GF-Operations-4 11/5/07 10:58 AM Page 2

GUIDE TO OPERATIONAL
RESEARCH IN PROGRAMS
SUPPORTED BY THE
GLOBAL FUND
GF-Operations-4 11/5/07 10:58 AM Page 3

What is operational research?

Operational research provides decision-makers Examples of research topics in
with information to enable them to improve the per- Global Fund-supported HIV,
formance of their programs. Operational research tuberculosis and malaria programs
helps to identify solutions to problems that limit
program quality, efficiency and effectiveness, or to • Systems for delivering antiretroviral treatment (ART);
determine which alternative service delivery strate- identifying needs and demand for ART and barriers to
treatment seeking among different population groups.
gy would yield the best outcomes. In simple terms,
Tanzania Round 4 HIV grant: www.theglobalfund.org/
it is described as “the science of better”. search/docs/4TNZH_824_0_full.pdf

Operational research focuses on factors which are • Integrating HIV and tuberculosis control. Ethiopia Round
under the control of programs. It seeks to improve 6 tuberculosis grant: www.theglobalfund.org/search/
the number and quality of services and program docs/6ETHT_1317_0_full.pdf

outputs and outcomes by optimizing program inputs

• Rapid diagnostic tests vs. microscopy for malaria in
(e.g., personnel, supplies) and processes (e.g., train- peripheral health facilities. Afghanistan Round 5
ing, supervision, promotion of services). Operational malaria grant: www.theglobalfund.org/search/docs/
research can also determine cost-effective and sus- 5AFGM_942_0_full.pdf
tainable ways to build service delivery capacity, test
financing alternatives and make advocacy and com-
munication strategies and tools more effective.
For example, a study to increase condom use among
patients on antiretroviral (ARV) treatment might
How to include operational research in a
experiment with changes in provider training or
proposal to the Global Fund?
client counseling and measure the impact on the
number of condoms distributed or frequency of Operational research should be described as an activity
consistent condom use. (with the associated budget) in the proposal form, with a
focus on explaining the link to program outcomes. Look
Why is operational research important? for further guidance in the Round 8 guidelines, available
as of 1 March 2008 on the Global Fund website.
The Global Fund believes that operational research
has an important role to play in the success of Further information is available in the Monitoring and
the programs it funds, and strongly encourages Evaluation Toolkit available at: www.theglobalfund.org/
proposals with an operational research component. en/apply/call7/documents/me/ and under Item 7 of
Everyone writing a Global Fund application, and www.theglobalfund.org/en/apply/call7/documents/
anyone concerned with improving their program’s technical/

performance, should think about whether opera-

tional research should be built into the application.
“Learning by doing” is an essential element of
improving program performance. Global Fund-
supported programs are recommended to spend
five to ten percent of their grant budget on
monitoring and evaluation (M&E), which can include
spending on relevant operational research.
research capacity and ability to learn from data
Operational research complements M&E and pro- and implementation, effectively serving as an
vides both the host program and the Global Fund investment in future program efficiency and
with solid information about which interventions performance. In Global Fund grants, operational
and service delivery models work (or do not work) in research components must be aligned with
HIV, TB, and malaria programs. Global Fund sup- national research priorities, matching the
port can also help build a program’s overall country’s evolving disease-control policies.
GF-Operations-4 11/5/07 10:58 AM Page 4

Who does operational research? How to do operational research

Operational research is commonly carried out by Either qualitative or quantitative methods can be
any health-care provider, including the public sector, used. Qualitative methods include focus group discus-
nongovernmental organizations (NGOs), and the sions or individual interviews with service providers or
for-profit sector as a means of identifying and clients or observational studies, e.g., observing health
solving problems in hospitals, health centers, and care workers. Quantitative methods include structured
community programs. Most studies need not be questionnaires or the analysis of service statistics.
elaborate, and they can be conducted within Formal epidemiological studies using qualitative or
almost any program. quantitative methods are commonly performed.

