Short-term disability in solid tumor patients with bone metastases and skeletal-related events

Abstract

Objective:

The skeleton is a common site of metastasis in patients with solid tumors. These patients often experience pain and reduced quality-of-life. This analysis evaluated the time and costs associated with short-term disability use among solid tumor patients with bone metastases (BM) and skeletal-related events (SREs).

Methods:

Data from patients 18–64 years old with solid tumors and BM, eligible for short-term disability benefits between January 1, 2002 and December 31, 2010, were extracted from MarketScan Research Databases. Short-term disability hours and costs associated with BM and SREs were evaluated.

Results:

Overall, 1098 patients met the criteria. For all patients with BM, the monthly mean short-term disability hours were 17.7 h pre-BM diagnosis and increased to 60.2 h post-BM diagnosis (p < 0.001). The corresponding mean monthly short-term disability costs were $277 and $963 in the pre- and post-BM diagnosis periods, respectively (p < 0.001). Monthly mean short-term disability hours were higher for the cohort of patients with SREs (21.2 h pre-SRE diagnosis and 67.4 h post-SRE diagnosis) than for those without an SRE (8.6 h pre-SRE diagnosis and 14.4 h post-SRE diagnosis) (p < 0.001). Similarly, the corresponding monthly mean short-term disability costs were higher for patients with SREs ($625 and $1259 pre- and post-SRE diagnosis, respectively) than for patients without an SRE ($452 and $612 pre- and post-SRE diagnosis, respectively) (p < 0.001). Results of a multivariate analysis indicated that SREs were associated with an additional 39.4 short-term disability hours and $613 in short-term disability costs per month (p < 0.001).

>Conclusion:

Short-term disability hours and costs increased significantly when patients with solid tumors developed BM and SRE.

Keywords::

• Bone metastases
• Productivity loss
• Solid tumors
• Short-term disability
• Skeletal-related events

Introduction

Bone is one of the most common sites of distant metastases among patients with solid tumors. Up to 75% of patients with advanced prostate and breast cancers and ∼40% of patients with advanced lung cancer will develop metastatic bone disease1. Bone metastases weaken bone, primarily through increased osteoclast activity, which places patients at risk of clinically relevant downstream events, including pain and skeletal complications (also termed skeletal-related events [SREs])2–4. Bone metastases are the most common cause of cancer-related pain5. Pain is generally the first symptom of metastatic disease to the bone5^,6 and is often intolerable, treatment-refractory, and linked to reduced quality-of-life and poor mobility and prognosis7^,8.

Results from placebo-controlled clinical trials showed that, if bone metastases are untreated, up to 49% of prostate cancer patients and 68% of breast cancer patients can experience an SRE such as pathological fracture, radiotherapy or surgery to bone or spinal cord compression9–11. A recent study estimated that women in the US with breast cancer and bone metastases, if not treated with a bone-targeted agent, could experience more than 300,000 SREs per year12.

SREs secondary to bone metastasis can be painful and debilitating and can result in reduced health-related quality-of-life (HRQoL)13. Pathological fractures may impair ambulation, and spinal cord compressions can result in numbness or weakness, urinary or fecal incontinence, and paralysis. Patients may need radiation therapy to bone to control local tumor burden and manage pain and may need to undergo surgery to bone to prevent or treat pathologic fractures2^,14.

The direct medical costs from bone metastases and SREs in solid tumor patients are well documented. Total healthcare costs were greater in men with prostate cancer and bone metastasis relative to men with prostate cancer that had not metastasized to bone, and further increases in medical costs were found for men with prostate cancer and bone metastases with an SRE15. Higher total medical care costs were also estimated for both breast cancer patients16 and lung cancer patients17 with bone metastasis and SREs compared to patients with bone metastasis but no SREs. One study estimated that in patients with prostate cancer and bone metastases who developed SREs, total SRE costs were $20,984/episode (range = $4749/episode–$88,838/episode)18. A second study reported similar cost estimates in patients with breast cancer and bone metastases who developed SREs19.

