Background:

Adolescents and young adults (AYA, age 15–39 years) with cancer may be at elevated risk for late morbidity following their cancer treatment, but few studies have quantified the excess burden of severe disease in this population. Using population-based data from Utah, we examined the risk of inpatient hospitalizations among AYA cancer survivors compared with their siblings and the general population.

Methods:

Survivors of AYA cancer who were ≥2 years from diagnosis and diagnosed from 1994 to 2015 (N = 6,330), their siblings (N = 12,924), and an age- and sex-matched comparison cohort (N = 18,171) were identified using the Utah Population Database (UPDB). Hospitalizations from 1996 to 2017 were identified from statewide discharge records in the UPDB. We estimated multivariable-adjusted hazard ratios (HR) for first hospitalization and rate ratios (RR) for total hospitalizations for survivors relative to the matched comparison cohort and siblings.

Results:

Overall, the risk of a first hospitalization was higher among AYA cancer survivors than the matched population-based cohort [HR = 1.93; 95% confidence interval (CI), 1.81–2.06]. Risk was most elevated for survivors of leukemia (HR = 4.76), central nervous system tumors (HR = 3.45), colorectal cancers (HR = 2.83), non-Hodgkin lymphoma (HR = 2.76), and breast cancer (HR = 2.37). The rate of total hospitalizations was also increased among survivors relative to the comparison cohort (RR = 2.05; 95% CI, 1.95–2.14). Patterns were generally similar in analyses comparing survivors to their siblings.

Conclusions:

AYA cancer survivors have a higher burden of inpatient hospitalization than their siblings and the general population.

Impact:

Results indicate the importance of long-term, risk-based follow-up care to prevent and treat severe morbidities after cancer treatment.

Each year in the United States, approximately 70,000 adolescents and young adults (AYA, ages 15–39) are diagnosed with cancer (1). With advances in cancer treatments, 5-year relative survival among AYAs with cancer has steadily increased over the past few decades, and is currently over 80% for all cancer types combined (2). Long-term cancer survivors, however, may be at increased risk for a number of adverse health outcomes as a result of late effects of their curative cancer therapy.

Numerous reports from the Childhood Cancer Survivor Study and other cohorts have examined late morbidity and hospitalization among survivors of childhood cancers (3–5). However, these studies have generally only included survivors diagnosed at ages 0 to 20 years with selected cancer types common among children [e.g., leukemia, central nervous system (CNS) tumors]. Patterns of excess hospitalization among AYA cancer survivors, particularly those diagnosed with cancer types more predominant in the older end of the AYA age range (e.g., breast cancer, colorectal cancer), have not been as well-characterized (6–8). A better understanding of the burden of hospitalizations after cancer treatment among AYA survivors may help to anticipate future healthcare utilization in more recently diagnosed patients.

Using population-based data from the state of Utah, the objective of this study was to describe hospitalizations among 2-year AYA cancer survivors, and to compare these with hospitalizations among siblings and a sex- and age-matched general population cohort.

Data source

The Utah Population Database (UPDB) is a unique resource which links statewide administrative records for the entire Utah population (vital records, drivers' licenses, voter registration, marriage, and divorce records) to cancer diagnosis and treatment information available from the Utah Cancer Registry (UCR). The UCR has been part of the NCI's Surveillance Epidemiology and End Results (SEER) program since 1973 and maintains records for patients diagnosed in Utah beginning in 1966. Patient information available from the UCR includes basic demographic, tumor, and treatment information. UPDB also links administrative records and cancer diagnosis information to statewide hospital discharge data from the Utah Department of Health. The hospital discharge data include a record for each inpatient discharged from any acute care hospital in Utah and has been available in the UPDB since 1996. UPDB can also determine the date at which each person was last known to be residing in Utah from UPDB-linked records such as drivers' licenses and voter registration records. UPDB can follow the entire cancer and health care utilization history of the Utah population, irrespective of disease status, during their residence in the state because of these record linkages. Demographic information (sex, birth year, race/ethnicity) in the UPDB comes from multiple record sources, including Utah birth certificates, death certificates, and other administrative records. This study was approved by the Institutional Review Boards of the University of North Carolina, the University of Utah, and the Utah Resource for Genetic and Epidemiologic Research.

Study population

UPDB data were used to identify a cohort of patients who were diagnosed with a first malignant primary cancer at ages 15 to 39 years from 1994 to 2015 and survived at least 2 years after their initial cancer diagnosis. We selected 2-year survivors under the assumption that most AYA patients would have completed active treatment within 2 years of diagnosis. We excluded those without a Utah birth certificate and those who either died or were no longer known to be residing in Utah as of January 1, 1996, the date when hospitalization data became available. We also excluded those who died or were no longer known to be in Utah as of 2 years after cancer diagnosis.

Population-based comparisons were randomly selected from Utah birth certificates within the UPDB, and were matched to cancer cases on birth year and sex using a 3:1 ratio. Individuals were eligible to be selected as comparisons if they had survived and lived in Utah up until the date of diagnosis of the matched case. Selection criteria also required that population-based comparisons had not been diagnosed with cancer before age 40. All siblings of included AYA cancer cases who had a Utah birth certificate and did not have a cancer diagnosis before age 40 were also identified in the UPDB. For our analyses, we excluded matched comparisons and siblings who had died or were no longer living in Utah as of January 1, 1996, or 2 years after the corresponding cancer case's diagnosis date, whichever came later. We also excluded siblings born before 1955, the earliest birth year among cancer cases and matched comparisons.

Hospitalization outcomes

Hospital discharge records from the UPDB were used to identify hospitalization events among AYA cancer survivors, siblings, and the matched comparison group. The primary diagnosis associated with each hospitalization event was determined from International Classification of Disease (ICD) version 9 or 10 codes. Codes were categorized into meaningful diagnostic groups for analysis based on ICD chapters (9, 10). Primary diagnosis codes related to pregnancy and delivery were not included as hospitalization events.

