Abstract
Background
Few studies have evaluated the medication burden borne by survivors of pediatric cancer. This study aimed to describe the drug utilization pattern of chronic medications in a cohort of young pediatric cancer survivors.
Methods
This was a population-based study of patients diagnosed with cancer at age 18 years or younger between 2000 and 2013 in Hong Kong and who had survived at least 5 years postdiagnosis. The primary outcome is the use of any chronic medication (medications that were prescribed for ≥30 consecutive days within a 6-month period). Multivariable log-binomial models were used to identify factors associated with chronic medication use. Kaplan-Meier analysis was used to present the cumulative proportion of survivors initiated on a chronic medication across time from cancer diagnosis.
Results
Of the 2444 survivors (median age = 22 years, interquartile range = 16-27 years), 669 (27.4%) required at least 1 chronic medication at least 5 years postdiagnosis. Survivors who developed a chronic health condition (CHC) had a 5.48 (95% confidence interval [CI] = 4.49 to 6.71) times higher risk of taking a chronic medication than those without CHC. At 10 years postdiagnosis, the cumulative proportion of survivors being initiated a chronic medication was 33.4% (95% CI = 31.1% to 35.6%) for the overall cohort. Higher cumulative proportions were observed in survivors with endocrine (74.6%, 95% CI = 68.4% to 79.6%), renal (68.8%, 95% CI = 54.2% to 78.7%), neurological (58.6%, 95% CI = 46.1% to 68.1%), and cardiovascular (54.7%, 95% CI = 44.0% to 63.4%) disorders.
Conclusion
Survivors with certain CHCs had a higher risk of starting a prescription medication in the early phase of survivorship. Future studies include examining the impact of medication burden on survivors’ functional status.
Advances in the treatment of cancer have led to remarkable improvements in the overall survival rate of children and adolescents diagnosed with cancer (1,2). Despite the reductions in morbidity and mortality in this population, the risk of chronic and lifelong adverse effects of cancer treatment remain a legitimate concern (3-5). Such late effects include cardiopulmonary, neurological, and musculoskeletal complications and endocrine problems such as thyroid dysfunction, growth and development delays, and infertility (5). These patients also have a risk of developing neurocognitive and psychological deficits that may affect their quality of life (6,7).
The management of cancer- and treatment-related late effects may require the lifelong use of medication (8,9). One study in the United States reported that, compared with non-cancer controls, survivors of pediatric leukemia, lymphoma, and central nervous system (CNS) and bone cancers have higher rates of prescription medication use within 3 years posttherapy, suggesting the potential of emerging morbidities during the early survivorship period (10). These survivors are at risk of multiple medication use resulting in an increased medication burden. Examining the pattern of medication use in cancer survivors is important to identify individuals who should be monitored closely for adverse health outcomes that may result from the use of multiple chronic medications.
Previous studies have typically focused on evaluating the association of chronic disease burden with functional outcomes and health-care utilization in survivors of pediatric cancer (3,4,11-14). However, few epidemiological studies have evaluated the medication burden borne by these patients. Studies on this topic have been either single-center studies (15) or confined to specific cancer diagnoses (10) or drug classes (16,17) (eg, the long-term use of psychoactive medications). To address these research gaps, the primary objective of this study was to describe the pattern of chronic medication use by a cohort of young Chinese pediatric cancer survivors, using a population-based electronic medical database in Hong Kong. The secondary objective was to identify patient characteristics associated with chronic medication use.
Methods
Study Design
This was a population-based observational study of survivors of pediatric cancer within the Hospital Authority health-care system in Hong Kong. The Hospital Authority is the sole public health-care provider in Hong Kong, and it serves more than 90% of Hong Kong residents. We conducted our study using data from the Clinical Data Analysis and Reporting System (CDARS), which is a centralized clinical data repository for all 43 public hospitals, 49 specialist outpatient clinics, and more than 70 general outpatient clinics within the Hospital Authority (18). CDARS captures patients’ clinical and treatment information and has previously been validated as a data source for epidemiological studies in Hong Kong (19). This study was reviewed and approved by the Joint Chinese University-New Territories East Cluster Clinical Research Ethics Committee (reference number 2018.427).
Study Population
The study population comprised patients aged 18 years or younger who were registered with an International Classification of Diseases, Ninth Revision (ICD-9) diagnosis code of 140.00 to 209.79 between January 1, 2000, and December 31, 2013. Both incident and prevalent cases of cancer were included. Patients were excluded if they were aged older than 18 years at the time of cancer diagnosis (ie, when the ICD-9 code for cancer was first registered) or with ambiguous or uncertain diagnosis.
Medication Use
The definition used for a chronic medication was a medication that was prescribed to a patient with an intended duration of 30 or more consecutive days within a 6-month period or a medication that was prescribed to be used as needed but with an intended cumulative duration of 180 or more days within a 12-month period (15).
