Despite current guideline recommendations, the empiric recommendation to place jaundiced infants “in the sun, by the window” is commonly issued to parents to prevent or treat jaundice.
Material and methodsSpectral irradiance (350−1150 nm) was measured with a spectroradiometer on sunny and cloudy days. Seven residential glazing configurations commonly used in dwellings were then interposed between the sun and the sensor. For each glass type, we obtained measurements without glass and with glass (5 and 30 cm from the sensor) and calculated the relative spectral transmission. We analyzed the irradiance in the blue band (≈ 460−490 nm) and compared it with neonatal phototherapy thresholds (30–65 µW/cm²/nm).
ResultsDirect solar irradiance reached a maximum of 2.8 W/m²/nm at 582 nm and mean values around 2.3 W/m²/nm in the 460–490 nm band on clear days, which were several times higher than intensive phototherapy thresholds. Even on cloudy days, blue irradiance remained clearly above these thresholds. Visible light transmission through glazing elements was high. Transmission did not exhibit relevant attenuation below 400 nm and increased again in the near-infrared region (>750 nm).
ConclusionsTypical residential glazing elements transmit very high levels of solar irradiance, with a broad spectrum that includes UVA and infrared radiation, and blue irradiance several times higher than used in phototherapy, but without spectral selectivity or dose control. Exposing healthy newborns to sunlight through home windows cannot be considered an appropriate or safe intervention to prevent jaundice in developed countries.
Pese a que las guías no recomiendan la exposición al sol, sigue siendo habitual el consejo empírico de colocar al neonato “al sol, detrás de la ventana” para prevenir o tratar la ictericia.
MétodosSe midió la irradiancia espectral solar (350–1150 nm) en días soleados y nublados. Posteriormente se interpusieron siete configuraciones de vidrio representativas de la edificación residencial. Para cada vidrio se realizaron medidas de transmisión espectral sin y con vidrio (5 y 30 cm). Se analizó la irradiancia en la banda azul (≈460–490 nm) y se comparó con los umbrales de fototerapia neonatal (30–65 µW/cm2/nm).
ResultadosLa irradiancia solar directa alcanzó un máximo de 2,8 W/m2/nm a 582 nm y valores medios de aproximadamente 2,3 W/m²/nm en la banda 460–490 nm en días despejados, varias veces por encima del umbral de fototerapia intensiva. Incluso en días nublados, la irradiancia azul se situó claramente por encima. Tras los acristalamientos, la transmisión visible fue alta. La transmisión por debajo de 400 nm no mostró una atenuación relevante y aumentó de nuevo en el infrarrojo (>750 nm).
ConclusionesLos acristalamientos residenciales dejan pasar niveles muy elevados de irradiancia solar, con un espectro amplio que incluye radiación UVA e infrarroja, y una irradiancia azul varias veces superior a la utilizada en fototerapia, pero sin selectividad ni control de dosis. La exposición de recién nacidos sanos a la luz solar a través de las ventanas no puede considerarse una intervención adecuada ni segura para prevenir la ictericia en países desarrollados.
Neonatal jaundice affects more than 50% of full term neonates and an even higher proportion of those born preterm. In most cases, it is a physiological and transient form of unconjugated hyperbilirubinemia; however, in a minority of cases, bilirubin levels can rise high enough to cause acute bilirubin encephalopathy and kernicterus, which can cause serious neurologic sequelae or death. Early identification of at-risk infants and timely initiation of phototherapy are essential to prevent these complications.1–4
International guidelines recommend the use of electric powered phototherapy devices (with fluorescent or LED blue light) delivering an irradiance of at least 30 μW/cm2/nm in the 460–490 nm range, which may reach up to 65 μW/cm2/nm in intensive phototherapy.1,3 Phototherapy is administered under controlled conditions, with a defined spectrum, known intensity, eye protection, and monitoring of body temperature and hydration status.
Despite this, in community practice, it is still common to encounter the informal advice to “put the baby in the sun, near the window” as a strategy to treat or prevent jaundice—a practice not supported by guidelines.1,4 At the same time, trials conducted in low-resource settings using sunlight filtered through films specifically designed to block ultraviolet (UV) and infrared (IR) radiation while maintaining adequate levels of blue light have yielded favorable results when these materials were used in canopy-style structures, in hospital settings, and under close monitoring.5–9
This raises the question whether sunlight passing through the standard single-paned or laminated glass commonly found in homes without any special coatings could provide a spectrum and irradiance comparable to those of conventional phototherapy.
