Late stage presentation of HIV-positive patients to antiretroviral outpatient clinic in Zambia

Background The World Health Organization (WHO) and the Zambian Ministry of Health set out new guidelines on combination antiretroviral therapy (cART) in 2013 expanding the eligibility criteria for patients with HIV. Objectives The primary objective were to determine when cART was initiated in HIV-positive outpatients according to clinical and immunological criteria, and to identify what proportion of patients who were eligible for cART according to 2013 WHO and 2013 Zambian cART guidelines were currently on cART. Methodology This was a clinical audit of HIV-positive outpatients attending the cART clinic at Ndola Central Hospital in Ndola, Zambia, with retrospective cross-sectional chart review and survey design. Data were collected from clinical records and interviews with patients. Results A total of 99% of patients eligible for cART according to 2013 guidelines were on treatment. Clinical staging of patients at initiated on cART (n = 206) was as follows: 28% clinical stage I, 21% clinical stage II, 36% clinical stage III and 15% clinical stage IV. The median CD4 count when patients were started on cART was 147 cells/mm3. Conclusion The results show that a majority of patients were initiated on cART late in their disease course according to immunological (CD4 < 200 cell/mm3) and clinical criteria (stage III or IV). However, the vast majority of patients eligible for cART were currently on treatment. The late initiation of cART appears to be a result of late diagnosis of HIV.


Introduction
Human immunodeficiency virus (HIV) and acquired immune deficiency syndrome (AIDS) are major causes of morbidity and mortality, with an estimated 36.7 million HIV-infected people worldwide in 2015. 1 Illness and symptoms associated with HIV are grouped into four clinical stages according to World Health Organization (WHO) criteria, ranging from asymptomatic disease (stage I) to AIDS defining illness (stage IV). 2 Immunological monitoring of HIV is achieved through a blood test called the CD4 count, which measures the CD4 lymphocytes. There is a strong relationship between CD4 count, immunodeficiency and risk of HIV-related illnesses. 3,4 Since the late 1980s, various antiretrovirals (ARVs) have been developed which can significantly improve outcomes for people with HIV. ARV drugs are given in combination in triple regimens known as combination antiretroviral therapy (cART). 3 Zambia is one of the countries most severely affected by the HIV epidemic, with an estimated prevalence of 13.3% in adults in 2013-2014. 5 The HIV epidemic in Zambia came to the fore in the 1990s; it was a generalised epidemic with the main method of transmission being heterosexual sex. 6 There are a variety of factors which could explain the HIV epidemic in Zambia. These include a tendency to have multiple concurrent partners, frequent intergenerational sex, low condom usage and transactional sex. 6,7,8 The prevalence of HIV in Zambia has decreased from 15.6% in 2001-2002 to 13.3% in 2013-2014; however, it remains higher in some regions such as the Copperbelt, where the prevalence was 18.2% in 2013-2014. 5 Over the past few years, there have been improvements in HIV testing and counselling (HTC) and HIV education programmes in Zambia. There was an increase in HIV testing in public health facilities from 7.4% of adults over 15 years tested in 2008 to 18.1% in 2010, and further increases have occurred since this date under a national HIV policy. 5 Access to cART in low-income countries has improved in recent years in large part because of assistance from The Global Fund to Fight AIDS, Tuberculosis and Malaria, and the President's Emergency Plan for AIDS Relief (PEPFAR). 9 From 2003, cART was provided free of charge at two centres in Zambia: The University Teaching Hospital in Lusaka and Ndola Central Hospital. This was subsequently expanded to other health facilities throughout Zambia with funding assistance from PEPFAR and the Global Fund, and by 2013, an estimated 82% of people eligible for cART in Zambia received treatment. 6,10 The national AIDS strategic framework (NASF) set a target of 90% of patients eligible for cART to be on treatment by 2015. 6 The 2013 WHO ART guidelines recommended that patients with clinical stage III or IV, or a CD4 count of less than 500 cells/mm 3 should be started on cART. 4 In 2013, the Zambian Ministry of Health produced a new set of cART guidelines in accordance with the 2013 WHO ART guidelines. 11 This study was performed in 2014; in 2015, the WHO released new recommendations, which stipulated that cART should be started in all patients with HIV regardless of their clinical stage or CD4 count. 12 Zambia offers universal free cART treatment to all HIV-positive people who meet eligibility criteria. The Zambian consolidated guidelines for treatment and prevention of HIV infection 2013 guidelines increased eligibility criteria in accordance with the WHO consolidated guidelines on the use of ARV drugs for treating and preventing HIV infection. 4,11 Earlier initiation of cART has been shown to improve health outcomes for patients with HIV. 13,14,15 It is important to audit cART use against these new national and international guidelines to ensure their effective use and identify any areas for improvement.

