Article Text

Short report
Comparison of two cohorts of patients presenting with AIDS: patients with previously known HIV diagnoses and true late presenters
  1. MJ Lee1,
  2. M Rayment2,
  3. A Scourfield2,
  4. B Gazzard2
  1. 1West Middlesex University Hospital, Isleworth, UK
  2. 2Directorate of HIV/GU Medicine, Chelsea and Westminster Hospital NHS Foundation Trust, London, UK
  1. Correspondence to Dr Michael Rayment, Chelsea and Westminster Hospital NHS Foundation Trust, 369 Fulham Rd, London SW10 9NH, UK; michaelrayment{at}


Objectives We aimed to identify the factors associated with developing AIDS 6 months or more after an HIV diagnosis, and to examine how post-HIV diagnosis AIDS (PHDA) patients differed from true late presenters (HIV diagnosed concurrent with the first AIDS presenting event) in their demographics and comorbidities.

Methods A retrospective analysis was undertaken of all inpatients admitted to a large HIV unit presenting with the following AIDS-defining infections: cryptococcal meningitis, cerebral toxoplasmosis or Pneumocystis jirovecii pneumonia between 1 January 2005 and 31 December 2010.

Results 114 HIV-positive patients presented with AIDS-defining infections. Compared with late presenters, PHDA patients had a larger proportion of migrants and visitors (53.7% vs 34.0%, p=0.047), were more likely to inject drugs (9.3% vs 0.0%, p=0.032), had more previous HIV-associated diseases (57.4% vs 12.8%, p=0.000), psychiatric comorbidities (35.2% vs 12.8%, p=0.009), rates of alcohol abuse (24.1% vs 4.3%, p=0.005) and reported social issues (25.9% vs 0.00%, p=0.000). 88.9% of PHDA patients were lost to follow-up for a period of at least 4 months since diagnosis. Common reasons for clinic non-attendance included travel, social issues, transfer of care and treatment avoidance. Common reasons for antiretroviral treatment breaks included drug side effects, negative beliefs about medication, incompatible lifestyles and social issues.

Conclusions Compared with late presenters, PHDA patients demonstrate clear demographical differences including higher rates of psychiatric comorbidities, social issues, alcohol and substance abuse. Many PHDA patients default follow-up. The retention of HIV patients in care and on treatment must be addressed by clinicians to prevent avoidable morbidity.

  • AIDS
  • HIV

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


Since the introduction of combined antiretroviral therapy (ART), the profile of hospital inpatients with HIV infections has evolved significantly. There is an increasing burden of cardiovascular, liver and metabolic diseases among these patients.1 ,2 The number of deaths due to AIDS is decreasing;3 however, there is still a population of HIV patients who continue to present with AIDS-defining illnesses. These patients can be broadly classified into two groups: (1) late presenters—patients who were diagnosed with HIV concurrent with their AIDS-defining disease and (2) patients with known HIV diagnoses who continue to progress despite previous medical follow-up (hereafter post-HIV diagnosis AIDS or ‘PHDA’ patients).

Our centre has previously reported that a substantial proportion of all AIDS diagnoses (62%) occur in this PHDA group (presenting with an AIDS event 6 months or more after HIV diagnosis). Sixty percent of these patients had previously commenced ART.4 It is unclear why these patients continue to present with AIDS. There is an element of missed opportunity for earlier clinical involvement for each group, with PHDA patients developing AIDS due to a failure of follow-up care and treatment; while late presenters develop AIDS due to a failure of timely HIV diagnosis. Risk factors associated with late presentation have been previously reported and include increased age at diagnosis, injecting drug use, heterosexuality in men and South American or African ethnicity.5 However, there are limited data regarding factors associated with progression to AIDS in PHDA patients.

In this paper, we aim to identify the factors associated with developing AIDS 6 months or more after HIV diagnosis, and to compare how the PHDA population differed from true late presenters in their demographics and comorbidities.


We performed a retrospective analysis of all individuals presenting to a large HIV inpatient service with the following late stage, AIDS-defining opportunistic infections (OIs): cryptococcal meningitis (CM), cerebral toxoplasmosis or Pneumocystis jirovecii pneumonia (PCP) between 1 January 2005 and 31 December 2010. The selection process of these patients via electronic clinical codes in our institute was previously described.4

Electronic case notes were reviewed to identify patient demographics. Any concurrent or previous diagnoses of HIV clinical indicator diseases (CIDs), previously diagnosed psychological comorbidities, alcohol or drug abuse and social issues were recorded.

For PHDA patients, history of ART use and clinical follow-up were also noted. The number of viral load measurements was used as a surrogate marker for clinic attendance. Lost to follow-up (LTFU) periods were defined as clinic non-attendance greater than 4 months. ART treatment breaks were also reviewed. Reasons for the longest LTFU period and ART breaks were obtained from electronic and case records.

Descriptive analysis was performed using the SPSS software (V.19.0; SPSS Inc. Chicago, Illinois, USA). In the univariate analysis, the Mann–Whitney U test was used for continuous variables, and the Pearson χ2 test was used for the categorical variables.


From a cohort of more than 6000 patients followed up at our centre, 114 HIV-positive patients were admitted with severe OIs in the study time frame: 8 cases of CM, 7 cases of toxoplasmosis and 99 cases of PCP. A total of 13 patients had missing data and were excluded from this analysis. There were 47 true late presenters (46.5%) and 54 PHDA patients (53.4%).

PHDA patients included a larger proportion of migrants and visitors from abroad compared with late presenters (53.7% vs 34.0%, p=0.047). Proportions of patients identifying as men who have sex with men were similar (66.7% vs 55.3%, p=0.243). PHDA patients were more likely to be intravenous drug users (IVDUs) (9.3% vs 0.0%, p=0.032). 17 late presenters and 10 PHDA patients had no risk factors for HIV infection recorded. In the PHDA group, the median time from HIV diagnosis to the current presentation was 7.3 years (IQR: 4.0–13.3 years) (Otable 1).

