PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
 
J Acquir Immune Defic Syndr. Author manuscript; available in PMC 2013 April 1.
Published in final edited form as:
PMCID: PMC3299935
NIHMSID: NIHMS350319

TRENDS IN REASONS FOR HOSPITALIZATION IN A MULTISITE UNITED STATES COHORT OF PERSONS LIVING WITH HIV, 2001 - 2008

Stephen A. Berry, MD PhD,1 John A. Fleishman, PhD,2 Richard D. Moore, MD, MHS,1 and Kelly A. Gebo, MD, MPH1, for the HIV Research Network

Abstract

Introduction

Hospitalization rates for comorbid conditions among persons living with HIV in the current HAART era are unknown.

Methods

Hospitalization data from 2001 – 2008 were obtained on 11,645 adults receiving longitudinal HIV care at 4 geographically diverse U.S. HIV clinics within the HIV Research Network. Modified clinical classification software from the Agency for Healthcare Research and Quality assigned primary ICD-9 codes into diagnostic categories. Analysis was performed with repeated measures negative binomial regression.

Results

During 2001 – 2008, the rate of AIDS defining illness (ADI) hospitalizations declined from 6.7 to 2.7 per 100 person years (PY), incidence rate ratio per year, 0.89 [0.87, 0.91]. Among the other diagnostic categories with average rates > 2 per 100 PY, cardiovascular hospitalizations increased over time (1.07 [1.03, 1.11]), while non-AIDS defining infection (0.98 [0.96, 1.00]), psychiatric (0.96 [0.93, 1.00]), and gastrointestinal/liver (0.96 [0.92, 1.00]) were slightly decreasing or stable. While less frequent overall, renal and pulmonary admissions also increased over time in univariate and multivariate analyses. Of all diagnostic categories, ADI admissions had the longest mean length of stay, 10.5 days.

Discussion

ADI hospitalizations have continued to decline in recent years but are still relatively frequent and potentially costly given long lengths of stay. Increases or stability in the rates of chronic end-organ disease admissions imply a need for broader medical knowledge among individual clinicians and/or teams who care for persons living with HIV and a need for long-term access to medications for these conditions.

INTRODUCTION

Persons living with HIV (PLWH) continue to be hospitalized at high rates.1-6 HIV clinicians may use knowledge of the current rates of illnesses in their efforts at prevention and early identification, and for planning management of patients with organ-specific specialists. Policy-makers and healthcare payers may use present trends to project future areas of clinical need and associated costs.

No clear pattern of trends in illness rates among U.S. PLWH in the current ART era has yet emerged. An examination of reasons for hospitalization in 2001 in the geographically diverse, multisite HIV Research Network (HIVRN) revealed that AIDS defining illnesses (ADI) were the most frequent reason for admission, followed by gastrointestinal/liver, psychiatric, and cardiovascular.7 Recent longitudinal analyses from two other U.S. cohorts, the HIV Outpatient Study (HOPS) and the military HIV Natural History Study (NHS), concurred in finding non-AIDS defining infections, ADI, gastrointestinal/liver, cardiovascular, and pulmonary among the leading categories.4, 5 However, these two studies did not show agreement on time trends within these categories. Factors which may be changing the pattern of morbidities since the early ART era include longer cumulative exposure to antiretrovirals and the aging of PLWH.8

This study evaluates time trends in hospitalization rates and associated lengths of stay (LOS) among diagnostic categories from 2001–2008 in the HIVRN. We include more recent years than were included in the HOPS analysis, and we examine a non-military population which may be more generally representative of U.S. PLWH than is the NHS. We have previously described a decline in the all-cause hospitalization rate in the HIVRN from 35 to 27 / 100 person years (PY) from 2002 to 2007.3

METHODS

Study Population and Data Collection

The HIVRN is a consortium of 12 sites providing longitudinal adult HIV care in 11 U.S. cities.3, 9 Sites abstract comprehensive demographic, laboratory, and inpatient and outpatient utilization data from clinical records, then strip these data of identifying characteristics and submit them to a data coordinating center where they are reviewed and combined. Institutional review boards at each site and at the data coordinating center at Johns Hopkins University have approved the collection and use of these data.

