Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2025 Jan 1.
Published in final edited form as: AIDS. 2023 Sep 29;38(1):85–94. doi: 10.1097/QAD.0000000000003740

Lower endoscopy, early-onset and average-onset colon cancer among Medicaid Beneficiaries with and without HIV: A cohort study

Corinne E Joshu 1,2,3, Keri L Calkins 1,4, Jacqueline E Rudolph 1, Xiaoqiang Xu 5, Eryka Wentz 1, Sally B Coburn 1, Maneet Kaur 1, Filip Pirsl 1, Richard D Moore 1,5, Bryan Lau 1,2,3,5
PMCID: PMC10841159  NIHMSID: NIHMS1933684  PMID: 37788111

Abstract

Background:

Studies suggest a lower CRC risk and lower or similar CRC screening among PLWH compared to the general population. We evaluated the incidence of lower endoscopy and average-onset (diagnosed at ≥50) and early-onset (diagnosed at <50) colon cancer by HIV status among Medicaid beneficiares with comparable sociodemographic factors and access to care.

Methods:

We obtained Medicaid Analytic eXtract (MAX) data from 2001-2015 for 14 states. We included 41,727,243 and 42,062,552 unique individuals with ≥7 months of continuous eligibility for the endoscopy and colon cancer analysis, respectively. HIV and colon cancer diagnoses and endoscopy procedures were identified from inpatient and other non-drug claims. We used Cox proportional hazards regression models to assess endoscopy and colon cancer incidence, controlling for age, sex, race/ethnicity, calendar year and state of enrollment, and comorbidities conditions.

Results:

Endoscopy and colon cancer incidence increased with age in both groups. Compared to beneficiaries without HIV, PLWH had an increased hazard of endoscopy; this association was strongest among those 18-39 years (HR:1.85 95%CI:1.77-1.92) and attenuated with age. PLWH 18-39 years also had increased hazard of early-onset colon cancer (HR:1.66, 95%CI:1.05-2.62); this association was attenuated after comorbidity adjustment. HRs were null among all beneficiaries <50 years. PLWH had a lower hazard of average-onset colon cancer compared to those without HIV (HR: 0.79 95%CI:0.66-0.94).

Conclusions:

PLWH had a higher hazard of endoscopy, particularly at younger ages. PLWH had a lower hazard of average-onset colon cancer. Early-onset colon cancer was higher among the youngest PLWH but not associated with HIV overall.

Keywords: Colonoscopy, Sigmoidoscopy, Lower Endoscopy, Colon Cancer, HIV, Medicaid

Introduction

People living with HIV (PLWH) have experienced an increase in the cumulative incidence of colorectal cancer (CRC),[1] which is particularly concerning because PLWH reportedly have higher CRC-specific mortality as compared to those without HIV.[2, 3] In the general population, CRC incidence has decreased among those 50 or older (average-onset), when incidence is the highest, and increased among those under 50 (early-onset).[4] Relative to the general population, most, but not all,[5] studies have reported either lower risk of CRC among PLWH or no difference.[6-10] One study reported lower colon cancer incidence among PLWH both above and below 50 years of age.[9] These findings are somewhat unexpected as PLWH in the US have a high prevalence of risk factors, like smoking and poverty, that are associated with reduced healthcare access and increased CRC risk.[11-15] Whether the observed lower incidence is a consequence of biologic mechanisms that differ among PLWH or non-biologic processes, including differences in sociodemographic and healthcare factors, remains unresolved,[9] and merits further investigation.

Some studies have evaluated HIV and CRC risk among those with comparable healthcare access and sociodemographic factors. In a privately insured population of predominantly male beneficiaries ≥18 years, there was no difference in CRC risk by HIV status overall, though risk increased among PLWH with decreasing values of recent CD4 counts.[16] In a Medicaid population of beneficiaries ≥40 years from 5 states in a single year, 2006, PLWH had a higher CRC risk compared to a control group matched on sex, 5-year age group and state, but this association was attenuated after adjustment for co-morbidities.[17] These studies did not report incidence by sex or race/ethnicity, although cancer incidence varies substantially by these factors in the general population.[4] They also did not report associations for average-onset and early-onset CRC separately, which may have different risk profiles.[18]

Population-wide CRC screening, historically recommended to begin at age 50 for average risk adults,[19] has been estimated to account for more than half of the substantial decline in CRC mortality experienced in the general population over the past several decades.[20] This is in part because colonoscopy can be used for both early detection and primary prevention, as polyps with malignant potential are removed.[21] Recently, rising CRC incidence and mortality among younger adults has led to updated screening guidelines that reduce the start age to 45.[21] Prior studies of CRC screening have been mixed with respect to whether PLWH receive screening less frequently or at similar rates as those without HIV.[17, 22-25] The reported lower or comparable CRC screening rates do not account for the observed deficit of CRC cases among PLWH. Investigators have suggested that PLWH may be more likely to experience gastrointestinal symptoms that lead to endoscopic investigation.[26, 27] Frequent lower endoscopy among PLWH, particularly among those below 50, could have a substantial impact on future CRC risk. Thus, exploration of lower endoscopy for both screening and diagnostic purposes can inform our understanding of both cancer prevention among PLWH and potentially CRC incidence rates.

Approximately 40% of PLWH in the US are covered by Medicaid,[28] a joint federal and state program that provides coverage to individuals that meet certain eligibility criteria including to individuals who are low-income.[29] The goal of this study was to expand on prior work by evaluating the incidence of lower endoscopy and colon cancer, both average-onset and early-onset, among more than 181,000 Medicaid beneficiaries with HIV compared to more than 42 million beneficiaries without HIV who were ≥18 and enrolled in 14 states between 2001 and 2015.

