This Community is Archived.

While this community is no longer active, we invite you to review and recommend past posts and resources. Membership for this community is closed, but we hope you'll join us in one of the many other communities on GHDonline.

Moderators of Adherence & Retention and GHDonline staff

You may use this brief for informational, non-commercial purposes with credit attribution: The Global Health Delivery Project,, Dec 02, 2010. Please see our Terms of Use for more information.

Lost To Follow Up as part of Adherence Denominator for HIV Care

Added on 02 Dec 2010
Last updated on 30 Aug 2013

Authors: Megan McLaughlin, Mischa Shattuck; Reviewed by Sophie Beauvais, David Bangsberg, MD, MPH

Frequently used to describe patients who stop coming to appointments and cannot be located, loss-to-follow-up (LTFU) is a critical issue for HIV care. Data from a diverse range of HIV treatment programs in resource limited settings show LTFU rates ranging from 5% to 40% within 6 month of antiretroviral therapy (ART) initiation, with extensive clinical impact and associated costs (Losina 2009).

In this discussion, members debate how to define LTFU in program evaluations and analyses of adherence. Practitioners and researchers share their experiences in the field and key references on LTFU, patient retention, and adherence.

Key Points

  • Moderator David Bangsberg, MD, offered several definitions to help distinguish among related concepts that are sometimes conflated in the literature:
    • Treatment access – It is important to distinguish between incomplete adherence and incomplete access, which is often related to logistical, economic, or structural barriers (such as pharmacy stock-outs).
    • Treatment adherence – The most common definition is the number of doses taken divided by number of doses prescribed over a defined period of time.
    • LTFU or visit adherence – Often defined as percent of scheduled visits that are attended, LTFU may be related to economic/structural barriers and less commonly behavioral-social factors.
    • Treatment failure – Treatment failure, which is commonly defined as virologic failure, may be related to incomplete adherence, absorption or metabolism of medication, or possibly overwhelming disease.
  • It may be useful to differentiate between visit adherence (percentage of visits attended) and true LTFU (dropping entirely out of care). Researchers are studying the extent to which poor visit adherence predicts LTFU.
  • Several members discussed the importance of using community health workers to track LTFU patients – both as a means of improving our understanding of this population and as a way of re-engaging these patients to prevent treatment non-compliance.
  • In tuberculosis care, patients are commonly classified as cured, failed, died, or defaulted; an increased number of defaulted patients results in a lower percentage of cured patients.
  • Some members argued that “defaulting” in rural HIV treatment programs does not necessarily equate bad outcomes. For example, a program found that 62% of patients LTFU were alive, and of those 83% had transferred to another clinic, often when a clinic closer to their home started providing ART as a result of successful treatment scale-up. Thus, LTFU can represent either program success or program failure.
  • Members raised a related question – how should patient transfers be handled in analysis? Transfers could be treated as censored data, or they could be classified as treatment failures, depending on their reason for transferring.
  • One member suggested performing sensitivity analyses, varying the estimated proportion of deaths among LTFUs in the sample and observing the effects on results.

Key References

Enrich the GHDonline Knowledge Base
Please consider replying to this discussion with the following information

  • How does your program define LTFU and/or visit adherence?
  • How do you treat LTFU and/or visit adherence in program evaluations and analyses?
  • Do you have measures/interventions in place to re-engage with patients LTFU and do you see any differences in outcomes?

Download: 12_02_10_LTFU_adherence_denominator_.pdf (43.7 KB)