Application of the UNAIDS Incidence Patterns Model to Determine the Distribution of New HIV Infection in Lagos State, Nigeria

Abstract

Background

Identifying patterns in the distribution of new HIV infections in the population is critical for HIV programmatic interventions. This study aimed to determine the distribution of New HIV infection by applying the incidence patterns mathematical model to data from Lagos state.

Methods

The incidence patterns model (IPM) software is a mathematical model developed by UNAIDS to estimate the demographic and epidemic patterns of HIV infections. This model was adapted in Lagos state to predict the distribution of new HIV infections among specified risk groups in the next 12 months.

Results

The IPM predicted a total HIV incidence of 37 cases per 100 000 individuals (3979 new infections) will occur among the 15 to 49 subpopulations. The results also showed that sero-concordant HIV-negative couples with external partners (29%), female sex workers (26%), men-having-sex-with-men (18%), and previously married females (6%) accounted for the majority of the estimated new HIV infections. Overall, key populations constitute almost half (48%) of the estimated number of new HIV infections.

Conclusion

The study helped to identify the population groups contributing significantly to new HIV infections. Therefore, priority interventions should be focused on these groups.

Background

Lagos state is in the southwestern part of Nigeria. It is the smallest state in the country in landmass but has the highest urban population, accounting for 27.4% of the national population. The state was reported to have about 24.6 million inhabitants in 2015. ¹ Lagos state remains the economic, financial, and commercial nerve center of Nigeria and the ECOWAS region, with gross national product (GNP) three times that of any West African Country, thus making Lagos state the ECOWAS economic hub and the springboard for innovation and development in Nigeria and sub-Saharan Africa (SSA). ¹ Lagos recorded its first two cases of HIV in 1985 in a sexually active 13-year-old female and a commercial sex worker. ² According to a 2018 Nigeria HIV/AIDS Indicator and Impact Survey (NAIIS) report, the prevalence of HIV among adults (15-64 years) in the state was 1.4%. ³

The State Response

HIV has been a major public health crisis in Nigeria. Since the first reported case of HIV in Lagos, the Lagos State Government has continued to intensify efforts at combating the epidemic at the state level. One example is the establishment, by an act of parliament, of Lagos State AIDS Control Agency (LSACA) in 2003. In 2016, the Lagos State Government committed to achieving the Global Targets on HIV epidemic control by 2025 (through the global 95-95-95 Initiative) and ending AIDS as a public health threat in 2030. ⁴ In 2019, the Lagos State Government partnered with the United States Government (USG) to provide free antiretroviral treatment (ART) to persons living with HIV in the state under the U.S. President's Emergency Plan for AIDS Relief (PEPFAR) and was named America's best partner in fighting HIV in Nigeria. ⁵

The government has also ensured that HIV testing and counselling are provided free in health facilities and ARVs are readily available to those who require it at no cost. Youth and Adolescent Friendly Services Centres were also established to educate and enlighten young citizens across all local government areas (LGAs) in the state.

The City of Lagos is one of 15 global high HIV-burden cities receiving support through the Fast-Track Cities Project (FTC), which aims to accelerate access to prevention, treatment and care services for people living with and at risk of HIV and key populations. The FTC project is implemented with funding from the United States Agency for International Development (USAID). The project contributed to the successes in moving the overall HIV response forward and culminated in the city being awarded the “Circle of Excellence Award” by the Fast-track Cities Institute in 2022. ⁶

Lagos state has recorded some successes in the fight against HIV/AIDS, however, when compared with the global 95-95-95 target—which is aimed at ensuring 95% of persons with HIV are diagnosed through testing, 95% of those diagnosed are placed on ART, and 95% of those on ART achieve viral load suppression—the state is still lagging, particularly in terms of achieving viral load suppression.^7,8 According to the Lagos State Ministry of Health 2021, the current HIV cascade for all age groups in Lagos state is estimated at 89-89-65. ⁶ An estimated 160 000 people are living with HIV in the state, out of whom 89% (142 224) are aware of their HIV status. Furthermore, 89% (126 700) of those aware of their status are on ART, and only 65% (82 577) have achieved viral load suppression. ⁶

Rationale for State Modelling and Synthesis

Nigeria like other countries in SSA continues to bear the burden of the HIV epidemic HIV interventions in Nigeria are largely funded by donors (80%). Unfortunately, this funding has continued to decline over the past years. The Lagos State Government has proactively continued to increase funding towards interventions to reduce the HIV epidemic at the state level and requires specifically tailored tools to inform HIV programmatic decision-making. To ensure that these interventions achieve their objectives fully, they must be implemented using an evidence-based approach. This requires identifying the population groups contributing the highest number of new cases (incident infections) of HIV together with strategically focused interventions to reduce the rate of new infections among these populations and by extension the state. Mathematical models provide a cost-effective alternative to prospective cohort studies to estimate HIV incidence.