Methods

Study Population

We obtained Medicaid Analytic eXtract (MAX) data from Centers for Medicare and Medicaid Services (CMS) from 2001-2015 for 14 states. MAX data was available through 2015 for California, Georgia, New York, and Pennsylvania; through 2014 for Massachusetts, Ohio, Texas, and Washington; and through 2013 for Alabama, Colorado, Florida, Illinois, Maryland, and North Carolina. Information on enrollment, inpatient care, long term therapy, and other therapy, including outpatient services, were available for the analysis. We restricted the data to individuals aged 18-64 who were enrolled in Medicaid during the study period and not dually covered by Medicare or private insurance. Individuals were considered eligible in a given month if they had ≥15 days of Medicaid coverage and eligible for study if they had ≥7 months of continuous enrollment, which allowed for 6 month ‘washout period’ for identification of covariates and ≥1 month of follow-up. The lower endoscopy analysis, hereafter endoscopy, (N=42,062,552) excluded beneficiaries with a colon cancer diagnosis (diagnosis codes 153.X) during the washout period to reduce procedures for colon cancer surveillance. The colon cancer analysis (N=41,727,243) excluded beneficiaries with evidence of an existing cancer diagnosis (diagnosis codes 140.X-239.X) to reduce prevalent cancers[17]. The Johns Hopkins Bloomberg School of Public Health Institutional Review Board determined that this secondary analysis of existing Medicaid claims data meets the criteria for exemption.

HIV infection and Key Covariates

We identified beneficiaries with HIV using a modified version of the CMS chronic condition definition of ≥1 inpatient claim or ≥2 long term or other care claims (i.e. non-drug claims) within a 1, instead of 2, year period.[30] Claims were identified using the following ICD-9 diagnosis codes: 042, 042.X, 079.53, 795.71, V08. Date of first HIV claim was defined as the service start date for the first claim. Individuals were classified as HIV positive if they ever met the claims-based algorithm during the washout period.

We obtained age, sex, state of residence, and race/ethnicity from the MAX personal summary file. Sex was categorized as male or female, and race/ethnicity was categorized as white, non-Hispanic Black, Hispanic/Latino, or other race and missing race, herafter other. We identified any diagnosis code for a Charlson index comorbidity, excluding HIV, that occurred during washout period[31], including myocardial infarction, congestive heart failure, peripheral vascular disease, cerebrovascular disease, dementia, chronic pulmonary disease, rheumatic disease, peptic ulcer disease, mild and moderate or severe liver disease, diabetes with and without chronic complication, paralysis, renal disease, and, for endoscopy only, non-colon malignancy and metastatic solid tumor.

Lower Endoscopy

We identified colonoscopy or sigmoidoscopy using the following ICD-9-PCS codes 45.23, 45.24, 45.25, 45.42, 45.43, 76.51, and the following CPT/HPCPS codes 45330-45350, 45355, 45378-45393, G0101, G0105, G0121 and without procedure code modifiers 52, 53, 73, 74.[32] Endoscopy date was defined as the service start date. The sensitivity and specificity of Medicare claims for measuring sigmoidoscopy and colonoscopy exceeds 90%.[33] CRC screening began at age 50 during this time period,[34, 35] thus we assumed the majority of endoscopies below age 50 were for diagnostic purposes. Because Medicaid coverage for CRC screening varied by state and over time during the study period,[36] we could not distinguish between diagnostic vs screening endoscopy among beneficiaries ≥50 years, though we classified symptoms previously identified as being associated with diagnostic endoscopy,[37] and barium enema and anoscopy, which may precede endoscopy (Table S2). Beneficiaries with and without HIV had a high prevalence of symptoms (<50 years no HIV: 84.1%, HIV: 86.7%; ≥50 years no HIV: 65.9% HIV: 56.3%); the prevalence of some indicators were higher among PLWH including anemia, weight loss and anoscopy. .

Colon Cancer

We defined colon cancer as having ≥1 inpatient claim or ≥2 long term or other care claims (i.e. non-drug claims) within a 1 year period.[30] Claims were identified using ICD-9 diagnosis codes 153.X. Because prior evidence suggests that PLWH have a higher risk squamous cell carcinoma of the rectum that is frequently misclassified as anal cancer,[38] we restricted to colon cancer only. Colon cancer date was defined as the service start date for the first claim. To further exclude misclassified anal cancers, beneficiaries with colon cancer who also had an anal cancer claim within 3 months of diagnosis were censored at diagnosis as a non-case. We also evaluated early-onset colon cancer, diagnosed at <50,[18] and average-onset colon cancer, diagnosed at ≥50.

Analysis

Because the distribution of age differs among people with and without HIV and prior work has illustrated the importance of careful age adjustment in analyses of HIV and cancer,[39] all analyses were run using age as the time scale. Beneficiaries began to contribute time at risk from the first day after their first 6 months of enrollment until they experienced the outcome of interest, died, experienced a lapse in coverage, were diagnosed with any cancer (colon cancer analysis only) or CRC (endoscopy analysis only), or reached the end of follow-up (9/30/15 for CA, GA, NY, and PA, 12/31/14 for MA, OH, TX, and WA, and 12/31/13 for AL, CO, FL, IL, MD, and NC). For analyses stratified by age (18-39, 40-49, 40-59, 60-64; <50 or ≥50 years) beneficiaries began to contribute time at risk from the first day after their first 6 months of enrollment in which they were within the age range of the specific category until they met the censoring criteria above or aged out of the category. If a beneficiary experienced the outcome of interest at an earlier age than the minimum age of the category, they were excluded from the analysis.