The objective of our study was to experimentally analyze the irradiance and spectrum of sunlight before and after filtering through various types of glazing commonly used in dwellings, and to assess, based on these results, whether or not it is advisable to recommend exposing healthy newborn infants to sunlight through home windows with the aim of preventing jaundice.
Material and methodsStudy designWe conducted a non-interventional, experimental field and laboratory study in two phases:
- 1
Measurement of the spectral irradiance of direct sunlight under different weather conditions.
- 2
Measurement of spectral irradiance through different types of residential glazing (C1–C7), representative of the glass typically used for home windows.
No patients participated in the study, and the study did not use any data from health records.
Measuring equipmentMeasurements were obtained with the StellarRAD handheld spectroradiometer (StellarNet, USA), calibrated and traceable to the National Institute of Standards and Technology (NIST). This instrument can perform optical and spectral measurements (Fig. 1).
Specifications:
- •
Spectrometer range: 350−1150 nm.
- •
Spectral resolution: <1 nm.
- •
Wavelength accuracy: <0.25 nm.
The device enables the measurement of spectral irradiance (W/m2/nm) for solar, LED and fluorescent light.
Measurements of direct solar irradianceSolar irradiance measurements were taken in a city in the Valencian Community with the following coordinates: latitude, 38.9858522; longitude, −0.1962664.
- •
On a sunny day in October, around solar noon
- •
Several days in May with the following sky conditions:
- -
Sunny
- -
Partly cloudy
- -
For each condition, we recorded the full spectrum (350−1150 nm) of incident irradiance.
We used the irradiance values recommended for neonatal phototherapy as a clinical reference1,3:
- •
Minimum recommended: 30 μW/cm2/nm (≈0.3 W/m2/nm) at 460−490 nm
- •
Typical intensive phototherapy values: up to 65 μW/cm2/nm (≈0.65 W/m2/nm)
We analyzed seven glazing configurations (C1–C7) that are representative of the glazing materials commonly used in residential buildings (Table 1).
Types of glass used in the study.
| Code | Type/model | Total thickness (mm) | Solar control | Energy-efficient | Sound insulating |
|---|---|---|---|---|---|
| C1 | Double-paned 4(6)4 | 8 | No | No | No |
| C2 | Double-paned 4(6)4 | 8 | Yes | No | No |
| C3 | Laminated 4 + 4 | 8 | No | No | Yes |
| C4 | Single-paned | 8 | No | No | No |
| C5 | Single-paned | 6 | No | No | No |
| C6 | Single-paned | 6 | No | Yes | No |
| C7 | Laminated 4 + 4 | 8 | No | Yes | No |
According to the manufacturer, and in accordance with UNE-EN 410:2011, the expected luminous transmittance is around 90% for glass without solar control or low-emittance (energy-saving) treatments and around 70%–75% for glazing with such treatments.
Protocol for measurement through glass
For each glazing configuration:
- 1
We placed the glass on a stable support between the source of light (sun) and the sensor of the spectroradiometer.
- 2
We made consecutive measurements of spectral irradiance:
- •
Without glass (reference).
- •
With glass (condition of interest).
- •
- 3
Measurements with glass were conducted placing the glass at two distances from the device (5 and 30 cm), and we found no relevant differences between the two, so we considered them equivalent.
For each type of glazing and wavelength λ, we calculated the spectral transmittance10 as the percentage of the incident solar radiation Ewithout glassλ transmitted by the glazing Ewith glassλ:
Study variables
- •
Direct solar spectral irradiance (W/m2/nm) in the 350−1150 nm range.
- •
Percent transmittance T(λ) for each type of glazing (C1–C7).
- •
Average transmittance in the wavelength ranges of interest:
- -
<400 nm (approximate value for UV radiation, primarily UVA in the measurement range of the spectroradiometer).
- -
400−750 nm (visible light), with emphasis on 460−490 nm (phototherapeutic window).
- -
750 nm (near infrared, as a proxy for solar heat gain).
- -
- •
Qualitative comparison of transmitted blue light irradiance with neonatal phototherapy thresholds.