The modes of transmission (MOT) model developed in 2003 was applied in Nigeria in 2010 to estimate the distribution of HIV incidence over the next 12 months and has been widely used for program and policy planning. ⁹ However, the model was limited in that it requires estimates of population sizes and behaviors (such as number of partners and acts per partner) that may not be readily available or poorly measured. In addition, since it focuses on a single point in time it may not fully represent the dynamics of the HIV epidemic and the changing contributions of different population groups over time to HIV transmission. ⁹

In response to the limitations of the MOT model, the incidence patterns model (IPM) was developed by Imperial College in 2016. ¹⁰

Overall, the Incidence pattern model is a proven tool useful in conceiving HIV preventive strategies, achieving the global 95-95-95 target and preventing misuse of limited financial and manpower resources. The model also aims to ensure that interventions are strategically focused on populations who require them most. The model stratified the population according to factors associated with HIV acquisition risk such as sexual activity, circumcision, etc. In 2020, UNAIDS applied the IPM in Nigeria to determine the distribution of HIV incidence in the different states of the country. In Lagos state, the contribution of key populations which include men having sex with men (MSM), female sex workers (FSW), and people who inject drugs (PWID) corresponds to almost half of the total new distribution of HIV infections reported.

Methods

Overview of the Incidence Patterns Model

The IPM is a mathematical model developed by UNAIDS to address the limitations of the earlier MOT studies and has been in use for several years. The IPM uses a generic approach for estimating the distribution of newly acquired HIV infections according to identifiable determinants of risk. It is a single-time-step compartmental deterministic model which builds on prior historical information on incidence patterns in the region to predict the distribution of new infections acquired in the next 12 months.

IPM provides an alternative to prospective cohort studies (which are costly and cannot be routinely implemented) to estimate HIV incidence. It offers an improved methodological framework and is better adapted to programmatic needs. ¹⁰

Data Inputs

The UNAIDS IPM was adapted for Lagos state to determine the distribution of new HIV infection by population group in the state. UNAIDS provided an Excel input file with pre-populated parameters already stratified to specific population groups. The research team then filled the inputs using numerous data sources in Lagos state and analyzed the data with IPM software using the R studio.

Inputs to the modelling software include population distribution, HIV prevalence, ART coverage, HIV incidence, etc. The population were further disaggregated into other parameters provided within the modelling software by UNAIDS as indicated below.

Stratification of the Population

In the IPM adapted, the adult population was stratified according to factors associated with HIV acquisition risk. The population is disaggregated by sex and marital/sexual activity status or membership in key populations, HIV and ART status for those in unions, and circumcision status. Men except MSM and men who are not sexually active were stratified by circumcision status, as being circumcised reduces the risk of infection. In the present study, estimates of the effects of male circumcision appeared to be rather unreliable as less than 4% of men in the 2018 NAIIS study were uncircumcised.

Married or cohabiting heterosexual unions were disaggregated according to Sero's status. Sero-discordant unions have either the man or the woman HIV-positive. The model allows for differences in the proportion of males and females in unions but does not explicitly consider polygamous unions. Unions were also disaggregated by ART status as HIV-positive individuals on ART have a reduced risk of onward transmission. However, the model did not stratify the individuals based on detectable or undetectable viral load.

Proportions were calculated for each parameter in the different subgroups—unions, never-married men and women, previously married men and women, and key populations which include men who have sex with men, FSW, and men and women who inject drugs. Mean and variance were also calculated for the duration of sexual activity.

The respective risk groups making up the adult population (15-49 years) considered by the model are defined in Table 1.

Table 1.

Stratification of Population Groups (Age 15-49 Years) by Sex and Marital/Sexual Activity Status or Membership in Key Populations and HIV Status for Those in Unions.