We calculated the age-stratified incidence rates of endoscopy and colon cancer and 95% confidence intervals (CI) among beneficiaries with and without HIV as the number of events per 1,000 (endoscopy) or 100,000 (colon cancer) person years at risk using Poisson regression. We used the Kaplan-Meier estimator to estimate time-to-outcome survival curves. We used Cox proportional hazards regression to estimate the relative hazard (HR) and 95% CI, comparing each outcome of interest by HIV status. Proportional hazards assumptions were assessed graphically. For colon cancer incidence, there was evidence of non-proportionality resulting in quantitative but not qualitative (i.e., curves crossing) violation of the assumption for analyses including all beneficiaries (18-64). Therefore, these HRs should be interpreted as time averaged. All models were adjusted for sex, race/ethnicity, calendar year and state of enrollment. In a second model, we additionally adjusted for comorbidity indicators as the presence of comorbid conditions are associated with both receipt of endoscopy and CRC risk. Models were stratified by age, sex, and race/ethnicity. In supplementary analyses, we (1) excluded beneficiaries with restricted benefits, which may preclude them from receiving endoscopy or cancer care, (2) extended the washout period to 12 months, and (3) censored individuals at receipt of anoscopy (endoscopy analysis only) to exclude the possibility of misclassified endoscopies; all point estimates were within 10% of the reported point estimates and there were no qualitative changes in findings.

Results

Compared to beneficiaries without HIV, a higher proportion of beneficiaries living with HIV (PLWH) were male, older, and Black; this patten was similar for both the lower endoscopy and colon cancer analytic populations (Table 1, Table S1).

Table 1.

Characteristics of Medicaid beneficiaries included in the colon cancer analysis from 14 states, 2001-2015,1 by HIV status.

Beneficiaries with HIV Beneficiaries without HIV
N 170,804 41,556,439
Follow-up time (years) 2
Mean (SD) 3.1 (3.7) 1.5 (2.3)
Female 36.1% 63.9%
Age at enrollment
18-29 14.8% 55.2%
30-39 26.9% 19.6%
40-49 33.6% 12.5%
50-59 17.1% 8.3%
60-65 7.6% 4.4%
Race
White 24.2% 33.0%
Black 48.3% 17.5%
Hispanic 10.4% 32.7%
Other/Missing 17.1% 16.7%
State
Alabama 1.0% 1.5%
California 17.5% 47.3%
Colorado 0.5% 1.0%
Florida 11.4% 4.4%
Georgia 4.4% 2.5%
Illinois 4.6% 4.2%
Maryland 4.1% 1.9%
Massachusetts 3.7% 3.1%
New York 37.4% 15.2%
North Carolina 3.9% 2.4%
Ohio 2.5% 4.1%
Pennsylvania 2.7% 4.3%
Texas 4.7% 4.9%
Washington 1.7% 3.1%
Enrollment Year
2001 36.1% 18.2%
2002 5.6% 6.7%
2003 4.7% 6.1%
2004 4.2% 5.8%
2005 3.9% 5.5%
2006 3.4% 5.1%
2007 3.0% 4.0%
2008 4.2% 5.7%
2009 4.7% 6.2%
2010 4.8% 6.3%
2011 4.6% 6.2%
2012 4.1% 6.0%
2013 3.8% 5.6%
2014 11.4% 9.7%
2015 1.5% 1.9%
Charlson Comorbidities
0 66.7% 91.0%
1 23.0% 6.6%
2+ 10.3% 2.4%
Open in a new tab
1

MAX data was available from 2001 through 2015 for California, Georgia, New York, and Pennsylvania; through 2014 for Massachusetts, Ohio, Texas, and Washington; and through 2013 for Alabama, Colorado, Florida, Illinois, Maryland, and North Carolina.

2

Follow-up time is measured after the initial 6-month washout period.

Lower Endoscopy

We identified 1,107,641 endoscopies in 62,659,999 person years among beneficiaries without HIV, and 19,356 endoscopies in 513,294 person years among PLWH. Endoscopy incidence rates increased substantially by age among both PLWH and beneficiaries without HIV (Table 2). By age 50, approximately 50% of male and 40% of female PLWH had an endoscopy compared to approximately 20% of males and females without HIV (Figure 1A,1B). In constrast, among beneficiaries ≥50, the cummalitive incidence was only modestly higher among male and female PLWH than their counterparts without HIV (Figure 1C, 1D).

Table 2.

Association between HIV and lower endoscopy and colon cancer overall and by sex and race among Medicaid beneficiaries from 14 states, 2001-2015.

Lower Endoscopy Colon Cancer
Age Group Cases IR per 1,000 PY
(95% CI)		 HR2 HR3 Cases IR per 100,000 PY
(95% CI)		 HR2 HR3
Age 18-39
No HIV 173,223 4.36 (4.34, 4.38) REF REF 1,388 3.48 (3.30, 3.67) REF REF
HIV 2,447 16.5 (15.9, 17.2) 2.32 (2.23, 2.42) 1.85 (1.77, 1.92) 19 13.0 (7.8, 20.3) 1.66 (1.05, 2.62) 1.46 (0.92, 2.32)
Age 40-49
No HIV 199,963 18.1 (18.0, 18.2) REF REF 2,501 22.2 (21.4, 23.1) REF REF
HIV 5,357 26.1 (25.5, 26.9) 1.37 (1.34, 1.41) 1.22 (1.18, 1.25) 45 22.3 (16.3, 29.8) 0.83 (0.61, 1.11) 0.79 (0.58, 1.06)
Age 50-59
No HIV 498,242 60.5 (60.4, 60.7) REF REF 6,580 72.0 (70.3, 73.8) REF REF
HIV 9,227 72.4 (71.0, 73.9) 1.25 (1.22, 1.28) 1.18 (1.15, 1.20) 82 57.4 (45.6, 71.2) 0.75 (0.60, 0.93) 0.76 (0.61, 0.94)
Age 60-64
No HIV 235,487 67.2 (66.9, 67.5) REF REF 4,645 122.6 (119.1, 126.2) REF REF
HIV 2,319 78.6 (75.5, 81.9) 1.12 (1.08, 1.17) 1.04 (1.00, 1.08) 41 127.3 (91.3, 172.7) 0.92 (0.68, 1.26) 0.90 (0.66, 1.22)
Open in a new tab
1