Statistical analysis

Person-time of follow-up was accrued from 2 years after the date of cancer diagnosis among AYA cancer survivors (and the same date for their siblings and matched population comparisons) until death, last date known to be in Utah, or end of follow-up on December 31, 2017, whichever occurred first. Time to first hospitalization was evaluated for all ICD diagnostic groups combined and for each individual diagnostic group using Cox proportional hazards regression models to estimate hazard ratios (HR). For analyses comparing AYA cancer survivors to their siblings, we used marginal Cox models with robust sandwich estimators (11) to account for clustering within families. We also estimated the cumulative incidence of first hospitalization from 2 years postdiagnosis to 5, 10, and 20 years postdiagnosis using nonparametric methods to account for death as a competing risk (12).

Hospitalization rate ratios (RR) for the total number of hospitalizations of any diagnostic group were estimated using zero-inflated Poisson (ZIP) regression models, which are robust for modeling count variables with a large number of zeros (13); person-time of follow-up was used as the offset. In analyses comparing survivors to their siblings, we accounted for clustering using mixed ZIP models with random effects (14). Multivariable regression models were adjusted for sex, birth year, and diagnosis year. In sensitivity analyses, we excluded AYA survivors diagnosed with distant stage disease or leukemia, who we considered most likely to still be in active treatment after 2 years postdiagnosis. As a secondary analysis, we also estimated HRs and RRs among 5-year survivors relative to matched comparisons and siblings, with follow-up beginning at 5 years after cancer diagnosis.

A total of 6,330 AYA cancer survivors, 12,924 siblings, and 18,171 matched comparison subjects were identified and included in these analyses. The majority of AYA cancer survivors were female, non-Hispanic White, and diagnosed at ages 30 to 39 years (Table 1). The most common cancer diagnoses overall included thyroid cancer, melanoma, breast cancer, and testicular cancer.

Table 1.

Characteristics of 2-year AYA cancer survivors, matched population comparisons, and siblings.

SurvivorsMatched comparisonsSiblings
N (%)N (%)N (%)
Total 6,330 (100%) 18,171 (100%) 12,924 (100%) 
Median follow-up for first hospitalization, years (range) 4.9 (0.0–21.9) 6.8 (0.0–22.0) 6.8 (0.0–22.0) 
Sex 
 Male 2,581 (41%) 7,394 (41%) 6,518 (50%) 
 Female 3,749 (59%) 10,777 (59%) 6,406 (50%) 
Year of birth 
 1955–1969 1,430 (23%) 4,290 (24%) 3,250 (25%) 
 1970–1979 2,745 (43%) 7,961 (44%) 4,813 (37%) 
 1980–2000 2,155 (34%) 5,920 (33%) 4,861 (38%) 
Race/ethnicity 
 Non-Hispanic White 6,186 (98%) 17,538 (97%) 12,565 (98%) 
 Other 144 (2%) 520 (3%) 233 (2%) 
 Missing 113 126 
Age at diagnosis 
 15–19 years 651 (10%)   
 20–24 years 1,014 (16%)   
 25–29 years 1,353 (21%)   
 30–34 years 1,632 (26%)   
 35–39 years 1,680 (27%)   
Year of diagnosis 
 1994–1999 1,319 (21%)   
 2000–2009 2,952 (47%)   
 2010–2015 2,059 (33%)   
Cancer type 
 Thyroid 1,175 (19%)   
 Melanoma 1,147 (18%)   
 Testicular 640 (10%)   
 Breast 603 (10%)   
 Hodgkin lymphoma 442 (7%)   
 CNS tumors 334 (5%)   
 Cervix/uterus 273 (4%)   
 Colon/rectum 263 (4%)   
 Non-Hodgkin lymphoma 257 (4%)   
 Leukemia 231 (4%)   
 Other 965 (15%)   
Summary stage 
 Localized 3,664 (58%)   
 Regional 1,420 (22%)   
 Distant 733 (12%)   
 Unstaged 505 (8%)   
 Missing   
SurvivorsMatched comparisonsSiblings
N (%)N (%)N (%)
Total 6,330 (100%) 18,171 (100%) 12,924 (100%) 
Median follow-up for first hospitalization, years (range) 4.9 (0.0–21.9) 6.8 (0.0–22.0) 6.8 (0.0–22.0) 
Sex 
 Male 2,581 (41%) 7,394 (41%) 6,518 (50%) 
 Female 3,749 (59%) 10,777 (59%) 6,406 (50%) 
Year of birth 
 1955–1969 1,430 (23%) 4,290 (24%) 3,250 (25%) 
 1970–1979 2,745 (43%) 7,961 (44%) 4,813 (37%) 
 1980–2000 2,155 (34%) 5,920 (33%) 4,861 (38%) 
Race/ethnicity 
 Non-Hispanic White 6,186 (98%) 17,538 (97%) 12,565 (98%) 
 Other 144 (2%) 520 (3%) 233 (2%) 
 Missing 113 126 
Age at diagnosis 
 15–19 years 651 (10%)   
 20–24 years 1,014 (16%)   
 25–29 years 1,353 (21%)   
 30–34 years 1,632 (26%)   
 35–39 years 1,680 (27%)   
Year of diagnosis 
 1994–1999 1,319 (21%)   
 2000–2009 2,952 (47%)   
 2010–2015 2,059 (33%)   
Cancer type 
 Thyroid 1,175 (19%)   
 Melanoma 1,147 (18%)   
 Testicular 640 (10%)   
 Breast 603 (10%)   
 Hodgkin lymphoma 442 (7%)   
 CNS tumors 334 (5%)   
 Cervix/uterus 273 (4%)   
 Colon/rectum 263 (4%)   
 Non-Hodgkin lymphoma 257 (4%)   
 Leukemia 231 (4%)   
 Other 965 (15%)   
Summary stage 
 Localized 3,664 (58%)   
 Regional 1,420 (22%)   
 Distant 733 (12%)   
 Unstaged 505 (8%)   
 Missing   

