mdlimodeling rf lreferral nt knetworks to maternal in ethiopia...referral network total population...

52
M d li Rf lN t k Modeling ReferralNetworks to Avert Maternal Death in Ethiopia Caleb Parker GIS Analyst Behavioral and Social Sciences FHI

Upload: others

Post on 21-Aug-2020

3 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: MdliModeling Rf lReferral Nt kNetworks to Maternal in Ethiopia...Referral Network Total Population in New Referral Network Population with Access to their Current Tier A Population

M d li  R f l N t k  Modeling Referral Networks to Avert Maternal Death in Ethiopia

Caleb ParkerGIS AnalystBehavioral and Social SciencesFHI

Page 2: MdliModeling Rf lReferral Nt kNetworks to Maternal in Ethiopia...Referral Network Total Population in New Referral Network Population with Access to their Current Tier A Population

Presentation Overview

• Referral backgroundReferral background• Setting: Ethiopia• Network AnalysisNetwork Analysis

– Objectives, data and methods– Defining referral networksDefining referral networks– Model building and preliminary results

• Next steps in EthiopiaNext steps in Ethiopia

Page 3: MdliModeling Rf lReferral Nt kNetworks to Maternal in Ethiopia...Referral Network Total Population in New Referral Network Population with Access to their Current Tier A Population

REFERRAL BACKGROUND

Page 4: MdliModeling Rf lReferral Nt kNetworks to Maternal in Ethiopia...Referral Network Total Population in New Referral Network Population with Access to their Current Tier A Population

What and why referral?

• Defined: upward movement in the seeking of care within p gthe health system.– Emergency Referral

f l i k i h fi ld f l d b• Referral is key in the field of maternal and newborn health– “under‐documented, under‐researched and under‐o , u de e e c ed d u de

theorized.”

• Referral utilizes current resources through linkages in a tsystem

– Cheaper– Short to medium‐term solutions

Page 5: MdliModeling Rf lReferral Nt kNetworks to Maternal in Ethiopia...Referral Network Total Population in New Referral Network Population with Access to their Current Tier A Population

Time between the beginning of a complication and death

Complication Hours DaysH h  Hemorrhage 

Postpartum  2

Antepartum 12

Ruptured uterus 1

Eclampsia 2

Obstructed labor 3Infection 6

Page 6: MdliModeling Rf lReferral Nt kNetworks to Maternal in Ethiopia...Referral Network Total Population in New Referral Network Population with Access to their Current Tier A Population

The 3 Delays Model3 y

Referral has the potential to reduce all 3 delays

DELAY #1 DELAY #2 DELAY #3Recei ing

Referral has the potential to reduce all 3 delays

Deciding to seek care

Reaching a facility

Receiving adequate

care

Onset of Recovery or deathComplication

Page 7: MdliModeling Rf lReferral Nt kNetworks to Maternal in Ethiopia...Referral Network Total Population in New Referral Network Population with Access to their Current Tier A Population

Pyramidal structure & bypassing

Regional Hospital Receiver

District Hospital

Transport

Health center/post

p

/dispensary

Sender

Adapted from Jahn & De Brouwere, 2001Community

Page 8: MdliModeling Rf lReferral Nt kNetworks to Maternal in Ethiopia...Referral Network Total Population in New Referral Network Population with Access to their Current Tier A Population

Requisites of a functioning referral systemq gCommunication

F ti iTransport

Functioning referral center

Protocols for senders &receiversreceivers

S M SF P SC M t it f l t i d l i t i C t k l d d f t hSource: Murray SF, Pearson SC. Maternity referral systems in developing countries: Current knowledge and future research needs. Soc Sci & Med 62, 2006.

Page 9: MdliModeling Rf lReferral Nt kNetworks to Maternal in Ethiopia...Referral Network Total Population in New Referral Network Population with Access to their Current Tier A Population

SETTING: ETHIOPIASETTING: ETHIOPIA

Page 10: MdliModeling Rf lReferral Nt kNetworks to Maternal in Ethiopia...Referral Network Total Population in New Referral Network Population with Access to their Current Tier A Population

Setting: Ethiopia

• Population: 73 millionp 73

• Area: 1.1 million km2

Twice the size of Texas– Twice the size of Texas

Page 11: MdliModeling Rf lReferral Nt kNetworks to Maternal in Ethiopia...Referral Network Total Population in New Referral Network Population with Access to their Current Tier A Population

Regions of Ethiopia

hAmhara

Page 12: MdliModeling Rf lReferral Nt kNetworks to Maternal in Ethiopia...Referral Network Total Population in New Referral Network Population with Access to their Current Tier A Population

