HIV-indicator-condition-driven HIV testing: clinically effective but still rarely implemented

HIV prevalence among cervical (pre)cancer diagnoses in Suriname: a retrospective population study

Objective

To determine the prevalence of HIV in women with (pre)cancerous cervical lesions in Suriname and their retention in care.

Methods

A retrospective population study including all women diagnosed with (pre)invasive cervical intraepithelial neoplasia (CIN I to III or cervical cancer) in the only pathology department, during 2010-2020. The HIV test coverage and the HIV positivity ratio were determined through matching pathology data with the national HIV test database. The relation between retention in HIV care up to 2022 and different covariates was determined through Kaplan-Meier survival analysis and log-rank tests.

Results

There were 2 901 (1 395 CIN I, 396 CIN II, 444 CIN III, and 666 cervical cancer) diagnoses of (pre)invasive cervical neoplasia. An overall HIV test coverage of 57.5% and a positivity ratio of 5.8% were found, with no difference among the (pre)cancer stages. The undiagnosed prevalence (women not previously known HIV-positive at cervical diagnosis) was 1.6% and 2.9% among precancer and cancer diagnoses, respectively. The median time in care of women with cervical cancer was 8 months for those not on antiretroviral therapy (ART) and 4 years for those starting ART. Among women with precancer stages this was 5 and 10 years, respectively (p < 0.05).

Conclusions

HIV testing, followed by treatment initiation when found HIV-positive, should be prioritized in women diagnosed with cervical neoplasia. This will enhance the individual clinical outcomes and facilitate the control of the HIV epidemic in Suriname.

Missed opportunities for early HIV diagnosis in Greece: The MORFEAS study, 2019 to 2021

BackgroundLate HIV diagnosis (CD4+ T-cell count < 350 cells/μL, or with an AIDS-defining event) remains a persistent challenge in Greece, indicating potential missed opportunities (MOs) for earlier testing.AimTo determine the frequency of HIV indicator conditions (ICs) preceding diagnosis and to quantify MOs for earlier testing at a nationwide level in Greece.MethodsThis multicentre retrospective study analysed data on 823 antiretroviral therapy-naive adults (≥ 18 years) diagnosed with HIV during 2019-21. Medical records were reviewed to identify pre-diagnosis healthcare contacts (HCCs) and ICs justifying HIV testing. Univariable and multivariable logistic regression identified factors associated with ≥ 1 MO. A Bayesian model estimated the time from seroconversion to diagnosis.ResultsAmong 517 participants with HCC data, 249 had ≥ 1 HCC. Of these, 59.0% (147/249) were late presenters. These cases had 365 HCCs, and 191 (52.3%) were MOs for testing. The most common ICs were sexually transmitted infections (39.8%; 76/191) and fever (11.0%; 21/191). Non-Greek origin was associated with lower odds of experiencing ≥ 1 MO (adjusted odds ratio: 0.48; 95% CI: 0.22─1.02), while higher education increased odds of MOs for early HIV diagnosis. Median time from seroconversion to diagnosis was 3.2 years for the full sample and 3.7 years for those with HCC, with about half of the latter reporting MOs post-estimated seroconversion. Recognising MOs would have potentially spared approximately 1 year of delay in diagnosis.ConclusionMOs for earlier HIV diagnosis were prevalent in Greece. Leveraging IC-guided testing and addressing barriers could support earlier diagnosis and treatment, limiting adverse health outcomes and preventing transmission.

Missed opportunities for HIV testing and sexual health-related challenges in an individual with intellectual disability: a case report

BACKGROUND

HIV testing remains an important tool in identifying people living with HIV/AIDS (PLWHA). An early diagnosis of HIV can lead to a prolonged life expectancy if treatment is initiated promptly. Indicator conditions can be the first sign of an HIV infection and should therefore be recognised and consequently a HIV test should be carried out. Testing should occur in all individuals as sexuality can be experienced by everyone, and stigma can lead to the exclusion of vulnerable groups, leading to a gap in diagnosis and treatment [1, 2].