Among AYA cancer survivors, there were 1,523 first hospitalization events over a median follow-up of 4.9 years (IQR = 1.8–9.8 years). The cumulative incidence of a first hospitalization by 10 years postdiagnosis (with follow-up beginning at 2 years postdiagnosis) was 25.5% for all cancer types combined (Supplementary Table S1). In multivariable-adjusted models, the risk of a first hospitalization among AYA cancer survivors was 1.93 [95% confidence interval (CI), 1.81–2.06] times that among the matched comparison group (Table 2). The elevation in risk was greater for male survivors (HR = 2.32; 95% CI, 2.08–2.60) than for female survivors (HR = 1.78; 95% CI, 1.65–1.92) and for those with distant stage disease (HR = 3.85; 95% CI, 3.39–4.38) and regional stage disease (HR = 2.22; 95% CI, 1.99–2.48) than those with localized stage disease (HR = 1.59; 95% CI, 1.47–1.72). HRs were significantly increased relative to matched comparisons for every cancer type, but were highest for leukemia, CNS tumors, colorectal cancers, non-Hodgkin lymphoma, and breast cancer, and lowest for cervical/uterine cancers and melanoma. Patterns were generally similar in analyses comparing survivors to their siblings, although HRs tended to be slightly attenuated (Table 2).

Table 2.

First hospitalization of any diagnostic group among 2-year AYA cancer survivors, matched population comparisons, and siblings.