While the clinical impact and direct costs associated with bone metastases and SREs are well documented, limited data are available on short-term disability hours and their related costs, and the contribution of these costs to the total economic burden of these conditions. The current study describes short-term disability hours and short-term disability costs in advanced cancer patients with bone metastases and SREs.

Methods

Data source

Data were extracted from Truven Health MarketScan® Commercial Claims and Encounters (Commercial) Database and the Health Productivity and Management (HPM) Database.

The Commercial database contains privately insured paid medical and prescription drug claims, derived primarily from self-insured employers. Annual enrollees in 2010 numbered ∼38 million individuals in the US. Healthcare was provided under a variety of health plans with fee-for-service and capitated payment arrangements, including preferred provider organizations, point of service plans, indemnity plans, and health maintenance organizations. The HPM database contains workplace absence, short-term disability, and workers’ compensation data for a subset of Truven Health’s employer clients. The database is fully linkable to the corresponding medical and pharmacy claims data for these employees. Because both the Commercial and HPM databases are de-identified in compliance with Health Insurance Portability and Accountability Act (HIPAA) regulations, Institutional Review Board approval was not required.

Patient selection

Patients were selected based on the presence of at least one inpatient or two outpatient claims for a solid tumor (International Classification of Diseases, 9th Revision, Clinical Modification [ICD-9-CM] diagnosis codes 140.xx–195.xx, 199.0, 199.2–202.38, 202.6x–202.98, 209.xx) and short-term disability eligibility for the study period between January 1, 2002 and December 31, 2010. Patients were assigned to the bone metastasis cohort if they had at least one bone metastasis diagnosis (ICD-9-CM diagnosis code 198.5) in this time window and were required to have at least one diagnosis of solid tumor prior to the first bone metastasis diagnosis (bone metastasis diagnosis date).

Patients with bone metastasis were required to have at least 6 months of continuous enrollment prior to the bone metastasis diagnosis date and at least 1 month subsequent to the bone metastasis diagnosis date, to be between the ages of 18–64 years, to have no evidence of pregnancy and/or delivery in the 6 months before and after the bone metastasis diagnosis date, to be active full-time employees, and to have lost no more than 248 hours of work due to short-term disability per month in the 6 months before and after the bone metastasis diagnosis date.

Patients in the bone metastases cohort were further divided into two cohorts: patients with and without SREs. SRE selection was based on the presence of at least one SRE diagnosis or procedure, with a diagnosis of bone metastasis prior to the SRE or no later than 30 days after the SRE diagnosis. The no SRE cohort included patients who had no evidence of SREs during the study period. As the sample size for patients without SREs was anticipated to be small, SRE-free time from the SRE cohort was included in the no SRE cohort if these patients had at least 10 months of data prior to their first SRE (a minimum of a 7-month follow-up [6-month pre-SRE + ≥1-month post-SRE] + a 3-month lag before the SRE to minimize the impact of the SRE).

The index date in the SRE cohort was the date of the first SRE claim (SRE index date). For each patient with an SRE, the number of days between the SRE index date and the bone metastasis index date was calculated, an ‘interval pool’ was created, and, for each no SRE patient, a randomly selected number from the pool was added to his/her bone metastasis index date to create the assigned index date. This was done to approximate the distribution of time between the first bone metastasis diagnosis and the index date of the patients in the SRE cohort and to account for natural disease progression.

Patients in all cohorts were followed until the end of continuous enrolment, end of short-term disability benefit eligibility, end of study period (i.e. December 31, 2010), or end of 6 months of follow-up, whichever was the earliest (Figure 1). For patients with SREs whose data was included in the no SRE cohort, only data prior to the SRE was included for the no SRE cohort analysis.

Figure 1. Schema for the study population. SRE, skeletal-related event.

