Evidence ledger
What each source supports, and where it stops.
12 sources stand behind 17 practices. Each entry records the exact part we used, what it supports, what it does not, and when we last checked it.
Three layers we keep apart
- A guideline or study supports a kind of care or a principle. For example, sleep specialists strongly recommend CBT-I for chronic insomnia.
- We wrote a practice card that points in that direction. The card is our own wording and design.
- Whether our card, delivered this way, helps anyone is a separate question that only research with real families can answer. We have not done that research yet.
The first layer being true does not make the third layer true. Each card is labelled as built on a guideline component, stating a clinical boundary, or our own design, so you can tell which is which.
The rules we hold ourselves to
- A card is shown only while every source it cites has been checked and has not been withdrawn. If a source lapses, its cards come down until it is re-checked.
- Information about getting urgent help is exempt from that rule. It stays up no matter what.
- Every change to a source or a card is recorded with its date.
- "Checked" means our editorial team confirmed the source's identity, the part cited and its scope. It does not mean a licensed clinician has approved the practice, and it does not mean the source's authors endorse us.
This approach follows LONG PROOF, LIGHT HOPE's evidence discipline: statements are checked against primary sources before publication and re-checked over time. It is a discipline, not a clinical validation or a certification.
How far a result travels
Before leaning on a study, we ask five questions. A mismatch on any one is stated on the card.
Sources
World Health Organization
- Published
- 2020-09-28
- Part we use
- Recommendation A (universally delivered psychosocial interventions)
- What it supports
- Recommendation A: universally delivered psychosocial interventions should be provided for all adolescents (strength: strong; certainty of evidence: low). Named components include emotional regulation, problem-solving, interpersonal skills, mindfulness, assertiveness and stress management.
- What it does not support
- That this product, or any single conversation card, is a WHO intervention or improves a teenager's mental health. We borrow component directions only and wrote original material.
- How we checked
- Publication page opened; recommendation wording checked against the WHO text hosted on NCBI Bookshelf (NBK565377).
- Last checked
- 2026-11-11 (editorial source mapping; identity and claim re-checked against the publisher on this date)
- Used in
- Listen first, decide together; Feelings can be said. Guesses can be checked.; Safety decides who to ask
National Institute for Health and Care Excellence (NICE)
- Published
- 2019-06-25
- Part we use
- Recommendations 1.1.1, 1.1.6, 1.1.7 and 1.1.21
- What it supports
- Care should rest on a supportive and collaborative relationship with the young person and family (1.1.1), consider the social, educational and family context (1.1.6), offer the young person the opportunity to discuss sensitive issues in private (1.1.7), and involve health and education professionals working together on bullying (1.1.21).
- What it does not support
- That a parent-facing script replaces assessment or treatment, or that bullying can be solved by asking a child to reinterpret it.
- How we checked
- Recommendation wording checked through the official NICE index; direct page access is restricted for automated readers.
- Last checked
- 2026-11-11 (editorial source mapping; identity and claim re-checked against the publisher on this date)
- Used in
- Listen first, decide together; Feelings can be said. Guesses can be checked.; Safety decides who to ask; Put the problem back in its setting
American College of Obstetricians and Gynecologists (ACOG)
- Published
- 2023-06
- Part we use
- Summary of recommendations
- What it supports
- Screening and diagnosis of perinatal mental health conditions belong inside a clinical system that covers depression, anxiety, bipolar disorder, psychosis and risk of self-harm.
- What it does not support
- Any self-scored result from this site. We do not screen, score or diagnose.
- How we checked
- Official summary checked; the full guideline text was not obtained.
- Last checked
- 2026-11-11 (editorial source mapping; identity and claim re-checked against the publisher on this date)
- Used in
- A screening is a doorway
American College of Obstetricians and Gynecologists (ACOG)
- Published
- 2023
- Part we use
- Implementation points for Clinical Practice Guideline No. 4
- What it supports
- Screening should be carried out with systems in place to ensure timely access to assessment and diagnosis, effective treatment, and appropriate monitoring and follow-up.
- What it does not support
- That a screening questionnaire alone is a diagnosis, or that a website can stand in for the follow-up.
- How we checked
- Checked through the official ACOG index; direct page access is restricted for automated readers.
- Last checked
- 2026-11-11 (editorial source mapping; identity and claim re-checked against the publisher on this date)
- Used in
- A screening is a doorway; When it is urgent: stop and get medical help
American College of Obstetricians and Gynecologists (ACOG)
- Published
- 2023-06
- Part we use
- Summary of recommendations
- What it supports
- Treatment of perinatal mental health conditions, including any decision about medication, is a clinical pathway.
- What it does not support
- Any advice from this site about starting, stopping or changing a medicine.
- How we checked
- Official summary checked; the full guideline text was not obtained.
- Last checked
- 2026-11-11 (editorial source mapping; identity and claim re-checked against the publisher on this date)
- Used in
- A screening is a doorway
American College of Obstetricians and Gynecologists (ACOG)
- Published
- FAQ091, last reviewed December 2025
- Part we use
- "I think I have postpartum depression. What should I do?"
- What it supports
- If you think you may have postpartum depression, call your ob-gyn or another health care professional right away. Do not wait until your postpartum checkup.
- What it does not support
- Waiting a set number of weeks before asking for help.
- How we checked
- Patient FAQ text checked through the official index.