Survivors vs. matched comparisonsaSurvivors vs. siblingsb
N eventsPerson-time (years)Unadjusted HR (95% CI)Adjusted HR (95% CI)cN eventsPerson-time (years)Unadjusted HR (95% CI)Adjusted HR (95% CI)c
Full sample 
 Matched comparisons/siblings (ref) 2,753 144,363 1,939 101,805 
 Survivors 1,523 40,552 1.93 (1.81–2.06) 1.93 (1.81–2.06) 1,205 32,875 1.89 (1.76–2.03) 1.78 (1.65–1.91) 
Females 
 Matched comparisons/siblings (ref) 2,001 81,229 1,193 48,761 
 Survivors 1,025 22,897 1.78 (1.65–1.92) 1.78 (1.65–1.92) 608 14,118 1.74 (1.57–1.92) 1.70 (1.54–1.88) 
Males 
 Matched comparisons/siblings (ref) 752 63,134 746 53,043 
 Survivors 498 17,655 2.33 (2.08–2.61) 2.32 (2.08–2.60) 283 10,806 1.84 (1.61–2.11) 1.87 (1.63–2.14) 
Leukemia 
 Matched comparisons/siblings (ref) 2,753 144,363 64 3,881 
 Survivors 94 1,138 4.24 (3.45–5.20) 4.76 (3.87–5.85) 69 925 4.29 (3.02–6.12) 4.45 (3.12–6.34) 
Non-Hodgkin lymphoma 
 Matched comparisons/siblings (ref) 2753 144,363 64 4,282 
 Survivors 75 1,581 2.45 (1.95–3.08) 2.76 (2.19–3.47) 68 1,370 3.21 (2.28–4.51) 3.38 (2.37–4.82) 
Hodgkin lymphoma 
 Matched comparisons/siblings (ref) 2,753 144,363 132 7,892 
 Survivors 96 3,189 1.57 (1.28–1.93) 1.77 (1.45–2.18) 74 2,613 1.66 (1.25–2.20) 1.66 (1.25–2.21) 
CNS tumors 
 Matched comparisons/siblings (ref) 2,753 144,363 119 5,697 
 Survivors 107 1,817 3.04 (2.51–3.69) 3.45 (2.84–4.19) 84 1,447 2.69 (2.06–3.51) 2.80 (2.14–3.68) 
Testiculard 
 Matched comparisons/siblings (ref) 752 63,134 67 5,956 
 Survivors 91 5,188 1.48 (1.19–1.84) 1.50 (1.21–1.87) 49 3,211 1.36 (0.95–1.95) 1.33 (0.94–1.90) 
Melanoma 
 Matched comparisons/siblings (ref) 2,753 144,363 348 18,792 
 Survivors 214 7,892 1.41 (1.23–1.62) 1.39 (1.21–1.60) 178 6,466 1.47 (1.23–1.76) 1.34 (1.12–1.61) 
Thyroid 
 Matched comparisons/siblings (ref) 2,753 144,363 331 17,075 
 Survivors 246 7,169 1.78 (1.56–2.03) 1.52 (1.33–1.73) 199 5,923 1.71 (1.45–2.02) 1.44 (1.21–1.72) 
Breaste 
 Matched comparisons/siblings (ref) 2,001 81,229 118 4,247 
 Survivors 197 3,224 2.42 (2.09–2.81) 2.37 (2.04–2.75) 114 1,896 2.07 (1.60–2.68) 2.02 (1.56–2.61) 
Cervix/uteruse 
 Matched comparisons/siblings (ref) 2,001 81,229 58 2,186 
 Survivors 69 2,173 1.30 (1.03–1.66) 1.28 (1.01–1.63) 40 1,302 1.16 (0.77–1.74) 1.13 (0.75–1.70) 
Colon/rectum 
 Matched comparisons/siblings (ref) 2,753 144,363 82 3,717 
 Survivors 70 1,244 2.88 (2.27–3.65) 2.83 (2.23–3.59) 59 1,022 2.51 (1.79–3.52) 2.39 (1.70–3.36) 
Localized stage 
 Matched comparisons/siblings (ref) 2,753 144,363 1,167 59,831 
 Survivors 784 25,584 1.60 (1.48–1.73) 1.59 (1.47–1.72) 614 20,651 1.52 (1.38–1.67) 1.41 (1.28–1.55) 
Regional stage 
 Matched comparisons/siblings (ref) 2,753 144,363 435 23,147 
 Survivors 373 8,444 2.28 (2.05–2.54) 2.22 (1.99–2.48) 299 6,912 2.23 (1.92–2.58) 2.05 (1.76–2.38) 
Distant stage 
 Matched comparisons/siblings (ref) 2,753 144,363 225 12,104 
 Survivors 258 3,817 3.45 (3.03–3.92) 3.85 (3.39–4.38) 204 3,126 3.27 (2.71–3.95) 3.34 (2.77–4.03) 
Unstaged 
 Matched comparisons/siblings (ref) 2,753 144,363 109 6,544 
 Survivors 105 2,663 2.01 (1.65–2.45) 2.04 (1.67–2.48) 87 2,158 2.37 (1.76–3.21) 2.32 (1.71–3.14) 
Survivors vs. matched comparisonsaSurvivors vs. siblingsb
N eventsPerson-time (years)Unadjusted HR (95% CI)Adjusted HR (95% CI)cN eventsPerson-time (years)Unadjusted HR (95% CI)Adjusted HR (95% CI)c
Full sample 
 Matched comparisons/siblings (ref) 2,753 144,363 1,939 101,805 
 Survivors 1,523 40,552 1.93 (1.81–2.06) 1.93 (1.81–2.06) 1,205 32,875 1.89 (1.76–2.03) 1.78 (1.65–1.91) 
Females 
 Matched comparisons/siblings (ref) 2,001 81,229 1,193 48,761 
 Survivors 1,025 22,897 1.78 (1.65–1.92) 1.78 (1.65–1.92) 608 14,118 1.74 (1.57–1.92) 1.70 (1.54–1.88) 
Males 
 Matched comparisons/siblings (ref) 752 63,134 746 53,043 
 Survivors 498 17,655 2.33 (2.08–2.61) 2.32 (2.08–2.60) 283 10,806 1.84 (1.61–2.11) 1.87 (1.63–2.14) 
Leukemia 
 Matched comparisons/siblings (ref) 2,753 144,363 64 3,881 
 Survivors 94 1,138 4.24 (3.45–5.20) 4.76 (3.87–5.85) 69 925 4.29 (3.02–6.12) 4.45 (3.12–6.34) 
Non-Hodgkin lymphoma 
 Matched comparisons/siblings (ref) 2753 144,363 64 4,282 
 Survivors 75 1,581 2.45 (1.95–3.08) 2.76 (2.19–3.47) 68 1,370 3.21 (2.28–4.51) 3.38 (2.37–4.82) 
Hodgkin lymphoma 
 Matched comparisons/siblings (ref) 2,753 144,363 132 7,892 
 Survivors 96 3,189 1.57 (1.28–1.93) 1.77 (1.45–2.18) 74 2,613 1.66 (1.25–2.20) 1.66 (1.25–2.21) 
CNS tumors 
 Matched comparisons/siblings (ref) 2,753 144,363 119 5,697 
 Survivors 107 1,817 3.04 (2.51–3.69) 3.45 (2.84–4.19) 84 1,447 2.69 (2.06–3.51) 2.80 (2.14–3.68) 
Testiculard 
 Matched comparisons/siblings (ref) 752 63,134 67 5,956 
 Survivors 91 5,188 1.48 (1.19–1.84) 1.50 (1.21–1.87) 49 3,211 1.36 (0.95–1.95) 1.33 (0.94–1.90) 
Melanoma 
 Matched comparisons/siblings (ref) 2,753 144,363 348 18,792 
 Survivors 214 7,892 1.41 (1.23–1.62) 1.39 (1.21–1.60) 178 6,466 1.47 (1.23–1.76) 1.34 (1.12–1.61) 
Thyroid 
 Matched comparisons/siblings (ref) 2,753 144,363 331 17,075 
 Survivors 246 7,169 1.78 (1.56–2.03) 1.52 (1.33–1.73) 199 5,923 1.71 (1.45–2.02) 1.44 (1.21–1.72) 
Breaste 
 Matched comparisons/siblings (ref) 2,001 81,229 118 4,247 
 Survivors 197 3,224 2.42 (2.09–2.81) 2.37 (2.04–2.75) 114 1,896 2.07 (1.60–2.68) 2.02 (1.56–2.61) 
Cervix/uteruse 
 Matched comparisons/siblings (ref) 2,001 81,229 58 2,186 
 Survivors 69 2,173 1.30 (1.03–1.66) 1.28 (1.01–1.63) 40 1,302 1.16 (0.77–1.74) 1.13 (0.75–1.70) 
Colon/rectum 
 Matched comparisons/siblings (ref) 2,753 144,363 82 3,717 
 Survivors 70 1,244 2.88 (2.27–3.65) 2.83 (2.23–3.59) 59 1,022 2.51 (1.79–3.52) 2.39 (1.70–3.36) 
Localized stage 
 Matched comparisons/siblings (ref) 2,753 144,363 1,167 59,831 
 Survivors 784 25,584 1.60 (1.48–1.73) 1.59 (1.47–1.72) 614 20,651 1.52 (1.38–1.67) 1.41 (1.28–1.55) 
Regional stage 
 Matched comparisons/siblings (ref) 2,753 144,363 435 23,147 
 Survivors 373 8,444 2.28 (2.05–2.54) 2.22 (1.99–2.48) 299 6,912 2.23 (1.92–2.58) 2.05 (1.76–2.38) 
Distant stage 
 Matched comparisons/siblings (ref) 2,753 144,363 225 12,104 
 Survivors 258 3,817 3.45 (3.03–3.92) 3.85 (3.39–4.38) 204 3,126 3.27 (2.71–3.95) 3.34 (2.77–4.03) 
Unstaged 
 Matched comparisons/siblings (ref) 2,753 144,363 109 6,544 
 Survivors 105 2,663 2.01 (1.65–2.45) 2.04 (1.67–2.48) 87 2,158 2.37 (1.76–3.21) 2.32 (1.71–3.14) 

aAnalysis compares full survivor cohort to full matched comparison cohort.

bAnalysis compares survivors with siblings to their siblings.

cAdjusted for sex, birth year, and diagnosis year.

dMales only.

eFemales only.

