Risks and seasonal pattern for mortality among hospitalized infants in a conflict-affected area of Pakistan, 2013-2016. A retrospective chart review. [version 1; peer review: 2 approved, 1 approved with reservations]

• Main Authors: Babette van Deursen, Annick Lenglet, Cono Ariti, Barkat Hussain, Jaap Karsten, Harriet Roggeveen, Debbie Price, Jena Fernhout, Ahmed Abdi, Antonio Isidro Carrion Martin
• Format: Article
• Language: English
• Published: F1000 Research Ltd 2019-06-01
• Series: F1000Research
• Online Access: https://f1000research.com/articles/8-954/v1

Background: In recent years, Médecins Sans Frontières has observed high mortality rates among hospitalized infants in Pakistan. We describe the clinical characteristics of the infants admitted between 2013 and 2016 in order to acquire a better understanding on the risk factors for mortality.  Methods: We analyzed routinely collected medical data from infants (<7 months) admitted in Chaman and Dera Murad Jamali (DMJ) hospitals. The association between clinical characteristics and mortality was estimated using Poisson regression. Results: Between 2013 and 2016, 5,214 children were admitted (male/female ratio: 1.60) and 1,178 (23%) died. Days since admission was associated with a higher risk of mortality and decreased with each extra day of admission after seven days. The first 48 hours of admission was strongly associated with a higher risk of mortality. A primary diagnosis of tetanus, necrotizing enterocolitis, prematurity, sepsis and hypoxic-ischemic encephalopathy were strongly associated with higher rates of mortality. We observed an annual peak in the mortality rate in September. Conclusions: The first days of admission are critical for infant survival. Furthermore, the found male/female ratio was exceedingly higher than the national ratio of Pakistan. The observed seasonality in mortality rate by week has not been previously reported. It is fully recommended to do further in-depth research on male/female ratio differences and the reasons behind the annual peaks in mortality rate by week.