Good ART clinic practices must include reliable ascertainment of outcomes of death, stopped treatment and loss to follow-up (attrition parameters), and the formal recording of transfer outs from one ART facility to another. Death may be ascertained pro-actively by a relative or friend providing information to the clinic, or it may be discovered as part of active tracing of patients who fail to attend their clinic appointment. In some countries, national death registers can also be used, if available. Similarly, patients who have stopped therapy may inform the clinic, but it is more likely that this information will be discovered through active tracing. Patient transfers need to be structurally managed, through formal recording, otherwise these patients may also be recorded as lost to follow-up. Thus, the ascertainment of true outcomes is a crucial programmatic activity.
In Malawi, as in many other African countries, patients are classified as lost to follow-up if they fail to attend clinic or obtain medications for 3 months or more. In an operational research study in Northern Malawi to determine the true outcomes of these patients, 50% had died, 15% had stopped therapy, and 8% had transferred to another clinic and failed to inform their original clinic of this move (Yu et al. 2007
). The results of this study were confirmed in a recent systematic review and meta-analysis (16 studies from sub-Saharan Africa), showing that 20% to 60% of patients who had been recorded as lost to follow-up had died (Brinkhof et al. 2009
Patients who are lost to follow-up can be identified through regular scrutiny of paper-based treatment cards or use of electronic data-base systems. Pharmacy-based records of medication collection by patients can also be used to monitor patient retention (Nachega et al. 2006
; Wood et al. 2008
). Electronic pharmacy records in particular may provide an efficient means of generating lists of patients who have failed to pick up medication and who can then potentially be traced in the community. For example, iDART is a computerized software that requires no licence and is freely available to download at URL http://www.cell-life.org/idart/download/
. This system allows easy management and follow-up of patients by pharmacists and programme management teams and has already been successfully used in seven large-scale ART programmes within four provinces in South Africa (Wood et al. 2008
Clinics will vary in terms of their ability to trace patients who are either late for their appointment or who have been recorded as lost to follow-up. Well-resourced clinics that are not overly burdened with high patient loads may have outreach teams that can trace patients and reduce their loss to follow-up rate. Poorly resourced clinics cannot do this, but correct recording of addresses or the location of homes within villages or towns, using community networks, and tapping in to the expanding cell phone networks (Lester et al. 2006
) should all be used to try and maintain patient contact. Where possible it is essential to conduct operational research to investigate and address the reasons for patient attrition, particularly deaths and loss to follow-up.