Telephone-Based Cognitive Behavioral Therapy for Insomnia in Perimenopausal and Postmenopausal Women With Vasomotor Symptoms

The researchers compared two interventions delivered entirely by telephone:

**CBT-I (Cognitive Behavioral Therapy for Insomnia):** Six 30–45 minute telephone sessions over 8 weeks. Each session included sleep restriction (limiting time in bed to match actual sleep time), stimulus control (going to bed only when sleepy, getting out of bed if awake for more than 20 minutes), sleep hygiene education (avoiding caffeine, alcohol, and screens before bed), cognitive restructuring (challenging unhelpful beliefs about sleep like "I'll never fall asleep"), and behavioral homework (practicing these techniques between sessions). Participants also submitted weekly electronic sleep diaries.

**MEC (Menopause Education Control):** Six telephone sessions of the same duration and schedule, but focused on general menopause and women's health topics (hormone changes, bone health, nutrition, exercise). No sleep-specific advice was given. Participants also submitted weekly sleep diaries.

The primary outcome was the **Insomnia Severity Index (ISI)** score at 8 weeks. Secondary outcomes included the **Pittsburgh Sleep Quality Index (PSQI)** , sleep diary variables (sleep latency, wake after sleep onset, total sleep time, sleep efficiency), hot flash frequency, and hot flash interference with daily life.

**106 women** aged 40–65 years (mean age ~55 years)

All were perimenopausal or postmenopausal (defined as having irregular periods or no period for at least 12 months)

All had **moderate-to-severe insomnia** (Insomnia Severity Index score ≥12 out of 28)

All had **at least 2 hot flashes per day** (self-reported)

Recruited from western Washington State (USA) via community advertisements and clinic referrals

Excluded: women with untreated sleep apnea, restless legs syndrome, current depression (PHQ-9 >14), bipolar disorder, substance abuse, night-shift work, or those already receiving CBT-I or taking sleep medications (except stable doses of antidepressants or hormone therapy)

**Insomnia Severity Index (ISI):** 7-item questionnaire, score 0–28. Scores ≥15 indicate moderate-to-severe insomnia; scores ≤7 indicate no clinically significant insomnia. Measured at baseline, 8 weeks, and 24 weeks.

**Pittsburgh Sleep Quality Index (PSQI):** 19-item questionnaire, score 0–21. Higher scores = worse sleep quality. Measured at same time points.

**Sleep diary:** Daily electronic entries for 2 weeks at baseline, 8 weeks, and 24 weeks. Recorded bedtime, wake time, sleep latency (minutes to fall asleep), wake after sleep onset (minutes awake during the night), total sleep time, and sleep efficiency (percentage of time in bed actually asleep).

**Hot flash diary:** Daily record of number of hot flashes and hot flash interference (rated on a 0–10 scale for how much hot flashes interfered with daily activities).

**Blinded assessments:** Research staff who administered questionnaires did not know which group participants were in. Participants were not blinded (they knew whether they were getting sleep therapy or general education).

**Design:** Single-site, parallel-group, randomized clinical trial (RCT). Participants were randomly assigned 1:1 to CBT-I or MEC using a computer-generated random sequence, stratified by age (<55 vs ≥55 years) and baseline insomnia severity (ISI 12–18 vs ≥19).

**Duration:** The intervention period was 8 weeks (6 sessions). Follow-up assessments occurred at 8 weeks (post-treatment) and 24 weeks (16 weeks after the last session). Total study duration per participant was 24 weeks.

**Blinding:** Outcome assessors were blinded to group assignment. Participants were not blinded (they knew which intervention they received), which is a limitation because expectation effects could influence self-reported sleep. However, the control group received an active, credible intervention (menopause education) rather than a placebo or waitlist, which partially mitigates this.

**Statistical approach:** Intent-to-treat analysis (all participants analyzed in the group they were assigned to, regardless of how many sessions they completed). Mixed-effects models with repeated measures were used to compare groups over time, adjusting for stratification variables. Missing data were handled using maximum likelihood estimation.

**What this design can prove:**

Causal inference: Because of random assignment, any differences between groups at 8 and 24 weeks can be attributed to the intervention (CBT-I vs MEC), not to pre-existing differences.

Generalizability to perimenopausal/menopausal women with insomnia and hot flashes who are willing to do telephone-based therapy.

Durability: The 24-week follow-up shows whether effects persist after treatment ends.

**What this design cannot prove:**

It cannot tell us whether CBT-I works better than other active treatments (e.g., medication, in-person therapy, digital sleep apps) because there was only one active comparator.

