High burden of metabolic comorbidities in a citywide cohort of HIV outpatients: evolving health care needs of people aging with HIV in Washington, DC

Transcription

1 DOI: /hiv ORIGINAL RESEARCH High burden of metabolic comorbidities in a citywide cohort of HIV outpatients: evolving health care needs of people aging with HIV in Washington, DC ME Levy, 1 AE Greenberg, 1 R Hart, 2 L Powers Happ, 1 C Hadigan 3 and A Castel 1 for the DC Cohort Executive Committee* 1 Department of Epidemiology and Biostatistics, Milken Institute School of Public Health at the George Washington University, Washington, DC, USA, 2 Research Department, Cerner Corporation, Kansas City, MO, USA and 3 Laboratory of Immunoregulation, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, MD, USA Objectives With the increasing impact of cardiovascular disease among populations aging with HIV, contemporary prevalence estimates for predisposing metabolic comorbidities will be important for guiding the provision of relevant lifestyle and pharmacological interventions. We estimated the citywide prevalence of hypertension, type 2 diabetes, dyslipidaemia, and obesity; examined differences by demographic subgroups; and assessed clinical correlates. Methods Utilizing an electronic medical record (EMR) database from the DC Cohort study a multicentre prospective cohort study of HIV-infected outpatients we assessed the period prevalence of metabolic comorbidities between 2011 and 2015 using composite definitions that incorporated diagnoses, pharmacy records, and clinical/laboratory results. Results Of 7018 adult patients (median age 50 years; 77% black), 50% [95% confidence interval (CI) 49 51] had hypertension, 13% (95% CI: 12 14) had diabetes, 48% (95% CI: 47 49) had dyslipidaemia, and 35% (95% CI: 34 36) had obesity. Hypertension was more prevalent among black patients, diabetes and obesity were more prevalent among female and black patients, dyslipidaemia was more prevalent among male and white patients, and comorbidities were more prevalent among older patients (all P < 0.001). For many patients, evidence of treatment for these comorbidities was not available in the EMR. Longer time since HIV diagnosis, greater duration of antiretroviral treatment, and having controlled immunovirological parameters were associated with metabolic comorbidities. Conclusions These findings underscore the pervasive burden of metabolic comorbidities among HIV-infected persons, serve as the basis for future analyses characterizing their impact on subsequent adverse cardiovascular outcomes, and highlight the need for an increased focus on the prevention and control of comorbid complications in this population. Keywords: diabetes, dyslipidaemia, HIV, hypertension, metabolic comorbidities, obesity Accepted 1 March 2017 Introduction The widespread availability of antiretroviral (ARV) therapy has led to the routine management of HIV infection as a chronic condition in which the most prevalent Correspondence: Matthew Levy, 950 New Hampshire Ave NW, Suite 500, Washington, DC 20052, USA. Tel: ; fax: ; mattelevy@gwu.edu *See Appendix. comorbidities have shifted from opportunistic infections and AIDS-related malignancies to noncommunicable diseases such as cardiovascular disease, liver disease, kidney disease, osteoporosis, neurocognitive decline, and non- AIDS-related malignancies [1]. HIV-infected persons confront a 1.5- to 2-fold increased risk of cardiovascular disease and, in recent years, populations of HIV-infected persons have experienced a rising proportionate mortality due to cardiovascular disease [2,3]. Predisposing metabolic comorbidities (i.e. hypertension, diabetes, dyslipidaemia 1

2 2 ME Levy et al. and obesity) that are known to decrease life expectancy and lower individuals quality of life have also been linked to changes induced by HIV infection and the use of ARV drugs, such as chronic inflammation, immune activation, immune deficiency, plasma lipid imbalances, and insulin resistance [4]. With the increasing number of people aging with HIV [5], the availability of contemporary populationlevel prevalence estimates for metabolic comorbidities will be increasingly important for characterizing the epidemiology of metabolic outcomes and guiding the promotion and provision of relevant lifestyle modification and pharmacological interventions to this population. Metabolic comorbidities among HIV-infected persons have been examined in other large observational cohort studies including the EuroSIDA study [6], Data Collection on Adverse Events of Anti-HIV Drugs (D:A:D) study [7,8], Swiss HIV Cohort Study [9,10], HIV Outpatient Study (HOPS) [11], Veterans Aging Cohort Study (VACS) and its Virtual Cohort (VACS-VC) [12,13], Multicenter AIDS Cohort Study (MACS) [14 16], Women s Interagency HIV Study (WIHS) [15,17,18], North American AIDS Cohort Collaboration on Research and Design (NA-ACCORD) [19,20], and an electronic medical record (EMR) study at a health care centre in New York City [21]. Despite recent publications, however, robust prevalence estimates based on recent data are largely unavailable: data collection for the aforementioned studies occurred mostly prior to 2010, metabolic variables were often reported as baseline covariates in longitudinal analyses or in smaller substudies focused on other health outcomes, and restrictive analytic inclusion/exclusion criteria have sometimes resulted in study samples that were not representative of the larger population of HIV-infected patients (Table S1). Thus, a comprehensive investigation of the epidemiology of metabolic comorbidities among HIV-infected persons is warranted, particularly as providers have increasingly prescribed newer ARV regimens with improved metabolic side effect profiles, such as integrase inhibitors [22]. In this study, we had the unique opportunity to investigate the prevalence and correlates of four key metabolic comorbidities in a large, citywide, racially diverse, and aging contemporary cohort of HIV-infected outpatients in Washington, DC using composite case definitions that incorporated clinical, laboratory, and pharmacy data. Our impetus was to raise awareness of the burden of metabolic comorbidities among people aging with HIV, to use this analysis as a springboard to assess how effectively metabolic comorbidities are being controlled, and to highlight the need for an increased focus on the prevention of cardiovascular disease among HIV-infected patients in DC and beyond. The objectives of the present analysis were to estimate the citywide prevalence of hypertension, type 2 diabetes, dyslipidaemia and obesity; to examine differences by sex, age, and race/ethnicity in the prevalence of these comorbidities; and to assess their demographic, behavioural and clinical correlates. Methods Study