Four HIVRN sites were chosen for this analysis because of complete collection of reasons for hospitalization in the form of International Classification of Diseases, Ninth Revision (ICD-9) codes. These codes are assigned by trained clinical abstractors in the generation of billing claims with the first-listed code generally required to be the primary reason for hospitalization. The four HIVRN sites are academically affiliated and are located in the West (2), the South (1), and the Northeast (1). All patients ≥ 18 years old receiving longitudinal HIV care at these sites 2001–2008 were included in this analysis.

Years of active outpatient care were defined by having at least one HIV clinician visit and one measured CD4 cell count (either of which could have been routinely scheduled or generated by an acute complaint). Loss to follow-up was defined as becoming inactive from outpatient care for any reason other than death. Subjects returned to observation if and when they returned to active care. Although sometimes available, hospitalization data from inactive years were excluded because some subjects who were lost to follow-up probably received care, including hospitalizations, at outside institutions.

Outcome Variables

Several steps were taken in assigning each hospitalization to a single diagnostic category. The first step was determining the primary ICD-9 code. Using a method similar to one we have previously employed,10, 11 the first listed code referring to neither HIV (042, V08, 795.71, V01.79) nor chronic Hepatitis C (070.44, 070.54, 070.70, 070.71) was defined as the primary code for the hospitalization. Hospitalizations (173 total) with a first-listed code for “chemotherapy encounter” (V58.11 and V58.12) were assigned to the first subsequent code for a type of cancer.

Secondly, Clinical Classification Software (CCS) developed by the Agency for Healthcare Research and Quality 12 was used to assign the primary ICD-9 code into one of 18 first-level categories, e.g. infection, cardiovascular. Hospitalizations (42 total, or 0.3%) for which the only ICD-9 code referred to HIV or to chronic Hepatitis C were classified as missing.

Lastly, we modified the CCS classification in several ways. First, the CCS assigns many infections to an organ system category rather than to the infection category. We reassigned ICD-9 codes falling into the following CCS sub-levels to the infection category: central nervous system infection; infection of the eye; otitis media; endocarditis; respiratory infection; intestinal infection; anal and rectal conditions; peritonitis and intestinal abscess; urinary tract infections; inflammatory conditions of the genitals; skin and subcutaneous tissue infections; infective arthritis and osteomyelitis; infection and inflammation of an internal prosthesis; postoperative infection. We also reassigned unspecified sepsis (995.91 and 995.92) from the injury/poisoning category to the infectious category, alcoholic cirrhosis (571.2) from psychiatric to gastrointestinal/liver, and hypertensive chronic kidney disease (403.00–403.91) from cardiovascular to renal/genitourinary (“renal”). Next, we constructed a separate category for ADI by reassigning appropriate admissions according to individual ICD-9 codes (Supplementary Table) as per the 1993 Centers for Disease Control and Prevention Revised Classification.13 Recurrent bacterial pneumonia was defined as any bacterial pneumonia admission occurring within > 30 but ≤ 365 days of a previous such admission.

In order to determine the most frequent individual diagnoses within each category, frequently-appearing individual ICD-9 codes were explored using an online ICD-9 description tool.14 Where appropriate, we grouped highly-similar codes into individual diagnosis groups. For example, within the cardiovascular category, unspecified chest pain (786.50); precordial pain (786.51); and discomfort, pressure, or tightness in the chest (786.59) were grouped together as ‘chest pain’. The Supplementary Table describes all such groupings.

For each diagnostic category, two LOS outcomes were examined. The first was the PY-specific mean LOS, and the second was the patient-specific mean LOS aggregated across all years. In creating these variables, the LOS for an individual admission was determined by subtracting the admission date from the discharge date and adding one. Hence, same-day admission and discharges counted as one day. PY with no admissions were not included in LOS analyses.

Independent Variables

We analyzed several covariates thought to be associated with hospitalization rate. We categorized race as non-Hispanic Black (“Black”), non-Hispanic White (“White”), Hispanic, and other. The racial groups Asian / Pacific Islander and American Indian comprised too few subjects (117 and 44, respectively) to perform meaningful statistical tests. For purposes of analysis, these groups were combined into the “other” category. Subjects reporting an HIV risk factor of injection drug use (IDU) in conjunction with another risk factor, e.g. same sex male contact (MSM), were coded as IDU. Age, CD4 count, and HIV-1 RNA level were time varying according to calendar year. Age was categorized as 18–35, 36–49, and ≥ 50 years old. For CD4 and HIV-1 RNA, the first measurements of each year were used.