Group	Population group	HIV status (man)	HIV status (woman)	Definition	HIV transmission
1	Female PWID		Includes both + and − individuals	Adult females who inject drugs	Transmission occurs from having external sexual/casual partners
2	Male PWID	Includes both + and − individuals		Adult males who inject drugs	Same as above
3	MSM	Includes both + and − individuals		Men who have sex with men	Same as above
4	FSW		Includes both + and − individuals	Female sex workers	Same as above
5	Previously married females		Includes both + and − individuals	Women who were married but are now divorced, widowed, or separated	Transmission occurs from having external/casual partners
6	Previously married men, uncircumcised	Includes both + and − individuals		Men who were married but are now divorced, widowed, or separated and are not circumcised	Same as above
7	Previously married men circumcised	Includes both + and − individuals		Men who were married but are now divorced, widowed, or separated and are circumcised	Same as above
8	Never married females		Includes both + and − individuals	Women who have never been married; will be young women	Transmission occurs from having external/casual partners
9	Never married males, uncircumcised	Includes both + and − individuals		Men who have never been married and are not circumcised	Same as above
10	Never married males, circumcised	Includes both + and − individuals		Men who have never been married and are circumcised	Same as above
11	Sero concordant unions, male uncircumcised	-	-	People in a union where both the man and the woman are HIV-negative, and the man is not circumcised	Transmission typically occurs from having external/casual partners
12	Sero concordant unions, male circumcised	-	-	People in a union where both the man and the woman are HIV-negative, and the man is circumcised	Same as above
13	Sero discordant unions, male positive	+	-	People living in a union where the man is HIV-positive, and the woman is HIV-negative	Transmission could occur from inside as well as outside the relationship
14	Sero discordant unions, female positive, male uncircumcised	-	+	People living in a union where the female is HIV-positive, and the man is uncircumcised	Same as above
15	Sero discordant couples, female positive, male circumcised	-	+	People living in a union where the female is HIV-positive, and the man is circumcised	Same as above

“+” = HIV-positive; “-” = HIV-negative.

Data Source for the Modelling

The fundamental principle behind the IPM is its reliance on available data. Data for the IPM were obtained by exploring key documents and past HIV studies within the state and the country. The National Demographic and Health Survey (NDHS) conducted in 2018 in Nigeria is a high-quality household survey that provides information on key topics including HIV prevalence, antenatal care, family planning etc among adults aged 15 to 49 years old. ¹¹ We used state-level data from the NDHS that covered population distribution, sociodemographic, marital status, HIV status, ART status, etc.

State-level data on population size and HIV prevalence among key populations, as well as duration of exposure to risk practices needed for this model, were obtained from the Nigeria HIV/AIDS Indicator, Impact Survey (NAIIS) conducted in 2018, and the Integrated Bio-Behavioral Surveillance Survey (IBBSS) conducted in 2020. ¹² Data on Key Populations were obtained from Mapping and Size Estimation of Key Populations in Nigeria: Six States and the Federal Capital Territory, 2019. ¹³

Data on the total number of new infections among the age group 15 to 49 years in the country in the past year were needed for the model and this was obtained from the National 2020 report. ⁹ Other additional data sources for incidence estimates were retrieved from the Lagos state HIV program data for 2021 and the available mid-year data for 2022. ¹⁴

The input parameters and data sources used for the IPM analysis are presented in Table 2.

Table 2.

Lagos State Data Inputs and Data Sources for IPM Analysis.

Parameter	Sample size	Proportion	Data source
Population distribution
Sero-discordant/w. pos	42	0.646153846	NAIIS 2018 3
Sero-discordant/w. pos/m. circ	42	1	Same
Sero-discordant/w. pos/m. uncirc	0	0	Same
Sero-discordant/m. pos	23	0.353846154	Same
Sero-concordant unions	1694	0.981798715	Same
Sero-concordant unions/m. neg	1684	0.994096812	Same
Sero-concordant/both neg/m. circ	537	0.31888361	Same
Sero-concordant/both neg/m. uncirc	7	0.00415677	Same
Sero-concordant/both pos	10	0.005903188	Same
Married or cohabiting men	1677	0.349666389	Same
Never married man/m. circ	1537	0.978357734	Same
Never married man/m. uncirc	34	0.021642266	Same
Previously married man/m. circ	114	0.991304348	Same
Previously married man/m. uncirc	1	0.008695652	Same
Not a sexually active man	1205	0.251251043	IBSSS 2020 12
Married or cohabiting women	2731	0.619555354	Same
Never married woman	1465	0.482542819	Same
Previously married woman	582	0.103817339	Same
Not sexually active woman	1205	0.21494827	Same
MSM	4828	0.000603367	Lagos State Mapping 13
FSW	40 863	0.005532314	Same
Male PWID	3995	0.000499265	Same
Female PWID	1332	0.000180335	Same
Biological parameters
Mean ART coverage	100	0.37	NDHS 2018 11
Mean duration of sexual activity never married men	7.951646	39.955434	Same
Mean duration of sexual activity never married women	6.72169	52.63282	Same
Additional parameters
Couples	1759		NAIIS 2018
Men	4796		Same
Women	5606		Same
Total new infections	3900		UNAIDS fact sheet 14
Total infections (prevalence)	120 000		Same
Total new infections, men	1677		Same
Total new infections, women	2223		Same
Never married men	1651	0.473982359	NAIIS 2018
Total population size	15 388 000		United Nations 15
Proportion aged 15 to 49	0.86		NAIIS 2018