Unadjusted incidence rate (IR) per 100,000 person-years.

2

Cox proportional hazards model used to estimate the hazard ratio (HR) and 95% confidence interval (CI); models adjusted for or stratified by age, sex, race, state, and enrollment year.

3

Cox proportional hazards model used to estimate the hazard ratio (HR) and 95% confidence interval (CI); models adjusted for or stratified by age, sex, race, state, enrollment year, and individual Charlson co-morbidities.

4

Counts ≤10 are not reported (NR).

Figure 1.

Figure 1.

Figure 1.

Open in a new tab

Cumulative probability of being free of an endoscopy by age and HIV status among Medicaid beneficiaries from 14 states, 2001-2015. Beneficiaries with HIV in dashed line; beneficiaries without HIV in solid line. Endoscopy at <50 years shown for males in panel A and for females in panel B. Endoscopy at ≥50 years shown for males in panel C and for females in panel D.

Across all ages, the hazard of lower endoscopy among PLWH was significantly higher than among beneficiaries without HIV (HR:1.35, 95%CI:1.33, 1.37), including after adjustment for comordibities (HR:1.25, 95%CI:1.23, 1.27). However, this association was strongest among beneficiaries 18-39, including after adjustement for comorbidities, and decreased substantially with age, such that the HR was close to the null for those 60-64 after adjustment for comordibities (Table 2). Collapsing across age categories <50, the association between HIV and endocopy was strongest among white and Black males, but null among white females (Table 3). Among beneficiaries ≥50, the association between HIV and endoscopy was strongest among white males and again null among white females (Table 3).

Table 3.

Association between HIV and lower endoscopy before and after age 50, and early-onset colon cancer (diagnosed before age 50) and average-onset colon cancer (diagnosed after age 50), overall and by sex and race among Medicaid beneficiaries from 14 states, 2001-2015.

Endoscopies Colon Cancer
Age 18-49 years Cases HR 1 HR 2 Cases HR 1 HR 2
Overall
No HIV 373,186 REF REF 3,889 REF REF
HIV 7,804 1.57 (1.54, 1.61) 1.38 (1.35, 1.42) 64 0.95 (0.74, 1.22) 0.90 (0.70, 1.15)
Male
No HIV 125,559 REF REF 1,548 REF REF
HIV 4,668 1.86 (1.81, 1.92) 1.66 (1.62, 1.72) 33 0.87 (0.62, 1.23) 0.83 (0.59, 1.18)
Female
No HIV 247,627 REF REF 2,341 REF REF
HIV 3,136 1.31 (1.26, 1.36) 1.13 (1.09, 1.17) 31 1.08 (0.75, 1.54) 1.00 (0.70, 1.43)
White Male
No HIV 62,154 REF REF 654 REF REF
HIV 1,389 1.81 (1.71, 1.91) 1.66 (1.57, 1.75) NR4 1.01 (0.54, 1.89) 1.02 (0.55, 1.92)
Black Male
No HIV 22,689 REF REF 388 REF REF
HIV 1,966 1.97 (1.88, 2.06) 1.81 (1.72, 1.90) 12 0.69 (0.36, 1.14) 0.61 (0.34, 1.09)
White Female
No HIV 123,895 REF REF 909 REF REF
HIV 623 1.21 (1.12, 1.31) 1.02 (0.94, 1.10) NR4 0.91 (0.34, 2.44) 0.83 (0.31, 2.23)
Black Female
No HIV 53,596 REF REF 693 REF REF
HIV 1,777 1.42 (1.36, 1.49) 1.29 (1.22, 1.35) 17 0.95 (0.58, 1.54) 0.91 (0.56, 1.49)
Age 50-64 years Overall
No HIV 734,454 REF REF 11,225 REF REF
HIV 11,552 1.21 (1.20, 1.24) 1.15 (1.13, 1.17) 123 0.79 (0.66, 0.95) 0.79 (0.66, 0.94)
Male
No HIV 285,001 REF REF 5,195 REF REF
HIV 7,120 1.28 (1.25, 1.31) 1.20 (1.17, 1.23) 76 0.75 (0.59, 0.94) 0.74 (0.59, 0.93)
Female
No HIV 449,453 REF REF 6,030 REF REF
HIV 4,432 1.14 (1.11, 1.18) 1.08 (1.05, 1.11) 47 0.91 (0.68, 1.21) 0.89 (0.67, 1.19)
White Male
No HIV 119,313 REF REF 2,249 REF REF
HIV 1,927 1.44 (1.38, 1.50) 1.38 (1.32, 1.44) 15 0.62 (0.37, 1.04) 0.63 (0.38, 1.05)
Black Male
No HIV 57,667 REF REF 1,326 REF REF
HIV 3,136 1.21 (1.16, 1.25) 1.15 (1.11, 1.20) 41 0.76 (0.55, 1.03) 0.75 (0.55, 1.02)
White Female
No HIV 176,530 REF REF 2,318 REF REF
HIV 695 1.01 (0.98, 1.13) 0.98 (0.91, 1.06) NR4 1.04 (0.52, 2.09) 1.02 (0.51, 2.04)
Black Female
No HIV 91,393 REF REF 1,681 REF REF
HIV 2,310 1.14 (1.09, 1.19) 1.10 (1.06, 1.15) 23 0.75 (0.49, 1.13) 0.76 (0.50, 1.14)
Open in a new tab
1

Cox proportional hazards model used to estimate the hazard ratio (HR) and 95% confidence interval (CI); all models adjusted for or stratified by age, sex, and race.