design This cross-sectional analysis utilized data collected in an ongoing, prospective, multicentre, observational cohort study of HIV-infected outpatients at 13 major community, government and academic clinical sites in Washington, DC. The methodology of the DC Cohort study has recently been described in depth [23]. In brief, HIV-infected patients were enrolled on an ongoing basis beginning in January 2011 and were prospectively followed from the visit at which informed consent was provided. Sociodemographic, clinical and laboratory data documented in outpatient EMR systems were routinely monitored and abstracted into the DC Cohort database. The protocol was reviewed and approved by multiple Institutional Review Boards including that of the George Washington University. This secondary analysis included participants who were 18 years of age and enrolled by 30 June Measures Variables for which data were abstracted included sociodemographic characteristics, behavioural risk factors, HIV disease parameters, ARV regimens during the period of active follow-up, and comorbidities. Data between the start and end of the observation period for each participant (i.e. the date of consent until 30 June 2015, study inactivity or death) were used to define variables including HIV parameters and ARV regimens. For many patients, the first regimen recorded in the study database was the regimen patients were taking at the time of consent; thus, lifetime ARV history was incomplete. Sociodemographic characteristics and histories of smoking, alcohol abuse and recreational drug use (each classified as current, previous or never) were abstracted via chart review at enrolment only. Key metabolic variables that represented outcomes of interest were hypertension, type 2 diabetes, dyslipidaemia and obesity. These outcomes were defined cross-sectionally using specific criteria between the start and end of each participant s study observation period. Participants were defined as having hypertension, diabetes, and/or dyslipidaemia if they met at least one of three criteria: (1) an ICD- 9 (International Classification of Diseases, 9th Revision) code that indicated a diagnosis, (2) a drug prescription that suggested receipt of treatment, and (3) values for clinical

3 Metabolic comorbidities in HIV patients 3 or laboratory results beyond a specified threshold. Criteria for clinical/laboratory results were specified as blood pressure 140/90 mm Hg on at least two occasions (for hypertension), serum glucose 200 mg/dl or glycated haemoglobin (HbA1c) 6.5% on at least two occasions (for diabetes), and total serum cholesterol 200 mg/dl or high-density lipoprotein cholesterol (HDL-C) < 40 mg/dl on at least one occasion (for dyslipidaemia). Data on whether patients had been fasting at the time of blood collection for laboratory testing were not available in the DC Cohort database. Laboratory data for one of the 13 sites were only available until September 2014, so data on serum glucose, HbA1c, total cholesterol, and HDL-C were censored at that date for patients at that site. Obesity was defined as having evidence of a relevant ICD-9 code or body mass index (BMI) 30 on at least one occasion. The detailed definitional criteria for metabolic variables are provided in Tables S2 S5. Multimorbidity was defined as having evidence of two or more of these four conditions. Statistical analysis Descriptive statistics were calculated for each variable among the total sample and by biological sex, age group and race/ethnicity. We also calculated period prevalence estimates and 95% confidence intervals (CIs) for metabolic outcome variables based on the binomial distribution among the total sample and by demographic categories. Differences in prevalence estimates by sex, age group and race/ethnicity were assessed using v 2 tests. For each metabolic outcome variable, we also assessed the proportions of participants who met the respective criteria for each definition based on diagnoses, prescriptions, or clinical/laboratory results alone, and based on combinations of these data sources. Correlates of metabolic comorbidities were identified using multivariable logistic regression, and all relevant sociodemographic, behavioural and clinical covariates were included in multivariable models. For the purpose of regression analyses, all missing data for independent variables were multiply imputed using 30 imputed data sets. All statistical analyses were completed using SAS version 9.3 (SAS Institute, Cary, NC, USA). Results Demographic and clinical characteristics Characteristics of the 7018 HIV-infected patients are summarized in Table 1 (characteristics stratified by sex, age group and race/ethnicity are included in Table S6). The median follow-up time was 2.7 years [interquartile range (IQR) years] and the median number of visits was 7 (IQR: 3 13). The median age was 50 years (IQR: years; range years), 73% were born male, and 77% were non-hispanic black. The median time since HIV diagnosis was 12.3 years (IQR: years) and 74% received primary care at their study site. A large proportion (41%) had been diagnosed with AIDS and nearly all (97%) had initiated an ARV regimen, including 50% with a recorded history of a protease-inhibitor-based (PI-based) regimen, 42% with a recorded history of a nonnucleoside reverse transcriptase inhibitor-based (NNRTI-based) regimen, and 38% with a recorded history of an integraseinhibitor-based (INSTI-based) regimen at at least one point in time during the period of active study follow-up. Prevalence of metabolic comorbidities The period prevalence was 50% (95% CI: 49 51%) for hypertension, 13% (95% CI: 12 14%) for type 2 diabetes, 48% (95% CI: 47 49%) for dyslipidaemia, 35% (95% CI: 34 36%) for obesity [68% (95% CI: 67 79%) for overweight/obesity and 8% (95% CI: 7 8%) for severe/morbid obesity], and 46% (95% CI: 44 47%) for metabolic multimorbidity (Tables 2 and 3; Fig. 1). When obesity was defined as having an ICD-9 code or BMI 30 on at least two occasions (instead of one), the prevalence of obesity was 35% among the subset of participants with at least two recorded measurements for BMI in their EMR (or with an ICD-9 code for obesity, regardless of the number of BMI measurements). As the prevalence of obesity was similar based on whether obesity was defined using one or two BMI measurements, we elected to use a single BMI 30 as the definition for obesity in order to avoid misclassifying participants with only one recorded BMI in their EMR during the follow-up period. Men were more likely than women to have dyslipidaemia (51 vs. 41%, respectively; P < 0.001), and women were more likely than men to have type 2 diabetes (16 vs. 12%, respectively; P < 0.001) and obesity (53 vs. 29%, respectively; P < 0.001). Metabolic comorbidities were more common among older age groups, with 86% (95% CI: 81 91%) of participants 70 years of age having hypertension and 74% (95% CI: 68 80%) having dyslipidaemia. Among racial/ethnic groups, non-hispanic black patients were most likely to have hypertension, diabetes and obesity, while non-hispanic white patients were most likely to have dyslipidaemia (all P < 0.001). Proportions of patients who met specific criteria for metabolic outcomes Proportions of participants who met the respective criteria for definitions of each metabolic outcome based on

