Lessons learned through implementing patienttracking systems in multiple settings in subSaharan Afri

1
Lessons learned through implementing
patient-tracking systemsin multiple settings in
sub-Saharan Africa Denis Nash, Peter Young, Batya
Elul, Matt Rosenthal, Cathy Maulsby, Deborah
Horowitz, Rosalind Carter, Jackie Kim, Monica
Katyal, David Hoos, Wafaa El-SadrInternational
Center for AIDS Care and Treatment Programs,
Columbia University, Mailman School of Public
Health, New York, NY USA
MONITORING EVALUATION
COMBINED ELETRONIC AND PAPER SYSTEMS
High-quality, comprehensive, and sustainable
monitoring and evaluation within and across HIV
care and treatment programs, with an aim towards
improving the quality of services, can play a key
role in the success of HIV care and treatment
program implementation.
Example of a longitudinal patient flow chart
FACILITY-LEVEL PATIENT TRACKING VS. PROGRAM
MONITORING AND EVALUATION

• Patient Tracking Collection of information on
  individual patients for clinical management and
  program planning, facilitating patient
  scheduling, care activities (both ART and
  non-ART), and follow-up at the facility-level.
• Fundamental component of care delivery that can
  and should be used for
• Program Monitoring and Evaluation Use of
  patient-level information and other relevant
  program data to monitor the quality, and evaluate
  the effectiveness, efficiency, equity, and
  acceptability of HIV service provision at the
  health facility, sub-national, national, and
  international levels.

PATIENT-LEVEL DATA TO BE CAPTURED
ELECTRONICALLY(should be kept to a minimum)
MODELS AND ATTTRIBUTES OF PATIENT-TRACKING
SYSTEMS IN HIV CARE AND TREATMENT PROGRAMS IN
RESOURCE-LIMITED SETTINGS

• Baseline/enrollment visit
• Unique ID
• Date of enrollment
• Age
• Sex
• WHO stage
• CD4 cell count
• OI diagnoses and co-morbidities
• ART history/status
• Pregnancy status
• Source of referral
• Cotrimoxizole use

• Follow-up visit
• Unique ID
• Date of visit
• WHO stage
• CD4 cell count
• OI diagnoses
• ART status
• ART regimen
• Pregnancy status
• Cotrimoxizole use

• Final disposition
• Unique ID
• Status and date
• Died Transferred Withdrew
  
• Lost to follow-up

• Paper-based systems (e.g., pre-ART and ART
  registers)
• Capture some data from patient medical records
  via hand transcription
• Data collection decentralized (e.g., registers
  completed at site)
• Capture up to 18 months of data on each patient
• Example WHO patient monitoring guidelines for
  HIV care and ART, Reed et al (South Africa)
• Systems that combine paper and electronic
  components
• Capture some/all data from patient medical
  records in database
• Data collection decentralized (e.g., forms and
  registers completed at site)
• Data entry can be centralized or decentralized
• Examples Siika et al (Western Kenya), Fraser et
  al (Haiti), Arakaki-Sanchez et al (Mozambique)

Supplemental information such as adherence,
non-clinical encounters, blood chemistry, side
effects, etc could also be collected, but would
require more resources (time)
PATIENT VOLUME, SITE VOLUME AND NUMBER OF
PATIENTS PER SITE AFTER 2 YEARS OF SCALE-UP
OBSERVATIONS AND ACTIVITIESOVER TWO YEARS OF
SCALE-UP
The ability of each site to track patients and
generate indicators varies substantially No
record systems initially Paper record systems
(ranging in quality) without electronic
components Paper record systems with electronic
components ME activities include Hiring
and training of clerical and data entry staff
On-site technical assistance Enhancement of
existing paper registers Computerization of key
information Quality assurance Capacity
building
LESSONS LEARNED
Average of 1069 patients per site

• Most HIV care and treatment programs supported
  by CU-ICAP will continue to rely primarily on
  paper-based systems due to lack of physical
  infrastructure
• Pre-ART and ART registers are not sustainable
  for large, mature programs
• In order to be most useful, key patient-level
  data for each visit recorded in paper-based
  systems must also be captured electronically on a
  routine basis (daily, weekly, monthly)
• An approach which retains the simplicity of
  these registers and captures only the minimum
  amount of longitudinal data necessary for patient
  tracking, reporting, and ME is ideal
• A simple, patient record-based approach, as
  opposed to a register-based approach would better
  facilitate computerization of key information
• Need a viable system of assigning unique patient
  identifiers
• Comprehensiveness and quality of data
  (completeness and accuracy) in the paper-based
  system must be assessed and maximized before
  attempting to computerize
• Data clerks dedicated to collecting, entering,
  reporting, and improving the quality of
  information are essential
• When available, invest time to capture
  retrospective data electronically using double
  key data entry to minimize errors
• It is useful to routinely capture site and
  programlevel attributes (e.g., urban vs. rural,
  provider patient ratio, new programmatic
  components)
• Computerization of key information from
  paper-based systems is feasible and sustainable

PRE-ART ART PAPER REGISTERS

• Observations
• Simple patient-driven registers (e.g., pre-ART
  and ART registers) have helped many new programs
  transition from registers that are episodic
  (capturing information on a single clinical
  encounter per patient) to longitudinal (capturing
  data from multiple clinical encounters per
  patient).
• Challenges
• Registers are often not well-maintained,
  incomplete, and inaccurate
• Patient medical records which serve as source
  document for registers can be rudimentary and
  incomplete
• Many of the data elements necessary for
  multipurpose use are missing
• Skilled staff (e.g., data clerks) to use and
  maintain systems are in short supply
• May not be sustainable in large, mature programs
• Difficult to computerize
• Requires substantial time and resources to use
  the data (hand tallying)
• Non-electronic registers limit flexibility in
  use of data

RECOMMENDATIONS

• A flexible, decentralized approach works best
• Flexibility should not compromise the ability to
  collect a minimum set of data elements across
  sites, or the ability to meet national and donor
  reporting requirements.
• Sustainable approaches require collection of
  program data that are useful to sites
• Responsibility for overseeing system shared by
  administrative and clinical staff
• Need staff dedicated to data capturing
• Before computerizing data, it is critical to
  assess and enhance data quality in the
  paper-based system
• Ongoing quality assurance mechanisms for both
  paper and electronic systems are also necessary
• Data security and confidentiality procedures
  should be in place prior to computerization
  Should be balanced with the need to
  efficiently track patient movement and transfers
  between care sites and services