Missed opportunities for HIV diagnosis: a three-year audit in the West of Scotland

Abstract

Background

National specialty guidelines for HIV testing aim to increase diagnosis and reduce late presentation. An audit of new HIV diagnoses in Glasgow was performed to assess local performance against these guidelines and estimate the proportion of patients presenting who had previous missed opportunities for diagnosis.

Methods

A retrospective case note review of 339 patients diagnosed from September 2008 to September 2011 was performed. Documented past medical history was assessed for HIV clinical indicator conditions prior to HIV diagnosis and prior review by medical services.

Results

Ninety (26%) individuals had at least one documented clinical indicator condition prior to HIV diagnosis, of whom 80 had prior contact with at least one speciality. This group also had a lower mean nadir CD4 count (258 cells/cmm versus 393 cells/cmm, p = <0.005) and were more likely to be severely immunocompromised at diagnosis, with a CD4 count below 50 cells/cmm (31% versus 9%, p = <0.005). AIDS-defining illnesses were also more common (31% versus 8%, p ≤ 0.005) as was HIV-related mortality (p ≤ 0.005).

Conclusion

Additional support and training are required to increase adherence to HIV-testing guidelines within primary and secondary care in order to prevent ongoing late presentation with both individual clinical and public health implications.

Keywords

HIV testing audit missed opportunities

With the advent of effective antiretroviral therapy (ART), HIV has become a treatable condition with mortality rates approaching those of HIV-negative matched populations.¹ However, in order to achieve this, HIV-positive individuals must be diagnosed as soon as possible after seroconversion to allow the initiation of therapy prior to compromised immunity. The UK CHIC study found that life expectancy was on average 15 years less in HIV patients diagnosed late, classified as a baseline CD4 count lower than the threshold for initiation of ART (CD4 < 350 cells/mm).² Unfortunately, late diagnosis remains common. In 2011, of the 6660 individuals diagnosed HIV-positive in the UK, 50% were classified as late diagnoses.³ Locally, the West of Scotland HIV cohort reflects national epidemiology with 57% of new diagnoses presenting with a CD4 count <350 cells/mm in the same time period.

In view of these findings, the British HIV Association (BHIVA) published national HIV testing guidelines in 2008 in an attempt to increase testing rates and reduce late diagnosis. Recommendations not only focus on high-risk groups and high prevalence communities but also highlight the need for HIV testing because of clinical indications. A list of adult and paediatric clinical indicator diseases (CIDs), which should prompt HIV testing, is described and it is recommended that all individuals presenting with any of these conditions should be offered a test regardless of risk factors.⁴ Nevertheless, despite this guidance, there is a continued need for education and policy to address this issue. The BHIVA national audit 2010 which reviewed 1112 new HIV diagnoses found that 50% of patients presented with a baseline CD4 count of <350 cells/mm and that 25% had at least one missed opportunity for HIV testing.⁵ In order to assess this issue locally and guide regional education and policy interventions, an audit of new diagnoses over a three-year period, since the introduction of the BHIVA guidelines, was performed.

Methods

From 1 September 2008 to 1 September 2011 there were 417 new cases of HIV reported at the regional unit in Glasgow. A total of 78 patients were excluded either due to a previous positive test prior to 2008 in another unit or because of transfer or death prior engaging in care with no clinical notes available. A retrospective case note review was therefore performed for 339 patients. Documented past medical history in clinical paper notes within the HIV centre and health board wide computerised clinical records were screened for HIV CIDs prior to HIV diagnosis. Prior review, investigation and procedures by medical services were also documented. Baseline demographics of patients with and without a prior CID were also collected. Data were analysed for significance with t test and chi-square statistics.

Results

Ninety patients (26%) had at least one documented clinical indicator condition prior to HIV diagnosis. Older individuals and those born in the UK were more likely to have a previous missed opportunity for diagnosis although this did not reach statistical significance (Table 1). Of note there was no significant difference in gender or likely route of transmission between the two groups.

Table 1.

Baseline demographics.

