Does Shorter Postnatal Hospital Stay Lead to Postdischarge Complications? An Instrumental Variables Approach

Mohammad Mahbubur Rahman, Saseendran Pallikadavath, Rabeya Khatoon

Research output: Contribution to journalArticle (Academic Journal)peer-review

2 Citations (Scopus)
122 Downloads (Pure)

Abstract

We exploit the difference in means of postnatal hospital stays between beneficiary mothers of Janani Suraksha Yojana (JSY) and control mothers to estimate causal effects of postnatal hospital stay on postdischarge complications of a mother and her newborn. We argue that JSY increases the demand for institutional deliveries in those hospitals, which are assigned by the government to provide such services to the JSY recipients. Given the limited supply of beds, health professionals and other facilities, an excess demand for institutional deliveries in those hospitals forces JSY recipients to stay for a shorter time after birth compared with nonrecipients of JSY who are free to deliver in any hospital. Thus, the dummy for JSY becomes a suitable instrument for postnatal hospital stay. Using instrumental variables (IV) regressions, we find that postnatal hospital stay has statistically significant negative causal effects on half of the available postdischarge complications in the data, including abdominal pain, vaginal discharge, convulsion and severe headache of a mother and fever or cold of the newborn.

Original languageEnglish
Pages (from-to)432-471
Number of pages40
JournalJournal of International Development
Volume31
Issue number5
Early online date15 Apr 2019
DOIs
Publication statusPublished - 1 Jul 2019

Research Groups and Themes

  • ECON Applied Economics
  • ECON CEPS Health

Keywords

  • Janani Suraksha Yojana
  • instrumental variable regressions
  • postnatal hospital stay
  • post-discharge complications

Fingerprint

Dive into the research topics of 'Does Shorter Postnatal Hospital Stay Lead to Postdischarge Complications? An Instrumental Variables Approach'. Together they form a unique fingerprint.

Cite this