The primary outcome was the use of any newly prescribed chronic medications at 5 years or more postdiagnosis, as well as chronic medications that were prescribed at the end of therapy and were continued for more than 5 years postcancerdiagnosis. Medications were excluded from this category if they 1) were not filled or dispensed by the pharmacy, 2) had no active pharmaceutical ingredients, 3) were parenterally administered and required inpatient admission, or 4) had no valid start date. As this analysis is focused on medications with a curative intent in survivors, medications that were dispensed 6 months prior to death were also excluded. We did not exclude chronic medications that were prescribed before cancer diagnosis because it is often difficult to tease apart preexisting health conditions or symptoms from the cancer- or treatment-related conditions.
Medications were categorized according to the British National Formulary (BNF) classification (20). The BNF is a widely accepted framework for the drug management of common diseases. All drugs within the BNF are organized by the body systems and therapeutic drug classifications. For simplicity, the drugs used by survivors of this study are presented by the therapeutic drug classifications.
The secondary outcomes were medication burden (the number of chronic prescription medications) and polypharmacy (the use of ≥5 concurrent chronic medications). These were descriptive outcomes as our previous work had shown that the rate of polypharmacy among young survivors is low (5.3%) (15).
Predictive Variables
Based on a literature review (1,8,10,15-17), several predictive factors for medication burden were identified a priori, namely primary cancer diagnosis, chronic health conditions (CHCs), and history of relapse. Primary cancer diagnoses were classified based on the Surveillance, Epidemiology and End Results Program of the National Cancer Institute classification system (21). The information on CHCs was captured electronically from CDARS. A CHC was defined as a condition with chronic implications requiring long-term management (eg, gonadal dysfunction, on-treatment epilepsy, stroke, and heart failure) or a condition that was first registered at least 5 years postdiagnosis (22).
Statistical Analysis
Descriptive statistics were used to summarize the distribution of relevant outcome variables and covariates according to reasonable groupings based on previous reports on childhood cancer (11,12,14,15,23). Multivariable models were used to identify clinical factors associated with chronic medication use, after adjustment for sex, time since diagnosis, and age at diagnosis. Consistent with other studies evaluating health outcomes in childhood cancer survivors (16,24,25), the log-binomial models (generalized linear models with Poisson error and log-link function) were used to minimize biased estimates after adjusting for confounders in cohort studies (26,27). The corresponding relative risk (RR) estimates and 95% confidence intervals (CIs) were reported. Because cancer relapse and its treatment are associated with a higher burden of comorbidities (4,28,29), a sensitivity analysis was conducted to evaluate the association between the clinical factors and chronic medication use in the cohort excluding relapsed patients.
Finally, to illustrate descriptively medication utilization from early survivorship (5 to 10 years postcancer diagnosis) to long-term survivorship (≥10 years postdiagnosis), the cumulative proportion of survivors who were initiated a chronic medication across time since diagnosis was estimated using Kaplan-Meier analysis in the overall cohort and by the type of CHC. Events including death and loss to follow-up were censored. The Schoenfeld residuals were calculated for each variable to ensure that they independently satisfied the assumptions of the Cox model. A P value less than .05 was considered statistically significant, and all tests were 2-sided. All analyses were conducted using the R program.
Results
Study Population
Between January 1, 2000, and December 31, 2013, a total of 3123 patients who were aged 18 years or younger had an ICD-9 diagnosis code of 140.00 to 209.79 registered on their records (Supplementary Figure 1, available online). Among them, 81 subjects were excluded from subsequent analyses because of the ambiguous or uncertain nature of their diagnoses. After further excluding 598 subjects who passed away and thus were lost to follow-up within 5 years postdiagnosis, 2444 subjects were included in the final analyzed cohort (Table 1).
Table 1.
Study population demographics
| Demographics | No. (%) |
|---|---|
| Total No. of patients | 2444 |
| Sex | |
| Male | 1413 (57.8) |
| Female | 1031 (42.2) |
| Age at diagnosis, y | |
| Median (IQR) | 8 (3-14) |
| 0 to younger than 1 | 144 (5.9) |
| 1 to younger than 5 | 666 (27.3) |
| 5 to younger than 10 | 521 (21.3) |
| 10 to younger than 15 | 550 (22.5) |
| 15 or older | 563 (23.0) |
| Age at the last follow-up date, y | |