Measurements on a sunny day in October yielded a maximum irradiance of approximately 2.8 W/m2/nm at 582 nm. In the 460−490 nm range, values were clearly above 0.3 to 0.65 W/m2/nm, that is, far above the thresholds applied in intensive phototherapy.
As for the measurements conducted in May:
- •
On a sunny day, the average irradiance in the 460−490 nm range was approximately 2.3 W/m2/nm, which is approximately seven to eight times the minimum threshold recommended for intensive phototherapy.
- •
On a cloudy day, the irradiance in this range was reduced by about half, but it remained well above the therapeutic threshold.
These data indicate that, even under partly cloudy conditions, outdoor sunlight provides blue irradiance levels far exceeding those required for conventional phototherapy, provided that exposure and timing are favorable.
Spectral transmittance of C1 to C7 glazingFig. 4 shows, as an example, the measurements for glass C1 (double-paned) and glass C5 (single-paned). The measurements were taken both with and without the glass for comparison. This allowed the quantification of the percentage of sunlight transmitted through the glass. Measurements were taken at distances of 5 and 30 cm from the glass, and the results were very similar for both.
Note that glass C1 (bottom row) blocked more light than glass C5 (single-paned), and both reduced the intensity of natural light.
Fig. 5 shows the transmittance results for all types of glass.
The relative spectral transmittance patterns of the seven glazing types (C1-C7) were similar, with some quantitative differences:
- •
Below 400 nm, filtering through glazing did not achieve significant attenuation, with transmittance values close to 100% for many types of glass, suggesting that a significant portion of UVA radiation can pass through these glazing configurations.
- •
In the visible light range (400−750 nm):
- -
The average transmittance of the C5 (6 mm) and C4 (8 mm) single-paned configurations was approximately 90%
- -
The transmittance of the C3 configuration (laminated glass, 4 + 4 mm with acoustic treatment) was very similar, indicating that the acoustic interlayer has little effect on light transmission.
- -
The average transmittance of the C1 double-paned 4(6)4 mm configuration without solar or low-emissivity treatments was approximately 85%.
- -
Glass with low-emissivity/solar control coatings (C7, C2, and C6) exhibited slightly lower transmittance, ranging from 70% to 72%.
- -
- •
In the range above 750 nm (near-infrared), we observed an increase in transmittance, indicating that, while somewhat lower in glazing with specific coatings, the proportion of infrared radiation passing through the glass remains significant.
Given that the direct solar irradiance in the 460−490 nm range was on the order of 2.3–2.8 W/m2/nm under favorable conditions, and given that untreated single-paned and laminated glass transmit approximately 85%–90% of visible light, we estimated that the blue light irradiance through these types of glazing on sunny spring and summer days is approximately 1.8–2.5 W/m2/nm, that is, several times greater than the 0.3 to 0.65 W/m2/nm used in intensive phototherapy.
Even in glass with solar control or low-emittance coatings (visible light transmittance ≈ 70 %–72 %), the resulting blue light irradiance would still be well above the minimum threshold for phototherapy during midday hours.
At the same time, transmission of wavelengths below 400 nm remained high with the tested configurations, and the near-infrared component was significant, implying a potentially high solar gain close to the window.
In short, sunlight passing through residential glazing types C1-C7 preserves very high irradiance levels in the blue light spectrum, in addition to UVA and infrared radiation, far exceeding the doses used in clinical phototherapy, while lacking the spectral selectivity or dose control offered by medical devices.
DiscussionAlthough the Cochrane review suggests that prophylactic sun exposure might slightly reduce the frequency or duration of jaundice, the evidence for this approach was of very low certainty, and there are no robust long-term safety data.11
The results of this experimental study show that standard residential glazing:
- •
Maintain irradiance levels in the blue light range (460–490 nm) that are several times higher than those required for intensive neonatal phototherapy.
- •
Does not significantly filter out UVA or near-infrared radiation.
- •
Does not allow for precise control of the administered dose, as it depends on a wide range of factors (time of day, season, weather, orientation, shadows, dirt on the glass, distance from the window, and surface area of skin exposed).
These findings support the major Spanish guidelines and consensus documents1 that agree on discouraging exposure to sunlight, directly or filtered through glass, as a standard strategy for the prevention or treatment of neonatal jaundice in infants born at or after 35 weeks of gestation in high-resource countries.12–14
Intensity and spectrum of sunlight through glassThe high visible light transmittance (70%–90%) of the studied glazing types implies that the sunlight transmitted to indoor spaces during periods of peak irradiance can far exceed the required thresholds for phototherapy. However, unlike electric powered phototherapy devices, sunlight filtered through glass is not limited to the optimal wavelength range of 460−490 nm, but rather encompasses a broad spectrum including significant UVA and IR components.
