Mapping of New HIV Infections in Morocco and Impact of Select Interventions

Supplementary Information

Mapping of new HIV infections in Morocco and impact of select interventions

Silva P. Kouyoumjian a, Houssine El Rhilanib, Amina Latific, Amina El Kettanic, Hiam Chemaitelly a, Kamal Alamib, Aziza Bennani c, and Laith J. Abu-Raddad a, d*

a Infectious Disease Epidemiology Group, Weill Cornell Medicine - Qatar, Cornell University, Qatar Foundation - Education City, Doha, Qatar

b Joint United Nations Programme on HIV/AIDS, Morocco Country Office, Rabat, Morocco

c Morocco National AIDS Programme, Ministry of Health, Rabat, Morocco

d Department of Healthcare Policy and Research, Weill Cornell Medicine, Cornell University, New York, New York, USA

*Corresponding author:Prof. Laith J. Abu-Raddad, Infectious Disease Epidemiology Group, Weill Cornell Medical College - Qatar, Qatar Foundation - Education City, P.O. Box 24144, Doha, Qatar. Telephone: + (974) 4492-8321. Fax: + (974) 4492-8333. E-mail:

1.MODES OF TRANSMISSION MODEL STRUCTURE

The MoTModel was used to estimate the expected short term incidence of HIV infections among the adult population stratified by mode of exposure.1-6The model was developed by the the Joint United Nations Programme on HIV/AIDS (UNAIDS) Reference Group on Estimates, Modelling and Projections.2It uses input data on the current prevalence of HIV infection, the number of individuals in particular risk groups, and the risk of exposure to HIV within each risk group to estimate the contribution of the different risk groups in the population to incidence over a period of one year. The model assumes the binomial function7 in terms of the number of partners and number of contacts with each partner to estimate the risk of infection for a susceptible individual in any particular risk group. Individuals are classified according to their levels of risk behavior. The definitions of each of these population groups are summarized in Table S1.1 and in previous publications.4-6Further details on the model structure can be found on the UNAIDS website.2

Table S1.1Definitions used in the Morocco National and Souss Massa DraaModes of Transmission Models to classify individuals according to risk behavior.

Risk group / Definition
Key populations / Individuals who are PWID, MSM, or FSWs
Most at risk network / Individuals who engage in a risk network of PWID, MSM or FSWs and their clients
Individuals engaging in casual heterosexual sex (intermediate risk group) / Individuals who have had more than one sexual partner in the last year
Partners of PWID, partners of clients of FSWs,
female partners of MSM, partners of individuals engaging in casual heterosexual sex / Individuals whose partners are allocated to the PWID, clients of FSWs, MSM or casual heterosexual sex risk groups
Individuals engaging in low-risk heterosexual sex (stable heterosexual couples) / Individuals who have had one heterosexual sex partner in the last year and their partners have also had only one heterosexual sex partner in the last year. This group consists of individuals in monogamous partnerships who are not only faithful to their partners and their partners are faithful to them, but their partners do not belong to PWID, MSM, or clients of FSWs
Individuals at no risk / Individuals who have had no sexual or injecting partners in the last year. This may be a result of primary abstinence (never had sexual intercourse) or secondary abstinence (had sexual intercourse in the past, but not currently sexually active)

PWID, people who inject drugs; MSM, men who have sex with men; FSWs, female sex workers.

2.NATIONAL MODES OF TRANSMISSION MODELPARAMETRIZATION

1

The main sources of data that were used to estimate the parameter inputs in the Morocco National and Souss Massa DraaModes of Transmission (MoT)Models are:

1)Three integrated bio-behavioral surveillance survey (IBBSS) studies conducted among people who inject drugs (PWID) in Tanger and Nador in 2011-12,8female sex workers (FSWs) in Agadir, Fes, Rabat and Tanger in 2011,9 and men who have sex with men (MSM) in Agadir and Marrakech in 2010-11,10 using respondent driven sampling methodology. Sentinel surveillance data, voluntary counseling and testing (VCT) data, notified HIV/AIDS and sexually transmitted infection (STI) cases, and other relevant national HIV data and studies provided through the Morocco Ministry of Health. Data and studies provided by the Morocco Ministry of Health are cited as references in the parameters Tables S2.1-S2.5 and S3.1-S3.4 below.

