RESEARCH Open AccessMale gender predicts mortality in a large cohortof patients receiving antiretroviral therapy inUgandaEdward J Mills1,2*, Celestin Bakanda3, Josephine Birungi3, Keith Chan2, Robert S Hogg2, Nathan Ford4,Jean B Nachega5,6 and Curtis L Cooper7AbstractBackground: Because men in Africa are less likely to access HIV/AIDS care than women, we aimed to determine ifmen have differing outcomes from women across a nationally representative sample of adult patients receivingcombination antiretroviral therapy in Uganda.Methods: We estimated survival distributions for adult male and female patients using Kaplan-Meier, andconstructed multivariable regressions to model associations of baseline variables with mortality. We assessedperson-years of life lost up to age 55 by sex. To minimize the impact of patient attrition, we assumed a weighted30% mortality rate among those lost to follow up.Results: We included data from 22,315 adults receiving antiretroviral therapy. At baseline, men tended to be older,had lower CD4 baseline values, more advanced disease, had pulmonary tuberculosis and had received lesstreatment follow up (all at p < 0.001). Loss to follow up differed between men and women (7.5 versus 5.9%, p <0.001). Over the period of study, men had a significantly increased risk of death compared with female patients(adjusted hazard ratio 1.43, 95% CI 1.31-1.57, p < 0.001). The crude mortality rate for males differed importantlyfrom females (43.9, 95% CI 40.7-47.0/1000 person-years versus 26.9, 95% CI 25.4-28.5/1000 person years, p < 0.001).The probability of survival was 91.2% among males and 94.1% among females at 12 months. Person-years of lifelost was lower for females than males (689.7 versus 995.9 per 1000 person-years, respectively).Conclusions: In order to maximize the benefits of antiretroviral therapy, treatment programmes need to begender sensitive to the specific needs of both women and men. Particular efforts are needed to enroll men earlierinto care.BackgroundAlthough the global prevalence of HIV among womenhas remained stable at 50%, in sub-Saharan Africa, it ismarkedly higher in females than in males [1]. Similarlyin Africa, young women have a higher incidence of HIVinfection [1-3]. As a focused public health strategy,efforts to improve access to treatment, research andhuman rights for women and girls have recentlyreceived particularly special attention. Substantially lessattention has been focused on men [4].Scale up in access to care and treatment has beencomparatively more successful for women than for men[5,6]. Recently, there has been growing recognition thatmen face important challenges for both access to careand delivery of care [4,7-9]. Men may be more difficultto provide care for due to issues of occupation, resi-dence or cultural beliefs. Reasons for health-seekingbehaviour differences between men and women arepoorly understood. Cultural norms that label sickness asa sign of weakness for men have fostered a reluctance ofcare seeking among men [2]. Men’s actions, and thoseof women, may be constrained by tradition and influ-enced by cultural beliefs and social norms [10].* Correspondence: edward.mills@uottawa.ca1Faculty of Health Sciences, University of Ottawa, Ottawa, CanadaFull list of author information is available at the end of the articleMills et al. Journal of the International AIDS Society 2011, 14:52http://www.jiasociety.org/content/14/1/52© 2011 Mills et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative CommonsAttribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction inany medium, provided the original work is properly cited.In addition, employment-related migration keeps menaway from their partners and families for prolonged per-iods and contributes to increased vulnerability by predis-posing