CASE PRESENTATION

A 63-year-old man, who identifies as bisexual and has had an intellectual disability since birth, presented at our health care centre for HIV testing. A decade ago, the patient was diagnosed with Stage III Diffuse Large B-cell Non-Hodgkin Lymphoma, an AIDS defining cancer. The patient presented at a Haematology and Oncology department 3 months prior, due to a weight loss of 10 kg over the past 5 months. Oral thrush, an HIV-indicator condition, had been diagnosed by the otolaryngologists shortly before. During this medical evaluation, pancytopenia was identified. Despite the presence of indicator conditions, the patient was never tested for HIV in the past. Staff members from the care facility for intellectually disabled suggested conducting a HIV test in our clinic through the public health department, where HIV positivity was revealed. The AIDS-defining diagnosis, along with a CD4 + cell count of 41/µl, suggests a prolonged period of HIV positivity.

CONCLUSION

Due to the presence of existing indicator conditions, an earlier HIV diagnosis was possible. We contend that most of the recent illnesses could have been prevented if earlier testing had been carried out. Therefore, patients presenting with AIDS indicator conditions, including those with mental disabilities, should be given the opportunity to be tested for HIV. HIV/AIDS trainings should be made available to health care professionals as well as to personnel interacting with vulnerable groups.

Characteristics and short- and long-term direct medical costs among adults with timely and delayed presentation for HIV care in the Netherlands

INTRODUCTION

In Europe, half of people living with HIV (PLWH) present late to care, with associated higher morbidity and mortality. This study aims to assess short- and long-term costs of HIV-care based on time of presentation and identify other factors contributing to higher costs in the first and fifth year after antiretroviral therapy (ART) initiation.

MATERIAL AND METHODS

We included ATHENA cohort data which prospectively includes 98% of PLWH in the Netherlands. PLWH who initiated ART in 2013 were included and followed over five years. PLWH were divided in three categories based on CD4 cell-count at time of ART initiation: timely presentation (CD4>350cells/μL), late presentation (CD4 200-350cells/μL or >350cells/μL with AIDS-defining illness) and very late presentation (CD4<200cells/μL). The total HIV-care cost was calculated distinguishing ART medication and non-ART medication costs (hospitalization, outpatient clinic visits, co-medications, and HIV-laboratory tests).

RESULTS

From 1,296 PLWH, 273 (21%) presented late and 179 (14%) very late. Nearly half of those who entered HIV-care in a very late stage were of non-Dutch origin, with 21% originating from sub-Saharan Africa. The mean cost per patient in the first year was €12,902 (SD€11,098), of which about two-thirds due to ART (€8,250 (SD€3,142)). ART costs in the first and fifth year were comparable regardless of time of presentation. During the first year on treatment, non-ART medication costs were substantially higher among those with late presentation (€4,749 (SD€8,009)) and very late presentation (€15,886 (SD€ 21,834)), compared with timely presentation (€2,407(SD€4,511)). Higher non-ART costs were attributable to hospitalization and co-medication. The total non-ART costs incurred across five years on treatment were 56% and 246% higher for late and very late presentation respectively as compared to timely presentation.

CONCLUSION

Very late presentation is associated with substantial costs, with non-ART costs nearly seven times higher than for those presenting timely. Hospitalization and co-medication costs are likely to continue to drive higher costs for individuals with late presentation into the future. Programs that identify individuals earlier will therefore likely provide significant short- and long-term health cost savings.

Promoting HIV indicator condition-guided testing in hospital settings (PROTEST 2.0): study protocol for a multicentre interventional study

Background

Late presentation remains a key barrier towards controlling the HIV epidemic. Indicator conditions (ICs) are those that are AIDS-defining, associated with a prevalence of undiagnosed HIV > 0.1%, or whose clinical management would be impeded if an HIV infection were undiagnosed. IC-guided HIV testing is an effective strategy in identifying undiagnosed HIV, but opportunities for earlier HIV diagnosis through IC-guided testing are being missed. We present a protocol for an interventional study to improve awareness of IC-guided testing and increase HIV testing in patients presenting with ICs in a hospital setting.