4 4 ME Levy et al. Table 1 Sociodemographic and clinical characteristics of HIVinfected patients in the DC Cohort; 1 January 2011 to 30 June 2015 (n = 7018) Characteristic Sex at birth Male 5135 (73.2) Female 1883 (26.8) Age < 40 years 1778 (25.3) years 1697 (24.2) years 2253 (32.1) years 1081 (15.4) 70 years 209 (3.0) Race/ethnicity Non-Hispanic black 5414 (77.1) Non-Hispanic white 972 (13.9) Hispanic 340 (4.8) Other/unknown* 292 (4.2) Age (years) [median (IQR)] 50 (39 57) Housing status Permanent/stable 4468 (63.7) Temporary/unstable 437 (6.2) Homeless 96 (1.4) Unknown 2017 (28.7) Employment status Employed 1955 (27.9) Unemployed 1580 (22.5) Unknown 3483 (49.6) Insurance status Medicare 955 (13.6) Medicaid 2331 (33.2) Ryan White/ADAP 85 (1.2) Public, other 896 (12.8) DC Alliance 86 (1.2) Private 1833 (26.1) Unknown 832 (11.9) Smoking history Current 2851 (40.6) Previous 1067 (15.2) Never 2297 (32.7) Unknown 803 (11.4) Alcohol abuse history Current 997 (14.2) Previous 990 (14.1) Never 3282 (46.8) Unknown 1749 (24.9) Recreational drug use history Current 873 (12.4) Previous 1602 (22.8) Never 2326 (33.1) Unknown 2217 (31.6) HIV transmission risk MSM 2693 (38.4) High-risk heterosexual 2201 (31.4) IDU 476 (6.8) MSM and IDU 82 (1.2) Perinatal 141 (2.0) Unknown 1425 (20.3) Time since HIV diagnosis (years) [median (IQR)] 12.3 ( ) AIDS diagnosis 2889 (41.2) Most recent CD4 count < 200 cells/ll 628 (8.9) cells/ll 2246 (32.0) > 500 cells/ll 4133 (58.9) Unknown 11 (0.2) Table 1 (Continued) Characteristic Nadir CD4 count < 200 cells/ll 2702 (38.5) cells/ll 3159 (45.0) > 500 cells/ll 1146 (16.3) Unknown 11 (0.2) Most recent HIV viral load < 200 copies/ml 5792 (82.5) 200 copies/ml 1177 (16.8) Unknown 49 (0.7) History of ARV therapy Any regimen 6813 (97.1) PI-based regimen 3499 (49.9) NNRTI-based regimen 2955 (42.1) INSTI-based regimen 2667 (38.0) Whether primary care is received at the site Yes 5194 (74.0) No 1741 (24.8) Unknown 83 (1.2) Anxiety/stress disorder 619 (8.8) Depression 1287 (18.3) Hepatitis C 1120 (16.0) Chronic kidney disease 517 (7.4) ADAP, AIDS Drug Assistance Program; MSM, men who have sex with men; IDU, injecting drug use; IQR, interquartile range; ARV, antiretroviral; PI, protease inhibitor; NNRTI, nonnucleoside reverse transcriptase inhibitor; INSTI, integrase inhibitor. Values are n (%) unless otherwise stated. *153 participants were of unknown race/ethnicity. 50 participants were missing a value for time since HIV diagnosis. 169 participants were ARV-na ıve and 36 participants were of unknown ARV experience. Categories for history of ARV therapy are not mutually exclusive. diagnoses, medication prescriptions, or clinical/laboratory results alone and based on combinations of these data sources are presented in Fig. 2. Each Venn diagram represents all participants classified as having one metabolic comorbidity and depicts the proportions of those who met the various diagnostic criteria. Based on the available EMR data, 38% lacked evidence of treatment for hypertension, 40% lacked evidence of treatment for diabetes, and 66% lacked evidence of treatment for dyslipidaemia (after excluding the 19% of dyslipidaemic patients whose only indication for dyslipidaemia was low HDL-C, 56% lacked evidence of treatment). Correlates of metabolic comorbidities Prevalent metabolic comorbidities were independently associated with traditional cardiovascular risk factors, greater times since HIV diagnosis and on ARV regimens during the study follow-up period, and specific immunovirological parameters (Table 4; see Table S7 for the full set of results). After adjusting for all other covariates, metabolic outcomes remained associated with sex, age and race/ethnicity. Hypertension was associated with

5 Metabolic comorbidities in HIV patients 5 Table 2 Prevalence of metabolic comorbidities among HIV-infected patients in the DC Cohort; 1 January 2011 to 30 June 2015 (n = 7018) Hypertension Type 2 diabetes Dyslipidaemia Obesity % (95% CI) P % (95% CI) P % (95% CI) P % (95% CI) P Total sample (n = 7018) 49.8 (48.7, 51.0) 12.9 (12.1, 13.7) 48.0 (46.8, 49.1) 35.2 (34.1, 36.3) By sex at birth Male (n = 5135) 49.3 (47.9, 50.6) (10.9, 12.6) < (49.4, 52.1) < (27.5, 30.0) < Female (n = 1883) 51.4 (49.2, 53.7) 15.9 (14.3, 17.6) 40.6 (38.4, 42.8) 52.8 (50.5, 55.0) By age group < 40 years (n = 1778) 24.9 (22.9, 26.9) < (2.3, 4.0) < (23.3, 27.3) < (25.9, 30.0) < years (n = 1697) 44.1 (41.8, 46.5) 10.2 (8.8, 11.6) 45.9 (43.5, 48.3) 41.7 (39.4, 44.1) years (n = 2253) 58.8 (56.7, 60.8) 15.6 (14.1, 17.1) 55.5 (53.4, 57.5) 39.0 (37.0, 41.0) years (n = 1081) 74.1 (71.5, 76.7) 24.4 (21.9, 27.0) 68.1 (65.3, 70.9) 31.1 (28.3, 33.8) 70 years (n = 209) 86.1 (81.4, 90.8) 28.7 (22.6, 34.8) 74.2 (68.2, 80.1) 23.9 (18.1, 29.7) By race/ethnicity Non-Hispanic black (n = 5414) 53.4 (52.1, 54.8) < (13.2, 15.1) < (45.1, 47.8) < (36.9, 39.5) < Non-Hispanic white (n = 972) 40.2 (37.1, 43.3) 7.9 (6.2, 9.6) 58.1 (55.0, 61.2) 23.2 (20.5, 25.8) Hispanic (n = 340) 33.2 (28.2, 38.2) 9.4 (6.3, 12.5) 49.4 (44.1, 54.7) 30.0 (25.1, 34.9) Other/unknown (n = 292) 34.3 (28.8, 39.7) 10.3 (6.8, 13.8) 42.1 (36.5, 47.8) 25.7 (20.7, 30.7) CI, confidence interval. Table 3 Prevalence of metabolic multimorbidity among HIV-infected patients in the DC Cohort; 1 January 2011 to 30 June 2015 (n = 7018) Multimorbidity (at least two comorbidities) Exactly two comorbidities Exactly three comorbidities Exactly four comorbidities % (95% CI) P % (95% CI) P % (95% CI) P % (95% CI) P Total sample (n = 7018) 45.6 (44.4, 46.8) 25.5 (24.4, 26.5) 14.9 (14.1, 15.8) 5.2 (4.7, 5.8) By sex at birth Male (n = 5135) 43.9 (42.6, 45.3) < (24.0, 26.3) (13.2, 15.2) (4.0, 5.2) < Female (n = 1883) 50.3 (48.1, 52.6) 26.3 (24.4, 28.3) 16.9 (15.3, 18.6) 7.1 (5.9, 8.2) By age group < 40 years (n = 1778) 21.5 (19.6, 23.5) < (14.3, 17.7) < (3.5, 5.4) < (0.6, 1.6) < years (n = 1697) 44.0 (41.6, 46.3) 25.0 (22.9, 27.1) 14.1 (12.4, 15.7) 4.9 (3.9, 5.9) years (n = 2253) 53.2 (51.2, 55.3) 27.4 (25.6, 29.2) 19.1 (17.5, 20.7) 6.8 (5.7, 7.8) years (n = 1081) 66.0 (63.1, 68.8) 34.1 (31.3, 37.0) 22.6 (20.1, 25.1) 9.3 (7.5, 11.0) 70 years (n = 209) 77.5 (71.9, 83.2) 44.0 (37.3, 50.8) 26.8 (20.8, 32.8) 6.7 (3.3, 10.1) By race/ethnicity Non-Hispanic black (n = 5414) 47.8 (46.5, 