Physiological Changes in Women’s Sleep Across their Lifespan

 Introduction

Sleep is unquestionably crucial for health and well-being. Disrupted sleep negatively impacts cognition, impairs memory consolidation, weakens the immune system and increases the risk of numerous diseases and indicators of health, such as obesity, cardiovascular disease, Alzheimer’s disease and psychiatric disorders. [1] Particularly, REM (rapid eye movement) sleep deprivation is associated with enhanced emotional reactivity, adverse health outcomes and higher mortality. [2]

However, there is a clear gender bias in reported sleep difficulties and disorders, with significant implications for women’s quality of life. Women usually report poorer sleep than men and have more sleep complaints. They are more likely to report longer sleep latency, more nocturnal awakenings, as well as more daytime sleepiness. Women are 41% more likely to experience insomnia than men, and this risk increases with age, and are also more than twice as likely to suffer from anxiety and depressive disorders, which are known causes of sleep disturbance. They also have twice the risk of experiencing restless legs syndrome (RLS). [1] Most patients with RLS also experience periodic limb movements of sleep (PLMS). The prevalence of PLMS increases with age and is also more prevalent in women. [3]

Therefore, to improve women’s health across their lifespan, it is crucial to understand if sleep objectively changes and how it changes during the different phases of a woman’s life. Additionally, more studies are needed to clearly identify the biological sources behind gender differences in sleep and, consequently, determine potential substrates accessible to clinical intervention. Many studies point out the role of ovarian hormones in mediating these gender differences in sleep and acutely regulating and influencing various parameters of women’s sleep. [4, 5, 6, 7]

REM sleep seems to be modulated during the menstrual cycle and premenstrual syndrome (PMS) may disrupt sleep severely. [8]

Pregnancy is also associated with sleep modulations. [8, 9]

Problems with sleep are also very common during menopause, particularly associated with vasomotor symptoms, and alleviating these symptoms improves perceived sleep quality. In premenopause, sleep-disordered breathing (SDB), including obstructive sleep apnea (OSA), is more common in men than women. However, women run the risk of being underdiagnosed for the disease, as they frequently present atypical symptoms, which may develop with a lower apnea-hypopnea index (AHI). [8]

In pregnancy, sleep disturbance may not be related to changes in ovarian hormones but rather to physiological factors, such as differences in sleeping position, increase in nocturnal micturition, gastroesophageal reflux or musculoskeletal discomfort, factors that possibly mask any effect of hormones. [2]

Nonetheless, estrogen, both endogenous and exogenous, as well as progesterone, are positively associated with sleep quality during the menopausal transition. Menopausal hormone therapy (MHT) improves sleep disturbances that characterize this period. [2]

However, ageing itself is related to a decrease in sleep quality and quantity for both men and women. [10] In fact, for women, both ageing and hormonal changes independently influence sleep architecture. Ageing is associated with decreased total sleep time (TST), more awakenings during the night and poorer sleep efficiency, regardless of menopausal state. [11]

Sleep disturbances increase with advancing age, affecting nearly half of the individuals with more than 65 years old, especially the female population, with a negative impact on women’s physical and emotional well-being. [2]

Therefore, the purpose of this investigation is to assess the effect of ageing on female’s sleep in our sample of Portuguese women. Understanding sleep changes across the lifespan is important to distinguish clinical sleep conditions or complaints as physiological or pathological and may ultimately lead to effective therapies that improve health and well-being for women, who were, until very recently, excluded from sleep studies.

Methods

This was an observational, analytical and retrospective study conducted in a differentiated Sleep Clinic. We included adult women of all ages who performed polysomnographic studies level I (in the sleep laboratory) and level II (in ambulatory) as well as level III with the artificial intelligence algorithm BodySleep® in the Clinic, between May 2022 and June 2024.

The following parameters were collected: age; BMI (Body Mass Index); TRT (Total Recording Time), the time from lights off to lights on; TST (Total Sleep Time), the time spent asleep while in bed; sleep efficiency, defined as the percentage of sleep during time spent in bed; WASO (Wakefulness After Sleep Onset), meaning the total number of minutes that a person is awake after having initially fallen asleep; sleep latency, the time from lights out to the first epoch of any stage of sleep; REM latency, the time between sleep onset and the first epoch of REM sleep; AHI (Apnea-Hypopnea Index), the combined average number of apneas and hypopneas that occur per hour of sleep; ODI (Oxygen Desaturation Index), the average number of desaturation episodes occurring per hour, where desaturation is defined as a decrease in the mean oxygen saturation of ≥3% for a minimum of 3 seconds; percentage of snore; PLMS (Periodic Limb Movements of Sleep) index; percentage of REM sleep; percentage of NREM sleep, the sum of the percentages of all stages of NREM – N1, N2 and N3; arousal index in TST, which indicates the number of times a person wakes up per hour of sleep; average pulse, maximum pulse and minimum pulse in TST.

Ethical considerations

The patients' anonymity and confidentiality were guaranteed. The study took place in accordance with ethical principles and good practice.

The research protocol was submitted and approved by the Ethics Committee for Research in Life and Health Sciences (CEICVS) of the Ethics Council of University of Minho.

Statistical analysis

The analyses were performed using IBM SPSS software (version 29) and the level of significance adopted was 0.05.

Normality of data was checked by Shapiro-Wilk tests. Quantitative data conforming to a normal distribution were represented as mean ± standard deviation. Non-normally distributed data were presented as medians (P25; P75). Count data were expressed in percentages (%). All results were presented using suitable statistics for qualitative and quantitative variables, using the Chi-square test and Student's t-test or Mann-Whitney U test, respectively.

The existence of a correlation between age and sleep parameters was assessed by computing the corresponding Pearson (R) or Spearman (rho) coefficient, according to the normality test. The strength of the relationship was defined as weak with a correlation coefficient of [0.2, 0.4[ or moderate if coefficient of [0.4, 0.6[.

Missing values in the measure were discarded and not considered for the calculations.

The effect size of statistically significant results was calculated using SPSS and Excel. For parametric tests, the effect size was measured by Cohen’s d, and the convention for a small (0.2 - 0.5), medium (0.5 - 0.8) and large (≥0.8) effect size was based on Cohen’s (1988) approach. [12] While for non-parametric tests, the effect size r was calculated as Z statistic divided by square root of the sample size (N) (Z/√N). The interpretation values used for r were the following: 0.1 - <0.3 (small effect), 0.3 - <0.5 (moderate effect) and ≥0.5 (large effect). [13]

Results

Construction of Study Sample

The study sample is constituted by 60 women. The overall sample ranges in age from 25 to 83 years, with a mean age of 50.4 years.

Sleep efficiency is used as an exclusion criterion. Cases with sleep efficiency below 80% were excluded, in order to rule out cases of insomnia and to ensure we only included women who have slept objectively well.

Table 1 summarizes the demographic and sleep efficiency characterization of the patients in the whole sample.

Table 1. Demographic and sleep efficiency characteristics of the total study sample

So, a total of 41 women with sleep efficiency higher than 80% were included in this study, spanning an age range from 26 to 82 years. Next, we categorized our sample into two age groups: group 1, with age below 50 years (Young adults: 26-49 years, n = 28) and group 2, with women aged 50 and over (Adults: 51-82 years, n = 13).

The socio-demographic characteristics of the two age groups are listed in Table 2.

Table 2. Socio-demographic characteristics.

Table 3 summarizes the sleep characterization of the female sample (N = 41).