Methods

We designed a multicentre interventional study to be implemented at five hospitals in the region of Amsterdam, the Netherlands. Seven ICs were selected for which HIV test ratios (proportion of patients with an IC tested for HIV) will be measured: tuberculosis, cervical/vulvar cancer or high-grade cervical/vulvar dysplasia, malignant lymphoma, hepatitis B and C, and peripheral neuropathy. Prior to the intervention, a baseline assessment of HIV test ratios across ICs will be performed in eligible patients (IC diagnosed January 2015 through May 2020, ≥18 years, not known HIV positive) and an assessment of barriers and facilitators for HIV testing amongst relevant specialties will be conducted using qualitative (interviews) and quantitative methods (questionnaires). The intervention phase will consist of an educational intervention, including presentation of baseline results as competitive graphical audit and feedback combined with discussion on implementation and opportunities for improvement. The effect of the intervention will be assessed by comparing HIV test ratios of the pre-intervention and post-intervention periods. The primary endpoint is the HIV test ratio within ±3 months of IC diagnosis. Secondary endpoints are the HIV test ratio within ±6 months of diagnosis, ratio ever tested for HIV, HIV positivity percentage, proportion of late presenters and proportion with known HIV status prior to initiating treatment for their IC.

Discussion

This protocol presents a strategy aimed at increasing awareness of the benefits of IC-guided testing and increasing HIV testing in patients presenting with ICs in hospital settings to identify undiagnosed HIV in Amsterdam, the Netherlands.

Trial registration

Dutch trial registry: NL7521. Registered 14 February 2019.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12879-021-06183-8.

Current evidence on the adoption of indicator condition guided testing for HIV in western countries: A systematic review and meta-analysis

BACKGROUND

Indicator condition (IC) guided testing for HIV is an effective way to identify undiagnosed people living with HIV, but studies suggest its implementation is lacking. This systematic review provides an overview of the adoption of IC-guided testing in Western countries.

METHODS

Seven ICs were selected: tuberculosis (TB), malignant lymphoma, hepatitis B, hepatitis C, cervical/vulvar carcinoma/intraepithelial neoplasia grade 2+ (CC/CIN2+, VC/VIN2+), and peripheral neuropathy (PN). Embase and Ovid MEDLINE were searched up to November 20th, 2020. Publications of all types, using data from ≥2009, reporting on HIV test ratios in patients ≥18 years in all settings in Western countries were eligible. HIV test ratios and positivity were reported per IC. A random effects-model for proportions was used to calculate estimated proportions (ES) with 95% CIs. This study was registered at PROSPERO, registration number CRD42020160243.

FINDINGS

Fifty-seven references, including 23 full-text articles and 34 other publications were included. Most (28/57) reported on HIV testing in TB. No reports on HIV testing in VC/VIN2+ or PN patients were eligible for inclusion. Large variation in HIV test ratios was observed between and within ICs, resulting from different testing approaches. Highest HIV test ratios (pooled ratio: 0·72, 95%CI 0·63-0·80) and positivity (0·05, 95% CI 0·03-0·06) were observed among TB patients, and lowest among CC/CIN2+ patients (pooled ES test ratio: 0·12, 95%CI 0·01-0·31, positivity: 0·00, 95%CI 0·00-0·00).

INTERPRETATION

IC-guided HIV testing is insufficiently implemented in Western countries. The large variation in test ratios provides insight into priority areas for implementing routine IC-guided HIV testing in the future.

FUNDING

HIV Transmission Elimination in Amsterdam (H-TEAM) consortium and Aidsfonds (grant number P-42,702).

Overlooked cases of HIV infection: An Italian tale of missed diagnostic opportunities

BACKGROUND

Late diagnoses are still a cause of increased HIV-related morbidity and mortality despite the availability of highly effective treatments. The aim of this study was to identify indicator conditions (ICs) in late presenters with HIV infection (LPs) that may represent missed opportunities of undertaking earlier HIV testing.