Short-term disability outcomes

In the bone metastasis cohort, short-term disability was evaluated during the 6 months prior to the date of bone metastasis diagnosis and during the 6-months subsequent to the bone metastasis diagnosis date. Similarly, short-term disability was evaluated for the 6-months before and after the SRE index date for the SRE cohort or the assigned SRE index date for the no SRE cohort. Short-term disability measures included: the mean monthly short-term disability hours; the number of patients with >0 short-term disability hours; and the costs associated with the short-term disability hours.

Costs associated with short-term disability were calculated by multiplying the number of monthly short-term disability hours by 60% of the 2012 average hourly wage from the US Bureau of Labor Statistics (BLS)20. According to the BLS, this represents the median fixed percentage of annual earnings paid by short-term disability plans to private industry workers21.

Variables

Demographic variables as of the bone metastasis diagnosis date or SRE index date included age and age group, gender, urban vs rural residency (population density), US Census Bureau Region of residence (geographic region), health plan type, and mean length of the follow-up period.

Clinical variables were evaluated for the 6 months prior to the bone metastasis diagnosis date or SRE index date and included tumor type; Deyo Charlson Comorbidity Index (DCI)22; comorbid conditions; number of days between the first solid tumor diagnosis and the bone metastasis diagnosis date; and the number of days between the bone metastasis diagnosis date and the SRE index date. The DCI is a numeric scale based on the presence or absence of specific ICD-9-CM diagnosis codes consistent with chronic disease (e.g. diabetes, heart disease), each assigned a weight, with higher scores indicating greater burden.

Analyses

Chi-square tests were used to evaluate the statistical significance of differences for categorical variables between SRE and non-SRE patients; Fisher’s exact test was used for categorical variables with rare events. T-tests were used to evaluate differences for normally distributed continuous variables.

The difference-in-difference method23 was used to summarize incremental short-term disability hours between SRE and non-SRE patients while considering the difference between pre-period and follow-up period (i.e., natural progression of the disease).

Multivariate analyses, adjusting for demographic (i.e., age, gender) and clinical characteristics (i.e., cancer type, DCI, hypertension, cardiovascular disease, pain, fatigue, renal disease, fever) and monthly short-term disability hours (or indirect costs) in the pre-period, were conducted to quantify the impact of SREs on monthly short-term disability hours, the percentage of patients with short-term disability hours >0, and the costs associated with short-term disability.

Sensitivity analyses

Sensitivity analyses for the SRE and no SRE cohort comparison included adjusting the wage rate to 50% and 100% of average hourly wage, changing the SRE index date (SRE cohort)/assigned SRE index date (no SRE cohort) to 1 month prior to the diagnosis date to capture more accurately the impact of SRE as symptoms of SREs often lead to and therefore precede the initial SRE diagnosis, and increasing follow-up for up to 12 months to explore the longer-term impact of SREs.

Results

Demographic characteristics

Of 149,147 patients with a solid tumor diagnosis and eligibility for short-term disability between January 1, 2002, and December 31, 2010, 1098 patients with bone metastases met the selection criteria for age, employment status, and pregnancy (Table 1). Over half of the population was between 45–64 years of age and slightly over half was male.

Table 1. Demographic characteristics.