Analysis

Rates of hospitalizations due to all causes and within each diagnostic category were calculated as the number of hospitalizations per 100 PY. We conducted comparative cross-sectional analyses. After graphic exploration of rates over time, calendar year was coded as a linear trend. Assessments of time trends in hospitalization rates and in yearly mean LOS were performed with repeated-measures negative binomial regression, separately by diagnostic category. Differences in time-aggregated mean LOS between diagnostic categories were assessed in a negative binomial regression model with dummy variables for each diagnostic category (with ADI as the reference). All models used generalized estimating equations and robust variance estimators. A two-sided type I error of 5% was considered statistically significant. All analyses were performed using Stata 11.0 (StataCorp LP, College Station, TX, USA).15

RESULTS

During 2001–2008, 11,645 subjects were observed in 40,499 PY of active outpatient care. The median number of active years was 3 (interquartile range (IQR), 1-5). At least one instance of drop out from active care occurred for 6,975 subjects (60%) with 1,482 subjects (21% of those who dropped out) returning to observation at least once and 1,128 (16% of those who dropped out) having their only instance of drop out due to death. In 2001, the cohort was 72% male, 43% Black, 39% White, 15% Hispanic, 28% IDU, and had a median age of 40 years (Table 1). By 2008, relative decreases in the percentages of Blacks (41%) and Whites (36%) were accompanied by an increase among Hispanics (19%) (chi-square P < 0.01 for each comparison); IDU decreased to 20% (chi-square P < 0.001); and median age increased to 45 years (rank sum P < 0.001). The median CD4 count on first measurement of the year increased steadily from 341 (177–550) to 416 (240–613) cells / μl between 2001 and 2008 (rank sum P < 0.001). The percentage of subjects with HIV RNA < 400 copies increased from 35% in 2001 to 59% in 2008 (OR per year [95% confidence interval], 1.19 [1.18, 1.20]).

Table 1
Patient characteristics

During active care, 4,423 subjects (38% of all subjects) were hospitalized at least once, resulting in a total of 13,323 hospitalizations. Among those hospitalized, the median number of admissions was 2 (1-4). The unadjusted all-cause hospitalization rate showed a decline from 37.1 / 100 PY in 2001 to 28.9 / 100 PY in 2008 (incidence rate ratio (IRR) per year 0.98 [0.97, 0.99], Figure 1).

Figure
Unadjusted hospitalization rates by diagnostic categories

Non-AIDS defining infection was the most frequent diagnostic category, totaling 26% of all admissions and having a mean rate across the study period of 8.6 / 100 PY. For all years combined, the three most frequent individual non-AIDS defining infection diagnoses were bacterial pneumonia, cellulitis, and septicemia (Table 2). Together, these three diagnoses were responsible for 13% of all admissions. The unadjusted hospitalization rate for non-AIDS defining infections showed a borderline significant decline over time, IRR per year 0.98 [0.96, 1.00] (Figure 1).

Table 2
Frequencies of diagnostic categories and of the three most frequent individual diagnoses within each category among 13,323 total admissions

ADI was the second most frequent diagnostic category, at 15% of all-cause hospitalizations over the study period. The ADI hospitalization rate declined from 6.7 in 2001 to 2.7 in 2008 (IRR per year 0.89 [0.87, 0.91]). This was the largest degree of change among all diagnostic categories.

Cardiovascular, psychiatric, and gastrointestinal/liver, were, respectively, the next most frequent diagnostic categories and had average rates across the study interval between 2 and 3/ 100 PY. In unadjusted analyses, cardiovascular hospitalizations became more frequent over time (IRR per year 1.07 [1.03, 1.11]) while gastrointestinal/liver and psychiatric admissions had declines of borderline significance (Figure 1).

Renal; non-AIDS defining cancer; pulmonary; and endocrine, nutritional, metabolic, immune (“endocrine/metabolic”) were next in frequency and had average rates across the study interval between 1 and 2 / 100 PY. In unadjusted analyses, renal admissions and pulmonary admissions increased in frequency while the other categories did not show significant changes.