M = men; w = women; pos = HIV positive; neg = HIV negative; circ = circumcised; uncirc = uncircumcised.

Model Fitting and Data Output

The Lagos Excel input data was analyzed on R studio using the IPM software. This then produced an output Excel file which estimated the number of new HIV infections among the various stratified population groups. The Excel file had several worksheets which included the maximum Likelihood, median and mean of the New HIV distribution. The IPM software also calculated uncertainty bounds (confidence intervals) around each estimate. These define the range within which the true value lies.

Ethical Considerations

This study exclusively made use of secondary data sources available in the region's public domain. No personal or sensitive information was disclosed or used in a manner that would infringe upon the rights or privacy of any individual. The data sources used in the study were anonymized and therefore cannot be linked to any individual. Permission was also obtained from the LSACA for their use. All sources of data utilized have been properly cited and referenced.

Results of Modelling

Figures 1 and 2 show the distribution of new HIV infections by population group in Lagos State. The estimated total number of new adult HIV infections across all population groups is three thousand nine hundred and seventy-nine (3979). Sero-concordant HIV-negative couples, FSW, and MSM account for the largest proportions of new adult HIV infections at 29.0%, 26.1%, and 17.6%, respectively. Previously married women contributed 6.4% of new infections while never married women contributed 6.1%. Men who inject drugs (MRDS) contribute 4.0% of new infections while females who inject drugs (fPWIDs) contribute 0.4% of new infections. The key populations in the study comprise FSW, MSM, mPWID, and fPWID. These groups, which constitute less than 1% of the total adult population in Lagos state, account for almost half (48.1%) of the total number of new HIV infections.

Figure 1.

Distribution of new HIV infections by population group in Lagos state.

Keys: F = Female; M = Male; mPWID = adult males who inject drugs; fPWID = adult females who inject drugs; MSM = Men who have sex with men; FSW = Female sex workers; FPrevMarried = Previously Married Female; MPrevMarriedCirc = Previously Married Male Circumcised; MPrevMarriedUnCirc = Previously Married Males Uncircumcised; FNeverMarried = Never Married; MNeverMarriedCirc = Never Married; CSCnegCircum = Sero-Concordant HIV Negative Couple Union Men Circumcised; CSCnegUnCirc = Sero-Concordant HIV Negative Couple Men Uncircumcised; CSDMpos = Sero-Discordant Union Male HIV Positive; CSDMnegCirc = Sero-Discordant Union Female HIV Positive Men Circumcised; CSDMnegUnCirc = Sero-Discordant Union Female Positive HIV Men Uncircumcised.

Figure 2.

Percentage distribution of new HIV infections in Lagos state.

Keys: F = Female; M = Male; mPWID = adult males who inject drugs; fPWID = adult females who inject drugs; MSM = Men who have sex with men; FSW = Female sex workers; FPrevMarried = Previously Married Female; MPrevMarriedCirc = Previously Married Male Circumcised; MPrevMarriedUnCirc = Previously Married Males Uncircumcised; FNeverMarried = Never Married; MNeverMarriedCirc = Never Married; CSCnegCircum = Sero-Concordant HIV Negative Couple Union Men Circumcised; CSCnegUnCirc = Sero-Concordant HIV Negative Couple Men Uncircumcised; CSDMpos = Sero-Discordant Union Male HIV Positive; CSDMnegCirc = Sero-Discordant Union Female HIV Positive Men Circumcised; CSDMnegUnCirc = Sero-Discordant Union Female Positive HIV Men Uncircumcised.

Table 3 shows that sero-concordant HIV-negative couples contributed the highest number of infections, at 1149 infections. This was significantly higher than new infections reported in couples in sero-discordant relationships, which was estimated at 92.

