This is an archived document. The links are no longer being updated.
TB Notes 1, 2002
Updates from the Research And Evaluation Branch
Evaluating Contact Investigation Programs
The contact investigation (CI) is a key strategy in
the elimination of TB, and TB programs have established a variety
of structures and practices to use in conducting these investigations.
Recent studies conducted by DTBE staff and others have shown that
TB control programs vary widely in their success in achieving outcomes
related to identifying, screening, and appropriately treating contacts.
These findings demonstrate the need for some programs to revise
their contact investigation procedures. But what changes should
be made? What aspects of the CI program work well as is and what
areas need enhancement?
To answer these questions, local TB programs need
a way to evaluate their current program processes and practices
to appropriately plan and implement improvements and recommendations.
CDC has developed a framework for public health program evaluation.1
This well-researched framework provides an adaptable six-step process
for public health programs to use in evaluating their programs and
operations. However, for those unfamiliar with evaluation concepts,
understanding how to apply the framework and identifying appropriate
resources can be difficult and time-consuming.
To help local TB programs build capacity for the evaluation
of contact investigation processes, the Prevention Effectiveness
Section (PES) of DTBE's Research and Evaluation Branch (REB) initiated
a project that involves partnering with two health departments in
the use of the CDC evaluation framework and applying the evaluation
concepts to their programs. The ultimate goal of the project is
to develop a translated version of the framework specific for TB-program
managers and staff. Such a translated guide of the evaluation framework
will enable even those new to program evaluation to easily identify
and apply the six-step process to a self-evaluation of their contact
Over the past year, the two grantees, the Massachusetts
and New Jersey health departments, have been designing and conducting
their evaluations. In Massachusetts, a partnership between state
and local TB staff to facilitate CIs has been in place for 5 years.
Case managers at local health departments coordinate with state-level
case managers to ensure that the CI protocol is implemented. The
protocol includes use of the concentric circle approach to identify
contacts, time lines for accomplishing tasks, and systems for ensuring
proper exchange of information among state, local, and private providers.
The state aggregates and oversees CI data to ensure nothing falls
through gaps. State staff led the self-evaluation effort in five
local TB programs.
In Newark, the New Jersey health department selected
to self-evaluate the implementation and effectiveness of its specially
developed protocol for conducting expanded CIs in congregate facilities.
The protocol aims to ensure that all potentially exposed persons
are included in the investigation, while not depleting resources
or unnecessarily testing the unexposed. This protocol involves collaboration
between the investigator and supervisor to determine if such a CI
is necessary, a systematic environment assessment to determine likelihood
of transmission, and a communication / education plan to allay fears
and promote effective testing and treatment. The evaluation process
will also document details of the protocol to facilitate translation
to other programs.
While each program may report findings from their
evaluations at a later time, the evaluators learned much about the
evaluation process and how it can work for contact investigation
program activities. The lessons they have learned may help guide
other TB programs in understanding how evaluation can be applied
to their contact investigation activities to provide them with useful,
practical information for program enhancement. Lessons learned from
each of the six steps in the framework are presented below.
Step 1: When using the CDC framework, the first step
in the process is to identify "stakeholders." These are the people
who are interested in contact investigations and will ultimately
be affected by the evaluation. By identifying the stakeholders,
and engaging key stakeholders throughout the process, the evaluators
ensure that the evaluation is relevant and provides useful information.
There are three main categories of stakeholders for contact investigations:
1) the state and local health department staff who conduct contact
investigations and their managers and supervisors; 2) the persons
receiving services, i.e., TB patients or suspects, contacts of cases,
TB care providers and other health service providers, managers or
employees at schools, jails, congregate facilities, or private businesses,
and the general public; and 3) those who will directly use evaluation
findings to enhance the programs and improve services.
Step 2: Once the key players have been identified,
the next step in the process is to clearly and thoroughly describe
the contact investigation system and its relationship to the general
TB control program. This description ensures that all participants
share the same understanding about what contact investigations are,
how they are conducted, why they are needed, what program resources
are used in conducting these investigations, and what effects are
anticipated for each activity. One key feature of CDC's framework
is developing a logic model that graphically describes how the different
aspects of contact investigation link together to achieve the expected
results. The graphic serves as a communication tool as well as explicitly
showing how activities logically result in outcomes.
Step 3: From the description and logic model, the
evaluators should then be able to focus evaluation questions and
develop their data collection strategies. The evaluation question
selected in Massachusetts was, "How well does our new case or suspect
assessment tool work in five participating local TB control programs?"
The evaluators decided to conduct a process evaluation to assess
how well the new tool is being implemented by the five programs
and an outcome evaluation comparing patient-level outcomes collected
before the new tool was implemented and after its adoption. In New
Jersey, the evaluation question was, "How useful and effective is
our protocol for expanding contact investigations in congregate
settings?" The evaluation methods centered on comparing case studies
of investigations in which the protocol was fully implemented with
case studies of investigations in which the protocol was not implemented.
Outcomes, including those related to resource use, were also compared.
Step 4: Identifying indicators and selecting sources
for data should be a simple process, given the work done in the
early steps. The evaluators selected similar indicators, including
process indicators: program organization and infrastructure, staff
training, adherence to protocol steps, use of forms or records,
staff perceptions, intensity of resource use, and outcome indicators:
number of contacts identified, percentage of contacts fully tested,
percentage of contacts testing positive for disease or LTBI, percentage
of contacts started on treatment, and percentage of contacts completing
treatment. The data sources that were selected matched each indicator,
and included policies, practice instructions, training agendas and
records, case files or records of contact investigations, completed
forms or other written documentation, interviews with investigators
and supervisors, and aggregate reports or forms. All data sources
were assessed to ensure that they were reliable and considered valid
by the stakeholders before data collection began.
Step 5: Analyzing data and determining findings also
flow from the previous steps. Content analysis, domain analysis,
and other methods were used to analyze qualitative data and statistical
methods were applied to the quantitative data. To ensure that the
evaluation findings could be easily understood by all stakeholders,
simple data analysis methods were selected. With data in hand, the
evaluators once again went back to the logic model to compare the
data collected on implementation and results against original intentions
and levels of success defined previously. These facts were then
discussed and interpreted by the stakeholders to provide meaning
and context. Findings were developed that identified what objectives
and goals were met or unmet and why.
Step 6: The most important step is an on-going process
to ensure that the findings from the evaluation are used. Currently,
the findings are being used to
- assess contact investigation process and practice as it is
- target areas for enhancement or improvement,
- develop standardized tools for data collection during interviews
and field visits,
- develop strategies to make necessary changes to operations,
- prioritize program activities and resources,
- identify effective policies, procedures, or practices for replication,
- organize key information for training staff and informing those
outside the program,
- garner political support by demonstrating effectiveness of
- understand the implications of policy and guidelines on the
investigation process, and
- identify areas for future research and evaluation.
The last use completes the circular nature of the framework. As
information is obtained, more information is needed and more questions
emerge. The process is repeated and will become routine for the
TB programs. A learning cycle is established as programs strive
for continuous quality improvement in their contact investigations
and in other aspects of their operations. For more information on
the Framework for Program Evaluation in Public Health, please see
or contact Maureen Wilce at 404-639-8123.
Submitted by Maureen Wilce, MS
Division of TB Elimination
1. CDC. Framework for program evaluation in public health. MMWR