2

Cox proportional hazards model used to estimate the hazard ratio (HR) and 95% confidence interval (CI); all models adjusted for or stratified by age, sex, race, state, enrollment year, and individual Charlson co-morbidities.

Colon Cancer

We identified 15,114 colon cancer cases in 64,063,192 person years among beneficiaries without HIV, and 187 colon cancer cases in 523,122 person years among PLWH. Colon cancer incidence rates increased substantially by age among both PLWH and beneficiaries without HIV (Table 2). The cumulative incidence of early-onset colon cancer was higher among male and female PLWH than their counterparts without HIV (Figure 2A, 2B). In contrast, the cumulative incidence for average-onset colon cancer was higher among male beneficiaries without HIV compared their counterparts with HIV (Figure 1C), but there was no notable difference by HIV status among female beneficiaries (Figure 2D).

Figure 2.

Figure 2.

Figure 2.

Open in a new tab

Cumulative probability of being free of colon cancer by age and HIV status among Medicaid beneficiaries from 14 states, 2001-2015. Beneficiaries with HIV in dashed line; beneficiaries without HIV in solid line. Colon cancer at <50 years shown for males in panel A and for females in panel B. Colon cancer at ≥50 years shown for males in panel C and for females in panel D.

Across all ages, the hazard of colon cancer among PLWH significantly lower than among beneficiaries without HIV (HR:0.85, 95%CI:0.73, 0.98), including after adjustment for comordibities (HR:0.83, 95%CI:0.72, 0.96). However, among beneficiaries 18-39 years old, PLWH had a higher hazard of colon cancer compared to beneficiaries without HIV (HR: 1.66, 95% CI:1.05, 2.62, Table 2), though this association was attenuated after adjustment for comorbidities (Table 2). Among beneficiaries 40-49 years old, there was no significant difference in colon cancer by HIV status. In contrast, among beneficiaries 50-59 years old, PLWH had a lower hazard of colon cancer compared to beneficiaries without HIV (HR: 0.76, 95% CI:0.61, 0.94, Table 2). Among beneficiaries 60-64 years old, there was no significant difference in colon cancer HIV status (Table 2).

Collapsing across age categories <50, there was no significant association between HIV and early-onset colon cancer among any sex-race category (Table 3). Among beneficiaries ≥50, The hazard of average-onset colon cancer among PLWH was 0.79 (95% CI: 0.66, 0.94) times the hazard among those without HIV, after adjustment for demographic and comorbidities (Table 3). When stratified by sex, this association remained statistically significant among male, but not female, beneficiaries. In all sex-race categories except white females, hazard ratios were below the null but not statistically significant (Table 3).

Discussion

We evaluated lower endoscopy and colon cancer incidence among Medicaid beneficiaries; for both outcomes, we observed large differences by HIV status that diminished with increasing age. Incidence of endoscopy and colon cancer increased with age among beneficiaries with and without HIV, particularly over the age of 50. Among beneficiaries 50 or older, the age at which CRC screening began for average risk adults during this time period, PLWH were more likely to receive lower endoscopy than beneficiaries without HIV, but the associations were attenuated compared to younger beneficiaires. PLWH also had a lower hazard of average-onset colon cancer than beneficiaries without HIV. Among beneficiaries under 40 years, PLWH were nearly twice as likely to receive lower endoscopy than beneficiaries without HIV; they also had an increased hazard of early-onset colon cancer, though this association was attenuated after adjusting for co-morbidities. Given lower endoscopy can both detect CRC and reduce future CRC risk, the high incidence of lower endoscopy among PLWH, particularly those <50, has important implications for understanding the burden of CRC among PLWH.

We observed a slightly higher incidence of lower endoscopy among PLWH ≥50 years even after adjustment for comorbidities; this difference decreased with age and was null among older females. These findings are generally consisitent with prior studies that have reported similar or small differences in colorectal cancer screening by HIV status. [17, 22-25, 40] In a prior study of Medicaid beneficiaries ≥40 years from 5 states, colorectal cancer screening was similar among those with and without HIV after matching for age and sex; though PLWH were slightly less likely to receive screening after adjusting for co-morbidities.[17] Other studies of colorectal cancer screening have reported that PLWH in the US receive screening less frequently or at similar rates as those without HIV.[22-25] These studies were restricted to a single clinic, included stool-based tests in their analysis, and had predominantly male study participants. Our study expands upon this prior work by including male and female beneficiaries from across the US.