In analyses of first hospitalization within individual diagnostic groups, the HR for survivors relative to matched comparisons was highest for neoplasms (HR = 11.55; 95% CI, 9.24–14.44), followed by blood diseases (HR = 3.54; 95% CI, 2.77–4.52; Table 3). The risk among survivors was also more than double that of comparisons for supplementary factors, infectious and parasitic diseases, nervous system diseases, circulatory diseases, skin diseases, endocrine and metabolic diseases, symptoms, signs, and ill-defined conditions, respiratory diseases, and injury and poisoning. Smaller, though significant, elevations in risk among survivors were observed for digestive diseases, mental diseases, musculoskeletal diseases, and genitourinary diseases. Rate ratios for the total count of all hospitalizations of all diagnostic groups were also elevated among AYA cancer survivors relative to matched comparisons (RR = 2.05; 95% CI, 1.95–2.14) and siblings (RR = 1.56; 95% CI, 1.45–1.68; Supplementary Table S2).

Table 3.

First hospitalization within individual diagnostic groups among 2-year AYA cancer survivors, matched population comparisons, and siblings.

Survivors vs. matched comparisonsaSurvivors vs. siblingsb
N eventsPerson-time (years)Unadjusted HR (95% CI)Adjusted HR (95% CI)cN eventsPerson-time (years)Unadjusted HR (95% CI)Adjusted HR (95% CI)c
Infectious and parasitic 
 Matched comparisons/siblings (ref) 97 161,497 58 112,902 
 Survivors 81 48,217 2.77 (2.06–3.72) 2.75 (2.05–3.69) 64 39,054 2.60 (1.83–3.70) 2.60 (1.82–3.71) 
Neoplasm 
 Matched comparisons/siblings (ref) 99 161,300 92 112,636 
 Survivors 345 47,189 11.42 (9.13–14.28) 11.55 (9.24–14.44) 270 38,204 8.39 (6.63–10.60) 8.08 (6.35–10.28) 
Endocrine/metabolic 
 Matched comparisons/siblings (ref) 397 160,146 278 112,011 
 Survivors 286 47,419 2.46 (2.12–2.87) 2.45 (2.11–2.86) 229 38,378 2.41 (2.02–2.86) 2.21 (1.86–2.64) 
Blood 
 Matched comparisons/siblings (ref) 124 161,285 77 112,757 
 Survivors 133 48,073 3.56 (2.79–4.55) 3.54 (2.77–4.52) 107 38,921 4.00 (2.98–5.37) 3.84 (2.83–5.21) 
Mental 
 Matched comparisons/siblings (ref) 561 159,016 376 111,401 
 Survivors 249 47,530 1.47 (1.27–1.71) 1.46 (1.26–1.69) 173 38,594 1.32 (1.10–1.58) 1.29 (1.07–1.55) 
Nervous system 
 Matched comparisons/siblings (ref) 184 161,150 129 112,758 
 Survivors 150 48,081 2.75 (2.21–3.41) 2.72 (2.19–3.38) 111 38,965 2.50 (1.94–3.24) 2.43 (1.87–3.15) 
Circulatory 
 Matched comparisons/siblings (ref) 223 160,715 186 112,293 
 Survivors 166 47,887 2.54 (2.08–3.11) 2.58 (2.11–3.15) 132 38,802 2.08 (1.66–2.59) 2.14 (1.72–2.68) 
Respiratory 
 Matched comparisons/siblings (ref) 171 160,931 109 112,729 
 Survivors 123 48,146 2.40 (1.90–3.03) 2.40 (1.90–3.03) 97 38,990 2.55 (1.94–3.36) 2.50 (1.89–3.31) 
Digestive 
 Matched comparisons/siblings (ref) 405 159,517 293 111,407 
 Survivors 235 47,526 1.95 (1.66–2.29) 1.95 (1.66–2.29) 182 38,508 1.79 (1.50–2.14) 1.75 (1.46–2.10) 
Genitourinary 
 Matched comparisons/siblings (ref) 499 158,146 274 111,300 
 Survivors 194 47,405 1.29 (1.10–1.53) 1.30 (1.10–1.54) 152 38,401 1.60 (1.32–1.95) 1.37 (1.12–1.66) 
Skin 
 Matched comparisons/siblings (ref) 51 161,707 37 113,059 
 Survivors 40 48,384 2.58 (1.71–3.91) 2.57 (1.70–3.89) 33 39,180 2.56 (1.61–4.09) 2.56 (1.58–4.14) 
Musculoskeletal 
 Matched comparisons/siblings (ref) 297 160,349 230 112,118 
 Survivors 117 47,995 1.35 (1.09–1.68) 1.37 (1.11–1.70) 96 38,874 1.22 (0.97–1.54) 1.19 (0.94–1.50) 
Congenital 
 Matched comparisons/siblings (ref) 26 161,707 27 113,059 
 Survivors 13 48,384 1.69 (0.87–3.29) 1.67 (0.86–3.26) 11 39,180 1.17 (0.58–2.36) 1.17 (0.57–2.41) 
Symptoms, signs, and ill-defined conditions 
 Matched comparisons/siblings (ref) 335 160,156 265 111,986 
 Survivors 246 47,701 2.45 (2.08–2.89) 2.44 (2.07–2.88) 193 38,624 2.11 (1.75–2.54) 2.02 (1.66–2.44) 
Injury and poisoning 
 Matched comparisons/siblings (ref) 403 159,382 272 111,614 
 Survivors 257 47,358 2.14 (1.83–2.50) 2.14 (1.83–2.51) 181 38,410 1.93 (1.59–2.33) 1.93 (1.59–2.33) 
Supplementary factors influencing health status and contact with health system 
 Matched comparisons/siblings (ref) 656 159,104 446 111,458 
 Survivors 613 45,905 3.23 (2.89–3.61) 3.20 (2.86–3.57) 480 37,187 3.21 (2.83–3.64) 3.00 (2.63–3.41) 
Survivors vs. matched comparisonsaSurvivors vs. siblingsb
N eventsPerson-time (years)Unadjusted HR (95% CI)Adjusted HR (95% CI)cN eventsPerson-time (years)Unadjusted HR (95% CI)Adjusted HR (95% CI)c
Infectious and parasitic 
 Matched comparisons/siblings (ref) 97 161,497 58 112,902 
 Survivors 81 48,217 2.77 (2.06–3.72) 2.75 (2.05–3.69) 64 39,054 2.60 (1.83–3.70) 2.60 (1.82–3.71) 
Neoplasm 
 Matched comparisons/siblings (ref) 99 161,300 92 112,636 
 Survivors 345 47,189 11.42 (9.13–14.28) 11.55 (9.24–14.44) 270 38,204 8.39 (6.63–10.60) 8.08 (6.35–10.28) 
Endocrine/metabolic 
 Matched comparisons/siblings (ref) 397 160,146 278 112,011 
 Survivors 286 47,419 2.46 (2.12–2.87) 2.45 (2.11–2.86) 229 38,378 2.41 (2.02–2.86) 2.21 (1.86–2.64) 
Blood 
 Matched comparisons/siblings (ref) 124 161,285 77 112,757 
 Survivors 133 48,073 3.56 (2.79–4.55) 3.54 (2.77–4.52) 107 38,921 4.00 (2.98–5.37) 3.84 (2.83–5.21) 
Mental 
 Matched comparisons/siblings (ref) 561 159,016 376 111,401 
 Survivors 249 47,530 1.47 (1.27–1.71) 1.46 (1.26–1.69) 173 38,594 1.32 (1.10–1.58) 1.29 (1.07–1.55) 
Nervous system 
 Matched comparisons/siblings (ref) 184 161,150 129 112,758 
 Survivors 150 48,081 2.75 (2.21–3.41) 2.72 (2.19–3.38) 111 38,965 2.50 (1.94–3.24) 2.43 (1.87–3.15) 
Circulatory 
 Matched comparisons/siblings (ref) 223 160,715 186 112,293 
 Survivors 166 47,887 2.54 (2.08–3.11) 2.58 (2.11–3.15) 132 38,802 2.08 (1.66–2.59) 2.14 (1.72–2.68) 
Respiratory 
 Matched comparisons/siblings (ref) 171 160,931 109 112,729 
 Survivors 123 48,146 2.40 (1.90–3.03) 2.40 (1.90–3.03) 97 38,990 2.55 (1.94–3.36) 2.50 (1.89–3.31) 
Digestive 
 Matched comparisons/siblings (ref) 405 159,517 293 111,407 
 Survivors 235 47,526 1.95 (1.66–2.29) 1.95 (1.66–2.29) 182 38,508 1.79 (1.50–2.14) 1.75 (1.46–2.10) 
Genitourinary 
 Matched comparisons/siblings (ref) 499 158,146 274 111,300 
 Survivors 194 47,405 1.29 (1.10–1.53) 1.30 (1.10–1.54) 152 38,401 1.60 (1.32–1.95) 1.37 (1.12–1.66) 
Skin 
 Matched comparisons/siblings (ref) 51 161,707 37 113,059 
 Survivors 40 48,384 2.58 (1.71–3.91) 2.57 (1.70–3.89) 33 39,180 2.56 (1.61–4.09) 2.56 (1.58–4.14) 
Musculoskeletal 
 Matched comparisons/siblings (ref) 297 160,349 230 112,118 
 Survivors 117 47,995 1.35 (1.09–1.68) 1.37 (1.11–1.70) 96 38,874 1.22 (0.97–1.54) 1.19 (0.94–1.50) 
Congenital 
 Matched comparisons/siblings (ref) 26 161,707 27 113,059 
 Survivors 13 48,384 1.69 (0.87–3.29) 1.67 (0.86–3.26) 11 39,180 1.17 (0.58–2.36) 1.17 (0.57–2.41) 
Symptoms, signs, and ill-defined conditions 
 Matched comparisons/siblings (ref) 335 160,156 265 111,986 
 Survivors 246 47,701 2.45 (2.08–2.89) 2.44 (2.07–2.88) 193 38,624 2.11 (1.75–2.54) 2.02 (1.66–2.44) 
Injury and poisoning 
 Matched comparisons/siblings (ref) 403 159,382 272 111,614 
 Survivors 257 47,358 2.14 (1.83–2.50) 2.14 (1.83–2.51) 181 38,410 1.93 (1.59–2.33) 1.93 (1.59–2.33) 
Supplementary factors influencing health status and contact with health system 
 Matched comparisons/siblings (ref) 656 159,104 446 111,458 
 Survivors 613 45,905 3.23 (2.89–3.61) 3.20 (2.86–3.57) 480 37,187 3.21 (2.83–3.64) 3.00 (2.63–3.41) 