| Median (IQR) | 22 (16-27) |
| 0 to younger than 12 | 306 (12.5) |
| 12 to younger than 18 | 434 (17.8) |
| 18 to younger than 24 | 659 (27.0) |
| 24 to younger than 30 | 614 (25.1) |
| 30 to younger than 38 | 431 (17.6) |
| Most common cancer diagnosisa | |
| Leukemia | 923 (37.8) |
| Acute lymphoblastic leukemia | 652 (26.7) |
| Acute myeloid leukemia | 164 (6.7) |
| Chronic myeloid leukemia | 56 (2.3) |
| Other leukemia | 51 (2.1) |
| Lymphoma | 274 (11.2) |
| Non-Hodgkin lymphoma | 222 (9.1) |
| Hodgkin lymphoma | 52 (2.1) |
| CNS tumor | 182 (7.4) |
| Peripheral nervous system | 181 (7.4) |
| Bone tumor | 144 (5.9) |
| Other soft tissue sarcoma | 130 (5.3) |
| Retinoblastoma | 106 (4.3) |
| Thyroid cancer | 80 (3.3) |
| Kidney tumor | 67 (2.7) |
| Germ cell tumor | 63 (2.6) |
| Liver tumor | 60 (2.5) |
| Other carcinoma | 44 (1.8) |
| Cancer relapse | |
| Yes | 467 (19.1) |
| No | 1977 (80.9) |
| Chemotherapy | |
| Anthracyclines | 1049 (42.9) |
| Alkylating agents | |
| Mustard gas derivatives | 999 (40.9) |
| Heavy metals | 415 (17.0) |
| Others | 127 (5.2) |
| Vinca alkaloids | 1013 (41.4) |
| Antimetabolites | |
| Folic acid antagonists | 906 (37.1) |
| Purine antagonists | 852 (34.9) |
| Pyrimidine antagonists | 698 (28.6) |
| Others | 43 (1.8) |
| Topoisomerase inhibitors | 638 (26.1) |
| Asparaginase | 535 (21.9) |
| Chronic comorbiditiesb | |
| No | 1670 (68.3) |
| Any | 774 (31.7) |
| Endocrine | 300 (12.3) |
| Behavioral | 162 (6.6) |
| Musculoskeletal | 161 (6.6) |
| Cardiovascular | 138 (5.6) |
| Respiratory | 129 (5.3) |
| Renal/Urinary tract | 94 (3.8) |
| Neurological | 91 (3.7) |
| Vision/Hearing | 80 (3.3) |
| Liver | 26 (1.1) |
More than one cancer diagnosis may be present in a patient. CNS = central nervous system; IQR = interquartile range.
Defined as either acute complication that occurred after the date of cancer diagnosis and chronic implications requiring long-term management (eg, gonadal dysfunction, on-treatment epilepsy, stroke, heart failure) or a condition that was first registered at 5 or more years postdiagnosis.
Leukemia was the most common cancer diagnosis (37.8%), followed by lymphoma (11.2%) and CNS tumors (7.4%) (Table 1). At 5 years or more postdiagnosis, at least 1 diagnosis of CHC was documented in 774 (31.7%) survivors (Table 2). The most common CHCs were endocrine (12.3%), behavioral (6.6%), and musculoskeletal complications (6.6%). Survivors with CNS tumors were most likely to develop at least 1 CHC (58.2%), followed by survivors of sarcoma and leukemia (Supplementary Figure 2, available online). An analysis of the distribution of CHCs across cancer diagnoses showed that endocrine disorders were the most common diagnoses among survivors of CNS tumors, lymphoma, and leukemia, and musculoskeletal disorders were the most prevalent in survivors of sarcoma (Supplementary Figure 3, available online).
Table 2.
Chronic medication utilization at 5 or more years postdiagnosis
| Characteristics | No. (%) |
|---|---|
| Total No. of patients | 2444 |
| Use of chronic medication(s) postcancer treatmenta | |
| No | 1775 (72.6) |
| Any | 669 (27.4) |
| 1 to 4 medications | 575 (23.5) |
| ≥5 medications | 94 (3.8) |
| Subgroup with prescribed chronic medication | 669 |
| Classification of chronic medications postcancer treatmentb | |
| Thyroids and antithyroids | 152 (22.7) |
| Antihistamines, allergy control | 118 (17.6) |
| Sex hormones | 103 (15.4) |
| Miscellaneous ophthalmic preparations | 81 (12.1) |
| Drugs acting on the nose | 79 (11.8) |
| Vitamins | 65 (9.7) |
| Corticosteroids (endocrine) | 64 (9.6) |
| Topical corticosteroids | 63 (9.4) |
| Minerals | 63 (9.4) |
| Bronchodilators | 60 (9.0) |
| Hypothalamic and pituitary hormones and anti-estrogens | 54 (8.1) |
| Anti-inflammatory eye preparations | 51 (7.6) |
| Antisecretory drugs and mucosal protectant for the gastrointestinal system | 47 (7.0) |
| Cytotoxic drugs | 38 (5.7) |
| Drugs used in anemia and other blood disorders | 37 (5.5) |
| Inhaled corticosteroids | 36 (5.4) |
| Immunomodulation | 30 (4.5) |
| Antidepressants | 29 (4.3) |
| Drugs for hypertension and heart failure | 28 (4.2) |
| Antipsychotics and related drugs | 28 (4.2) |
| Laxatives | 25 (3.7) |
| Drugs for acne rosacea | 25 (3.7) |
| Fluids and electrolytes | 25 (3.7) |
| Analgesics | 24 (3.6) |
| Contraceptives | 22 (3.3) |
| Antidiabetics | 21 (3.1) |
| Lipid-regulating drugs | 19 (2.8) |
| Drugs used in rheumatic diseases and gout | 18 (2.7) |
| Drugs used in dyspepsia and gastro-oesophageal reflux disease | 17 (2.5) |
| Beta-adrenoceptor blocking drugs | 17 (2.5) |
| Cromoglycate, leukotriene, and phosphodesterase type-4 inhibitors | 17 (2.5) |
| Nitrates, calcium-channel blockers, and other antianginal drugs | 16 (2.4) |
| Mucolytics | 16 (2.4) |
Any newly prescribed chronic medications at ≥5 years postdiagnosis as well as chronic medications that were prescribed at end of therapy and were continued beyond ≥5 years postdiagnosis.