No previous CID	Previous CID
Demographic	249	90
Gender (Male:Female)	172:77 (69%:31%)	66:24 (73%:27%)	p = 0.44
Age (mean in years)	38	41	p = 0.02
Place of birth	144 (58%)	60 (67%)	p = 0.02
UK	84 (34%)	28 (31%)
Africa	5 (2%)	1 (1%)
Western Europe	11 (4%)	1 (1%)
Eastern Europe	5 (2%)	0
Asia
Transmission	127 (51%)	44 (49%)	p = 0.19
Sexual intercourse	111 (45%)	40 (44%)
Heterosexual	8 (3%)	2 (2%)
Homosexual	0	1 (1%)
IVDU	3 (1%)	3 (3%)
Iatrogenic
Unknown

CID: clinical indicator disease; IVDU: intravenous drug use.

The most common CIDs were blood dyscrasia, bacterial pneumonia, weight loss, dermatological presentations, lymphadenopathy and chronic diarrhoea (Table 2). Of note, a number of patients were diagnosed with more than one CID. In 80 individuals prior contact with at least one speciality was documented, most commonly primary care in 61 patients followed by general medical in-patients, ENT, gastroenterology and general surgery (Table 3). In 10 patients, a previous CID was documented; however, it was not clear from the clinical notes which speciality had delivered care for this condition. A further analysis matched CID to speciality (Table 4). In 17 individuals, potentially unnecessary investigations were also performed including bronchoscopy, colonoscopy, endoscopy, lymph node biopsy and open lung biopsy (Table 5).

Table 2.

Clinical indicator diseases.

CID	No of patients
Blood dyscrasia (lymphopenia, neutropenia, thrombocytopenia)	28
Bacterial pneumonia	19
Weight loss	18
Lymphadenopathy	11
Diarrhoea	11
Herpes zoster	8
Previous STI (HSV and genital HPV)	7
Oral candida	5
Mononucleosis type syndrome	3
Hepatitis B	3
Hepatitis C	2
Seborrhoeic dermatitis	2
Varicella zoster	2
TB	2
Dementia/memory impairment	2
Peripheral neuropathy	1
Crypotococcal meningitis initially diagnosed as CVA	1
Viral meningitis	1
Cervical intraepithelial neoplasia	1
Molluscum contagiosum virus (facial)	1
Nodular prurigo	1

CID: clinical indicator disease; CVA: cerebrovascular accident: HPV is human papillomavirus: HSV is herpes simplex virus: STI is sexually transmitted infection: TB is tuberculosis.

Table 3.

Speciality interactions in individuals with a previous CID.

Speciality	No of patients
General practice	61
General medicine inpatients	8
ENT outpatients	7
Gastroenterology outpatients	5
General surgery outpatients	5
Dermatology outpatients	4
Accident and Emergency	2
General medicine outpatients	1
Haematology outpatients	1
Respiratory outpatients	1
Hepatitis C outpatients	1
Breast surgery outpatients	1
Maxillofacial inpatients	1
Missing data	10

CID: clinical indicator disease; ENT: ear, nose and throat.

Table 4.

Clinical indicator diseases by speciality.

Speciality	CID
General practice	All conditions listed in Table 1 with the exception of viral meningitis, CIN and misdiagnosed CVA.
General medicine inpatients	Blood dyscrasia
	Bacterial pneumonia
	Diarrhoea
	Weight loss
	Oral Candida
	Varicella zoster
	CVA - subsequently diagnosed as cryptococcal meningitis
	Viral meningitis
ENT outpatients	Lymphadenopathy
ENT outpatients	Recurrent sinusitis (in presence of pancytopenia)
Gastroenterology outpatients	Diarrhoea
	Weight loss
	Oral candida
	HBV
General surgery outpatients	Diarrhoea
	Lymphadenopathy
	Perianal HPV
Dermatology outpatients	Molluscum contagiosum (facial)
Dermatology outpatients	Severe eczema in presence of lymphopenia
Accident and Emergency	Bacterial pneumonia
	Oral candida
	Herpes Zoster
General medicine outpatients	Blood dyscrasia
	Bacterial pneumonia
	Diarrhoea
	Weight loss
	Oral candida
Haematology outpatients	Blood dyscrasia
Respiratory outpatients	Bacterial pneumonia
	Weight loss
	Blood dyscrasia
Hepatitis C outpatients	HCV
Breast surgery outpatients	Axillary lymphadenopathy
Maxillofacial inpatients	Seroconversion illness

CIN: cervical intraepithelial neoplasia; CVA: cerebrovascular accident; ENT: ear, nose and throat; HBV: hepatitis B virus; HCV: hepatitis C virus; HPV: human papillomavirus.