aAnalysis compares full survivor cohort to full matched comparison cohort.

bAnalysis compares survivors with siblings to their siblings.

cAdjusted for sex, birth year, diagnosis year.

In sensitivity analyses, HRs for first hospitalization among 2-year survivors relative to matched comparisons were 1.87 (95% CI, 1.75–2.00) and 1.77 (95% CI, 1.66–1.90) when AYAs with leukemia and distant stage disease, respectively, were excluded. Corresponding RRs for total hospitalizations were 1.98 (95% CI, 1.89–2.08) excluding leukemia, and 1.78 (95% CI, 1.69–1.87) excluding distant stage disease. In secondary analyses restricted to 5-year survivors (N = 4,579) and their matched comparisons (N = 12,990) and siblings (N = 9,220), HRs for first hospitalization were 1.62 (95% CI, 1.50–1.76) and 1.47 (95% CI, 1.34–1.61) for survivors versus matched comparisons and siblings, respectively (Supplementary Table S3). Across cancer types, HRs for survivors versus matched comparisons ranged from 1.25 for cervical/uterine cancers to 3.33 for leukemia. RRs for all hospitalizations were 1.77 (95% CI, 1.67–1.88) and 1.48 (95% CI, 1.33–1.64) for survivors relative to matched comparisons and siblings, respectively (Supplementary Table S4).

