Prostate Cancer Care Before and After Medicare Eligibility.

Prior studies suggest Medicare eligibility confers significant and substantial reductions in mortality and beneficial increases in health service utilization. We compared 13,882 patients diagnosed with prostate cancer at ages 63 to 64 years with 14,774 patients diagnosed at ages 65 to 66 (controls) in 2004 to 2007. Compared with controls, patients diagnosed with prostate cancer before Medicare eligibility had no statistically significant or meaningful differences in cancer stage, time to treatment, or type of treatment.

Introduction

Health insurance coverage is generally associated with better health outcomes and with receipt of appropriate care,[1-3] but whether this is true among the near elderly[4-6] or among cases of common, serious cancers diagnosed around the 65th year of age[7] is less well understood. In this brief report, to be considered in conjunction with a larger accompanying study of lung cancer, we examine prostate cancer and seek to understand whether the magnitude of beneficial insurance effects is as high as suggested by the prior literature,[8,9] and whether paradoxical harms caused by patient and physician moral hazard exist.

Background and Hypotheses

Prostate cancer is typically a more slowly progressing cancer, one of the most common cancers among US men,[10] prevalent among patients around the time of access to relatively low cost, generous Medicare health insurance. We hypothesized that, as for lung cancer, diagnoses made before or after Medicare eligibility would be associated with different staging and treatment.

More treatment may also be obtained by some minorities who tend to present with more advanced disease,[11] and in whom stage at diagnosis is inversely correlated with insurance status and income.[7,12,13] Access to Medicare may also allow the use of potentially unwarranted aggressive therapy such as radical prostatectomy or radiotherapy in low-risk strata in response to the ability of patients to request such services or physicians to provide and bill for them.[14-16]

Moreover, in the absence of clearly superior treatment modalities,[17-21] any variation in treatment may harm patients and result in the inefficient use of scarce health care resources.

Better access to care may increase screening in response to Medicare’s prostate-specific antigen (PSA) and digital rectal exam screening services.[22,23] We hypothesized that patients without symptoms on turning 65 years of age might receive more screening due to increased access to primary and preventive care. Access to intensive surgical or radiotherapeutic treatment options might then be more available to these patients than their younger cohort neighbors leading to higher utilization of both surgery and radiotherapy.

Data and Methods

We analyzed a national convenience samples based on Surveillance, Epidemiology, and End Results (SEER) registry data maintained by the National Cancer Institute (NCI). The NCI administers 15 SEER registries, which cover approximately 26% of the national population. We used birth month and year and first diagnosis of prostate cancer diagnosis month and year to construct 2 cohorts. We included 13 882 patients in a pre-Medicare-eligibility 2-year cohort aged between 63 and 64, and 14 774 patients in a post-65-year-old 2-year cohort aged between 65 and 66. We tested the significance of changes in categorical variables using chi-square tests. We used Kruskal-Wallis equality of population tests for changes in continuous variables such as tumor marker variables or ordinal biopsy score, and median tests for equality of medians. We compared proportions receiving types of treatment using Fisher’s exact tests of crude risk ratios and adjusted risk ratios.

This study was approved by the institutional review board (IRB) of the study institution’s Health System and declared exempt from IRB review under 45CFR46.101(b)(4).

Results

Baseline characteristics and diagnostic characteristics were small and generally not clinically meaningful (results shown in the appendix). Risk stratification was very similar across the 2 cohorts. Testing for trend across low-risk, intermediate-risk, and high-risk strata was barely significant (P = .039). The proportion of patients classifiable as low risk decreased slightly from 27.6% to 27.3% (P = .54), the proportions classifiable as intermediate risk rose slightly from 22.0% to 23.0% (P = .028), while the high-risk category saw a small decrease from 42.3% to 41.1% (P = .036).

Overall, we found small and inconsistent, albeit significant changes in treatment comparing the 2-year-older cohort with the younger one (Table 1). In the older cohort aged 65 to 66, there was a 3.7% point decrease in the use of surgery of any sort (P < .001); this reduction came chiefly through declining use of radical prostatectomy from 44.3% to 39.6% (P < .001) and was present both overall and among a low-risk subset of patients (lower panel).

Table 1.

Prostate Cancer, Type of Therapy Received, and Crude and Adjusted Risk Ratios.