In our study, HIV was modestly associated with a lower hazard of colon cancer overall. When restricted to average-onset colon cancer, or colon cancer diagnosed ≥ 50 years, the association between HIV and colon cancer incidence was in the inverse direction for all sex and race/ethnicity categories, except white females. Our findings for colon cancer overall and average-onset colon cancer are consistent with prior studies reporting no difference or lower risk of colon cancer among PLWH compared to the general population. [6-10] This includes a study among Medicaid beneficiaries >40 years from 5 states in 2006 which reported a modestly increased risk of colorectal cancer among PLWH that was attenuated after adjustment for comorbidities.[17]

The reason for the observed lower risk of colon cancer among PLWH is unknown, though healthcare utilization has been a hypothesized factor. Colonoscopy is both a tool for early detection and a method of primary prevention, as polyps with malignant potential are removed.[21] We also found that approximately 50% of male and 40% of female PLWH had an endoscopy by age 50 compared to 20% among those without HIV. This profound discrepancy in rate of lower endoscopy between those with and without HIV may help explain the lower risk of colon cancer incidence observed among PLWH 50-59 years old and similarly among PLWH 40-49 years old, albeit with wider confidence intervals.

One prior study reported a lower risk of colon cancer among PLWH under 50 compared to the general population.[9] Here, we similarly saw a lower (albeit not significant) hazard of colon cancer among PLWH under 50, compared to beneficiaries without HIV. The high incidence of lower endoscopy among PLWH <50 may contribute to a lower risk of colon cancer in these age groups. In addition, early-onset colon cancer incidence has been increasing in the general population,[4] though whether early-onset incidence is changing among PLWH is unknown. In contrast to our observations for early-onset colon cancer overall, we found a higher hazard of colon cancer among PLWH under 40 years compared to their counterparts without HIV; this association was attenuated after adjustment for comorbidities and was based on 19 cases. Several factors could help explain this difference. A previous study reported a higher risk of colorectal cancer incidence among privately insured PLWH with lower CD4 counts as compared to those without HIV.[16] Prior work has noted poorer viral suppression and/or retention in care among those with public insurance and younger PLWH.[41, 42] Therefore, it is possible that PLWH (especially younger PLWH) in our study are less likely to be retained in care as compared to prior studies in which PLWH were ascertained from HIV registries. Medicaid enrollment does not necessarily correspond with the usual definition of retention in care among PLWH.[43] More studies are needed to explore the burden of colon cancer and its risk factors among PLWH.

We evaluated lower endoscopy and colon cancer incidence among more than 40 million Medicaid beneficiaries from 14 states enrolled over a 15-year period. Approximately 40% of PLWH in the US are covered by Medicaid.[28] Thus, our study population reflects a sizable proportion of PLWH in the US. In our study, PLWH were similar to those without HIV with respect to some sociodemographic characteristics and insurance status. However, access to care may still differ by state, which we adjusted for, and by unmeasured factors. The size of our population allowed us to evaluate outcomes by age, sex, and race/ethnicity, but we were unable to assess differences in viral suppression and CD4 count, which may influence receipt of endoscopy and cancer incidence. We were unable to distinguish endoscopy for diagnostic versus screening purposes, which is important to understanding patterns of cancer prevention care among PLWH. However, we found both beneficiaries with and without HIV had a high prevalence of symptoms associated with a diagnostic endoscopy in both age groups, which may reflect state level reimbursement policies for lower endoscopy, including procedures for colorectal cancer screening and for solely diagnostic purposes, during this time frame . We were unable to evaluate cancer-specific factors, including stage at diagnosis, which is important to cancer outcomes. Finally, we idenitified HIV status within the first 6 months of enrollment. This is similar to an intent to treat type of analyses and therefore, there are individuals who may have become infected with HIV during follow-up. If these individuals followed the same risk patterns as the PLWH identified , then our estimates of the association between HIV status and endoscopy and colon cancer incidence are conservative.

Our study is among the largest to evaluate lower endoscopy and colon cancer incidence among PLWH enrolled in Medicaid, and in particular among those below 50 and by sex and race/ethnicity . Incidence of lower endoscopy and colon cancer increased with age among beneficiaires with and without HIV. Compared to beneficiaries without HIV, PLWH had higher endoscopy hazard among those under 50 years, and slightly higher among those 50 or older. PLWH had a lower hazard of average onset colon cancer. While PLWH <40 had a higher hazard of early-onset colon cancer, HIV was not associated with early-onset colon cancer overall. Future studies are needed to confirm these findings in other study populations. In addition, more work is needed to understand how receipt of endoscopy at younger ages influences colorectal cancer screening and colorectal cancer incidence at older ages.

Supplementary Material

Supplemental Data File (.doc, .tif, pdf, etc.)

Acknowledgements:

This study was funded by R01 CA250851, U01 AI069918, P30 CA006973. This research was also funded in part by a 2018 developmental grant from the Johns Hopkins University Center for AIDS Research, an NIH funded program (1P30AI094189), which is supported by the following NIH Co-Funding and Participating Institutes and Centers: NIAID, NCI, NICHD, NHLBI, NIDA, NIA, NIGMS, NIDDK, NIMHD. C Joshu was supported by RSG-18-147-01. M Kaur was supported by the NIDDK Clinical Research and Epidemiology in Diabetes and Endocrinology Training Grant (T32DK062707). SB Coburn was supported by F31CA247610. F Pirsl was supported by the NCI Cancer Epidemiology, Prevention, and Control Grant (T32CA0093140).

Role of the funding source:

The funding sources had no role in the design, methods, analysis, or preparation of paper. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH.

Footnotes

Disclosures: The authors disclose no potential conflicts of interest.

Data sharing:

The data utilized for the proposed study is under the authority of Centers for Medicaid & Medicare Services (CMS) and administered by ResDAC. Investigators can “re-use” the data in this study if they (1) independently meet the CMS standards of use and receive approval for reuse, and (2) have permission from the NIH program officer who oversees this study.