Many AYA cancer survivors have unique healthcare needs in the years following their cancer diagnosis and treatment. Using population-based hospitalization discharge data from Utah, we found that the risk of a first hospitalization among 2-year AYA cancer survivors was nearly twice that of a sex- and age-matched comparison cohort. The elevation in risk was greatest for AYAs survivors of leukemia, lymphomas, CNS tumors, breast cancer, and colorectal cancer. Results were generally similar in analyses comparing AYA cancer survivors to their siblings, further reinforcing the robustness of our conclusions. Our findings thus highlight the importance of risk-based survivorship care in this population.

Although the need for AYA-focused survivorship research has been increasingly recognized, AYAs cancer survivors continue to represent an understudied population, with current recommendations for follow-up care in this age group based largely on studies of patients diagnosed as children. Indeed, the most recent guidelines for AYA Oncology from the National Comprehensive Cancer Network specifically acknowledge the lack of large cohort studies to address survivorship issues among AYAs who exceed the upper age limit of most childhood cancer survivor studies (20 years; ref. 1). AYAs with cancer are distinct from childhood and adult cancer patients in their cancer type distribution, disease and host biology, patterns of treatment, and psychosocial issues (15), all of which may contribute to a unique risk profile for late effects of cancer treatment and other health outcomes and support the need for research specific to survivors diagnosed as AYAs.

The highly elevated risk of hospitalization for neoplasms among AYA cancer survivors in our cohort is expected, given the potential for recurrence of the primary cancer combined with the excess risk of a second malignancy among survivors (16). However, the risk of hospitalization among survivors was also significantly increased for nearly every other diagnostic group, with risk more than double that of the matched comparison group for, among others, infectious and parasitic diseases, endocrine and metabolic diseases, blood diseases, nervous system diseases, circulatory diseases, respiratory diseases, and skin diseases. These findings are generally consistent with studies of AYA survivors in Europe and Canada (6–8).

Given the potential for several cancer therapies to cause acute and/or late adverse effects on the heart and vascular system (17), it is critical to document morbidity from circulatory diseases among cancer survivors to identify high-risk patient groups. To our knowledge, only one prior U.S.-based study has examined hospitalization from circulatory diseases among AYAs with various cancer types. Though lacking general population rates for comparison, Keegan and colleagues reported rates of cardiovascular disease (CVD), defined as hospitalization or death from coronary artery disease, congestive heart failure, and stroke, among 2-year AYA cancer survivors in California (18). In their study, cancer types with the highest incidence of CVD included CNS tumors, acute lymphoid leukemia, acute myeloid leukemia, and non-Hodgkin lymphoma. Our smaller sample size in Utah did not allow for extensive cancer type-specific analyses of hospitalization for CVD defined as in analyses by Keegan and colleagues. However, our results indicate that hospitalization for circulatory diseases is significantly increased among AYA cancer survivors compared with the general population, underscoring the importance of long-term follow-up for prevention, detection, and management of circulatory diseases among patients with a history of AYA cancer.

In our analyses, some of the most elevated risks of hospitalization were observed among AYAs with leukemias, lymphomas, and CNS tumors, cancer types more predominant among younger AYAs and often associated with intensive and protracted cancer treatment regimens. Other studies of AYA and childhood patients with these cancer types have also reported a high risk of severe morbidity in the years following cancer diagnosis and treatment. In a study of 5-year AYA cancer survivors in Denmark, relative risks for first hospitalization were 2.21, 1.93, 1.87, and 1.64 for survivors of leukemia, brain cancer, Hodgkin lymphoma, and non-Hodgkin lymphoma, respectively, versus matched comparison subjects (6). For leukemia, the exceptionally high relative risk of a first hospitalization in our study probably reflects, in part, the greater likelihood for these patients to still be in active cancer treatment beyond 2 years postdiagnosis than AYAs with other malignancies. However, late effects of treatment or other conditions among long-term survivors are also likely important contributors to the high rates of hospitalization in this group.

Few AYA-specific studies have examined hospitalizations for AYAs with cancers more typically diagnosed in older adults, such as breast cancer and colorectal cancer. In our analyses, the risk of a first inpatient hospitalization among AYAs with breast and colorectal cancer was more than double that of both the matched comparison group and siblings, indicating a high burden of severe morbidity among survivors of these cancer types. In addition, the rate of total hospitalizations over the follow-up period among colorectal cancer survivors was more than three times that of matched comparisons, a finding which may be informative for anticipating the healthcare needs of individuals with a history of AYA colorectal cancer. Our results, considered in the context of recently reported increases in colorectal cancer incidence among young adults (19), emphasize the need for future research in larger samples to better understand the specific diagnoses that contribute to high rates of hospitalization among survivors.

Our study has several strengths. The unique data resources available in the UPDB allowed us to estimate the cumulative incidence of hospitalization among AYA cancer survivors, and to compare hospitalizations in the AYA survivor cohort to those among a sex- and age-matched population comparison group. We were also able to compare survivors to their siblings, an analysis which may provide some control for confounding due to unmeasured genetic, socioeconomic, and cultural factors. However, there are also certain limitations to our analyses. The number of hospitalizations was too small to perform cancer type-specific analyses of individual diagnostic groups or to conduct analyses focused on more specific diagnosis codes within diagnostic groups. Furthermore, we could only account for hospitalizations that occurred in Utah; information on care received out of state was not available. We were also unable to identify patients who were undergoing active cancer treatment during our follow-up period. Thus, for some conditions, such as blood disorders, hospitalizations among AYA cancer survivors could reflect toxicities associated with recent receipt of certain cancer therapies. Nevertheless, our findings provide information on the burden of hospitalization among AYAs who have survived at least 2 years from their cancer diagnosis.

In conclusion, our analyses using population-based data from Utah suggest that AYA cancer survivors have a higher risk of first hospitalization and higher rates of total hospitalizations than the general population and siblings. Cancer types at high risk included not only hematologic cancer and brain tumors, but also cancers nearly exclusive to older AYAs such as breast and colorectal cancer, indicating the importance of long-term, risk-based follow-up care to prevent and treat severe morbidities among survivors of these cancers. Future studies may be warranted to investigate associations between specific cancer therapies or lifestyle factors and hospitalization outcomes, and to identify strategies to reduce the burden of hospitalization among AYA cancer survivors.

