Trial Review

Please note the ANZCTR will be unattended from Friday 20 December 2024 for the holidays. The Registry will re-open on Tuesday 7 January 2025. Submissions and updates will not be processed during that time.

Registering a new trial?

To achieve prospective registration, we recommend submitting your trial for registration at the same time as ethics submission.

The safety and scientific validity of this study is the responsibility of the study sponsor and investigators. Listing a study does not mean it has been endorsed by the ANZCTR. Before participating in a study, talk to your health care provider and refer to this information for consumers
Trial registered on ANZCTR


Registration number
ACTRN12622000940774
Ethics application status
Approved
Date submitted
27/06/2022
Date registered
1/07/2022
Date last updated
5/10/2024
Date data sharing statement initially provided
1/07/2022
Type of registration
Prospectively registered

Titles & IDs
Public title
SHINE: Sleep Health In Perinatal Care
Scientific title
A Scalable Insomnia Program during the Pregnancy and Postpartum Periods: An Effectiveness-Implementation Hybrid Trial
Secondary ID [1] 307120 0
None
Universal Trial Number (UTN)
Trial acronym
SHINE
Linked study record

Health condition
Health condition(s) or problem(s) studied:
Insomnia 326290 0
Sleep Disturbance 326291 0
Condition category
Condition code
Mental Health 323601 323601 0 0
Other mental health disorders
Reproductive Health and Childbirth 323602 323602 0 0
Childbirth and postnatal care
Neurological 323603 323603 0 0
Other neurological disorders
Public Health 323604 323604 0 0
Health service research

Intervention/exposure
Study type
Interventional
Description of intervention(s) / exposure
The Healthy Sleep program uses therapist-assisted self-help Cognitive Behavioural Therapy for Insomnia (CBT-I) to address maladaptive sleep-related cognitions and behaviors, which play key roles in the maintenance of poor sleep during the perinatal periods. Content is adapted from our previous intervention (Bei et al., 2021, https://doi.org/10.1017/S0033291721001860), and aims to address 3 types of perinatal sleep complaints identified in the literature: (1) symptoms of insomnia, (2) pregnancy and infant-related sleep disturbance, and (3) daytime impairments (e.g., sleepiness, fatigue). Feedback on intervention content, format, readability, visual appeal, mode of delivery was sought and incorporated from a Community Reference Group.

The following evidence-based therapeutic components are included: (1) General skills to increase resilience to sleep challenges: sleep hygiene, addressing unhelpful thoughts and beliefs about sleep, relaxation, and managing night-time worries. (2) Identifying and managing insomnia symptoms (e.g., stimulus control, time-in-bed restriction), and when to seek professional help. (3) Fostering realistic expectations and normalise some sleep loss via early education on sleep patterns of new parents and infants. (4) Mindfulness-based strategies targeting physical discomfort, pain, and cognitive arousal. (5) Age-appropriate and evidence-based infant sleep/settling skills to reduce awakenings and increase maternal sense of control. (6) Prioritising one’s own sleep and rest. (7) Smart naps based on circadian principles. (8) Managing sleepiness/fatigue. (9) Enlisting support. (10) Avoiding supine going-to-sleep position in pregnancy to reduce risk for late stillbirth.