References

  • 1.Silverberg MJ, Lau B, Achenbach CJ, Jing Y, Althoff KN, D'Souza G, et al. Cumulative Incidence of Cancer Among Persons With HIV in North America: A Cohort Study. Ann Intern Med 2015; 163(7):507–518. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Coghill AE, Shiels MS, Suneja G, Engels EA. Elevated Cancer-Specific Mortality Among HIV-Infected Patients in the United States. J Clin Oncol 2015; 33(21):2376–2383. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Coghill AE, Han X, Suneja G, Lin CC, Jemal A, Shiels MS. Advanced stage at diagnosis and elevated mortality among US patients with cancer infected with HIV in the National Cancer Data Base. Cancer 2019; 125(16):2868–2876. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Siegel RL, Miller KD, Goding Sauer A, Fedewa SA, Butterly LF, Anderson JC, et al. Colorectal cancer statistics, 2020. CA Cancer J Clin 2020; 70(3):145–164. [DOI] [PubMed] [Google Scholar]
  • 5.Patel P, Hanson DL, Sullivan PS, Novak RM, Moorman AC, Tong TC, et al. Incidence of types of cancer among HIV-infected persons compared with the general population in the United States, 1992-2003. Ann Intern Med 2008; 148(10):728–736. [DOI] [PubMed] [Google Scholar]
  • 6.Frisch M, Biggar RJ, Engels EA, Goedert JJ. Association of cancer with AIDS-related immunosuppression in adults. Jama 2001; 285(13):1736–1745. [DOI] [PubMed] [Google Scholar]
  • 7.van Leeuwen MT, Vajdic CM, Middleton MG, McDonald AM, Law M, Kaldor JM, et al. Continuing declines in some but not all HIV-associated cancers in Australia after widespread use of antiretroviral therapy. Aids 2009; 23(16):2183–2190. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Hessol NA, Whittemore H, Vittinghoff E, Hsu LC, Ma D, Scheer S, et al. Incidence of first and second primary cancers diagnosed among people with HIV, 1985-2013: a population-based, registry linkage study. Lancet HIV 2018; 5(11):e647–e655. [DOI] [PubMed] [Google Scholar]
  • 9.Coghill AE, Engels EA, Schymura MJ, Mahale P, Shiels MS. Risk of Breast, Prostate, and Colorectal Cancer Diagnoses Among HIV-Infected Individuals in the United States. J Natl Cancer Inst 2018; 110(9):959–966. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.O’Neill TJ, Nguemo JD, Tynan AM, Burchell AN, Antoniou T. Risk of Colorectal Cancer and Associated Mortality in HIV: A Systematic Review and Meta-Analysis. Journal of acquired immune deficiency syndromes (1999) 2017; 75(4):439–447. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.White A, Thompson TD, White MC, Sabatino SA, de Moor J, Doria-Rose PV, et al. Cancer Screening Test Use - United States, 2015. MMWR Morb Mortal Wkly Rep 2017; 66(8):201–206. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Leung CW, Ding EL, Catalano PJ, Villamor E, Rimm EB, Willett WC. Dietary intake and dietary quality of low-income adults in the Supplemental Nutrition Assistance Program. Am J Clin Nutr 2012; 96(5):977–988. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Weinberger AH, Smith PH, Funk AP, Rabin S, Shuter J. Sex Differences in Tobacco Use Among Persons Living With HIV/AIDS: A Systematic Review and Meta-Analysis. Journal of acquired immune deficiency syndromes (1999) 2017; 74(4):439–453. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Bryan L, Westmaas L, Alcaraz K, Jemal A. Cigarette smoking and cancer screening underutilization by state: BRFSS 2010. Nicotine & tobacco research : official journal of the Society for Research on Nicotine and Tobacco 2014; 16(9):1183–1189. [DOI] [PubMed] [Google Scholar]
  • 15.Liang PS, Chen TY, Giovannucci E. Cigarette smoking and colorectal cancer incidence and mortality: systematic review and meta-analysis. Int J Cancer 2009; 124(10):2406–2415. [DOI] [PubMed] [Google Scholar]
  • 16.Silverberg MJ, Chao C, Leyden WA, Xu L, Horberg MA, Klein D, et al. HIV infection, immunodeficiency, viral replication, and the risk of cancer. Cancer Epidemiol Biomarkers Prev 2011; 20(12):2551–2559. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Keller SC, Momplaisir F, Lo Re V, Newcomb C, Liu Q, Ratcliffe SJ, et al. Colorectal cancer incidence and screening in US Medicaid patients with and without HIV infection. AIDS care 2014; 26(6):716–722. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Siegel RL, Jakubowski CD, Fedewa SA, Davis A, Azad NS. Colorectal Cancer in the Young: Epidemiology, Prevention, Management. Am Soc Clin Oncol Educ Book 2020; 40:1–14. [DOI] [PubMed] [Google Scholar]
  • 19.Bibbins-Domingo K, Grossman DC, Curry SJ, Davidson KW, Epling JW Jr., García FAR, et al. Screening for Colorectal Cancer: US Preventive Services Task Force Recommendation Statement. Jama 2016; 315(23):2564–2575. [DOI] [PubMed] [Google Scholar]
  • 20.Edwards BK, Ward E, Kohler BA, Eheman C, Zauber AG, Anderson RN, et al. Annual report to the nation on the status of cancer, 1975-2006, featuring colorectal cancer trends and impact of interventions (risk factors, screening, and treatment) to reduce future rates. Cancer 2010; 116(3):544–573. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.United States Preventive Services Task Force. Final Recommendation Statement: Colorectal Cancer: Screening. In; 2021.