Table 1

Differences between post-HIV diagnosis AIDS patients (PHDA patients) and late presenters

Prior to admission, PHDA patients had more CIDs diagnosed than late presenters (57.4% vs 12.8% p=0.000), more psychiatric comorbidities (35.2% vs 12.8% p=0.009) and a larger proportion reporting alcohol (24.1% vs 4.3% p=0.005) and substance abuse (22.2% vs 2.1% p=0.003). A quarter of PHDA patients had known social care issues, compared to none of the late presenting group. Concurrent diagnoses of new CIDs during admission (see online supplementary appendix table A) were not significantly different.

About 88.9% of PHDA patients were LTFU for at least one period greater than four months since their diagnosis of HIV. Of the PHDA group, the median length of LTFU periods was 11 months (IQR: 6–19 months). Reasons recorded were: travelling abroad (13%), social issues (13%), transfer of care to other centres (7.4%), treatment avoidance (5.6%) and religious views (3.7%). One patient (1.85%) had been seen by at least two treatment centres for their HIV follow-up during their LTFU period. About 63% of PHDA patients had previously been started on ART, of which the median duration was 48 months (IQR: 33–84 months). Twenty patients (58.9%) had discontinued ART at presentation. The median ART break each patient had was 12 months (IQR: 4–23 months). The reasons for stopping ART were: drug side effects (26.5%), incompatible lifestyles (11.8%), religious views (8.8%), travel abroad (5.9%), social issues (5.9%), negative views on ART (5.9%), belief in alternative medication (2.9%) and ART stopped on medical advice (2.9%).


More than half of all patients presenting to our inpatient unit with severe OIs had established HIV diagnoses.

Late presenters had few previously reported CIDs or comorbidities. PHDA patients had four times as many CIDs including previously treated OIs. The rates of alcohol and IVDUs in the PHDA group were higher than in the late presenting group, and also higher than previously reported rates of alcohol and IVDU in general cohorts of patients living with HIV.6 ,7 The rate of IVDU was almost twice that observed in the broader, UK HIV population.8

A majority of the PHDA group (88.9%) had previously defaulted HIV follow-up care. The most common reasons for LTFU periods were travel abroad and social issues; perhaps reflective of the high migrant/visitor population in the PHDA patients. Black African ethnicity and acquiring HIV infection abroad have been associated with LTFU rates.9 It was unclear if these patients had access to, and continued to take, ART during their time abroad. Social issues were common in the PHDA group and rare in the late presenter group. Reported social issues included financial difficulties, housing issues, receipt of disability benefits and reported domestic abuse. One patient was seen to change clinics frequently but this behaviour was rare (1.85%). The Health Protection Agency's Survey of Prevalent HIV Infections Diagnosed database (and the upcoming HIV/AIDS Reporting Section datasets) could be useful in tracking these patients between services.10

Strategies to reduce the burden of late presenting HIV infection must be developed.11 ,12 The low prevalence of CIDs in the late presenting group suggests that CID-based testing may not have identified earlier stage HIV infection in this cohort. Demographic/risk factor-based testing, and screening in high prevalence populations, may reduce the number of late presenters. However, the main finding of this paper remains that the PHDA cohort contributes the larger burden of AIDS events in our centre. This cohort has potentially modifiable factors contributing to their disease progression: psychosocial issues, drug and alcohol dependency problems and poor clinic attendance rates. HIV care centres should establish links with social services for social issues in PHDA patients to be highlighted and made known to them to ensure adequate social support. Patients who discontinue medications due to side effects or non-conventional belief systems should be offered adherence support and ART switches where appropriate.13 For patients with extended LTFU periods, assertive outreach should be established to engage these individuals in care. In our centre, for example, LTFU exercises are being undertaken to identify individuals either lost or at risk of being LTFU, and thereafter, provide an outreach service to help these patients re-access HIV clinical services. In a third of the PHDA cohort, reasons for LTFU or treatment discontinuation were undocumented. Documentation could be improved to inform and tailor targeted strategies to encourage clinic attendance and treatment adherence.

It must be noted that these are two very different groups presenting with AIDS, and a direct comparison may not be valid without a control group of HIV patients who do not progress to AIDS. However, we have shown that within the inpatient population admitted with AIDS, PHDA patients are a distinct subgroup from late presenters. Future case–control studies would be beneficial to identify the differences between PHDA patients and HIV patients who do not develop AIDS and to investigate factors associated with treatment access and effectiveness.

In conclusion, PHDA patients are a subgroup of patients who develop AIDS, with clear differences in demographics; with higher rates of psychiatric comorbidities, social issues, alcohol and substance abuse than both HIV late presenters and the broader cohort of patients living with HIV. Many default from clinical care and have poor ART compliance. Non-attendance and non-compliance in HIV care remain challenges that must be addressed by clinicians to prevent avoidable morbidity.

Key messages

  • Half of the patients presenting with AIDS-defining illnesses have previously diagnosed HIV infections (post-HIV diagnosis AIDS patients).

  • These patients have higher rates of psychiatric comorbidities, social issues and alcohol and substance abuse.

  • These patients often have treatment breaks and clinic non-attendance due to potentially modifiable factors such as social issues, drug side effects and treatment avoidance.


Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

    Files in this Data Supplement:


  • Contributors MJL and MR contributed to the study design, data collection and drafting of the paper and revisions. MJL performed the data analysis. MR supervised the project. AS and BG contributed to the drafting of the paper and editorial review.

  • Competing interests None.

  • Provenance and peer review Not commissioned; externally peer reviewed.