Content of the intervention is delivered via the following means, combined:
- A 60-min standardised telephone or telehealth session is delivered by a trained researcher at program entrance, within 2 weeks of enrolment in pregnancy, to (a) introduce core components (factors contributing to sleep, managing insomnia and sleep deprivation, addressing unhelpful thoughts/beliefs about sleep), (b) discuss personalised strategies using intervention materials (additional components that are relevant to each individual will be added, such as sleep restriction, managing nighttime worries, managing physical discomfort), and (c) motivate adherence and sustainable behavioural change. Only key components and those relevant to the individual (i.e., not all components) will be delivered in this session.
- Up to 3 optional mini-consultations via phone or telehealth (~15 minutes) before childbirth if required;
- 1 mini-consultation via phone or telehealth (~ 30 minutes) when the newborn is 3 months.
- Partner or another person who will also be involved in caring for the newborn is encouraged to attend at least one of the above consultations.
- Multimedia intervention materials (including written text, images, audio, and videos) are delivered digitally via email at enrolment, 5 weeks after enrolment, and 2 weeks, 1.5 months, 3 months, and 6 months postpartum. These electronic materials are timed according to probable sleep challenges at each stage of the perinatal period (e.g., managing insomnia, physical discomfort, and expectation of postpartum sleep at late pregnancy, managing daytime sleepiness during early postpartum). Each electronically delivered module is designed to be succinct and easy to read on a computer, tablet, or phone, and will take no more than 10 minutes to read. Content accessing information such as email opening rate will be captured to assess compliance.
- Participants who require additional support applying intervention materials can request brief email or telephone clarification from our team.
- All sessions will be audio recorded for assessing treatment fidelity.
Intervention code [1] 323580 0
Behaviour
Intervention code [2] 323883 0
Lifestyle
Comparator / control treatment
Sleep Hygiene Booklet: This condition will account for the non-specific effects of attention and participation in a sleep program (e.g., receiving health information, and expectations of benefit). Participants in the control condition will receive an information booklet at program entry containing the psychoeducation and sleep hygiene information from the Healthy Sleep Program without other components. Sleep education and sleep hygiene are widely used in insomnia trials as a control condition with good face validity, because these two components on their own have shown substantially inferior effects on symptoms of insomnia. This booklet will be based on non-active components of previously tested intervention: https://doi.org/10.1017/S0033291721001860
Control group
Active

Outcomes
Primary outcome [1] 331797 0
The primary outcome measure is insomnia symptom severity measured via the Insomnia Severity Index total score (ISI; Bastien, Vallieres & Morin, 2001)
Timepoint [1] 331797 0
There are two primary endpoints because sleep patterns differ substantially during pregnancy and postpartum:
1. Pregnancy endpoint is ISI measured five weeks after baseline.
2. Postpartum endpoint is the average score of the ISI measures across 3, 6, and 12 months postpartum representing overall symptom burden.
Secondary outcome [1] 411167 0
Sleep-related impairment measured using the PROMIS Sleep-Related Impairment (Buysse et al., 2012).
Timepoint [1] 411167 0
T2 - ~5 weeks post baseline (T1)
T3 - 3 months postnatal
T4 - 6 months postnatal
T5 - 12 months postnatal
T6 - 24 months postnatal
Secondary outcome [2] 411375 0
Total quality-adjusted life-years (QALYs) in the birthing parent from baseline to 24-months postpartum, calculated based on AQoL-4D (Hawthorne et al, 1999) index scores at T1 to T6 in the main analysis, and based on PROMIS-Pref (PROPr) index (Dewitt et al, 2018; Hanmer et al, 2018) scores at T1-T6 in sensitivity analyses.
Timepoint [2] 411375 0
T2 - ~5 weeks post baseline (T1)
T3 - 3 months postnatal
T4 - 6 months postnatal
T5 - 12 months postnatal
T6 - 24 months postnatal
Secondary outcome [3] 411376 0
Total quality-adjusted life-years (QALYs) in the infant from birth to 24-months of age, calculated based on parent completion of the Infant Quality of Life Instrument (IQI) at T3-T6 (Jabrayilov et al, 2019; Krabbe et al, 2020).
Timepoint [3] 411376 0
T3 - 3 months postnatal
T4 - 6 months postnatal
T5 - 12 months postnatal
T6 - 24 months postnatal
Secondary outcome [4] 411377 0
Cost-effectiveness: Total cost per birthing parent-infant dyad will be based on parental report, calculated based on (i) per dyad cost of the intervention and usual care conditions, (ii) birthing parent self-report of health service utilization for the birthing parent at T1-T6 and for the infant at T3-T6, and (iii) productivity and activity impairment costs related to insomnia symptoms in the birthing parent during the trial period based on WPAI-SHP [insomnia] data at T1-T6 (Reilly et al, 1993). Treatment effects with respect to total cost will be estimated using a one-part GLM models with gamma variance function and a log link, controlling for birthing parent characteristics at baseline. Results will be expressed as (1) cost per QALY gained, and (2) cost per point improvement on the primary clinical outcome.
Timepoint [4] 411377 0
T2 - ~5 weeks post baseline (T1)
T3 - 3 months postnatal
T4 - 6 months postnatal
T5 - 12 months postnatal
T6 - 24 months postnatal
Secondary outcome [5] 411378 0
Qualitative information regarding the reach, adoption, implementation and maintenance of the Healthy Sleep program. This will be assessed using qualitative interviews with participants from the intervention groups, as well as stakeholders (e.g., hospital staff) and project personnel. These interviews (up to 60 min) will be conducted one-on-one or via small groups, via telephone or videoconferencing. Interviews with different individuals will collectively inform the understanding of the facilitators and barriers to the intervention.
Timepoint [5] 411378 0
T4 - 6 months postnatal (when intervention delivery is completed)