Includes the therapeutic classes prescribed to more than 2% of the cohort prescribed with chronic medications.
One-fifth of the cohort developed a relapse (19.1%; Supplementary Table 1, available online). Among survivors who relapsed, cancer recurrence occurred within 10 years from primary cancer diagnosis (77.5%). More than half of the relapsed survivors had developed a CHC (47.3%) during the time of evaluation.
Use of Chronic Medications
A subgroup of 669 survivors (27.4%) required at least 1 chronic medication (Table 2). The most commonly prescribed class of chronic medication was thyroid replacement therapy (22.7%), followed by antihistamines (17.6%) and sex hormones (15.4%). A minority (3.8%) of patients were prescribed at least 5 chronic medications (ie, polypharmacy) during the study period (Supplementary Table 2, available online).
The most commonly prescribed medications in survivors with endocrine disorders were thyroid hormone (28.0%), sex hormone (27.7%), and hypothalamic-pituitary hormone (16.7%) replacement therapies (Figure 1). A subgroup of survivors with behavioral disorders received antipsychotics (14.2%) and antidepressants (13.0%), and survivors with respiratory disorders were most commonly treated with antihistamines and medications for allergy control (23.3%) and bronchodilators (24.8%).
Figure 1.
Chronic medication burden by drug therapeutic classes against different groups of CHC. Each number represents the proportion in percentage of the subgroup with the indicated group of CHC in the header. Black cell signifies zero subjects. CHC = chronic health conditions.
Factors Associated With Chronic Medication Use
Compared with survivors of leukemia, survivors of CNS tumors were more likely to be prescribed a chronic medication (RR = 2.58, 95% CI = 1.83 to 3.64; Table 3). Among survivors of CNS tumors, the most commonly prescribed chronic medications included thyroid hormone (25.3%), hypothalamic-pituitary hormone (18.1%), and sex hormone (16.5%) (Supplementary Figure 4, available online). Survivors who developed a CHC had a 5 times higher risk of taking a chronic medication than those who did not have a CHC (RR = 5.48, 95% CI = 4.49 to 6.71). In particular, higher risks were observed in survivors with endocrine (RR = 4.95, 95% CI = 3.71 to 6.63), musculoskeletal (RR = 4.01, 95% CI = 2.74 to 5.88), and respiratory (RR = 3.92, 95% CI = 2.57 to 6.00) disorders.
Table 3.
Factors associated with chronic medication use
| Predictive factors | Estimated RR (95% CI) | P a |
|---|---|---|
| Chronic conditionsb | ||
| Any chronic comorbidities | 5.48 (4.49 to 6.71) | <.001 |
| Types of chronic conditionsb | ||
| Liverc | 0.803 (0.312 to 2.06) | .65 |
| Respiratory | 3.92 (2.57 to 6.00) | <.001 |
| Renalc | 1.30 (0.793 to 2.13) | .29 |
| Endocrine | 4.95 (3.71 to 6.63) | <.001 |
| Vision/Hearingc | 1.83 (1.05 to 3.20) | .03 |
| Cardiovascular | 1.31 (0.851 to 2.02) | .21 |
| Neurologicalc | 2.30 (1.34 to 3.94) | .003 |
| Behavioral | 3.50 (2.37 to 5.18) | <.001 |
| Musculoskeletal | 4.01 (2.74 to 5.88) | <.001 |
| Groups of primary cancer diagnosesb,d | ||
| Central nervous system | 2.58 (1.83 to 3.65) | <.001 |
| Germ cell tumor | 0.759 (0.381 to 1.42) | .41 |
| Hodgkin lymphoma | 0.497 (0.218 to 1.03) | .08 |
| Liver tumor | 0.796 (0.379 to 1.54) | .52 |
| Non-Hodgkin lymphoma | 0.551 (0.371 to 0.802) | .002 |
| Sarcoma | 0.744 (0.531 to 1.03) | .08 |
| Non-CNS solid tumor | 0.957 (0.727 to 1.26) | .75 |
| Cancer relapseb | ||
| Any cancer relapse | 2.97 (2.38 to 3.71) | <.001 |
Log-binomial models (generalized linear models with Poisson error and log-link function) were used to calculate the P values. All tests were 2-sided. CI = confidence interval; CNS = central nervous system; RR = relative risk.
Models are adjusted for sex, time since diagnosis, and age at diagnosis.
Interpreted with caution because of the small sample (n < 100).
Reference group is leukemia.
Survivors who had developed a relapse demonstrated a higher risk of being prescribed a chronic medication than survivors who did not (RR = 2.97, 95% CI = 2.38 to 3.71). The sensitivity analysis showed that the predictors of chronic medication use among survivors without any history of relapse (Supplementary Table 3, available online) were similar to the primary analysis on the overall cohort.