2)Literature review of available data on HIV in Morocco. Specific studies from this review are cited as references in the parameters Tables S2.1-S2.5 and Tables S3.1-S3.4 below.

3)Synthesis review of HIV epidemiology in other Middle East and North Africa (MENA) countries as extracted from the MENA HIV/AIDS Epidemiology Synthesis Project.11, 12 This source of data is referred to in the parameters tables as ‘SMENA: Studies in other MENA countries’

4)Global HIV data, referred to as ‘GA: Global analyses’

5)Consensus agreement following discussions among national experts and stakeholders and the research team during the Morocco National and the Souss Massa DraaMoT workshops that were held in Morocco in the preparatory phases of the MoTproject, referred to as ‘CA’.

2.1Biological parameters and demographic attributes

Table S2.1Biological HIV transmission parameters used in the Morocco National and Souss Massa Draa Modes of Transmission Models.a

Parameter / Value / Justification / Reference
Transmission probability per HIV exposure from needle sharing / 0.008 / Systematic review and meta-analysis of HIV transmission probability per injecting act in observational studies / 13, 14
Transmission probability per non-commercial vaginal sex act / 0.0034 / Systematic review and meta-analysis of HIV transmission probability per coital act in observational studies / 14, 15
Transmission probability per commercial vaginal sex act / 0.0051 / Systematic review and meta-analysis of HIV transmission probability per coital act in observational studies / 14, 15
Transmission probability per anal sex act / 0.01 / Systematic review and meta-analysis of HIV transmission probability per coital act in observational studies / 14-16
Reduction in female to male transmission probability per vaginal sex act due to male circumcision / 60% / Finding of three clinical trials / 17-19
Prevalence of male circumcision / 99% / More than 99% of the population of Morocco practice male circumcision as a religious obligation and based on CA
Transmission probability per HIV exposure from medical injections / 0.004 / Systematic review and meta-analysis of HIV transmission probability per injection in observational studies / 13, 14
Transmission probability per HIV exposure from blood transfusions / 0.925 / Systematic review and meta-analysis of HIV transmission probability per transfusion in observational studies / 13, 14
ART coverage (Fraction of people receiving ART) / 20% / 20% coverage applies to all groups given that there are 5,301 people on ART out of about 30,000 HIV infected persons in Morocco at the end of 2012. This is also in consideration of the roughly similar testing rates among the different groups. / 20
Efficacy of ART in reducing HIV heterosexual transmission / 0.96 / Based on clinical trials of treatment for prevention and other observational data / 2, 21, 22
Efficacy of ART in reducing transmission among men who have sex with men / 0.90 / Based on clinical trials of treatment for prevention and other observational data / 2, 21, 22
Efficacy of ART in reducing transmission among needle injectors / 0.80 / Based on clinical trials of treatment for prevention and other observational data / 7, 22, 23

ART, antiretroviral therapy; CA, consensus agreement.

aSame biological parameters and assumptions were used for the national MoT andthe Souss Massa DraaMoTModel.

The transmission probabilities were adjusted for male circumcision (MC) and for antiretroviral therapy (ART) in each specific sub-population in Morocco. The adjustment for MC was done according tothe formula:

This applies only for HIV transmission between an infected female and a susceptible circumcised male. The value for the efficacy of MCin reducing HIV transmission,as well as the coverage of MC among males in Morocco, are found in Table S2.1.The unadjusted probability for MCis the probability that HIV is transmitted during a sex act in absence of MC.

The adjustment for ART was done according to the formula:

The values for the efficacy of ART in reducing the heterosexual, homosexual, and injecting transmission of HIV, as well as the coverage of ART () among each sub-population in Morocco, are found in Table S2.1.is the probability that HIV is transmitted during an act of exposure (sexual, injecting, and medical) in absence of ART.

Table S2.2Demographic attributes for Morocco.

