. Data from APPLES indicate that recreational physical activity is less in those with SDB. However, this finding appears to be principally explained by concomitant obesity. 

Weight loss frequently results in an improvement and sometimes resolution in SDB. This is most evident in those who undergo bariatric surgical procedures.^24,25 Persons with SDB are frequently counseled to treat their SDB by losing weight through diet and exercise,²⁶ an approach that is usually unsuccessful.²⁵ Failure to primarily address SDB in conjunction with a weight reduction program may diminish the latter’s success. However, evidence to date indicates that treatment of SDB does not consistently result in weight loss. In a sample of clinical patients with SDB, treatment with CPAP did not result in weight loss. Moreover, in female patients, there was actually an increase in weight.²⁷ In addition, consistent weight reduction was not observed in a small number of patients with severe OSA who underwent tracheostomy.²⁸ Thus, it appears that weight gain engendered by the presence of OSA is not easily reversed despite therapy. Prospective studies will be required to determine whether primary treatment for OSA enhances weight loss programs in those with OSA.

Although this analysis demonstrated a positive association of severity of SDB on five-year increase in BMI, there are several caveats that deserve consideration. The BMI of participants tended to be lower than that seen in clinical SDB populations and a relatively small number of subjects had large changes in BMI. As previously noted, the mean BMI increase was, at best, quite modest. When converted for illustrative purposes to weight using an average height of 167 cm of the participants, those with an AHI between ≥ 5 to < 15 had a mean adjusted increase in BMI of 0.21 kg/m² equal to 0.59 kg or 1.30 lbs. Similarly, those with AHI ≥ 15 had an adjusted BMI increase of 0.51 kg/m² equal to 1.42 kg or 3.13 lbs. Thus, the magnitude of the changes we observed may not be applicable to clinical populations where patients with SDB may have a higher BMI. In addition, it is not known when the participants developed SDB, thus definitive inference of causality cannot be made. However, following a large undiagnosed cohort over an extended period of time to determine incidence of SDB onset and subsequent change in weight would be exceedingly difficult and costly.  Additionally, the model only accounted for a small amount of the total variance in five-year BMI increase, suggesting that there are likely other unmeasured variables influencing the amount of BMI increase over time in this cohort. Finally, while not statistically significant, the unadjusted mean change in BMI was slightly less in the high RDI group in comparison to the lower RDI groups. This observation underscores the biological complexity of the interactions among weight change, SDB, age, gender and other factors.

In conclusion, our findings suggest that although weight gain is a risk factor for developing or worsening SDB, SDB may, in a reciprocal fashion, lead to increased weight gain. This may help explain why patients with SDB find it difficult to lose weight.

Acknowledgements

This work was supported by National Heart, Lung and Blood Institute cooperative agreements U01HL53940 (University of Washington), U01HL53941 (Boston University), U01HL53938 (University of Arizona), U01HL53916 (University of California, Davis), U01HL53934 (University of Minnesota), U01HL53931 (New York University), U01HL53937 and U01HL64360  (Johns Hopkins University), U01HL63463 (Case Western Reserve University), and U01HL63429 (Missouri Breaks Research).

Sleep Heart Health Study (SHHS) acknowledges the Atherosclerosis Risk in Communities Study (ARIC), the Cardiovascular Health Study (CHS), the Framingham Heart Study (FHS), the Cornell/Mt. Sinai Worksite and Hypertension Studies, the Strong Heart Study (SHS), the Tucson Epidemiologic Study of Airways Obstructive Diseases (TES) and the Tucson Health and Environment Study (H&E) for allowing their cohort members to be part of the SHHS and for permitting data acquired by them to be used in the study.  SHHS is particularly grateful to the members of these cohorts who agreed to participate in SHHS as well. SHHS further recognizes all of the investigators and staff who have contributed to its success. A list of SHHS investigators, staff and their participating institutions is available on the SHHS website, www.jhucct.com/shhs.

The opinions expressed in the paper are those of the author(s) and do not necessarily reflect the views of the Indian Health Service.

These data have been presented in part at the Annual Meeting of the Associated Professional Sleep Societies, June 11, 2009, Seattle, WA.