49.2) < (24.3, 26.6) (15.4, 17.4) < (5.4, 6.6) < Non-Hispanic white (n = 972) 40.8 (37.8, 43.9) 27.6 (24.8, 30.4) 11.1 (9.1, 13.1) 2.2 (1.3, 3.1) Hispanic (n = 340) 35.6 (30.5, 40.7) 23.5 (19.0, 28.0) 8.8 (5.8, 11.8) 3.2 (1.4, 5.1) Other/unknown (n = 292) 32.5 (27.2, 37.9) 21.2 (16.5, 25.9) 7.5 (4.5, 10.6) 3.8 (1.6, 6.0) CI, confidence interval. former [adjusted odds ratio (aor) 1.43; 95% CI 1.19, 1.72] and current smoking (aor 1.49; 95% CI 1.27, 1.75). All metabolic comorbidities were independently associated with either depression or anxiety/stress disorder and were strongly associated with each other. Greater time since HIV diagnosis was associated with greater odds of having type 2 diabetes and dyslipidaemia [each aor: 1.01 (95% CI: ) per 1-year increase]. Greater time on ARV regimens during the active study follow-up period (per 1-year increase) was associated with greater metabolic risk: use of PI-based regimens was associated with dyslipidaemia (aor: 1.11; 95% CI: ); use of NNRTI-based regimens was associated with hypertension (aor: 1.03; 95% CI: ) and dyslipidaemia (aor: 1.09; 95% CI: ); and use of any other regimen was also associated with hypertension (aor: 1.05; 95% CI: ) and dyslipidaemia (aor: 1.07; 95% CI: ). Having a nadir CD4 count < 200 cells/ll (vs. > 500 cells/ll) was associated with greater odds of hypertension (aor: 1.35; 95% CI: ) and lower odds of type 2 diabetes (aor: 0.59; 95% CI: ); having a recent CD4 count < 200 cells/ll (vs. > 500 cells/ll) was associated with lower odds of dyslipidaemia (aor: 0.56; 95% CI: ) and obesity (aor: 0.64; 95% CI: ). Having a recent HIV viral load 200 HIV-1 RNA copies/ml was associated with lower odds of dyslipidaemia (aor 0.70; 95% CI 0.59, 0.82).

6 6 ME Levy et al. Fig. 1 Bar graphs of the prevalence of metabolic comorbidities (a) among all HIV-infected patients in the DC Cohort and by (b) sex at birth, (c) age group and (d) race/ethnicity; 1 January 2011 to 30 June 2015 (n = 7018). Discussion In an aging citywide cohort of largely non-hispanic black HIV-infected outpatients, we found a very high prevalence of metabolic comorbidities despite access to health care and the availability of newer ARV regimens with improved metabolic side effect profiles. The contemporary estimates obtained in this study were based on data collected between 2011 and 2015, and thus are complementary to those of previous studies that largely reported on data collected between 2000 and As a consequence of the diverse demographic characteristics of this large study sample, we were able to calculate precise estimates for the prevalence of hypertension, diabetes, dyslipidaemia and obesity by biological sex, racial/ethnic group, and age group (five age groups were used, which included 1290 people aged 60 years). The findings of this study underscore the persistent and pervasive public health burden of metabolic comorbidities among people currently living with HIV in real-world clinical settings. The period prevalence estimates obtained for the sample of adult HIV-infected patients in the DC Cohort are generalizable to the broader HIV-infected outpatient population in a large urban US city. Half of HIV-infected patients in the DC Cohort were classified as hypertensive, which represents a higher prevalence compared with the 32% of participants in the EuroSIDA study [6], 14% of participants in the D:A:D study [8], 45% of participants in the Swiss HIV Cohort Study [10], 32% of male patients in the VACS-VC [12], 32% of female patients in the WIHS [17], 43% of HIV-infected outpatients at an urban New York health centre [21], and 44% of male patients in the MACS with hypertension [14], although it also represents a marginally lower prevalence compared with the 54% of male patients and 57% of female patients in the HOPS with hypertension [11]. Among DC Cohort patients, 12% of male patients and 16% of female patients had diabetes, which was comparable to the 12% of male patients and 19% of female patients in the HOPS [11], 13% of male patients in the MACS [14], 14% of male patients in the VACS-VC [12], and 18% of female patients in the WIHS with diabetes [17], but higher than the 5% of participants in the EuroSIDA study [6], 3% in the D:A:D study [8], and 7% in the Swiss HIV Cohort Study with diabetes [10]. More than half of male patients and more than 40% of female patients in the DC Cohort were classified as dyslipidaemic, and these prevalences are higher than the 37% of male patients and 31% of female patients at an urban New York health centre [21], 45% of EuroSIDA study participants [6], and 42% of D:A:D

7 Metabolic comorbidities in HIV patients 7 Fig. 2 Venn diagrams of the proportions of HIV-infected patients with (a) hypertension (n = 3497), (b) type 2 diabetes (n = 902), (c) dyslipidaemia (n = 3370) and (d) obesity (n = 2469) who met criteria for these metabolic comorbidities based on diagnoses, medications, and/or test results. Of the 902 patients classified as having type 2 diabetes, 30% had evidence of elevated serum glucose, 40% had evidence of elevated glycated haemoglobin (HbA1c), and 51% had evidence of either elevated serum glucose or elevated HbA1c; the latter of these is depicted by the blue circle in (b). Of the 3370 patients classified as dyslipidaemic, 52% had evidence of elevated total cholesterol, 46% had evidence of low high-density lipoprotein cholesterol (HDL-C), and 80% had evidence of either elevated total cholesterol or low HDL-C; the latter of these is depicted by the blue circle in (c). BMI, body mass index. study participants with dyslipidaemia [8], but lower than the 58% of male patients in the VACS-VC [12], 81% of male patients and 67% of female patients in the HOPS [11], and 92% of male patients in a MACS substudy with dyslipidaemia [16]; these large differences might be attributed to different histories of ARV therapy and/or the use of different quantitative cut-offs for various lipid parameters to define dyslipidaemia. Nearly twice as many women as men had obesity (53 vs. 29%, respectively) in the DC Cohort; aside from an obesity prevalence of 32% among male patients in the MACS [14], the prevalence was lower in other studies: 18% among male patients and 32% among female patients in the HOPS [11], 32% among female patients in the WIHS [17], 14% among male patients in the VACS-VC [12], and 7% among participants in the Swiss Cohort HIV Study [10]. In addition, 46% of DC Cohort patients with a non-missing value for smoking history were current smokers; previous studies have similarly found a high prevalence of current smoking, ranging from 41 to 57% among HIV-infected participants [7,9,24 26]. Prevalence estimates stratified by demographic subgroups further elucidate the very high burden of metabolic comorbidities among older HIV-infected patients.