Diagnostic categories with average rates less than 1 / 100 PY included injury/poisoning, symptomatic, hematologic, neurologic, orthopedic, obstetric/gynecologic, dermatologic, unclassified, congenital, perinatal, and missing (comprising 1% of all admissions).

For frequent diagnostic categories, Table 3 shows multivariate analyses of hospitalization rates. As in univariate analysis, ADI hospitalizations were associated with a declining time trend; gastrointestinal/liver with a borderline declining time trend; and cardiovascular, renal, and pulmonary hospitalizations with increasing time trends. In multivariate analysis, all-cause, non-AIDS defining infections, and psychiatric were no longer associated with declines over time. Categories with stable time trends in univariate analyses (non-AIDS defining cancer, endocrine/metabolic, and injury/poisoning) continued to show stability in multivariate analyses.

Table 3
Multivariate analysis of factors associated with hospitalization rate by diagnostic category

In multivariate analyses, several patient characteristics were associated with higher hospitalization rates among the most frequent diagnostic categories (Table 3). Age ≥ 50 years was associated with a higher hospitalization rate than age ≤ 35 in all categories except ADI. The association for age ≥ 50 was particularly strong for cardiovascular (IRR vs. ≤ 35 years old, 5.01 [3.41, 7.38]). Among subjects ≥ 50, cardiovascular surpassed ADI to become the second most frequent category and accounted for 13% of all admissions. Blacks, women, and IDUs tended to have higher hospitalization rates across most diagnostic categories. Black race was strongly associated with renal admissions (IRR vs. White race, 3.45 [2.63, 4.52]). Lower CD4 count strata were associated with hospitalization within every frequent diagnostic category.

For cardiovascular hospitalizations, we evaluated whether codifying age as a categorical variable might have failed to capture the full effect of this potential strong confounder of the relationship with calendar time. In alternate models, we codified age as both a linear and a quadratic variable. In both cases, the multivariate calendar time trends (IRR 1.05 per year [1.01, 1.09]) were similar to the trend produced by the categorical age variable (1.06 [1.02, 1.10]).

Because of the known association of IDU with viral hepatitis, the rate of GI hospitalizations was investigated by IDU and non-IDU subgroups. Among 2775 IDU, the univariate (IRR per year 0.94 [0.88, 1.00]) and multivariate (0.93 [0.88, 0.99]) time trends were declining. Among 8870 non-IDU, the univariate (0.98 [0.93, 1.03]) and multivariate (0.99 [0.94, 1.04]) time trends were not significantly changing. There was no statistically significant interaction between IDU and calendar time (P > 0.20 for each interaction term in a model containing the interactions between IDU and year codified categorically 2001–2008).

The yearly mean LOS did not change significantly over time within diagnostic categories (either in univariate or multivariate analyses, results not shown). For the most frequent diagnostic categories, Table 4 shows the mean LOS, calculated across all years. The result for ADI was 10.5 days, significantly longer than for any other category. The result for cardiovascular was 5.9 days, which ranked near the lowest. The mean time-aggregated LOS across all categories was 7.2 days.

Table 4
Length of Stay (LOS) by Diagnostic Category

DISCUSSION

Our study has several important findings. The ADI hospitalization rate declined sharply, with the 2008 rate being less than half of the 2001 rate and the category decreasing from 18% of all admissions in 2001 to 8% of all admissions in 2008. Non-AIDS defining infection was the most frequent category across all years and was not significantly changing over time in multivariate analysis. Cardiovascular, renal, and pulmonary hospitalization rates increased significantly. Other relatively frequent categories; psychiatric, gastrointestinal/liver, non-AIDS defining cancer, and endocrine/metabolic; did not have clear trends. Finally, LOS differed substantially across diagnostic categories and was longest for ADI.

The strong decline in ADIs represents continued progress. The trend coincided with substantial immunologic and virologic improvements in the cohort. Shorter duration with low CD4 count because of more timely and/or more tolerable ART may have been an important unmeasured contributor. This decline resembles the decline apparent for ADIs during 2000–2005 in the HOPS and contrasts with the relative stability of ADIs in the NHS 1999–2007.4, 5 In the latter study, the average annual CD4 count was higher (554 cells / mm3) and stable across time. From a public health perspective, most ADIs are theoretically preventable through early HIV case finding, engagement in care, and use of ART. The long LOS associated with ADI admissions likely increases the cost of ADI admissions compared with other categories. Despite the decline, as of 2008, ADIs remain relatively frequent, and persistent efforts to further reduce incidence are needed.