them to high-risk behaviours [10]. These riskfactors make men both vulnerable to HIV infection anddiscouraged from seeking testing and treatment. InAfrica, antenatal care services act as a common entrypoint to HIV testing, treatment and care services forwomen, as pregnant women are routinely tested forHIV. No such opportunity exists for men [4,8].In this paper, we report on outcomes and life expec-tancy of men and women receiving combination antire-troviral therapy (cART) in a nationally representativecohort in Uganda.MethodsParticipantsOur analysis includes routinely collected data on allpatients aged 14 and older who initiated antiretroviraltherapy at clinics run by The AIDS Support Organiza-tion (TASO) in Uganda. This cohort has been describedin detail elsewhere [11].Founded in 1987, TASO provides psychosocial sup-port, clinical care and cART to individuals infected withHIV. TASO began providing widespread cART in 2004and now provides cART to more than 24,000 patients.Since its inception, TASO has expanded the scope of itswork to include nutritional support, vocational training,HIV education and capacity building of healthcareworkers. TASO’s mission is to restore hope and improvethe quality of life of individuals, families and commu-nities affected by HIV and related diseases. TASO pro-vides support to more than 100,000 HIV-infectedpatients and supports family members through educa-tion, counselling and educational stipends. TASO pro-vides services through 11 regional TASO centres and 35mini-TASO centres in rural districts.TASO receives its core cART funding through theUnited States President’s Emergency Plan for AIDSRelief (PEPFAR), as well as from Irish AID and SwedishSIDA. The Ugandan Ministry of Health refers newlydiagnosed HIV-infected children and adults to TASOclinical sites for care. Many of the populations servicedby TASO represent marginalized and neglected groups,including infants and children, orphans, conflict-affectedpopulations, internally displaced people, widows, prison-ers and family members of HIV-infected patients thatmay require urgent support. TASO programmes empha-size adherence and retention, and include innovativeapproaches to maintaining patient interest, includingdrama and social groups, diary writing and involvementof patients in clinical duties to become “expert patients”.TASO provides a range of services, including HIV test-ing, clinical care, provision of cART and psychosocialsupport. Laboratory services are limited, but include CD4analysis, complete blood analysis, tuberculosis and malariadiagnosis, and urine assessments. Criteria used for initia-tion of cART at TASO include World Health Organiza-tion (WHO) Stage III or IV or a CD4 cell count below 250cells/mm3. Criteria for children’s clinical admittance intothe TASO cART programme are based on Ugandan Min-istry of Health Guidelines. Children are eligible for cARTif they have WHO paediatric Stage III, advanced Stage II,or Stage I with CD4 cell percentage (%) < 15% for thoseover 18 months of age, and < 20% for those under 18months of age [12]. The Uganda Ministry of HealthNational Antiretroviral Treatment and Care Guidelines forAdults and Children have not yet been updated to reflectWHO’s newest recommendations for clinical staging andimmunological classification [13].When a patient attends a TASO clinic, clinicians willcomplete standardized patient forms detailing patientdemographics, as well as clinical, psychosocial and drugutilization data. These data are then hand entered, induplicate, into the TASO clinical administrative datacollection. Patients are provided with a unique confiden-tial identifying number. Patients requiring cART typi-cally receive an initiation regimen based on a non-nucleoside reverse transcriptase inhibitor with first-linetreatment