We note the interest generated in this journal [1], on our paper [2], and wish to provide the following clarifications for the benefit of the readership. Dr. Mathew’s opinion was that our study had a high risk of bias principally because: (i) specific method used for the sequence generation was unclear; (ii) the primary outcome (rate of bilirubin decline) assessors were not blinded; and (iii) ‘treatment days’, rather than ‘number of infants enrolled’ was chosen as unit of measurement.

First, Dr Matthew apparently overlooked details provided in our study protocol on the computer-generated block randomization done independently by the USA-based study statistician (provided as supplementary data to the main article [2]. The randomization method was indeed ‘adequate’ [3]. Second, as in most clinical settings, bilirubin levels were objectively determined by the laboratory technician using duly calibrated bilirubinometer on blood samples from the infants at stated intervals to monitor need for continuation or withdrawal of treatment by filtered sunlight or conventional phototherapy. As reported [2,4], the laboratory technician responsible for measuring serum bilirubin levels was unaware of the treatment allocation sequence prior to bilirubin determination for eligible infants. As our primary outcome was objectively measured and the risk of bias minimal, blinding of the participating parents, or the hospital personnel was considered unnecessary [2,3,5]. Third, the stated aim was to compare the rate of bilirubin decline in babies able to tolerate filtered sunlight or conventional phototherapy for at least 5 hours. As interruptions in the management of temperatures outside the acceptable range were not predetermined, treatment days were variable. Hence, the need to appropriately define the unit of measurement as a ‘treatment day’ rather than ‘number of infants’ randomized. This formed the basis of the required sample size. There was indeed no statistical or ethical justification for continuing with enrolment once the required treatment days had been achieved.

1. Mathew JL,　Kumar A,　Khan AM. Filtered sunlight for treatment of neonatal hyperbilirubinemia. Indian Pediatr. 2015;52:1075-9.

2. Slusher TM, Olusanya BO, Vreman HJ, Brearley AM, Vaucher YE, Lund TC, et al. A randomized trial of phototherapy with filtered sunlight in African neonates. N Engl J Med. 2015;373:1115-24.

3. Higgins JP, Altman DG, Gøtzsche PC, Jüni P, Moher D, Oxman AD, et al. Cochrane Bias Methods Group; Cochrane Statistical Methods Group. The Cochrane Collaboration’s tool for assessing risk of bias in randomised trials. BMJ. 2011;343:d5928.

4. Slusher TM, Olusanya BO, Vreman HJ, Wong RJ, Brearley AM, Vaucher YE, et al. Treatment of neonatal jaundice with filtered sunlight in Nigerian neonates: study protocol of a non-inferiority, randomized controlled trial. Trials. 2013;14:446.

5. Boutron I, Tubach F, Giraudeau B, Ravaud P. Blinding was judged more difficult to achieve and maintain in non-pharmacologic than pharmacologic trials. J Clin Epidemiol. 2004;57:543-50.

6. Olusanya BO, Ogunlesi TA, Kumar P, Boo NY, Iskander IF, de Almeida MF, et al. Management of late-preterm and term infants with hyperbilirubinaemia in resource-constrained settings. BMC Pediatr. 2015;15:39.