Tigray and Amhara – Key Indicators

Tigray Amhara NationalTigray Amhara  National

Population (2007 Census) 4,400,000 17,200,000 73,900,000

Number of hospitals 14 18 112Number of hospitals  14 18 112

Percent minimum recommended (CEmOC) 55% 18% 39%

Institutional delivery rate 9% 5% 7%Institutional delivery rate  9% 5% 7%

Cesarean delivery rate 0.7% 0.2% 0.6%

Percent of health centers with desired staffing 45% 2% naPercent of health centers with desired staffing 45% 2% na

Page 13: MdliModeling Rf lReferral Nt kNetworks to Maternal in Ethiopia...Referral Network Total Population in New Referral Network Population with Access to their Current Tier A Population

OVERVIEW OF DATA AND OVERVIEW OF DATA AND METHODS

Page 14: MdliModeling Rf lReferral Nt kNetworks to Maternal in Ethiopia...Referral Network Total Population in New Referral Network Population with Access to their Current Tier A Population

Background for this activity

• Ethiopia Needs Assessment for EmONC conductedp

– 797 Healthcare facilities

Geocoded– Geocoded

• Staff saw potential GIS use

• Emergency Referral Network Analysis

– Strengthening referral systems could save lives in short and medium term, even as longer term strategies are implemented 

Page 15: MdliModeling Rf lReferral Nt kNetworks to Maternal in Ethiopia...Referral Network Total Population in New Referral Network Population with Access to their Current Tier A Population

Objectives:

1. Model current referral system y

2. Identify facilities 2+ hours from closest receiving facility

3. Model short and medium term solutions to reduce travel time

Page 16: MdliModeling Rf lReferral Nt kNetworks to Maternal in Ethiopia...Referral Network Total Population in New Referral Network Population with Access to their Current Tier A Population

Data

1. Ethiopia EmONC Assessmentp2. Population Data – 2007 Census projected to 20083. Digital Elevation Model3 g4. Road Network Data

Page 17: MdliModeling Rf lReferral Nt kNetworks to Maternal in Ethiopia...Referral Network Total Population in New Referral Network Population with Access to their Current Tier A Population
Page 18: MdliModeling Rf lReferral Nt kNetworks to Maternal in Ethiopia...Referral Network Total Population in New Referral Network Population with Access to their Current Tier A Population

Ethiopia Elevation

Page 19: MdliModeling Rf lReferral Nt kNetworks to Maternal in Ethiopia...Referral Network Total Population in New Referral Network Population with Access to their Current Tier A Population
Page 20: MdliModeling Rf lReferral Nt kNetworks to Maternal in Ethiopia...Referral Network Total Population in New Referral Network Population with Access to their Current Tier A Population

Ethiopia Road Network

Page 21: MdliModeling Rf lReferral Nt kNetworks to Maternal in Ethiopia...Referral Network Total Population in New Referral Network Population with Access to their Current Tier A Population

Road Network Discrepancies

Page 22: MdliModeling Rf lReferral Nt kNetworks to Maternal in Ethiopia...Referral Network Total Population in New Referral Network Population with Access to their Current Tier A Population

Road Network Discrepancies

Page 23: MdliModeling Rf lReferral Nt kNetworks to Maternal in Ethiopia...Referral Network Total Population in New Referral Network Population with Access to their Current Tier A Population

Methods

1 Define Sender and Referral Facilities1. Define Sender and Referral Facilities

2. Model Current Referral Network

3. Model Referral Scenarios

B ild f ilit   t h t   – Build facility catchment areas 

Page 24: MdliModeling Rf lReferral Nt kNetworks to Maternal in Ethiopia...Referral Network Total Population in New Referral Network Population with Access to their Current Tier A Population

Defining facilities

• Tier A (Receivers)– conducted cesarean delivery in 3 months prior to the assessment

• Tier B (Senders)– all other facilities

• Identified referral networks based on • Identified referral networks based on shortest travel time to a Tier A facility

Page 25: MdliModeling Rf lReferral Nt kNetworks to Maternal in Ethiopia...Referral Network Total Population in New Referral Network Population with Access to their Current Tier A Population
Page 26: MdliModeling Rf lReferral Nt kNetworks to Maternal in Ethiopia...Referral Network Total Population in New Referral Network Population with Access to their Current Tier A Population
Page 27: MdliModeling Rf lReferral Nt kNetworks to Maternal in Ethiopia...Referral Network Total Population in New Referral Network Population with Access to their Current Tier A Population

Model Current Referral Network

Key Variable: Travel Timel l d di l i f h iCalculated direct travel time for each Tier B 

facility based on:Di t     d  t kDistance on road networkRoad surfaceSeason

Calculated adjusted travel time based on:Motorized transportation available on siteCommunication (if no transportation)