comprising nevirapine, lamivudine and stavu-dine and boosted lopinavir, didanosine and zidovudineas second-line treatment [14]. Patients requiring treat-ment for TB co-infection will receive their combinationcare at a TASO clinic.Patients aged 14 years and older who initiated antire-troviral therapy at TASO clinics in Uganda between 1January 2004 and 1 January 2010 were included in thisstudy. These patients were followed until either the timeof death or the end of the study period (1 January2010). For this analysis, we extracted the followinginformation: age at the start of the antiretroviral therapy;gender; baseline CD4 count; WHO clinical disease stage;presence of tuberculosis or sexually transmitted infec-tions at treatment initiation; date when they were lastseen; and, where applicable, date of all-cause mortality,non-disease mortality, or defaulting from care (definedas a three-month untraceable absence from a clinic).AnalysisPatient characteristics by gender are described usingmedians and interquartile ranges for continuous vari-ables and counts and percentages for categorical dataand compared using a chi-squared test. Survival distri-butions for male and female patients are estimatedusing the Kaplan-Meier method and compared by log-rank test. Survival was calculated from the start date ofantiretroviral therapy to the date of death. Patients whowere lost to follow up were censored at the date whenMills et al. Journal of the International AIDS Society 2011, 14:52http://www.jiasociety.org/content/14/1/52Page 2 of 7they were last seen. Patients who were still alive at thedate when the study ended were censored at this date.Survival times were expressed in months. We applied aweighted analysis whereby 30% of patients lost to followup were assumed to be dead, weighted by baseline CD4and age, as suggested by Egger et al [15,16]. To accountfor missing baseline CD4 cell counts, we conducted ouranalyses using the multiple imputation method of SAS[17].Potential years of life lost (PYLL) before age 55 wereused to examine the effect of HIV on premature mortal-ity. PYLL represent the sum of years that individuals lostbecause of premature mortality. PYLL are a convenientsummary measure that account for not only the numberof deaths, but also the ages at which death occurs. Toobtain PYLL, we grouped deaths according to age atdeath by five-year increment categories. The total num-ber deaths for a particular cause in each five-year agegroup are multiplied by the average number of yearsremaining in that age group to age 55 years, as follows:PYLL = di (55− Yi)where, Yi is the age at the midpoint of age group i.Survival times were expressed in months. Unadjustedand adjusted Cox proportional hazards regression wasconducted in order to quantify the effect of gender onsurvival, adjusting for age, CD4 status and WHO clinicaldisease stage. This analysis included point and confi-dence interval estimates for the hazard ratios of deathfor each factor. Hazard proportionality was assessed byanalysis of scaled Schoenfeld residuals. All significancetests were two-sided with a p value of < 0.05 consideredsignificant. All analyses were conducted using SAS ver-sion 8 (SAS Institute, Cary, NC). Additional file 1 pre-sents results for the overall analysis based on onlydocumented deaths.Institutional approvalApproval to conduct this study was received from theadministrative headquarters and Research Ethics Boardof TASO Uganda, an approved Ugandan NationalScience and Technology Ethics Review Board, and theResearch Ethics Boards of the University of BritishColumbia and the University of Ottawa. Because thisanalysis was based on routine clinical data, retrospectiveindividual patient consent was not required.ResultsWe included data from 22,315 adults receiving antire-troviral