8 8 ME Levy et al. Table 4 Correlates of metabolic comorbidities among HIV-infected patients in multivariable analysis (n = 7018) Hypertension Type 2 diabetes Dyslipidaemia Obesity aor (95% CI) aor (95% CI) aor (95% CI) aor (95% CI) Smoking history Never 1.00 ( ) 1.00 ( ) 1.00 ( ) 1.00 ( ) Previous 1.43 (1.19, 1.72)*** 1.07 (0.85, 1.36) 0.96 (0.80, 1.14) 1.00 (0.84, 1.19) Current 1.49 (1.27, 1.75)*** 0.91 (0.73, 1.14) 0.96 (0.82, 1.12) 0.79 (0.68, 0.93)** Alcohol abuse history Never 1.00 ( ) 1.00 ( ) 1.00 ( ) 1.00 ( ) Previous 1.17 (0.98, 1.41) 1.00 (0.79, 1.26) 1.22 (1.02, 1.46)* 1.04 (0.87, 1.25) Current 1.43 (1.17, 1.75)*** 0.68 (0.50, 0.91)* 1.11 (0.90, 1.36) 0.94 (0.78, 1.14) Anxiety/stress disorder 1.48 (1.21, 1.81)*** 0.96 (0.73, 1.25) 1.21 (1.00, 1.47)* 1.04 (0.86, 1.26) Depression 1.13 (0.98, 1.31) 1.22 (1.01, 1.48)* 1.26 (1.09, 1.46)** 1.19 (1.04, 1.38)* Hepatitis C 1.46 (1.22, 1.75)*** 1.05 (0.84, 1.31) 0.54 (0.45, 0.64)*** 0.70 (0.59, 0.84)*** Chronic kidney disease 4.82 (3.64, 6.38)*** 1.26 (1.00, 1.60) 1.15 (0.93, 1.43) 0.93 (0.75, 1.14) Hypertension 3.04 (2.48, 3.72)*** 1.89 (1.68, 2.12)*** 2.13 (1.89, 2.41)*** Diabetes 2.94 (2.40, 3.60)*** 2.53 (2.11, 3.03)*** 2.01 (1.71, 2.37)*** Dyslipidaemia 1.87 (1.66, 2.11)*** 2.57 (2.14, 3.09)*** 1.73 (1.54, 1.95)*** BMI < ( ) 1.00 ( ) 1.00 ( ) 25< (1.26, 1.67)*** 1.24 (0.98, 1.56) 1.41 (1.27, 1.66)*** (2.23, 2.97)*** 2.33 (1.88, 2.90)*** 2.10 (1.83, 2.43)*** Time since HIV diagnosis (per year increase) 1.00 (1.00, 1.01) 1.01 (1.00, 1.02)* 1.01 (1.00, 1.02)* 1.00 (0.99, 1.01) Time on a PI-based ARV regimen (per year increase) 1.01 (0.99, 1.04) 1.01 (0.98, 1.04) 1.11 (1.09, 1.14)*** 0.98 (0.96, 1.00) Time on an NNRTI-based ARV regimen (per year increase) 1.03 (1.01, 1.06)** 1.01 (0.97, 1.04) 1.09 (1.06, 1.12)*** 0.95 (0.93, 0.98)*** Length of time on any other ARV regimen (per year increase) 1.05 (1.01, 1.08)** 1.04 (1.00, 1.08) 1.07 (1.04, 1.11)*** 0.95 (0.92, 0.98)** AIDS diagnosed 0.95 (0.75, 1.20) 1.27 (0.94, 1.71) 1.04 (0.83, 1.30) 1.10 (0.88, 1.37) Most recent CD4 count > 500 cells/ll 1.00 ( ) 1.00 ( ) 1.00 ( ) 1.00 ( ) cells/ll 0.92 (0.80, 1.06) 0.92 (0.76, 1.11) 0.82 (0.72, 0.94)** 0.72 (0.63, 0.82)*** < 200 cells/ll 0.82 (0.64, 1.04) 1.08 (0.77, 1.52) 0.56 (0.44, 0.70)*** 0.64 (0.50, 0.81)*** Nadir CD4 count > 500 cells/ll 1.00 ( ) 1.00 ( ) 1.00 ( ) 1.00 ( ) cells/ll 1.20 (1.02, 1.42)* 0.83 (0.66, 1.05) 0.87 (0.74, 1.02) 0.90 (0.77, 1.06) < 200 cells/ll 1.35 (1.01, 1.79)* 0.59 (0.41, 0.86)** 1.08 (0.82, 1.42) 0.88 (0.67, 1.15) Most recent viral load < 200 copies/ml 1.00 ( ) 1.00 ( ) 1.00 ( ) 1.00 ( ) 200 copies/ml 0.91 (0.77, 1.07) 1.09 (0.86, 1.38) 0.70 (0.59, 0.82)*** 0.91 (0.78, 1.07) aor, adjusted odds ratio; BMI, body mass index; CI, confidence interval; PI, protease inhibitor; ARV, antiretroviral; NNRTI, nonnucleoside transcriptase inhibitor. *P < 0.05; **P < 0.01; ***P < The full set of results from univariable and multivariable logistic regression analyses are included in Table S7. All multivariable analyses adjusted for age, sex at birth, race/ethnicity, HIV transmission risk category, housing status, employment status, smoking history, alcohol abuse history, recreational drug use history, whether primary care is received at the clinical site, anxiety/stress disorder, depression, hepatitis C, chronic kidney disease, hypertension, diabetes, dyslipidaemia, overweight/obesity, time since HIV diagnosis, time on PI-based regimens, time on NNRTI-based regimens, time on other (non-pi-, non-nnrti-based) regimens, most recent CD4 cell count, nadir CD4 cell count, AIDS diagnosis, and most recent viral load. There was evidence of hypertension, dyslipidaemia and multimorbidity among more than half of those aged years and a large majority of those aged 60 years, and there was evidence for type 2 diabetes in approximately one-quarter of those aged 60 years. Among older HIV-infected persons, the prevalence of hypertension was notably higher in the DC Cohort than in other studies: 74% of DC Cohort patients aged years and 86% of those aged 70 years had hypertension, compared with 60% of HIV-infected patients aged 60 years in an Italian study [27] and 45% in a study of HIVinfected US Veterans aged 60 years [28]. The prevalence of type 2 diabetes in the DC Cohort was 24% among patients aged years and 29% among those aged 70 years, which was higher than the 12% of HIVinfected Veterans aged 60 years with diabetes [28], but lower than the 39% of HIV-infected Italian patients aged 60 years with diabetes [27]. A significant statistical interaction between HIV status and age has also been reported previously, whereby the risk of comorbidity was greater among older HIV-infected persons than would have been expected from the independent effects of HIV infection and age alone [28,29]. In terms of sex and race/ ethnicity, male and white individuals were more likely to have dyslipidaemia [11,15], female and black individuals were more likely to have diabetes and obesity [11,15], and black individuals were more likely to have hypertension [11,21]. Furthermore, our use of composite metabolic outcome measures that incorporated data on diagnoses, medication

9 Metabolic comorbidities in HIV patients 9 prescriptions and clinical/laboratory results maximized the sensitivity of definitions for metabolic comorbidities. Large proportions of patients classified as having metabolic comorbidities lacked documented evidence in their EMR of having received a prescription for treatment. This finding should be further investigated as it could point to unmet treatment opportunities for hypertension, diabetes and dyslipidaemia, and also to the reality that care for comorbidities is often sought elsewhere such as a separate primary care provider. Some studies of HIV-infected patients have reported gaps in treatment, with only 56% treated for hypertension, 56% treated for diabetes, and 39% and 13% treated for dyslipidaemia in various patient populations [30 32]. Other studies, however, have found higher treatment rates, including 75% for hypertension and 81 87% and 90% for dyslipidaemia [21,33]. Reported barriers to the treatment of hypertension, diabetes and