Parameter / Value / Reference
Total population size in Morocco (TP) / 34 309 932 / 20, 23
Total 15-49 yrs. adult population size (TAP) / 19 149 878 / 20, 23
Male (TMP) / 9 574 939 / 20, 23
Female (TFP) / 9 574 939 / 20, 23
Adult (15-49 yrs.) HIV prevalence / 0.14% / 23

1

2.2Risk group size estimates

Table S2.3Risk group size estimatefor each HIV risk group in the Morocco National Modes of Transmission Model, 2013.

Population / Prevalence of risk / Risk group size estimate
Total (TAP) / Male (TMP) / Female (TFP) / Total (TAP) / Male (TMP) / Female (TFP)
PWID / 0.026% / 0.047% / 0.005% / 5000 / 4480 / 520
Calculation / # PWID/ TAP / (total # PWID x (1-fraction of PWID who are female))/ TMP / (# PWID x fraction of PWID who are female)/ TFP / Total # PWID x (1- fraction of PWID who are female) / Total # PWID x fraction of PWID who are female
Justification &/or reference / Based on size estimate work done for the IBBSS study on PWID,8based on triangulation and extrapolation from different sources of national data / Fraction of PWID who are female: 10.4%24 / Fraction of PWID who are female: 10.4%24 / Based on size estimate work done for the IBBSS study on PWID,8 based on triangulation and extrapolation from different sources of national data / Fraction of PWID who are female: 10.4%24 / Fraction of PWID who are female: 10.4%24
Stable sexual partners of PWID / 0.015% / 0.003% / 0.027% / 2890 / 301 / 2589
Calculation / (Fraction of the TMP who are stable sexual partners of PWID + Fraction of the TFP who are stable sexual partners of PWID)/2 / Prevalence of PWID among females x fraction of PWID who are in a stable sexual partnership / Prevalence of PWID among males x fraction of PWID who are in a stable sexual partnership / # of males in a stable sexual partnership with PWID + # of female in a stable sexual partnership with PWID / # of female PWID x fraction of PWID who are in a stable sexual partnership / # of male PWID x fraction of PWID who are in a stable sexual partnership
Justification &/or reference / 57.8% are in a stable sexual partnership: RV based on 20.1% (Tanger) and 21.1% (Nador) are married,8 also 60.6% (Tanger) and 54.9% (Nador) had sexual intercourse in the last 12 months,8 on SMENA,11on CA, and on GA / 57.8% are in a stable sexual partnership: RV based on 20.1% (Tanger) and 21.1% (Nador) are married,8 also 60.6% (Tanger) and 54.9% (Nador) had sexual intercourse in the last 12 months,8 on SMENA,11 on CA, and on GA / 57.8% are in a stable sexual partnership: RV based on 20.1% (Tanger) and 21.1% (Nador) are married,8 also 60.6% (Tanger) and 54.9% (Nador) had sexual intercourse in the last 12 months,8 on SMENA,11 on CA, and on GA / 57.8% are in a stable sexual partnership: RV based on 20.1% (Tanger) and 21.1% (Nador) are married,8 also 60.6% (Tanger) and 54.9% (Nador) had sexual intercourse in the last 12 months,8 on SMENA,11 on CA, and on GA
FSWs / 0.444% / -- / 0.888% / 85 000 / -- / 85 000

Table 2.3 cont.Risk group size estimate for each HIV risk group in the Morocco National Modes of Transmission Model, 2013.