Eligibility
Key inclusion criteria
(a) Expectant mothers and birthing parents who are aged 18 years or older;
(b) At least 26 weeks but no more than 32 weeks gestation at enrolment;
(c) Able to communicate (read/write/speak) in English;
(d) Have regular access to email and the internet.
(e) Currently live in Australia
(f) Score above 7 on the Insomnia Severity Index (ISI; Bastien, Vallières, & Morin, 2001)
Minimum age
18 Years
Maximum age
No limit
Sex
Females
Can healthy volunteers participate?
No
Key exclusion criteria
a) Participants with stable use (five or more times per month) of medications or substances (prescription and over-the-counter) that directly affect sleep; non-stable use will be assessed on a case-to-case basis. Stable use of psychotropic medications for the treatment of non-sleep-related conditions (e.g., antidepressants, anxiolytics) is not exclusionary and any changes in doses will be documented.
b) Participants with unstable medical conditions (e.g., severe diabetes, reflux) that directly and significantly affect sleep will be excluded. Medical conditions that do not directly or significantly impact sleep are not exclusionary.
c) Participants with mental health conditions that significantly affect sleep, including severe current posttraumatic stress disorder, current substance abuse/dependence disorders, lifetime bipolar or psychotic disorders, and current suicidal ideation/self-harm behaviours or individuals who pose a risk of harm to others.
d) Participants who self-report a current diagnosis of the following sleep disorders, with symptoms significantly affecting sleep:
- Sleep apnoea: loud snoring, or observed gasping or pauses in breathing, or previously diagnosed with apnoea/hypopnea index >15/hr but not or inadequately treated;
- Periodic limb movement disorder with arousal index > 15 per hour;
- Self-reported diagnosis of restless legs syndrome (RLS) occurring three times/week, with duration of at least one month and onset prior to pregnancy. Not exclusionary if RLS increased or emerged during pregnancy (as long as pre-pregnancy frequency was no more than once a week);
- Severe circadian rhythm disorders: Irregular Sleep Wake Disorder, Non-24-Hour Sleep-Wake Syndrome, Advance Sleep-Phase Syndrome (if habitual bed time is earlier than 8 pm and habitual wake time is earlier than 4 am. Occasional deviation from this schedule is allowed.), Delayed Sleep-Phase Syndrome (if habitual bed time is later than 3 am and habitual wake time is later than 11 am. Occasional deviation from this schedule is allowed.)
- Narcolepsy;
- Other previously diagnosed sleep disorders if severe, assessed on a case-to-case basis.
e) Participants who are undertaking fixed night shift work (between midnight and 5 am) or rotating work schedules that require night shifts at enrolment;

Study design
Purpose of the study
Treatment
Allocation to intervention
Randomised controlled trial
Procedure for enrolling a subject and allocating the treatment (allocation concealment procedures)
Randomization scheme will be generated prior to project commencement, and uploaded to REDCap, which includes a secure database and implements role-based access of specific data. To randomize, research staff logins to REDCap, enters eligibility and stratification information, and receives group allocation. Staff carrying out allocation will not be able to alter randomization algorithm.
Methods used to generate the sequence in which subjects will be randomised (sequence generation)
Randomisation will be stratified by:
1. Site (hospital coded as 0; community coded as 1)
2. Baseline insomnia severity index (low [less than 14] coded as 0; high [greater than or equal to 14] coded as 1)
3. Parity (first coded as 0; second or more coded as 1)
4. Baseline PROMIS depression symptoms (low [less than or equal to 50] coded as 0; high [great than 55] coded as 1)

A Big Stick Design (BSD) with maximally tolerated imbalance = 2 was selected as the approach to randomization. Unique sequences of length 384 with different seeds were generated for each stratum. These sequence lengths were chosen as it is not known how many participants will be in a specific stratum.
Masking / blinding
Open (masking not used)
Who is / are masked / blinded?