The frequency and relative stability of non-AIDS defining infection admissions indicates that despite the decline in ADIs, management of infections remains a large component of HIV healthcare. Our finding of a stable multivariate time trend was distinct from the moderate declines seen in the HOPS and NHS. Non-AIDS defining infections comprised a relatively larger proportion of hospitalizations than in the HOPS, the NHS, or in our prior HIVRN study. These three studies classified many infections with organ system categories (e.g. pneumonia classified with pulmonary). In the present study, we made non-AIDS defining infection the more inclusive category because antimicrobials are the primary therapy and because many cases require infectious diseases specialist involvement.

The frequency and the increasing time trends in cardiovascular, renal, and pulmonary hospitalizations are notable findings but ultimately require confirmation in other studies. Most recent studies among PLWH have not found similar increases. In one instance of similarity, cardiovascular hospitalizations among PLWH in Denmark increased from 2.5 to 4.4 / 100 PY during 1995–2007.16 In contrast, renal and pulmonary hospitalization rates did not increase in this cohort, and no statistically significant increases were seen in any of these three diagnostic categories in either the HOPS or NHS.4, 5 With approximately 5000 patients under observation annually, our analytic sample was larger than the other cohorts (1500–3000 patients annually). It is also possible that the increases we have identified in cardiovascular, renal, and pulmonary admissions may simply reflect trends in the general population rather than any effect specific to PLWH. To the contrary, reports of nationwide data through 2008 suggest decreases in hospitalizations for myocardial infarctions and heart failure and uncertainty surrounding obstructive pulmonary disease hospitalization trends.17-19 However, a slight increase in general population renal hospitalizations has been described.20

Assuming increases in one or more of these three categories do represent effects particular to PLWH, then identification of underlying causes and consideration of the implications for HIV healthcare are warranted. Although our cohort was aging over the study interval, our multivariate analyses support the conclusion that age was not the sole factor underlying the increasing rates. Cumulative exposure to certain antiretrovirals (e.g. protease inhibitors and tenofovir) and/or to uncontrolled HIV disease has previously been associated with cardiovascular and renal illnesses21-28. A limitation of our dataset is that we could not evaluate medication exposures, cumulative HIV exposure, nor smoking in the case of pulmonary disease. Presently in the U.S., there may be substantial variation in whether primary care for PLWH is provided by HIV specialists (some of whom have infectious diseases training) or by separate primary care practitioners.29, 30 The growing breadth of complications may signify need to emphasize the latter model or to reconsider what training components are needed to make an HIV specialist. Finally, increases in admissions in these end-organ disease categories may forecast increasing chronic medication costs (e.g. anti-platelet and lipid-lowering agents in the case of cardiovascular disease.)

Unlike a rise noted in the first five years after ART in several cohorts,6, 10, 31 gastrointestinal/liver admissions did not increase during 2001–2008. Unexpectedly, gastrointestinal/liver rates declined among the subgroup of IDU. We have previously speculated that with longer survival from ADIs, PLWH coinfected with Hepatitis C (HCV) would have had a persistent increase in hepatic illness over time.11 Similar to our results, no increase occurred in the HOPS, the NHS, nor Spanish, Danish and Australian studies.4, 5, 16, 31, 32 While not increasing, gastrointestinal/liver still represents a leading category of illness in all of these cohorts. Given the strong etiologic role for HCV, access to HCV medicines will be important for the foreseeable future.

The overall frequency of psychiatric admissions and their relatively long LOS highlights persistent need for access to outpatient psychiatric care and coverage of psychotropic medicines.

We did not detect an increase in hospitalizations for non-AIDS defining cancers 2001–2008. This is generally consistent with recent studies of cancer incidence which have noted increases in specific cancers (e.g. anal cancer, Hodgkin's Disease) but no clear increase in overall non-AIDS defining cancer incidence since 2001.33-37

A possible limitation of our study is the use of ICD-9 codes. This may be less accurate than physician chart review, although two validation studies using similar methodology at one site found > 95% concurrence with chart review.10, 11 Another limitation of this study is that we are not certain of data collection from hospitalizations occurring outside the HIVRN site hospitals. While each site makes attempts to capture utilization data from neighboring clinicians, this may be incomplete. At one site, an analysis of statewide insurance claims revealed that 91% of all admissions occurred at the home hospital.10 We expect that this lack of data from outside institutions generates a slight underestimation of hospitalization rates; however we do not expect that it substantially affects the relative pattern across diagnostic categories.

