T. Nkurunziza1, G. Toma1,2, J. Odhiambo1, R. Maine2,3, R. Riviello2,4,6, N. Gupta1,4, A. Bonane5,6, C. Habiyakare7, T. Mpunga7, B. Hedt-Gauthier1,2,5 1Partners In Health/Inshuti Mu Buzima,Clinical,Kigali, , Rwanda 2Harvard School Of Medicine,Department Of Global Health And Social Medicine,Brookline, MA, USA 3University Of California – San Francisco,Surgery,San Francisco, CA, USA 4Brigham And Women’s Hospital,Boston, MA, USA 5University Of Rwanda,College Of Medicine And Health Sciences,Kigali, , Rwanda 6Kigali University Teaching Hospital,Surgery,Kigali, , Rwanda 7Ministry Of Health,Kigali, , Rwanda
Introduction:
In low-and middle-income countries, nine out of ten patients lack access to timely, safe and affordable surgical care. Most patients seek care at district hospitals with limited surgical capacity, creating a need for referral. Weaknesses in referral systems lead to delays that contribute to substantial disability and death. This study assesses the predictors of delayed referrals for injured patients.
Methods:
This retrospective cohort study included all injured patients between January 1 and December 31, 2013 from three rural district hospitals in Rwanda, with a focused analysis on those recommended for referral. We defined delayed referral as non-execution of referral or execution of referral more than two days after referral recommendation. We performed a multivariate logistic regression using stepwise backward selection to identify the risk factors for delayed referral.
Results:
Of the 1,227 patients with injuries evaluated, 23% (n=282) were recommended for referral. Of these patients, 36.5% were injured through road traffic accidents and 53.6% were diagnosed with closed fractures. Overall, 46.5% (n=107) of the patients recommended for referral had a delay in referral execution. Reasons for delay that were documented in 57 patients’ files included awaiting appointment (45.6%, n=26), lack of space at referral hospitals (40.4%, n=23) and financial (14%, n=8).
In the multivariate model, the major risk factors for delayed referral included district hospital (OR:3.77, 95% CI: 1.5- 9.18), age >35 years (OR=2.45, 95%CI: 1.09-5.50), closed fractures/dislocation (OR=16.37, 95%CI: 3.13-85.78), admission to surgical wards (OR=10.25, 95%CI: 2.70-38.82) and admission for at least seven days prior to referral recommendation (OR=4.80, 95%CI:1.38-16.63).
Conclusion:
Although it is promising that over fifty percent of referrals were completed in a timely fashion, services at district hospitals should be improved to minimize the need for referral. Specifically, improved availability of surgical infrastructure, equipment, supplies, and trained staff at district hospitals may improve the execution of timely and appropriate referrals. Further, patient-centered programs to facilitate referral and support patient expenses may contribute to timely referral and should be further studied.