Table 3.

Distribution of Sero-Concordant Negative Couples and Sero-Discordant Couples by Population Size, Incidence, and Number of New Cases.

Population group	Average population size	Incidence rate estimates (%)	New infections	Incidence rate per 100 000
Sero-Concordant HIV-Negative Couples
Males Uncircumcised	73 395	0.04	33	46
Males Circumcised	4 367 524	0.03	1116	26
Sero-Discordant Couples
Males Positive	43 661	0.22	46	105
Females Positive, Male Uncircumcised	22 576	0.34	38	170
Females Positive, Male Circumcised	48 627	0.03	8	15

Table 4 shows the incidence rate in key populations which ranged between 0.7% and 2.1%. The highest number of new infections was in FSW, while the lowest was recorded in females who inject drugs.

Table 4.

Key Population Distribution by Population Size, Incidence, and Number of New HIV Infections.

	Average population size	Incidence rate estimates (%)	New HIV infections	Incidence rate per 100 000
fPWID	2224	0.71	15	682
mPWID	21 228	0.85	159	751
MSM	136 408	0.55	702	514
FSW	71 713	2.11	1039	1449

As displayed in Table 5, incidence rates were similar in the previously married and never-married populations, with 255 new infections in previously married females and 242 new infections in never-married females. The incident rates were lower in circumcised males across both groups.

Table 5.

Distribution of the “Previously Married” and “Never Married” Population by Population Size, Incidence, and the Number of New Infections.

Population group	Average population size	Incidence rate estimates (%)	New infections	Incidence rate per 100 000
Previously married
Females	435 728	0.06	255	59
Males uncircumcised	9335	0.17	12	129
Males circumcised	1 191 630	0.02	182	15
Never married
Females	2 043 984	0.01	242	12
Males uncircumcised	65 291	0.15	98	150
Males circumcised	2 201 444	0.00	34	2

Discussion

This study was aimed at identifying drivers of the HIV epidemic in Lagos State by providing estimates of incidence by population group as well as the distribution of new infections. In this study, the total new HIV infections for Lagos state was estimated at 3979, which is slightly higher than the 3662 reported by the UNAIDS finding in 2020. The sero-concordant HIV-negative couples with external partners, FSW, previously married circumcised men, and previously married females account for the majority of the new infections observed in the adult population. Overall, the key populations comprising FSW, MSM, and PWID, which constitute less than 1% of the total population in Lagos state, account for almost half (48.1%) of the total number of new HIV infections.

Sero-concordant HIV-negative couples (with HIV transmission occurring through external unions) contributed to over one-quarter (29.0%) of new HIV infections. In Lagos state, the impact of male circumcision status on the distribution of HIV new infections is less relevant due to the small number of uncircumcised men in the state as reported by NAIIS 2018. ³ As for the sociodemographic characteristics of the selected population in the group, data sources revealed that a considerable proportion of the males in sero-concordant HIV negative relationships were between 30 and 40 years and 60% had completed senior secondary school education or below. Over four-fifths of them were reported to have ever been married and 70% were employed at the time of the survey. Regarding their sexual characteristics, more than half (58.3%) of the male population were sexually active before 21 years with over four-fifths already sexually active by 25 years. Furthermore, about 17.3% of the male populations in the group were reported to have had at least two sexual partners in the last 12 months preceding the data collection and only 5.6% were reported to have consistently used condoms in the sexual encounters.

In this study, key populations contributed the highest percentage of new infections at 48.1% of new infections. This is similar to the findings reported in the national IPM study conducted in 2020, where key populations contributed 42.1% of new infections in Lagos. The study suggests that MSM contribute about 17.6% of new HIV infections, a slightly higher increase than the 13.4% reported by the 2011 Lagos MOT study but significantly higher than the 5.1% reported by the 2020 National IPM study for Lagos state. The reason for this disparity in figures may be attributed to the data utilized in both studies as the current study applied actual figures obtained from the 2019 Lagos state mapping and size estimation of key populations. This contrasted with the 2020 study which utilized standard estimates obtained from studies fitting the model to cohort studies in the sub-Saharan region.^10,13,16,17

Data sources used for the input, such as the mapping of key populations in Lagos state show that over half (56%) of the participating MSM were between the ages of 25 and 34 years and approximately two-thirds of the population captured engage in sex work. ¹³ The mapping revealed a sparse hotspot distribution and nonspecific hang-out locations among the population which could indicate a highly mobile and dynamic group which socialize exclusively via online platforms. Previous key population mapping conducted in the state in 2015 revealed that approximately forty-five (45.2%) of the MSM community have had sex with a woman in the last 12 months and about 36.3% did not use condoms in the last sexual act. This indicates the bisexual behavior of a viable proportion of the target population with possible sexual transmission among their female heterosexual partners.