Table 5.

Procedures performed in individuals with a previous CID prior to HIV diagnosis.

Procedure	Number of patients
Bronchoscopy	5
Colonoscopy	5
Lymph node biopsy	4
Endoscopy	2
Open lung biopsy and VATS	1
Tonsil biopsy	1

CID: clinical indicator disease; VATS: video assisted thoracic surgery.

Considering clinical outcome, individuals with a previous CID were eventually more likely to be diagnosed during an acute admission (34% versus 20%, p = 0.007, Table 6). This group also had a lower baseline CD4 count (258 cells/cmm versus 393 cells/cmm, p ≤ 0.005) and were more likely to be severely immunocompromised at diagnosis with a CD4 count <50 cells/cmm (31% versus 10%, p ≤ 0.005) (Table 7). AIDS-defining illnesses were also more common than in patients without a prior CID (35% versus 8%, p ≤ 0.005) as was HIV-related mortality with three deaths due to PCP in the previous CID group (p ≤ 0.005).

Table 6.

Place of diagnosis.

Place of diagnosis	Previous CID	No previous CID
Screening	37	106
GUM services	1	14
Antenatal	4
Blood transfusion	5
Occupational health	1
Prison services	1
Assisted Conception Unit
Acute admission	16	30
General medicine	10	17
Infectious diseases	4	1
Respiratory	1	1
Gastroenterology
Accident and Emergency
Outpatients	6	33
General practice	6	3
Haematology	2	1
Neurology	1	1
General medicine	1	1
Respiratory
Dermatology
Oncology
Urology
Brownlee counselling	3	23
Not documented	5

GUM: genitourinary medicine.

Table 7.

Baseline CD4 count, CDC category, AIDS-defining diagnosis and HIV-related mortality.

No previous CID	Previous CID
CD4 at presentation • Mean • Median	393 363 (4–1281)	258 164 (0–954)	p < 0.005
CD4 < 200 CD4 < 50	66 (26%) 25 (10%)	47 (52%) 28 (31%)	p < 0.005
CDC • A • B • C	217 (87%) 13 (5%) 19 (8%)	43 (48%) 15 (17%) 32 (35%)	p < 0.005
AIDS-defining diagnosis	19 (8%)	32 (35%)	p < 0.005
HIV-related mortality	0	3	p < 0.005

AIDS: acquired immune deficiency syndrome; CDC: centre for disease control.

Discussion

Our findings reflect national data with 26% of patients having a previous CID which should have prompted an HIV test.⁵ Nevertheless, the limitations of this retrospective analysis have to be considered and our results may underestimate missed opportunities for testing due to lack of documentation and unavailable sources, or possibly overestimate because of the verbal discussion and refusal of an HIV test, which was not documented. However, despite these issues it is clear there remain a significant proportion of individuals who should have been offered testing.

The baseline demographic trends found are consistent with previous research, with a higher proportion of missed opportunities in older individuals.⁶ However, the trend for individuals born in the UK to be diagnosed later is inconsistent with previous studies. This finding may reflect local clinician’s assumption of low risk in this group and adds further support to testing on the basis of clinical indication alone.