METHODS

The medical records of LPs referred to a specialist clinic in Milan, Italy, between 2011 and 2017 were reviewed to assess the frequency of ICs during the five years preceding diagnosis. Logistic regression analysis was used to investigate the factors associated with missed opportunities of making an earlier diagnosis.

RESULTS

The analysis considered 203 LPs (60.6% of the patients newly diagnosed as having HIV infection during the study period). Most had had ≥1 medical encounter in the five years before diagnosis, and 54 (26.6%) had been diagnosed as having ≥1 IC without undergoing HIV testing. The most frequent ICs were herpes zoster (19.8%), constitutional symptoms (17.4%) and lympho/thrombocytopenia (12.8%), and the missed opportunities for testing occurred in the settings of primary care (44.9%), specialist medical (38.2%) or surgical services (11.3%), and emergency departments (5.6%). Twenty-five (53.2%) of the 47 subjects with a non AIDS-defining IC had AIDS at the time of the diagnosis of HIV infection. Subjects aged >60 years were at increased risk of missed diagnostic opportunities (aOR 4.80, p = 0.008).

CONCLUSION

Implementing IC-guided HIV testing in non-specialist settings is an essential means of reducing late diagnoses of HIV infection even in the case of older subjects.

Monitoring HIV-indicator condition guided HIV testing in Estonia

OBJECTIVES

The aim of the study was to assess indicator condition (IC) guided HIV testing in Estonia from 2012-2015.

METHODS

We used Estonian Health Insurance Fund (EHIF) data. EHIF is the core purchaser of health care services in Estonia, covering health care costs for insured people (94% of the total population). After health care services' provision, the provider sends an invoice to EHIF, which includes patient information (e.g. age, gender, diagnoses based on ICD-10) and services provided (e.g. what tests were performed).

RESULTS

Among the ICs analysed, the highest proportion of patients tested was among those presenting with infectious mononucleosis-like illness (27-33% of patients) and viral hepatitis (28-32%), the lowest proportion of patients tested was among those presenting with herpes zoster (4-5%) and pneumonia (4-8%). Women were tested somewhat less than men, especially in cases of sexually transmitted infections (9-13% and 18-21%, respectively).

CONCLUSIONS

Our data shows that IC-guided HIV testing rates are low in Estonia. Therefore, it is critical to follow Estonian HIV testing guidelines, which recommend IC-guided testing. In general, health insurance data can be used to monitor IC-guided HIV testing.

Indicator condition-guided HIV testing with an electronic prompt in primary healthcare: a before and after evaluation of an intervention

OBJECTIVE

Indicator condition (IC)-guided HIV testing is a strategy for the diagnosis of patients with HIV. The aim of this study was to assess the impact on the proportion of HIV tests requested after the introduction of an electronic prompt instructing primary healthcare (PHC) physicians to request an HIV test when diagnosing predefined IC.

METHODS

A prospective interventional study was conducted in 2015 in three PHC centres in Barcelona to assess the number of HIV test requests made during the implementation of an electronic prompt. Patients aged 18-65 years without HIV infection and with a new diagnosis of predefined IC were included. The results were compared with preprompt (2013) and postprompt data (2016).

RESULTS

During the prompt period, 832 patients presented an IC (median age 41.6 years [IQR 30-54], 48.2% female). HIV tests were requested in 296 individuals (35, 6%) and blood tests made in 238. Four HIV infections were diagnosed (positivity rate 1.7%, 95% CI 0.5% to 4.4%). The number of HIV tests requested based on IC increased from 12.6% in 2013 to 35.6% in 2015 (p<0.001) and fell to 17.9% after removal of the prompt in 2016 (p<0.001). Younger patient age (OR 0.97, 95% CI 0.96 to 0.98), birth outside Spain (OR 1.53, 95% CI 1.06 to 2.21) and younger physician age (OR 0.97, 95% CI 0.96 to 0.99) were independent predictive factors for an HIV test request during the prompt period. The electronic prompt (OR 3.36, 95% CI 2.70 to 4.18) was the factor most closely associated with HIV test requests. It was estimated that 10 (95% CI 3.0 to 26.2) additional new cases would have been diagnosed if an HIV test had been performed in all patients presenting an IC.