Page 28: MdliModeling Rf lReferral Nt kNetworks to Maternal in Ethiopia...Referral Network Total Population in New Referral Network Population with Access to their Current Tier A Population

Key Variable – Travel Time

Factors Tier B Facility #1 Tier B Facility #2Factors Tier B Facility #1 Tier B Facility #2Direct travel time (min) to closest Tier A

48  118

‐ Road Surface (in GIS) (in GIS)

‐ Season Dry Dry

Transportation on site No (48x2=96) Yes (118x1=118)Communication No (96+30=126) N/AAdj t d t l ti   6 8Adjusted travel time (min)

126 118

Page 29: MdliModeling Rf lReferral Nt kNetworks to Maternal in Ethiopia...Referral Network Total Population in New Referral Network Population with Access to their Current Tier A Population
Page 30: MdliModeling Rf lReferral Nt kNetworks to Maternal in Ethiopia...Referral Network Total Population in New Referral Network Population with Access to their Current Tier A Population
Page 31: MdliModeling Rf lReferral Nt kNetworks to Maternal in Ethiopia...Referral Network Total Population in New Referral Network Population with Access to their Current Tier A Population

Percentage of facilities within 60 and 120 minutes to closest Tier A facilityminutes to closest Tier A facility

C l i i Cl CCumulative Minutes to Closest Tier A Facility

Count Percentage

Under 60 77 35.8%Under 60 77 35.8%61 to 120 64 29.8%More than 120 74 34.4%3

Page 32: MdliModeling Rf lReferral Nt kNetworks to Maternal in Ethiopia...Referral Network Total Population in New Referral Network Population with Access to their Current Tier A Population

Model building

• Build Catchment AreasM d l 0  St t  Q• Model 0: Status Quo

• Model 1: All facilities have motorized transportation  and communicationand communication

• Model 2: Upgrading strategic Tier B facilities to Tier A facilitiesfacilities

• Model 3: Improved road surfaces• Optimal model: what is the best outcome with the • Optimal model: what is the best outcome with the 

lowest input (additional phones/vehicles)

Page 33: MdliModeling Rf lReferral Nt kNetworks to Maternal in Ethiopia...Referral Network Total Population in New Referral Network Population with Access to their Current Tier A Population
Page 34: MdliModeling Rf lReferral Nt kNetworks to Maternal in Ethiopia...Referral Network Total Population in New Referral Network Population with Access to their Current Tier A Population

Creating Catchment Areas

• An area that “should” be served by a facilityAn area that  should  be served by a facility• Cost Distance Raster in ArcGIS

– SlopeSlope– Rivers/terrain

• Once created, calculated populationOnce created, calculated population

Page 35: MdliModeling Rf lReferral Nt kNetworks to Maternal in Ethiopia...Referral Network Total Population in New Referral Network Population with Access to their Current Tier A Population
Page 36: MdliModeling Rf lReferral Nt kNetworks to Maternal in Ethiopia...Referral Network Total Population in New Referral Network Population with Access to their Current Tier A Population
Page 37: MdliModeling Rf lReferral Nt kNetworks to Maternal in Ethiopia...Referral Network Total Population in New Referral Network Population with Access to their Current Tier A Population
Page 38: MdliModeling Rf lReferral Nt kNetworks to Maternal in Ethiopia...Referral Network Total Population in New Referral Network Population with Access to their Current Tier A Population

Facility Catchment Areas with Road Network

Page 39: MdliModeling Rf lReferral Nt kNetworks to Maternal in Ethiopia...Referral Network Total Population in New Referral Network Population with Access to their Current Tier A Population

Referral Networks with Facility Catchment Areas

Referral Networks

Page 40: MdliModeling Rf lReferral Nt kNetworks to Maternal in Ethiopia...Referral Network Total Population in New Referral Network Population with Access to their Current Tier A Population

Population Density by Catchment Area

Population Density (sq km)

Page 41: MdliModeling Rf lReferral Nt kNetworks to Maternal in Ethiopia...Referral Network Total Population in New Referral Network Population with Access to their Current Tier A Population

Model building

• Build Catchment Areas• Model 0: Status Quo• Model 1: All facilities have motorized transportation 

and communication• Model 2: Upgrading strategic Tier B facilities to Tier A 

facilities

Page 42: MdliModeling Rf lReferral Nt kNetworks to Maternal in Ethiopia...Referral Network Total Population in New Referral Network Population with Access to their Current Tier A Population