Operational research is most successful when it The research often follows a well-defined process:
is carried out by a team of program implementers
and researchers who work closely together during 1. Identification of the program problem
every stage of the research, from the identification Identifying the problem is the most critical step
of the problem to the dissemination and utilization in the operational research process. Unless a
of results. The role of the program implementers problem is clearly defined it is impossible to
who know the program and its clients, includes: develop good solutions.
(1) taking the lead in defining the program problem,
(2) specifying when the study result is needed for 2. Identification of possible reasons and solutions
decision-making, (3) ensuring that providers and and the testing of potential solutions
facilities cooperate with the researchers, (4) utilizing Research is required when either the reason for the
the findings in program decision-making and problem or the solution to the problem is not self-evi-
(5) disseminating research findings. Researchers are dent. Once the problem has been identified, it is the
responsible for translating the program problem into job of the program implementer and researcher to
a researchable problem and for the quality of the determine the reasons for the problem and to gener-
research. It could also be useful to involve stakehold- ate possible solutions. Sources for thinking of possi-
ers as advisors throughout the operational research ble reasons and solutions include program staff and
process, for example representatives from civil clients, community members and literature on the
society groups, NGOs, affected communities, the topic. Good solutions must be measurable, easy to
government and technical assistance agencies. implement and sustainable.
GF-Operations-4 11/5/07 10:58 AM Page 5

In some cases, operational research is used

to determine the effectiveness of a proposed Many operational research projects follow a
solution through comparison methods in an pattern similar to the one shown here.
intervention study.

Designs used for this are:

• quasi-experimental design: one where the situ-
1
ation is analyzed before and after introducing Consider
9
the possible solution to the service delivery further Identify
ways of problem 2
unit, or between service delivery units with and improving

without the proposed solution; and Monitor Identify

changes in possible
• true experiment: one where service delivery, the revised reasons
8 programs
or health outcomes, are compared between
3
randomly assigned “experimental” and “control”
Identify
Disseminate
groups. results
possible
solutions

3. Results utilization 7
Research
Act on protocol
Prior to beginning the study, it is necessary to findings of to test 4
research
decide how its results are meant to be used. Conduct
solution(s)

This determines, to some extent, the informa- 6

research
5
tion that will be collected. For instance, a study
might find that one way of delivering insecticide-
treated bed nets is superior to another way in
getting households to buy and use the nets, but
the decision-maker will (and should) want to
know, for example, if the new strategy is more Operational research is successful
expensive; if information about cost was not col- if its findings are used in making a
lected, a good decision cannot be made. program decision
• Let the program identify the problem to be solved or the
4. Results dissemination decision to be made
Early in the operational research process, a • Know who will make the decision and when
strategy must be developed for disseminating
• Include a results utilization and dissemination plan in the
results to stakeholders. This dissemination could
research proposal and budget
take the form of a seminar if there are multiple
• Identify program stakeholders, their information needs,
decision-makers or stakeholders, or in a meeting
and the best way of providing information to each.
in the decision-maker’s office. Researchers
should also assist the decision-makers and
stakeholders in devising ways to widely imple-
ment the decisions made. Informing the CCM will
facilitate the process of including any lessons
learned in any country request for continued
funding (or grant reprogramming, if relevant).
GF-Operations-4 11/5/07 10:58 AM Page 6

PRACTICAL EXAMPLES OF OPERATIONAL RESEARCH

EXAMPLE 1 EXAMPLE 3

Determining target locations and populations Comparing alternative approaches to malaria

for delivering HIV prevention services treatment regimen adherence
In Kawempe district of Uganda, the AIDS Control Project In Uganda, a new first-line malaria treatment option
conducted a study to identify areas of potentially high was needed, and the program considered switching to
HIV transmission in order to efficiently target prevention artemether-lumefantrine (A-L). Since the new therapy
efforts. The study used the “PLACE methodology 1”, which requires a relatively complicated three-day course of
identifies locations where individuals go to meet new sex- treatment, there was concern that treatment effective-
ual partners. Information is collected on the locations and ness could be compromised by poor adherence. One
their users and on factors that could facilitate or hinder option to maximize adherence was to do in-hospital
program activities in the venues. supervised therapy, but this would be expensive for the
program and complicated for patients, which could limit
Trained interviewers talked to taxi drivers, youth and utilization of the drug. A randomized trial comparing
police who identified 255 meeting places, mostly bars treatment outcomes of patients receiving supervised and
and nightclubs. They visited the locations, talked to own- unsupervised treatment was conducted in the Mbarara
ers and employees and recorded the number of persons University Hospital. A total of 957 malaria patients were
visiting the sites at peak business hours. Interviews with randomly assigned to either supervised or unsupervised
customers confirmed that patrons visited the locations to treatment. Supervised treatment consisted of hospitaliz-
meet new partners. Most patrons were young and had not ing patients for three days and closely supervising treat-
used a condom with their last partner. Only 11 percent of ment. The unsupervised group was counseled on how to
sites had an AIDS prevention poster, and only 20 percent use the drug at home. Twenty-eight day cure rates were
had condoms available. Information used by the program 97.7 percent for the supervised group, and 98.0 percent
included the specific locations where people gathered, for the unsupervised group, indicating that an A-L
the number of people who gathered at those locations, regimen could be based on unsupervised treatment.3
and times when the sites were most crowded. Additionally,
managers who were willing to have AIDS prevention
activities in their establishment and those willing to sell
condoms were identified.

