Research report
A research report is the sourced material a domain
dossier is synthesized from — generated on a plan and a cadence, one topic per file.
A report carries no confidence tags. Its bracketed markers say who might have
an incentive to shade a line; none of them says anyone checked it. To reach a score, a line has
to be drafted onto a candidate as unverified and pass the Verifier or the
Corroborator, like everything else.
Section clinical-evidence · Version 2026-09-01 · Cadence quarterly · Evidence class mixed · Sources 9 · Supersedes none
Domain: neonatal-hyperbilirubinemia-home-phototherapy · Scope: peer-reviewed and trial-registry sources reachable by this repo's connectors, WebSearch and WebFetch; a registered outcome with no results posted is a sponsor's intention, not a finding. Excludes CPT descriptors (AMA-licensed) and any jurisdiction outside the US and EU.
Sourcing: three independent US guideline-implementation studies now agree the 2022 AAP threshold revision cut phototherapy use without a measured safety cost, while the home-vs-inpatient comparative literature itself has moved from a small favorable 2020 meta-analysis to two larger, more recent ones that grade the evidence "very low"/"low" and flag higher readmission risk — the guideline's own text remains unretrieved
The domain dossier's §4 already covers the 2022 AAP threshold revision at one US site (Cook County) and the two most recent home-vs-inpatient meta-analyses; this pass targets what that section flagged as thin or unretrieved. Three additional, independent US studies of the 2022 guideline's effect — a national children's-hospital database study, an 8-hospital network study, and a two-cohort chart review — agree the revision reduced phototherapy use and bilirubin testing without a measured increase in readmissions, exchange transfusion, or kernicterus [2][3][4]. The guideline's own companion technical report, read here at full abstract depth rather than as a title pointer, states plainly that "neurotoxicity does not occur until bilirubin concentrations are well above the 2004 exchange transfusion thresholds" and that its own systematic review found only "limited and/or inconsistent evidence" of late phototherapy adverse effects including cancer and epilepsy [1]. Against that, the home-vs- inpatient comparative evidence base has moved in the opposite evidentiary direction over time: a small 2020 meta-analysis (259 neonates) found home phototherapy favorable on bilirubin decline [5], while two larger, more recent meta-analyses grade the combined evidence "very low" to "low" quality and report a higher readmission risk with home phototherapy [8][9]. The AAP guideline's own full text remains unretrieved (HTTP 403), unchanged from the dossier's prior finding.
_Baseline (v1). No prior version; this establishes the starting point for future diffs._
The dossier's §4 recorded only that the 2022 AAP guideline's companion technical report described the threshold change as "narrow," from a title-level pointer. Read at abstract depth, the technical report (PMID 35927519) states the update reviewed English-language RCTs and observational studies published through March 2022, addressing both topics carried over from the 2004 guideline and two topics addressed for the first time — phototherapy-associated adverse effects and the effectiveness of intravenous immune globulin (IVIG) for preventing exchange transfusion. Its stated conclusion: "Accumulated evidence justified narrowly raising phototherapy treatment thresholds in the updated clinical practice guideline" [1]. This is one document's own summary of its evidence base, not an independently retrieved evidence table — mark [single-source]. The guideline's own graded evidence tables, and its stated conditions for home phototherapy specifically (monitoring frequency, admission triggers), remain unretrieved: a fetch of the PDF at publications.aap.org returned HTTP 403 on this pass, the same result the dossier already recorded for both the PDF and the HTML article page.
The dossier's §4 anchors its guideline-effect evidence on one single-center study (Cook County, n=1,577). Three further, independent US analyses were located and each reports the same direction of effect — fewer interventions after the 2022 revision, without a measured increase in the safety outcomes each study tracked:
Because three independent, non-affiliated primary sources agree on the same direction of effect — reduced testing and phototherapy use with no measured safety cost, at three different US care settings and data sources — this is stated plainly rather than carrying a single-source marker; the specific magnitude of each effect is attributed to its own citation because no two studies report the same number. Neither [3] nor [4]'s structured abstract states the study's country explicitly; [2] is explicitly a US freestanding-children's-hospital database study by design (PHIS is a US-only database).