Objectives
The first primary objective of this study was to audit the proportion of patients eligible for cART according to 2013 WHO ART and 2013 Zambian cART guidelines who had been initiated on cART. The second primary objective was to find when HIV-positive outpatients attending the cART clinic at Ndola Central Hospital were initiated on cART according to CD4 count and clinical criteria. CD4 count at initiation of cART will also be stratified against clinical stage. In addition, a subgroup analysis of those started before 01 January 2010 and those started after 01 January 2010 will be performed. The secondary objectives were to (1) establish which cART regimens patients were currently on, (2) ascertain if these regimens were first-line, (3) audit first-line regimens against guidelines, (4) find changes in regimens and reasons for switching, (5) note side effects from the current regimen, (6) find adherence to cART, (7) find prevalence of risk factors for HIV and (8) check reported condom use.

Study design
A clinical audit of cART regimens with retrospective crosssectional chart review and survey design.

Data collection
Information was collected from clinical records and interviews with patients at the adult cART outpatient clinic in Ndola Central Hospital in Ndola, Copperbelt, Zambia. The study population was HIV-positive outpatients attending the adult cART clinic between 11 March 2014 and 07 April 2014. All patients attending the clinic during the study period were invited to participate in the study. The purpose of the study was explained to the patients and written or verbal consent was obtained from patients who wished to participate. In cases where the patient was unable to write their signature but could give verbal informed consent, the data collector noted that verbal consent had been obtained. Patients who did not provide consent were excluded from the study. Information on clinical stage, CD4 count, cART regimens, reasons for changing regimens and side effects from cART was collected from clinical records. Additional information on risk factors for HIV, reported condom use and adherence to medications was collected from interviews with patients. Where patients did not speak English, a translator was used. Information from the data collection sheets was transferred to an Excel spreadsheet for analysis.

Adherence to guidelines
The 2013 Zambian consolidated guidelines for treatment and prevention of HIV infection and the 2013 WHO consolidated guidelines on the use of ARV drugs for treating and preventing HIV infection both recommend starting cART in any HIV-positive patients with WHO clinical stage III or IV. In addition, these guidelines recommend that patients of any clinical stage with a CD4 count of below 500 cells/mm 3 should also be initiated on cART. 4,11 Patients were classified as eligible for treatment if they currently met eligibility criteria according to the 2013 guidelines for cART, or if had previously been started on cART according to guidelines at the time of initiation ( Table 1). The eligibility criteria for cART have been expanded; therefore, all patients eligible under previous guidelines would be eligible under the 2013 guidelines with the addition of patients who meet the expanded eligibility criteria. This audit monitored the proportion of patients eligible for cART who were currently on cART to establish the adherence to the 2013 guidelines.
ART regimens used in this audit were compared against those recommended as first-line by the 2013 Zambian consolidated cART guidelines and 2013 WHO ART guidelines. The WHO recommended regimens were zidovudine (AZT) + lamivudine (3TC) + nevirapine (NVP); tenofovir disoproxil fumarate (TDF) + 3TC [or emtricitabine (FTC)] + efavirenz (EFV) and TDF + 3TC (or FTC) NVP; with a preference for TDF + 3TC (or FTC) + EFV. 4 The 2013 Zambian consolidated cART guidelines for first-line regimens were in accordance with WHO guidelines with the addition of abacavir (ABC) + 3TC (or FTC) + EFV regimen. 11

Statistics
The sample size required was calculated using the public health statistical programme Epi Info TM 7. The population size was 9000 as this is the number of HIV-positive patients who attend the cART outpatient clinic at Ndola Central Hospital. The expected frequency of adherence to WHO and Zambian Ministry of Health cART guidelines was set at 90% for initiation of cART. This expected frequency was chosen as 90% was the NASF target for the percentage of patients eligible for cART to be on treatment by 2015. 6 Using this expected frequency and confidence limits of 5% and a confidence level of 95%, it was calculated that a sample size of 136 was required. Public health statistical programme Epi Info TM 7 was used for descriptive statistical analyses of this study, including calculating interquartile ranges (IQR) and standard deviations (SD).