Intervention assignment
Parallel
Other design features
Phase
Not Applicable
Type of endpoint/s
Efficacy
Statistical methods / analysis
For Sample Size;
Sample size calculations were informed by statistical, empirical, and clinical considerations, given cautions on basing power solely on pilot data. Factors considered were as follows:
1. In a previous trial (Bei et al., 2021), Cohen’s d effect sizes for ISI and sleep-related impairment in women with baseline elevated ISI (>7) ranged from 0.43 to 1.15.
2. Trial investigators with clinical experience discussed the smallest meaningful effect and identified a Cohen’s d = 0.40.
3. We calculated bootstrap confidence intervals for Cohen's d from the previous trial data. Results showed that >75% of the time, the pilot effect size was d >0.40, bolstering confidence that this clinically meaningful effect is highly plausible from our intervention.
There are hypothesis tests across the primary and secondary outcomes, counting different measures and the pregnancy/postpartum endpoints, therefore we set a = .01 to adjust for multiple comparisons. Specifically, conventional a = .05. With four primary and secondary endpoints (two for ISI, two for sleep-related impairment), gives:
a = .05 / 4 = .0125. For ease, we rounded this down, conservatively, to a = .01.

For the pilot and for identifying a clinically important difference, we used Cohen's d effect size. However, for power analyses, we wanted to additionally consider that analyses will include baseline (T1) scores and strata as covariates. Collectively, the models will have the following parameters estimated:
- one for the intercept
- one for covarying for the outcome at baseline
- one for the dummy coded treatment condition
- 15 for the dummy coded strata
resulting in a total of 18 estimated coefficients. Cohen's d = 0.40 translates to an Cohen's $f^2$ of 0.04.
The `pwr.f2.test()` function from the `R` package `pwr` was used to calculate the required sample size for the following conditions:
- two-sided a = .01 (accounts for multiple primary and secondary endpoints)
- 1 numerator degree of freedom, testing the dummy coded treatment condition
- 85% power
- 10% dropout (90% retained)
- 18 degrees of freedom used up in the regression models

To prevent overstratification and risk imbalance. As we are using a BSD with maximally tolerated imbalance = 2, the block size approach does not directly apply. However, a maximally tolerated imbalance = 2 for a two group design would yield the same possible maximum imbalance as permuted blocks with block size B = 4. With our sample size of N = 384, we could have up to 24 strata, which is greater than the planned 16 strata, again suggesting that N = 384 is an appropriate sample size for this trial.
Thus, the final planned sample size is N = 384, which will provide at least 85% power to detect the a Cohen's d = 0.40 (Cohen's $f^2$ = 0.04) with two-sided a = .01 to account for multiple comparisons and the outcome variable at baseline and all strata included as covariates.

UPDATED INCREASED SAMPLE SIZE: When recruitment of proposed sample size (n=384) was complete, attrition and completeness of data was considered to assess the appropriateness of sample size. Completion rates for T2, T3, T4 varied between 74% - 84%, which was lower than the anticipated 90%. To account for this increased attrition a increased sample size is required. The additional participant calculation is based on the following: If 70% of sample complete at least 2 postpartum time points we need 346 / 0.7 = 495, this is 111 more than planned, this is 78 more than planned Thus, an increased sample size of n=495 is proposed to account for potential impacts of attrition.