dyslipidaemia in the general population have included provider disagreement with clinical recommendations, the absence of an adequate patient care plan, overestimation of the effectiveness of diet control as a treatment modality, complex treatment regimens, poor patient education, failure to appreciate the severity of potential complications, perceived and actual drug side effects, and depression or stress/anxiety hindering the adoption of a new treatment regimen [34 36]. Older HIV-infected patients often face additional barriers to non-hiv medication adherence such as neurocognitive dysfunction, substance use disorders, food insecurity, and limited social support [37]. As a next step, we intend to assess whether patients met specific criteria for the treatment of hypertension, diabetes and dyslipidaemia based on published recommendations for the management of these conditions. Regarding interventions to mitigate the burden of metabolic disease, the promotion of lifestyle modification offers additional beneficial impact to patients, as interventional programmes focused on diet, exercise and/or smoking cessation have improved metabolic and cardiovascular outcomes in people living with HIV [38 40]. Information on the management of metabolic conditions using behavioural change was not available in our study, but such lifestyle modification may account for high proportions of participants lacking documented treatment. In cross-sectional analyses, metabolic comorbidities were independently associated with traditional cardiovascular risk factors, greater times since HIV diagnosis and on ARV therapy during the active study follow-up period, and immunovirological parameters representative of controlled HIV infection. Patients with depression, anxiety/stress disorder, and/or a history of alcohol abuse had a greater risk of various metabolic outcomes. Although not focused on metabolic risk, previous research among people with HIV found that depressive symptoms were correlated with comorbid conditions [41] and that alcohol abuse and dependence were associated with a higher prevalence of cardiovascular disease [42]. Our finding that greater times on ARV regimens during the study period were associated with greater metabolic risk is supported by previous evidence that various ARV drugs and regimens were associated with an increased risk of diabetes [43,44], hypertension [45,46] and dyslipidaemia [47]. Greater time since one s HIV diagnosis was also independently associated with increased risk of diabetes and dyslipidaemia, which contributes to the growing body of evidence that the natural course of HIV infection induces specific atherogenic changes such as imbalances in serum concentrations of triglycerides, HDL-C and glucose [48 50]. While low nadir CD4 cell count was associated with hypertension in our analysis and in one other study [51], higher CD4 cell count and lower HIV viral load were, surprisingly, associated with the presence of diabetes, dyslipidaemia and obesity. Given these conflicting findings and the number of statistical tests conducted, it is possible that these associations were spurious. Additional research on the relationships between immunovirological and metabolic parameters, particularly research utilizing longitudinal methods, is warranted. These study findings should be interpreted within the context of several limitations. In defining variables crosssectionally based on each patient s available EMR data, patients with shorter follow-up periods and fewer visits had less opportunity to meet criteria for metabolic variable definitions; however, this approach allowed for the complete utilization of EMR data. Laboratory data for patients at one of the 13 sites were unavailable between September 2014 and June 2015, so the prevalences calculated for diabetes and dyslipidaemia may slightly underestimate the true prevalences of these conditions, and the proportions classified as having diabetes and dyslipidaemia based on laboratory results may also be underrepresented. Data on whether participants were fasting were unavailable, so we were unable to use measurements for biomarkers that are less reliable in nonfasting states, such as triglycerides. This analysis was cross-sectional and, thus, temporality and causality of associations cannot be determined. In addition, regression analyses consisted of multiple comparisons of covariates with four metabolic outcome variables, so spurious associations are possible and associations should therefore be interpreted with appropriate caution. Regarding ARV regimens, information on adherence was not collected. Although data on regimens were abstracted at enrolment, lifetime exposure to ARV therapy was incomplete; thus, associations between length of time on ARV

10 10 ME Levy et al. regimens and metabolic outcomes probably underestimate true associations. Moreover, specific behavioural and social variables were only measured via chart review at enrolment and there were missing data for these variables; however, values were multiply imputed for the purpose of regression modelling. Conclusions These findings underscore the pervasive public health burden of metabolic comorbidities in a contemporary citywide cohort of HIV-infected persons and have critical implications for the evolving health care needs of people aging with HIV in our community. Models of HIV care that more centrally incorporate primary and secondary prevention of metabolic disease may be warranted to most effectively monitor metabolic risk and prevent subsequent adverse cardiovascular outcomes among HIVinfected patients [52,53]. Our results also serve as the basis for future analyses to characterize how effectively metabolic comorbidities are controlled in the DC Cohort and to elucidate barriers to their successful management in HIV clinical care settings. This study contributes to a better understanding of the epidemiology of metabolic comorbidities in HIV-infected populations and demonstrates the increasingly important public health priority that managing metabolic and cardiovascular risk represents for HIV-infected patients and their providers. Acknowledgements We would like