Characteristic	All patients with bone metastases (N = 1098)		Patients with bone metastases and SRE (N = 854)		Patients with bone metastases and without SRE* (N = 701)		p Value (SRE vs no SRE)
	n	%	n	%	n	%
Age category (years)							0.991
18–24	1	0.09	0	0	1	0.14
25–34	35	3.19	28	3.28	29	4.14
35–44	135	12.30	111	13.00	95	13.55
45–54	445	40.53	354	41.45	279	39.80
55–64	482	43.90	361	42.27	297	42.37
Age (years), mean (SD)	52.4 (7.8)		52.1 (7.8)		51.6 (8.0)		0.197
Gender							0.651
Male	620	56.47	470	55.04	408	58.20
Female	478	43.53	384	44.96	293	41.80
Population density							0.975
Urban	1000	91.07	771	90.28	639	91.16
Rural	98	8.93	82	9.60	61	8.70
Unknown	0	0	1	0.12	1	0.14
Geographic region							0.992
Northeast	200	18.21	143	16.74	135	19.26
North Central	357	32.51	273	31.97	219	31.24
South	375	34.15	311	36.42	241	34.38
West	165	15.03	125	14.64	105	14.98
Unknown	1	0.09	2	0.23	1	0.14
Health plan type							0.999
Health management  organization	136	12.4	109	12.8	101	14.4
Preferred provider  organization	680	61.9	538	63.0	419	59.8
Other	282	25.7	207	24.2	181	25.8
Length of follow-up period, mean days (SD)	206.47 (226.51)		192.17 (201.55)		311.32 (136.74)		<0.001

*No SRE cohort includes patients with no SREs during the study period plus the SRE-free time from patients in the SRE cohort with ≥10 months of data prior to their first SRE.

N, number of patients meeting study selection criteria; n, number of patients meeting study selection criteria with specified characteristic; SD, standard deviation; SRE, skeletal-related event.

Overall, 1098 patients had bone metastasis. Among the patients with bone metastasis, data from 854 patients who developed SREs and met the 6-month pre-SRE and 1-month post-SRE criteria are included in the SRE cohort. For the no SRE cohort, data from 701 patients are included: 94 patients had no evidence of an SRE, while 607 patients met the criteria allowing for the inclusion of SRE-free time in the no SRE cohort; the remainder of the study was based on analysis of the 701 patients in this cohort. There were no significant differences in the demographic variables among the patients in the SRE and no SRE cohorts (Table 1). In general, no significant differences were found in the demographic characteristics for the 94 patients with no evidence of SRE and those with an SRE, other than the mean age (54.4 years vs 52.1 years, respectively, p = 0.008).

Clinical characteristics

Compared to patients without SREs, patients with SREs had significantly higher rates of prostate or breast cancer (Table 2), consistent with the higher prevalence of bone metastases in patients with these tumor types2. Patients with SREs also had significantly higher rates of specific co-morbid conditions reflected in the higher mean DCI score (6.49 vs 2.05, p < 0.001).

Table 2. Clinical characteristics.

Characteristic	All patients with bone metastases (N = 1098)		Patients with bone metastases and SRE (N = 854)		Patients with bone metastases and without SRE* (N = 701)		p Value (SRE vs no SRE)
	n	%	n	%	n	%
Tumor type†
Prostate	309	28.14	236	27.63	162	23.11	0.042
Lung	116	10.56	69	8.08	61	8.70	0.659
Breast	300	27.32	263	30.80	94	13.41	<0.001
Other‡	577	52.55	461	53.98	479	68.33	<0.001
Deyo Charlson Comorbidity Index, mean (SD)	4.39 (3.31)		6.49 (3.16)		2.05 (2.86)		<0.001
Co-morbid conditions
Hypertension	231	21.04	188	22.01	100	14.27	<0.001
Cardiovascular disease	218	19.85	209	24.47	82	11.70	<0.001
Diabetes	15	1.37	13	1.52	5	0.71	0.138
Anemia	38	3.46	29	3.40	12	1.71	0.039
Depression	43	3.92	40	4.68	21	3.00	0.088
Pain	405	36.89	378	44.26	91	12.98	<0.001
Fever	28	2.55	23	2.69	3	0.43	0.001
Fatigue	68	6.19	64	7.49	25	3.57	0.001
Renal disease	91	8.29	97	11.36	23	3.28	<0.001
Days between first solid tumor diagnosis and first bone metastasis diagnosis, mean (SD)	93.92 (70.84)		91.23 (70.89)		91.92 (70.69)		0.847
Median	104.00		93.00		98.00
25%	15.00		14.00		15.00
75%	168.00		167.00		167.00
Days between first bone metastasis diagnosis and first SRE, mean (SD)			22.59 (104.02)		2.37 (25.05)		0.061
Median			0.00		n/a
25%			−7.00^$		n/a
75%			14.00		n/a

*No SRE cohort includes patients with no SREs during the study period plus the SRE-free time from patients in the SRE cohort with ≥10 months of data prior to their first SRE.