CONCLUSIONS

A significant increase in HIV test requests was observed during the implementation of the electronic prompt. The results suggest that this strategy could be useful in increasing IC-guided HIV testing in PHC centres.

Routine HIV testing program in the University Infectious Diseases Centre in Lithuania: a four-year analysis

BACKGROUND

HIV transmission remains a major concern in Eastern Europe, and too many people are diagnosed late. Expanded testing strategies and early and appropriate access to care are required. Infectious disease departments might be targets for expanded HIV testing owing to the intense passage of key patient populations that carry indicators of HIV disease. Our objective was to evaluate the feasibility and clinical effectiveness of a fully integrated, opt-out routine, rapid HIV testing program.

METHODS

A retrospective four-year study of a screening program was conducted from 2010 through 2014. The program was divided into two periods: from 2010 to 2012 (pilot study) and from 2013 to 2014. The pilot study consisted of routine HIV testing of patients aged 18-55 that were hospitalized in one department. In the second period, all inpatients aged 18-65 were eligible. Targeted testing was conducted in the other inpatient department during the pilot study and the outpatient department during both periods.

RESULTS

During the pilot study, 2203 patients were hospitalized, 1314 (59.6%) were eligible, 954 (72.6%) were tested, and 3 (0.31%) were newly diagnosed HIV-positive. In the second period, 4911 patients were hospitalized, 3727 (75.9%) were eligible, 3303 (88.6%) were tested, and 7 (0.21%) were HIV-positive. In total, 2800 targeted tests were performed, and 4 (0.14%) patients tested positive with newly discovered HIV. All 14 newly diagnosed patients were provided with care. Comparing cumulative groups of routine and targeted testing, the HIV prevalence was 0.23% vs. 0.14% (p = 0.40) and was above the reported cost-effectiveness threshold of 0.1% (p = 0.012). A lower proportion of advanced disease and a higher proportion of heterosexually transmitted infection were found in the routine testing group.

CONCLUSION

Routine HIV testing in admissions of infectious diseases is acceptable, feasible, sustainable and clinically effective. Compared to targeted testing, routine testing helped to discover more patients in earlier stages and those with heterosexually transmitted HIV infection.

A feasibility study for a clinical decision support system prompting HIV testing

OBJECTIVES

Levels of undiagnosed HIV infection and late presentation remain high globally despite attempts to increase testing. The objective of this study was to evaluate a risk-based prototype application to prompt HIV testing when patients undergo routine blood tests.

METHODS

Two computer physician order entry (CPOE) systems were modified using the application to prompt health care workers (HCWs) to add an HIV test when other tests selected suggested that the patient was at higher risk of HIV infection. The application was applied for a 3-month period in two areas, in a large London hospital and in general practices in Teesside/North Yorkshire. At the end of the evaluation period, HCWs were interviewed to assess the usability and acceptability of the prompt. Numbers of HIV tests ordered in the general practice areas were also compared before and after the prompt's introduction.

RESULTS

The system was found to be both useable and generally acceptable to hospital doctors, general practitioners and nurse practitioners, with little evidence of prompt/alert fatigue. The issue of the prompt appearing late in the patient consultation did lead to some difficulties, particularly around discussion of the test and consent. In the general practices, around 1 in 10 prompts were accepted and there was a 6% increase in testing rates over the 3-month study period (P = 0.169).

CONCLUSIONS

Using a CPOE-based clinical decision support application to prompt HIV testing appears both feasible and acceptable to HCWs. Refining the application to provide more accurate risk stratification is likely to make it more effective.