At 10 years postdiagnosis (ie, when survivors transit into long-term survivorship), the proportion of survivors with a chronic medication was 33.4% (95% CI = 31.1% to 35.6%) for the overall analyzed cohort. Figure 2 summarizes the cumulative proportion of survivors who were initiated a chronic medication, stratified by the type of CHC. For those survivors without a CHC, the proportion was small (20.4%, 95% CI = 18.0% to 22.7%). Higher cumulative proportions were observed in those with endocrine (74.6%, 95% CI = 68.4% to 79.6%), renal (68.8%, 95% CI = 54.2% to 78.7%), vision and hearing (63.2%, 95% CI = 48.5% to 73.6%), neurological (58.6%, 95% CI = 46.1% to 68.1%), respiratory (56.2%, 95% CI = 45.9% to 64.6%), behavioral (55.6%, 95% CI = 46.3% to 63.3%), and cardiovascular (54.7%, 95% CI = 44.0% to 63.4%) disorders.
Figure 2.
Cumulative proportion of survivors being initiated a chronic medication since cancer diagnosis. Kaplan-Meier plots for the initiation of chronic medication across time from cancer diagnosis (stratified by the type of CHC). CHC = chronic health conditions.
Discussion
Using real-world data from electronic health records, we sought to characterize the medication burden in a population of pediatric cancer survivors who were treated in the health-care system of Hong Kong after 2000. A subgroup (27.4%) of survivors were taking at least 1 chronic prescription medication. Overall, only a minority (3.8%) of survivors had polypharmacy. Survivors with endocrine, musculoskeletal, respiratory, and behavioral complications had the highest risk of starting a prescription medication early in the cancer survivorship continuum. This is one of the first studies to report the treatment burden of late effects borne by survivors of pediatric cancer.
Our results concur with findings from the literature showing that even contemporary treatment protocols carry a risk of late effects and may have implications on the medication burden during the early survivorship period (3). Survivors who developed at least 1 CHC were 5.48 times more likely to take a chronic medication than those who did not. The use of multiple medications in young survivors may have clinical implications, as physiological processes and neurodevelopment may be sensitive to the effects of pharmacological interventions during adolescence and young adulthood (30). For example, a previous study showed that the use of psychoactive medications during adolescence is associated with neurocognitive impairment and poorer social attainment later in life (16). Other studies have also reported worse functional outcomes in childhood cancer survivors with endocrine complications, although the comparative contribution of hormonal and metabolic abnormalities vs medications on adverse outcomes is often hard to distinguish (13,14,31). Admittedly, the lifelong use of chronic medications is likely warranted in certain conditions, such as hormonal replacement therapy in cancer survivors with hypothalamic or gonadal dysfunction. However, the screening and early identification of modifiable risk factors for certain late effects (eg, cardiovascular diseases, diabetes, and behavioral disorders) may facilitate timely lifestyle interventions before the condition becomes chronic and requires pharmacological interventions.
A higher medication burden was observed in survivors of CNS tumors and leukemia compared with survivors of other cancers. A similar observation was found in survivors who had a relapse. This finding is consistent with previous findings that these groups of survivors demonstrate the highest cumulative burden of CHCs (4,12,15,22,32); hence, they require the use of multiple medications. It is also important to note that concurrent medication use may confound the study of cancer-related symptoms in these survivors. For example, anticonvulsants and hormonal replacement therapy may exacerbate neurological and behavioral symptoms in CNS tumor survivors (32,33). Our results reinforce the need for closer monitoring of adverse health outcomes from multiple medications in these high-risk groups. Future studies should also evaluate the well-documented burden associated with multiple medications in patients with chronic diseases, such as nonadherence to medication therapy (34,35) and inappropriate drug prescribing and drug interactions (36,37), as well as increased health-care utilization costs (9,38).
The steep slope observed on the Kaplein-Meier curves within 5 years postdiagnosis seemed to suggest that survivors tended to be prescribed a chronic medication during the early posttherapy period rather than at a later phase of the survivorship period (ie, ≥10 years postdiagnosis). This is contrary to well-established evidence that the incidence of chronic morbidities increases as survivors age (22). Our findings may reflect the initiation of medications to treat chronic conditions that occur early in the posttreatment period (eg, hormonal replacement therapy for secondary endocrinopathy). A longer follow-up time on our cohort is required to assess medication utilization in survivors as they age and as late-occurring chronic conditions emerge. Respiratory chronic conditions were recorded in 5% of our survivors, hence, the use of prescribed systemic antihistamines may be for the symptomatic relief of respiratory conditions, or it may reflect the high prevalence of allergies in the local pediatric population due to environmental factors (39). As we do not expect the prevalence of allergic conditions to differ between cancer survivors and the general population (40), these conditions might be present even prior to the cancer diagnosis and may not be related to the cancer or its treatment. Future studies should include prospectively capturing the specific indications of chronic medications and the verification with consultation notes to characterize medication use patterns in individuals with preexisting conditions.
