Understanding health worker data use in a South African antiretroviral therapy register

OBJECTIVE

To evaluate how electronic data management systems affect data use practices in antiretroviral therapy (ART) programs within local health districts, and individual health facilities.

METHODS

We used a data quality audit to establish a baseline of the quality of data in the electronic register alongside in-depth interviews with health workers and managers, to understand perceptions of data quality, data use by facility staff and challenges affecting data use.

RESULTS

The findings provide a four-level continuum of data use that can be applied to other settings and recommendations for optimising facility-level data use.

CONCLUSION

By defining four levels of data use our findings suggest the potential to encourage a structured process of moving from passive data use, to more active and engaged data use, where data could be used to anticipate patient behaviour and link that behaviour to differentiated care plans.

A three-tier framework for monitoring antiretroviral therapy in high HIV burden settings

The provision of antiretroviral therapy (ART) in low and middle-income countries is a chronic disease intervention of unprecedented magnitude and is the dominant health systems challenge for high-burden countries, many of which rank among the poorest in the world. Substantial external investment, together with the requirement for service evolution to adapt to changing needs, including the constant shift to earlier ART initiation, makes outcome monitoring and reporting particularly important. However, there is growing concern at the inability of many high-burden countries to report on the outcomes of patients who have been in care for various durations, or even the number of patients in care at a particular point in time. In many instances, countries can only report on the number of patients ever started on ART. Despite paper register systems coming under increasing strain, the evolution from paper directly to complex electronic medical record solutions is not viable in many contexts. Implementing a bridging solution, such as a simple offline electronic version of the paper register, can be a pragmatic alternative. This paper describes and recommends a three-tiered monitoring approach in low- and middle-income countries based on the experience implementing such a system in the Western Cape province of South Africa. A three-tier approach allows Ministries of Health to strategically implement one of the tiers in each facility offering ART services. Each tier produces the same nationally required monthly enrolment and quarterly cohort reports so that outputs from the three tiers can be aggregated into a single database at any level of the health system. The choice of tier is based on context and resources at the time of implementation. As resources and infrastructure improve, more facilities will transition to the next highest and more technologically sophisticated tier. Implementing a three-tier monitoring system at country level for pre-antiretroviral wellness, ART, tuberculosis and mother and child health services can be an efficient approach to ensuring system-wide harmonization and accurate monitoring of services, including long term retention in care, during the scale-up of electronic monitoring solutions.

Incomplete registration and reporting of culture-confirmed childhood tuberculosis diagnosed in hospital

SETTING

The South African National Tuberculosis Programme (NTP) recommends the registration of tuberculosis (TB) patients at community clinics. TB in children is often diagnosed at referral hospitals, and there are concerns as to whether these children are accurately reflected in routine NTP reporting.

OBJECTIVE

To assess the completeness of registration of children with culture-confirmed TB diagnosed in a referral hospital, in the routine provincial electronic TB register (ETR.Net), and to describe TB treatment outcomes.

DESIGN

A retrospective cohort study including children aged <13 years diagnosed with culture-confirmed TB at Tygerberg Children's Hospital from July 2007 to June 2009. Data on demographic, clinical and referral factors were collated from hospital data sources. Electronic matching was used to identify children in the provincial ETR.Net.

RESULTS

Only 166 of 267 (62%) children were registered in ETR.Net. Children with TB meningitis and death prior to referral were significantly less likely to be registered. Treatment outcome data were available for only 70% of children; favourable outcomes were reported in 56%.

CONCLUSIONS

A large proportion of children diagnosed with confirmed TB at a referral hospital were not registered, resulting in underreporting of the burden and severity of childhood TB. Routine surveillance of childhood TB should include linkage of hospital data.