With regard to CIDs, our findings again reflect national findings with blood dyscrasia, bacterial pneumonia, weight loss, lymphadenopathy and diarrhoea amongst the most common presentations.⁵ Again this supports the need for increased training highlighting the need for testing in individuals presenting with these conditions. Furthermore, patients with a previous CID were often reviewed by medical services on multiple occasions, with a significant proportion having unnecessary investigations and in some cases invasive procedures prior to an HIV test. This delay in diagnosis does not only affect individual morbidity and mortality, it also adds additional expense to healthcare costs with a combination of potentially avoidable investigations in addition to the complex care of undiagnosed patients who inevitably present with severe immune compromise and opportunistic infections.⁷ Our findings demonstrate this with significantly more individuals presenting with severe immunocompromise and AIDS-defining diagnoses in addition to three HIV-related deaths in the group with a previous CID. This mortality and morbidity is completely preventable with early diagnosis and ART.

HIV testing is also important from a public health perspective. Individuals unaware of their HIV status continue to have unprotected sex and, if untreated, have high HIV viral loads increasing transmission likelihood to sexual partners. Research in the USA estimates that 54% of new infections are caused by the undiagnosed 25% of HIV-infected persons and suggests that the HIV epidemic will only be tackled with expanded testing.⁸ This issue becomes even more important with increasing focus within the HIV community regarding ‘treatment for prevention,’ with one study showing early initiation of ART in HIV-positive individuals, conferring a 94% reduction in HIV transmission to sexual partners.⁹

It is therefore imperative that both local and national training and policy interventions address the issue of missed opportunities for HIV testing before transmission rates can decline and the life expectancy for untested HIV-positive individuals can be maximised. There has been additional national guidance published by NICE to increase testing specifically in homosexual and African groups and internationally from the HIV Europe, ECDC and WHO Europe HIV who recently published HIV indicator condition guidance, reflecting the wider recognition of the need for increased testing and awareness. ^10,11 However, until guidance is translated into clinical practice, the aims of these initiatives can and will not be achieved. We urge clinicians working in Scotland to offer HIV tests to patients presenting with HIV clinical indicator conditions without being deterred by the absence of traditional risk factors.

Footnotes

Declaration of conflicting interests

None declared.

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Statement of contributorship

DB designed the audit, collected and analysed the data and drafted the paper. AM, RN, DJB, FF, RF, EP, RAS and AW made comments, revised the draft and contributed patients to the audit.

References

Zwahlen

Harris

Hogg

. Mortality of HIV-infected patients starting potent antiretroviral therapy: comparison with the general population in eight industrialized countries. Int J Epidemiol 2009; 38: 1624–1633.

May

Gompels

Delpech

. Impact of late diagnosis and treatment on life expectancy in people with HIV-1: UK Collaborative HIV Cohort (UK CHIC) study. BMJ 2011; 343: d6016–d6016.

Health Protection Agency. HIV in the UK: 2011, http://www.hpa.org.uk/webc/HPAwebFile/HPAweb_C/1317131685847 (2011, accessed 8 November 2012).

British HIV Association. UK national guidelines for HIV testing 2008, www.bhiva.org (2008, accessed 8 November 2012).

Ellis

Curtis

Ong

ELC

. HIV diagnosis and missed opportunities. Results of the British HIV Association (BHIVA) national audit 2010. Clin Med 2012; 12: 430–434.

Girardi

Sabin

Monforte

. Late diagnosis or HIV: epidemiological features, consequences and strategies to encourage earlier testing. JAIDS 2007; 46: S3–S8.

Krentz

Gill

. The direct medical costs of late presentation of HIV infection over a 15 year period. AIDS Res Treat 2012; 2012: 757135–757135.

Marks

Crepaz

Janssen

. Estimating sexual transmission of HIV from persons aware and unaware that they are infected with the virus in the USA. AIDS 2006; 20: 1447–1450.

Cohen

Chen

McCauley

. Prevention of HIV-1 infection with early antiretroviral therapy. N Engl J Med 2011; 365: 493–505.

10.

NICE guidance. Increasing the uptake of HIV testing to reduce undiagnosed infection and prevent transmission among men who have sex with men, http://egap.evidence.nhs.uk/ph34 (2011, accessed 8 November 2012).

11.

HIV Europe Steering group, ECDC and WHO Europe. HIV indicator conditions: guidance for implementing HIV testing in adults in health care settings, www.hiveurope.eu (2012, accessed 15 November 2012).