References

1. Bixler EO, Vgontzas AN, Lin HM, Ten Have T, Rein J, Vela-Bueno A, et al. Prevalence of sleep-disordered breathing in women: Effects of gender. Am J Respir Crit Care Med 2001;163:608-13.
2. Bixler EO, Vgontzas AN, Ten Have T, Tyson K, Kales A. Effects of age on sleep apnea in men: I. prevalence and severity. Am J Respir Crit Care Med 1998;157:144-8.
3. Carmelli D, Swan GE, Bliwise DL. Relationship of 30-year changes in obesity to sleep-disordered breathing in the western collaborative group study. Obes Res 2000;8:632-7.
4. Duran J, Esnaola S, Rubio R, Iztueta A. Obstructive sleep apnea-hypopnea and related clinical features in a population-based sample of subjects aged 30 to 70 yr. Am J Respir Crit Care Med 2001;163:685-9.
5. Phillips BG, Hisel TM, Kato M, Pesek CA, Dyken ME, Narkiewicz K, et al. Recent weight gain in patients with newly diagnosed obstructive sleep apnea. J Hypertens 1999;17:1297-300.
6. Tishler PV, Larkin EK, Schluchter MD, Redline S. Incidence of sleep-disordered breathing in an urban adult population: The relative importance of risk factors in the development of sleep-disordered breathing. JAMA 2003;289:2230-7.
7. Young T, Shahar E, Nieto FJ, Redline S, Newman AB, Gottlieb DJ, et al. Predictors of sleep-disordered breathing in community-dwelling adults: The sleep heart health study. Arch Intern Med 2002;162:893-900.
8. Young T, Palta M, Dempsey J, Skatrud J, Weber S, Badr S. The occurrence of sleep-disordered breathing among middle-aged adults. N Engl J Med 1993;328:1230-5.
9. Young T, Peppard PE, Taheri S. Excess weight and sleep-disordered breathing. J Appl Physiol 2005;99:1592-9.
10. Newman AB, Foster G, Givelber R, Nieto FJ, Redline S, Young T. Progression and regression of sleep-disordered breathing with changes in weight: The sleep heart health study. Arch Intern Med 2005;165:2408-13.
11. Ip MS, Lam B, Tang LC, Lauder IJ, Ip TY, Lam WK. A community study of sleep-disordered breathing in middle-aged chinese women in hong kong: Prevalence and gender differences. Chest 2004;125:127-34.
12. Millman RP, Carlisle CC, McGarvey ST, Eveloff SE, Levinson PD. Body fat distribution and sleep apnea severity in women. Chest 1995;107:362-6.
13. Barvaux VA, Aubert G, Rodenstein DO. Weight loss as a treatment for obstructive sleep apnoea. Sleep Med Rev 2000;4:435-52.
14. Grunstein RR, Stenlof K, Hedner JA, Peltonen M, Karason K, Sjostrom L. Two year reduction in sleep apnea symptoms and associated diabetes incidence after weight loss in severe obesity. Sleep 2007;30:703-10.
15. Peppard PE, Young T, Palta M, Dempsey J, Skatrud J. Longitudinal study of moderate weight change and sleep-disordered breathing. JAMA 2000;284:3015-21.
16. Traviss KA, Barr SI, Fleming JA, Ryan CF. Lifestyle-related weight gain in obese men with newly diagnosed obstructive sleep apnea. J Am Diet Assoc 2002;102:703-6.
17. Quan SF, Howard BV, Iber C, Kiley JP, Nieto FJ, O'Connor GT, et al. The sleep heart health study: Design, rationale, and methods. Sleep 1997;20:1077-85.
18. Redline S, Sanders MH, Lind BK, Quan SF, Iber C, Gottlieb DJ, et al. Methods for obtaining and analyzing unattended polysomnography data for a multicenter study. sleep heart health research group. Sleep1998;21:759-67.
19. Ball K, Crawford D, Ireland P, Hodge A. Patterns and demographic predictors of 5-year weight change in a multi-ethnic cohort of men and women in australia. Public Health Nutr 2003;6:269-81.
20. Chaput JP, Despres JP, Bouchard C, Tremblay A. The association between sleep duration and weight gain in adults: A 6-year prospective study from the Quebec Family Study. Sleep 2008;31:517-23.
21. Spiegel K, Tasali E, Penev P, Van Cauter E. Brief communication: Sleep curtailment in healthy young men is associated with decreased leptin levels, elevated ghrelin levels, and increased hunger and appetite. Ann Intern Med 2004;141:846-50.
22. Taheri S, Lin L, Austin D, Young T, Mignot E. Short sleep duration is associated with reduced leptin, elevated ghrelin, and increased body mass index. PLoS Med. 2004;1:e62.
23. Vasquez MM, Goodwin JL, Drescher AA, Smith TW, Quan SF. Associations of dietary intake and physical activity with sleep disordered breathing in the apnea positive pressure long-term efficacy study (APPLES). J Clin Sleep Med 2008;4:411-8.
24. Rasheid S, Banasiak M, Gallagher SF, Lipska A, Kaba S, Ventimiglia D, et al. Gastric bypass is an effective treatment for obstructive sleep apnea in patients with clinically significant obesity. Obes Surg 2003;13:58-61.
25. Fritscher LG, Canani S, Mottin CC, Fritscher CC, Berleze D, Chapman K, et al. Bariatric surgery in the treatment of obstructive sleep apnea in morbidly obese patients. Respiration 2007;74:647-52.
26. Veasey SC, Guilleminault C, Strohl KP, Sanders MH, Ballard RD, Magalang UJ. Medical therapy for obstructive sleep apnea: A review by the medical therapy for obstructive sleep apnea task force of the standards of practice committee of the american academy of sleep medicine. Sleep 2006;29:1036-44.
27. Redenius R, Murphy C, O'Neill E, Al-Hamwi M, Zallek SN. Does CPAP lead to change in BMI? J Clin Sleep Med 2008;4:205-9.
28. Haapaniemi JJ, Laurikainen EA, Halme P, Antila J. Long-term results of tracheostomy for severe obstructive sleep apnea syndrome. ORL J Otorhinolaryngol Relat Spec 2001;63:131-6.