Article Text

Download PDFPDF

To nap or not to nap: more questions than answers
  1. Yue Leng1,
  2. Kristine Yaffe2
  1. 1 Department of Psychiatry, University of California San Francisco, San Francisco, California, USA
  2. 2 Department of Psychiatry, Neurology and Epidemiology, University of California San Francisco, San Francisco, California, United States
  1. Correspondence to Dr Yue Leng, University of California San Francisco, San Francisco, CA 94121, USA; yue.leng{at}

Statistics from

We spend a third of our lives in sleep. One of the most common yet understudied sleep behaviours in human beings is daytime napping. While napping is traditionally viewed as a countermeasure to sleepiness and as a strategy to boost performance, especially in healthy younger adults or among shift workers,1 the effects of napping in middle-aged to older-aged populations are largely unknown.

One key question is whether naps are protective or harmful for heart health among older adults. A pioneering Greek case–control study from the late 1980s2 compared 97 men with an acute episode of coronary heart disease (CHD) and 90 control subjects, and showed a 30% reduction in the incidence of CHD associated with a 30 min afternoon nap. Over the past decade, growing epidemiological evidence has pointed to napping as a risk factor for cardiovascular disease (CVD), and a recent meta-analysis3 summarised 11 prospective studies and concluded that there was a J-shaped dose–response relationship, in which the risk of CVD decreased with increasing nap duration until it reached 30 min/day, but started to increase with longer naps after the 30 min/day threshold. While this meta-analysis suggests that napping duration could be an important factor to consider in studies of naps, most previous studies have only compared nappers and non-nappers, or have considered only the duration but not the frequency of napping. Furthermore, the majority of studies only examined CVD mortality, whereas non-fatal CVD events are much less studied in relation to napping. Many studies have failed to fully address the influence of potential confounders or the independent effects of napping.

In their Heart paper, Hausler et al 4 examine the longitudinal association between self-reported napping frequency and average napping duration and incidence risk of both fatal and non-fatal CVD among 3462 men and women (mean age 57.1) with no history of CVD in the CoLaus study, a Swiss population-based prospective cohort. Over a median follow-up period of 5.3 years, 155 fatal and non-fatal CVD events developed. The authors found a 48% lower CVD risk (HR=0.52, 95% CI 0.28 to 0.95) for those who napped for one to two times per week compared with the non-nappers, both in the unadjusted and adjusted models (for age, sex, education, smoking, sedentary behaviour, body mass index, night-time sleep duration, and other primary CVD risk factors such as hypertension, diabetes and dyslipidaemia). However, the increased unadjusted HR (1.67, 95% CI 1.10 to 2.55) associated with a napping frequency of six to seven times per week was attenuated after adjustment for the above covariates (HR=0.89, 95% CI 0.58 to 1.38). No association was found between napping duration and incident CVD. While it remains premature to conclude on the appropriateness of napping for maintaining optimal heart health, the findings by Hausler et al 4 offer some reassurance that the answer is probably more than a simple ‘yes’ or ‘no’, and that we have much more to learn about napping.

The biggest challenge in epidemiological studies of napping is how we define and measure naps. Are they planned or unplanned? What is the purpose of the naps? Are they taken occasionally when needed or habitually as a cultural practice? Are they taken to compensate for insufficient or poor night-time sleep, or do they indicate underlying ill health? Is night-time sleep quality taken into account? What is the timing, duration and frequency of the naps? Do we count in a 5 min ‘dozing-off’ as a nap? What is the best way to measure naps? Until we get to the answers to some of these questions, the implications of napping cannot be fully addressed.