therapy, representing 59,436 person-years. Themajority (19,885; 89%) were aged between 14 and 49years and 2430 (11%) were aged 50 years or older. Fig-ure 1 displays the distribution of patients by age andsex. Baseline CD4 cell count, WHO stage at initiation,presence of tuberculosis and duration of follow up alldiffered significantly between male and female patients(Table 1). Over the course of the study, 918 femaleswere lost to follow up (5.9%) and 515 males (7.5%) (p <0.001). Patients lost to follow up had a lower medianCD4 than those not lost (105 interquartile range 73-207vs. 144, interquartile range 34-181, p < 0.001).There were 1498 deaths that were accounted for in ouranalysis. We imputed a further 445 deaths from lost-to-follow-up patients. Additional file 1 provides the resultswithout adjusting for loss to follow up. Overall, men had asignificantly increased risk of death compared with femalepatients (hazard ratio 1.53, 95% confidence interval [CI]1.40-1.68, p < 0.001). Most deaths (n = 740) occurredwithin the first three months of initiating antiretroviraltherapy (302 in men and 438 in women). There was nodifference in non-disease mortality by gender (odds ratio,0.95, 95% CI 0.59-1.54, p = 0.86). The probability of survi-val among males compared with females was 95.5% (95%CI 95.0-96.0) and 97.2 (95% CI 96.9-97.4%) at threemonths; 93.2% (95% CI 92.6-93.8%) and 95.8% (95% CI95.5-96.1%) at six months; 91.2% (95% CI 90.6-91.9%) and94.1% (95% CI 93.7-94.5%) at 12 months; and 89.1% (95%CI 88.4-89.9%) and 92.7% (95% CI 92.3-93.1%), respec-tively, at 24 months (Figure 2).Univariate and multivariable Cox proportional hazardsregressionTable 2 presents the unadjusted and adjusted hazardratios (HRs) for each variable. We found that genderFigure 1 Distribution of included patients by age and sex. Greyshading, female; black shading, male.Mills et al. Journal of the International AIDS Society 2011, 14:52http://www.jiasociety.org/content/14/1/52Page 3 of 7remained independently associated with mortality afteradjusting for other factors (adjusted HR 1.43, 95% CI1.31-1.57). We additionally found that lower CD4 statusat baseline was associated with mortality.The crude mortality rate was 31.8 (95% CI 30.3-33.2)per 1000 person-years for the overall cohort. Femaleshad a lower crude mortality rate compared with males:26.9 (95% CI 25.4-28.5) per 1000 person-years versus43.9 (95% CI 40.7-47.0) per 1000 person-years, respec-tively. Potential years of life lost (PYLL) was 795.0 per1000 person-years for the overall cohort. Similarly,PYLL was lower for females than males (689.7 versus995.9 per 1000 person-years, respectively).DiscussionOur study demonstrates that male HIV patients inUganda have consistently worse survival outcomes com-pared with their female counterparts. Our study findingsbuild upon evidence suggesting that male outcomes areconsistently worse in Africa [5,18-20]. Given the lowcoverage of antiretroviral therapy among men in Ugandaand other parts of Africa, an emphasis on involving menin HIV testing and the route to treatment is critical ifwe are serious about addressing the vulnerability ofwomen for HIV acquisition.Public health planning can frequently be counterintuitive [21]. In the context of global developmentalprogramming, the concepts of “gender and develop-ment” and “women and development” have been fre-quently constructed as one and the same [22], andincreasing calls for consideration of gender relations inthe AIDS response have been mostly met with a focuson girls and women. However, the specific vulnerabil-ities of boys and men in this new configuration of gen-der relations are rarely addressed. Indeed, very little isknown about how to engage men with directed effortsto change their risk-taking and health-seekingbehaviours.It is