to thank the site Principal Investigators (PIs), Research Assistants (RAs), the Community Advisory Board (CAB), the patients themselves, the DC Department of Health, and the National Institutes of Health for their contributions to the DC Cohort. The authors are also grateful for the methodological expertise and guidance offered by Dr Manya Magnus regarding multiple imputation of missing data. Funding and support: This work was supported by the National Institute of Allergy and Infectious Diseases at the National Institutes of Health (NIH; grant number UM1AI069503) and in part by the National Institute of Allergy and Infectious Diseases Intramural Research Program (to C.H.). This work was facilitated in part by the infrastructure and services provided by the District of Columbia Center for AIDS Research, an NIH-funded programme (grant number P30AI117970), which is supported by the following NIH Co-Funding and Participating Institutes and Centers: National Institute of Allergy and Infectious Diseases; National Cancer Institute; Eunice Kennedy Shriver National Institute of Child Health and Human Development; National Heart, Lung, and Blood Institute; National Institute on Drug Abuse; National Institute of Mental Health; National Institute on Aging; Fogarty International Center; National Institute of General Medical Sciences; National Institute of Diabetes and Digestive and Kidney Diseases; and Office of AIDS Research. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH. Conflicts of interest: The authors have no conflicts of interest to declare. Appendix: the DC Cohort Executive Committee Debra Benator, Veterans Affairs Medical Center; Princy Kumar, Georgetown University; Deborah Goldstein and David Hardy, Whitman-Walker Health; David Parenti, George Washington Medical Faculty Associates; Maria Elena Ruiz, Washington Hospital Center; Angela Wood, Family and Medical Counseling Service; Lawrence D Angelo, Burgess Adolescent Clinic, Children s National Medical Center; Natella Rakhmanina, Special Immunology Service Pediatric Clinic, Children s National Medical Center; Sohail Rana, Pediatric Clinic, Howard University Hospital; Saumil Doshi, Adult Infectious Disease Clinic, Howard University Hospital; Annick Hebou, MetroHealth; Ricardo Fernandez, La Clinica Del Pueblo; Stephen Abbott, Unity Health Care; Michael Kharfen, HIV/AIDS, Hepatitis, Sexually Transmitted Diseases, Tuberculosis Administration (HAHSTA), DC Department of Health; and Henry Masur, National Institutes of Health. References 1 Deeks SG, Lewin SR, Havlir DV. The end of AIDS: HIV infection as a chronic disease. Lancet 2013; 382: Islam FM, Wu J, Jansson J, Wilson DP. Relative risk of cardiovascular disease among people living with HIV: a systematic review and meta-analysis. HIV Med 2012; 13: Feinstein MJ, Bahiru E, Achenbach C et al. Patterns of cardiovascular mortality for HIV-infected adults in the United States: 1999 to Am J Cardiol 2016; 117: Lake JE, Currier JS. Metabolic disease in HIV infection. Lancet Infect Dis 2013; 13: Lazarus J, Nielsen K. HIV and people over 50 years old in Europe. HIV Med 2010; 11: Shahmanesh M, Schultze A, Burns F et al. The cardiovascular risk management for people living with HIV in Europe: how well are we doing? AIDS 2016; 30:

11 Metabolic comorbidities in HIV patients 11 7 Friis-Moller N, Weber R, Reiss P et al. Cardiovascular disease risk factors in HIV patients association with antiretroviral therapy. Results from the DAD study. AIDS 2003; 17: Friis-Moller N, Ryom L, Smith C et al. An updated prediction model of the global risk of cardiovascular disease in HIVpositive persons: the data-collection on adverse effects of anti-hiv drugs (D:A:D) study. Eur J Prev Cardiol 2016; 23: Glass T, Ungsedhapand C, Wolbers M et al. Prevalence of risk factors for cardiovascular disease in HIV-infected patients over time: the Swiss HIV Cohort Study. HIV Med 2006; 7: Hasse B, Tarr PE, Marques-Vidal P et al. Strong impact of smoking on multimorbidity and cardiovascular risk among human immunodeficiency virus-infected individuals in comparison with the general population. Open Forum Infect Dis 2015; 8: ofv Buchacz K, Baker RK, Palella FJ Jr et al. Disparities in prevalence of key chronic diseases by gender and race/ ethnicity among antiretroviral-treated HIV-infected adults in the US. Antivir Ther 2013; 18: Sico JJ, Chang C-CH, So-Armah K et al. HIV status and the risk of ischemic stroke among men. Neurology 2015; 84: Armah KA, McGinnis K, Baker J et al. HIV status, burden of comorbid disease, and biomarkers of inflammation, altered coagulation, and monocyte activation. Clin Infect Dis 2012; 55: Monroe AK, Fu W, Zikusoka MN et al. Low-density lipoprotein cholesterol levels and statin treatment by HIV status among Multicenter AIDS Cohort Study men. AIDS Res Hum Retroviruses 2015; 31: Kaplan RC, Kingsley LA, Sharrett AR et al. Ten-year predicted coronary heart disease risk in HIV-infected men and women. Clin Infect Dis 2007; 45: Abraham AG, Darilay A, McKay H et al. Kidney dysfunction and markers of inflammation in the multicenter AIDS cohort study. J Infect Dis 2015; 212: Sobieszczyk ME, Hoover DR, Anastos K et al. Prevalence and predictors of metabolic syndrome among HIV-infected and HIV-uninfected women in the Women s Interagency HIV Study. J Acquir Immune Defic Syndr 2008; 48: Schwartz JB, Moore KL, Yin M et al. Relationship of vitamin D, HIV, HIV treatment, and lipid levels in the Women s Interagency HIV Study of HIV-infected and uninfected women in the United States. J Int Assoc Provid AIDS Care 2014; 13: Koethe JR, Jenkins CA, Lau B et al. Rising obesity prevalence and weight gain among adults starting antiretroviral therapy in the United States and Canada. AIDS Res Hum Retroviruses 2016; 32: Koethe J, Jenkins C, Lau B et al. Body mass index and early CD4 T-cell recovery among adults initiating antiretroviral therapy in North America, HIV Med 2015; 16: Myerson M, Poltavskiy E, Armstrong EJ, Kim S, Sharp V, Bang H. Prevalence, treatment, and control of dyslipidemia and hypertension in 4278 HIV outpatients. J Acquir Immune Defic Syndr 2014; 66: Srinivasa S, Grinspoon SK. Metabolic and body composition effects