Population / Prevalence of risk / Risk group size estimate
Total (TAP) / Male (TMP) / Female (TFP) / Total (TAP) / Male (TMP) / Female (TFP)
Calculation / Half the prevalence of sex work among females / Prevalence of sex work among females x TFP / Prevalence of sex work among females x TFP
Justification &/or
reference / Based on size estimate work done for IBBSS study on FSWs,9 based on triangulation and extrapolation from different sources of national data
Clients of FSWs / 4.439% / 8.877% / -- / 850 000 / 850 000 / --
Calculation / Half of the fraction of males seeking FSWs / Reasonable value chosen on the assumption of 10 clients per FSW / Fraction of males seeking FSWs x TMP / Fraction of males seeking FSWs x TMP
Justification &/or
reference / Based on CA
Stable sexual partners of FSWs clients / 1.775% / -- / 3.551% / 340 000 / -- / 340 000
Calculation / Half the fraction among females / Fraction of TMP who are married x fraction of the TMP seeking FSWs / Fraction of the TFP who are partners of FSWs clients x TFP / Fraction of the TFP who are partners of FSWs clients x TFP
Justification &/or reference / 40% of FSW clients are married in Morocco,9 based on average value of 36.4% (Agadir), 36.9% (Fes), 27.4% (Rabat), and 27.1% (Tanger)
MSM / 0.289% / 0.578% / -- / 55 319 / 55 319 / --
Calculation / Half the prevalence of MSM among males / Prevalence of MSM among males x TMP / Prevalence of MSM among males x TMP

Table S2.3 cont.Risk group size estimate for each HIV risk group in the Morocco National Modes of Transmission Model, 2013.

Population / Prevalence of risk / Risk group size estimate
Total (TAP) / Male (TMP) / Female (TFP) / Total (TAP) / Male (TMP) / Female (TFP)
Justification &/or reference / Based on size estimate work done for the IBBSS study on MSM,10 based on triangulation and extrapolation from different sources of national data
Stable female partners of MSM / 0.122% / -- / 0.243% / 23 278 / 23 278
Calculation / Half the fraction of women who are partners of MSM / Fraction of TMP who are MSM x Fraction of MSM who have female partners / Fraction of females who are partners of MSM x TFP / Fraction of females who are partners of MSM x TFP
Justification &/or reference / 42.08% of MSM have female partners.10Based on average of 3.5% (Agadir) and 0.9% (Marrakech) of MSM are married, and 80.9% (Agadir) and 83.0% (Marrakech) of MSM reporting sex with a woman, and on CA
People engaged in casual HS / 5.5% / 10.0% / 1.0% / 1 053 243 / 957 494 / 95 749
Calculation / (Prevalence of casual HS among males + prevalence of casual HS among females) /2 / # of males engaged in casual HS + # of females engaged in casual HS / Prevalence of casual HS among males x TMP / Prevalence of casual HS among females x TFP

Table S2.3 cont.Risk group size estimate for each HIV risk group in the Morocco National Modes of Transmission Model, 2013.