The dossier's §4 already carries the two most recent home-vs-inpatient meta-analyses (998 patients, 2024, "very low" quality, RR 4.61 readmission; and 31 studies, 2026, "low" quality for all outcomes, 3.5% pooled readmission). An earlier, smaller meta-analysis exists that the dossier did not carry: a 2020 review of 259 neonates across studies located via PubMed, Scopus, Embase, Cochrane Library and two Chinese databases, assessed by its own authors as "low risk" on the Cochrane Risk of Bias Tool, found home-based phototherapy "more effective... in reducing the rate of total serum bilirubin (standard mean difference = 0.32, 95% confidence interval = -0.22 to 0.86, P = .04)," while reporting "no significant difference in duration of phototherapy (standard mean difference = 0.59, 95% confidence interval = 0.28 to 0.90, P = .06)" — a confidence interval that does not cross zero paired with a non-significant p-value, an inconsistency present in the abstract itself and not resolved by anything retrieved this pass [5]. Its overall conclusion was that home phototherapy is "an effective, feasible, safe, and alternative" option [5], the opposite valence from the two later, larger meta-analyses' "very low"/"low" quality grading and elevated-readmission-risk finding [8][9]. Each of these three meta-analyses rests on a body of evidence rated by its own authors, and only one (2020) reaches a favorable overall conclusion; mark [5] [single-source] for its specific effect sizes, since no other source located this pass reports the same 229–259-patient early evidence base.
The RCT contributing individual patient data to the pooled meta-analyses above is registered as NCT03536078 ("Home Phototherapy for Term Newborns With Icterus," Region Örebro County, Sweden), a randomized, parallel, single-masked multicenter trial with 147 actual enrollees, gestational age 36–42 weeks, bilirubin 300–400 µmol/L at ≥48 hours of age; its primary outcome is change in parent-child bonding from discharge to 4 months, with parental stress, quality of life, and breastfeeding experience as secondary outcomes. The registry record itself carries hasResults=false — no outcome, positive or negative, has been posted to ClinicalTrials.gov for this trial, so every outcome named above is the sponsor's stated intention, not a finding [sponsor-reported] [6]. A separate, published cost-effectiveness analysis drawing on the same trial population does report a finding — that home phototherapy was cost-effective relative to inpatient care in this Swedish cohort — but that is a class-2 published finding from a secondary publication, not something the registry record itself states, and no dosage-response or bilirubin-trajectory outcome from the trial was located independent of that cost-effectiveness framing [single-source] [7]. The distinction matters for this evidence base specifically: the trial that anchors much of the meta-analytic pooling above never posted its own primary bonding/stress outcomes to the registry, only a derivative economic analysis reached print.
[1] Technical Report: Diagnosis and Management of Hyperbilirubinemia in the Newborn Infant 35 or More Weeks of Gestation — Pediatrics (published 2022, month not given in the PubMed record; accessed 2026-09-01). PMID 35927519 — https://pubmed.ncbi.nlm.nih.gov/35927519/ [guideline] [2] Impact of the 2022 AAP Guidelines on Neonatal Hyperbilirubinemia Admissions: A PHIS Study — Hospital Pediatrics (published 2025, month not given in the PubMed record; accessed 2026-09-01). PMID 40484396 — https://pubmed.ncbi.nlm.nih.gov/40484396/ [peer-reviewed] [3] Bilirubin Measurement and Phototherapy Use After the AAP 2022 Newborn Hyperbilirubinemia Guideline — Pediatrics (published 2024-04-01; accessed 2026-09-01). PMID 38482582 — https://pubmed.ncbi.nlm.nih.gov/38482582/ [peer-reviewed] [4] Effect of the 2022 AAP Guideline for Managing Hyperbilirubinemia in the Newborn — American Journal of Perinatology (published 2026-04, day not given in the PubMed record; accessed 2026-09-01). PMID 40701575 — https://pubmed.ncbi.nlm.nih.gov/40701575/ [peer-reviewed] [5] Home-Based Phototherapy Versus Hospital-Based Phototherapy for Treatment of Neonatal Hyperbilirubinemia: A Systematic Review and Meta-Analysis — Clinical Pediatrics (published 2020-06, day not given in the PubMed record; accessed 2026-09-01). PMID 32423343 — https://pubmed.ncbi.nlm.nih.gov/32423343/ [peer-reviewed] [6] Home Phototherapy for Term Newborns With Icterus — ClinicalTrials.gov (record last updated 2021-03-11; accessed 2026-09-01). NCT03536078 — https://clinicaltrials.gov/study/NCT03536078 [trial-registry] [7] The cost-effectiveness of home phototherapy for hyperbilirubinemia in neonates: results from a randomized controlled trial — Scientific Reports (published 2023, month not given in the PubMed record; accessed 2026-09-01). PMID 37400561 — https://pubmed.ncbi.nlm.nih.gov/37400561/ [peer-reviewed] [8] Efficacy of home phototherapy versus inpatient phototherapy for neonatal hyperbilirubinemia: a systematic review and meta-analysis — Italian Journal of Pediatrics (published 2024-03-04; accessed 2026-09-01). PMID 38433210 — https://pubmed.ncbi.nlm.nih.gov/38433210/ [peer-reviewed] [9] Home phototherapy for neonatal hyperbilirubinemia: a mixed methods systematic review and meta-analysis — Pediatric Research (published 2026-05, day not given in the PubMed record; accessed 2026-09-01). PMID 40715357 — https://pubmed.ncbi.nlm.nih.gov/40715357/ [peer-reviewed]
Well established: the direction of effect from the 2022 AAP threshold revision — fewer bilirubin draws, fewer phototherapy episodes, no measured increase in readmission, exchange transfusion, or kernicterus — rests on three independent, non-affiliated US analyses [2][3][4] using three different data sources (a national children's-hospital database, an 8-hospital network, and a two-cohort chart review), stated plainly in §3.