Epidemiology
A total of 221 patients attended the cART clinic during the study period. Information was collected from a total of 209 patients. The remaining 12 patients did not provide consent to participate, so they were excluded from the study. Of the patients, 60% were females and 40% males ( Table 2). The age of the patients ranged from 15 to 70 years; the median age was 43 years (IQR: 36-49); and the mean age was 42 years (SD ± 11). The median age that patients were started on cART was 36 years (IQR: 30-43); for men, this was 40 years (IQR: 34-48) and for women it was 34 years (IQR: 29-40).

Primary outcomes
Of the 209 patients in this audit, 204 met the eligibility criteria through clinical stage or immunological criteria for cART according to the 2013 WHO and Zambian cART guidelines. This was based upon whether patients currently met eligibility criteria or had previously been started on cART according to the guidelines at the time of initiation (Table 1). Four patients had been started on cART previously despite never meeting 2013 cART eligibility criteria. The remaining patient was clinical stage I, but did not have a CD4 count since diagnosis; therefore, it was impossible to determine their cART eligibility. One hundred and two patients met both immunological and clinical criteria, three patients met only clinical criteria and 99 patients were eligible on immunological criteria alone (Table 3).
Of the 204 patients who met eligibility criteria, 202 were currently on or had just been started on cART. For the two remaining patients, one was on tuberculosis (TB) treatment and a clinical decision had been made to delay cART initiation; and the other was presenting to the clinic for the first time. CD4 count results at initiation of cART were available for 181 of 206 patients and the range was from 1 cells/mm 3 to 626 cells/mm 3 ; the median was 147 cells/mm 3 (IQR: 84-242) and the mean was 168 cells/mm 3 (SD ± 117) ( Figure 2). Ninety five per cent (172/181) of the patients had a CD4 count of below 350 cells/mm 3 at initiation of cART, the 2010 WHO criteria for starting cART. Ninety eight per cent (178/181) of patients had a CD4 of less than 500 cells/mm 3 ; this is the level set by the 2013 WHO ART guidelines and the 2013 Zambian cART guidelines. The majority of patients were started on cART soon after diagnosis as they were eligible for treatment at diagnosis, 73% of patients were started on cART within one month of diagnosis.    (Table 1). There were seven patients started on cART since 2013 for which the Zambian 2013 cART guidelines would apply; because of the small number, they were put in the subgroup started after 01 January 2010 rather than being given a separate subgroup. One hundred and fifty-three patients had been started on cART before 2010 and 53 since 01 January 2010; there was information on clinical stage at initiation on cART for 152 patients started before 2010 and all 53 patients in the second group ( Figure 3). This was a descriptive study and no statistical analyses were performed for comparison between the two groups.