Statistical Analyses for Aim 1
Effectiveness analyses will be conducted according to a pre-specified Statistical Analysis Plan. The first aim is that the level of insomnia and daytime symptoms will be lower in the CBT than Control condition. Participants missing more than 2 of the 3 assessments during postpartum will be considered missing the postpartum endpoint. Primary analyses will be linear regression with insomnia or daytime symptoms at pregnancy and postpartum as the outcome, and treatment condition (CBT vs Control) as the focal predictor. The model will control for recruitment source and any participant characteristics with baseline imbalance between conditions or that differ between participants missing and non-missing on the outcome. Analyses will be intent-to-treat, with participants analyzed in the condition to which they were assigned regardless of actual intervention receipt. The test will be two-sided with a = 0.01. We will test an interaction between condition and recruitment source to assess intervention effect heterogeneity across sources. Sensitivity analyses include: (1) a complier average causal effect analysis, which estimates the impact of an intervention in the population subgroup that complies with its assigned treatment, which will illuminate the intervention impact in women who meet the minimum recommended number of sessions and amount of recommended home practice; (2) multiple imputation for missing outcomes by random forests, which allow non-linear changes over time, and (3) pattern-mixture modeling to explore missing not at random scenarios. Effect sizes along with 95% confidence intervals will be calculated for group differences at each endpoint and any non-normality will be addressed using a non-parametric bootstrap. Finally, we will conduct secondary analysis using linear mixed models to compare trajectories of insomnia and daytime symptoms between conditions.

Economic Evaluation for Aim 2
Cost-effectiveness and cost-utility analyses will be conducted from a societal perspective, summarising the health and productivity effects of the intervention during the trial period. Usual care is selected as comparator to directly inform treatment and funding decisions. Direct costs of the active control condition will be excluded from analyses, with estimates of treatment effects for primary and secondary outcomes interpreted as conservative relative to usual care.
Primary and secondary outcomes for the economic evaluation are designed to capture hypothesised health gains during pregnancy for the pregnant participant and for the birthing parent and infant during the postpartum period. The primary outcome for the economic evaluation will be total quality-adjusted life-years in the birthing parent and infant from baseline to 24-months postpartum. For the birthing parent, total QALYs from baseline to 24-months postpartum will be calculated based on AQoL-4D (Hawthorne et al, 1999) index scores at T1 to T6 in the main analysis, and based on PROMIS-Pref (PROPr) index (Dewitt et al, 2018; Hanmer et al, 2018) scores at T1-T6 in sensitivity analyses. For infants, total QALYs from birth to 24-months will be calculated based on parent completion of the Infant Quality of Life Instrument (IQI) at T3-T6 (Jabrayilov et al, 2019; Krabbe et al, 2020). The secondary outcome for the economic evaluation will be the primary clinical outcome during the postpartum period (average ISI during the postpartum period); representing overall symptom burden for the birthing parent and infant. Treatment effects with respect to QALYs will then be estimated using one-part generalized linear models (GLM), controlling for index scores at baseline and specifying appropriate variance and link functions (Glick et al, 2014).
Total cost per birthing parent-infant dyad will be calculated based on (i) per dyad cost of the intervention and usual care conditions, (ii) birthing parent self-report of health service utilization for the birthing parent at T1-T6 and for the infant at T3-T6, and (iii) productivity and activity impairment costs related to insomnia symptoms in the birthing parent during the trial period based on WPAI-SHP [insomnia] data at T1-T6 (Reilly et al, 1993). Treatment effects with respect to total cost will be estimated using a one-part GLM models with gamma variance function and a log link (rather than transformed OLS or two-part models (Butin & Zaslavsky, 2004)), controlling for birthing parent characteristics at baseline. Results will be expressed as (1) cost per QALY gained, and (2) cost per point improvement on the primary clinical outcome. We will summarise sampling error and parameter uncertainty using the bootstrap acceptability method to calculate confidence intervals and generate cost-effectiveness acceptability curves (Glick et al., 2014).

Implementation Evaluation for Aim 3
Descriptive statistics will be used to describe survey responses concerning reach, adoption, implementation and maintenance of the Healthy Sleep program. Qualitative interviews with participants, research staff and perinatal professionals will be audio recorded, transcribed and thematically analysed to describe factors (barriers, enablers) perceived to influence wider implementation and sustainability of the program. Two experienced qualitative researchers will independently code the qualitative data, agree on the coding frame and discuss discrepancies until agreement is achieved