According to a landmark study, impaired pulmonary, cardiac, endocrine, and nervous system functions were most prevalent (detected in ≥20% of participants at risk) (22). Our results showed that patients with endocrine, respiratory, and neurological conditions were at the highest risk of being prescribed a chronic medication. However, cardiovascular conditions were not associated with chronic medication use in our multivariable analysis. Our cohort is still relatively young. The estimated prevalence of clinically significant cardiovascular conditions in survivors aged 40 years and younger was around 5% (22,41). Therefore, the non-statistically significant finding should be interpreted in light of the low proportion of survivors with documented cardiovascular conditions (5.6%) and with consideration that 54% of these survivors were on a chronic medication by the time they were 10 years postdiagnosis. On the contrary, even though the prevalence of renal late effects (5%) was substantially less common (22), our study found that the medication burden in this minority is high; 68% of patients with a renal condition were initiated a chronic medication at long-term survivorship. The medication burden for this group is likely to be higher when their renal health declines as they age. Future studies with a longer follow-up should investigate the evolving profile of morbidities and medication burden in high-risk groups of survivors.
The major strength of this study is the use of a large sample size of pediatric patients diagnosed with cancer in the public health-care network of Hong Kong. However, the findings of this study should be considered in the context of several limitations. First, this study did not include an age-matched healthy comparison control group. However, the prevalence of CHCs identified here (13.0%) is higher than the rates reported by the latest Hong Kong Census and Statistics Department (3.5% to 5.1% for individuals aged 15 to 39 years) (42). Given the high prevalence of CHCs reported in childhood cancer survivors (4,7,22), it is reasonable to speculate that the medication burden may be considerably higher in cancer survivors than in the age-matched general population. Notably, our reported prevalence of CHCs was substantially lower than previously published estimates, as most previous studies have performed systematic screening of late effects (4,22) or captured self-reported late effects (11,23). The confounding effects of the different chemotherapy drugs cannot be accounted for. This is because although CDARS has a high reported level of completeness for demographic and prescription data (19), there may be a certain degree of inaccuracy in cancer treatment data because of the use of paper forms for chemotherapy orders in the pediatric wards of some hospitals in the early 2000s. Therefore, some chemotherapy drugs may not have been captured electronically. Moreover, CDARS does not include dosimetry data; hence, we were unable to evaluate the effects of radiation on late effects and chronic medication use.
This is the first study harnessing population-based electronic health data to examine drug use patterns in a cohort of relatively young Chinese survivors of childhood cancer. Although polypharmacy was observed in a minority of the survivors in this study, chronic medication use was a common occurrence and has the potential to contribute to the future medical burden, particularly in survivors with endocrine, behavioral, and musculoskeletal late effects, as well as individuals who survived a relapse. These high-risk groups should be monitored closely for drug-drug interactions and adverse events during follow-up care, as well as nonadherence to drug therapy and increased cost associated with health-care services utilization. Future studies should also include examining the impact of multiple chronic medication use on the health status of survivors as they age.
Funding
No funding was received for conducting this study.
Notes
Role of the funder: Not applicable.
Disclosures: All authors do not have any conflict of interest to declare that are relevant to the content of this article.
Authors contributions: CLYE: Conceptualization, Methodology, Investigation, Data Curation, Project administration, Writing—Review & Editing. KHH: Conceptualization, Methodology, Software, Formal analysis, Investigation, Writing- Original draft preparation, Writing—Review & Editing, Visualization. SLKL: Methodology, Investigation, Writing—Review & Editing. AWKL: Methodology, Investigation, Writing—Review & Editing. GLHW: Conceptualization, Methodology, Investigation, Data Curation, Writing—Review & Editing. CKL: Conceptualization, Methodology, Investigation, Writing—Review & Editing. YTC: Conceptualization, Methodology, Software, Formal analysis, Investigation, Writing- Original draft preparation, Writing—Review & Editing, Supervision, Project administration.
Supplementary Material
Contributor Information
Celeste Lom-Ying Ewig, School of Pharmacy, Faculty of Medicine, The Chinese University of Hong Kong, Hong Kong SAR, China; Department of Pharmaceutical Outcomes and Policy, College of Pharmacy, University of Florida, Gainesville, USA.
Ka Ho Hui, School of Pharmacy, Faculty of Medicine, The Chinese University of Hong Kong, Hong Kong SAR, China.
Samantha Lai Ka Lee, Department of Paediatrics and Adolescent Medicine, Hong Kong Children’s Hospital, Hong Kong SAR, China.
Alex Wing Kwan Leung, Department of Paediatrics and Adolescent Medicine, Hong Kong Children’s Hospital, Hong Kong SAR, China; Department of Paediatrics, Faculty of Medicine, The Chinese University of Hong Kong, Hong Kong SAR, China; Hong Kong Hub of Paediatric Excellence, The Chinese University of Hong Kong, Hong Kong SAR, China.
Grace Lai-Hung Wong, Medical Data Analytics Centre (MDAC), Department of Medicine and Therapeutics, Faculty of Medicine, The Chinese University of Hong Kong, Hong Kong SAR, China.
Chi Kong Li, Department of Paediatrics and Adolescent Medicine, Hong Kong Children’s Hospital, Hong Kong SAR, China; Department of Paediatrics, Faculty of Medicine, The Chinese University of Hong Kong, Hong Kong SAR, China; Hong Kong Hub of Paediatric Excellence, The Chinese University of Hong Kong, Hong Kong SAR, China.