As with many previous studies, Hausler et al 4 assessed napping by asking the participants to self-report the frequency and duration of napping during the previous week. The item ‘Sieste ou repos au lit l’après-midi’ (Nap or bed rest in the afternoon) was used to define naps, and participants were considered as nappers if they reported to have napped at least once during the previous week. The self-reported napping likely only reflects intentional or planned naps and could lead to an underestimation of total napping or has different implications to objectively measured napping.5 Furthermore, while the average daily napping duration was derived based on reported napping duration each time and the frequency over the week, it is unclear if most of these ‘infrequent’ naps happened to be long naps, and if the effects are different for ‘frequent short naps’ versus ‘infrequent long naps’. To date, there is no ‘gold standard’ for measuring naps. While polysomnography (PSG) is the agreed method for evaluating night-time sleep duration and fragmentation and to differentiate sleep staging, it is impractical for the measure of daytime naps in population-based studies. Actigraphy is a widely used means to evaluate 24-hour sleep–wake activities and has been validated against PSG for measuring daytime naps in healthy young adults, with high sensitivity and reliability for predicting total sleep time, sleep efficiency and sleep latency.6 However, the measure is limited in its capability in differentiating naps from inactive wakefulness, especially in sedentary elderly populations. There is an urgent need for the development of a valid and feasible measure that best defines napping before meaningful conclusions about related health risks can be made.

Age is another important consideration in determining the effects of napping on CVD. In their sensitivity analysis, Hausler et al 4 found attenuation of the effects (HR=0.82, 0.37–1.83) for those who napped for one to two times per week after restricting the sample to the older adults (65 years and older). While the authors speculated that the decreased sample size has led to reduced statistical power, this is unlikely to be the main explanation given that effect size was also reduced. Older adults might take naps for different reasons than younger ones. Notably, previous findings on the association between napping and increased CVD risk were mostly found among older adults,3 and it is often hypothesised that napping in the elderly could be indicative of underlying health problems. Thus, the benefits of occasional napping might become less evident in this population who are already at high risk for CVD. Furthermore, older adults might take longer naps, given that they have the time and flexibility, and it might be the length rather than the frequency of the naps that is driving the increased CVD risk. Unintentional naps are also common in older adults and are usually under-reported. Better understanding of the nature of the naps taken by the elderly will help to inform the association between napping and CVD risk in this population.

It is increasingly recognised that daytime napping should not be viewed separately from night-time sleep.7 8 Recent findings in 116 632 participants aged 35–70 years (with 52% ≥50 years) from the multiregional Prospective Urban Rural Epidemiology study suggested an increased risk of major CVD and all-cause mortality associated with increasing daytime nap duration (0–1 hour and >1 hour vs no nap) only among individuals with at least 6 hours of sleep at night but not in those with shorter night-time sleep duration.7 This is in line with another study which also found an interaction between daytime napping and night-time sleep, with the association between longer napping and increased risk of cognitive impairment only significant among those with 6–8 hours of night-time sleep duration or those with a sleep efficiency of over 70%.8 This evidence suggested that napping might be excessive and particularly problematic among those who are already getting good quality sleep at night but could play a more compensatory role in those with insufficient night-time sleep. Interestingly, Hausler et al 4 reported that the protective effects of infrequent napping remained in the subsample with at least 6 hours of night-time sleep. While the authors were unable to compare the association between napping and CVD among those with <6 hours vs ≥6 hours of night-time sleep due to insufficient number of people in the former group, the association stayed robust after further adjustment for PSG-measured obstructive sleep apnoea or excessive daytime sleepiness. Therefore, it is unlikely that infrequent napping helps to reduce CVD risk only through compensating for poor night-time sleep.

While the exact physiological pathways linking daytime napping to CVD risk is not clear, the study by Hausler et al 4 contributes to the ongoing debate on the health implications of napping and suggests that it might not only be the duration but also the frequency that matters. However, further investigation is needed to take into account the cultural context and population demographics, to examine daytime napping in conjunction with night-time sleep, to identify the role of potential confounders, mediators and moderators, and to include assessment of biomarkers in order to understand biological mechanisms. The study of napping is a challenging but also a promising field with potentially significant public health implications. While there remain to be more questions than answers, it is time to start unveiling the power of naps for a supercharged heart.


View Abstract


  • Contributors YL drafted the editorial. KY reviewed and edited the draft.

  • Funding YL is supported by the National Institute on Aging (NIA) (1K99AG056598), and received funding from GBHI, Alzheimer’s Association and Alzheimer’s Society (GBHI ALZ UK-19-591141). Support was provided by NIA grant K24AG031155, awarded to KY.

  • Competing interests None declared.

  • Patient consent for publication Not required.

  • Provenance and peer review Commissioned; internally peer reviewed.

Request Permissions

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

Linked Articles