No potential conflicts of interest were disclosed.

Conception and design: C. Anderson, J.Y. Ou, A.C. Kirchhoff, H.B. Nichols

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): J.M. Ramsay, A.C. Kirchhoff, H.B. Nichols

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): C. Anderson, J.G. Trogdon, A.C. Kirchhoff, H.B. Nichols

Writing, review, and/or revision of the manuscript: C. Anderson, H.K. Kaddas, J.Y. Ou, J.M. Ramsay, J.G. Trogdon, A.C. Kirchhoff, H.B. Nichols

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): H.K. Kaddas, J.M. Ramsay, H.B. Nichols

Study supervision: A.C. Kirchhoff, H.B. Nichols

Other (financial support): A.C. Kirchhoff

H.B. Nichols received funding from the St. Baldrick's Foundation (Scholar Award 523803).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1.
Coccia
PF
,
Pappo
AS
,
Beaupin
L
,
Borges
VF
,
Borinstein
SC
,
Chugh
R
, et al
Adolescent and young adult oncology, version 2.2018. NCCN Clinical Practice Guidelines in Oncology
.
J Natl Compr Cancer Net
2018
;
16
:
66
97
.
2.
Barr
RD
,
Ferrari
A
,
Ries
L
,
Whelan
J
,
Bleyer
WA
. 
Cancer in adolescents and young adults: a narrative review of the current status and a view of the future
.
JAMA Pediatr
2016
;
170
:
495
501
.
3.
Goldsby
R
,
Chen
Y
,
Raber
S
,
Li
L
,
Diefenbach
K
,
Shnorhavorian
M
, et al
Survivors of childhood cancer have increased risk of gastrointestinal complications later in life
.
Gastroenterology
2011
;
140
:
1464
71
.
4.
Perkins
JL
,
Chen
Y
,
Harris
A
,
Diller
L
,
Stovall
M
,
Armstrong
GT
, et al
Infections among long-term survivors of childhood and adolescent cancer: a report from the Childhood Cancer Survivor Study
.
Cancer
2014
;
120
:
2514
21
.
5.
Kurt
BA
,
Nolan
VG
,
Ness
KK
,
Neglia
JP
,
Tersak
JM
,
Hudson
MM
, et al
Hospitalization rates among survivors of childhood cancer in the Childhood Cancer Survivor Study cohort
.
Pediatr Blood Cancer
2012
;
59
:
126
32
.
6.
Rugbjerg
K
,
Olsen
JH
. 
Long-term risk of hospitalization for somatic diseases in survivors of adolescent or young adult cancer
.
JAMA Oncol
2016
;
2
:
193
200
.
7.
Richardson
DP
,
Daly
C
,
Sutradhar
R
,
Paszat
LF
,
Wilton
AS
,
Rabeneck
L
, et al
Hospitalization rates among survivors of young adult malignancies
.
J Clin Oncol
2015
;
33
:
2655
9
.
8.
Zhang
Y
,
Lorenzi
MF
,
Goddard
K
,
Spinelli
JJ
,
Gotay
C
,
McBride
ML
. 
Late morbidity leading to hospitalization among 5-year survivors of young adult cancer: a report of the childhood, adolescent and young adult cancer survivors research program
.
Int J Cancer
2014
;
134
:
1174
82
.
9.
World Health Organization
. 
International classification of diseases, 9th revision (ICD-9)
.
Available from:
cdc.gov/nchs/icd/icd9cm.htm.
10.
World Health Organization
. 
International classification of diseases, 10th revision (ICD-10)
.
Available from:
https://www.cdc.gov/nchs/icd/icd10cm.htm.
11.
Lin
DY
. 
Cox regression analysis of multivariate failure time data: the marginal approach
.
Stat Med
1994
;
13
:
2233
47
.
12.
Lin
G
,
So
Y
,
Johnston
G
. 
Analyzing survival data with competing risks using SAS software
.
SAS Global Forum 2012; Paper 344–2012
.
Available from:
support.sas.com/resources/papers/proceedings12/344-2012.pdf.
13.
Long
JS
.
Regression models for categorical and limited dependent variables
.
Thousand Oaks (CA)
:
Sage Publications
; 
1997
.
14.
Hall
DB
. 
Zero-inflated Poisson and binomial regression with random effects: a case study
.
Biometrics
2000
;
56
:
1030
9
.
15.
Sender
L
,
Zabokrtsky
KB
. 
Adolescent and young adult patients with cancer: a milieu of unique features
.
Nat Rev Clin Oncol
2015
;
12
:
465
80
.
16.
Lee
JS
,
DuBois
SG
,
Coccia
PF
,
Bleyer
A
,
Olin
RL
,
Goldsby
RE
. 
Increased risk of second malignant neoplasms in adolescents and young adults with cancer
.
Cancer
2016
;
122
:
116
23
.
17.
Curigliano
G
,
Cardinale
D
,
Dent
S
,
Criscitiello
C
,
Aseyev
O
,
Lenihan
D
, et al
Cardiotoxicity of anticancer treatments: epidemiology, detection, and management
.
CA Cancer J Clin
2016
;
66
:
309
25
.
18.
Keegan
THM
,
Kushi
LH
,
Li
Q
,
Brunson
A
,
Chawla
X
,
Chew
HK
, et al
Cardiovascular disease incidence in adolescent and young adult cancer survivors: a retrospective cohort study
.
J Cancer Surviv
2018
;
12
:
388
97
.
19.
Lui
RN
,
Tsoi
KK
,
Ho
JM
,
Lo
CM
,
Chan
FC
,
Kyaw
MH
, et al
Global increasing incidence of young-onset colorectal cancer across 5 continents: a joinpoint regression analysis of 1,922,167 cases
.
Cancer Epidemiol Biomarkers Prev
2019
;
28
:
1275
82
.