Recruitment
Recruitment status
Recruiting
Date of first participant enrolment
Anticipated
1/09/2022
Actual
20/02/2023
Date of last participant enrolment
Anticipated
30/11/2024
Actual
Date of last data collection
Anticipated
28/02/2027
Actual
Sample size
Target
495
Accrual to date
461
Final
Recruitment in Australia
Recruitment state(s)
ACT,NSW,NT,QLD,SA,TAS,WA,VIC
Recruitment hospital [1] 22627 0
Monash Medical Centre - Clayton campus - Clayton
Recruitment hospital [2] 22628 0
Casey Hospital - Berwick
Recruitment hospital [3] 22629 0
Dandenong Hospital- Monash Health - Dandenong
Recruitment hospital [4] 22630 0
The Royal Women's Hospital - Parkville
Recruitment hospital [5] 22631 0
The Women’s at Sandringham - Sandringham
Recruitment postcode(s) [1] 37901 0
3168 - Clayton
Recruitment postcode(s) [2] 37902 0
3806 - Berwick
Recruitment postcode(s) [3] 37903 0
3175 - Dandenong
Recruitment postcode(s) [4] 37904 0
3052 - Parkville
Recruitment postcode(s) [5] 37905 0
3191 - Sandringham

Funding & Sponsors
Funding source category [1] 311429 0
Government body
Name [1] 311429 0
National Health Medical Research Council
Country [1] 311429 0
Australia
Primary sponsor type
University
Name
Monash University
Address
18 Innovation Walk, Monash University Clayton VIC 3800
Country
Australia
Secondary sponsor category [1] 312823 0
None
Name [1] 312823 0
Address [1] 312823 0
Country [1] 312823 0

Ethics approval
Ethics application status
Approved
Ethics committee name [1] 310905 0
Monash Health Human Research Ethics Committee
Ethics committee address [1] 310905 0
Ethics committee country [1] 310905 0
Australia
Date submitted for ethics approval [1] 310905 0
22/06/2022
Approval date [1] 310905 0
13/09/2022
Ethics approval number [1] 310905 0
RES-22-0000-379A

Summary
Brief summary
Trial website
Trial related presentations / publications
Public notes

Contacts
Principal investigator
Name 119298 0
A/Prof Bei Bei
Address 119298 0
School of Psychological Sciences
Monash University
18 Innovation Walk
Clayton Campus, Clayton VIC 3800
Country 119298 0
Australia
Phone 119298 0
+61 3 9905 3903
Fax 119298 0
Email 119298 0
bei.bei@monash.edu
Contact person for public queries
Name 119299 0
Bei Bei
Address 119299 0
School of Psychological Sciences
Monash University
18 Innovation Walk
Clayton Campus, Clayton VIC 3800
Country 119299 0
Australia
Phone 119299 0
+61 3 9905 3903
Fax 119299 0
Email 119299 0
bei.bei@monash.edu
Contact person for scientific queries
Name 119300 0
Bei Bei
Address 119300 0
School of Psychological Sciences
Monash University
18 Innovation Walk
Clayton Campus, Clayton VIC 3800
Country 119300 0
Australia
Phone 119300 0
+61 3 9905 3903
Fax 119300 0
Email 119300 0
bei.bei@monash.edu

Data sharing statement
Will individual participant data (IPD) for this trial be available (including data dictionaries)?
Yes
What data in particular will be shared?
The research database will be made publicly available no more than 7 years after the final publication, although we will de-identify the data by removing names, date of birth, addresses/contact details, and any information that may link the data to individual participants. These personally identifying data will be completely erased and destroyed.
When will data be available (start and end dates)?
Data will be available no more than 7 years after the final publication. Data may also be requested by contacting the principle investigator.
Available to whom?
The de-identified database will be made available through Monash Bridge (or similar repositories) to maximize the potential benefit to the scientific and research community.
Available for what types of analyses?
For non-commercial purpose.
How or where can data be obtained?
The database will be accessible by contacting the principle investigator (bei.bei@monash.edu), or at a later stage via a data repository.


What supporting documents are/will be available?

Doc. No.TypeCitationLinkEmailOther DetailsAttachment
16481Study protocol    Via an open access journal publication.
16482Statistical analysis plan    Via an open access journal publication.
16483Informed consent form    Online public repository.



Results publications and other study-related documents

Documents added manually
No documents have been uploaded by study researchers.

Documents added automatically
No additional documents have been identified.