1647. Preventive Cardiovascular/Metabolic Health Screening in an HIV+ Cohort by Type of Primary Care Model
Session: Poster Abstract Session: HIV: Cardiovascular Disease in HIV
Saturday, October 10, 2015
Room: Poster Hall
Posters
  • Rhodes IDWeek.pdf (336.4 kB)
  • Background: Preventive screening needs of persons living with HIV/AIDS (PLWHA) may be addressed differently in different care models.  We examined three observed care models:  infectious disease provider only (ID), generalist provider only (G), and both ID & generalist providers (ID&G) to investigate if guideline-based metabolic health screening differed.

    Methods: We developed an adult PLWHA cohort from Partners HealthCare System Research Database Patient Registry (RPDR) using validated ICD9 HIV codes.  Visit pattern in the previous 2 years determined care model. Outcomes were screening for hypertension (HTN) by blood pressure, obesity by BMI, hyperlipidemia (HL) by lipids, and diabetes (DM) by A1C and glucose proxy.  We conducted multivariate logistic analyses for each outcome controlling for age, sex, race, viral load (VL) suppression, antiretroviral (ARV) use, recent & nadir CD4, and total 2012 clinic visits. 

    Results: The ID group was more likely to be male, white, and speak English.  G and ID&G groups had more minorities and clinic visits. Groups had no significant differences in age, recent or nadir CD4, VL suppression, or ARV use.

    Table 1: Demographics

     

    ID  n=864

    G  n=90

    ID&G n=621

    p value

    Male %

    77

    63

    63

    .01

    Race %:  White

                      Black

                      Hispanic

    56

    25

    14

    34

    24

    40

    43

    36

    16

    .01

    English language %

    93

    60

    87

    .01

    2012 Visits: Mean (SD)

    8 (5)

    10 (8)

    12 (9)

    .01

    VL <400 copies* %

    89

    80

    88

    .054

    Median CD4*

    583

    561

    576

    .86

    ARV use, %

    97

    97

    96

    .26

    *most recent 2012 value, when available

    G group patients had 60% lower odds of obesity screening but 2.9-fold increased odds of HL screening. There were no differences for HTN or DM screening by care group.

    Table 2: Outcomes

    ID

    G

    ID&G

     

    % Screened

    OR (95% CI)

    % Screened

    OR (95% CI)

    % Screened

    OR (95% CI)

    HTN

    99.5

    -

    100

    -

    99.8

    Reference

    Obesity

    88

    .8 (.5-1.3)

    87

    .4 (.2-.9)

    93

    Ref.

    HL

    28

    1.0 (.8-1.3)

    48

    2.9 (1.8-4.8)

    30

    Ref.

    DM A1C

    5

    .7 (.4-1.3)

    8

    1.3(.5-3.7)

    7

    Ref.

    DM glucose

    69

    1.2 (.8-1.7)

    66

    .9 (.5-1.7)

    70

    Ref.

    Conclusion: We demonstrated demographic composition varies by patient choice of primary care model.  After adjusting for demographics, HIV characteristics, and visits there are no significant screening differences in ID and ID&G groups. The G group had higher odds of HL screening and lower odds of obesity screening and non-significantly lower VL suppression.  Further research is needed to define optimal care model to address PLWHA’s HIV and general medicine care.

    Corinne Rhodes, MD1, Susan Regan, PhD2, Daniel Singer, MD1 and Virginia Triant, MD, MPH3, (1)Division of General Internal Medicine, Massachusetts General Hospital, Boston, MA, (2)Medical Practice Evaluation Center, Massachusetts General Hospital, Boston, MA, (3)Medicine/Infectious Diseases, Massachusetts General Hospital, Boston, MA

    Disclosures:

    C. Rhodes, None

    S. Regan, None

    D. Singer, None

    V. Triant, None

    Findings in the abstracts are embargoed until 12:01 a.m. PDT, Wednesday Oct. 7th with the exception of research findings presented at the IDWeek press conferences.