likely that much of the successes of engagingwomen in clinical care and their more positive out-comes stem from directed efforts to target women, aswell as logistic efforts to access women at a health-Table 1 Characteristics of included patients at baselineCharacteristics Category Total Femalen (%)Malen (%)p valueAge 14-19 333 233 (1.5) 100 (1.5) < 0.00120-29 3486 2866 (18.5) 620 (9.1)30-39 9774 6966 (45) 2808 (41.2)40-49 6292 4035 (26) 2257 (33.1)50+ 2430 1392 (9) 1038 (15.2)Total (n) 22,315 15,492 6823CD4 < 50 3452 2173 (16.9) 1279 (22.8) < 0.00150-99 2942 1944 (15.1) 998 (17.8)100-149 3410 2391 (18.5) 1019 (18.2)150-249 5740 4152 (32.2) 1588 (28.3)250+ 2954 2233 (17.3) 721 (12.9)Total (n) 18,498 12,893 5605WHO Stage at antiretroviral therapy initiation Stage 1 465 339 (3.4) 126 (2.7) < 0.001Stage 2 7985 5615 (56) 2370 (51.2)Stage 3 4982 3294 (32.9) 1688 (36.5)Stage 4 1220 778 (7.8) 442 (9.6)Total (n) 14,652 10,026 4626TB at antiretroviral therapy initiation No 21,207 14,847 (95.8) 6360 (93.2) < 0.001Yes 1108 645 (4.2) 463 (6.8)Total (n) 22,315 15492 6823Sexually transmitted infection diagnosed at antiretroviral therapy initiation No 17,634 11,579 (74.7) 6055 (88.7) < 0.001Yes 4681 3913 (25.3) 768 (11.3)Total (n) 22,315 15,492 6823Switch from first antiretroviral therapy No 20,675 14,313 (92.4) 6362 (93.2) 0.024Yes 1640 1179 (7.6) 461 (6.8)Total (n) 22,315 15,492 6823Median months of follow-up time (interquartile range) - - 32 (20-47) 30 (18-39) < 0.001Mills et al. Journal of the International AIDS Society 2011, 14:52http://www.jiasociety.org/content/14/1/52Page 4 of 7seeking moment, such as in antenatal clinics. Efforts toengage men at antenatal services have had poor out-comes. It seems more likely that efforts aimed at theirplaces of work or aimed at peer educators may have bet-ter success [8].Circumcision clinics are gaining prominence in muchof Africa as an intervention to reduce male infections[23]. This may provide an important opportunity to testmen for their serostatus and direct them into care.Retention in programmes will be a major challenge asmen frequently travel for work, and therefore targetedand novel interventions to maintain or increase reten-tion in care is also badly needed [8].Several studies have found that males fare worse thanfemales in terms of cART access and outcomes[5,18-20,24]. While this variable has typically beenincluded as a covariate in regression analysis, it is infre-quently examined in detail. A paper from South Africalooked at a small cohort of 2196 patients receivingcART, 33% of whom were male [19]. As with our study,men presented at a later age and with more advanceddisease. Two studies from Malawi examined mortalityoutcomes between males and females with a very highcrude mortality rate (123.2/1000 person years) andfound a heightened mortality among males regardless ofclinical features (HR 1.90, 95% CI 1.57-2.29) as well asincreased loss to follow up (HR 1.66, 95% CI 1.43-1.92)[20,25]. A study involving more than 12,000 patients inTanzania found an increase in mortality among men(HR 1.19, 95% CI 1.05-1.30), immunologic non-responsedefined as a CD4 cell count of less than 100 cells/mm3after at least six months of cART (HR 1.74, 95% CI1.44-2.11) and loss to follow up (HR 1.19, 95% CI 1.10-1.30) than that in women [6]. The largest evaluationuntil now, examining 11,153 patients across four coun-tries with close clinical monitoring, found an HR of 1.17(95% CI 1.02-1.35) for male mortality after adjustmentsfor other expected covariates [18]. A previous analysisusing this cohort examined life expectancy and foundTable 2 Proportional hazards regression for time to deathVariable Unadjusted hazard ratio(95% CI)p value Adjusted hazard ratio(95% CI)p valueGender (male versus female) 1.53 (1.40-1.68) <0.001 1.43 (1.31-1.57) <0.001Age14-19 1.00 1.0020-29 0.98 (0.67-1.42) 0.895 1.05 (0.72-1.53) 0.79430-39 0.90 (0.62-1.29) 0.566 0.96 (0.67-1.39) 0.83540-49 0.87 (0.60-1.26) 0.459 0.93 (0.65-1.35) 0.71350+ 1.20 (0.82-1.75) 0.338 1.31 (0.90-1.91) 0.164CD4 at antiretroviral therapy initiation (per 100 cells/mm3) 0.65 (0.61-0.68) <0.001CD4 count<50 1.00 1.0050-99 0.83 (0.73-0.94) 0.005 0.82 (0.72-0.94) <0.001100-149 0.63 (0.55-0.72) <0.001 0.63 (0.55-0.73) <0.001150-249 0.44 (0.39-0.50) <0.001 0.45 (0.39-0.51) <0.001250+ 0.37 (0.31-0.45) <0.001 0.38 (0.32-0.46) <0.001World Health Organization Stage at ART initiationStage 1 1.00Stage 2 0.99 (0.67-1.47) 0.957Stage 3 2.04 (1.38-3.03) <0.001Stage 4 3.78 (2.52-5.67) <0.001Log rank p<0.001            !%$)" !$!$)  ! #&' & $' #&!(  &$% !!''&("#  %)(%$"%#  ""!# !"!&  !&' '&&    Log rank p<0.001 Figure 2 Kaplan-Meier survival curves on time to death.Mills et al. Journal of the International AIDS Society 2011, 14:52http://www.jiasociety.org/content/14/1/52Page 5 of 7dramatic differences between sexes: males at the age of20 could expect to live an additional 19.1 years whilefemales could expect 30.6 years [26].Strengths of our study include the large and nationallyrepresentative sample involving patients across manyage groups. As TASO engages active retention ofdefaulting patients, we have minimized loss to follow upcompared with other AIDS service organizations thatfrequently experience about 40% losses [27]. Attritiontypically occurs at two distinct times, pre- and post-anti-retroviral therapy initiation [28]. A TASO study foundthat approximately 26% of 637 patients eligible for anti-retroviral therapy did not initiate treatment, and mostoften, these were males [16]. We applied a weightedanalysis to adjust for an attrition effect on mortality[15].Our study represents just one country. A previous sys-tematic review, of early experiences providing antiretro-viral therapy, found a ratio of 2.3:1 of women receivingantiretroviral therapy compared with men across 13countries in sub-Saharan Africa [29]. A further limita-tion is that we do not have data on viral failure and can-not be sure if treatment failures are associated withgender, although a large study of better resourced clinicsfound results similar to ours [18]. Finally, as an observa-tional study, we recognize that our estimates are subjectto potential confounding by unmeasured variables.Other studies have found that level of education andpregnancy status at initiation influence mortality, andmay partially explain our mortality estimates [5,30,31].ConclusionsIn conclusion, our study has important implications forfuture generations of individuals infected and affectedby HIV/AIDS. Funding agencies should recognize thatmales and females are necessary components of mosthouseholds, and plan for their interventionsappropriately.Additional materialAdditional file 1: Analysis according to unadjusted data.AcknowledgementsThe Canadian Institutes of Health Research sponsored this study. The studysponsors had no role in the design, conduct, collection of data, analysis orinterpretation of this study. Edward Mills had full access to the data and isresponsible for the decision to submit the manuscript for publication.Author details1Faculty of Health Sciences, University of Ottawa, Ottawa, Canada. 2BritishColumbia Centre for Excellence in HIV/AIDS, Vancouver, Canada. 3The AIDSSupport Organization (TASO), Headquarters, Kampala, Uganda. 4Centre forInfectious Disease Epidemiology and Research, University of Cape Town,South Africa. 5Centre for Infectious Diseases, Stellenbosch University,Stellenbosch, South Africa. 