of newer antiretrovirals in HIV-infected patients. Eur J Endocrinol 2014; 170: R185 R Greenberg AE, Hays H, Castel AD et al. Development of a large urban longitudinal HIV clinical cohort using a webbased platform to merge electronically and manually abstracted data from disparate medical record systems: technical challenges and innovative solutions. J Am Med Inform Assoc 2016; 23: Data Collection on Adverse Events of Anti-HIV drugs (D:A:D) Study Group. Factors associated with specific causes of death amongst HIV-positive individuals in the D: A: D Study. AIDS 2010; 24: Lifson AR, Neuhaus J, Arribas JR, van den Berg-Wolf M, Labriola AM, Read TR. Smoking-related health risks among persons with HIV in the strategies for management of antiretroviral therapy clinical trial. Am J Public Health 2010; 100: Crothers K, Goulet JL, Rodriguez-Barradas MC et al. Impact of cigarette smoking on mortality in HIV-positive and HIVnegative veterans. AIDS Educ Prev 2009; 21: Guaraldi G, Orlando G, Zona S et al. Premature age-related comorbidities among HIV-infected persons compared with the general population. Clin Infect Dis 2011; 53: Goulet JL, Fultz SL, Rimland D et al. Aging and infectious diseases: do patterns of comorbidity vary by HIV status, age, and HIV severity? Clin Infect Dis 2007; 45: Rodriguez-Penney AT, Iudicello JE, Riggs PK et al. Comorbidities in persons infected with HIV: increased burden with older age and negative effects on health-related quality of life. AIDS Patient Care STDS 2013; 27: Freiberg MS, Leaf DA, Goulet JL et al. The association between the receipt of lipid lowering therapy and HIV status among veterans who met NCEP/ATP III criteria for the receipt of lipid lowering medication. J Gen Intern Med 2009; 24: Estrada V, Bernardino JI, Masia M et al. Cardiovascular risk factors and lifetime risk estimation in HIV-infected patients under antiretroviral treatment in Spain. HIV Clin Trials 2015; 16:

12 12 ME Levy et al. 32 Reinsch N, Neuhaus K, Esser S et al. Are HIV patients undertreated? Cardiovascular risk factors in HIV: results of the HIV-HEART study. Eur J Prev Cardiol 2012; 19: Lichtenstein KA, Armon C, Buchacz K et al. Provider compliance with guidelines for management of cardiovascular risk in HIV-infected patients. Prev Chronic Dis 2013; 10: E Ross SA. Breaking down patient and physician barriers to optimize glycemic control in type 2 diabetes. Am J Med 2013; 126: S38 S Khatib R, Schwalm J-D, Yusuf S et al. Patient and healthcare provider barriers to hypertension awareness, treatment and follow up: a systematic review and meta-analysis of qualitative and quantitative studies. PLoS One 2014; 9: e Chee YJ, Chan HHV, Tan NC. Understanding patients perspective of statin therapy: can we design a better approach to the management of dyslipidaemia? A literature review Singapore Med J 2014; 55: Edelman EJ, Gordon KS, Glover J, McNicholl IR, Fiellin DA, Justice AC. The next therapeutic challenge in HIV: polypharmacy. Drugs Aging 2013; 30: Fitch KV, Anderson EJ, Hubbard JL et al. Effects of a lifestyle modification program in HIV-infected patients with the metabolic syndrome. AIDS 2006; 20: Fleetwood C, Campa A, Martinez S et al. Nutritional intervention decreases risk of metabolic syndrome and improves quality of diet. FASEB J 2015; 29: Moscou-Jackson G, Commodore-Mensah Y, Farley J, DiGiacomo M. Smoking-cessation interventions in people living with HIV infection: a systematic review. J Assoc Nurses AIDS Care 2014; 25: Havlik RJ, Brennan M, Karpiak SE. Comorbidities and depression in older adults with HIV. Sex Health 2011; 8: Freiberg MS, McGinnis KA, Kraemer K et al. The association between alcohol consumption and prevalent cardiovascular diseases among HIV infected and uninfected men. J Acquir Immune Defic Syndr 2010; 53: Butt AA, McGinnis K, Rodriguez-Barradas MC et al. HIV infection and the risk of diabetes mellitus. AIDS 2009; 23: Capeau J, Bouteloup V, Katlama C et al. Ten-year diabetes incidence in 1046 HIV-infected patients started on a combination antiretroviral treatment. AIDS 2012; 26: Seaberg EC, Munoz A, Lu M et al. Association between highly active antiretroviral therapy and hypertension in a large cohort of men followed from 1984 to AIDS 2005; 19: Crane HM, Van Rompaey SE, Kitahata MM. Antiretroviral medications associated with elevated blood pressure among patients receiving highly active antiretroviral therapy. AIDS 2006; 20: Dube MP, Cadden JJ. Lipid metabolism in treated HIV Infection. Best Pract Res Clin Endocrinol Metab 2011; 25: Oliviero U, Bonadies G, Apuzzi V et al. Human immunodeficiency virus per se exerts atherogenic effects. Atherosclerosis 2009; 204: Bonfanti P, De Socio GL, Marconi P et al. Is metabolic syndrome associated to HIV infection per se? Results from the HERMES study. Curr HIV Res 2010; 8: Paula AA, Falcao MC, Pacheco AG. Metabolic syndrome in HIV-infected individuals: underlying mechanisms and epidemiological aspects. AIDS Res Ther 2013; 10: Manner IW, Troseid M, Oektedalen O, Baekken M, Os I. Low nadir CD4 cell count predicts sustained hypertension in HIVinfected individuals. J Clin Hypertens (Greenwich) 2013; 15: Justice AC. HIV and aging: time for a new paradigm. Curr HIV/AIDS Rep 2010; 7: Chu C, Selwyn PA. An epidemic in evolution: the need for new models of HIV care in the chronic disease era. J Urban Health 2011; 88: Supporting Information Additional Supporting Information may be found in the online version of this article at the publisher s web-site: Table S1. Previous publications of the prevalence of hypertension, diabetes, dyslipidaemia, and obesity in large-scale HIV-infected cohort studies. Table S2. Clinical criteria used to define hypertension. Table S3. Clinical criteria used to define type 2 diabetes. Table S4. Clinical criteria used to define dyslipidaemia. Table S5. Clinical criteria used to define obesity. Table S6. Sociodemographic and clinical characteristics of HIV-infected patients in the DC Cohort by sex, age group and race/ethnicity; 1 January 2011 to 30 June 2015 (n = 7018). Table S7. Correlates of metabolic comorbidities among HIV-infected patients in the DC Cohort; 1 January 2011 to 30 June 2015 (n = 7018).