Download StudyGeneral

Study Overview

Health Impacts of Emergency Response and Post-Crash Medical Care in Malawi and Tanzania
Study is 3ie funded:
Study ID:
Initial Registration Date:
Last Update Date:
Study Status:
Change History for Status
Changed On Previous Value
10/19/2021 In Development
10/19/2021 In Development

Road traffic accidents (RTAs) are a rapidly growing health problem in developing countries, but evidence on the best interventions to reduce the burden of mortality and morbidity from RTAs in these settings is limited. The government of Malawi and Tanzania are implementing pilot programs aimed at reducing adverse health outcomes from RTAs through improved access to emergency medical services, most notably emergency ambulance services. We plan to evaluate the impact and effectiveness of these two pilot programs aimed at reducing adverse health outcomes from road traffic injuries (RTIs) through improved access to emergency medical services. Both pilots contain the following components 1.) training local community members to act as first responders on the scene; 2.) setting up an ambulance system with a centralized dispatch system; and 3.) improving trauma treatment capability and equipment in health facilities. We will use high frequency data and a difference-in-differences strategy to evaluate the impact of this program. The evaluation of this project will contribute to policy decisions to scale up the program in both countries.

Registration Citation:

Milusheva, S. and Kevin, 2019. Health Impacts of Emergency Response and Post-Crash Medical Care in Malawi and Tanzania. Registry for International Development for Impact Evaluations (RIDIE). Available at: 10.23846/ridie168

Health, Nutrition, and Population
Additional Keywords:
Road safety, Emergency medical services, Health systems
Secondary ID Number(s):

Principal Investigator(s)

Name of First PI:
Sveta Milusheva
World Bank
Name of Second PI:

Study Sponsor

World Bank
Study Sponsor Location:
United States

Research Partner

Name of Partner Institution:
Ministry of Health Malawi, Ministry of Health Tanzania, Malawi Roads Authority, Ministry of Works Transport and Communication Tanzania
Type of Organization:
Government agency (eg., statistics office, Ministry of Health)

Intervention Overview


The main intervention will be implemented along a section of major highways in Malawi (M1) and Tanzania (A7). In both cases the pilot intervention comprises of five components -- o Creation of an ambulance dispatch center and activation of an emergency access telephone number; o Training community first responders in villages adjacent to the highway; o Training paramedics, fire safety professionals, and drivers; and procurement and management of ambulances and EMS equipment; o Designing and rolling out an information campaign for using the emergency access telephone number; o Renovating and improving the physical trauma facility infrastructure , procuring equipment and consumables of laboratory equipment, and increasing capacity with refresher training on Advanced Trauma Life support. The beneficiaries are all those that experience a trauma and live along the section of each highway that receives the treatment or pass through that section while experiencing a trauma (especially road traffic crash).

Private Intervention Details:
Theory of Change:

The theory of change is that by improving care at the site of accidents, reducing the time from the occurrence of trauma to medical treatment at a hospital (via the ambulance system), and improving care en route to the hospital and at facilities, victims of trauma will receive better and more timely medical care and will ultimately have better health outcomes. Along this causal chain, there are several intermediate steps which we will seek to observe in the data. First, the time elapsed between the time when the trauma occurs and the patient receives medical care should decrease, given that ambulances with trained professionals will provide immediate care, and that an ambulance, rather than ad-hoc private transport, will transport patients to hospitals. Faster and higher quality care should then lead to better health status, both upon arrival in hospital, and over the medium term. Short run improvements in health conditions should be observable in the trauma evaluation scores and vital signs. Over the medium term, positive changes in health condition should translate into shorter in-patient stays, better prognosis upon discharge from hospitals, and decreased likelihood of readmission to health facilities with complications. In the long term, we should see lower mortality and lower morbidity. 

Multiple Treatment Arms Evaluated?

Implementing Agency

Name of Organization:
Ministry of Health Malawi and Ministry of Health Tanzania
Type of Organization:
Public Sector, e.g. Government Agency or Ministry

Program Funder

Name of Organization:
World Bank
Type of Organization:
Foreign or Multilateral Aid Agency

Intervention Timing

Intervention or Program Started at time of Registration?
Start Date:
End Date:
Evaluation Method

Evaluation Method Overview

Primary (or First) Evaluation Method:
Difference in difference/fixed effects
Other (not Listed) Method:
Additional Evaluation Method (If Any):
Other (not Listed) Method:

Method Details

Details of Evaluation Approach:

In this study, we will estimate intention-to-treat effects of the emergency trauma program on the treatment and health of trauma victims in Malawi and Tanzania. We will estimate treatment effects on both proximate outcomes (time elapsed between trauma and medical treatment, vital signs upon arrival in hospital, health status upon hospital discharge) and final outcomes (morbidity and mortality). The proximate outcomes are the main study endpoints. We will use a difference-in-difference model which includes high frequency data both before and after in treatment and control areas. The main data used will be trauma registry data collected on an on-going basis from treatment and control facilities, with data collection beginning prior to program implementation. We will use this data to generate a set of outcome variables and to compare cases in the treatment and control facilities before and after the rollout of the program, using the high frequency nature of the data (we can aggregate it at both weekly and monthly resolution) to control extensively for pre-trends. Standard errors will be clustered at facility level.