†Patients could have multiple cancer types.

‡Other includes lip, oral cavity, pharynx; digestive organs and peritoneum; respiratory and intrathoracic organs; neuroendocrine, bone, connective tissue, skin, other; genitourinary organs; lymphatic.

$In some patients, bone metastases were diagnosed after the patient experienced an SRE.

N, number of patients meeting study selection criteria; n, number of patients meeting study selection criteria with specified characteristic; SD, standard deviation; SRE, skeletal-related event; n/a, not applicable.

Short-term disability hours and costs

Among patients with bone metastasis, regardless of SRE status, the monthly mean short-term disability hours were 17.7 h in the pre-period compared to a monthly mean of 60.2 h in the 6 months following diagnosis of bone metastasis, a difference of 42.5 h (Table 3). The percentage of patients within the bone metastasis cohort who had >0 short-term disability hours increased from 34% in the pre-period to 48% in the 6-month follow-up period.

Table 3. Short-term disability hours and costs.

Measure	All patients with bone metastases (N = 1098)		Patients with bone metastases and SRE (N = 854)		Patients with bone metastases and without SRE* (N = 701)		p Value (SRE vs no SRE)
	Mean	SD	Mean	SD	Mean	SD
Pre-period
Monthly short-term disability hours	17.7	37.6	21.2	40.6	8.6	27.9	<0.001
Patients with short-term disability  hours >0, n (%)	372	33.88	362	42.39	102	14.55	0.003
Monthly short-term disability costs ($)	277	599	327	625	135	452	<0.001
6-month follow-up period
Monthly short-term disability hours	60.2	78.8	67.4	77.9	14.4	38.3	<0.001
Patients with short-term disability  hours >0, n (%)	522	47.54	446	52.22	134	19.12	0.001
Monthly short-term disability costs ($)	963	1284	1068	1259	230	612	<0.001

*No SRE cohort includes patients with no SREs during the study period plus the SRE-free time from patients in the SRE cohort with ≥10 months of data prior to their first SRE.

N, number of patients meeting study selection criteria; n, number of patients meeting study selection criteria with specified measure; SD, standard deviation; SRE, skeletal-related event.

For patients with or without SREs, the monthly short-term disability hours in the 6-month pre-SRE index date and the 6-month post-SRE index date are shown in Figure 2. While the no SRE cohort had similar monthly short-term disability hours before and after SRE the index date, the monthly short-term disability hours for the SRE cohort increased from 39.2 h in the month immediately preceding their SRE diagnosis to 70.3 h in the month following the SRE diagnosis. Over the 6-month post-diagnosis period, the monthly short-term disability hours gradually decreased from a peak of 70.3 h at month 1 post-diagnosis to 52.4 h at month 6 post-diagnosis, but did not return to pre-SRE diagnosis hours.

Figure 2. Short-term disability hours at 30-day intervals from the index date of the SRE. SRE, skeletal-related event.

On average, patients with SREs had monthly means of 21.2 and 67.4 short-term disability hours in the pre-period and in the 6 months of follow-up, respectively, a difference of 46.1 h per month (Table 3, Figure 3). Patients in the no SRE cohort used fewer short-term disability hours, with monthly means of 8.6 and 14.4 short-term disability hours, a difference of 5.8 h. The percentage of patients in the pre-period and in the 6 months of follow-up who had >0 short-term disability hours increased from 42% to 52% in the SRE cohort and from 15% to 19% in the no SRE cohort. Comparing the difference in monthly short-term disability hours in the SRE cohort between the pre-period and 6 months of follow-up with the difference in hours in these time periods in patients without SRE (difference-in-difference), we found 40.3 more monthly short-term disability hours for the SRE cohort, reflecting an increase of $647 in corresponding costs (p < 0.001 for both).