	No. (%)		P value	Risk ratios (95% CI) of 65- to 66-year cohort vs 63- to 64-year cohort
	63- to 64-year cohort (n = 13882)	65- to 66-year cohort (n = 14774)		Crude	Adjusted[a]
Surgery received
None	7068 (50.9)	8068 (54.6)	<.001	1.07 (1.05-1.10)	1.07 (1.05-1.10)
Local destruction[b]	152 (1.1)	236 (1.6)	<.001	1.46 (1.19-1.79)	1.43 (1.17-1.75)
TURP ± local destruction[b]	347 (2.5)	429 (2.9)	.04	1.16 (1.01-1.34)	1.15 (0.99-1.33)
Radical prostatectomy	6146 (44.3)	5846 (39.6)	<.001	0.89 (0.87-0.92)	0.90 (0.87-0.92)
Radiation received
None	8670 (62.5)	8910 (60.3)	<.001	0.97 (0.95-0.98)	0.96 (0.94-0.98)
Beam	2675 (19.3)	3108 (21.0)	<.001	1.09 (1.04-1.14)	1.09 (1.04-1.14)
Brachytherapy	1518 (10.9)	1675 (11.3)	.28	1.04 (0.97-1.11)	1.05 (0.99-1.13)
Beam and brachytherapy	643 (4.6)	698 (4.7)	.71	1.02 (0.92-1.13)	1.06 (0.95-1.17)
Neither surgery nor radiation	2353 (17.0)	2766 (18.7)	<.001	1.10 (1.05-1.16)	1.08 (1.03-1.14)
Low-risk stratum[c]	(n = 3835)	(n = 4034)
No surgery	2680 (69.9)	2931 (72.7)	.007	1.04 (1.01-1.07)	1.05 (1.02-1.08)
Radical prostatectomy	991 (25.8)	882 (21.9)	<.001	0.85 (0.78-0.92)	0.83 (0.77-0.90)
No radiation	1873 (48.8)	1858 (46.1)	.01	0.94 (0.90-0.99)	0.93 (0.88-0.97)
Beam radiotherapy	776 (20.2)	908 (22.5)	.01	1.11 (1.02-1.21)	1.12 (1.02-1.22)
Neither surgery nor radiation	775 (20.2)	833 (20.6)	.63	1.02 (0.94-1.12)	1.00 (0.92-1.10)

Note. CI = confidence interval; TURP = transurethral resection of the prostate; PSA = prostate-specific antigen.

Adjusted for registry locations, year of diagnosis, Hispanic ethnicity, race, and marital status.

Cryoprostatectomy, laser ablation, hyperthermic, microwave, ultrasound, needle, or other local tumor destruction.

PSA ≤ 10 ng/mL, and T2a or lower, and Gleason score of 6 or lower, where T2 NOS (not otherwise specified) staged disease classified as ≤T2a.

Conclusions

This study used cancer registry data to detail disease, treatment, and outcome differences among the near elderly and elderly around the age of 65 years for a common and important cancer.

Our approach was designed to identify harms from underinsurance, harms from insurance due to overdiagnosis and overtreatment, and benefits from insurance due to better and more timely access to care. Nevertheless, our study failed to show substantial, consistent, or clinically meaningful differences between patients diagnosed before and immediately after eligibility for Medicare.

Improved access to insurance has been found to be associated with better access to health care and improvement in health,[24] with substantially improved survival after acute conditions,[9] and with substantially increased utilization of care.[8,22,25] Yet this adequately powered study was unable to detect differences in utilization of the magnitude found by prior studies.

On the contrary, access to insurance could lead to moral hazard in the setting of prostate cancer. Of particular concern is the risk that provider preferences or provider financial self-interest could bias treatment toward more aggressive care. Widespread and growing concern exists about aggressive therapy in men who may do similarly well under more conservative “active surveillance” approaches.[26-28] Closely related to this is concern about the harm caused by overdetection and overtreatment of screen-detected prostate cancer.[29]

If easier, cheaper, and more frequent access to more cancer specialists were to lead to increased detection of low-risk disease and more aggressive treatment, then access to Medicare may tend to harm some men with indolent cancers and increase system costs. Our results cast doubt on such effects in the context of prostate cancer; no substantive evidence was found that access to Medicare led to harmful differences in care.

Our study has several major limitations, given the narrowness of our study in one condition and using data that span only a little more than 1 in 4 cancer patients nationally without sufficient power to understand regional variations in the relationship of interest.

However, the most important limitation is that our study was unable to observe prior insurance coverage, the quality of such coverage, prior health care treatment, educational levels, income, prior health status, and utilization of the comparison groups, among other relevant variables.

In consequence, while these negative results appear to imply that with prostate cancer neither the beneficial effects of more generous insurance nor the detrimental effects of overutilization were as pronounced as hypothesized, we refrain from claiming these results to be causal.