FSW also contribute significantly to new infections. The data sources inputted in the study showed that 62% of the FSW mapped were between 25 and 34 years old. Over four-fifths had only senior secondary education or below. About 75% of the FSW population was non-brothel based while hotels and lodgings accounted for about two-thirds of the total non-brothel based FSW. Mapping of key populations conducted showed a sharp influx of FSW was noticed in the state in the month of December when there are many celebrations around the end of the year and there is an influx of visitors from the diaspora and other states into Lagos.

For PWID, a significant percentage were over 25 years and a little above half had senior secondary or higher school as the highest level of education. Only a minority of the population are women (14%) of which 45% are sex workers. The 2015 mapping reported that about one-fourth (25.4%) of PWID were identified in Lagos Island LGA while Ojo LGA has the least estimated population (0.63%). In many hotspots for PWID, males are more likely to be observed compared to females.

The never-married female population in our study contributed only 6.1% of the total new infections. This is significantly lower than the 29.5% reported for the population in the 2020 study. This difference can be explained on the basis that this present study utilized several data sources such as raw data from both NAIIS and IBBSS. These data were cleaned, and missing data were excluded before stratification was done based on the required IPM input variables. The disparity observed may also be attributed to effective HIV prevention efforts aimed at the never-married female population and/or an increase in ART coverage thereby limiting HIV spread from HIV-positive sexual partners.

The 2011 MOT study conducted in Lagos revealed that low-risk heterosexuals contributed 55.3% of new infections at the time. In this present study, low-risk heterosexuals represent about 28.0% of new infections in adults (15-49 years). Though it is a low-risk group they make up a large part of the adult population. Our findings suggest that Lagos State has a mixed HIV epidemic with a disproportionate concentration among key population. ¹³ This is in keeping with the research by Tanser et al ¹⁸ which revealed the occurrence of concentrated sub-epidemics of HIV among key populations within the generalized epidemic contexts in Sub-Saharan African countries. Borquez A et al ¹⁰ using the IPM also reported a significant contribution of the key population, particularly FSW, to the distribution of New HIV infections in SSA.

Our findings reiterate the importance of targeted interventions for key populations but highlight a need for programming targeted at groups conventionally thought of as low risk, but who often engage in high-risk sex with external partners.

Study Limitations

There are some limitations encountered in the study. The IPM software uses a Bayesian approach for estimation, therefore the results can slightly vary if the code is run on different computers or at different times. There were missing data in the NAIIS study, which was the major data source used in the study, hence additional sources of data were explored as discussed in the methodology. The IPM does not include partners of individuals that make up the key population and other population groups such as sero-concordant HIV-negative couples where the transmission occurs from external partners. Also, the data sources used did not take into consideration prevalent cultural practices such as polygamy. Additionally, the majority of the data sources used in the research represented adults aged 15 to 49 years except for NAIIS study which covered all age groups.

Conclusion and Recommendations

Findings from the IPM showed that in Lagos State, Nigeria, Sero-concordant HIV-negative couples and the key populations contributed significantly to the projected New HIV infections in the next 12 months. These population groups according to data sources exhibited high-risk sexual behaviors, which can increase the risk of HIV infection. Therefore, tailored sexual health interventions and innovative strategies using behavioral change communication, information, education, and communication as well as capacity building on differentiated service delivery (DSD) should be implemented and scaled up.

Data availability was a major obstacle encountered in the study, therefore adequate efforts should be made to improve data collection at subregional levels, especially among key populations through the utilization of modern technology and internet tools for mappings and identification. This will enable relevant stakeholders to better understand the epidemic, identify remaining gaps and barriers to service uptake and provide adequate response. Furthermore, to achieve a robust and all-inclusive prediction of HIV incidence by the model, we recommend the inclusion of other typologies such as transgender people, disabled and people in confined spaces into future HIV incidence model tools.

Finally, studies as such which show where new infections are occurring in the community should be adopted and conducted in other states/cities in the country to help identify the gaps in the National HIV prevention efforts requiring strategically targeted programs.