A final limitation is that our data come from only four sites and may not be widely generalizable. Compared to the 26,929 subjects who received care during 2001-2008 at the 8 non-selected HIVRN sites, the 11,645 subjects included in this analysis were comparable with respect to age, gender, median CD4 count and HIV RNA and number of years in care but were less likely to be Black (38% vs. 53%) or Hispanic (19% vs. 23%), and were more likely to be MSM (41% vs. 36%) or IDU (24% vs. 18%). Our multivariate analyses accounting for the effects of demographic and clinical variables may help to increase the generalizability of our findings. Nevertheless, we believe further studies will be needed to confirm if the trends we have seen are widely representative among PLWH.

In summary, we have found that a strong decline in ADI hospitalizations has been coupled with relative stability or increases in non-AIDS defining infection, cardiovascular, psychiatric, gastrointestinal/liver, renal, non-AIDS defining cancer, pulmonary, and endocrine/metabolic admissions in our large, mostly inner-city cohort. Further improvements in access to care and use of ART may help ensure a continued decline in ADIs. However, there is also an increasing need for preventing, managing, and paying for general medical conditions among PLWH.

Supplementary Material

ACKNOWLEDGEMENTS

We thank the patients and providers of the HIV Research Network.

Sponsorship: Agency for Healthcare Research and Quality (290-01-0012) and the National Institutes of Health K23AI084854, R01 AG026250, R01 DA011602, R01 AA16893, K24 DA00432

APPENDIX – HIVRN Details

Participating Sites

Alameda County Medical Center, Oakland, California (Howard Edelstein, M.D.)

Children's Hospital of Philadelphia, Philadelphia, Pennsylvania (Richard Rutstein, M.D.)

Community Health Network, Rochester, New York (Roberto Corales, D.O.)

Drexel University, Philadelphia, Pennsylvania (Sara Allen, C.R.N.P., Jeffery Jacobson MD)

Johns Hopkins University, Baltimore, Maryland (Kelly Gebo, M.D., Richard Moore, M.D)

Montefiore Medical Group, Bronx, New York (Robert Beil, M.D.)

Montefiore Medical Center, Bronx, New York (Lawrence Hanau, M.D.)

Nemechek Health Renewal, Kansas City, Missouri (Patrick Nemechek, D.O.)

Oregon Health and Science University, Portland, Oregon (P. Todd Korthuis, M.D.)

Parkland Health and Hospital System, Dallas, Texas (Laura Armas-Kolostroubis, M.D.)

St. Jude's Children's Hospital and University of Tennessee, Memphis, Tennessee (Aditya Gaur, M.D.)

St. Luke's Roosevelt Hospital Center, New York, New York (Victoria Sharp, M.D.)

Tampa General Health Care, Tampa, Florida (Charurut Somboonwit, M.D.)

University of California, San Diego, La Jolla, California (Stephen Spector, M.D.)

University of California, San Diego, California (W. Christopher Mathews, M.D.)

Wayne State University, Detroit, Michigan (Jonathan Cohn, M.D.)

Sponsoring Agencies

Agency for Healthcare Research and Quality, Rockville, Maryland (Fred Hellinger, Ph.D., John Fleishman, Ph.D., Irene Fraser, Ph.D.)

Health Resources and Services Administration, Rockville, Maryland (Alice Kroliczak, Ph.D., Robert Mills, Ph.D.)

Data Coordinating Center

Johns Hopkins University (Richard Moore, M.D., Jeanne Keruly, C.R.N.P., Kelly Gebo, M.D., Cindy Voss, M.S., Bonnie Cameron, M.S.)

Footnotes

Conference Presentation: The findings herein were presented, in part, at the 18th International AIDS Conference, July 2010, Vienna, Austria and at the 1st International Workshop on HIV and Aging, October 2010, Baltimore, USA.

Disclaimer: The views expressed in this paper are those of the authors. No official endorsement by the Department of Health and Human Services, the National Institutes of Health, or the Agency for Healthcare Research and Quality is intended or should be inferred.