From a safety perspective, this “improvised phototherapy” approach carries significant risks15–17:
- •
Thermal risk: The combination of a very high blue light irradiance and abundant infrared radiation can lead to hyperthermia and dehydration, especially if the infant is scantily dressed to increase the exposed surface area. Due to weather variability, episodes of very high irradiance may go unnoticed by caregivers.
- •
Cutaneous and ocular risk: Exposure to UVA is unnecessary for the treatment of hyperbilirubinemia and is associated with skin photoaging and an increased risk of skin cancer and long-term eye damage.10,14 Exposing the eyes of a neonate to high levels of visible light and UVA rays without protection can be particularly problematic.
- •
Organizational risk: The belief that jaundice can be “treated” or “prevented” using natural light from a window may delay a visit to the doctor or the initiation of conventional phototherapy when it is actually indicated during a period (the first few days of life) when bilirubin levels can rise rapidly.1–4
- •
Other risks: Potential adverse reactions to phototherapy, such as hemolysis, allergic diseases, DNA damage, and cancer, must be taken into account. To prevent serious harm to the health of infants, the use of phototherapy in clinical settings must be standardized, optimized, and controlled. Further studies are also needed to elucidate the mechanism underlying adverse reactions to phototherapy in infants and to explore and optimize new treatment regimens in the future.
- •
Other practical risks: set posture, inadequate supervision, falls when placing infant on booster seats or surfaces near windows, etc.
It is estimated that 481 000 newborns develop severe jaundice each year; of this total, 114 000 die and more than 63 000 are left with permanent neurologic impairment. More than 75% of cases occur in sub-Saharan Africa and South Asia, where access to diagnosis and treatment is very limited.
Conventional electric powered phototherapy is the standard of care, but in many low-income countries it poses significant challenges due to a lack of appropriate equipment, insufficient irradiance caused by poor maintenance, an unstable power supply, and high costs that may be unaffordable.
For this reason, in many settings, newborns are exposed to direct sunlight as a traditional practice because there is no alternative (the choice is sunlight or nothing), which carries risks such as sunburn, hyperthermia, or dehydration. In fact, some studies suggest that the optimal window for neonatal exposure to sunlight is in the morning, between 8am and 10am, with a duration of 30−60 min.17
Several trials of filtered-sunlight phototherapy as an alternative to electric phototherapy have been conducted in low-resource settings using special films, and they have demonstrated the non-inferiority of this approach in terms of bilirubin reduction and selected clinical outcomes, provided that nearly all UV radiation is blocked and infrared radiation is significantly reduced while maintaining adequate blue light irradiance, and that there is close monitoring of temperature, hydration, and bilirubin levels.5–9,18
However, these approaches differ fundamentally from home exposure through standard residential glazing:
- •
They use films selected based on detailed spectral and thermal specifications, as opposed to conventional residential glazing.5,9,18
- •
They are implemented using specific structures (canopies) and in hospital settings, under continuous supervision.6–8
- •
They are used to treat clinically significant jaundice, not to prevent physiological jaundice in all healthy newborns.
Thus, this approach should not be confused with at-home sunlight exposure through glass: the concept of filtered-sunlight phototherapy (FSPT) in itself emphasizes that clinical efficacy depends on the careful control of the spectrum and irradiance within the therapeutic range (blue-green light), as well as the specific mitigation of thermal and cutaneous risks. The evidence on FSPT does not support “sunlight for jaundice” at home: it actually demonstrates that, if sunlight is to be used for therapeutic purposes, it must be in the form of a controlled intervention (UV/IR filtering, selection of light in the therapeutic range, monitoring).19,20
Exposure to sunlight through residential glazing varies greatly in terms of spectral transmission and irradiance depending on the type of glazing, the orientation, and weather conditions, and there is no way to measure or systematically monitor these factors, which compromises both effectiveness and safety. Therefore, the use of sunlight through windows as an attempt to prevent or treat neonatal jaundice lacks sufficient scientific basis and may delay effective interventions (transcutaneous bilirubin screening and controlled phototherapy), does not guarantee efficacy, and introduces avoidable risks. The positive results of these FSPT trials do not, therefore, support the recommendation to expose healthy newborns to sunlight through windows in the home as a prophylactic measure.