  • 22.Corrigan KL, Wall KC, Bartlett JA, Suneja G. Cancer disparities in people with HIV: A systematic review of screening for non-AIDS-defining malignancies. Cancer 2019; 125(6):843–853. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Iqbal S, Browne-McDonald V, Cerulli MA. Recent trends for colorectal cancer screening in HIV-infected patients. Dig Dis Sci 2010; 55(3):761–766. [DOI] [PubMed] [Google Scholar]
  • 24.Campbell J, Young B. Use of screening colonoscopy in ambulatory HIV-infected patients. J Int Assoc Physicians AIDS Care (Chic) 2008; 7(6):286–288. [DOI] [PubMed] [Google Scholar]
  • 25.Guest JL, Rentsch CT, Rimland D. Comparison of colorectal cancer screening and diagnoses in HIV-positive and HIV-negative veterans. AIDS care 2014; 26(12):1490–1493. [DOI] [PubMed] [Google Scholar]
  • 26.Reinhold JP, Moon M, Tenner CT, Poles MA, Bini EJ. Colorectal cancer screening in HIV-infected patients 50 years of age and older: missed opportunities for prevention. Am J Gastroenterol 2005; 100(8):1805–1812. [DOI] [PubMed] [Google Scholar]
  • 27.Antoniou T, Jembere N, Saskin R, Kopp A, Glazier RH. A population-based study of the extent of colorectal cancer screening in men with HIV. BMC Health Serv Res 2015; 15:51. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.The Henry J. Kaiser Family Foundation. Medicaid and HIV. In. Menlo Park, California; 2016. [Google Scholar]
  • 29.Centers for Medicare & Medicaid Services. Medicaid Eligibility. In.
  • 30.Center for Medicare and Medicaid Services. Chronic Conditions Warehouse. In; 2020.
  • 31.Quan H, Sundararajan V, Halfon P, Fong A, Burnand B, Luthi JC, et al. Coding algorithms for defining comorbidities in ICD-9-CM and ICD-10 administrative data. Med Care 2005; 43(11):1130–1139. [DOI] [PubMed] [Google Scholar]
  • 32.Bonafede MM, Miller JD, Pohlman SK, Troeger KA, Sprague BL, Herschorn SD, et al. Breast, Cervical, and Colorectal Cancer Screening: Patterns Among Women With Medicaid and Commercial Insurance. Am J Prev Med 2019; 57(3):394–402. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Schenck AP, Klabunde CN, Warren JL, Peacock S, Davis WW, Hawley ST, et al. Data sources for measuring colorectal endoscopy use among Medicare enrollees. Cancer Epidemiol Biomarkers Prev 2007; 16(10):2118–2127. [DOI] [PubMed] [Google Scholar]
  • 34.United States Preventive Services Task Force. Screening for colorectal cancer: recommendation and rationale. Ann Intern Med 2002; 137(2):129–131. [DOI] [PubMed] [Google Scholar]
  • 35.United States Preventive Services Task Force. Screening for colorectal cancer: U.S. Preventive Services Task Force recommendation statement. Ann Intern Med 2008; 149(9):627–637. [DOI] [PubMed] [Google Scholar]
  • 36.National Conferene of State Legislators. Colorectal Cancer Screening: What are States Doing? In; 2011.
  • 37.Garcia-Albeniz X, Hsu J, Bretthauer M, Hernan MA. Effectiveness of Screening Colonoscopy to Prevent Colorectal Cancer Among Medicare Beneficiaries Aged 70 to 79 Years: A Prospective Observational Study. Ann Intern Med 2017; 166(1):18–26. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Coghill AE, Shiels MS, Rycroft RK, Copeland G, Finch JL, Hakenewerth AM, et al. Rectal squamous cell carcinoma in immunosuppressed populations: is this a distinct entity from anal cancer? Aids 2016; 30(1):105–112. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Shiels MS, Pfeiffer RM, Engels EA. Age at cancer diagnosis among persons with AIDS in the United States. Ann Intern Med 2010; 153(7):452–460. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Goedert JJ, Hosgood HD, Biggar RJ, Strickler HD, Rabkin CS. Screening for Cancer in Persons Living with HIV Infection. Trends in cancer 2016; 2(8):416–428. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Ludema C, Cole SR, Eron JJ Jr., Edmonds A, Holmes GM, Anastos K, et al. Impact of Health Insurance, ADAP, and Income on HIV Viral Suppression Among US Women in the Women's Interagency HIV Study, 2006-2009. Journal of acquired immune deficiency syndromes (1999) 2016; 73(3):307–312. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Yehia BR, Stephens-Shields AJ, Fleishman JA, Berry SA, Agwu AL, Metlay JP, et al. The HIV Care Continuum: Changes over Time in Retention in Care and Viral Suppression. PloS one 2015; 10(6):e0129376. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Lesko CR, Tong W, Moore RD, Lau B. Retention, Antiretroviral Therapy Use and Viral Suppression by History of Injection Drug Use Among HIV-Infected Patients in an Urban HIV Clinical Cohort. AIDS Behav 2017; 21(4):1016–1024. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplemental Data File (.doc, .tif, pdf, etc.)
NIHMS1933684-supplement-Supplemental_Data_File___doc___tif__pdf__etc__.docx (19.6KB, docx)

Data Availability Statement

The data utilized for the proposed study is under the authority of Centers for Medicaid & Medicare Services (CMS) and administered by ResDAC. Investigators can “re-use” the data in this study if they (1) independently meet the CMS standards of use and receive approval for reuse, and (2) have permission from the NIH program officer who oversees this study.

RESOURCES