Figure 3. Difference in mean short-term disability hours: pre-period and follow-up period for SRE and no SRE cohorts. SRE, skeletal-related event.

Multivariate analysis indicated that, controlling for baseline characteristics, the development of SREs in bone metastasis patients was associated with an increase of 39.4 in monthly short-term disability hours and $613 in monthly short-term disability costs. This increased the odds of the patient with an SRE using short-term disability hours by 239% (odds ratio = 3.39, p < 0.001; Table 4). Overall, the development of an SRE resulted in the use of 236.5 more short-term disability hours and $3677 more in indirect costs during the first 6 months of follow-up.

Table 4. Multivariate analyses results.

Difference between	SRE vs no SRE (N = 1555)
	Marginal impact	p Value
Core analysis
6-month follow-up and pre-periods
Monthly short-term disability hours	39.4	<0.001
Short-term disability costs ($)	613	<0.001
% of patients with short-term disability hours >0 (odds ratio)	3.39	<0.001
Sensitivity analysis
Index date moved 1 month prior to SRE diagnosis: 6-month follow-up and pre-periods
Monthly short-term disability hours	45.3	<0.001
Short-term disability costs ($)	704	<0.001
% of patients with short-term disability hours >0 (odds ratio)	3.85	<0.001
Up to 12-month follow-up and no SRE index date change: Up to 12-month follow-up and pre-periods
Monthly short-term disability hours	37.0	<0.001
Short-term disability costs ($)	572	<0.001
% of patients with short-term disability hours >0 (odds ratio)	2.17	<0.001
SRE Index date moved 1 month prior to SRE diagnosis and up to 12-month follow-up period: Up to 12-month follow-up and pre-periods
Monthly short-term disability hours	42.2	<0.001
Short-term disability costs ($)	655	<0.001
% of patients with short-term disability hours >0 (odds ratio)	2.56	<0.001

SRE vs no SRE models control for SRE, baseline short-term disability hours (or indirect cost, or % with short-term disability >0), age, gender, Deyo Charlson Comorbidity Index, hypertension, cardiovascular disease, pain, fatigue, and renal disease.

SRE, skeletal-related event; N, number of patients in the SRE and no SRE cohorts combined.

Sensitivity analyses

As expected, shifting the SRE index date to 1 month before the SRE diagnosis date resulted in an additional 14.0 short-term disability hours per month for the difference-in-difference analysis, and an additional 5.8 short-term disability hours per month for the multivariate analysis. Using 50% and 100% of the average BLS hourly wage, additional costs per month translated into $539 and $1078, respectively.

To capture the longer-term burden of SREs, we increased the follow-up period for up to 12 months. On average, patients with SREs had 37.0 short-term disability hours per month and $572 in costs per month in the 12-month follow-up period (Table 4). Although the mean monthly short-term disability hours were lower for the 12-month follow-up period than for the 6-month follow-up period, the total short-term disability hours during the 12-month follow-up period were higher than those reported for the 6-month follow-up period. The SRE cohort used 443.4 more short-term disability hours and had $6865 more in indirect costs during the 12 months of follow-up than did patients in the no SRE cohort.

Discussion

In this study of patients with solid tumors and bone metastases, short-term disability hours and costs were significantly increased by the presence of SREs in patients. Monthly short-term disability hours and costs increased by 39.4 h and $613, respectively, in the 6-month post-SRE period. Moving the index to 1 month prior to the diagnosis date increased the short-term disability hours and costs. Increasing the follow-up period for up to 12 months facilitated the capture of the full impact of bone metastasis and SREs on short-term disability.