Yin Ting Cheung, School of Pharmacy, Faculty of Medicine, The Chinese University of Hong Kong, Hong Kong SAR, China.
Data Availability
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to identifiers in the dataset and are hence restricted from sharing by the Hospital Authority of Hong Kong.
References
- 1. Erdmann F, Frederiksen LE, Bonaventure A, et al. Childhood cancer: survival, treatment modalities, late effects and improvements over time. Cancer Epidemiol. 2021;71(pt B):101733. [DOI] [PubMed] [Google Scholar]
- 2. Force LM, Abdollahpour I, Advani SM, et al. The global burden of childhood and adolescent cancer in 2017: an analysis of the Global Burden of Disease Study 2017. Lancet Oncol. 2019;20(9):1211-1225. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Williams AM, Liu Q, Bhakta N, et al. Rethinking success in pediatric oncology: beyond 5-year survival. J Clin Oncol. 2021;39(20):2227-2231. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Bhakta N, Liu Q, Ness KK, et al. The cumulative burden of surviving childhood cancer: an initial report from the St Jude Lifetime Cohort Study (SJLIFE). Lancet. 2017;390(10112):2569-2582. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Landier W, Skinner R, Wallace WH, et al. Surveillance for late effects in childhood cancer survivors. J Clin Oncol. 2018;36(21):2216-2222. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Williams AM, Cheung YT, Hyun G, et al. Childhood neurotoxicity and brain resilience to adverse events during adulthood. Ann Neurol. 2021;89(3):534-545. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Yeh JM, Hanmer J, Ward ZJ, et al. Chronic conditions and utility-based health-related quality of life in adult childhood cancer survivors. J Natl Cancer Inst. 2016;108(9):djw046. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Murphy CC, Fullington HM, Alvarez CA, et al. Polypharmacy and patterns of prescription medication use among cancer survivors. Cancer. 2018;124(13):2850-2857. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Babcock ZR, Kogut SJ, Vyas A.. Association between polypharmacy and health-related quality of life among cancer survivors in the United States. J Cancer Surviv. 2020;14(1):89-99. [DOI] [PubMed] [Google Scholar]
- 10. Smitherman AB, Mohabir D, Wilkins TM, Blatt J, Nichols HB, Dusetzina SB.. Early post-therapy prescription drug usage among childhood and adolescent cancer survivors. J Pediatr. 2018;195:161-168.e7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Suh E, Stratton KL, Leisenring WM, et al. Late mortality and chronic health conditions in long-term survivors of early-adolescent and young adult cancers: a retrospective cohort analysis from the Childhood Cancer Survivor Study. Lancet Oncol. 2020;21(3):421-435. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Chang WH, Katsoulis M, Tan YY, Mueller SH, Green K, Lai AG.. Late effects of cancer in children, teenagers and young adults: population-based study on the burden of 183 conditions, in-patient and critical care admissions and years of life lost. Lancet Reg Health Eur. 2022;12:100248. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Vuotto SC, Krull KR, Li C, et al. Impact of chronic disease on emotional distress in adult survivors of childhood cancer: a report from the Childhood Cancer Survivor Study. Cancer. 2017;123(3):521-528. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Cheung YT, Brinkman TM, Li C, et al. Chronic health conditions and neurocognitive function in aging survivors of childhood cancer: a report from the childhood cancer survivor study. J Natl Cancer Inst. 2018;110(4):411-419. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Ewig CLY, Cheng YM, Li HS, et al. Use of chronic prescription medications and prevalence of polypharmacy in survivors of childhood cancer. Front Oncol. 2021;11:642544. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Cheung YT, Liu W, Brinkman TM, et al. Prescription psychoactive medication use in adolescent survivors of childhood cancer and association with adult functional outcomes. JNCI Cancer Spectr. 2020;4(5):pkaa057. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Brinkman TM, Ullrich NJ, Zhang N, et al. Prevalence and predictors of prescription psychoactive medication use in adult survivors of childhood cancer: a report from the Childhood Cancer Survivor Study. J Cancer Surviv. 2013;7(1):104-114. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Hospital Authority (Hong Kong). Clusters, hospitals & institutions. https://www.ha.org.hk/visitor/ha_visitor_index.asp?Content_ID=10036&Lang=ENG&Ver=HTML. Accessed July 5, 2022.
- 19. Wong MCS, Jiang JY, Tang J-L, Lam A, Fung H, Mercer SW.. Health services research in the public healthcare system in Hong Kong: an analysis of over 1 million antihypertensive prescriptions between 2004-2007 as an example of the potential and pitfalls of using routinely collected electronic patient data. BMC Health Serv Res. 2008;8(1):138. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Joint Formulary Committee. British National Formulary 81. London: BMJ Publishing and the Royal Pharmaceutical Society; 2021. [Google Scholar]
- 21. . National Cancer Institute Surveillance, Epidemiology and End Results Program. SEER cancer classification. https://training.seer.cancer.gov/disease/categories/classification.html. Accessed November 1, 2021.