EXAMPLE 2

A before-and-after study to determine the

effectiveness of community health workers
Tuberculosis treatment at the Malteser hospital in
Sudan was disrupted by ethnic conflict. To increase
treatment access, the hospital tested the effectiveness
1. MEASURE Evaluation. 2003. “PLACE in Uganda: Monitoring
of community health workers (CHWs) who were given
AIDS-Prevention Programs in Kampala Uganda Using the PLACE
training in providing tuberculosis care. Before the inter- Method,” Carolina Population Center, University of North Carolina
vention, tuberculosis patients spent six months in the at Chapel Hill, Chapel Hill NC.

hospital until they completed a full course of directly- 2. Hughes S. 2006. “Community based tuberculosis-control pro-
observed treatment (DOTS). After intervention, patients gram in Yei county: experience in a war situation in southern
Sudan.” Operational Research in Tropical and Other Communicable
were discharged at two months and referred to CHWs for Diseases Final Report of Summaries 2003-2004 implemented
continued treatment. Patients treated in-hospital in during 2004-2006. Results portfolio 3. World Health
Organization, Regional Office for the Eastern Mediterranean.
2004 were compared with patients who received CHW
care in 2005. The respective cure rates were 76 percent 3. Piola P Fogg C, Bajunirwe F et al. “Supervised vs. unsupervised
intake of six-dose artemether –lumefantrine for treatment of
and 82 percent. Researchers recommended rapid acute, uncomplicated Plasmodium falciparum malaria in Mbarara,
expansion of the CHW program.2 Uganda: a randomized trial. Lancet 2005;365:1467-73.
GF-Operations-4 11/5/07 10:58 AM Page 1

F U R T H E R I N F O R M AT I O N

Scano F et al.. TB/HIV research priorities in resource-

limited settings. Report of an expert consultation. 2005,
World Health Organization, Stop TB Department: Geneva.
WHO/HTM/TB/2005.355, WH/HIV/2005.03,
whqlibdoc.who.int/hq/2005/WHO_HTM_TB_2005.355.pdf

WHO, Dept of HIV/AIDS and Special Programme for Research and

Training in Tropical Diseases (WHO/TDR). Operational research to
support HIV treatment and prevention in resource-limited settings –
Summary of activities, July 2004 to January 2006.
www.who.int/hiv/topics/or/en/index.html

Special Programme for Research and Training in Tropical Diseases

(TDR), Report of the Scientific Working Group on Malaria. 2003,
World Health Organization: Geneva. TDR/SWG/03.
www.who.int/tdr/publications/publications/pdf/malaria_swg.pdf

Special Programme for Research and Training in Tropical Diseases

(TDR). Report of the Scientific Working Group meeting on
Tuberculosis. 2005, Geneva, Switzerland. TDR/SWG/06.
www.who.int/tdr/publications/publications/pdf/swg_tub.pdf

Fisher AA, Foreit JR. Designing HIV/AIDS Intervention Studies:

An Operational Research Handbook. Population Council 2002.
www.popcouncil.org/horizons/orhivaidshndbk.html

WHO, Stop TB Dept. “Preparing TB proposals for the Global

Fund to fight AIDS, Tuberculosis and Malaria) (2007 edition)
at index no. 6.1.
www.who.int/tb/dots/planningframeworks/en/

L I N KS TO PA RT N E R S F O R T E C H N I CA L
A S S I S TA N C E A N D A D V I C E

WHO Special Programme for Research and Training in

Tropical Diseases (TDR)
www.who.int/tdr/topics/ir/default.htm

WHO Stop TB Department

www.who.int/tb/en

WHO HIV/AIDS Department

www.who.int/hiv/en

WHO Global Malaria Programme

www.who.int/malaria/en

This guide was written by Jim Foreit (Population Council)

and George Schmid (WHO, Dept of HIV/AIDS), with input
from Eline Korenromp, Daniel Low-Beer and Serge Xueref
(Global Fund to Fight AIDS, Tuberculosis and Malaria)

Cover Photo: The Global Fund / John Rae

Inside Photo: The Global Fund / Oliver O’Hanlon