Secondary outcomes
The majority of patients (87%; 180/206) were currently on first-line treatment regimens. Eighty three per cent (149/180) of first-line regimens were those recommended by the 2013 Zambian ART guidelines and 81% (145/180) by WHO 2013 guidelines (Table 5). Sixty three per cent (114/180) of patients on first-line regimens were on the TDF + FTC + EFV regimen which is recommended in preference to other first-line regimens by both the 2013 WHO guidelines and the 2013 Zambian cART guidelines.
Fifty nine per cent of patients had their cART regimens changed at least once. Fifty per cent of regimen changes were because of side effects of which peripheral neuropathy and lipodystrophy were the most common. The second most frequent reason for regimen change was treatment failure (22%), with immunological treatment failure making up 72% of this total. The third largest group (21%) had their treatment changed to keep regimens in accordance with the Zambian cART guidelines.
Twenty nine per cent of the patients reported having one or more side effects from their current cART regimens. The most common side effects were peripheral neuropathy (in 6% of patients) followed by nausea and/or vomiting (5%) and then dizziness (4%) Peripheral neuropathy was most common in patients on cART regimens containing stavudine.
Of the patients, 202 reported their adherence to cART regimens; 78% reported perfect adherence, indicating that they had not missed a single dose in the last month; 13% reported good adherence, meaning that they had missed one dose in the last month; 6% reported average adherence, meaning that they had missed one dose in the last week; and 2% reported poor adherence, meaning that they had missed more than one dose in the last week ( Table 6).
The most prevalent risk factor amongst the study population was multiple concurrent sexual partners, which was reported by 28% of participants; other risk factors were first sexual contact < 15 years (13%) and transactional sex (11%) ( Table 7). Forty eight per cent of participants said they always or usually used condoms, while 48% reported they occasionally or never used condoms, with the remaining patients reporting abstinence over condom use (Table 8).

Discussion
Ninety eight per cent of patients were eligible for cART using current guidelines, and of these, 99% were currently on cART.
These results show that the cART clinic was adhering closely to the 2013 WHO and Zambian national guidelines on initiation of cART.
At initiation on cART, the majority of patients were stage III or IV ( Figure 1) and thus eligible for treatment on clinical stage alone. 4,11 The CD4 counts when patients were started on cART were generally very low (Figure 2), the median being 147 cells/mm 3 , well below 500 cells/mm 3 the level of eligibility for cART under the 2013 guidelines. These results show that most patients had advanced disease before they were started on treatment, which is likely to have a negative impact on health outcomes. 13, 14,15 CD4 count was stratified against clinical stage to assess if clinical stage was consistent with the degree of immunosuppression. It would be expected that the degree of immunosuppression (measured by decreasing CD4 count) would increase as the clinical stage increased; however, there did not appear to be a clear correlation between the clinical stage and CD4 count in this study (Table 4). This could be because patients all clinical stages had severe immunosuppression (CD4 < 200 cells/mm 3 ), meaning that even those who were asymptomatic would also be at high risk of developing diseases or symptoms of clinical stage III     Patients attending the clinic had been on cART for varying time periods; therefore, a subgroup analysis was performed. No statistical tests were performed to compare the subgroups; so it is not possible to make meaningful comparisons between the two subgroups. However, it appeared that a lower proportion of patients started on cART after 01 January 2010 were clinical stage III or IV than those started before 2010 ( Figure 3); this could suggest that patients were being diagnosed earlier.
Further research is needed with statistical power to test if there has been progress towards earlier initiation of cART.
Since this audit was conducted, the WHO released additional guidelines in 2015 which stipulate that cART should be started in all patients with HIV irrespective of their clinical stage or CD4 count. This would have affected the cART eligibility for five patients in this audit; however, four of these patients had already been started on cART despite not meeting eligibility criteria; the reasons for these patients being initiated on cART were not clear. It appears that late presentation to the cART clinic is the main factor delaying initiation of cART, as once patients attended the cART clinic their eligibility was assessed according to current guidelines and those that were eligible started on cART within a short period, the majority within a month of diagnosis.
Late initiation on cART may occur because of delay in referral to cART services from HIV counselling and testing (HCT) facilities, or late diagnosis as a result of low uptake of HCT programmes. This may mean that patients are only tested for HIV once they become symptomatic by which stage the patient's illness has already progressed significantly. Recent data have shown an increase in HIV testing in public health facilities. Further research is needed to investigate if these improvements in HCT result in earlier initiation on cART. It is important that these improvements in HIV testing are monitored to ensure they are maintained, and further progress made. In addition, good education about HIV is needed to improve uptake of HCT by informing people of the importance of early diagnosis and assisting early identification of symptoms associated with HIV. HCT facilities also need to have good referral systems to cART services for HIV-positive people as this can be another source of delay in initiation of cART. Early initiation of treatment can improve morbidity and mortality in patients with HIV and additionally reduce the infectivity, minimising further transmission. 13,14,15,16 A large majority (87%) of patients on cART were currently on first-line regimens of which 83% were in accordance with the Zambian Ministry of Health guidelines (Table 3). There seems to be a concerted effort to change patients onto regimens recommended by the national guidelines, with patients recently started on cART being initiated on the TDF + FTC + EFV regimen and patients on older regimens being changed to regimens in accordance with the Zambian cART guidelines; this is a common reason for regimen change. Just under one third of patients experienced at least one side effect on their current cART regimen, the most common being peripheral neuropathy, which was most common in those on stavudine regimens. The 2013 guidelines no longer recommend regimens with stavudine; this should help reduce the incidence of peripheral neuropathy as a side effect of ARVs.
The most common reported risk factor for HIV in this study was having multiple concurrent partners with 28% of patients reporting this (Table 5). In the Zambian demographics and health survey (ZDHS) from 2013 to 2014, 1.0% of women and 12.8% of men aged 15-49 reported having concurrent sexual partners within the last 12 months. 17 The prevalence of concurrent sexual partners appears higher in patients attending the ART clinic than in the general population in Zambia. In this study, 11% of patients reported transactional sex; in the ZDHS from 2013 to 2014, 12.8% of men reported paying for sex; however, there is limited information on the number of sex workers in Zambia. 17 Thirteen per cent of patients reported early sexual contact, the ZDHS from 2013 to 2014 found 11.7% of women and 16.2% of men aged 15-24 years reported sexual contact below the age of 15 years; this is similar to the prevalence in our study. Two per cent of patients in this study reported men who have sex with men (MSM) and 1% intravenous drug use (IVDU); there is currently very limited information on these key populations. However, the population council and partners are currently conducting research in this area of which the results are awaited. 18 In addition, almost half of the patients in the study reported that they only used condoms infrequently (  This may make the sample less representative of the population who attend the cART clinic, particularly as there could be bias towards patients who regularly attend the cART clinic. Some interviewees had problems understanding the questions because of language difficulties; some of the interviews were conducted with the help of translators; it is possible that these factors could lead to inaccurate answers. In some cases, information was missing from the notes, and where it was given, dates were sometimes omitted; this made it difficult to obtain a full data set for every patient, thus reducing reliability of results. There were no statistical analyses of reported risk factors which reduce the significance which can be attached to these results. Results on self-reported adherence to cART did not use a prior validated measure of adherence; this may reduce the reliability and comparability of these results. In addition, self-reported adherence often overestimates medication adherence. 21

