Improving neonatal outcomes in regional hospitals in Ghana using an integrated approach to systems change Medge Owen, MD Professor of Obstetric Anesthesiology Wake Forest School of Medicine Executive Director, Kybele www.kybeleworldwide.org
Founded in 2001 Non-profit 501 (c)(3) organization to promote safe childbirth worldwide through medical education partnerships
Kybele s Educational Model On-site Education Cost Efficient International Trainers Multidisciplinary Teams Leadership Involvement / Development Monitoring / Evaluation Kybele fertility goddess 7,000 BC
Context Regional Referral Hospitals Mortality rates are much higher than national average 1,2 They have high work volumes Many high risk cases are referred late They depend on resources: equipment, medication and blood Staff numbers are inadequate Lack organization and problem solving skills 1 Acta Obstet Gynecol Scand 2012;91:87-92 ; 2 Perinatol 2012;36:79-83
Context Regional Referral Hospitals MMR in Ghana decreased from 470 to 380 maternal deaths/100,000 live births between 2005 and 2013 (MGD target 185) In tertiary hospitals MMR is higher than the national average: Koforidua MMR 957 (2004-2009) Semin Perinatol 2012;36(1):79-83 Tamale Teaching Hospital MMR 842 (2008) Ghana Med J 2011;45(3):105-110 Korle Bu Teaching Hospital MMR 840 (2012) Internal source Komfo Anokye Teaching Hospital MMR 1004 (2008-2010) Acta Obstet Gynecol Scand 2012:91(1):87-92
Aim: 2007 to Present Dr. George Yankee, Ghana Health Minister with Dr. Medge Owen, WFUSM Aim: 50% Reduction of Maternal & Newborn Mortality in Tertiary Hospitals.
Two Systems Strengthening Approaches Picking the low hanging fruit (2007-2011) Building a stronger foundation (2013-2015)
Theory of Change : 2007 to 2011 Identify systematic challenges Strengthen leadership Motivate & empower staff Improve knowledge & skills Initiate triage & patient flow processes Partnership Srofenyoh,Int J Gynecol Obstet 2012;116:17-21 Maximize physical workspace Improve resources & logistics Improve service quality & clinical standards Improve communication within & between departments Facilitate communication & feedback with referral centers Monitor implementation Process Map Personnel-Based Bundles Systems-Management Bundles Quality & Communication Bundles Process Mapss Map Advocacy at all levels Frequent monitoring visits Appropriate & timely referrals Improved patient surveillance OUTPUTS Clinical protocol & guideline use Timely intervention & reduction of delay Improved responsiveness towards patients Improve patient satisfaction SECONDARY OUTCOMES Improve institutional reputation Produce local trainers & experts PRIMARY OUTCOME Reduce maternal & newborn morbidity & mortality
Theory of Change : 2013-2015
INTEGRATED MODEL (2013-2015) Clinical Excellence Operational Excellence Leadership Excellence Integrated Health Systems Strengthening Approach
QI Projects (2013-2015) Obstetric Theater Delay Obstetric Triage NICU Hand Hygiene NICU 5S Of 926 admissions, median waiting time to be seen 40 min; max. 1 day, 2.5 hours Median Decision to Delivery time Emergency CS = 4 hrs, Elective CS = 3 days
Leadership Activities (2013-2015) Emotional Intelligence Workshop Leadership Styles Training Accountability Workshop Clinical Champions Leadership Training Leadership Ambassador Training Compassionate Care Workshop Individual Coaching Sessions
Clinical Training (2013-2015) Obstetric Triage Neonatal Resuscitation Labor and Delivery NICU CPAP Training
% Adherence % Adherence Process Results Hand Hygiene Adherence (4/22/15-5/18/15) 1211 observations with 92% compliance (from 67% compliance baseline) Post intervention data shows that Diaper Change (lowest adherence at baseline) had highest % increase 120% 100% 80% 60% 40% 20% 67% 92% 89% 53% 74% 98% 89% 88% PRE %Adh POST %Adh 120% 100% 80% 60% 40% 20% 0% 92% 89% 79% 67% Total Observations 71% 92% 96% 67% Morning Afternoon Night NICU Shifts 0% Total Observations PRE %Adh POST %Adh Diaper Change Feeding NICU Activities Examination Post intervention data shows that Night Shift (lowest hand hygiene adherence at baseline) had highest % increase
Number of CS operations performed by Senior House Officers Process Results: Obstetric Theater Delay 300 250 200 150 100 50 0 CS performed by Senior House Officers 67 181 248 All Senior House Officers were trained to perform cesarean sections when Specialists are attending another case or are otherwise unavailable Percentage of mothers waiting an unacceptable length of time for CS: Before intervention: 9% After intervention: 2% Both the theater staff and mothers benefit from this increase in available surgeons
Waiting Time (mins) Process Results: Obstetric Triage 45 Median Wait Time from Arrival to Assessment 40 35 30 25 20 15 10 5 0 2011 2012 2013 2014 2015 2016 2017 2018 Year Waiting time for assessment reduced from 40 minutes to 5 minutes and sustained
Process Results: NICU 5S BEFORE AFTER
Case Fatality Rates (%) Case Fatality Rates 2007-2015 16 14 12 10 8 6 Hemorrhage Pre-eclampsia 4 2 0 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 Year
Neonatal Outcomes 2012-2016 Year Stillbirth Rate Neonatal Death Rate 2012 12.0 31.1 43.1 2013 10.2 23.8 33.9 2014 9.9 17.1 27.0 2015 9.4 23.6 33.0 2016 8.3 25.1 33.3 Institutional Perinatal Mortality Rate Estimated 392 newborn lives were saved
Lessons Learned Health Systems Strengthening is fundamental to progress Change can be slow There can be setbacks There may be different avenues to success
Deliveries per Year Maternal Mortality Rate Greater Accra Regional Hospital 2007-2014 MMR Deliveries and MMR - Ridge Hospital: 2007 to 2014 12000 600 10000 550 8000 500 6000 4000 450 400 Deliveries MMR 2000 350 0 1 2 3 4 5 6 7 8 Year 300 Photo Source: Kyebele Worldwide
Per 1000 Live Births Neonatal Outcomes 2012-2016 35 Stillbirth and Neonatal Death Rate - Greater Accra Regional Hospital 2012-2016 30 25 20 15 Stillbirth Rate Neonatal Death Rate 10 5 0 2012 2013 2014 2015 2016 Year
Newborn Statistics Ridge Regional Hospital Years 2011 2012 2013 2014 2015 2016 Neonatal Death Rate % NICU Deaths 38.0 31.1 23.8 17.1 23.6 25.1 38% 29% 19% 13% 15% 16% Neonatal Death Rate = deaths/live births x 1000 % NICU Deaths = deaths/admissions x 100