It must be emphasized that, in the context of phototherapy for neonatal hyperbilirubinemia, spectral irradiance is crucial because:
- •
Bilirubin absorbs light in the 460−490 nm range (blue-green spectrum), with an absorption maximum at 478 nm.
- •
The administered radiation should not include ultraviolet light: UVA (320−390 nm), or, more generally, radiation below 400 nm. Infrared radiation, above 750 nm, should also be absent.
- •
For phototherapy to be effective, a minimum spectral irradiance of 30 μW/cm2/nm in that therapeutic range is required, with exposure of 35%–80% of the body surface area, in order to achieve an effective reduction within 4–6 h.21 In addition, the maximum is typically set at 65 μW/cm2/nm (0.65 W/m2/nm).
- •
The higher the spectral irradiance in that range, the faster the conversion of bilirubin into excretable isomers.22
Current strategies for preventing kernicterus are based on:
- •
Universal bilirubin screening prior to discharge1,13
- •
Risk stratification based on postnatal age, bilirubin levels, and risk factors1,4
- •
Standard electric powered phototherapy when bilirubin levels reach the established thresholds1–4
Systematic exposure to sunlight through windows is not part of these strategies and offers no proven benefit, while introducing potential risks and a false sense of security. The results of this study provide quantitative support for the recommendation against using sunlight through residential glass as a substitute for or a prophylactic adjunct to phototherapy.
LimitationsThere are limitations to this study:
- •
The number of glazing configurations analyzed in the study, while representative, did not cover the full range of available glazing options.
- •
Measurements were taken on a limited number of days and under specific conditions; data were not collected throughout the entire year or for every orientation. Spectral irradiance values were not calculated for clear days around solar noon during the equinoxes and the summer solstice, which represent conditions of maximum irradiance.
- •
The measurement range of the spectroradiometer (starting at 350 nm) limited the accurate assessment of the UVB fraction; UV radiation values should be interpreted with caution.
- •
It did not include any clinical data, so the clinical risk to infants has been extrapolated based on physical principles, phototherapy thresholds, and previous evidence regarding the risks of UV/IR radiation and FSPT.5–9
- •
Due to the potentially toxic effects of blue light therapy, green LED light may also be used to reduce total serum bilirubin levels, as it may cause fewer adverse reactions compared to blue light. This aspect has not been studied specifically. Therefore, green light therapy should be investigated more thoroughly in the future.23
Despite these limitations, the magnitude of the measured irradiance levels and the high transmittance throughout the spectrum strongly support the conclusion that sunlight through residential glass cannot be considered a safe form of at-home phototherapy.
Conclusions- 1
Direct sunlight irradiance levels in the blue spectrum (460−490 nm) are much higher than those required for intensive phototherapy in neonates, even under partly cloudy conditions.
- 2
Commonly used residential glazing types (single-paned, laminated, and double-paned, with or without solar control, low-emissivity, or acoustic treatments), transmit a high percentage of visible light, including the phototherapeutic range, and also significant transmission of UVA and infrared radiation.
- 3
The indoor blue-light irradiance through these types of glazing can far exceed the clinical thresholds for phototherapy, but with significant temporal variability, and is associated to exposure to undesirable wavelengths, increasing the risk of hyperthermia and skin or eye damage in newborn infants.
- 4
In the absence of specifically designed and tested spectrally selective films or coatings and of a professional monitoring system, exposing healthy newborns to sunlight through the windows of their homes should not be recommended as a strategy for preventing jaundice in developed countries.
- 5
An explicit recommendation should be made that health education for families discourage the practice of “placing the baby in the sun entering through a window” as a substitute for screening or follow-up.
We conducted an experimental physics study without participation of human subjects or animals, so it was not necessary to obtain prior approval from a research ethics committee.
FundingThe study received funding from the Unisalut program (https://unisalut.uji.es/), specifically from the POLISABIO subprogram, a Universidad Politécnica de Valencia-FISABIO cooperation program under the 2023 grants for innovation projects (IPs). Project: Fototerapia neonatal domiciliaria (PI2023-04).
The authors have no conflicts of interest to declare.
