One in five patients at London clinic are lost to follow-up and do not attend elsewhere in the UK

This article is more than 16 years old. Click here for more recent articles on this topic

The numbers of patients at a busy south London HIV clinic who do not return for appointments or to collect prescriptions is comparable to rates reported in sub-Saharan Africa, according to a study presented by Sarah Gerver at the International AIDS Conference in Mexico City on August 5th. Four in ten patients are lost to follow-up, and more than half of these patients could not be traced to other HIV clinics in the UK.

Regular contact with medical providers is important to allow for monitoring of CD4 cell count and viral load, ongoing prescription of HIV treatment, identification of other health needs and discussion of HIV prevention issues. However the phenomenon of “loss to follow-up” (patients who do not attend clinic and cannot be contacted) is poorly studied in Europe and North America.

In the study led by Professor Philippa Easterbrook, researchers from King’s College Hospital in London and the Health Protection Agency looked at all adult patients who had attended the HIV clinic at King’s College Hospital at least once between 1995 and 2005. Those who had not been seen by the clinic at any time during 2006 were considered lost to follow-up (LFU).


loss to follow up

In a research study, participants who drop out before the end of the study. In routine clinical care, patients who do not attend medical appointments and who cannot be contacted.

adjusted odds ratio (AOR)

Comparing one group with another, expresses differences in the odds of something happening. An odds ratio above 1 means something is more likely to happen in the group of interest; an odds ratio below 1 means it is less likely to happen. Similar to ‘relative risk’. 

detectable viral load

When viral load is detectable, this indicates that HIV is replicating in the body. If the person is taking HIV treatment but their viral load is detectable, the treatment is not working properly. There may still be a risk of HIV transmission to sexual partners.

multivariate analysis

An extension of multivariable analysis that is used to model two or more outcomes at the same time.


A healthcare professional’s recommendation that a person sees another medical specialist or service.

Information on those individuals was then compared with national databases of patients receiving HIV care in the UK (SOPHID and an equivalent Scottish database), in order to identify people who had transferred their care to a different HIV clinic. UK death registers were also searched in order to identify people who had died. However there was no systematic attempt to contact patients by phone or by mail, because many patients have worries around confidentiality and request not to be contacted.

There had been 2070 patients registered at the clinic during the ten-year period. Almost 60% were black African or black Caribbean, and 30% were gay or bisexual men.

A total of 836 people (40%) were lost to follow-up. Just under half of these were attending other UK clinics, leaving 432 people who were lost to follow-up and not attending any UK clinic (referred to as UK-LFU). Those identified as UK-LFU were 52% of those lost to follow-up, or 21% of all patients.

Death registers showed that just 18 of those UK-LFU were known to have died in the UK.

Those UK-LFU tended to have been attending the clinic for a relatively short period of time (median 4 months), and this was shorter for those not on treatment. Those who were UK-LFU also tended to be younger.

One third of those UK-LFU were taking antiretroviral drugs, and their interruption of treatment must be of particular concern. Looking specifically at the factors associated with being UK-LFU and on treatment, it was more common in those who were not doing well on treatment and may have had difficulties with adherence. The following figures are all drawn from multivariate analysis. Being UK-LFU was four times more likely for people who had a detectable viral load compared to those who were undetectable (Adjusted odds ratio (AOR): 4.06), and was almost twice as likely in those with a CD4 cell count below 200 cells/mm3, compared to those above 350 cells/mm3 (AOR: 1.84).

Looking at those who were not on treatment, being UK-LFU was also more common in those with a CD4 cell count below 200 cells/mm3 (AOR: 3.59).

The risk of being UK-LFU varied strongly according to population groups. White gay and bisexual men were less likely to be UK-LFU, and they form the comparison group for the following figures. Looking first at those who were UK-LFU and on treatment, it was much more likely to occur in African heterosexual women (AOR: 2.30). For those not on treatment, it was more likely to be African heterosexual men (AOR: 4.36) and Caribbean heterosexual men (AOR: 2.64).

There were relatively few intravenous drug users in the study population. While their attendance at appointments did tend to be unreliable, LFU rates were not particularly high.

When available, the medical notes for those UK-LFU were reviewed to see if there was any information that could explain why they subsequently dropped out of view. Notes were available for almost three-quarters of patients who were identified as UK-LFU, and a potential reason was identified for 71% of these patients. Where a reason was identified, it was established that just over half of the group may have left the UK, having either mentioned plans to leave, or having reported immigration difficulties. It was reported that one in ten patients in this group had not been seen after initial diagnosis, and 3% were reported to be in denial of their status.

However Sarah Gerver did note that studies of this type are subject to methodological difficulties, in particular errors in the way identifying information is recorded in the national databases and the lack of data on people leaving the UK. Moreover, given the stigma surrounding HIV in different communities, some patients may have chosen to register at different clinics with different names. All of these factors may mean that her figures over-estimate the numbers lost to follow-up.

But she pointed out that the rates of loss to follow up in this clinic with a high proportion of African patients are comparable with those seen in some clinics in sub-Saharan Africa, where four in ten are LFU or known to have died. She concluded that her findings highlight the need to better understand the health-seeking behaviours of patients, and to develop strategies to avoid loss to follow-up happening.

The researchers suggest some changes to clinic procedures in order to limit LFU:

  • Regular updating of contact details by clinic staff
  • More frequent tracking of patients who are LFU
  • Formally recording transfers of patients to other clinics
  • More intensive monitoring and support of patients in their first six months at the clinic, including more counselling, peer support and linkages to other services.