Private Details of Evaluation Approach:

For impact on health status as measured by the trauma registries in the selected facilities, the unit of analysis will be the patient. The specification below represents this model, with T_h for treatment, P_m a dummy for the period, and b_3 is the coefficient of interest. g_m are month fixed effects and e_(i,h,m) represents robust standard errors clustered at the facility level calculated using the Wild cluster-t bootstrap method, for patient i in hospital h in month m. Basic patient level demographic controls will be used as controls in alternative specifications. Y_(ihm)=b_0+b_1 T_h+b_2 P_m+b_3(T_h x P_m) +g_m+e_(ihm) Given trauma-related mortality is a rare event, we will not have enough statistical power to detect changes in mortality linked to the program in our 25 facilities. Therefore, we will also use nationwide data on RTAs and mortality to test for program impact on RTA-linked mortality. We will use the countries’ Health Monitoring Information System and police records on crashes. The specification will be the same as above except: a) outcomes are measured at the facility (for HMIS) or road segment (for police); b) both month and year FE can be controlled for.

Outcomes (Endpoints):

Primary outcomes: 1. Amount of time between occurrence of trauma and first contact with health-affiliated person (including first responder) 2. Amount of time between accident and initial care (provided by a trained nurse or clinician) 3. 0-4 scale of the overall condition of the patient upon arrival at facility 4. Trauma score on arrival at hospital (Glasgow Coma Score, Kampala Trauma Score, AVPU) 5. Basic vital signs on arrival in hospital: Body temperature, blood pressure, respiratory rate, pulse rate. 6. Basic vital signs 24 hours after entering hospital: Body temperature, blood pressure, respiratory rate, pulse rate. Secondary outcomes: 1. Mortality: percentage of RTA victims who are deceased within 1 month of accident

Unit of Analysis:
Unit of analysis will be an individual patient

1. The implementation of the program improves the medical care available to victims of trauma in Malawi and Tanzania, by reducing the time between incidence of trauma and access to medical treatment. 2. The implementation of the program leads to better health status upon arrival in hospital, as measured by trauma scores and vital signs, for victims of trauma in Malawi and Tanzania. 3. The implementation of the program leads to better health status upon discharge from hospital for victims of trauma in Malawi and Tanzania 4. The implementation of the program potentially reduces mortality for victims of trauma in Malawi and Tanzania 5. The implementation of the program changes the geographic composition of trauma patients in treatment and control hospitals 6. The implementation of the program increases facility utilization for non-trauma cases for which ambulances may also be used (such as emergency obstetric services, cerebrovascular disease/stroke, or out of hospital cardiac arrest)

Unit of Intervention or Assignment:
Unit of intervention is at the level of hospital.
Number of Clusters in Sample:
Since the unit of intervention is the hospital, we have a total of 25 clusters (11 hospitals in Malawi, 14 in Tanzania).
Number of Individuals in Sample:
88,500 observations
Size of Treatment, Control, or Comparison Subsamples:
12 clusters in control (3540 expected observations per cluster, 42480 total) 13 in treatment (expected 3540 observations per cluster, 46020 total)

Supplementary Files

Analysis Plan:
Other Documents:

Outcomes Data

Our main source of data collection will be through the trauma registries. This trauma registry data will provide high frequency information on the trauma cases coming to these health facilities, including the time elapsed between trauma occurrence and arrival in hospital and treatment, severity of trauma, vital signs upon entrance, number of days in the hospital, and outcome of the case upon departure from the hospital.
Data Already Collected?
Data Previously Used?
Data Access:
Data Obtained by the Study Researchers?
Data Approval Process:
Approval Status:

Treatment Assignment Data

Participation or Assignment Information:
The data collected from trauma registries will enable us to confirm the treatment assignment, since the assignment is done at the level of the facility and registries will record whether project ambulances are operating in treatment facilities. Additionally, within the trauma registry there are detailed questions regarding where the trauma occurred which we can also use to ensure that the case coming is within the geographic area of the treatment.
Data Obtained by the Study Researchers?
Data Previously Used?
Data Access:
Data Obtained by the Study Researchers?
Data Approval Process:
Approval Status:

Data Analysis

Data Analysis Status:

Study Materials

Upload Study Materials:

Registration Category

Registration Category:
Prospective, Category 1: Data for measuring impacts have not been collected

Completion Overview

Intervention Completion Date:
Data Collection Completion Date:
Unit of Analysis:
Clusters in Final Sample:
Total Observations in Final Sample:
Size of Treatment, Control, or Comparison Subsamples:


Preliminary Report:
Preliminary Report URL:
Summary of Findings:
Paper Summary:
Paper Citation:

Data Availability

Data Availability (Primary Data):
Date of Data Availability:
Data URL or Contact:
Access procedure:

Other Materials

Survey Instrument Links or Contact:
Program Files:
Program Files Links or Contact:
External Link:
External Link Description:
Description of Changes:

Study Stopped


This study was stopped because the main part of the intervention in Tanzania did not move forward (the centralized dispatch system), and in Malawi, the centralized dispatch pilot faced many challenges along the way that prevented its full implementation. This has made it not possible to conduct an impact evaluation since the interventions were not fully implemented.