- Last checked
- 2026-11-11 (editorial source mapping; identity and claim re-checked against the publisher on this date)
- Used in
- Get ready for one real conversation
National Institute of Mental Health (NIMH)
- Published
- 2023 revision
- Part we use
- Signs and symptoms; How can family and friends help?
- What it supports
- Perinatal depression can be treated. Postpartum psychosis is described as a psychiatric emergency that requires hospitalization. Family and friends can help by encouraging a talk with a health care provider, helping the person get to appointments, and assisting with daily tasks.
- What it does not support
- Using the medication passages of the publication as current prescribing advice; we do not reproduce them.
- How we checked
- Full page opened.
- Last checked
- 2026-11-11 (editorial source mapping; identity and claim re-checked against the publisher on this date)
- Used in
- Get ready for one real conversation; Turn kindness into one task; When it is urgent: stop and get medical help
National Institute for Health and Care Excellence (NICE)
- Published
- 2015-11-12; last updated 2026-04-15
- Part we use
- Recommendations 1.5.22 to 1.5.24 (depressive symptoms and sleep problems associated with menopause)
- What it supports
- Menopause-specific cognitive behavioural therapy can be considered for depressive symptoms that fall short of a depression diagnosis and occur with vasomotor symptoms such as hot flushes and night sweats (1.5.22), and for sleep problems associated with vasomotor symptoms (1.5.24). Suspected or diagnosed depression should follow NICE depression guidance (1.5.23).
- What it does not support
- That every low mood in midlife is menopause, that one approach suits every sleep problem, or that a short card equals menopause-specific CBT.
- How we checked
- Recommendation wording and numbering re-checked on the NICE page. The CBT recommendations date from the 2024 update; the 2026 update concerns other sections.
- Last checked
- 2026-11-11 (editorial source mapping; identity and claim re-checked against the publisher on this date)
- Used in
- Describe the nights; Low mood is not automatically "just menopause"; Your preferences belong in the decision; Comfort tonight, care over time
American Academy of Sleep Medicine (AASM)
- Published
- 2021-02-01
- Part we use
- Recommendations 1 and 6
- What it supports
- Multicomponent cognitive behavioral therapy for insomnia (CBT-I) carries a strong recommendation for adults with chronic insomnia. Sleep hygiene used alone is conditionally recommended against as a treatment.
- What it does not support
- That bedtime tips, or anything on this site, is CBT-I. We point to CBT-I; we do not deliver it.
- How we checked
- Full text opened. Edinger JD et al., J Clin Sleep Med 2021;17(2):255-262, doi:10.5664/jcsm.8986.
- Last checked
- 2026-11-11 (editorial source mapping; identity and claim re-checked against the publisher on this date)
- Used in
- When poor sleep keeps going: ask about CBT-I
National Heart, Lung, and Blood Institute (NHLBI), NIH
- Published
- page last updated 2022-03-24
- Part we use
- Healthy lifestyle changes; Cognitive behavioral therapy for insomnia
- What it supports
- Healthy sleep habits may help. For long-term insomnia, CBT-I is usually the first treatment recommended. A sleep diary can help when you talk with a provider.
- What it does not support
- Replacing treatment with lifestyle tips.
- How we checked
- Full page opened.
- Last checked
- 2026-11-11 (editorial source mapping; identity and claim re-checked against the publisher on this date)
- Used in
- Set the day down; When poor sleep keeps going: ask about CBT-I; Sleepy in the day: safety first; A sleep note for your appointment
Cochrane Database of Systematic Reviews
- Published
- 2022-08-24
- Part we use
- Main results and authors' conclusions
- What it supports
- Across 13 trials with 1,007 adults, listening to music probably improves self-reported sleep quality (10 trials, 708 participants; Pittsburgh Sleep Quality Index mean difference -2.79, 95% CI -3.86 to -1.72; moderate-certainty evidence). Music may not improve objectively measured sleep (3 trials, 136 participants), and the effect on insomnia severity is very uncertain (2 trials, 63 participants).
- What it does not support
- That the short original sounds on this site improve sleep. Trials typically used 25 to 60 minutes of music daily for 3 days to 3 months, with different music and different listeners.
- How we checked
- Plain-language summary and full review opened. Jespersen KV et al., doi:10.1002/14651858.CD010459.pub3. Search date December 2021.
- Last checked
- 2026-11-11 (editorial source mapping; identity and claim re-checked against the publisher on this date)
- Used in
- Comfort tonight, care over time; Quiet or sound: your choice
Journal of Midwifery & Women's Health
- Published
- online 2024-10-17; J Midwifery Womens Health 2025;70(3):387-395
- Part we use
- Results
- What it supports
- In an online trial, 98 first-time pregnant women were randomised to music plus sleep hygiene or sleep hygiene alone for four weeks (71 completed). Both groups improved, with no significant difference between the groups.
- What it does not support
- Reading a before-and-after improvement as proof that the music caused it, or extending the result to postpartum depression.
- How we checked
- Confirmed from the PubMed abstract (PMID 39417384); the publisher page restricts automated readers. Hoegholt NF et al., doi:10.1111/jmwh.13699.
- Last checked
- 2026-11-11 (editorial source mapping; identity and claim re-checked against the publisher on this date)
- Used in
- Quiet or sound: your choice
Found a problem?
If a citation is wrong or out of date, write to hello@lighthope.ai. Corrections are made in the ledger and the affected cards are withdrawn until fixed.