The National Institutes of Health estimated that, of the $216.6 billion in cancer costs in the US in 2009, $86.6 billion were attributable to direct medical costs and $130 billion to indirect mortality costs24. Indirect morbidity or lost productivity costs include absenteeism (absences, short- and long-term disability, and workers’ compensation) and presenteeism25. Similarly, the National Institute for Health and Care Excellence (NICE) reported that there were substantial indirect costs associated with cancer care in the European Union. Of the €126 billion total cancer care costs in 2009, healthcare accounted for €51 billion (40%), with informal care or productivity loss due to lost working days or death accounting for the remaining €75 billion (60%)26.

Several studies have evaluated work absences and short-term disability use in breast cancer patients. Nearly two thirds of patients with breast cancer reported missing work (44% up to 1 month and 24% over 1 month), while one third reported stopping work altogether27. Patients with metastatic breast cancer also had significantly more short-term disability days and costs (41.2 days and $6166) than patients with early-stage breast cancer (24.7 days and $3690)28. The indirect costs, including absenteeism and early retirement, and direct costs for treating the disease were found to be similar in patients with metastatic breast cancer29.

To our knowledge, this is the first report of short-term disability in solid tumor patients with bone metastasis and SREs. Our study indicates that SREs as a result of bone metastasis have an even greater impact on short-term disability, as demonstrated by the number of short-term disability days and costs for the patients with bone metastasis and SREs vs those without SREs. The costs associated with short-term disability vary between published studies28^,29 and our study, possibly reflecting differences in the study populations or methods used to analyze short-term disability costs. However, all studies consistently showed an increase in the use of short-term disability and costs associated with the development of cancer, bone metastasis, and skeletal complications.

There are a few limitations to this study. The overall study sample was drawn from a patient population covered by large self-insured employers. Thus, the results may not reflect short-term disability usage for patients covered by small employers. Also, the HPM database does not include actual cost data; therefore, short-term disability costs were imputed using average wage rates from the BLS, which contains data from both small and large employers. The average wage rate is often lower for small employers than for larger employers30; thus, applying the average wage rates from the BLS in the cost calculation might have underestimated short-term disability costs for this study population. Additionally, productivity loss associated with bone metastasis and SREs was likely underestimated because absence hours, due to a small sample size of patients with this data in the HPM database, and data relating to other indirect costs, such as presenteeism, are not available in the HPM database. ICD-9-CM diagnosis or procedure codes were used to identify the patients with solid tumors, bone metastasis, SREs, and co-morbidities, and these could have been miscoded, resulting in patients being misclassified. Finally, selection bias may have occurred with the inclusion criteria that all patients have a minimum of 6-month continuous enrollment in the pre-period and a minimum of 1-month continuous enrollment in the follow-up period. Patients not meeting these criteria, possibly reflecting a patient population with more advanced disease who may have used more short-term disability hours, were excluded from the study.

Conclusions

Short-term disability hours and costs in patients with bone metastasis secondary to solid tumors were significantly increased by the presence of SREs. In addition to the direct medical cost savings and improvement in patient quality-of-life, therapies that prevent or delay the development of these events may reduce work loss and associated indirect costs.

Transparency

Declaration of funding

This study was funded by Amgen Inc., Thousand Oaks, CA. The sponsor provided medical writing assistance to the authors.

Declaration of financial/other relationships

Yi Qian, Arun Balakumaran, and Jorge Arellano are employees and shareholders of Amgen Inc. Xue Song and Kui Zhang are employees of Truven Health Analytics Inc, who conducted the study on the behalf of Amgen Inc.

Acknowledgments

Writing assistance was provided by Amy Foreman-Wykert, PhD (Amgen Inc.). Data included in this manuscript was previously presented at the ISPOR 16th Annual European Meeting; Dublin, Ireland; November 2–6, 2013.