It cannot tell us which component of CBT-I (sleep restriction, stimulus control, cognitive restructuring) is most important — the "package" was tested as a whole.

It cannot tell us whether results generalize to men, younger women, or women without hot flashes.

Because participants were not blinded, some of the benefit could be due to expectation effects (though the control group also received attention and education).

The sample was predominantly white (86%) and well-educated (college degree or higher in 70%), limiting generalizability to more diverse populations.

**Methodological weaknesses:**

Single-site study (all participants from one geographic region)

No objective sleep measurement (e.g., actigraphy or polysomnography) — all outcomes were self-reported, which can be biased by memory and expectation

Relatively small sample (106 women) — subgroup analyses (e.g., by age or hot flash severity) are underpowered

24-week follow-up is good but not long-term (no data beyond 6 months)

The MEC control group received active attention and education, but it's possible that some MEC participants inadvertently applied sleep hygiene principles from general health advice

**Primary outcome (Insomnia Severity Index at 8 weeks):**

CBT-I group: ISI decreased from 18.3 at baseline to 8.4 at 8 weeks (mean change: -9.9 points)

MEC group: ISI decreased from 18.1 to 13.4 (mean change: -4.7 points)

Between-group difference: **5.2 points** (95% CI: -6.1 to -3.3, p < 0.001)

This difference was sustained at 24 weeks: CBT-I ISI = 7.5, MEC ISI = 11.7 (difference: -4.2 points, p < 0.001)

**Secondary outcome (PSQI at 8 weeks):**

CBT-I: PSQI decreased from 12.3 to 8.3 (change: -4.0 points)

MEC: PSQI decreased from 12.1 to 10.7 (change: -1.4 points)

Between-group difference: **2.7 points** (95% CI: -3.9 to -1.5, p < 0.001)

Sustained at 24 weeks: CBT-I = 7.7, MEC = 10.1 (difference: -2.4 points, p < 0.001)

**Clinical remission (ISI score ≤7, indicating no insomnia):**

At 8 weeks: 33/47 (70%) of CBT-I participants vs 10/41 (24%) of MEC participants

At 24 weeks: 37/44 (84%) of CBT-I participants vs 16/37 (43%) of MEC participants

Number needed to treat (NNT) to achieve one additional remission at 8 weeks: approximately 2.2

**Sleep diary outcomes (at 8 weeks, CBT-I vs MEC):**

Sleep latency: CBT-I decreased by 18.4 minutes vs MEC decreased by 5.5 minutes (difference: -12.9 minutes, p < 0.001)

Wake after sleep onset: CBT-I decreased by 38.6 minutes vs MEC decreased by 14.8 minutes (difference: -23.8 minutes, p < 0.001)

Sleep efficiency: CBT-I increased from 76.5% to 87.4% vs MEC from 76.1% to 80.2% (difference: +7.2 percentage points, p < 0.001)

Total sleep time: CBT-I increased by 27.2 minutes vs MEC increased by 6.4 minutes (difference: +20.8 minutes, p = 0.02)

**Hot flash outcomes:**

Hot flash frequency: No significant difference between groups at 8 or 24 weeks (both groups decreased by about 1–2 hot flashes per day)

Hot flash interference (0–10 scale): CBT-I decreased by 15.7 points vs MEC decreased by 7.1 points at 8 weeks (p = 0.03); at 24 weeks, CBT-I decreased by 22.8 vs MEC decreased by 11.6 (p = 0.003)

**Adherence:**

87% of CBT-I participants completed all 6 sessions; 91% of MEC participants completed all 6 sessions

No serious adverse events were reported in either group

The 5.2-point greater reduction in ISI for CBT-I is considered a **large clinical effect**. For context, the minimal clinically important difference (MCID) for the ISI is typically 6–7 points for a within-group change, but between-group differences of 3–4 points are considered meaningful. A 5.2-point difference means the average CBT-I participant moved from "moderate insomnia" (ISI ~18) to "subthreshold insomnia" (ISI ~8), while the control group remained in the "moderate insomnia" range (ISI ~13).

Sleep efficiency improved by **7.2 percentage points more** in CBT-I. This means that for every 8 hours in bed, the CBT-I group gained about 35 more minutes of actual sleep compared to the control group.

Wake after sleep onset decreased by **~24 minutes more** in CBT-I. This is roughly equivalent to eliminating one long middle-of-the-night awakening.

Sleep latency decreased by **~13 minutes more** in CBT-I. This is about the time it takes to read a short article or listen to one song — a noticeable reduction in time spent lying awake before falling asleep.

