Comparison of State Prevention of Mother-To-Child Transmission of HIV and National Biennial ANC HIV Sentinel Surveillance Data: Ten Year Experience from Gombe State, North East Nigeria

There are significant variations in PMTCT programme implementation in the country. Biennial serosentinel survey among pregnant women attending antenatal clinics provides estimates of HIV and for monitoring the epidemic. The objective of this work was to compare HIV prevalence trend using National ANC HSS data and PMTCT programme data in Gombe state over the last 10 years. Methodology: Cross-sectional comparative study. The HIV prevalence among pregnant women in Gombe State obtained from Gombe State PMTCT programme data from 2004-2014 was compared with the National Biennial sentinel survey for HIV in pregnant women attending ANC in the state over the same period. Results: Women tested for HIV during ANC in Gombe state increased from 4689 in 2004 to 74,737 in 2014. 447,732 women were cumulatively tested for HIV with a positivity rate of 2.1% (9543). ANC HIV positivity rates from PMTCT programme data witnessed a decline from 8.2% (385/4689) in 2004 to 0.6% (497/74,737) in 2014. Conversely, the the national sero-prevalence survey averaged 4.2%. Conclusion: While routine yearly Gombe state PMTCT programme data showed a declining HIV trend, biennially conducted seroprevalence in the state was consistent over the period. Implications are unclear to us; accurate estimation of HIV prevalence is a prerequisite for planning.


Introduction
Trends in HIV prevalence are necessary to monitor the course of the epidemic, measure the effectiveness of control and prevention interventions, and plan further HIV control efforts. Timely and reliable data on HIV prevalence are gathered by HIV surveillance systems through the on-going systematic collection, analysis and reporting of data at different points in the HIV disease process [1] [2].
Since 1991, Nigeria's estimate of the HIV prevalence and burden has been through the conduct of biennial HIV sero-sentinel survey among pregnant women attending clinics throughout the country [3] [4]. In addition, National AIDS and reproductive health surveys (NARHS) provide further understanding of the dynamics of the disease in the country [5]. The National HIV prevalence declined from 4.1% in 2010 to 3.0% in 2014 with state, zonal and urban-rural variations [4] [5].
Antenatal clinics (ANCs) provide an accessible cross-section of healthy, sexually active women in the general population, and data from ANC HIV sero-sentinel survey (HSS) are considered to be generally representative of the underlying community [1] [2]. In many countries, national HIV prevalence estimates are substantially based on annual or biennial ANC HSS [1] [2] [3].
The Joint United Nations Programme on HIV/AIDS (UNAIDS)/World Health Organization (WHO) 2000 [6] and 2013 guidelines on second generation HIV surveillance [2] recommend conducting sero-survey among pregnant women as a core surveillance activity in concentrated and generalized HIV epidemics. Prevention Mother-to-Child Transmission of HIV (PMTCT) programmes in some sub Saharan Africa countries have witnessed substantial expansion in coverage (global report) [7] [8] [9]. As PMTCT programmes expand coverage and capture socio demographics, syphilis and HIV testing data similar to those collected by ANC HSS, many countries are considering the use of routinely collected PMTCT programme data to complement or replace ANC HSS The World Health Organization (WHO) recommends that five areas be evaluated before a successful transition from ANC HSS to surveillance using routine antenatal HIV testing can be considered. These are: agreement between ANC SS and routine antenatal HIV test results; magnitude of selection bias in routine antenatal testing compared to ANC SS data; all ANC HSS sites that should provide routine antenatal HIV testing; the quality of routinely collected antenatal data and the state of routine HIV testing quality assurance practices [1].
Nigeria accounts for the 28% of the global burden of mother to child transmission of HIV with about 58,000 of infants delivered with HIV infection in 2014 [10] and there are significant regional and state variations in PMTCT programme implementation in the country [3] [4] [11]. Ebonyi State (Southeast Nigeria) was able to supersede the 80% coverage target while Nassarawa (Northcentral) Gombe (Northeast) and the Federal capital territory-Abuja had over 70% coverage. Seven states were unable to counsel and test up to 20% of pregnant women [11].
Since the inception of the PMTCT programme in Nigeria in 2002, of the 36 states in the country, Gombe State has consistently maintained its position as one of the three leading states in PMTCT coverage nationwide [11]. Also, the dried blood spot DNA-PCR for early infant diagnosis (EID) of HIV was first introduced within the State [11].

Aim
The aim of this report is to compare HIV prevalence trends using National ANC HSS and with that obtained from PMTCT programme data in Gombe state between 2004 and 2014.