6Johns Hopkins Bloomberg School of PublicHealth, Baltimore, MD, USA. 7Division of Infectious Diseases, The OttawaHospital, Ottawa, Canada.Authors’ contributionsEM, CB, JB and KC conceived the study design, analyzed and interpreted thedata and wrote the manuscript. EM, CB and JB contributed to the design ofthe study and revised the manuscript. KC, RH, NF, JN and CC participated inthe analysis and interpretation of the data and revised the manuscript. All ofthe authors approved the final version of the manuscript submitted forpublication.Competing interestsThe authors declare that they have no competing interests.Received: 22 June 2011 Accepted: 3 November 2011Published: 3 November 2011References1. World Health Organization (WHO): HIV/AIDS epidemiological surveillancereport for the WHO African Region, 2007 update World Health Organization,Regional Office for Africa; 2008.2. United Nations Integrated Regional Information Networks: Uganda: Men’sunion encourages men to be more open about HIV United NationsIntegrated Regional Information Networks; 2006.3. Ulin PR: African women and AIDS: negotiating behavioral change. Soc SciMed 1992, 34:63-73.4. Mills EJ, Ford N, Mugyenyi P: Expanding HIV care in Africa: making menmatter. Lancet 2009, 374:275-276.5. Braitstein P, Boulle A, Nash D, Brinkhof MW, Dabis F, Laurent C,Schechter M, Tuboi SH, Sprinz E, Miotti P, Hosseinipour M, May M, Egger M,Bangsberg DR, Low N, Antiretroviral Therapy in Lower Income Countries(ART-LINC) study group: Gender and the use of antiretroviral treatment inresource-constrained settings: findings from a multicenter collaboration.J Womens Health 2008, 17:47-55.6. Hawkins C, Chalamilla G, Okuma J, Spiegelman D, Hertzmark E, Aris E,Ewald T, Mugusi F, Mtasiwa D, Fawzi W: Sex differences in antiretroviraltreatment outcomes among HIV-infected adults in an urban Tanzaniansetting. AIDS 2011, 25:1189-1197.7. Birungi J, Mills EJ: Can we increase male involvement in AIDS treatment?Lancet 2010, 376:1302.8. Kuwane B, Appiah K, Felix M, Grant A, Churchyard G: Expanding HIV carein Africa: making men matter in Johannesburg. Lancet 2009, 374:1329.9. Johannessen A: Are men the losers of the antiretroviral treatment scale-up? AIDS 2011, 25:1225-1226.10. Mane P, Aggleton P: Gender and HIV/AIDS: What Do Men have to Dowith it? Current Sociology 2001, 49:23-37.11. Bakanda C, Birungi J, Nkoyooyo A, Featherstone A, Cooper CL, Hogg RS,Mills EJ: Cohort Profile: The TASO-CAN Cohort Collaboration. Int JEpidemiol 2011.12. Uganda Ministry of Health: National Antiretroviral Treatment and CareGuidelines for Adults, Adolescents and Children 2009.13. World Health Organization (WHO): WHO Case Definitions of HIV forSurveillance and Revised Clinical Staging and Immunological Classification ofHIV-related Disease in Adults and Children 2008.14. Kiboneka A, Nyatia RJ, Nabiryo C, Anema A, Cooper CL, Fernandes KA,Montaner JS, Mills EJ: Combination antiretroviral therapy in populationaffected by conflict: outcomes from large cohort in northern Uganda.BMJ 2009, 338:a2662.15. Egger M, Spycher BD, Sidle J, Weigel R, Geng EH, Fox MP, MacPhail P, vanCutsem G, Messou E, Wood R, Nash D, Pascoe M, Dickinson D, Etard JF,McIntyre JA, Brinkhof MW, IeDEA East Africa, West Africa and SouthernAfrica: Correcting mortality for loss to follow-up: a nomogram applied toantiretroviral treatment programmes in sub-Saharan Africa. PLoS Med2011, 8:e1000390.16. Amuron B, Namara G, Birungi J, Nabiryo C, Levin J, Grosskurth H,Coutinho A, Jaffar S: Mortality and loss-to-follow-up during the pre-treatment period in an antiretroviral therapy programme under normalhealth service conditions in Uganda. BMC Public Health 2009, 9:290.Mills et al. Journal of the International AIDS Society 2011, 14:52http://www.jiasociety.org/content/14/1/52Page 6 of 717. Yuan YC: Multiple Imputation for Missing Data: concepts and NewDevelopment. SAS Code [http://support.sas.com/rnd/app/papers/multipleimputation.pdf].18. May M, Boulle A, Phiri