The **84% remission rate** at 24 weeks means that 8 out of 10 women who completed CBT-I had no clinically significant insomnia symptoms 4 months after treatment ended. This is a durable effect, not just a temporary fix.

**Acknowledged by authors:**

Single-site study limits generalizability

No objective sleep measures (actigraphy or polysomnography)

Participants were predominantly white and well-educated

The control group received active attention, which may have reduced the apparent effect size compared to a no-treatment control

Hot flash frequency was self-reported and may be subject to recall bias

**Critical reader notes:**

**No blinding of participants:** Women knew whether they were getting sleep therapy or general education. This could inflate the CBT-I effect due to expectation bias. However, the MEC group also showed improvements (ISI dropped 4.7 points), suggesting that attention alone has some benefit, but CBT-I added substantial extra benefit.

**Self-report bias:** Sleep diaries are known to correlate only moderately with objective measures (actigraphy). People with insomnia tend to overestimate how long it takes them to fall asleep and underestimate total sleep time. The reported improvements could partly reflect changes in perception rather than actual sleep physiology.

**No long-term follow-up beyond 6 months:** We don't know if benefits persist at 1 year or longer. Some CBT-I studies show that effects can fade if people stop practicing the techniques.

**Exclusion of women with depression or sleep apnea:** This limits applicability to the many menopausal women who have comorbid depression or undiagnosed sleep apnea (which becomes more common after menopause).

**Small sample size for subgroup analyses:** The study was not powered to detect differences by age, hot flash severity, or hormone therapy use. It's possible that CBT-I works better for some subgroups than others.

**No measurement of sleep medication use:** Although women on stable doses were included, changes in sleep medication use during the study were not reported. If CBT-I participants reduced their medication use, that would be an important additional benefit.

For someone running their own n=1 experiment:

**What to test:**

A structured CBT-I protocol delivered via telephone or video call (or self-guided using a workbook or app). The key components to implement are:

1. **Sleep restriction:** Calculate your average total sleep time from a 1-week sleep diary, then limit your time in bed to that amount (minimum 5 hours). Gradually increase time in bed by 15 minutes each week when sleep efficiency (time asleep / time in bed × 100) exceeds 85%.

2. **Stimulus control:** Only go to bed when sleepy. If you're awake for more than 20 minutes, get out of bed and do something relaxing in dim light until you feel sleepy again. Use the bed only for sleep and sex.

3. **Sleep hygiene:** No caffeine after 2 PM, no alcohol within 3 hours of bedtime, no screens (phone, tablet, TV) for 1 hour before bed, keep bedroom cool (65–68°F / 18–20°C) and dark.

4. **Cognitive restructuring:** Write down negative thoughts about sleep (e.g., "I'll never fall asleep," "I'll be exhausted tomorrow") and challenge them with evidence (e.g., "I've had bad nights before and still functioned," "Even 5 hours of sleep is enough to get through the day").

**Minimum meaningful duration:**

Run the experiment for **at least 8 weeks** (the duration of the study's intervention period). Sleep restriction can cause initial worsening of sleep (and daytime sleepiness) in the first 1–2 weeks, so don't judge results before week 3.

For a self-experiment, a reasonable timeline: 1 week of baseline measurement, 8 weeks of intervention, 2 weeks of post-intervention measurement. Total: 11 weeks.

**What to measure (specific metrics):**

**Insomnia Severity Index (ISI):** Take the free online questionnaire (7 questions, ~3 minutes) at baseline, weekly during the intervention, and at the end. Target: score ≤7 (no insomnia).

**Sleep diary (daily):** Record bedtime, wake time, estimated time to fall asleep, estimated time awake during the night, total sleep time, and a 1–10 rating of sleep quality. Calculate sleep efficiency nightly.

**Hot flash interference:** Rate on a 0–10 scale how much hot flashes interfered with your daily activities (separate from frequency).

**Daytime functioning:** Rate your energy, mood, and concentration on a 1–10 scale each day.

**Key confounds to control for:**

**Hormone therapy:** If you're on hormone therapy (estrogen, progesterone, or combination), keep the dose stable throughout the experiment. Starting or stopping hormones will confound results.

**Sleep medications:** If you take over-the-counter sleep aids (melatonin, diphenhydramine) or prescription sleep medications, either keep the dose stable or (ideally) discontinue them 2 weeks before starting (with medical supervision if prescription).

**Caffeine and alcohol:** Track your intake daily. Aim to keep caffeine consistent (e.g., 1 cup of coffee before noon) and avoid alcohol entirely during the experiment, as it fragments sleep.

**Stress:** Major life events (