Gombe State
Gombe state is one of the 6 states in North east Nigeria and one of the 36 in the country and has an estimated population of 3.0 million ( Figure 1) [12].
There are 594 health facilities, of these, 21 general hospitals, 1 state owned specialist Hospital, 1 Federal Teaching Hospital, 68 private health facilities, and 517 government owned primary health facilities [13].
Women of child bearing age (15 -49 years) are estimated as 666,260. The population of children aged 5 years and below is 605,691 (20% of total population). The annual growth rate is 3.2%. The total fertility rate is 7 [13]. Antenatal care provided by skilled Health workers is at 58.2% [14].
Study design This is a retrospective prevalence-based comparative study of national HIV sero-prevalence sentinel survey among pregnant women attending ANC in Gombe State and PMTCT routine data for the State from 2004-2014. Gombe State agency for the control of AIDS (GOMSACA) was established in 2002 [15] [16]. The national HIV sero-prevalence sentinel survey was conducted by the Federal Ministry of Health, Nigeria among pregnant women attending ANC in Gombe state and initially involved two then subsequently four sites; with Gombe and Kaltungo as urban and Kwami and Zambuk as rural sites. The  [4]. Women who participated in the sentinel survey were pregnant women of all ages, pregnancy was confirmed by health provider on-site, the women were attending ANC for the first time for that particular pregnancy and they accepted routine ANC test.
PMTCT services in Gombe State started in September 2004 in 9 health facilities and has been scaled up to involve 201 PMTCT sites currently. PMTCT data was collected from each facility records and reviewed the by State agency for the control of AIDS on a yearly basis from 2004 till 2014 [17]. HIV Group information and testing for PMTCT was followed by post-test counselling and opt-out was the strategy in all ANC clinics in the state [18]. Non laboratory staffs have been trained on HIV testing with quality control/assurance processes put in place [19]. HIV testing algorithm is serial testing using recommended HIV test kits with high sensitivity and specificity including Determine R, Unigold and Stat Pak [20].
A multi-disciplinary team formed in 2004 coordinates PMTCT services in Gombe State [17]. Coordinating teams were formed in all health facilities offering PMTCT services and regular coordination meeting, programme review sessions, quarterly monitoring and supportive supervisory meetings are held and  Table 1). The number of sites for the National HIV sero-sentinel survey in Gombe state remained 2 rural and 2 urban sites over the last 10 years. Also, the number of pregnant women tested during the HSS remained at 900 (Table 2). Although the prevalence of HIV among pregnant women in 2005 obtained from Gombe State PMTCT data was higher than Sero-sentinel survey (7.5% vs 4.9%); the State PMTCT HIV prevalence rate demonstrated a steady decline till 2014 ( Figure 2). From 2005-2012, the National HSS prevalence for Gombe state was consistently up to 4%, with a slight decline to 3.4% in 2014 ( Figure 2). HIV prevalence for Gombe state PMTCT programme data and ANC HSS were 0.6% and 3.4% respectively in 2014 ( Figure 2).

Discussion
To our knowledge, this study is the first to compare the State HIV PMTCT programme data with ANC HSS data over a 10-year period in Nigeria. Nigeria is currently witnessing a scale up in PMTCT programmes with Gombe state as one of the three states in the country with PMTCT coverage of over 70% as at 2015 [11]. National PMTCT coverage is 30.2% [43].
Comparing ANC sentinel surveillance and PMTCT programme is relatively simple and can provide important information [43].  [50]. While the WHO recommends that five areas be evaluated before a successful transition from ANC HSS to surveillance using routine antenatal HIV testing can be considered, mature PMTCT programmes at subnational levels can provide reliable data to inform decision making at the national levels [50].
The Gombe state PMTCT program has achieved over 70% [11] coverage; therefore comparing trends in PMTCT programme data and ANC HSS becomes The scaling up of HIV testing by non-laboratory staff is a major factor in uptake and demonstrated increased coverage of over 70%; this was supported by effective quality assurance and quality control processes supervised by laboratory scientists. Mirkuzie et al. [52] compared PMTCT and ANC HSS data in Ethiopia over 2 years and concluded that PMTCT data can be used to monitor HIV trends. HIV prevalence rate in the 2 groups studied by the Ethiopian researchers were 6.2% and 4.5% and 6.1% and 5.5% in 2008 and 2009 respectively [52].
Their study was however limited by the short duration (2008-2009) and low coverage. Similarly in a study whose duration was short and PMTCT coverage varied between 6% and 76% Kumar et al. [53] [55], Kenya [56] and Cameroon [57] have however suggested the need for additional preparation including strengthening quality assurances of HIV rapid testing and data quality practices before transitioning to PMTCT data for HIV surveillance.

Conclusion & Recommendations
Gombe state PMTCT programme coverage is high and increasing with high HIV testing acceptance rate. As HIV testing service quality and regular data quality assessment and validation are strengthened in the wake of decentralization, the HIV prevalence data of the Gombe state PMTCT programme has progressively become lower than the National ANC HSS. While the implications for these observations are unclear, accurate estimate of HIV prevalence and burden is critical determinants for planning of impactful interventions.
A sub-regional wide evaluation of PMTCT programme data and ANC HSS in countries with high PMTCT coverage is recommended to guide transitioning from the use of ANC HSS to PMTCT programme data for HIV trend estimation.

Limitations
A larger national or sub-regional study would have been better able to demonstrate the comparison between ANC HSS and PMTCT data across a larger subset of Nigeria's population. However, due to the wide disparity in PMTCT coverage between states and various regions in the country, that was not feasible.