- 22. Hudson MM, Ness KK, Gurney JG, et al. Clinical ascertainment of health outcomes among adults treated for childhood cancer. JAMA. 2013;309(22):2371-2381. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Streefkerk N, Tissing WJE, van der Heiden-van der Loo M, et al. The Dutch LATER physical outcomes set for self-reported data in survivors of childhood cancer. J Cancer Surviv. 2020;14(5):666-676. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Dixon SB, Chen Y, Yasui Y, et al. Impact of risk-stratified therapy on health status in survivors of childhood acute lymphoblastic leukemia: a report from the childhood cancer survivor study. Cancer Epidemiol Biomarkers Prev. 2022;31(1):150-160. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Zheng DJ, Krull KR, Chen Y, et al. Long-term psychological and educational outcomes for survivors of neuroblastoma: a report from the Childhood Cancer Survivor Study. Cancer. 2018;124(15):3220-3230. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. McNutt L-A, Wu C, Xue X, Hafner JP.. Estimating the relative risk in cohort studies and clinical trials of common outcomes. Am J Epidemiol. 2003;157(10):940-943. [DOI] [PubMed] [Google Scholar]
- 27. Skov T, Deddens J, Petersen MR, Endahl L.. Prevalence proportion ratios: estimation and hypothesis testing. Int J Epidemiol. 1998;27(1):91-95. [DOI] [PubMed] [Google Scholar]
- 28. Ehrhardt MJ, Williams AM, Liu Q, et al. Cumulative burden of chronic health conditions among adolescent and young adult survivors of childhood cancer: identification of vulnerable groups at key medical transitions. Pediatr Blood Cancer. 2021;68(6):e29030. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Essig S, von der Weid NX, Strippoli MP, et al. ; for the Swiss Pediatric Oncology Group (SPOG). Health-related quality of life in long-term survivors of relapsed childhood acute lymphoblastic leukemia. PLoS One. 2012;7(5):e38015. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Horace AE, Ahmed F.. Polypharmacy in pediatric patients and opportunities for pharmacists’ involvement. Integr Pharm Res Pract. 2015;4:113-126. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Tonning Olsson I, Alberts NM, Li C, et al. Pain and functional outcomes in adult survivors of childhood cancer: a report from the St. Jude Lifetime Cohort study. Cancer. 2021;127(10):1679-1689. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Alemany M, Velasco R, Simó M, Bruna J.. Late effects of cancer treatment: consequences for long-term brain cancer survivors. Neurooncol Pract. 2021;8(1):18-30. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Schiff D, Lee EQ, Nayak L, Norden AD, Reardon DA, Wen PY.. Medical management of brain tumors and the sequelae of treatment. Neuro Oncol. 2015;17(4):488-504. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Calip GS, Xing S, Jun DH, Lee WJ, Hoskins KF, Ko NY.. Polypharmacy and adherence to adjuvant endocrine therapy for breast cancer. J Oncol Pract. 2017;13(5):e451-e62. [DOI] [PubMed] [Google Scholar]
- 35. Bouwman L, Eeltink CM, Visser O, Janssen J, Maaskant JM.. Prevalence and associated factors of medication non-adherence in hematological-oncological patients in their home situation. BMC Cancer. 2017;17(1):739. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. Rambhade S, Chakarborty A, Shrivastava A, Patil UK, Rambhade A.. A survey on polypharmacy and use of inappropriate medications. Toxicol Int. 2012;19(1):68-73. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37. Elbeddini A, To A, Tayefehchamani Y, Wen CX.. Importance of medication reconciliation in cancer patients. J Pharm Policy Pract. 2021;14(1):98. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38. Babcock ZR. Polypharmacy in Cancer Patients: Health-Related Quality of Life, Expenditures, and Adverse Events [Open Access Dissertations. Paper 869.]. Kingston, RI: University of Rhode Island; 2019. https://digitalcommons.uri.edu/oa_diss/869. Accessed July 5, 2022.
- 39. Wong GWK, Leung TF, Ko FWS.. Changing prevalence of allergic diseases in the Asia-Pacific region. Allergy Asthma Immunol Res. 2013;5(5):251-257. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40. University of Hong Kong. Survey Findings on Prevalence of Allergic Diseases among Hong Kong Primary and Secondary Schoolchildren [press release ]. January 13, 2022. https://www.hku.hk/f/upload/23935/Appendix_eng.pdf. Accessed July 5, 2022.
- 41. Feijen E, Font-Gonzalez A, Van der Pal HJH, et al. ; for the DCOG‐LATER Study Group. Risk and temporal changes of heart failure among 5-year childhood cancer survivors: a DCOG-LATER Study. J Am Heart Assoc. 2019;8(1):e009122. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42. Census and Statistics Department. Persons with Disabilities and Chronic Diseases. Thematic Household Survey Report No. 63. 2020. https://www.censtatd.gov.hk/en/data/stat_report/product/C0000055/att/B11301632021XXXXB0100.pdf.
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to identifiers in the dataset and are hence restricted from sharing by the Hospital Authority of Hong Kong.