M d l 0 STATUS QUO

Facility Catchment Areas Beyond 2-hour Travel Time

Model 0: STATUS QUO Total Population   

Population  Currentlywith Access

10,019,395 5,626,099 56.2%

Page 43: MdliModeling Rf lReferral Nt kNetworks to Maternal in Ethiopia...Referral Network Total Population in New Referral Network Population with Access to their Current Tier A Population

Model building

• Build Catchment AreasBuild Catchment Areas• Model 0: Status Quo• Model 1: All facilities have motorized transportation Model 1: All facilities have motorized transportation 

/communication• Model 2: Upgrading strategic Tier B facilities to Tier A Model 2: Upgrading strategic Tier B facilities to Tier A 

facilities

Page 44: MdliModeling Rf lReferral Nt kNetworks to Maternal in Ethiopia...Referral Network Total Population in New Referral Network Population with Access to their Current Tier A Population

M d l 0 STATUS QUO

Facility Catchment Areas Beyond 2-hour Travel Time

Model 0: STATUS QUO Total Population   

Population  Currentlywith Access

10,019,395 5,626,099 56.2%

Page 45: MdliModeling Rf lReferral Nt kNetworks to Maternal in Ethiopia...Referral Network Total Population in New Referral Network Population with Access to their Current Tier A Population

Facility Catchment Areas Beyond 2-hour Travel Time

Page 46: MdliModeling Rf lReferral Nt kNetworks to Maternal in Ethiopia...Referral Network Total Population in New Referral Network Population with Access to their Current Tier A Population

Model 1: Increase in population served by ensuring that every Tier B facility has a motorized vehicle 

Total  Population Currently  Population with  Overall Increase in Referral Network Population with Access* Access in Model 1 Population Access

1412 – Pawe 1,338,664  646,721  48.3% 646,721  48.3% ‐ 0.0%

1505 – Debre Tabor 2 723 313 2 182 465 80 1% 2 447 996 89 9% 265 531 9 8%1505  Debre Tabor  2,723,313  2,182,465  80.1% 2,447,996  89.9% 265,531  9.8%

1543 – Felege Hiwot 2,156,109  1,151,324  53.4% 1,994,981  92.5% 843,657  39.1%

1623 – Debre Markos 3,801,309  1,645,589  43.3% 2,660,041  70.0% 1,014,452  26.7%

TOTAL 10,019,395 5,626,099 56.2% 7,749,739 77.3% +2,123,640  21.2%

*Only for persons at the Tier B facility

Page 47: MdliModeling Rf lReferral Nt kNetworks to Maternal in Ethiopia...Referral Network Total Population in New Referral Network Population with Access to their Current Tier A Population

Facility institutional delivery rates

Institutional delivery rate

Page 48: MdliModeling Rf lReferral Nt kNetworks to Maternal in Ethiopia...Referral Network Total Population in New Referral Network Population with Access to their Current Tier A Population
Page 49: MdliModeling Rf lReferral Nt kNetworks to Maternal in Ethiopia...Referral Network Total Population in New Referral Network Population with Access to their Current Tier A Population

M d l 0 STATUS QUO

Facility Catchment Areas Beyond 2-hour Travel Time

Model 0: STATUS QUO

Page 50: MdliModeling Rf lReferral Nt kNetworks to Maternal in Ethiopia...Referral Network Total Population in New Referral Network Population with Access to their Current Tier A Population

Facility Catchment Areas Beyond 2-hour Travel Time

Page 51: MdliModeling Rf lReferral Nt kNetworks to Maternal in Ethiopia...Referral Network Total Population in New Referral Network Population with Access to their Current Tier A Population

Model 2: Increase in population served by upgrading Fenote Selam District Hospital to a Tier A 

Total Tier B Facilities in New Tier A 

Referral Network

Total Population in New Referral Network 

Population with Access to their Current Tier A

Population with Access to New Tier A

Overall Increase in Population with 

Access

16 1 945 023 293 895 15 1% 781 640 40 2% 487 745 25 1%16 1,945,023  293,895  15.1% 781,640  40.2% +487,745  25.1%

Page 52: MdliModeling Rf lReferral Nt kNetworks to Maternal in Ethiopia...Referral Network Total Population in New Referral Network Population with Access to their Current Tier A Population

Next Steps in Ethiopia

Further modelingg• Combinations of models (ex. Motorized vehicles plus upgrading)• Improved road surfaces• Optimal model: what is the best outcome with the lowest input • Optimal model: what is the best outcome with the lowest input 

(additional phones/vehicles)• Costing

Federal MoHWorking Group on Referral• Site selection for new ambulancesSite selection for new ambulances• Testing new management models• Transfer of methods and validation of assumptions