Potential Conflicts of Interests: R.D.M has been a consultant for Bristol-Myers Squibb and has received research funding from Merck, Pfizer, and Gilead. K.A.G. has been a consultant and received research funding from Tibotec. Both other authors: no conflicts.

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

REFERENCES

1. Krentz HB, Dean S, Gill MJ. Longitudinal assessment (1995-2003) of hospitalizations of HIV-infected patients within a geographical population in canada. HIV Med. 2006;7(7):457–466. [PubMed]
2. Krentz HB, Gill MJ. Cost of medical care for HIV-infected patients within a regional population from 1997 to 2006. HIV Med. 2008;9(9):721–730. [PubMed]
3. Yehia BR, Fleishman JA, Hicks PL, et al. Inpatient health services utilization among HIV-infected adult patients in care 2002-2007. J Acquir Immune Defic Syndr. 2010;53(3):397–404. [PMC free article] [PubMed]
4. Buchacz K, Baker RK, Moorman AC, et al. Rates of hospitalizations and associated diagnoses in a large multisite cohort of HIV patients in the united states, 1994-2005. AIDS. 2008;22(11):1345–1354. [PubMed]
5. Crum-Cianflone NF, Grandits G, Echols S, et al. Trends and causes of hospitalizations among HIV-infected persons during the late HAART era: What is the impact of CD4 counts and HAART use? J Acquir Immune Defic Syndr. 2010;54(3):248–257. [PMC free article] [PubMed]
6. Gebo KA, Fleishman JA, Moore RD. Hospitalizations for metabolic conditions, opportunistic infections, and injection drug use among HIV patients: Trends between 1996 and 2000 in 12 states. J Acquir Immune Defic Syndr. 2005;40(5):609–616. [PubMed]
7. Betz ME, Gebo KA, Barber E, et al. Patterns of diagnoses in hospital admissions in a multistate cohort of HIV-positive adults in 2001. Med Care. 2005;43(9 Suppl):III3–14. [PubMed]
8. CDC [11/24, 2010];HIV/AIDS surveillance report, 2005. http://www.cdc.gov/hiv/surveillance/resources/reports/2005report/.
9. HIV Research Network Hospital and outpatient health services utilization among HIV-infected patients in care in 1999. J Acquir Immune Defic Syndr. 2002;30(1):21–26. [PubMed]
10. Gebo KA, Diener-West M, Moore RD. Hospitalization rates in an urban cohort after the introduction of highly active antiretroviral therapy. J Acquir Immune Defic Syndr. 2001;27(2):143–152. [PubMed]
11. Gebo KA, Diener-West M, Moore RD. Hospitalization rates differ by hepatitis C satus in an urban HIV cohort. J Acquir Immune Defic Syndr. 2003;34(2):165–173. [PubMed]
12. Elixhauser A, Steiner C, Palmer L. Clinical Classifications Software (CCS), 2008. U.S. Agency for Healthcare Research and Quality; 2008.
13. 1993 revised classification system for HIV infection and expanded surveillance case definition for AIDS among adolescents and adults. MMWR Recomm Rep. 1992;41(RR-17):1–19. [PubMed]
14. [11/28, 2010];Free only searchable 2009 ICD-9-CM. http://icd9.chrisendres.com/.
15. StataCorp Stata statistical software: Release 11. College Station, TX: 2009.
16. Engsig FN, Hansen AB, Gerstoft J, Kronborg G, Larsen CS, Obel N. Inpatient admissions and outpatient visits in persons with and without HIV infection in denmark, 1995-2007. AIDS. 2010;24(3):457–461. [PubMed]
17. Centers for Disease Control and Prevention (CDC) QuickStats: Rate of hospitalizations for coronary atherosclerosis and acute myocardial infarction (MI), by year - national hospital discharge survey, united states, 1996 - 2005. MMWR Morb Mortal Wkly Rep. 2007;56(26):659.
18. Chen J, Normand SL, Wang Y, Krumholz HM. National and regional trends in heart failure hospitalization and mortality rates for medicare beneficiaries, 1998-2008. JAMA. 2011;306(15):1669–1678. [PubMed]