Indeed, further study is needed to build on the seminal work of Card and colleagues[8,9] to better understand the objective impact of insurance on treatment and outcomes in near elderly. Further study is also needed to understand whether subjective benefits of greater access and coverage nevertheless contribute to patient well-being.

Table A1.

Prostate Cancer, Baseline Characteristics of Pre-Medicare and Post-Medicare Eligible Cohorts.

	No. (%)		P value
	63- to 64-year cohort (n = 13882)	65- to 66-year cohort (n = 14774)
Age at diagnosis, mean (SD), y	63.5 (0.5)	65.5 (0.5)	<.001
Caucasian	10 965 (79.0)	11°575 (78.4)	.03
Black	1813 (13.1)	1893 (12.8)
Asian Pacific Islander	603 (4.3)	743 (5.0)
Other or unknown	501 (3.6)	563 (3.8)
Hispanic	1122 (8.1)	1414 (9.6)	<.001
Never married	1259 (9.1)	1344 (9.1)	<.001
Married	9865 (71.1)	10°399 (70.4)
Separated	107 (0.8)	111 (0.8)
Divorced	991 (7.1)	931 (6.3)
Widowed	323 (2.3)	426 (2.9)
Unknown marital status	1337 (9.6)	1563 (10.6)
Diagnosed, y
2004	3131 (22.6)	3551 (24.0)	.004
2005	2965 (21.4)	3240 (21.9)
2006	3752 (27.0)	3816 (25.8)
2007	4034 (29.1)	4167 (28.2)
Registries[a]			.002
California[b]	3095 (22.3)	3465 (23.5)	.02
New Jersey	1958 (14.1)	1872 (12.7)	<.001
Los Angeles	1375 (9.9)	1577 (10.7)	.03
Detroit, metropolitan area	907 (6.5)	916 (6.2)	.25
Seattle, Puget Sound	898 (6.5)	882 (6.0)	.08

Not shown: Alaska natives, Atlanta, Connecticut, Hawaii, Iowa, Kentucky, Louisiana, New Mexico, rural Georgia, San Francisco/Oakland, San Jose-Monterey, Utah.

Excluding Los Angeles, San Francisco/Oakland, and San Jose-Monterey.

Table A2.

Prostate Cancer, Extent of Disease, Staging, and Risk Stratum.

	No. (%)		P value
	63- to 64-year cohort (n = 13882)	65- to 66-year cohort (n = 14774)
PSA, median (IQR), ng/mL	6.1 (4.6-9.2)	6.3 (4.7-9.8)	<.001
≤4	1770 (12.8)	1691 (11.5)	.001
≤10	9577 (69.0)	9832 (66.6)	<.001
>10, ≤20	1511 (10.9)	1784 (12.1)	.002
>20	1132 (8.2)	1309 (8.9)	.03
Gleason score, median (IQR)	6 (6-7)	7 (6-7)	.04
≤6	6745 (48.6)	7001 (47.4)	.04
7	5050 (36.4)	5432 (36.8)	.49
≥8	1644 (11.8)	1857 (12.6)	.06
AJCC clinical T stage
T1	5093 (36.7)	5689 (38.5)	.03
T2	7178 (51.7)	7447 (50.4)
T3	1122 (8.1)	1111 (7.5)
T4	156 (1.1)	162 (1.1)
Unknown	331 (2.4)	357 (2.4)
Stage categories[a]
Screen-detected disease, T1c	4853 (35.0)	5411 (36.6)	.003
≤T2a	8956 (64.5)	9888 (66.9)	<.001
T2b	294 (2.1)	292 (2.0)
≥T2c	4632 (33.4)	4591 (31.1)
Risk stratum
Low[b]	3835 (27.6)	4034 (27.3)	.04
Intermediate[c]	3048 (22.0)	3404 (23.0)
High[d]	5876 (42.3)	6073 (41.1)
Unclassifiable[e]	1123 (8.1)	1263 (8.6)

Note. PSA = prostate-specific antigen; IQR = interquartile range; AJCC = American Joint Committee on Cancer.

Clinical T2 NOS staged disease classified as ≤T2a; unknown stage disease as ≥T2c.

Stage ≤ T2a, and PSA ≤ 10 ng/mL, and Gleason score ≤ 6.

Stage = T2b, or PSA > 10-20 ng/mL, or Gleason score = 7.

Stage ≥ T2c, or PSA > 20 ng/mL, or Gleason score = 8-10.

Stage ≤ T2a, missing PSA and/or missing Gleason: not meeting low-risk stratum criteria.