Thin: the technical report's own characterization of its evidence base, including the "limited and/or inconsistent" late-adverse-effects finding, rests on that single document [single-source] [1]; the underlying primary cohort/case-control studies behind that finding were not independently retrieved this pass. The 2020 meta-analysis's favorable effect sizes rest on that single review [single-source] [5], as does the cost-effectiveness finding from the Örebro RCT [single-source] [7]. NCT03536078's own registered outcomes carry no posted results and are marked [sponsor-reported] [6].
Rescoped from class 3: none in this report. Every proposition here is either a registry fact (NCT03536078's registered design and its hasResults=false status) or a published measured quantity; no question about what a clinician, parent, or payer would do was posed or answered.
Out of scope: the AAP guideline's own full text and its graded evidence tables — a publications.aap.org fetch of the guideline PDF returned HTTP 403 on this pass, the same result the dossier already recorded; CPT descriptors; any jurisdiction outside the US and EU.
Not searched vs. not found: Not searched this pass — the individual primary studies underlying the technical report's cancer/epilepsy adverse-effects statement (a targeted PubMed search for "phototherapy neonatal jaundice childhood cancer epilepsy risk cohort" within the last 10 years returned zero records, so the search was attempted and came back empty rather than not run); the country/site of the 8-hospital network [3] and the two-cohort study [4], neither of which is stated in the structured abstract retrieved. Not found — a US population-level home-vs-inpatient readmission comparison directly analogous to the two larger meta-analyses in dossier §4; a primary bilirubin-trajectory or bonding/stress outcome from NCT03536078 itself, independent of the derivative cost-effectiveness publication.
[inference] None beyond what is marked at the sentence level in §3; every comparative statement above (e.g., "the opposite valence from the two later, larger meta-analyses") compares two directly retrieved sources rather than synthesizing a new claim.
| Proposition | Evidence class | Resolvable identifier | Dossier section |
|---|---|---|---|
| A PHIS-database study found hyperbilirubinemia hospitalizations at US freestanding children's hospitals fell from 5,051 to 3,778 in the 12 months following the 2022 AAP guideline's publication (P<.001), with no change in IVIG use, exchange transfusion, or kernicterus incidence | 2 published | PMID 40484396 | 4 |
| An 8-hospital US network study of >22,000 newborns ≥35 weeks' gestation found a 47% relative decrease in phototherapy utilization (3.9% to 2.1%, P<.001) after the 2022 AAP guideline, without an increase in readmissions for phototherapy | 2 published | PMID 38482582 | 4 |
| A two-cohort chart review (1,058 pre- vs. 1,029 post-2022-guideline newborns) found a 51% decrease in phototherapy sessions and a 35% reduction in readmissions for phototherapy | 2 published | PMID 40701575 | 4 |
| The AAP 2022 guideline's companion technical report states its systematic review found only "limited and/or inconsistent evidence" of late phototherapy adverse effects, including cancer and epilepsy | 2 published | PMID 35927519 | 4 |
| A 2020 meta-analysis of 259 neonates found home-based phototherapy associated with a greater reduction in total serum bilirubin than hospital-based phototherapy (SMD 0.32, 95% CI -0.22 to 0.86, P=.04) | 2 published | PMID 32423343 | 4 |
| NCT03536078, a 147-patient RCT of home vs. hospital phototherapy in Swedish term newborns, has posted no results for its registered parent-child bonding and parental stress outcomes | 1 registry | NCT03536078 | 4 |