Conclusion
Early initiation of cART can stop disease progression and improve long-term health outcomes. In 2013, new WHO and Zambian national guidelines on cART expanded the eligibility criteria for cART treatment. This study evaluated and audited the use of cART at the outpatient cART clinic at Ndola Central Hospital in Ndola, Zambia, comparing and analysing results against the 2013 Zambian cART guidelines and the 2013 WHO ART guidelines.
The study showed that almost all patients eligible for cART according to the 2013 cART guidelines were on treatment, the majority on appropriate cART regimens; however, many patients were initiated on cART late according to clinical and immunological criteria. This late diagnosis may be because of a lack of uptake of HCT; however, there have been improvements in HCT in Zambia over the past few years. Further research is also needed to assess the impact of increased HCT on earlier initiation of cART. It is important that HIV testing is monitored closely to ensure that testing levels are maintained and further increased, as well as good referral systems to cART clinics.
In conclusion, almost all eligible patients who attended the clinic were started on cART and in the majority of cases cART treatment was initiated soon after diagnosis. Initiation of cART tends to be late according to immunological and clinical criteria, which is probably because of late diagnosis of HIV. Interviews of patients highlighted the limited awareness of the influence of sexual practices and the need to take precautions to avoid the spread of HIV. Progress in tackling HIV requires continued expansion of HCT coupled with education programmes targeting groups at high risk.

Ethical consideration
A research proposal was approved by the hospital ethics committee. Each patient was assessed to check that they understood what participation in the study would involve. Written or verbal consent was obtained before participation in the study as explained above. Patients were informed that data that would be collected both from interviews and from their clinical records. All information was kept confidential and anonymous.