S, Messou E, Myer L, Wood R, Keiser O, Sterne JA,Dabis F, Egger M, IeDEA Southern Africa and West Africa: Prognosis ofpatients with HIV-1 infection starting antiretroviral therapy in sub-Saharan Africa: a collaborative analysis of scale-up programmes. Lancet2010, 376:449-457.19. Cornell M, Myer L, Kaplan R, Bekker LG, Wood R: The impact of genderand income on survival and retention in a South African antiretroviraltherapy programme. Trop Med Int Health 2009, 14:722-731.20. Chen SC, Yu JK, Harries AD, Bong CN, Kolola-Dzimadzi R, Tok TS, King CC,Wang JD: Increased mortality of male adults with AIDS related to poorcompliance to antiretroviral therapy in Malawi. Trop Med Int Health 2008,13:513-519.21. Shelton JD: Ten myths and one truth about generalised HIV epidemics.Lancet 2007, 370:1809-1811.22. Chant SH, Gutmann MC: Mainstreaming men into gender and development:Debates, reflections, and experiences Oxfam; 2000.23. Kiboneka A, Wangisi J, Nabiryo C, Tembe J, Kusemererwa S, Olupot-Olupot P, Joffres M, Anema A, Cooper CL, Montaner JS, Mills EJ: Clinicaland immunological outcomes of a national paediatric cohort receivingcombination antiretroviral therapy in Uganda. AIDS 2008, 22:2493-2499.24. Braitstein P, Boulle A, Nash D, Brinkhof MW, Dabis F, Laurent C,Schechter M, Tuboi SH, Sprinz E, Miotti P, Hosseinipour M, May M, Egger M,Bangsberg DR, Low N, Antiretroviral Therapy in Lower Income Countries(ART-LINC) study group: Gender and the use of antiretroviral treatment inresource-constrained settings: findings from a multicenter collaboration.J Womens Health (Larchmt) 2008, 17:47-55.25. Taylor-Smith K, Tweya H, Harries A, Schoutene E, Jahn A: Genderdifferences in retention and survival on antiretroviral therapy of HIV-1infected adults in Malawi. Malawi Medical Journal 2010, 22:49-56.26. Mills EJ, Bakanda C, Birungi J, Chan K, Ford N, Cooper CL, Nachega JB,Dybul M, Hogg RS: Life Expectancy of Persons Receiving CombinationAntiretroviral Therapy in Low-Income Countries: A Cohort Analysis FromUganda. Ann Intern Med 2011, 155:209-216.27. Rosen S, Fox MP, Gill CJ: Patient retention in antiretroviral therapyprograms in sub-Saharan Africa: a systematic review. PLoS Med 2007, 4:e298.28. Rosen S, Fox MP: Retention in HIV Care between Testing and Treatmentin Sub-Saharan Africa: A Systematic Review. PLoS Med 2011, 8:e1001056.29. Muula AS, Ngulube TJ, Siziya S, Makupe CM, Umar E, Prozesky HW,Wiysonge CS, Mataya RH: Gender distribution of adult patients on highlyactive antiretroviral therapy (HAART) in Southern Africa: a systematicreview. BMC Public Health 2007, 7:63.30. Mutevedzi PC, Lessells RJ, Heller T, Barnighausen T, Cooke GS, Newell ML:Scale-up of a decentralized HIV treatment programme in rural KwaZulu-Natal, South Africa: does rapid expansion affect patient outcomes? BullWorld Health Organ 2010, 88:593-600.31. Melekhin VV, Shepherd BE, Stinnette SE, Rebeiro PF, Barkanic G, Raffanti SP,Sterling TR: Antiretroviral therapy initiation before, during, or afterpregnancy in HIV-1-infected women: maternal virologic, immunologic,and clinical response. PLoS One 2009, 4:e6961.doi:10.1186/1758-2652-14-52Cite this article as: Mills et al.: Male gender predicts mortality in a largecohort of patients receiving antiretroviral therapy in Uganda. Journal ofthe International AIDS Society 2011 14:52.Submit your next manuscript to BioMed Centraland take full advantage of: • Convenient online submission• Thorough peer review• No space constraints or color figure charges• Immediate publication on acceptance• Inclusion in PubMed, CAS, Scopus and Google Scholar• Research which is freely available for redistributionSubmit your manuscript at www.biomedcentral.com/submitMills et al. Journal of the International AIDS Society 2011, 14:52http://www.jiasociety.org/content/14/1/52Page 7 of 7