19. Stein BD, Charbeneau JT, Lee TA, et al. Hospitalizations for acute exacerbations of chronic obstructive pulmonary disease: How you count matters. COPD. 2010;7(3):164–171. [PubMed]
20. Centers for Disease Control and Prevention (CDC) Hospitalization discharge diagnoses for kidney disease - united states, 1980-2005. MMWR Morb Mortal Wkly Rep. 2008;57(12):309. [PubMed]
21. Henry K, Melroe H, Huebsch J, et al. Severe premature coronary artery disease with protease inhibitors. Lancet. 1998;351(9112):1328. [PubMed]
22. Lang S, Mary-Krause M, Cotte L, et al. Impact of individual antiretroviral drugs on the risk of myocardial infarction in human immunodeficiency virus-infected patients: A case-control study nested within the french hospital database on HIV ANRS cohort CO4. Arch Intern Med. 2010;170(14):1228–1238. [PubMed]
23. Mary-Krause M, Cotte L, Simon A, Partisani M, Costagliola D, Clinical Epidemiology Group from the French Hospital Database Increased risk of myocardial infarction with duration of protease inhibitor therapy in HIV-infected men. AIDS. 2003;17(17):2479–2486. [PubMed]
24. D:A:D Study Group. Sabin CA, Worm SW, et al. Use of nucleoside reverse transcriptase inhibitors and risk of myocardial infarction in HIV-infected patients enrolled in the D:A:D study: A multi-cohort collaboration. Lancet. 2008;371(9622):1417–1426. [PMC free article] [PubMed]
25. Lichtenstein KA, Armon C, Buchacz K, et al. Low CD4+ T cell count is a risk factor for cardiovascular disease events in the HIV outpatient study. Clin Infect Dis. 2010;51(4):435–447. [PubMed]
26. Gallant JE, Parish MA, Keruly JC, Moore RD. Changes in renal function associated with tenofovir disoproxil fumarate treatment, compared with nucleoside reverse-transcriptase inhibitor treatment. Clin Infect Dis. 2005;40(8):1194–1198. [PubMed]
27. Mocroft A, Kirk O, Gatell J, et al. Chronic renal failure among HIV-1-infected patients. AIDS. 2007;21(9):1119–1127. [PubMed]
28. Mocroft A, Kirk O, Reiss P, et al. Estimated glomerular filtration rate, chronic kidney disease and antiretroviral drug use in HIV-positive patients. AIDS. 2010;24(11):1667–1678. [PubMed]
29. 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 [PMC free article] [PubMed]
30. Wilson IB, Landon BE, Hirschhorn LR, et al. Quality of HIV care provided by nurse practitioners, physician assistants, and physicians. Ann Intern Med. 2005;143(10):729–736. [PubMed]
31. Martin-Carbonero L, Sanchez-Somolinos M, Garcia-Samaniego J, et al. Reduction in liver-related hospital admissions and deaths in HIV-infected patients since the year 2002. J Viral Hepat. 2006;13(12):851–857. [PubMed]
32. Falster K, Wand H, Donovan B, et al. Hospitalizations in a cohort of HIV patients in australia, 1999-2007. AIDS. 2010;24(9):1329–1339. [PMC free article] [PubMed]
33. Crum-Cianflone NF, Hullsiek KH, Marconi V, et al. Trends in the incidence of cancers among HIV-infected persons and the impact of antiretroviral therapy: Authors’ reply. AIDS. 2009;23(13):1791–1792. [PubMed]
34. Long JL, Engels EA, Moore RD, Gebo KA. Incidence and outcomes of malignancy in the HAART era in an urban cohort of HIV-infected individuals. AIDS. 2008;22(4):489–496. [PMC free article] [PubMed]
35. Powles T, Robinson D, Stebbing J, et al. Highly active antiretroviral therapy and the incidence of non-AIDS-defining cancers in people with HIV infection. J Clin Oncol. 2009;27(6):884–890. [PubMed]
36. Seaberg EC, Wiley D, Martinez-Maza O, et al. Cancer incidence in the multicenter AIDS cohort study before and during the HAART era: 1984 to 2007. Cancer. 2010;116(23):5507–5516. [PMC free article] [PubMed]
37. Simard EP, Pfeiffer RM, Engels EA. Cumulative incidence of cancer among individuals with acquired immunodeficiency syndrome in the united states. Cancer. 2011;117(5):1089–1096. [PMC free article] [PubMed]