Population / Prevalence of risk / Risk group size estimate
Total (TAP) / Male (TMP) / Female (TFP) / Total (TAP) / Male (TMP) / Female (TFP)
Justification &/or reference / RsV based on CA after considering the totality of evidence including measures of sexual behavior among the general population, and a new study among youth which showed that 10.0% of females and 30.3% of males are sexually active.25 / RsV based on CA after considering the totality of evidence including a study among family planning and ANC attendees indicating that 0.7% of women reportedextra-marital sex, a study among youth where the prevalence of casual HS among males was reported at 15%,and 36.0% of males and 5.2% of females reportedever engaging invaginal sex,26and a new study among youth showingthat 10.0% of females and 30.3% of males are sexually active.25
Stable partners of people engaged in casual HS / 2.525% / 0.050% / 5.000% / 483 534 / 4787 / 478 747
Calculation / (Fraction of males who are partners of casual HS people + Fraction of females who are partners of casual HS people) / 2 / Fraction of females engaged in casual HS x fraction of them having steady/spousal partners / Fraction of males engaged in casual HS x fraction of them having steady/spousal partners / Fraction of males who are stable partners of people engaged in casual HS x TMP / Fraction of females who are partners of people engaged in casual HS x TFP
Justification &/or reference / Assuming a RsV of 5% of females engaged in casual HS having steady/spousal partners, and based on CA (same choices as 2010 MoT) / Assuming a RsV of 50% of males engaged in casual HS having steady/spousal partners, motivated also by the fact that 52.3% of females in 15-49 are married by DHS,27 and based on CA (same choices as 2010 MoT)
Table S2.3 cont.Risk group size estimate for each HIV risk group in the Morocco National Modes of Transmission Model, 2013.
Population / Prevalence of risk / Risk group size estimate
Total (TAP) / Male (TMP) / Female (TFP) / Total (TAP) / Male (TMP) / Female (TFP)
People with no risk HS this year / 27.50% / 20.00% / 35.00% / 5 266 216 / 1 914 988 / 3 351 228
Calculation / (Prevalence of no risk HS in males + prevalence of no risk HS among females) /2 / # of males with no risk HS + # females with no risk HS / Fraction of the TMP with no risk HS x TMP / Fraction of the TFP with no risk HS x TFP
Justification &/or reference / CA and RsV based on levels of reported sexual activity among youth and marriage rates in Morocco26 (same choices as 2010 MoT) / CA and RsV based on levels of reported sexual activity among youth and marriage rates in Morocco.26 (same choices as 2010 MoT)
Stable heterosexual couples / 57.365% / 60.445% / 54.286% / 10 985 397 / 5 787 570 / 5 197 827
Calculation / # of males engaged in stable heterosexual couple + # of females engaged in stable heterosexual couple / Prevalence of stable heterosexual couple among males x TMP / Prevalence of stable heterosexual couple among females x TFP
Justification &/or reference / Those are defined as the population remainder; that is those who are not in any of the previous risk group categories nor their sexual partners/clients / Those are defined as the population remainder; that is those who are not in any of the previous risk group categories nor their sexual partners/clients / Those are defined as the population remainder; that is those who are not in any of the previous risk group categories nor their sexual partners/clients
Population at risk of medical injections / 100% / 100% / 100% / 19 149 878 / 9 574 939 / 9 574 939
Justification &/or reference / Assuming all the adult population is at risk / Assuming all the adult population is at risk / Assuming all the adult population is at risk / Assuming all the adult population is at risk / Assuming all the adult population is at risk / Assuming all the adult population is at risk
Population at risk of blood transfusions / 0.364% / 0.364% / 0.364% / 69 750 / 34 875 / 34 875
Table S2.3 cont.Risk group size estimate for each HIV risk group in the Morocco National Modes of Transmission Model, 2013.
Population / Prevalence of risk / Risk group size estimate
Total (TAP) / Male (TMP) / Female (TFP) / Total (TAP) / Male (TMP) / Female (TFP)
Calculation / (# of blood donations per year / # of transfusions per person)/ TP / (# of blood donations per year / # of transfusions per person)/ TP / (# of blood donations per year / # of transfusions per person)/ TP / # of blood donations per year /# of transfusions per person / (# of blood donations per year /# of transfusions per person)/2 / (# of blood donations per year /# of transfusions per person)/2
Justification &/or reference / Based on 125 000 people receiving blood transfusion per year, of which 55.8% are among the 15-49 years age group. This estimate is based on an estimate of 250 000 blood containers being used every year for blood transfusions with an average of about 2 per person. Of notice that there are 384 745 blood containers through blood donation sites. Number of donations per blood donor is about 3 / Based on 125 000 people receiving blood transfusion per year, of which 55.8% are among the 15-49 years age group. This estimate is based on an estimate of 250 000 blood containers being used every year for blood transfusions with an average of about 2 per person. Of notice that there are 384 745 blood containers through blood donation sites. Number of donations per blood donor is about 3 / Based on 125 000 people receiving blood transfusion per year, of which 55.8% are among the 15-49 years age group. This estimate is based on an estimate of 250 000 blood containers being used every year for blood transfusions with an average of about 2 per person. Of notice that there are 384 745 blood containers through blood donation sites. Number of donations per blood donor is about 3

Cont., continued; ANC, antenatal care; CA, consensus agreement in the national modesof transmission workshop; FSWs, female sex workers; HS, heterosexual sex; GA, global analyses; PWID, people who inject drugs; MSM, men who have sex with men; RV, representative value; RsV, reasonable value; TAP, total adult population (15-49 yrs.); TFP, total adult female population (15-49 yrs.); TMP, total adult malepopulation (15-49 yrs.); SMENA, studies in other Middle East and North Africa countries; IBBSS, integrated bio-behavioral surveillance survey.